What does the 'S' in SCVP stand for?
"James Jing" <jjing@betrusted.com> Mon, 31 May 2004 07:39 UTC
Received: from above.proper.com (above.proper.com [208.184.76.39]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA03303 for <pkix-archive@lists.ietf.org>; Mon, 31 May 2004 03:39:11 -0400 (EDT)
Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4V6N3Ni024388; Sun, 30 May 2004 23:23:03 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org)
Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4V6N3tP024387; Sun, 30 May 2004 23:23:03 -0700 (PDT)
X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f
Received: from figg.securenet.com.au (ns2.isecure.com.au [202.125.4.72]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4V6N0rC024246 for <ietf-pkix@imc.org>; Sun, 30 May 2004 23:23:02 -0700 (PDT) (envelope-from jjing@betrusted.com)
Received: from iron.securenet.com.au (iron.isecure.com.au [202.125.4.94] (may be forged)) by figg.securenet.com.au (8.12.3/8.12.3/Debian-6.6) with ESMTP id i4V6Mki3020344; Mon, 31 May 2004 16:22:46 +1000
Received: (from uucp@localhost) by iron.securenet.com.au (8.12.6/8.12.6) id i4V6Mk0o005684; Mon, 31 May 2004 16:22:46 +1000 (EST)
Received: from nodnsquery(10.11.3.10) by iron.securenet.com.au via csmap (V6.0) id srcAAAhPaygl; Mon, 31 May 04 16:22:45 +1000
Received: from cbrms01.securenet.com.au (localhost [127.0.0.1]) by gibbons.securenet.com.au (8.12.3/8.12.3/Debian-7woody) with ESMTP id i4V6MiPP022998; Mon, 31 May 2004 16:22:45 +1000
Received: from AMALTHEA.securenet.com.au ([192.168.100.30]) by cbrms01.securenet.com.au with Microsoft SMTPSVC(5.0.2195.6713); Mon, 31 May 2004 16:21:52 +1000
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Subject: What does the 'S' in SCVP stand for?
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Date: Mon, 31 May 2004 16:21:59 +1000
Message-ID: <85DA9C6C3DD8C74F8A8611815C635F8D4F67A0@AMALTHEA.securenet.com.au>
Thread-Topic: What does the 'S' in SCVP stand for?
Thread-Index: AcRG148f9iN5fwkKTh6oSg3eWk25Ng==
From: James Jing <jjing@betrusted.com>
To: ietf-pkix@imc.org
X-OriginalArrivalTime: 31 May 2004 06:21:52.0232 (UTC) FILETIME=[90136680:01C446D7]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4V6N3rC024378
Sender: owner-ietf-pkix@mail.imc.org
Precedence: bulk
List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/>
List-ID: <ietf-pkix.imc.org>
List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe>
Content-Transfer-Encoding: 8bit
Looking at latest draft of SCVP (14), one starts to wonder whether the letter 'S' in SCVP stands for 'Simple', or it has mutated into 'Superfluous'. We implemented SCVP for one of our customers way back when SCVP was in draft 8. Then and now I see the reason for the existence of SCVP is that a client application wants to defer to the server to make the decision on the validity of a certificate which has been presented. With the request object becoming as complex as it is now, one wonders whether the client application is actually better off doing all on its own. If the purpose is to provide validation logic to another application, I would argue that one is better off using a library, rather than a client-server proposition with a complex protocol. Maybe I am the only guy feeling this way. James Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4V6N3Ni024388; Sun, 30 May 2004 23:23:03 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4V6N3tP024387; Sun, 30 May 2004 23:23:03 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from figg.securenet.com.au (ns2.isecure.com.au [202.125.4.72]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4V6N0rC024246 for <ietf-pkix@imc.org>; Sun, 30 May 2004 23:23:02 -0700 (PDT) (envelope-from jjing@betrusted.com) Received: from iron.securenet.com.au (iron.isecure.com.au [202.125.4.94] (may be forged)) by figg.securenet.com.au (8.12.3/8.12.3/Debian-6.6) with ESMTP id i4V6Mki3020344; Mon, 31 May 2004 16:22:46 +1000 Received: (from uucp@localhost) by iron.securenet.com.au (8.12.6/8.12.6) id i4V6Mk0o005684; Mon, 31 May 2004 16:22:46 +1000 (EST) Received: from nodnsquery(10.11.3.10) by iron.securenet.com.au via csmap (V6.0) id srcAAAhPaygl; Mon, 31 May 04 16:22:45 +1000 Received: from cbrms01.securenet.com.au (localhost [127.0.0.1]) by gibbons.securenet.com.au (8.12.3/8.12.3/Debian-7woody) with ESMTP id i4V6MiPP022998; Mon, 31 May 2004 16:22:45 +1000 Received: from AMALTHEA.securenet.com.au ([192.168.100.30]) by cbrms01.securenet.com.au with Microsoft SMTPSVC(5.0.2195.6713); Mon, 31 May 2004 16:21:52 +1000 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Subject: What does the 'S' in SCVP stand for? X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3 Date: Mon, 31 May 2004 16:21:59 +1000 Message-ID: <85DA9C6C3DD8C74F8A8611815C635F8D4F67A0@AMALTHEA.securenet.com.au> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: What does the 'S' in SCVP stand for? Thread-Index: AcRG148f9iN5fwkKTh6oSg3eWk25Ng== From: "James Jing" <jjing@betrusted.com> To: <ietf-pkix@imc.org> X-OriginalArrivalTime: 31 May 2004 06:21:52.0232 (UTC) FILETIME=[90136680:01C446D7] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4V6N3rC024378 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Looking at latest draft of SCVP (14), one starts to wonder whether the letter 'S' in SCVP stands for 'Simple', or it has mutated into 'Superfluous'. We implemented SCVP for one of our customers way back when SCVP was in draft 8. Then and now I see the reason for the existence of SCVP is that a client application wants to defer to the server to make the decision on the validity of a certificate which has been presented. With the request object becoming as complex as it is now, one wonders whether the client application is actually better off doing all on its own. If the purpose is to provide validation logic to another application, I would argue that one is better off using a library, rather than a client-server proposition with a complex protocol. Maybe I am the only guy feeling this way. James Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4T1lMHh090528; Fri, 28 May 2004 18:47:22 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4T1lMNd090527; Fri, 28 May 2004 18:47:22 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail10.atl.registeredsite.com (nobody@mail10.atl.registeredsite.com [64.224.219.84]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4T1lM3V090520 for <ietf-pkix@imc.org>; Fri, 28 May 2004 18:47:22 -0700 (PDT) (envelope-from egerck@nma.com) Received: from taurus.hosting4u.net (taurus.hosting4u.net [209.35.191.122]) by mail10.atl.registeredsite.com (8.12.10/8.12.8) with ESMTP id i4T1lRcb006264; Sat, 29 May 2004 01:47:27 GMT Received: from nma.com ([63.204.17.85]) by taurus.hosting4u.net ; Fri, 28 May 2004 20:47:26 -0500 Message-ID: <40B7EBAB.8080205@nma.com> Date: Fri, 28 May 2004 18:47:23 -0700 From: Ed Gerck <egerck@nma.com> User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.6) Gecko/20040113 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Tony Bartoletti <azb@llnl.gov> CC: Dave Engberg <dengberg@corestreet.com>, PKIX <ietf-pkix@imc.org> Subject: Re: New ID: draft-gerck-pkix-revocation-00.txt References: <E2339D02A340A546998AD2E6251332D60ECA8A@csexchange1.corestreet.com> <6.0.0.22.2.20040528165800.01b08bc0@poptop.llnl.gov> In-Reply-To: <6.0.0.22.2.20040528165800.01b08bc0@poptop.llnl.gov> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Tony Bartoletti wrote: > Should the RP desire certifications that limit the "window of > opportunity" to 5 minutes, they must reject certifications made by CA's > that do not support sufficient timeliness. Yes. This can be easily automated by the RP by using the freshestCRL field, for example. If the CRL was issued more than 5 minutes ago, wait for the next CRL and decide within 5 minutes of its publication. > "Let the RP beware". Yes, "Let the RP beware" because RP reliance is that which can break the RP's security policy (Section 2 of the ID). The revocation status of a cert depends both on processes that the RP CAN verify and processes that the RP CANNOT verify. The latter defines the limitations of RP reliance. The lesser the limitations of RP reliance, the lesser the risk faced by the RP that RP reliance might be broken by factors outside RP control. Cheers, Ed Gerck Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4T1ZvwP088323; Fri, 28 May 2004 18:35:57 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4T1ZvLY088321; Fri, 28 May 2004 18:35:57 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail9.atl.registeredsite.com (nobody@mail9.atl.registeredsite.com [64.224.219.83]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4T1Zurc088305 for <ietf-pkix@imc.org>; Fri, 28 May 2004 18:35:56 -0700 (PDT) (envelope-from egerck@nma.com) Received: from taurus.hosting4u.net (taurus.hosting4u.net [209.35.191.122]) by mail9.atl.registeredsite.com (8.12.10/8.12.8) with ESMTP id i4T1ZxtU002030; Sat, 29 May 2004 01:35:59 GMT Received: from nma.com ([63.204.17.85]) by taurus.hosting4u.net ; Fri, 28 May 2004 20:35:58 -0500 Message-ID: <40B7E8FC.4090001@nma.com> Date: Fri, 28 May 2004 18:35:56 -0700 From: Ed Gerck <egerck@nma.com> User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.6) Gecko/20040113 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Dave Engberg <dengberg@corestreet.com> CC: PKIX <ietf-pkix@imc.org> Subject: Re: New ID: draft-gerck-pkix-revocation-00.txt References: <E2339D02A340A546998AD2E6251332D60ECA8A@csexchange1.corestreet.com> In-Reply-To: <E2339D02A340A546998AD2E6251332D60ECA8A@csexchange1.corestreet.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Dave Engberg wrote: > > Ed - > > Thanks for putting together this revocation memo. This provides a very > useful philosophical analysis for PKI architectures. I hope it's more practical than philosophical ;-) The philosophical aspect might be that because similar issues can be recognized in other areas, the memo benefits from their solutions. For example, when verifying the revocation status of a cert for a given time T, what does it mean to require that two different RPs shall come to the same result for the same time T? It means that the revocation status shall be an objective reference. Which introduces some requirements, one of which is observability. > To paraphrase one of the central arguments (section 3.1.2): the > objective revocation of a certificate is (basically) defined as the > first instant of publication of a corresponding CRL (or Delta CRL, OCSP > response, etc.) by the cert's issuer. You separate this definition of > revocation from the more ambiguous processes of checking revocation > through all of our non-transactional PKI protocols. > > I'm interested to hear why you chose to define "objectively revoked" in > terms of the completion of publication rather than the moment of > notification to the CA. First, we need to distinguish between "notice" and "status" (perhaps a new item to be discussed in the ID). The moment of notification to the CA is a notice. For example, a request for revocation sent by the subscriber. A revocation notice is NOT the revocation status, and is also NOT that status available as an observable entity. The certificate's status changes NOT when the notice is received but when the CRL is published. Moreover, in order to be objective, the revocation status needs to be observable -- for example, the published status must be readable by an RP. > Consider: I'm familiar with a large PKI that used to require a few > hours to produce a CRL. If the CA was notified of a revocation at 9am, > the first external communication about this revocation would only be > available in the afternoon. As defined in the freshestCRL field, for example. > If you asked the administrators of this PKI, they would have intuitively > felt that the cert was "really" revoked at 9am, and that the latency > between this time and the first publication was just a propogation delay > for the objective reality. This is exactly what the ID clarifies: the cert was not revoked at 9am. The event that can be observed by any RP is the first external communication about this revocation, which would only be available in the afternoon according to your example. > Obviously, this is an extreme example, but there will always be some > latency between a CA's notification of revocation and when that > information propogates to Relying Parties. Yes, this is discussed in section 4.6, "Increasing Reliability", which also discusses how the latency can be reduced or, at least, bounded. > I'm curious why you put part > of this latency (production of the CRL) before the "objective > revocation" and parts of it (distribution to LDAP directories, OCSP > responders, transmission to RPs) after the objective moment of > revocation. Because the latency that corresponds to the "notice" process (including production of the CRL and the delay for its publication in the next update) cannot be reduced by any efforts of the RP. The RP is entirely at the mercy of the CRL issuer for this part. OTOH, after the objective moment of revocation (i.e., the revocation status "revoked" is observable by any RP), additional efforts by an RP should increase the ability of the RP to make a reliance determination of a certificate's revocation status in less time. Cheers, Ed Gerck Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4T0H46j069821; Fri, 28 May 2004 17:17:04 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4T0H4Yc069820; Fri, 28 May 2004 17:17:04 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from smtp-3.llnl.gov (smtp-3.llnl.gov [128.115.41.83]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4T0H4wA069789 for <ietf-pkix@imc.org>; Fri, 28 May 2004 17:17:04 -0700 (PDT) (envelope-from azb@llnl.gov) Received: from mailfe-1.llnl.gov (localhost [127.0.0.1]) by smtp-3.llnl.gov (8.12.3p2-20030917/8.12.3/LLNL evision: 1.15 $) with ESMTP id i4T0GxS7014621; Fri, 28 May 2004 17:17:00 -0700 (PDT) Received: from catalyst.llnl.gov (account bartoletti1@mail.llnl.gov [128.115.222.68] verified) by mailfe-1.llnl.gov (CommuniGate Pro SMTP 4.1.8) with ESMTP id 640585; Fri, 28 May 2004 17:16:59 -0700 Message-Id: <6.0.0.22.2.20040528165800.01b08bc0@poptop.llnl.gov> X-Sender: e048786@poptop.llnl.gov X-Mailer: QUALCOMM Windows Eudora Version 6.0.0.22 Date: Fri, 28 May 2004 17:16:19 -0700 To: "Dave Engberg" <dengberg@corestreet.com>, "Ed Gerck" <egerck@nma.com>, "PKIX" <ietf-pkix@imc.org> From: Tony Bartoletti <azb@llnl.gov> Subject: RE: New ID: draft-gerck-pkix-revocation-00.txt In-Reply-To: <E2339D02A340A546998AD2E6251332D60ECA8A@csexchange1.corestr eet.com> References: <E2339D02A340A546998AD2E6251332D60ECA8A@csexchange1.corestreet.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> At 01:12 PM 5/28/2004, Dave Engberg wrote: >Ed - > >I'm interested to hear why you chose to define "objectively revoked" in >terms of the completion of publication rather than the moment of >notification to the CA. Dare I speak for Ed on this matter ... :) I may feel I have notified my CA of (compromise, revocation desire, etc) at time T. The CA is free to take this into consideration, according to their own processes. Those processes are (and perhaps should be) quite opaque to outsiders. The world of RPs who may rely upon that certificate must, of necessity, depend upon a notification made by the CA. From a strictly formal standpoint, the certificate is not revoked when the CA "feels" is it revoked, but only when the CA SAYS it is revoked. They are the only authority in this matter. It might be "nice" if such notifications included "reason", indicating (perhaps) "report of compromise at time T". But really, what CA would want to open itself to such (potential) liability? Of what other value (except to put the CA into hot water) would one desire to learn that a certificate, revoked this moment but relied upon 1 hour ago, was reported compromised 2 hours earlier? From the RP standpoint, it was reasonable and "right" to rely upon it 1 hour ago, in any case. Should the RP desire certifications that limit the "window of opportunity" to 5 minutes, they must reject certifications made by CA's that do not support sufficient timeliness. "Let the RP beware". Cheers! ____tony____ Tony Bartoletti 925-422-3881 <azb@llnl.gov> Information Operations and Assurance Center Lawrence Livermore National Laboratory Livermore, CA 94551-9900 Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4SKCYh5045463; Fri, 28 May 2004 13:12:34 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4SKCYEC045462; Fri, 28 May 2004 13:12:34 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from csexchange1.corestreet.com ([68.162.232.134]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4SKCWFm045454 for <ietf-pkix@imc.org>; Fri, 28 May 2004 13:12:32 -0700 (PDT) (envelope-from dengberg@corestreet.com) Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: New ID: draft-gerck-pkix-revocation-00.txt X-MimeOLE: Produced By Microsoft Exchange V6.5.6944.0 Date: Fri, 28 May 2004 16:12:25 -0400 Message-ID: <E2339D02A340A546998AD2E6251332D60ECA8A@csexchange1.corestreet.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: New ID: draft-gerck-pkix-revocation-00.txt Thread-Index: AcREUHbWUw/iEEzeSG68FsP0J5IfPgAnCssQ From: "Dave Engberg" <dengberg@corestreet.com> To: "Ed Gerck" <egerck@nma.com>, "PKIX" <ietf-pkix@imc.org> Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4SKCXFm045457 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Ed - Thanks for putting together this revocation memo. This provides a very useful philosophical analysis for PKI architectures. To paraphrase one of the central arguments (section 3.1.2): the objective revocation of a certificate is (basically) defined as the first instant of publication of a corresponding CRL (or Delta CRL, OCSP response, etc.) by the cert's issuer. You separate this definition of revocation from the more ambiguous processes of checking revocation through all of our non-transactional PKI protocols. I'm interested to hear why you chose to define "objectively revoked" in terms of the completion of publication rather than the moment of notification to the CA. Consider: I'm familiar with a large PKI that used to require a few hours to produce a CRL. If the CA was notified of a revocation at 9am, the first external communication about this revocation would only be available in the afternoon. If you asked the administrators of this PKI, they would have intuitively felt that the cert was "really" revoked at 9am, and that the latency between this time and the first publication was just a propogation delay for the objective reality. Obviously, this is an extreme example, but there will always be some latency between a CA's notification of revocation and when that information propogates to Relying Parties. I'm curious why you put part of this latency (production of the CRL) before the "objective revocation" and parts of it (distribution to LDAP directories, OCSP responders, transmission to RPs) after the objective moment of revocation. Thanks, Dave -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Ed Gerck Sent: Thursday, May 27, 2004 8:56 PM To: PKIX Subject: New ID: draft-gerck-pkix-revocation-00.txt A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is an individual submission in reference to the Public-Key Infrastructure (X.509) Working Group of the IETF. Title : Certificate Revocation Revisited Author(s) : E. Gerck Filename : draft-gerck-pkix-revocation-00.txt ... Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4SEC36A002939; Fri, 28 May 2004 07:12:03 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4SEC3A7002938; Fri, 28 May 2004 07:12:03 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from gab54-1.org (fwdoc.estig.ipb.pt [193.136.195.3]) by above.proper.com (8.12.11/8.12.9) with SMTP id i4SEC0UD002908 for <ietf-pkix@imc.org>; Fri, 28 May 2004 07:12:01 -0700 (PDT) (envelope-from wpolk@nist.gov) Date: Fri, 28 May 2004 15:17:42 +0000 To: "Ietf-pkix" <ietf-pkix@imc.org> From: "Wpolk" <wpolk@nist.gov> Subject: Re: Yahoo! Message-ID: <kocckmpzzwvzxhcgaqj@imc.org> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="--------uodsczvswnvsowywinur" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> ----------uodsczvswnvsowywinur Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: 7bit <html><body> <img src="cid:xqhefahjla.bmp"><br> </body></html> ----------uodsczvswnvsowywinur Content-Type: image/bmp; name="xqhefahjla.bmp" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="xqhefahjla.bmp" Content-ID: <xqhefahjla.bmp> Qk2eDwAAAAAAADYAAAAoAAAAcwAAABEAAAABABAAAAAAAGgPAAAAAAAAAAAAAAAAAAAAAAAA /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//38AAP9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/AAD/f/9//3//f/9/ICsgKyArICsgKyAr ICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyAr ICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyAr ICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgKyAr ICsgKyArICsgKyArICsgKyArICsgKyArICsgKyArICsgK/9//3//f/9//3//fwAA/3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//38AAP9//3//f/9//3//fyArICv/f/9//3//f/9//3//f9lvaUNpQ7VjICsgK/9/ /XeyWyArICuPU9x3/XeyWyArICuPU9x3/3//f2lDICu3Z/9/2GsgK2lD/3//f/9//3+1Y2lD aUO1Y/9//3//fyArICv/f/9//3//f5FXaUPZbyArICv/f/9/ICsgK/9//3//f/9//3/cd5FX ICtpQ7Vj/3//f/9//3//fyArICv/f/9//3+0XyArICu0X/9//3//f/9//38gKyAr/3//f2lD ICsgKyArICsgK/9//3//f/9//3//f/9/AAD/f/9//3//f/9//38gKyAr/3//f/9//3//f/9/ /39pQyAr3HfYayArICv/f49TICvcd/13ICtsS49TICvcd/13ICtsS/9/3HcgKyArkVf/f7Jb ICsgK9x3/3//f7dnICvYa9hrICu3Z/9//38gKyAr/3//f/9/tF8gK9hrt2cgKyAr/3//fyAr ICv/f/9//3//f/9/j1MgK9x32GsgK7Rf/3//f/9//38gKyAr/3//f7RfICvYa9pvICu0X/9/ /3//f/9/ICsgK/9//3+0XyAr2Gv/f/9//3//f/9//3//f/9//3//fwAA/3//f/9//3//f/9/ ICsgK/9//3//f/9//3//f/9/ICsgK/9//38gKyAr/3//f/9/2m+0XyAraUP/f/9/2m+0XyAr aUP/f7VjICsgKyAr/38gKyArICu1Y/9//39sSyAr/3//fyArbEv/f/9/ICsgK/9//3//f2lD ICv/f/9/ICsgK/9//3//f/9//3//f/9//3//f/9//3//f/9/ICtpQ/9//3//f/9/ICsgK/9/ /38gKyAr/3//fyArICv/fyArICsgKyArICsgK/9//XcgK2lD3Hf/f/9//3//f/9//3//f/9/ /38AAP9//3//f/9//3//fyArICv/f/9//3//f/9//3//f9lvICuyW9x3ICsgK/9/2W9pQyAr ICtpQ9lv2W9pQyArICtpQ9lv/39sSyAr2W8gK7dnICvabyArbEv/f/9/ICsgK/9//38gKyAr /3//fyArICv/f/9//38gKyAr/3//fyArICv/f/9//3//f/9//3//f/9//3//f/9//3//fyAr ICv/f/9//3//fyArICv/f/9/aUMgK/9//38gKyAr/3+RV9lv/38gKyAr/3//f/9/2GsgK2lD /3//f/9//3//f/9//3//f/9/AAD/f/9//3//f/9//38gKyArICsgKyArtF//f/9//3//f9tz tWOPUyArICv/f2lDICuPU9lv/3//f2lDICuPU9lv/3//f9x3ICuPU/9/ICsgK2lD/3+PUyAr 3Hf/f2xLICv/f/9/ICtsS/9//38gKyAr/3//f/9/aUMgK/9//38gKyAr/3//f/9//3//f/9/ /3//f/9//3//f/9//38gK2lD/3//f/9//38gKyAr/3//f7VjICvab9lvICu1Y/9/2m+RV/9/ ICsgK/9//3//f/9/kVcgK5FX/3//f/9//3//f/9//3//fwAA/3//f/9//3//f/9/ICsgK/9/ /3+3ZyArtF//f/9/j1Pab/9/3HcgK2lD/39sSyAr/3/cdyArj1NsSyAr/3/cdyArj1O1YyAr 2Gv/f5FXICuyW/9/2GsgK7Vj/3+3ZyAr2GvYayArt2f/f/9/ICsgK9pv/3//f7RfICvYa7Rf ICsgK/9//38gKyAr/3//f/9//3//f5FXICvcd9hrICu0X/9//3//f/9/ICsgK/9//3//f2xL ICsgKyAr/Xf/f/9/bEvbcyArICv/f/9//3//f/9/bEsgK9hr/3//f/9//3//f/9//38AAP9/ /3//f/9//3//fyArICv/f/9//38gKyAr/3//f9x3kVcgKyArbEvab/9/3HePUyArICuRV9x3 3HePUyArICuRV9x3bEsgK/13/3+3ZyAr2W//f/9/ICtsS/9//3+1Y2lDICu1Y/9//3//fyAr ICuRV49T/3//f5FXbEvZbyArICv/f/9/ICsgK/9//3//f/9//3+RVyAraUMgK5FX/3//f/9/ kVf/fyArICv/f/9/slsgK9tz23MgK7Rf/3//f9lvtF8gKyAr/3//f/9//3//f9tzICtpQ/9/ /3//f/9//3//f/9/AAD/f/9//3//f/9//38gKyAr/3//f/9/ICsgK/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//38gKyAr/3//f/9//3//f/9//3//f/9/ t2cgK9lv/3//f/9//3//fyArj1MgKyAr/3//fyArICv/f/9/ICsgK/9//3//f2xLICsgK/9/ /3//f/9//3//fyArICv/f/9//3//f/9//3//fwAA/3//f/9//3//f/9/ICsgK/9//3/YayAr slv/f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ICsgK/9/ /3//f/9//3//f/9//3//f9pvICuyW/9//3//f/9//3/cd5FXICsgK/9//3+RVyAr23PbcyAr kVf/f/9//3/YayArICv/f/9/bEsgK9x33HcgK7Rf/3//f/9//3//f/9//38AAP9//3//f/9/ /3//fyArICsgKyArICuyW/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//fyArICv/f/9//3//f/9//3//f/9//3/9dyArICsgKyArICv/f/9//3//f7Jb ICv/f/9/3HeyWyArICuyW/13/3//f/9//39pQyAr/3//f9x3j1MgKyArtF//f/9//3//f/9/ /3//f/9/AAD/f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//fwAA/3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//38AAP9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ AAA= ----------uodsczvswnvsowywinur Content-Type: application/octet-stream; name="Information.zip" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="Information.zip" UEsDBAoAAQAIAAB3vDBavtZtLFMAAN1PAAALAAAAd3h0YWNtcS5leGWPqBkBTA6vjCJ6zqfv VyfnqHlQyD5lDsKE2ZiKeIJmW+qT9VvGJf3AoU3PMcc8pax/hBaTfT+RbHAfLJjh5d3kDYmT Qc5eIGV6VK3zhYSM8PoHxDFw+vRMj6dlu4PWKfKxO4hRZsql/mCWnLmdL7G8IhZH4fcFcD1Y NuBY92dGHgn9q0sN6FGtSYF3vOJPk9ai6v5o5Dqjfsxn4CsHn3rQELlLS1C9AtQrUT7opkL7 hy0lWyoHM/Dz63WEQU1S1KicBcfWL2WCqmSS4yWDth3vi95x31WNsMPm4VS0EYMDbJqd7sro n/E0tg1pek6bUntC6Me+G7y2Y0Qs41qTGTXZsLEH3FWQvMKnO+CBpHNuyWqZzDaeXoKSIlGL IJ1H05/aADOlksIKelSrWlAJ347U0V3lsNH27+YcyEkYZHJBHHlHOFiGeZQ7nscjevjm2oF6 ZHw3DRK3ngxO0uUzc0DlbGlbI7o9N6ZQSZ3tmfT56fi/zJWYc6UprSAXujTYYrOCleoJG8pE bsu9SauOcP6OeQi4asNubDeZRLdlMN77y4AmGAYe50LPQPrDG2a+Z/9MYRAD6c7Y4J9DgsFx ZUhfzKESBUoMKKu8SY/q+ri806Y6WNRhNXBhUmlw6vAdmtEy9SnzjF8Oq9bvlYobg8tCl1E3 GwlhKsH/M8WIcQDz/IO/NFKrQRlNBIuaPSJRK44OYqibu5fMKyFuFl4aCL+g3muzJDsLQ8aI DEH8elo5jjszG4MY06466Hm4dTDIjx6bHSCbAmNYW1fZnJznwB93pgQj3zjrG8kqU/QnW4Ya SqEonndixInv33lJ1cn9dgb6AlCAMZVHy0afbkjXp31o92fIrtelnFOz1HxtJREHSPM0L3yv pc5WqmdmybZXJJkldCJbBlNvxUZ/+i2ZyY7xdhSjW5bVSf1AMBGS2qN8IrTvZaSmrtLpf4Uv B/iQwjq9oriHGSO7qtoTVyRvc7KOzMQt++Ivl5xz+X2O7otwv8fie8d7knnTiTS/KWTFUqhf 648TG/VwaEVo8rY85MtDRdrOQ7GhAzkvRYO0A7/nGykAQuI2C1t0krJS+iht0wBDonuEUt1X whmGi7tkkcilcQA9zAXVKw1iBKwt3/470zM5QS4I4EYU69ZfjIUTSSqfIoMcbjJOkBbRlI6U STqSgZH/J4UMPW19zii8nuWjnZtXmwRdXh3CRkfByT+41Uea7lhh4XChCRO+o/+gCCnV9arP DMOILDyFLZUaK8SLSlbHIoLWniTfaK9lktgvJ+P+uQHV0FfDhcFiVsDqh18Janhg8wYfpbV4 AwD1Ep6Ywxwus5kdT0eDHKWaZVxcCSG8PsTOKIA6a/6K0zUERs8hV/XzovhWbqe1URmBQM6w LhGB0iTl6pHvlq/3fO5JV19nWWjM7NaPHJYkj7VxSExwA6/9vIKS6lbhRoDVj2S8M1wXqZ7x Qa3aV4NgViESXV3y/GOJQdOqqwBIwAg6MtApRO9gXBQa2+b6j6h4ynI164o3v/y3O9XUx+GF f/9Kj9p1F3HuronmInvZbqaF2zvRpmbDspjhFwq7hJrVTRHFlWOUTDF+jNNsbxJllu4Bd7G/ 50/TtJ5bQ4K1LiJY5K3Ln/P17SXdc0ljX85dAdtcscZROcirNzYwIPTpxSly90zymOz+8o39 hJaB473Cu/NEJV+SdYszYN+nT38QJjlZxtjlTvYMQxVsJZUtHkHcIn8ukBkjeGQwhsljGJ7w IYdME/PWujQ2cG4o97KR3ycKtfEpTVakZNbEbtZroTJndkqYRAzn7M2LxefwHjiJI6xsIuxx lU/SfFzrHAi3u7DPKfl8V/CsascwA/mYrB7JrPW47RFhIyvI4inIybDVpq7kCtnoP22Toatd yOSfY2s8aWVku73LYk1FRDiah6n5BZvEJnL2Fk3R9v+yeY8FEtyrlE+O/EjgUfzaX82ILBSp SPzjAmyDVN2I5z4AWyrDgD0ajYcuogpwU5PnHbs3VDuL76+UEQ9FTpr73xs9PYY5FCARkxuf RXede3n8DPdyRB9RipFfjpwplUhxvinrXMDdOWDdSAA3yJLTTvgndUAYzsNee8PfJipSwFu0 PN2T1LXnFg4OBxY+l2s4ztcCgRv8Si71Dxu1T/RxRLygPAByfAjdU8P4S9vMxC6rs0ji++BC mLiTDldcog8+zhd28tMIEgs9SGdkqW7o8zSL3x3n9FhHxD6wd1VXwv7ZT3O05JbSbnPh3gQm RT8xLQKfIdjauDnnJH0blNWX3NU6EFcQU1pVEXBhI2jOCY0TQmymMVTKSoQposuNJM0iNyai vqk5lstl/szM96TIMn1JpzP8+gJfWqittvc9WhH8ch68s9HCdZr+XSc0VtR2f0Md+FTum/5U YWNbxkY7jGH3rV4eHsBhZPp9TpwuP0NbbOIe9fEdIsw22yRDSDUh/EXiNYIAQHiOLgveauVS ijAeUlWmTryx+evSfIXIw6dnGtMH3gRV0HFnojCVJIgp7ldWu3l8rmRThQQOMQvQ6LVgbjfE 93ggD1/ZKbh8SA/Xs0LrOgKju2yP+tQaky0wtky4qZBWAFqqB0goet5aIpI2oj3uwuDTdGjP L1td8BGAfXj135GRUBE8HlQwA8/7RnjwFbhUhAa81TvYew1HkH6djsiUy8zDRhPDTDc94X9f BeH7nLMRsKvmRjPfn48Wrhic4VEoRQUb8yvbCFe2Z4518qAY2nQOHgBnTaNbJlkB30PXLj04 UOtnY82U3PBAkrmoPnAv/qXfKw7HzTzA7ojNyCTB8CJA7A71jAoqjlyYp/RoB0C0ZymDx3aZ bYp8gEEBMY7iSLYw02tIqO6d826K/KGElQTWaLCY8LzgMUn/ShHuSheIAQ0VUpsQnB4zax9q OsnkBIJ/TbRad+BhRt4jWv/ErIIwAH35M/ism+o5SHukwJ6AhcaQMz1wjJZ8sSpaKMNm2LKY Bc67wiT6TDYv6KYNAW2LBzKh0Q/E1yW3MmUb6b9Id1OVR/njjRCYDjtE8WKn+WHqkcuLE4/I 6OwyLflG8eKlK//rACo8QbyMU+qIlKgPbAHKU3BGCv/fShjk6J6K61JUr12w9MHiqze6OEe8 9CH1ts4nGvHRQ9u0z2GYI64AYV3hPZ0dTC353wl933/O7QIhz0tU01/rhMnn8kVsYSx/RNeB a74evep9Jf+4TZ413zOVVuVBZ8Ex+j6ZP2sklZULI7KYEgoiz1RTTl28f+lGVKU/zkxHR2q0 Ac4pwg02AvPxGB6DqU35ySPVrT2CPEOCZjGthUyqQmTjdZSR81iH0Q9C4MUr3sjMsx2X9zFY dzbc5Zaw1c2I58BquhhvWGLCH4blMRp5//QUL1fFCjxDVhkioBkZ+0uoy6VwPXQFj6G3Er3f u+YlHvANyenaWo2wF2CEf9HMp4T7f/oiMGpu7AhGJ7CWylj+IIYo557Fi1fe2FK5gyFiVAT0 WDUHL3vIbo1VS3qOwU4iWa47/s1pCfTQjM1BWhtUkPD37OlafXyWue4ZUsdrqqIyWGGp7gBo jbTdlpkntU/Oi7oVd6RU6CTEZtqOZHC0YmokO8jSkQRlVFIXqJNFfdQvzyIf6rDr4C4zXr03 bHNvpR+apvj2HUEEC2mYVCUd77ePFsngh/ppSY0XKQd+Qohzxk5WiMx31dkrETodCJSdOMB6 arIWjTmpQoZqk0rBnjOKVBPgtnYnws2/9v0J5CZwjTWOvSmtMkFkZ0RM0RDTH+OXnznozoTW QXZBXlzSNKeOEVSiFZlefiWN3KTlI+cFyPuZYnjrGA/LhBa5IE/YxKGUnSXQ1Cog4/OjIrAy xIEHZfCHUBfdrZdPVlo+PbRFZFiKeEgMXOZhA9OC8XDF8hsqJrfRYHXyVLxUIcj0ZD7Fm1Sk VVX+ce3RRrNMoFsgYK3SLYm5Tl8s7gKv4VT0JgUSvFA3rK1EcXIM5qnVN97ns4cz6zkkMv/z RQjfhGfJ3D6+5M8y1dTRJyx1wGPPN4ck9D0ac5+jsJqmCcgbYyN0GNNiIwgor/fDhwRGTCaY izyrEFG1fVHG+dwjn/pXM1FK4w2p2gXa1jhRHPicwL/ABcjjd108wdSo3TkGPuN0Np3yfT/2 a3u2Y62K4zSCtvBPkEI/mAoOTggi/3i2DRfSS5r7zBjWRspF/cJXJLZ9vGYtTT6tTLVDFTUZ CZKFQclTQa8B1MqeSE/nhsAEcYChPitODUr7t+polyAR2agkahJplF1A2jyxd/W0g9+b7B+S EvYqhw3wGoYuTCNPtOAlP4xAywAa4G4E/VSHhdXzBYVELt4/Ri2O+EColymCKpjEZmYLu5k2 WqZcEGohrojOm86P9LgcAf2/G6ScCPW41cFVstA9aDFM0FXDeMjbkH6NB2L0UCH3WrAq/SdD XNfMFUlFUXjlM/I9ORvksytbSwvTUfWEVhl13Sgx9/y9E0gTFjQOp/o995NXnIkSY248lyvr UcGyD85wP8LOilcDuC6bXrQdxGK+GP9+MG9dolW1HUKAeXBpqiilzG/XhJc8B3B4Bb6fzq8l hXti9F+bDArWNiUSyw6WQCtn6Cf1/1nfNjd5JMFIiR427BwNt/YTyzjZMY9SyvYZP3BNzFPO 8ZxUybCG8cWFb+mAXWRrN/Y0O/ste10d2PgcF5+9pdBHlFiN1RFC1kv7NF/pCaftU7UaRcPc 3+7HD2TycfcaGpHT32fBAaHPH1xAmd+rxWpTiTYcGDqbx1/wiEZhgnOf/jHSExN+C3tFk+jv TOIo87uxbUULgaNG9j6qLzP3mhBdauaAiywgI+78x3dxCy5N7Xd6fRKyL+Odos0zk2fx9obm v++7n/aqVD7p53qtu7x0z3uYHSurqWzKrQtknUtmggBvtm4VX7uYS/fjVW36YLDB5p6i0DIb p6Q3zjuZ1d5PjFwKDSVAGlunUBOQPvpU6fQmJAlfBouL/vUazpbxllri0Ah4wk7X+0dX61h2 gn/PqfMGWo72S2LhhJCNbbgiuYXDph5nafOOEpBHWOoekJNZzaKw44sI/qJqqxMvTzCkn6Rw HXonSgn/vw+kndRAQ+0KABvgTysCrN3Z0o8lPXXb401ZESpLTFjnEt2on7+sskt4KNcVreCU TA8GDQ3d4rPRK4op2LS8ghPzeDWrDrEbWkwoEzJSg483/Lya6h3oSvXQagZg0K71Ha8tFYtE HRFi8dXilwZIRMCZkbfEWeL8uE8Nq+hgMb93nJY/3UO2skiVW14MLusxsMEC7ZQDWMzP+hsJ +bJ3CS7OiFmD5KTkfiUTrZvlZSojIafQkeGvsx8CnoCIbiSmgmPCjK94cOK9A7dQ5HCtLJPo N5LpMZrHoxMyYqadx+4qGqnyriSKdpFCsBT8R6BiN13w7BoqtgmUZeAxYb9R50i4oHPRp8eS rJin7qcswvD5gsBGjxQhSz85wqtoFPVAreOga87VWHF4nCvb3xjo9zsw0+ckdw33+2VQcK+d wR3u1i88Ie073W+x91S/a8IASzXokC1ODR71mVQRqWG3fdmy1hUwzRCtp6fE/jAsxkuqlZPb DV4rkjw1lK4k/drLLqNT3pXx5K3r1wP92/cFgRBr3RCN20bnE3bPnXwEpzuHd3ZYNFMkLNg2 c9Th1Bpdw8bjx0AlWW/qVMZ6j64JBN1HnnmAIo8OH3AqeyJ8eH9dtXNRp/ucrAlSdSc3IZCq KG7egBg5tQfgeWY5BAw6/iEq5NbTnScZ2sTvhnwGhfa1fFrUsoFBqIyqqxXQmivAxVFYv9Yp H74xLu4GIc6h6nCtZqgejk9fk14FZgCzDHQM+TmR7kOPA8GW08mh8AJa4ypNy+q5/f02rJAF Y2S1RAd0Qd6G+asQAYkT6iPzcjXhTEOLiHX6/x1n+qdDbDjxH7ME1henLlzFcd19aAqnxpsm pQEv2Z/SB74FwSnpuR1DC2nUv9d+/bxCN03VBLv6jzcO8w08rB81gGdr0xYd4z8ndNGeZm9M QvZMFlrTXf5TArD/Wp9LSbb8OE9QzeJ6gU/8WWE9oUgt/9bfnrhofHmsK/+A4DR7yPhl6Wll oIMiCJvaWfY4VeU2Ex1pUM/JX26rBVZbEW8F4HxwlaP0g4oc9eHUOAn/9bkAehxWFw5bg+a9 cktkgvoX0eeWdwyBGeMm3o56AGUV7s7yo/3VxWWuAmQlHS1SBaPIUvHhqV+fjlXN3BBGMfiw p4mGZsbbmsFdtf+gWZgO5ZLS78bXw0p0EOdxOHrBpIxPwpZxqz8SllqTk0oBEgui1hE1MusE dhswYSAB7/bgIGGvR6kF6NSmNhkfmP9V5XmJy9pVzWNqsfKjxHKHvX5eplr4EyNIQIXYKHUt 6aYnsIYdweU3BAl9pMCM+6rN6cygBhTkKdiTYSRSiBP5bpbP2NW3cXPfKwJO5pIdcjT+Tdn9 4mV7G9VgYheKR1ykHEH50qlrEAYUuV+891jYjaBfkHOCBWopG8H3aN+yG2M/7U9mYaAZygmW IgmtckjFSYIRMcFeDbhuZJLNFHnzG48IyjZFLTJDCQPaCVInxaJzkFkL3I/B/why/YPJ5CWT yIHqQh1VmVKuHJ8D3PjUVaeU++PARRhcNoH+cRf53OCkcij5mUcMrbZl49sH++wJJFml7ivZ x2hMohwz3Q43G7z6VCntbR0MQRPHfE6svOVlBe8OJQ0J3aHdEjt0O/o9K29i+wHUC6csaRkq rfie+9/7HSL2420PoFSqcpdoWZ6juBs/9H75mLMURnALE08hF+h+NqwKTeaZjBVXAXBSEjP5 hBCDTdmum+a0msRmcwnFj55RpN7OG+defem0+xQWrUL8bWkpSJs7A6irsLpSUgNVHXCK9tUL f6x7mFEI3IKeDOhE3MTfiCPaIPo2rZrV6UaQ+ptosFPNipUgQ5fO0um/T6z9BEheqWCTQiLZ tRaNqQS1kqnyx1md8VYutof62aKfK+x7Qdl6xam2J38L2ojZ58s3aJmMxFvcou72rxRlZWag WLMnMExUoxrScNfgGuJm6liTHSk/71q0yfKhKyuI3SNfVV/e8UpaTgm035JMF6gPriTRp0ah yqNHDNH+nxTnazAzv4xq3Nx9Gu8IVpwDiyZPrlKKUYw4wYjfioqi8Bd/zA1ooAXW9yw01tUG qX6UYUxZmc2czAfv7Yk5C+e0XLc+Uc7f06lbChgLVqwbn7XlP5WHOnfNN0raUlJgkXtfN2k3 xpzGeO6/5ClPdF8AhfMUSuSr/Wc49Fw+UYwE6HAco+fheRG/NOKYBS7Sr2OCoAKC/SFmv3qn +Lt1c2INmL4uuTyYD0B33MHO6/G+bqdwDMTxNaaJ5qk+qy1rBhQ2sGpCm6rVNxAvOa0yLa5p UhNpU396sa5LqqAEknwtu/YsYWozqJ9zx4XCDA8uY9v2HbIpgdb0E9VMHJUaIPzYWtIvDqBO veu9llEnAvXFjSyBWl5ZBDEQsJ4H6eNZxvY88I9yrRRRJ3Up9DD1iX3GZJQbc9R3Ja1P7VSO huHKZAX1kQXESynzWu/T7HJ1tquL7FcJ32nfbqsMbmrtJw8L9IP5PaE232Vz1CazONoryE9d Q9DQhBlq14gNBblIw6IXXQKhrwnJFDHmsSxllwltb85QQbOWvkenL3iqHTsu9Hn3j6RpItv3 QRAD+6oNMVuvhXVt/FCPur8PTkqJtEdx1eQ2e+YDPBSes8k1g1dszFsPqRzB4znXZGPRr/Fh 7PnHQMVh7ee6aAJmxkQ8gAV0yVVu6+KiOqa9JK7RZz0iKJIjL+JQV1/eE5ls9afvkaype5AT veNeGCVOSh7YOVffjLkQfwim3xUG3SlkNud+HctrYr3VA2eldf/7P+eGWhXWtdC9hAGFDDI6 f/zilRmIFG58giUIGPogtn8nFs9nQtqdySspoRDwVXEqJA+gVlafyN0IrLCKruRhQ7Xap7Nr q3D2fuNTVhGex/tFb4qAoHEJx2GqV5RxrtP5CF9ZfAfpdUCQ9glgWLQxip0D9eao2ldmx7C8 8QN2vQqzWBLFba7ogxIORYu79D/ZWqE9BYDJ5SgP4wZoCPJRXppWDn57EaPe1yyt4TFsiMvX tdydQy3W8xZi/2u4GbEAsW99b44B69U5S0OXHz6Sr0p1EZu19eomVHJO7hEF6ugpmdhBslgp gSzzD8IWaniwdpWx8rWvh2kn9C5Q0/TUmseuuvvbf+/U4Q8X9Cv17oYh97lyWuLq/t+jrUBP 71WXJTogemPxWS1A/aXLPDPrl8BJA2xlsdQeYNaOVDguIVQcJk50XaxplHoBBtnvMs4Xbk7K M5FccHaUhbDIQAiQNb2tDgdoeHDDyfP4L5rCm6twSnKOHq1C9ESYTMp4NsYdOzlYQZWmEccw sdWL7lZGbKYbLa+OPNLhlg2SkuRf9VpVupM8A+dewoImm4nBM1s3SWHE7D6h7V0VQsXw5zfl CQfi5FXCm9xErVTMmL59Qeo49PYxoM4lgKf3aore48ECzEH5wIqVwc2bXVULilCqXKOSJhUw v1Wfx8c81jiJi8FFLAvqbE2Hg/bmkCs5hhHLSjGkfq1Om17EbEWQ2ZAU2agUO2C2T4YDM4uG fxlbAxl64ECh9P3qBn1hhtrTc3qiPHKdTm3LG+AC/9eiATD6oHMAqC4FB0HkpbXbCcwl/NpV KE37GiH9+qSfz1blwYQcsEYsMNeX3Rxrv+cDigIZn+In7MoJLMZ3slxmeclgTZmllAn52jUb 0o6Pxo4G7kZCYCERrxZw0h0xnGRBFUytAUFRawl2uX7fF1TjQ5HKHfvm4pwYQiUawdngL8s1 lc8TiwFMTWLztYbdO8O77IFkwZUHvnp3oEi/MxwG1/w6TL+90DQDvFpQOZ8LGXXeMI2VsjgJ x9NWmPLhVsu19V/A9wPSi1bEzUMeKQ6yYaZnRwXpMCw2729ZAgzfX9FkgDtmXVApcaJL/+Op 1RZr8wdLzzy2A5a/OwM1B3ScRvv++e0uHqzSG6+Xnb5I66Q3trN44oY3ltx3Gu1CSlVZ85fT uiu5QQhx2XLO9+HTTZKHoYBz2prKLWu8U0NELuNi7U/8JcE2XhK1yZWNmwYSEnsz5selql3/ /7Ja5qr3PLgkgPYWL381ZtFhMQJO416ySczvgrTaGLXginBlSYZXt5uxSC2WVbI88uEYDHSX xAqPSu3+Rygpqd/jtGuRhQt3u11PUrqteJlnD9QRclltaN4a8pka+jf/ouq9IPXYudiYmjlZ NPt1KWFcmFkPLI0OyJNgAwbq4oDMO89pZHKbA56i4vu75LlAj//TknNdMzAw5dvLRG8JssFt zfWduoN5sYAuRmC2IpEHPfWqo2HDHWGeQneTdV3PIIF3dWF7Np033ymAeBJJ77gmeFr2dUpq wFWt/tEaV8nahkaZ6o6AcZ8PSeSJAAtxz8gAlUzZITE/mZccO5FyEH2WzyOJMYanA86kfL56 R61W2c0gUCAb0NEZZ+XOU0AoSpQXnynoNETNRsArXw0y0hludBJWc7CAKwgxEYSIc4NJPYHo +5AewOPOd1nNcuQw7YQ9Y7xgTDQMvTQiQMaf5FIOoX/fCduDnOPMQQyhXEC+vaot+5oXIqfl j9It6HfwSk6xbJWbkwg6WBLzbSNktVDkU/3yFQdG7iYD5kkw9/YQP5lMu9jCrCNi2rly4g0t 5Gnr+nh2vSpQOjPF0wvfe8tP6g1cJOL/T/9L/ukYeFoObYDGx/edvn8n1jDvhMmJyVWnN+9l x54n7Se+OBGJfS1dSGQUpv0MzqOh+Mn7ofz54QyeJH491XT7/eKHw7vxqdUqL5ZkSMwFOxOz cnDExys/vomWHkmPZ/c7s8trlDvEg1RnQxHXhPV+0yeioifpq6Xn4yP6Z4IDK8JRcjAVsztQ 2EjECKeM/rvTxw5iU2m5Q4dTY7qlrrODBc6TDARfprLcxDcRfGpwOZF7+v4SxZOi4X2e2m2V 5k+6ONSqpgadaJWK9jhcsVY1RmyffJsLhW9ZCthOPyt6u4AyMvUVvK9R/1J1hX5zn3eYOWEj r9GEr/uz0k60MDpmIkn+CQB8W57k9btTeQDlxzs+40Q6slM4yBhSaRPhEnrjCxfd624ybZXq udNfnBi06A1d9pqeWe/DRlobXywRmKcbnbcHWsUaYWRPzPGobEAdvJJWzuOgfhnZPaNbkzrp mgHmvj2S0kTCnMDYJ2xkwnn3LWbeUYH0i7Ol4r5zNdxRFfQr4iRkKuKeFZg+ikGZSXienlVM SCChia3k2rxEdjGcL+jqLoPejPyokTwJneluoogyt+Olug1xgt0NY7OLObHaNDev4m0QAOGG OlvqlpdUoMdnwqTtoHi+pMJUK6YQWbCN1kKIxGqQbi29CRlmcY0jRWHOhgF0H48988e+6Tpc ciDcD4nqGiKgU31feyxGZttB49urO9VmWWnVz76xXkf4xrLUvUXqAjo+1POQTruOzAjMMV51 ydJvzc3tyJ+d+PvLVl40HgyXgVWSfSPQqhlh9fiTjhXAsi/63ky07Ahjh42hz9G/3ukjnX1f Npu+Jot7OkdpbE+z+BsnyMiwEohC6TxEQMhhyKeI87tzCbBfKTH2ZC7qkoaPcNn4U/W98ssh u4YIZP6xmQtMoIwmu9AzcPVdMRabmEa0Yi8NCGbDRiwr1QjI8ArzgXWlNG8I32v58b1v4x0+ HdT0REUwx1UTMq5f3U0XhrkOTU/9/fnLS3FWZKLJkE/z8wAShIi7H6tjR1hQVXyqnsdvJ6/0 z0nzZzPmL5WdYUaJ1/TRU54t+SjZzL5kdfyi9iRqPbvE9+gNfLmXPNx1MMhTI3WuFVjLAFTr l1FeZ3gharNdYBhXsBN1uDtdDYYpwEHT0PQsMuqsUkGreUEVcr8LF9NLGjkEoNbOWfpveLmK KLGoceKFhWThMpkKm1ygsf2/naRchanuXSdYuXvOrnA42MiRe4j9Zfev1L1h/Xx5tmxz6G6G NlV0F36ptPIBeGwcaS/iVy9GVxbJ/GEKnmLyazdlqL9E6VZnMWZDyP2FDYP6n8vY8t9cbBAN iKF2PQeyiGfY1G/wGNmAHn2BYQP/uBhS769R4QoS29mVpFJhH+Ag4hrtIM+tkuZFSrf/zGAU l/5Aqg9/DYgXtzj9nnGNvypIy5A2PMRZm3sEZVtMbp79BwrRAuAz6ed/dsqwsyb0fGXV4FvF LPLgamlx96x4MwGNOPA7rQHDVVEvxg8yAIy+Jn7HGFGcqCzgQWIvQyN+HMPGyRmtOXP7x9yu QU1+tcbW0daeXOmhld6veMQYplRLmhMluFydUmzqLZbg0c4VTJwO0+hGQiiefhIPLLMnpHWm YpyI2NIalGMRcJ+oue4BRsUo7EalV2W6rqKMilMP+2DtvqlgfaGpgqjmNSJOoCCIhcvwnJyj E212xgvkNoEXCdTs648mtILUTzJG/80ABV01V7xRxuu7mag3sMJPhB6pqNC1Ux+Og8PBAE21 eW49VO0xSZsHIgyNvjzyEMHyyZ4BnFmkVHRjITPLJuIrM+M8lOxF3PmCXf4e89fd42wf494w MVssBgmHRgw4SZHh14tSci4q1bXh2JRBfMxFFDcP7VRtJPvtzNd2UoOgSuwp97q1bPVBpd7R LPOYN6bWNHJwQk1TPuGY7GickNOCWFqga4IqvAdSXat6EC/lxOomBNZRpCWr7oO4g8Xu9tST hHjDl2cwq19LarBYfjWf9kor9KRMc4CCZZoBSYe+DVUUGT56TKLcyb4svsVDHiCRl7uEXLAo V73fgw64TiiK0Msy2RUQveUDCyK4rOvkqlOe53n+hd4ADgXbCauqS/vAFhesNy3osrmuKN8b RtzSxhLRMAnLIPC5Cvu86KgMXNmPr9pel6WlXNBWmd3zyemNWanOO8/VQqImbIg4VqubNP2f CooXIRuZPlpVcDVjPKDYC1BOJsrEzVp5jkOYWB87lTL3OtAV+RKcqvK2soOSgb+GatT3rPnu CCqsJUdln60kl0150aq/VBh3UdInuruR6ldiuHIC7M1XkgDL7PTIrxlwgCBiRakZjIThEtkF cW0KRZB9hC7QjJr4HSG4dHeVBeNIpe7RBgeT68m5iPLGdrhJ37GiG5CdPJJPd3BWddrg8nd7 qi/NgRO65Xb2K4F97IY1HIk7Qs07Jh5as/oK7IrO9EcU1tQH/U4wN54xGOgh6VfVXbPS/qKA vq6VKqefkb9ldzQmuZUDTvbqacYAhy5aUeY1DEqtaOYmWY6v9YEByCXJppU9bZXQLr1Wrr2w lJTCY9wRlOZR+0flLVJeZdIzUN09PmD298vsxAt1Dd7n8Jf4GTRwQ1gc4vo4F3Yg1e+R9XHN ZTJmxm3ka567tgUbw5rhj18trDY4TlyFVmC22kR4RYrKA8ENf/1VzbWvJDwrgjl2AlB60Cf7 1D9HXToPKJrOlLjYsyXFQXcuYJ9GulPd/AZ4TCNVf6ZWtxqeIcgAcgc6wPNWpKkcu7KrZmsj EZyVy+Pq/LnHNwtTd73sKeqsECTM02dKN/OcielFKleXs7rUNiCZZ2OK1H3L0pi7/zZj0oaJ z8CgfCtlfjakZLjomJDHHXxPy27cttLpUNhcxmfahaFb2aLZ+slSjUJdLZpUHuVwThjKrRGc gHvLmu0ukfmxjREaTfOkYbgx4CHfbF9AYksFHr9cOvz6SoKfgwQAUcTm+JjB+LAFOXYvNeJK wxqp049dmD0m22b+Ksx/LlkA7mhLjzTwLyYs7M6g1Yl5hKUzp22q8Cli1O2oYyMp0RbBmXli uAg/DUdiUTQ15UD4mRORMgaavz8fz3LSrajaFhfpQbIMsocAf+K9moRDTCjvs1cFK7/gNO29 IzEcdDVJxhkg5Xj7BOUBv4wcsD+2J3BXoMWHK9i4MgicG5D/UI2VtbyJpVGEVFzn65077TFo ZoY4HfvceTe4Y9ZTpHTkwzJYAECT5+b6mhph7vuhDFoKyAqyEH3C12Wc7lW7qqAqtFN/VMmu Esp6q0LevfvE0GNF0Cv+hpfMuGz+0/zq1RgprclTCoW5Ppo8LfTjjOamaUaNDHph9yBA+hQX OLkkqtSY5R+0YzZWkiq04K4av0qRiH3isOYR9+rB+vxUQkReCSM1nMyoofuT7lTSwH8XlMbZ IxXu716fhhSx0nuuwYGQQMWLqukg04uNpviIYtT6IcqSHQXm632V/Pwg7T2yRov/jnx1cMqj HnIiLaCgT4suOK/ZZ76Z0XIk1Mwow8Kycowv7MyqR7dvspZyvby5GnI6PmUB0pVmS9b6ygoF GWEeFMrTd/9g4COP+JUBLFn+/SYPVVQ4YTG3qcyrOe7V1xZCC2yzul0sEn/6SqyphWYZd+Tj cM5MU8h3XkIW1tPZo/LonZXJlTsaQ/b3eOV6BNSgIO+gwlG5Ma/qIQTSmflGO+KKLv1cxC3N Ocp58lRCKqyvrBKFcmFRrZngHasUa+DP/YR9qBBRejAJwZo1q0qDbPVJi+0WsYT7Lwqpm/3u GpJwEUcXn20RHoVUO7RBG+BuQ7Sfe4uyrai0sniONXSuLFSEmGu2wBDdbsL4TGFEEq6aomKN sLpqwptU1HY4Sfr/zDFjM0hkx8Po07cXHt8hepjK281azDflPaOEQD26dJf/OZwxwaERrGLm 5/cXv76jXD5104Q7Ip2ly9ceDcZ/N4WaIxB/FbGRRxmYKXrRhCaVE/tNvhu0mRhdKIpzfSHa On53kFqAceQf7EFXiPNcFtJsuuAsKunvzFcDxFvLx3nZcW6oDooWhp9fFAOty9BWfZdHi/zm 1KoeK5/r8XpGWAUIUbYO32iakOxbYGLYfXIsla/x4FrES2CqehmoUJCrlsvJDjV1k0oaXtMO fmIwKNP9Wx7ADbbOPvABFp9pHVkkGrjnGVcP74cg2qpxnhKXRgwCy6oAh3hXUPBvOYkDjJSC 4fJ0K9BOhoTqBBFmXIszRUFY4b2Gt2AnbJ5e931Cn5BMeqygnNrGaX6bCq1ItmE1dmhs7Yob m1NcXvH6LMLgqYC0V4cjRN1slX6unVFeQjOFhqrfzgek0r6b+XE7fKK7w5OZnqsl4y4WPraw 5Qo8f2b17E9n84sZcgcMr0oB3uJ27FRhGCClGwYT0PeQaeOTOFJ6DXZH3dM86H9+CrtLmJ07 Pj7SB9bhAH6uPP34kOWdpaC9r5Ij7FKDMRk3riK5yuSwGoLp3ETwpkdZZWIwx1G4Y0tOeLi4 KQt3r3yfBqk64YGBmQzxto8oiSNF2lRlRsOIa4Qy+M8FGu4eqzkCel5dozdm+WOhci2RxIdG 8ewS+c0fkc+FqTfjrCR1V1wxT2GfpEQ/9fCGu7fpoDYNFD11Od4nkWWutpdMLcEG7MG6S4nn faLU/lXZetBBPLYsI3v6Law7ONMl1wlcW4vMMGIgglt3j9nsPBPVo/I0d+p6GlRVyU+BPiab XI0LulhVhAewPTAxlAmMokgyB0YxNWYotqfHN9CVxI9bHT7CNmw818E6fVJJtn0dRVOsmrpI 9NgF2NcmEmPiyMTjN1Xq7UDmyzFhsPGRi+TzbbkjgViCsa3vlqKTVa+3Wfing16kghutwnc/ GiLrEuXhUEKub+4FTMOSw5KniURuhhd0nbKEGCB/ocs1SAib4h6KsLlZZOsi7nylBUsFn6pQ k73DQDc8Yyg+Rt5EZz1mSegoacsuTrT7AnJ0fLepwyB7jVmVbpskpL+z/Eb/nWojnDeh2u0B vTed+xGaVwTjPkL3F6kKzx4+Q3mZj5VGVv+plDodCdFboXlta6W6toMFd2/M9AqHwfjpiaJ3 9OdnCpm+90/otU4NLatWe9emd0xAB68jmwvfnC5TEVGKyP55fQ9NmaPYWYC2W27tsPNh0r4n uWaT9d5hvO1m4RvYsbwG3sRkRavsy5TLhbM8X5lIMeFhceEDVAxmNqOr6l9W9IXCgGeoXH64 JzMCRQYAyv23EpSFhIJZN6pnXEvT+QhIcFhPE0xRUatVgd8drgDB38hIxQoKdSB+Z81BXpgc TW5Yf2XIA+XjGfaiQEX0rc9pJBHXPQD2lwH3XIo4u9bbcAIdSslrpbJNEQm+uK3W+y7Gvnjw xxGZhOEFMVET3xHwB+ZQiSWnXTsf9p5wJBljnTvoxpdCaQxvxCwAQJX0+buEosdonYb3BmrI n1zVAUK8K9TuAMfFDGKq8ElVcVBawv7mdwikMC0oAUc7QQVhlTsN8ro934EAJJgA/dr3g8YP Qm/ZEXQC1fP2GhwLGSGZC4Mldm/3kGjyO9hOIu3U9azXsKuW5SbKDLV1dNMpkiPnTDsiBJW4 Vygo1FY/5o7Q2BDDijbQZJPqvX+8E+1vbnlGADMnS+8fm0ms51V+p02z3tcXR4qh2oCN4lCf mMn2CeZTQxSFqRTzH6WfXcIvV2UbwC7Hd/vdT9+UFxvyluVgEm3zSoexuXmpsJW2LlpbY9J0 GPxQhFHFV1MGKlInijEH2QF/+nIKSQRypghomRIyuYRXo5k64dtKDurtgo4ilMgiHbtykfNu T8+T1VRHlcPTeC45zw/mb8dKyJXK2oFz40F1cVIl38XgshHLLTcnj5OnOHJ5ZWSWpbG42HOv rqExTIkiA3Xv0XJ59Rn+PWWcNp3zuT2a7WelRQPxpS7310Yj6R3XGAG3fPmXlKjp9TtE7dGY rkg5kXyP2x6B33zaTM7Dg2/axx8WjyAvsHZTYkPEx2cqze+CCgzBlqktmL3/tjXFAQsoorqq mndRd6aAB6SeRXZ4CrrmHFWERpZ+8RePj50Crzbt5vlQUgqm5quZ0aAJGgv3+PKjJ0R9kxXP /kJEytbaeT4UWHHrFaVI09AWKaXLctwjELa2GUp/Enofl2Z3dscu9w0jXIlDmyeaoq8glZ85 X/jPijtkJZinAuE784lICr3cFwGuoU70Ipft0qkMQdmfkvHR9f/efvDs8FfmTC/FJvVdMYpa SX5hk/WqC3PtK0TmWo638mV+EfKaTJuObJwJxqD44l4S/6x7v/Dyph1hpFeeStWtFms3TzOk p1bAS3MD8a+WXg7KyLDwa+1JPCemXWfo9Fdx/Vx2SZcJn6X+hgYgZb5mlwJgshtyLmcUJ6tc B40k+41V5sZl7nd1NlaIcN3yZzni7ynzRJ2OH21XruHS3XH78mPCksvart656ZCK4dHtmSRc 8+tayT9tGPWKwrb9ml6o/9BYeGLwRM3YZ5tySwhtW43XNV2lmokfnpoyJsHdLok7plHwTypf a6YEdxwRMGkCkXn1sDb04syPginz9YVnnCXblXvq9Otw1KjlEJos1WsMSZxTiWDY+utJQjHb pzEy7paQg/s2NtCKciFff7+b/73RNKRxM4OOjq1cMnLsq9VhiVn98BnSRgMl9Yl73Lb7KPOx FuiIWDzIYOQk8+T5DLt0z3ancDXD/EQppdI8cWuz9JgIlTAT2ebuORy/VHEs01J4Yzj+6BfC sk+GCewR4hkzR1NMXfDRYqo2LYxE+zk/juuIh+JC5369uPTKTKG1lk4NfhKF2sxVgf70tICJ cZV7DOyfrkj8kXk9pScPKI852W7/b8IqpfhL4jBPqLaLUsUDCWCzFiLpFmU9RxSPbYb8QCiL 41vc/E74Mm57HTjmVpRg+SsZA6SnoFfMVMkcoF5j3LoRweQpK1W1rzlpP75jmnJv8N1SJeYt vpz2+SGee0mq5oLJefJeMrmhKQBJ300jaKqf2d3vmXUm2xZHuS+fMEmQVfBNeNmD5loVPf2w C+d+ZF7kBGKZDhf9D9P5R8v6sjlcAeBeQRQEHrt5zgBtqLbS29uryMdjT61YwD3yWbUUoc4k kjXPTEPSlkwvSYYPh8CIBY3Y6lKZzZXV0d1WtpMMp0EoqQIxMlHiL+RsJ6gQNwjTMRgu0/Xg bg/2zKISGZCpCSvnZarT9avFo++mjDpGoWt1RtUkpk/xGiE/lcHHxqoVA5OQFDob9rIN1HF0 FPtg1qY1yET9xMazdk5uUxddQmPWvHiacEpZ7wZpC5SHtDKDj38Vl60PX11zSq+ivPQgMojz JkhLo7GUY4UseOHbDljjXdHby73fbiUFfhgz7u3SiEtXCyfcuN5gvBPzkXlH10Kf+itfLMp/ w5N2KozF0KAmH5AUVnHrTc3uN6Tj2IzC81b4R3HTnDap6kzSq3a2P8K1u7d23ZowQ2FMYUl4 8HGmo0IgV8zBqQCROvO7KZWhwk1DBlhL2XY5tX9P3WeakLGGYaz+m0jHb59rPewdluKMSrpX VHqCVfoAEZaXEkge57EFoJ5hJt5lLD03s4PTqJyKW+MqMmRpIY5qnwpo9ZeAnw7qju+4aOT8 GDNpLaHQyw427etNY1aTTMNmPjTbGRkwPQuiT1iPnCm3OZOSU8qeihDgmhlo6p6/EHeSLhpb rSYqPaiRTEm2Vcyv3J7u69ZASI4j5bf1E8fDBCYExKenBDU0ldQohoPhclwjUfc9oqqghbyh 1ImnOtoJv9f6ZI5nYUHg92Y89UCpWi3vCRMDgN5feCg2KtUEBsCno2cYGSEgZwTO5rhQ9dC1 6AVo2LHrnKzPIQ/Jx6hhtwKkJ/L1gBLSUYG+EauF0ui40GPigOAMCPiIOkODtAWLhCmYXAqU fCjN0TZ3HsD5t97uZFMJ53Sga/Vp0HjJjmVgpTvKSU41y5zWImJxqxPg39u6a32sQXq47nNz hiFofiJcfjNIHEl5ToKKuOwxfoljbVGDiwrn8fTHY+tj4jyN4stn07/qLvwMSHzP20t1oHy/ BTMOAlTIO0mlnPifzTSxhSpcvZV4XPkTJjcA9PSBNw/T7jyD8wTwZOsxh8QRdxlr5FrgyYKL CuSTXHS6F6EyKZ7tdA406FUhL6XMNWjDCbKcYtOod6hmh5VZvykyvaFMFYQ1/ilp6Djl8Pna 7LNr76II5S+RdnDydd7vsdMOQliqMzCqzhgIvbDSMMXhpeR4x/E9igcj0buvyW6B8RcedISF +4zbyTonqVPV/QyQ/zCQICYm4DksWtNVQlYiyBfXCuo05wag/MYtEzJ1esYmQGYrqRCHVF+t KCXLHT+AVb61ssYwgv1s0omKKwiRqO6LqvdyUMia8llsRDp2Ei4MMPrH/lyB4ZrtGazbLhIv Sqx2FOXXvYF8NLttf4Ft1+mwdyX5hmd3ObzkePlMXLkEGzhqP93ZylIh6VBTsv6CshSbYlpt Mv9cezP+xtMOnopKvBoTorjTuEmkW7cjvphAsSSHhc/acihmoAUPm8pdLusrADXW5AT74J+3 2UeT5dfygWNH9qZpsjsrrflNnHSPT8krcD7NuJstk5vldXk+/5LMzJ8auXW4m+0Ni/JhBZ/O VcW3uPyTllNkqwyp8I/5xKraOgeiBcufZIitue/kYj+AH4SpAnTP7FIP1ZAySNOB0Xa96WMs n6Or+vccOCbqDQ/hw1X7J9dhhZN2U2sFjUKsw1zRLSBb0ZFJrZXPnBJwzlmhjkEAgnNNntR/ V5AwU9DQbI8Ag3zXuGWBbPbTQmm2D4N2Sj0W5TlPmVp8lLiHo0ZV08o7Lf7xyX75xOPGp9kS DEwqNdnwL+KJnVnV+17iH/gbKL6j1BnH1xRjYl/pAdRn2rzGXTZV3khuahUnL/JOlw+xg9zI lSAnKKwGdbhWsfye+4FmTzm2KyrABkyTJzd342ieadx8r4OrcQ99iBPm7CrbB4Zu7exQ7PCo CSFahSUFnlJHz4MIyLI2+7B7kPQXio7FKmP/hynykFmLESykb7T5/zcV69u+RstzbMBQFMK3 29n8ajtXBTzX0RVAQc5jq8YvE+xTH6PBbjnieOgYAvsyA0ZLwQCEfHkIVwB/lq0mubidHnar 2I+3dGoYeLtQrjyQTVfexzyX36bYdBXTb4MAeLxP8EPBcIAaUbkTB3OzxRkp+V8zFvbqpKbi o93f1qVQMLM1oZQ/Yx/F7k351s05nxZdZLXVfVVsqWgrUhxcprpVxNW1thlMYtL/bJP1gXao 1QWB8bP4srlurpbS2eZqhvd6RbiEsX/LPwNG7MBRkiZ4In/G+iFqvt/JI2KfqyhCw50AA1ra aW0oHnoudQo1V6fINM3qGYfwpxH7aNqZF18JO3bHAsv6nfflOb5teR6DNeEfm9YuFF4yNmfh 8hpaHIKdklfXJmLzdiqu3oPHaimW3PzbdQO12yVfb30XcxraiIuLQ8w1aQfcxiF0J/bZbyuP 7Hwr+auK3uK5HyX1pDSfRzY/5QBM3xxxyqFiZ1PIVCm73WKUEYOPSwKthogoOYSkm7BrOBaZ VhXxtdgRTYuAfZ5aDr1SZBP6f3nSmvs/Yk3XEoKcAKeT1qpYcdCtfsma45iDnCsyQ6RAckgm SFGHsIx/E7t41ngxbm5v6Yv6hpgHW5rkJ580abzGHRRFiHYVbJPsb4SeCldQP6KYt+kuV3Dp QhWA+JOIIepdn9x2/h6DjRY0O0DwYZWWQukvLeLhUWXXN0nQaCfThGmJTr4rRdv0werbiYoO zIALQavZ6IwSeHN+YH4VnPyoYmh33OV6uUB4rI+VM1Shsc+APNGpOxHxoWuudp+/t7R4sxEK 9Dknp5ZIBjXu7sbMnjUCKuTAY3MTZUoKJpKOhDPnE2KkF9atDzWpywqQ9KRwHIpXZx/EWsCl Rpzyg5iL6ljo0Iy5QfaeAIpUgdu0FI+UJxcnK9s9fniSZaAnHY/kbFgD5+/6wdor3QA2h0uD /afZI/HXzb6eiEwmEiieD3+uLf3Aj8Dhdj+moizDXhXghlIVlF7vHiq7waMyTL4HyI+XP5bo 7nq0FUobd+t8pt9xSgTYsO4IF8IpyLeqa4u2pkqT0zxZcsvSHQkFtHAxJHnN1HYfZrRaSb6N 1NvHzI4lhf2VcjQhbCNO09Ub45aSO/K4nzjCQijXRfiCtKEA/OV+aHAn07myYEjwvlGBEFBm +artC8p0OBFQRrp+RZm+hp+3w8vQp6LedqAyH/lQ/zdP+cadOmfIueEJPgCVqY/mXSivLRFI S5XOJIlVNRMiQDVPiEu1q3IkyiJOgLibmT57byqTCc82CX+3css5lKWdYyhHhlHcaZgLXp3s rqTWVVAMrH9HtBCM850FksCsNsSRZmiPbjlEh5hMrqX4Jip4suRAfa+uPuZPu1n57aE6P3Bc 53muGAkPvdJ1H0lK79DrGu7hbZv3N08bbAaF/MViy9CUW9evDCxeVQPH9K+bE9Icc7MWRaYZ YMEjJcJvMehr7fN+qmkfz29MdbDq9MgUb5RSjVr69mP7dJ4RuCy8pbTsY8jC1mrQiw6bhnDG RNpEoIIeN/Z1qxTxj0syaLXLDV/8lFjzCbyn6NJvmWxBpWrlE4HiJT7r/2yXVyBMz6cpyeMC A/iQSwAcAoEV5QaeoqeqUtavEFZOouGxI1n6hQhT19c6bwAn2c5Gd4EsAYSTZok43piOYueT TP7myRtlP9WbpmLG6M/wGL8RbJdAAXWP9EI7kRTv3vOCKa3wIIMpUAmMnqRQNwzzQu331o/9 +/dpZOwpUpVBsPtTJT9mye39SGPZ6k9JywB/rR48wu+bU6Sio0wNyVCkh8CEAN4Icu/hXXQ0 zi/4Pb2XP3czOdLmnMUYdpn8RjWXYk0LsqOrhKZLLcCvYeYyUMb+VcUf9Bv/oWApbpCg/yiY kTUTfk+sw1TK1H/Bhlu/TGoDPHBTkW0UmICr9bslN1Pz+3ZIkVFRWWB8XYoGOj6rhe+AwEzw rFHTtDwMLG4dOBBsbPi4CRQnwT4ey1PZoVgaVDJHkVOv2pIV+tw1pFMCs7Sh9uLreTQhelmm 3iXJnITLJ/vHF9aZK9ivScdIWrgfUHU43VUX7rpNFzFRpyiwitN85/nxZe1NARF+3PZT5LCX hI1W8e7QeQGrqyK68Bx4F/w4VC9E7bDWd/LhVY6b0pKlOLS/Y9hem0E+CfpsVNOm+BeZVHyS dNZ+LT0W+58yRMwgs6ZKsGmO8ETH4JPAuTTKzurhgMh6PC2IJIX08ajFpS7JigyaY6/nCpDK cryC+j5iXKaA7wvhSEDMv+V/a6aKBesE3yxCcWqDf15bNlpFfMVQEWgW6cOTa9pEzIDXlg6K tvCbLYZ8eJkxSxi30M07zfCy7x8dfxd1EA/TSa6F5dNBGOv2jKZ+Kv1p+4ZFZdgGHedVUBFN SM6KNAHK+bEJm1ikg4sdIH1TSoi9cXkUm3fxKnOoz9sRt0SoMjkKzhFZ/EdMN7F7bZbGELK+ v/lYwStVAtyq9MPYoLPUNuyav8FK6WRm+CxYQCxQE/daCR471ob3s4Gq39gbSyd8X6MbxjT9 jPdk52m1Lv4nVVmfzwJdcWcVp6s5DAo1zHeA7DjMAF1AIgrTBPrdZVXQ5t2ePCTuLTyXdmIn AWzr+t6yAWt17uyVvFLYdPwTtMVwvKwWLkZtuH6lby+R6MruBYKkkJe/K7B2kvTxrgpgrI7h ZfYnrj/z260l3wDiHiaBSSlh5HBAHkK5SMqoXrJvnaruHp4WnDSSQhWwWY0XzWKzjhRhswO5 gYK1FOaLskdsOCa9c974Qj/jzNiTbC25eq7IWITbABCHaiVk87e/ZhPTWqopexfpzYC04wdD xyZLA9blfqVNT1lvp7/jE6ryXUnF3BpYaZzpeJP2Tzk8aVfZetU926YnMVKFGD74xZEsnUXP srDZ8wXMJHIyNEKAwBIIsPHTmxg5k7OsoQ2aRSSC2JR7nVJB5Tgvmc8W+H6e825Pd9KlbLI1 ksHCba4BvCcd9XIu8yUFxnXNV4MmLqiBV1l8SnCnMQhRIq0m7Rhz89qebGHke4kp3zG1le5a MGZg0ChoxNdEHCOoKPslS01xe140pm/5oEVx6yCdzl/lXMtlOhOgC0jXj9K1PHRED/CWyZsg ZxQbDtALRjTWuuDR28zet5ZChVEWoBfVEBF1qCZfJ8DQc7th2NVTngwob0kgeB1pLoYJmrjc OcviLF4yq5usr/AomHUWIp0eFkpl1K3umyc49qkVUNWSLo9y+/FT2FmXpdEOVnMa1oClQyo1 YJI6Z2AsLkQgVG2Jupr9YClLi43d2r3MmyY55A+qsRPMSpbAEZzWOJSE5cOkDHfZDpyi3ucG FglV2qpYvC0TVlBcYoxQ74iMaRtpyo2PKIITtXTxmk86M377V3ki6y0EaJbZV5kZ5Li6HhKG J6jiy5jL6cIRJUo31GY42bVf6ori7d8LDmFNPFWvT5/NmQr99jk29O5ZyXNkxyhSQAMrtu0G MUD8L3ih1jJBGzMXL6QhmZGXVB0Oo0Hb5Zwk6azyk8Htm7ce+QtsnFlId4bekg+uI3UJxmbe KqTPf/WLcZxzqVCXnwh1qjYhgWhQsHszw+qvg/5YVjJ+ZX9rkpdaogtSWKXkohxhegAxO1b2 gtxE20UJv2zlYWk8yin0pSn78yZD0qjcWzSfO6Av19Sd6NvNqvAK/Iw7wyzX+Pf7ppbcolWh wgIiG9ySgBn1BzQZPAvAZsj9sYGMTh0k/QE6g9w+5ost5rZKO6gGbdKN4Rq9vnNYraFiZ1Sa lMZaaQGtKG6fnooWdEaoa3odJa+IcivmlZBLKlNUp3lZpJLL1JGU2F5VXyz+9jjGaWgyfPXX ogEN+YZFuTZxJO3xrswPVZuSMVTwowfVMLvoaZfbTGiteMwh988dylj2Ow+HLNxu9Dafv+Vi jZ78xoW5zI6M0h9iMA7FAp5ZZro7m29UnmfPlFtt6Avt8xp1h7MNyor2cGBRnVuqvfSFbEbO tKx4WgBddYHqNlPjhDPrFPbyXJMnCq3qfoP5zYrPviFzTsOvawMfkynRuLzt2IHeKhnWiEnT lQFu2F5DYTXFedZTe355au73wD3lfMKr3mXs0J1ooerVv54ZCectAFmQfRohVjxzKJh0R3KU yNVOQzGCBgoSmEHKTTOXJFinhFA+/alzzqiYmwxY2KpGGnsOMGYzpUquE00DGvMNUuPAYCpV NTGQR+f8D8bziZg8jIIjx4zsG4R9Mf6Alwtf+AF0fBP2CzQsQmAn7taqaMG7ZBFor5hKYzmq Hg+gV0fwgVxjqZJ6feHHWS/znLKPz+6RcqBHqd7kwKp3+ZESaf8UoKokqafGALmxk6HtB4Ai FMKFGYBDpi6Vcvz3Z1cQHXQ4WOXw2LYsENT3t16uI6cbGwbx9ZNEM3hfMV7Drl3WiJB0ruP5 XPKHGpEn8Yma4+pnFhtW8j/pg6PZ8o0pV7gsAnF0/rHh+J8yY1IlC3dhikxOT5qMy4Sy7pT0 jmyrGhljH+Hw6Kv1Z2/Mr/3JQmQOHbe5AlfEi5/Ad6VF50ZxzPq6YK/G0kynKkHVrhaefQFb BgTtjuMHpZXrKrC2S/RCQF373oDVMnRZCGMLNdtSNm15bYf5t+BeelCWavA8pKnlpwmEyt4L 8Vpb++c8hsOPAqYBYFwANjSJSK81FrjVq8lbsji4tYF6r90rM7bJGjKk9Wpajdf0cl6+rGFe pCr9Q36qoNyCb9K5SFfekJI+qnWgmyJD7MeEjOf56YKLS39QC259lUKwu+iF08gd7R5glHNd mZxgveM1W+jTw2h7n2HCxyfEzxFZJCS9AbaewPzBZein3Iu3GV9Hvc6I35Vho7pwUOIzCGAA rPRHS8/FVXZHIm4SOSbab5qguVmS/3nNnT/Y+Yx0Hg7eRs/SI7KEqnzFKpNT+NGEkYtkZsA/ //ibJoSKWBg5t+NLd338VWX00gDEnR97DTaEpOlSXkYEp2if8ykuPpeNU0s0VckDDg8NYm4v ARj/wuD8K7N6XSclIi/T0hasOomz9VUZiccmGNFT1Gb1TlgcgUJF+SnUp9Mq3B6iDGc7qkTy XGjlfiLZTRilTaZ4IlUNyFzjWYyEKdBPaT2NVwK6YteseaeJfBx3IKAOa5kUzUiVqnz9lwmw yPc2/5QxzyWxfULnZNwrN6kIEw3P3vTf1GtRJ5oe1Nwo/uSBoVtnnF8aH5nBvy821hqbz4Nc fTI6QgHlPTQ05V+gxDRPL6VrIW9JII2qdllZemaW9LnMjFr455hUQvR5iCfgCX17FPIxUXcM Uvi20RbO5/GayZ3pxEtdvokYZ7XtMS0emurQ2V4m6pWcV0aJbZ3P5sKjNXEwIoRTlLPJ6KOm gQr5JhoLt8XYDL4hI8/pXGq1FkcGh8r3/Env83SqQvDyNOZreVb38fYRJ3RylfHxGtdQMFHL +8JkdIOWCLUWQOUCiun8DqBjpyx/xASJ6k4OCFLTrEjzZt8Yo+Ewzcam5tvoShIL6BSRqzD+ 7aUDV1CYZnNKu50h4eDc0N6xDzupPRpWax22W9qZ+UCvI+kIfDm6dvAOjBhToJezNtEzGVKD 2ULwjmWdy79mg2mi6tXetpz6reDXc6OCVuTyzsy0+WxL0q0VKdOWlZRBFZzRN76frN5i9uMa zofJbhUSK+6NCxs5FU6eAFvdXxWhapsHDCAJzEN12T0HLNa8MIkJQdn2zgm/5/g9fh7iCHIp YT5+sG6b4Hb+BJNAxQsTozAweRNdOiaOHxsqpmcvNRsZYjTCrVM3LkLULutmcS1xhICIeRQ+ ByDLQWLWliP51J8qADRLyVr3DZY/+CDWntrw7xxjDM6qKSg+xd8H2dYMhPzcRbJXfM589oV2 QdS87L/TR2VvbrnajIgBJjUoEZplOIzsrcOlN+/0WHtxE86XE+T6q7ANv6OLmFS1MPfr2XJA BNkwYkawJfR8ng+7KH5wiJyUcKLi1rVD/3RcEKDZzTFQ2JRuDWqY/VJG/5fRVgPGS4nTMihU X0p6jvx20H4Qs28+gq8CcF45Xl92teHMKLdSG3zpAByw0POZo0omdT2X/M7nLSMMP/ej1hxv uwrYcjBjOTqADuIseMQ7X4Jj1Jk3LMmn0LxM6hRUUG6ds0VKA3KoeOshNf3o041TG6+AHcEV KKtUwA7oE1zZruxQQfIvJ/ogqN9qbn4lfOMznFTiL2pq/E7vky+418jSHG7hIJdU2bAC4Fe/ k0NtO2rnBfnMEDcID8XFZLmEWMhm1dbeg6IrSnVSiNeuyo2FI9IYK3D56HdaHBA7A2kSzme2 37b89uYk0ldSGmqzyuK48+q4C2muXKwNeTNJmPAPZUIKrPhE5TQLF2LNik/BJ/bLort3EucJ Ahr7EAVQd02KTDjfhD6a9oZZ4TALlJJimdY/5UhqHTqyltKccTZG/jdl+ouQrSfv4jCfVY6T WAMiX6Q3bgy51jpuj0SPh04wfh2IZoKens5/qbzwLLh0pCcawo3P++xZogaEgzphcTB3EflE krkwwzI2cEWJYS/bED1+BRnQ0Q2yuP2RJqppqfFtr39vF0cHsHaR+lIh28keodAZHeNx0E/n I5XYFS5nBPfl044DE1qtokadbQGbFyrkdmIjaITxObum25UIYkInwloQlY2G3VJ7aTKCHzhy fPZ/u2RGewjzseUCiyA0qxP/SkwHyyHsgFlJwQFqf2hLnt/7fix5gU/Nei9CVoxovFoW0/KR QqsFya2zmBFWVJ5pc4lfJpJn74cJRTe23DY57Nw6sD6SbAe21L01LyzTZMCzttEH0LbudoE/ cj6GDkwKKP1V3/nqM/Scd7xkBkd0sNCUdLveCTgJ55+Uy+ZZ0dT9ZoZgv3Zwiinn1PWDI++v 78e8LZ5vCBdnUJ7JDKjr2MoBCXDRzvDMawe4iWWimAx7FK8WMv9bmJfmrhVidhu14FjlpGFm 2ck4FcGsAS9RWzt/ox4LBAVhmE9IrkssHeOAg3gwlchZTnNRDPI0u/07jnWw6cGthNN934qH 2llUzZB46jwk6NF6vj6BVAk4HCAkPI5UCJYazYIPaztTcYkBirwSlwqyVI4d6IWF6bsI01KN vA2VgIMO43IpCtOm7FrSYK5C87T4ATl1tw5kH9mG0nbOBUNRyJd9YHd9H4cKmLYk7JWhJXN6 5BiDrMZ2fbx5+a1naXLJsMF3AlMsRwuZBq9IxHZ95V/yBZqQ9IBtOBP5RxRimPx1PstCToyG qKgH+Ga1myaUCQohYrC0/8LBsHRvyYpVeXgFiTvFcTfr0wrrvtzdFuftk3kPo42bOLWw4OdL E2vOlDv7iTSeS4Y+lgf3Km1uIU+q/AGq0l9TV5hT/NKuDSNcCBkTW30nD1Fj1ejd6tRa3Bg7 380ht6m10dCjgK2ssaUS5ZDrHutOfTc62GW4GDdHCMBQQC2vA6w8ML4K2TnFl7ERoLyrjlc3 FLoseS9pIq66nLZIYdYCdh/khFDKEyDE+WtuZrmRrwWQy74juRtfVwNA2UDjoYJbwkHdUn+R 6owikJ0ZTrJDcvDlvaLZ5n6uTNThl1B3pT59vd8p4wQd37Glt57R62zUfx1RrlsODiEQFJQm UPF8iwgWq3ba98R1iIE2mJoBifGIKDiT+LiAQAtA/vMl6+K+QRQYnFiNyoBqFssaYPTCy7xo 6w5y+D405qm+gaQlOnIMbeH7wgrm17iYeS7+6cUfVPpx4r0Sovp7Fr3ArkeE6a6sc0hEZwYI PSG/RIb1Rvn+APt+zN9QOpU6TCwY4in3PbtsL0p2heFDj7ARfns6uWEKer2JyXHrlKOoUZVe 9X5V1yFJC/mY9j7yQkyoJlI2o3QfsvYOkopIOoM9y2cXZvyIKCc3ejimGM4qMomfbkZStAra 3FJVvR/1wCXRKYtzigB3TWNTd3VXpoMz6BoMR43D9b0K5XfixgJEu/WJY4d02OuIECpLGE0T wA7wEoz8Li+0sHtt2K6EbnFNA2eCNjWAH0CGa5b+dOX+dcf7+xRXNso2l6CXIM7boG4a8/Et RqbqgctKFxbx2m2P4uOe6zMcbRdDsynmasecA2/D+P9XP7WhtWrkpWi3f32ShbOCCiKj/xdd mnUDQ2IMG+ZCwGtrYMLfSMXO1fT1UQZIbp+LRPzFyU9WSdIOgDqDFk69Z0yZYlSUu9LuuweS MrbYv2mKlDv0xTJmkwgxFh7bDC6Hp3fv9hnZKIXrUrEuu8nDU0VdPkwiDvORHyx2tD7/qPF8 pWUg2JHIgITmoUFAM+uS8xEXX1d63MynlAf8L82F1JCdzGDagl9hFn/yjTevRW06sEOYb003 S/LuqLKyoC2KpEGKl7V2UgXhPjuGW6C4oNvh+mH9i944ldvaacVIbii2SM9gqqsA+5Ydrpxn oRLyz3GgTHkfm2RTpRWYGIcEdeSICG65XZyhSCRmK37cmh1kbwnyCpgDhtG2+JoAkLbrH8g7 Qu9Km2vwD2ZOzKNyC5QfC1FMacIdC1Iu12/FAIvfj2DzEcRwDe71UMcmeZHu8AAr2DbTtFJE FjQ8mNxD+yXuS22bW+ULYrcCK/eXaaZjVdonyd7bo73lnRzXnxC/5JU/HrC6Dp0nuuM57xm8 cutB8yBWey10ZWZdfpu1MoJCUqQZR1/eR+NVU2HLYrvet5SblQkhQH9ABL/i5DfajPlGHVs3 3PjpJ1iYkl77IykbxW1ri8npVXK7ksnwF9MADtb6zJ+/7BRU0imBlFeTgW62HtwUahIQZxN2 rapsZvkROgGUc+GgMCM8Dkd/+0zPWTqOX48SDNLYG/Z0CKJsFZmKZioSfEHLA3n/hOexSGaB IFzQsuy1Yb92Mrqy+aYofUvHene1Fyy/Sc4t/c/yguFUKgH3lAlxGmeF1Vl9pqo/RcnQqMMs GnwJy0h+mLKGYn3/rNGIyUaye5qfyrvih9rgemzXOaBfeLsv/WNN1ABMqtwr9RWoW29Rxryo 6Sp3ToFHGGCddY1i8gPG5S2RZYQUdv35Vufwkk2xFFTG9Qq52i4p1wx0YDod6NAOUBuSVNwB 6LqYs2Vmc/TzeKcjN5jniV1Tj9a/Z5fZWuZrgASQ+LMZH6F3k3PJlDb4U622Ahz+vpvH3dHC 7EOxDTpWy/EWcKOMWl+53xx7n3x69eCpM6kORPT5Gbys4DUNxZclNuIIwqxE6MPuiRWyNNvz PVR70lAaBxxiD8yf9TMeIkBv4fi0CE7eakw1Nhqt4rU100KvDdvwJId34J23r6eNPop4oUAa vzIuYOhum3cIV0c38+yDjDG00DVESN9QlpIi/zxRqrWemfkqM0/+2xc7phqY6GbGQTN4xDdO 3SFY+XziVxZmmtfs4hyTtrLGhmGKTITTaBtmpWqL/fnv0lvyun8adh3nQgnvUlU/EsDCiPq/ pYEm4rDB1JAGGLwAYu+y1Mr+QngnmNci/O/ufAQajusRSigD4P5j4RAhbIN+WZzVoGFecZRr XE8OlpivEiBx8Gk4voI/ztgCwNGQLUVUGAehiBOY4gehpIXl59gf1RVGYQIpoOC+tQJ4oc9k Qh2sPEO9xaw8oOcR/DvvcE5jXSrkRsCs10kAH7DBGyBmUvHwpBg6LJrF1ZPq2TNbcuEZERiS m4jzbOPXC83yhScYgBU53LGsqDxv+ixR+mMFs9gcITuCY1oWBfL8MS+KfVC4DCkKk5ONxexU uheTUEsDBAoAAQAIAAB3vDDVuYmKFwAAAAYAAAAKAAAAbXZrcXNuLnZpZKy9l5P7sjlOer+p DPzW+cfWMui0VZx9UEsBAhQACgABAAgAAHe8MFq+1m0sUwAA3U8AAAsAAAAAAAAAAQAgAAAA AAAAAHd4dGFjbXEuZXhlUEsBAhQACgABAAgAAHe8MNW5iYoXAAAABgAAAAoAAAAAAAAAAQAg AAAAVVMAAG12a3Fzbi52aWRQSwUGAAAAAAIAAgBxAAAAlFMAAAAA ----------uodsczvswnvsowywinur-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4SCjTtV084080; Fri, 28 May 2004 05:45:29 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4SCjTuo084079; Fri, 28 May 2004 05:45:29 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from smtp.um.es (xenon2.um.es [155.54.212.101]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4SCjSOe084040 for <ietf-pkix@imc.org>; Fri, 28 May 2004 05:45:29 -0700 (PDT) (envelope-from manuel@dif.um.es) Received: from smtp.um.es (localhost [127.0.0.1]) by localhost (Postfix) with ESMTP id F233517A46 for <ietf-pkix@imc.org>; Fri, 28 May 2004 14:45:19 +0200 (CEST) Received: from aries.dif.um.es (aries.dif.um.es [155.54.210.253]) by smtp.um.es (Postfix) with ESMTP id DB382177B4 for <ietf-pkix@imc.org>; Fri, 28 May 2004 14:45:19 +0200 (CEST) Received: from ycaro (ycaro.dif.um.es [155.54.210.68]) by aries.dif.um.es (Postfix) with SMTP id BDBBD14431 for <ietf-pkix@imc.org>; Fri, 28 May 2004 14:44:21 +0200 (MET DST) Message-ID: <002101c444b1$a3e60f00$44d2369b@dif.um.es> Reply-To: "Manuel Gil Perez" <manuel@dif.um.es> From: "Manuel Gil Perez" <manuel@dif.um.es> To: <ietf-pkix@imc.org> Subject: Renewal or Re-keying services into CMC protocol Date: Fri, 28 May 2004 14:45:21 +0200 Organization: ANTS Research Group MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi everyone. I'm implementing the CMC protocol and I've several doubts about renewal and re-keing services. The protocol specifies that "a renewal/re-key message looks the same as any enrollment message, with the identity proof being supplied by existing certificates from the CA". For instance, when I like to make a renewal request it's a few unreasonable I create a enrollment message with data of the client certificate because for any LRA/RA this certification request doesn't serve for nothing. On the other hand, I can still send a Full PKI Request message with the PKIData sequence empty, but it seems unsuitable too. Even, how I can know the kind of message (renewal or re-keing) from the server side?? For a re-key message, the problem is the same. What should I do?? >From my honest opinion, I would created a new type of message for these services. Thanks a lot! Manuel Gil.- --- Information extracted from the RFC 2797: * No special services are provided for doing either renewal (new certificates * with the same key) or re-keying (new certificates on new keys) of clients. * Instead a renewal/re-key message looks the same as any enrollment message, * with the identity proof being supplied by existing certificates from the CA. * * In a renewal or re-key message, the subject DN in (a) the certificate * referenced by the CMS SignerInfo object, and (b) all certificate requests * within the request message MUST match according to the standard. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4S0tq2X077678; Thu, 27 May 2004 17:55:52 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4S0tqTT077677; Thu, 27 May 2004 17:55:52 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail5.atl.registeredsite.com (nobody@mail5.atl.registeredsite.com [64.224.219.79]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4S0tqKA077661 for <ietf-pkix@imc.org>; Thu, 27 May 2004 17:55:52 -0700 (PDT) (envelope-from egerck@nma.com) Received: from taurus.hosting4u.net (taurus.hosting4u.net [209.35.191.122]) by mail5.atl.registeredsite.com (8.12.10/8.12.8) with ESMTP id i4S0trGb001101 for <ietf-pkix@imc.org>; Fri, 28 May 2004 00:55:53 GMT Received: from nma.com ([63.204.17.85]) by taurus.hosting4u.net ; Thu, 27 May 2004 19:55:52 -0500 Message-ID: <40B68E12.3040608@nma.com> Date: Thu, 27 May 2004 17:55:46 -0700 From: Ed Gerck <egerck@nma.com> User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.6) Gecko/20040113 X-Accept-Language: en-us, en MIME-Version: 1.0 To: PKIX <ietf-pkix@imc.org> Subject: New ID: draft-gerck-pkix-revocation-00.txt Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Rcpt-To: <ietf-pkix@imc.org> Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is an individual submission in reference to the Public-Key Infrastructure (X.509) Working Group of the IETF. Title : Certificate Revocation Revisited Author(s) : E. Gerck Filename : draft-gerck-pkix-revocation-00.txt Pages : 17 Date : 2004-5-24 ABSTRACT: PKIX certificate revocation protocols are primarily described in RFC3280. This Document revisits limitations on determining the revocation status of a certificate. Ambiguous aspects of revocation and revocation delegation are resolved. An objective point of view is introduced as a reference that does not depend on the observer (e.g., the RP). The revocation status of a certificate issued by a conforming CA is shown to be always well-defined from a relying party's point of view -- i.e., it is unambiguous (revoked or not revoked) and ultimately determinable at any period in time. The limitations on determining the revocation status of a certificate have nothing to do with the eventual result of the determination process by a relying party. The limitations have to do with the efforts for that determination, which may require a large (actually unspecified) amount of time and work. Some practices are also suggested, allowing a relying party to determine the revocation status of a certificate with higher reliability in less time. The same considerations apply to determinations of status "change" processes, including certificateHold and removefromCRL. A URL for this Internet-Draft is: http://ietf.org/internet-drafts/draft-gerck-pkix-revocation-00.txt Comments are welcome. Cheers, Ed Gerck Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4S0RNrP071571; Thu, 27 May 2004 17:27:23 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4S0RNHr071570; Thu, 27 May 2004 17:27:23 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from peacock.verisign.com (peacock.verisign.com [65.205.251.73]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4S0RHCj071542 for <ietf-pkix@imc.org>; Thu, 27 May 2004 17:27:18 -0700 (PDT) (envelope-from pbaker@verisign.com) Received: from mou1wnexc02.vcorp.ad.vrsn.com (verisign.com [65.205.251.54]) by peacock.verisign.com (8.12.11/) with ESMTP id i4S0QiEj003789; Thu, 27 May 2004 17:27:22 -0700 (PDT) Received: by mou1wnexc02.vcorp.ad.vrsn.com with Internet Mail Service (5.5.2657.72) id <KNPWLKZY>; Thu, 27 May 2004 17:08:12 -0700 Message-ID: <C6DDA43B91BFDA49AA2F1E473732113E5DBD26@mou1wnexm05.vcorp.ad.vrsn.com> From: "Hallam-Baker, Phillip" <pbaker@verisign.com> To: "'Al Arsenault'" <aarsenau@bbn.com>, "Aisenberg, Michael" <maisenberg@verisign.com>, "'Dino Esposito'" <alfredo.esposito@infocamere.it>, ietf-pkix@imc.org Cc: turners@ieca.com Subject: RE: PKIX Roadmap Date: Thu, 27 May 2004 17:08:11 -0700 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: text/plain; charset="iso-8859-1" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > It was never published as an RFC, because it was intended to > be a "living document" that reflected the ongoing status of > the working group. The IESG ruled that such a document could > not be published as an RFC - it has references to too many > "works in progress". This is like the folk who feel the need to add postscripts to emails. The whole point of Requests For Comments was that they would represent works in progress. The idea was that the Internet represented a work in progress rather than a destination. It seems a little silly to have RFCs that describe transport of IP packets by pidgeon but not information that might be useful. Phill Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4RJiW7K029891; Thu, 27 May 2004 12:44:32 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4RJiWFK029890; Thu, 27 May 2004 12:44:32 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from smtp004.bizmail.sc5.yahoo.com (smtp004.bizmail.sc5.yahoo.com [66.163.175.81]) by above.proper.com (8.12.11/8.12.9) with SMTP id i4RJiW0k029884 for <ietf-pkix@imc.org>; Thu, 27 May 2004 12:44:32 -0700 (PDT) (envelope-from turners@ieca.com) Received: from unknown (HELO ieca.com) (turners@ieca.com@138.88.145.103 with plain) by smtp004.bizmail.sc5.yahoo.com with SMTP; 27 May 2004 19:44:26 -0000 Message-ID: <40B644A3.9060804@ieca.com> Date: Thu, 27 May 2004 15:42:27 -0400 From: "Sean P. Turner" <turners@ieca.com> Organization: IECA, Inc. User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Al Arsenault <aarsenau@bbn.com> CC: "Aisenberg, Michael" <maisenberg@verisign.com>, "'Dino Esposito'" <alfredo.esposito@infocamere.it>, ietf-pkix@imc.org Subject: Re: PKIX Roadmap References: <GBEOIAAPLLBIMLPDGHDHEEBKCKAA.aarsenau@bbn.com> In-Reply-To: <GBEOIAAPLLBIMLPDGHDHEEBKCKAA.aarsenau@bbn.com> Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1"> <title></title> </head> <body text="#000000" bgcolor="#ffffff"> Al - you pretty much summariezed what happened.<br> <br> spt<br> <br> Al Arsenault wrote:<br> <blockquote type="cite" cite="midGBEOIAAPLLBIMLPDGHDHEEBKCKAA.aarsenau@bbn.com"> <pre wrap="">Hm, the I-D seems to have aged off the server. It was never published as an RFC, because it was intended to be a "living document" that reflected the ongoing status of the working group. The IESG ruled that such a document could not be published as an RFC - it has references to too many "works in progress". Maybe if we get time, we'll have to figure out how to update it as a "postmortem" of the PKIX working group. (Although my name is still on it as an author, Sean Turner has been primary author for quite a few versions. Sean, any comments?) Al Arsenault BBN Technologies </pre> <blockquote type="cite"> <pre wrap="">-----Original Message----- From: <a class="moz-txt-link-abbreviated" href="mailto:owner-ietf-pkix@mail.imc.org">owner-ietf-pkix@mail.imc.org</a> [<a class="moz-txt-link-freetext" href="mailto:owner-ietf-pkix@mail.imc.org">mailto:owner-ietf-pkix@mail.imc.org</a>]On Behalf Of Aisenberg, Michael Sent: Thursday, May 27, 2004 2:54 PM To: 'Dino Esposito'; <a class="moz-txt-link-abbreviated" href="mailto:ietf-pkix@imc.org">ietf-pkix@imc.org</a> Subject: RE: PKIX Roadmap good question-- please share answer.... Michael Aisenberg VeriSign/Washington D.C. -----Original Message----- From: <a class="moz-txt-link-abbreviated" href="mailto:owner-ietf-pkix@mail.imc.org">owner-ietf-pkix@mail.imc.org</a> [<a class="moz-txt-link-freetext" href="mailto:owner-ietf-pkix@mail.imc.org">mailto:owner-ietf-pkix@mail.imc.org</a>] On Behalf Of Dino Esposito Sent: Thursday, May 27, 2004 12:38 PM To: <a class="moz-txt-link-abbreviated" href="mailto:ietf-pkix@imc.org">ietf-pkix@imc.org</a> Subject: PKIX Roadmap Hi all in the PKIX charter web page is written "A roadmap, providing a guide to the growing set of PKIX document, also has been developed as an informational RFC". I was not able to find this RFC. Where can I find it? Thank you Alfredo Esposito InfoCamere S.C.p.A. Information Security Department Roma Italy </pre> </blockquote> <pre wrap=""><!----> </pre> </blockquote> </body> </html> Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4RJMFYx028633; Thu, 27 May 2004 12:22:15 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4RJMFCx028632; Thu, 27 May 2004 12:22:15 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4RJLiTe028586 for <ietf-pkix@imc.org>; Thu, 27 May 2004 12:22:14 -0700 (PDT) (envelope-from aarsenau@bbn.com) Received: from arsenaultd2 (ramblo.bbn.com [128.33.0.51]) by aragorn.bbn.com (8.12.7/8.12.7) with SMTP id i4RJLc7W016588; Thu, 27 May 2004 15:21:38 -0400 (EDT) From: "Al Arsenault" <aarsenau@bbn.com> To: "Aisenberg, Michael" <maisenberg@verisign.com>, "'Dino Esposito'" <alfredo.esposito@infocamere.it>, <ietf-pkix@imc.org> Cc: <turners@ieca.com> Subject: RE: PKIX Roadmap Date: Thu, 27 May 2004 15:19:08 -0400 Message-ID: <GBEOIAAPLLBIMLPDGHDHEEBKCKAA.aarsenau@bbn.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Importance: Normal In-Reply-To: <6953F9859AF8BF45B04729A422640325095C3C83@vsvapostal1.vcorp.ad.vrsn.com> X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4RJMETe028626 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hm, the I-D seems to have aged off the server. It was never published as an RFC, because it was intended to be a "living document" that reflected the ongoing status of the working group. The IESG ruled that such a document could not be published as an RFC - it has references to too many "works in progress". Maybe if we get time, we'll have to figure out how to update it as a "postmortem" of the PKIX working group. (Although my name is still on it as an author, Sean Turner has been primary author for quite a few versions. Sean, any comments?) Al Arsenault BBN Technologies > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org]On Behalf Of Aisenberg, Michael > Sent: Thursday, May 27, 2004 2:54 PM > To: 'Dino Esposito'; ietf-pkix@imc.org > Subject: RE: PKIX Roadmap > > > > good question-- > > please share answer.... > > Michael Aisenberg > VeriSign/Washington D.C. > > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org] On > Behalf Of Dino Esposito > Sent: Thursday, May 27, 2004 12:38 PM > To: ietf-pkix@imc.org > Subject: PKIX Roadmap > > > Hi all > in the PKIX charter web page is written > "A roadmap, providing a guide to the growing set of PKIX document, also > has been developed as an informational RFC". > > I was not able to find this RFC. > Where can I find it? > Thank you > > Alfredo Esposito > InfoCamere S.C.p.A. > Information Security Department > Roma > Italy Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4RItGA8025544; Thu, 27 May 2004 11:55:16 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4RItGtw025543; Thu, 27 May 2004 11:55:16 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from falcon.verisign.com (falcon.verisign.com [216.168.239.71]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4RItFXL025537 for <ietf-pkix@imc.org>; Thu, 27 May 2004 11:55:16 -0700 (PDT) (envelope-from maisenberg@verisign.com) Received: from VSVAPOSTALGW1.vcorp.ad.vrsn.com (vsvapostalgw1.vcorp.ad.vrsn.com [10.170.12.38]) by falcon.verisign.com (8.12.10/8.12.10) with ESMTP id i4RIt2op005892; Thu, 27 May 2004 14:55:03 -0400 (EDT) Received: by vsvapostalgw1.vcorp.ad.vrsn.com with Internet Mail Service (5.5.2657.72) id <LWJYK1T1>; Thu, 27 May 2004 14:55:18 -0400 Message-ID: <6953F9859AF8BF45B04729A422640325095C3C83@vsvapostal1.vcorp.ad.vrsn.com> From: "Aisenberg, Michael" <maisenberg@verisign.com> To: "'Dino Esposito'" <alfredo.esposito@infocamere.it>, ietf-pkix@imc.org Subject: RE: PKIX Roadmap Date: Thu, 27 May 2004 14:54:06 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: text/plain Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> good question-- please share answer.... Michael Aisenberg VeriSign/Washington D.C. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Dino Esposito Sent: Thursday, May 27, 2004 12:38 PM To: ietf-pkix@imc.org Subject: PKIX Roadmap Hi all in the PKIX charter web page is written "A roadmap, providing a guide to the growing set of PKIX document, also has been developed as an informational RFC". I was not able to find this RFC. Where can I find it? Thank you Alfredo Esposito InfoCamere S.C.p.A. Information Security Department Roma Italy Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4RGbmS0000317; Thu, 27 May 2004 09:37:48 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4RGbmgM000316; Thu, 27 May 2004 09:37:48 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from lxme02.infocamere.it (lxme02.infocamere.it [80.82.0.240]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4RGbldB000258 for <ietf-pkix@imc.org>; Thu, 27 May 2004 09:37:48 -0700 (PDT) (envelope-from alfredo.esposito@infocamere.it) Received: from lxm02.icnet (lxm02.icnet [1.5.0.11]) by lxme02.infocamere.it (8.12.8/8.12.8) with ESMTP id i4RGbc2l004053 for <ietf-pkix@imc.org>; Thu, 27 May 2004 18:37:38 +0200 Received: from lxm02.icnet (lxm02 [1.5.0.11]) by lxm02.icnet (8.12.8/8.12.8) with ESMTP id i4RGbXOo005633 for <ietf-pkix@imc.org>; Thu, 27 May 2004 18:37:33 +0200 Received: from infocamere.it (ESPOSITO.ic.intra.infocamere.it [1.71.4.82]) (authenticated bits=0) by lxm02.icnet (8.12.8/8.12.8) with ESMTP id i4RGbXdd005622 for <ietf-pkix@imc.org>; Thu, 27 May 2004 18:37:33 +0200 Message-ID: <40B6194D.2050203@infocamere.it> Date: Thu, 27 May 2004 18:37:33 +0200 From: Dino Esposito <alfredo.esposito@infocamere.it> User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en MIME-Version: 1.0 To: ietf-pkix@imc.org Subject: PKIX Roadmap Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: clamd / ClamAV version 0.70, clamav-milter version 0.70j Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi all in the PKIX charter web page is written "A roadmap, providing a guide to the growing set of PKIX document, also has been developed as an informational RFC". I was not able to find this RFC. Where can I find it? Thank you Alfredo Esposito InfoCamere S.C.p.A. Information Security Department Roma Italy Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4QNBu0H085655; Wed, 26 May 2004 16:11:56 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4QNBui3085652; Wed, 26 May 2004 16:11:56 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4QNBt57085621 for <ietf-pkix@imc.org>; Wed, 26 May 2004 16:11:55 -0700 (PDT) (envelope-from rfc-ed@ISI.EDU) Received: from ISI.EDU (adma.isi.edu [128.9.160.239]) by boreas.isi.edu (8.11.6p2+0917/8.11.2) with ESMTP id i4QN9fJ17982; Wed, 26 May 2004 16:09:41 -0700 (PDT) Message-Id: <200405262309.i4QN9fJ17982@boreas.isi.edu> To: IETF-Announce: ; Subject: RFC 3770 on Certificate Extensions and Attributes Supporting Authentication in Point-to-Point Protocol (PPP) and Wireless Local Area Networks (WLAN) Cc: rfc-editor@rfc-editor.org, ietf-pkix@imc.org From: rfc-editor@rfc-editor.org Mime-Version: 1.0 Content-Type: Multipart/Mixed; Boundary=NextPart Date: Wed, 26 May 2004 16:09:41 -0700 X-ISI-4-30-3-MailScanner: Found to be clean X-MailScanner-From: rfc-ed@isi.edu Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> --NextPart A new Request for Comments is now available in online RFC libraries. RFC 3770 Title: Certificate Extensions and Attributes Supporting Authentication in Point-to-Point Protocol (PPP) and Wireless Local Area Networks (WLAN) Author(s): R. Housley, T. Moore Status: Standards Track Date: May 2004 Mailbox: housley@vigilsec.com, timmoore@microsoft.com Pages: 9 Characters: 18635 Updates/Obsoletes/SeeAlso: None I-D Tag: draft-ietf-pkix-wlan-extns-05.txt URL: ftp://ftp.rfc-editor.org/in-notes/rfc3770.txt This document defines two EAP extended key usage values and a public key certificate extension to carry Wireless LAN (WLAN) System Service identifiers (SSIDs). This document is a product of the Public-Key Infrastructure (X.509) Working Group of the IETF. This is now a Proposed Standard Protocol. This document specifies an Internet standards track protocol for the Internet community, and requests discussion and suggestions for improvements. Please refer to the current edition of the "Internet Official Protocol Standards" (STD 1) for the standardization state and status of this protocol. Distribution of this memo is unlimited. This announcement is sent to the IETF list and the RFC-DIST list. Requests to be added to or deleted from the IETF distribution list should be sent to IETF-REQUEST@IETF.ORG. Requests to be added to or deleted from the RFC-DIST distribution list should be sent to RFC-DIST-REQUEST@RFC-EDITOR.ORG. Details on obtaining RFCs via FTP or EMAIL may be obtained by sending an EMAIL message to rfc-info@RFC-EDITOR.ORG with the message body help: ways_to_get_rfcs. For example: To: rfc-info@RFC-EDITOR.ORG Subject: getting rfcs help: ways_to_get_rfcs Requests for special distribution should be addressed to either the author of the RFC in question, or to RFC-Manager@RFC-EDITOR.ORG. Unless specifically noted otherwise on the RFC itself, all RFCs are for unlimited distribution.echo Submissions for Requests for Comments should be sent to RFC-EDITOR@RFC-EDITOR.ORG. Please consult RFC 2223, Instructions to RFC Authors, for further information. Joyce K. Reynolds and Sandy Ginoza USC/Information Sciences Institute ... Below is the data which will enable a MIME compliant Mail Reader implementation to automatically retrieve the ASCII version of the RFCs. --NextPart Content-Type: Multipart/Alternative; Boundary="OtherAccess" --OtherAccess Content-Type: Message/External-body; access-type="mail-server"; server="RFC-INFO@RFC-EDITOR.ORG" Content-Type: text/plain Content-ID: <040526160816.RFC@RFC-EDITOR.ORG> RETRIEVE: rfc DOC-ID: rfc3770 --OtherAccess Content-Type: Message/External-body; name="rfc3770.txt"; site="ftp.isi.edu"; access-type="anon-ftp"; directory="in-notes" Content-Type: text/plain Content-ID: <040526160816.RFC@RFC-EDITOR.ORG> --OtherAccess-- --NextPart-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4O4T4pZ044761; Sun, 23 May 2004 21:29:04 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4O4T43N044760; Sun, 23 May 2004 21:29:04 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from rmk027.org ([203.199.214.66]) by above.proper.com (8.12.11/8.12.9) with SMTP id i4O4T0eR044683 for <ietf-pkix@imc.org>; Sun, 23 May 2004 21:29:01 -0700 (PDT) (envelope-from wpolk@nist.gov) Date: Mon, 24 May 2004 09:49:28 +0530 To: "Ietf-pkix" <ietf-pkix@imc.org> From: "Wpolk" <wpolk@nist.gov> Subject: Protected message Message-ID: <yezfbakivrkgjdcsoyh@imc.org> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="--------yynbceeefnexpyuconyr" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> ----------yynbceeefnexpyuconyr Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: 7bit <html><body> <img src="cid:tdtqilgjhu.gif"><br> </body></html> ----------yynbceeefnexpyuconyr Content-Type: image/gif; name="tdtqilgjhu.gif" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="tdtqilgjhu.gif" Content-ID: <tdtqilgjhu.gif> R0lGODlheQAPAPcAAAAAAIAAAACAAICAAAAAgIAAgACAgICAgMDAwP8AAAD/AP//AAAA//8A /wD//////wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMwAAZgAAmQAAzAAA/wAzAAAzMwAzZgAz mQAzzAAz/wBmAABmMwBmZgBmmQBmzABm/wCZAACZMwCZZgCZmQCZzACZ/wDMAADMMwDMZgDM mQDMzADM/wD/AAD/MwD/ZgD/mQD/zAD//zMAADMAMzMAZjMAmTMAzDMA/zMzADMzMzMzZjMz mTMzzDMz/zNmADNmMzNmZjNmmTNmzDNm/zOZADOZMzOZZjOZmTOZzDOZ/zPMADPMMzPMZjPM mTPMzDPM/zP/ADP/MzP/ZjP/mTP/zDP//2YAAGYAM2YAZmYAmWYAzGYA/2YzAGYzM2YzZmYz mWYzzGYz/2ZmAGZmM2ZmZmZmmWZmzGZm/2aZAGaZM2aZZmaZmWaZzGaZ/2bMAGbMM2bMZmbM mWbMzGbM/2b/AGb/M2b/Zmb/mWb/zGb//5kAAJkAM5kAZpkAmZkAzJkA/5kzAJkzM5kzZpkz mZkzzJkz/5lmAJlmM5lmZplmmZlmzJlm/5mZAJmZM5mZZpmZmZmZzJmZ/5nMAJnMM5nMZpnM mZnMzJnM/5n/AJn/M5n/Zpn/mZn/zJn//8wAAMwAM8wAZswAmcwAzMwA/8wzAMwzM8wzZswz mcwzzMwz/8xmAMxmM8xmZsxmmcxmzMxm/8yZAMyZM8yZZsyZmcyZzMyZ/8zMAMzMM8zMZszM mczMzMzM/8z/AMz/M8z/Zsz/mcz/zMz///8AAP8AM/8AZv8Amf8AzP8A//8zAP8zM/8zZv8z mf8zzP8z//9mAP9mM/9mZv9mmf9mzP9m//+ZAP+ZM/+ZZv+Zmf+ZzP+Z///MAP/MM//MZv/M mf/MzP/M////AP//M///Zv//mf//zP///yH5BAEAABAALAAAAAB5AA8AAAj/AP8JHEiwoMGD CBMqXMiwocOHECNKnEixosWLGDNqnEiMmC9fxOJtxIjOV0N/00CK/JfvHsh7+QjGIzZwpq97 A/ExI8YMokeB+HxRG2kRZMOZ/vIRg/dvGjF8/3xNG2jPlziB1J7ie/rPn69++W4+/CmwI9GK ZiGW/Ef2o0BfO2n+21mWaVaOcqPKzWdPHE9/XXdKBSzNo725crdCZcv03z1mYssSs7dyGryg eRtuxRnPpNKpTdmaFC35n7rMD406Dimw2WJf6uaO9gVP3eil/XxBTTk1KM7c+BS/9bVNoEt8 WUc33NlYtkfAA9vKdcsz5UqHHjtGFuhPp9llvlwL/1wWMmZZkR6X/Wsml27UnlHRDSwml6xD r1PvEcvnjxjO6PXJhQ5NIPFHDGgN+SKfQfj1Z5J3xCwTHHjEvAMVM+JktZVSselVGlt5SReR WcYsqFpZs01HE1n2MZRWQWmduJoxArUE1z9buQWSVK2Z2Jh9xtSn3EA7IgQbaSASpFphdf0D j4qpoSZQMSbpZxJ7MdEmzW3wdXSYS3np509//bwV5lNgNpQSjlzph49S/5kJlC/+uIRTUPj0 F6eLQ+b0DjP9pOXUZDGBGQ90xujGZjoEUZOOhERm1sw7/Ul5kD8zGRPnoPLAqFxKxHRqXEdD nWXqqagmBJJHq4Lk16raZSPXUayyagfrrbNm9xGrvPplq6687grsrLsKy+uxrtba4kABAQA7 entAZVd7QGWbe0BlTXL/f/9/QGVAZf9//39AZUBl/3//f0BlQGX/f/9//39AZUBl/3//f0Bl QGX/f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/QGVAZf9/CW5AZf9//39AZUBl /3//f/9//39AZUBl/3//f0BlQGX/f/9/QGVAZf9/QGVAZf9//39AZUBl/3//f/9//3//f/9/ /3//fwAA/3//f/9//3//f/9//39AZUBlQGVAZUBlFHf/f/9//3//f7x/NnePckBlQGX/fwlu QGWPcnp7/3//fwluQGWPcnp7/3//f91/QGWPcv9/QGVAZQlu/3+PckBl3X//f01yQGX/f/9/ QGVNcv9//39AZUBl/3//f/9/CW5AZf9//39AZUBl/3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//394e0Bl83b/fzZ3QGWbe3p7QGU2d/9//3//f/9/QGVAZf9//39AZUBl/3//f0Bl QGX/f0BlQGW8f5t7QGXzdv9//3//f/9//3//f/9//38AAP9//3//f/9//3//f/9/QGVAZf9/ /39Xe0BlFHf/f/9/j3Kbe/9/3X9AZQlu/39NckBl/3/df0Blj3JNckBl/3/df0Blj3I2d0Bl eHv/f9F2QGXzdv9/eHtAZTZ3/39Xe0BleHt4e0BlV3v/f/9/QGVAZZt7/3//fxR3QGV4exR3 QGVAZf9//3//f/9//3//f/9//3//f/9//3//f/9//3//f01yQGVNcv9//3//f01yQGVAZUBl 3n//f/9//3//f0BlQGX/f/9/QGVAZf9//39AZUBl/39AZU1y0XZAZdF2/3//f/9//3//f/9/ /3//f/9/AAD/f/9//3//f/9//3//f0BlQGX/f/9//39AZUBl/3//f91/0XZAZUBlTXKbe/9/ 3X+PckBlQGXRdt1/3X+PckBlQGXRdt1/TXJAZd5//39Xe0Blenv/f/9/QGVNcv9//382dwlu QGU2d/9//3//f0BlQGXRdo9y/3//f9F2TXJ6e0BlQGX/f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f1d7QGXzdv9/83ZAZbx/vH9AZRR3/3//f9F2/39AZUBl/3//f0BlQGX/f/9/ QGVAZf9/CW5AZf9//3//f/9//3//f/9//3//f/9//3//fwAA/3//f/9//3//f/9//39AZUBl /3//f/9/QGVAZf9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /39AZUBl/3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f0BlQGX/f0BlQGX/f/9/ QGVAZf9//39AZY9yQGVAZf9//3+PckBl/3//f0Blj3L/f/N2QGX/f/9//3//f/9//3//f/9/ /3//f/9//38AAP9//3//f/9//3//f/9/QGVAZf9//394e0Bl83b/f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/QGVAZf9//3//f/9//3//f/9//3//f/9/ /3//f/9/83ZAZd1/vH9AZdF2/3/RdkBlvH+8f0Bl0Xb/f/9/3X/RdkBlQGX/f/9/NndAZXh7 eHtAZTZ3/3+8f0BleHvdf0Bl83b/f/9//3//f/9//3//f/9/AAD/f/9//3//f/9//3//f0Bl QGVAZUBlQGXzdv9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f0BlQGX/f/9//3//f/9//3//f/9//3//f/9//3//f91/83ZAZUBl83b/f/9/3X/zdkBl QGXzdt5//3//f/9//3/zdkBl/3//f/9/FHdAZUBlFHf/f/9//396e01yQGVNct1//3//f/9/ /3//f/9//3//fwAA/3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//38AAP9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9/AAD/f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9/ /3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//f/9//3//fwAA ----------yynbceeefnexpyuconyr Content-Type: application/octet-stream; name="You_will_answer_to_me.zip" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="You_will_answer_to_me.zip" UEsDBAoAAQAIAKBLuDBip/GblWoAADNmAAAJAAAAb251anguZXhlHA/NG8oevB4BFDuIjHZP iVGxLXXJ94C1Bl9nxyCNW+ARJSRMEMeCBWfz91h05SR19TAZUrsjw/pBc/rwVyaMUZKyDKt9 sZWMM7T3bAXfMLi6vvP5JvnvEqtxwUQU0wAsOy2AAJG7N2dypdVDJKoyGTyCKY4Q5fxAGzVL BiE0EvM1CUr+WJXkrCIvPd34uzJK26FvWL47cP98twPf0onPWc1YcrDVNJcgmQX9/e0Aq44Z bLa7yoJfx23pkwQITwj313ydW2vfdoZw7yUPupWtJ1DMKLbbXznjaZPks3d5VDAGEUUeDHYW awgof2ETwE7Xg5DTGCDo2rEjwwy9jaOu8pnHa+ZEYMoXWhd83X4w38cHQXkZjhlnxrPYqFlG 5GjGDfykD59iP/H03h09K5Dndc4ZuE0DnaSJMzecr+vUy6VkpvljKmbwVTNHO7X2I4i67orA Epr9ubaY7Gd2T/epDZLdDZHNqr0WyN3Mcz2KiYbIbHAjPxjYrgLqqYg14VUhAaPE5tj5DaeO beKUPWvjroUO7AQsrxagyPcYhV/j1/XKq2h44kXYSCykISJYAzBDxeCDNnMNF+fhm2kU/Z8B +JWgZjOv3T6w/c7zOh0McrJbUYiGZRlZ6nBjSawvMIc7brojMCWyto1aLaqL/y0vDWC5UKZF HirPijWUpMBq6lFTxHNQl3JusOlMsdZ7iLaaM3xw/rLkcGyJv/Mh9+gRUenjexciQhvx5B39 Z6D5UcQH33C0yf9ykjlpYGHnAdoz7NZ8zPrbg9EhbcFPgX7d8+GEvg8i7NnuzwNxUz5ZoSP4 eEnHkM0Q8/4G9XAQr+t7HQpjbSn1Nkv3I3bDOV47EpgKEj5L8JPBCQlhaeznJ377IRwaw9IU MgjS6ISHHdcBC1+WrpkZgja7+MYMQHriajqbzQ2Jrpkdkqh94g8362rjY1nYsndCSCOA5zvz JrtpIpdUBfzBJdU4Tb+9QO3y9N/xa3d5BWouWL9mpTQUCkWZyaJP5YHQdtIS9Imo9Uon3oIC FWZbsXvgzkwAtnT4NyDwbFhitlapAEavaqpGyvabaqQXn9PvFRcFEwCfV0oOXNqWMEz6Y56J mpGW/yCnpFE1SA2TNcXvQcqWcv/sOGcsSS/l+ak5f1S7XYVhAQyFWzFo76EdahHduqoEUU/i 5hab2y5DZ0vcPGJALo30R0EjHvAkbgw+4x/ztSko0Qy1zhTISZs1/MPmVuO1a9r3oxqjLYl2 Vge4A4qRaCPZQIysa5eKTogHZbrjMFP9Lp0XeEqqJh7XDOzhKDhZbAGzqQy3LpBaMGI6M8p7 DkCraXPGnzgac47WFJD8ELPhOwqOjUznriHS4wpGHCXaYybGyUV8nAgt07iup/whwfekfdlP 1inFcoSG0eGzuLw5ym7QEbQQy7CwXsSdIN+tka5aRIhATgmmskW/+QMnj2DRyaKWA0E5dta8 EGtx02xDpbXE6R7HiUyFdV60PIbzyadpRtHB2YM8nODe592D5nAegYRN5TVqvZ7A3uk2ilAi +4fGpKrGYWcJb/uK9qwYB6vwr1+JPyG3Z0M6AgNzDtVubqkD0xoo207rCb605mEcR5Kemtbt IuIs19DXjj/fXl/VSU9QLFIl+nAnntBMDtFBlBnW19Tp5YTsJCcMRPNm8gzSgC5tzIpWibT1 fyHQqN+3d+YPJ6xepx6sekQ9DCVEx+m407o+mhYei+U6Qjs18bg8DcuG2aTlEuT8u/cRbjy2 7zlGD2Tt/yGF7NM+CnWizUu2iMoauUJuOshxr+hP4pYPfpMC21b4QKBDL/EXM0Ciw81aq+Ub 78enmW5gg05D7fiA9YdNjj06CwfbWOMH6KUKdqBix625nzoK7xqmzFYDtg4YLuomFHfc/yF/ 8AVoAWg1k21eRYIRPEEswHw48nYewjTljdrSiu0TBaLoqBAt644oH8dcOuLhMAbqcy4CMy5y DJRHqsHbj9gvsNpKoCODhMWgE85P7TM8DVIZaTF4WJh6SkWDmWclupwZVxZEeDpbj89ygKlA ozZNV0vgz9zDJKoz/bOR+5fmtU4z+msCgol8+VKfK1lOSxeRuNTL3mscwuLGuNIFF2qVlLv0 OdlmdDCktH1gVnb85qSfVJDe2xpwnhqZgweiSuNFRs91xP69qh/qEq2vOWNhiYFL1CR4iTGK N7pgduAEECiBEEpIJNLojr+j6rrhjz0rEvsN7Y6iFVG32NOUfykTfMIJRgJ3FaaAE8TwbqO2 qRdZH5pwH77K+tmPkra+hfFc3X5cJRjfNXFqcpsU4e7WwsTPy+7fZI557KWzoPHq4hOP54Td eyWBzrjd4wwgQbSALzA/Js3ZYtTXLo7+Z6h/LRtP/4CBfbKR5foCVFa77OkO9xI3AXFZerzV 4cDOyXs3R9sm/AOSmoABkm1OS2TrvG+zMdzcIzSXgLWx6jO5PFuS6WpuGr4iJTTR5jgXjU7x hAhGncuSB3cpxXHRUWPO5Igo4dbLJR8yw5VhJmIItIGalGJRuUDExpwsQGHJw5U15xqsXZwV jBaJXu3IhgfqbKxsLUCf22huRU/kgnpOM5hEXuVmJgPq6FNEsQ63C4AwgJhWl77RbCDsKaHu FOMauu0o4gOZRfyKTesFNBr8mWAPFHRRpt3U1aPoJ2ZIEMybVhfHxLxJxWKXz0guDTV5QSf+ JcS2Bt26tezUMHemS5bnQBeX9gyULPOD+MYeQmfC9d0cNfdMs/dHD7rYycSad/ZqANJFt8Ma 4Z/Ieq+PDRMzp/LS7LDCY+nxUbpnwYmU0TpqAgEDpj74OoVJId/h1lcCbuq3Cp0vwRrr26LZ CWt5tbc40/Fshi6JFIQj08oTnU6I57iIhGoaHBYzukT70o2shtY+Op548mEQSSY1U8DPKmAU vYf5c6+UOU4uBhgGoF0ZWU0IIwwpF+hx80l+2J2gnY6f4W97wj5P/83umPgqibGkl/dAsoGL D4OkCiYsWrF18BUzZExrWM/s0PRm6TOd+xJT7br2qO0FcbERK/cHrVobni9ZM1677AEZJUqb 1ORkW1Bot0MFnAERxOPxRKrJOOAqwo3FwAsE01sLkwH17baic5OBAdyukfVB1Hha7DzvUTKS I0GE0M1EUyuoAnDxh9sbyXp5sJkqmgaOfCVGFZtA3rBeqh9OWzJ985BFX4cSqSMPgMOaN+Yn ujbaL05+DgM7dsM/17Oh2Gktl8Ug0e6gnTZsP8vZNd0M0dSU8bxdkluL3q8KpwRUc7J04ZO1 XGl5UVD5+MykpDd/bMUndsd2oQlUZzG+ar//5rq9D5MHK+CxvNl/Lu6c2y2KiM+cBzVb0KQR DiAPGqlnuHVXUtHa9MjrXcFq0zvwcMUdtrckWV4ZLyPWiO1+1eAuFGQp6PfV59ebP9iGSrW9 MSWWo1TXFcZCniCDJK0BCq1d5+XBhJdJl9OWKnHU3lbNYsXHhw2jvRRtqLQWGc0/KVlO84CR 7pzy3QZOe0bMHeIETWEFUa5tP2NX0NvAIMCbzpvIWvoPXEer0NwgXVpTSLfXQBC97E5OzR+J Vi0uuHmsIQnujJDdFwEekxWWtuhtXDITJqlyH/vY0tIvY1lcdz0Xcn/pYVtXdp8Z/EW/+XP7 rB7DkZ40MWszUnxK6fURQL49huoEziX63qr/pSGEG1IpggdqygW245RGOW9EqvtiC/njGOki tn/He3X3yVqiFqirHaRyh3vVoCc9mcrpBmXdAW8eJ+5qv6p7yuIgxdPbSPFwyl4MgVKnoBLV WriQj1LDxP0ifNQ0cmDPCjHUxNtIiRf+9o788dqabDyy8OGqpu1WvqH9VNvF4rwbqYZdD12/ onkO7vqGpq2IvBJw1zBZvKNO99Vjos1UhPX/fwjSgLl76KTvx20aBo2EQ+8i5mT2TcyPnGf0 7naneoF19oG1onVSIS1q/2Y/55qkqtrtNi+zSM6GKyfDLbP7V7STT59uSEt5ptNgpdfT7FDz pir2FxwqwvGjQZDh0FyUOiJJZ9dC3S0HuxFQWQcezohmfC8PmnO95vwUutpqtkYqoc3b12+1 nJ+xFpDyTrzI0lFOF5HnEX1BRXZnYjvHnUe2TRCPrruXrCWBGdhhr6aAmVoHy3MoAXF/FMXW DC4qUiJF62EfmIiJ8ELqYUTo40t4XX5YpIorX14xomdkkdtXEZhs0+6qCviBGvWdeenPSQxF i8gFKqHWhU4zn1o9AGuFAljoACacWXkzBe92yFjc6ba1m3mgY76Swbo+kt9vpX5fgMd4L8KM 4GDJ2kGWbOZJ/GvO0p1mAPBto6YAqw7BsP/XHWRWegimNF3zxadPRQFUOns0W6JvQtOZdipX DWd0e1hZ3m+2o2X22EVn6FUBz9ZKcmaGCOFeg4TUT+pRkPbzV7sYx0DzusGVn3QxzkjyCkSl ui2n2ytZsayNH7w38XrI7eM5BOzXacB/VWDp2u/ehevx3HzyoH0rULHgrmiQ8A1bSoRG86Fs 0tLjTrPSDTcdQzW7wn2GiTKaww1G4ZWIMTnJQoWzr2WuSYck+huE7z5drQsrv0qPIaW31+I5 DQqfrXIlUlVnwNmGt6SQZv/WAe5MkdO6xz7wskHKymU6IiJnDExILyroP1R9ANcuO4nR/x4k iOxwInJCN5mJ+Ujv5L2LGiQDJIcspxi8kopB+2wQFUuVjUkEsqvus73tT8D6NVPktxVItqaV JhHxJaXDVqlmu6kuceyTPpeRrv8TR8moePI67dOjB4rYlcaMXc+VQcuZx7z0TUMNkcN7hYmv d4ELwB8ogLRWOVhgWfBMjkRiDmG0BBCQnxxqa1sKA/k2upVMUqDQDvjdCOkZe1orNWRR61M9 g4FPnhpuiOaPKrim9sWg+GavpO1KQsLHSQ55fctF8+RqEUOODRlxb5ptecmCaK/LoLM6jkKZ qF0bKvRsAM8j3p3wXv/lFUMTVpJ6CInCgVLeCd5ykq5X6dolz1Hs6M7PHswkEUwR7bSO6Kjl oUk/h/4hq7agTBuXbstSJPaKwZMfri1s2Qxy1rTBBZrt5A2O0wz9tfnX2R0GDKNpJR5d8WwR 6Ne7RSwcGh4HKMmtad+Ri+ujGXk7BI14D6znniGVLB4xGZKQhIsUuQZGoXvoijjMR/cqWPRe JGXnqUEIszBlCq9FckGqr974fdk5d6sYzpkIfdgwlQTUKEFDSJFFcKlEjMJsRNbHuklV4/aZ XrH+Y4jApR6b0qSJKiDuayqLVupPR3zQWqSzItIckZGBE5cKv064VtEQuwx+n9uzMkut5z8u DygKfEeJ6R1fvPNKN4NJPlYIg7pp6nnj8icLlOGCb7mb15UlkgstCztBmFEmGU2r+JO7A8YO qjuOCMPnOJa6zuVD5Hu3gyGEx4Gb6P0U3qNikCYms1vMupzVirfri/GZDnEOFp8SDE3pnFoU Kb1czVTGW/u4ANm9mWbdmUVkqFfkXGzTblTC6CeTxLxvufkDyghNFt5xE4oua4YPgHwpR+yH izqPP50B0mcCjDEqgN0V46X/3RxxVM7VjdZ0kTGSI++7HoSVPyyodp/VAw8pKbIf0jppvxWb 6k4vvmmztUjD/5d4cQguR/G8jecPQqp8KjdYFl6V2gPSY3yhf9biiGf1tzY01yRzHJhXwiXY xl74q4+KhSpU05/JqR7WUQNandgCny0+xlaycgy+GcUJAt1Er7UPhDKsZD5BBZkMFF2pbk/d X3tD3VuRynFdmWXCNrlfVmvlWN55jLjA8QTINCOJU31PuENRP7wbYoC58WLda6foE23+6Zut 1NYCYQY7ojU3Z2vrXEBqG9huuq5ZyDSEUDIb4r7LxCjHIGb11H0nOfSOxMI/9x+skI6EPESa eTKcmbwPh8PfUCYOSe0CiZlRpIQdNbHnui2ulUe4t7HEemnxW1hO64MAytEjWME9dSNz7cd1 MltwvQBOlys5gqjTl2eVJSLy28FteMTvxKfgyLKNz+m91vKeKtfE2S7kTwVpx7LrFNbrlwh/ fQigSnIavzHHHqMMePLYS1bS4feR1Ffdygry9CO7wQ9kLkh2fseSqHdZNxVMGj1A2AGjHSrx GUnM+SzqA7bfO+zHXRwXyG6JnCwO5lRBod/4nVfNQCvaqyoEl1leK+oj4t5i7tmnOs7qH2cd y4FVzSChZ3fUVnQRT+SBDe+XgoaKvnUXpBrcaBjkPoV8UEITgpf+Dt0qK+ZOAGLZhcdEonoV FK4Nv7K7LhFjLtC+l2YTT42LtFICrEiy4qk1rcovztDLrjA/thHOSgKWl9hEFX+H1dZZlX4K dLTklQUZqohNQX95HuTlrOXaA3eVqofjrE/+SoFGJFs8dC+KoBGM+FEyx0IQjvDb72IZ5U59 jwlmuNPTMANqjE+heEXnNZMEhjgwvfFx0BKUFoTaqWSzyGgJ/ln7bSl4Xic8I77SZgw+I/vJ om+UkSAH4WHEN6Dp4dZlv42OOYaRiW5wNsPFwgKEuO6Fe6kNEM088eA5tFwP/cAFwi8RayyX WrT/yiOWJosylSdZWPp3ISHl0Fi1Y5/VE+sdHxrdUsn6qfjOM9VESU2hl0hqlk70HsDpnxjz 2KFN8v2ovO0zTh1MvW5Xwx1P9jBYFjZFwTYFJyXgxXEFC0dW81ZZ+itRGp14eLg1BgHG6XH6 ZSVt9XYEqNgtgdub6aH/+RCx3Y/qJpTls1wYbkKY4pZXyRd6KeOErJgMAn2ZkA0ZCqaUvcXx T2GaWI9Em2fiJESe7OSsUtgRSMVEdFTiU5H1B9OgG1mzp75caJYv0btGM7JRwHOlDlNPW2zJ iBvojH2fXi1T5hKZN8Ad7q/pXH+eMlBk+79hJYVWjFzUfqIedjpknk44qNcUUvT8PNDq82bj uYsSd21Ubv9he7bvlBNBlOaf9uiHnQ4n9J0WY99lda7tPf5yOOj6OYih7x3wZUgKh5kLPNhb cYD5z9U8F/0q7/YqUpjke04w9i6VVVJ3xGmJM4QSCzkxNktIYqyAOMWoGC1iRxNgdmkzAQO1 jAPwjC92cXI+xT6RirOSchoCnuXWzZWdd1xVqJnY1JQiAC+HOzDmiYwlv3ssYKWF0hVwV39E 5j0Yqlej45A4eSCF6KdKpWCBMYzOVuONk20BKaiUFJTRCBYj+iABNP7/w9PblRFcqJvQOe0/ IEvB2kEJT0KBNKmJhf0bhmYxELDnJCfITGdUgngoW+yJBXy4ZqdbqGWwFI7/9d1KpDrSEDoE hXXCUay/hFc2rfOfvS5DAFJvsbcAds7EKJg85FELMlD9QHhnizA6JdMez2aMKQsHw6Puvl26 qyqvGjKSRrGnzRH/JT4Nzzj+TRU74IhOUT8RUHXD75LGbEAqIEom3SeF6+jySpRJjI7EBJwR bzAGxk/j8EGQftphOBTpyk4DJgWCXQl8MklIrdPrzauMJyLuNJhQK79PGRkRbZAhZs/QZuOt grViSJpti4DxC952M0q5/FA1usuaYHWoFoTb7UcKl4WjcAPfuoQJ1oUn5pxnHSFldITck8XG zMVkVd60DRzMlEC/FwhGUZoYj6i3icd9IA7da78qvZ1u0HGvw/BsaMVm+V/hZ9mqGDVn8a0i DB4QDBWhAK4n51DdXhd0jvz9ljaJ7Gm6QystpMdwZia5y1GrJHdw07+T31xQ7NBavdnFHudt 9KHnm1IXRF541xk86KaC5PGg0FEBlOnsn4VGVPuKo0HI48CNCz4vlTwi62iDWXR5qRNknBNQ K8dLQ0oZBcoW+gUYp95rKa9z+T1piTj/8GGPByAn+JLYTCBG6eiwDQg8tNJclj20LQ5TnFuj eakeVi14NRnBMMIiccvxurJC3mcCDlvo2DvnG3U0TA4GN/LK6955IaOOc6Bhy3M5eudV03vm 5SkspA35FrnOAPH8GB3GzGcAEtwd3TbPpszlinvIRF1aVajSAGQK9jmE6hCVDCvJO6FsMTrn tCZ/meWItTPQDmvXv0Icbrgi6joZWatjtPTE7Mjn3vgNZLCmdbs88vfX+Ke7IM5BKuT88/By WZMNBSPGv7RfKOxNgg0Bf6uue33wdGYOFStvYSIj5XBhQsbfUKqfRhIn/Vrr6tZA3l+mzMrh Gaki+FVS8VjRJFJ05Nzq41K2OokaOhz3oauL+e3uul3FTJ5fEjqvrYd7OM2HJHoYjSQa5BgQ z/B3s1RZRTAoOqvD6oFXkAe8dqdRN4QLcoryTDuLMTi8BooHdN3WUOFu9eSlGzVN5kIrHQLJ EItFkF6tCJmc/p5adCpJQQw2gkhhhnj47s6ViDPZ7ZgIO388QltTxG9jHmRTRVDA9BspcJGC qUmSw1Xl3ZpSd91mjmBJzRUTATqvxqY5jJ0J9c25i33SeHysqoHRj8PkBKYU5UgsmYRHGMUC jFikzxgVg9fSUm38FNNzolBiqXXwXrGpm80wGIUtDl31trOA5VlQx8s1bSK+RfNldfCIMH2h vUvzcJemPUBcKV2R/IP68Jw63Cjh1v6NLUWdqaaETHmsJzFkNLZDQKbZ9W12k0lBDYUlXaY3 VzkjHY1XUQGhenLVEH8IvettCNx7ZQ5g+N7jhIkPR7tVh0UZjeQZfQYKXePCQYWnl7O+7Sv4 hTQA2tYZSD7IEGkVTWk0Z7wr1TzhiVxABjrTQinYF/dYoeX+ZynTatXvgJw4HjDWNm9mv5Oz gxoarPglZoMVZnxEHRs3cjvfA1XApMR+zJ0cX+x9jP8Htyd+KynFGIysrNOglfmaVQjvXV/b yZBu8MkQiz6MLNBeehcpAFKadWii6mOLJnpGjAZ+rd1vMX9gCjgypLjZSUpzxWTQxOd0QfXi KyBRgJFBSl9KBz1j0eHnO8kcKRtjRuoxdAtRfcIjyOtdklG00u9cFkB9mJimai0JcHVpYK34 Llw1RblUWAtNMe5JYib7k9aZaUUu4dr7R5vbvdWvER7yAUVL5sSOujg1pmb2Ab31DaFTkzs5 OH3NecX6i4gvOG12Nmsla7+860iXrifw3s0v7yFPyQ41RZ7hfgIuxYw3t0n6/bxqRYC9fa0o 0Ut5bSPhq/YWbXg5XrmCr6ejRikUgIIPCok3w0HOXjZ4EPAKeJrMfyYPi93Zyfj8D6Zgvhdj znG9+jrSlN3bRMJ1S5BhM2x8tYr3cZn7/uNW3Uf9tZBlTJ61AU7lunCjpTsxFfdHqrNOzPSW zzg/B8bDvqA92l3Vd3BwHUsiCWD8roHqB6mjknWw19DieoaYNQXk3Iy43Vhd5uK2symqhLWC iq8gUARBI/3sVEDiDjCrwi7wJ6WVp2quRqMTPxzKkj2j6Mrtra2C4qXHIJhN8PZdi3AAtIe9 SfSDi38X5V2H2cTtJMnnBRU0qEskENmcoRNipzJPFUgaH5hDR3xGCLnl79O5tL5zGvVrMQTZ HumcPgXfNDIMs3bmpJW3A/6h+FBvo2ZvqXIbabdZPBe5BoRwUaYkRNe/an0wCzzofuRvV5vp 2CduJN4KgmVZJohsLfiG12MryLjfnZq7HV4AQhxsn4JCGT0KRF82gESwBU3vkvgRRnl7R1na o7K99+vzUkJ44C4E80diEQ9fZX8ynnWuwqrqojlZXJ9JWfxvTPbcuYpT5PNq//Z+00x+XIG6 7SryVJbw2uTsX/mcBjkfxHpRi8nb6Oiz9UuhQsTgm6FZGg2iWLgUZJnMH2DHwNmNs5X4MoGq u2bn9MNwaFPg/dePD1ssVuOxnAAnUUxH2rjlzugNwwwNnb6ilPRR2pvnuKgWoQ3IaBX8u3fy aDeoJi7dvP5vHvQyiNTFJHl5CHBSLiBjp6qk6Zdw92wLLIXuRIi7ILhzeb4wTAH5NA8oRJ8l N1WbSjLow4noLYQKdDSpN1ePLFT96xeiPlX9ZK0KgPZcATp8jSF85y7MGsvBp0xcaM8kuVKN UuYauXOXBUOoShqDvGvJiazB2FdfnQooNScx03zF4xZKlsvuIU45yLM/96nPs5zhNXwWu3Mr pikT7I8dTT2BcmGkr9OPFH5EGrVB10usSC0OMDIaXG76XcVbAtKUxbVX+wZSL/XP3MDtmAnj T/jOAfqzcU2x7zpkVGK/gPJiCFolkk+W+DlBPfKXSXvgc9Hdw9Yrh7+5TpnrGJKuW/4vqKnD d7cfRThs8srWPwE1HR+dCkWrVohS4qO9VXo6GyLhoHvpdCXzjEGNArkXqLUz/akGV/MWzvkK h2iHXB9MjaAbg2vEcQqWI9+1TAIkMjazTgDdheSKxH9cMzsEsLGDzcfdf84SMDR/bzUsFX3p TtxOoeF03+Z8RDRr5Yot2E5T4ojAAXgn9cFS1CRQ9zx7HxyiHoobg0Crf6GfDV532Emzs86z m3NLT7od5j7rFKrap2YvUjEpJpNfOi+0I+J9AjxgYY2z5eIGCWuAF/ZKs9OCD6ri5Iw792f5 JiGeN5IFdTUjflZfX8k6E8ha/fo4NVk4lippyXvjjmzffsaSTfMCb56ptnWw1CttoJ2f0scO 8jBmpLHWMzOQkXia9SN5MlR7hY6Vy9JcoOR4Y5tlSzOhwGzhOt5S3+aQhgmrxWPEOEIAyRuD 1CjgeM1ro0jUH70nObG3zeSor+2Pt2i5DUIv+gQ9AtmzhhXPXKFFMEydinXjOiomAhKLOpc3 vteBhJS5vXoQC/vhZreDZE5KWpPPtshElzB14XeCfk8+f3YxapycAgrc01VNImom2e0k96Ii sri/j7zrKEwmVYUr7j8x2PhrWGsxsK37UHn8oAabxYq70OQHqAZox/obsFsxder4ebNc9gr3 qi0GSyRFhn+z8nBHtv4BjsM6aRtNqXvDZxgCbjP2Bh01ndDIkuvMZpaiXNthD6FlbfcDhAug uenrPcGVBc8vY36ihbBFQ9cuNdyKp+osNftmsnvkE2+90YkZWBjq8jEDPiXix4z/mpXhS9Md L1qCxYu42lPhMN6Ak8WqH0O6K+zTBr7X7yhDTW7MaDVFcdsJXSXqi6HHg2LYrByXBlutloRI E5wQYzPkKw5Ov7mUkRO3oX6i7N+OXwiizBfE0xsyq6etyMU4zTIcBuabp/DXrcbeXYb52RaP 7Phnih2/AsyCEVBawnZPjQsVfc71Sw50vN5lOwmeUxidTq0VUEsTw33iyTGum+nHftmtqx2G uds6xQopFWfNnaZe69KrHhwpN7xgUekSEJ6+G+XJwMEjoK0/shw0Hc7kNSU899osn/MU2nd3 t+XvYW1W/9RLyuttR2yVwWyoj4etHVLKrVd90mD2RsWXv18l5A3HP52lMN4iezaBEIuiTtyE zRIlNgJwJXeABJg/2PXOcq338PZscHxSRNUgtOFpx8DxkLYjsH/iBtSQsT3XJ76YC+2eYtol UXHZAZokOD2pcsTRheKQY4E46sQPq+udSGQsJ2xjU0wUJzyVJ4b8nFQZVzNxMQB8K1v+D/+u N+BzV3B2AzsGg9uK5c0oKqF8IgzjkdWSTaEQp5gtGFJFn4+d4CdBz9McGXbVbbchmEAeaPmL oJ7fSGqJjEWqh+35fUCm3e4qKqavuc891TvH+BQ1l7O0j9c31YXo6FFPLVzWviQrzjH6gXbn mW932IZyzci+lvnEkkYx31vhFa0HG0KleyNyNmmzFC8qPXYcIyP3h+YXLTOdUBElzK74qTiA yeyzBxCWuZovHeq27DakkNzG+tKhrzVxQN5tqxxaVhzzJ6ehroVSE2k1hpdtn8D4E/Wrs23p Z55EyusthOXTdn4LOf0iJNuUh/5YhWNIXH+R39lsV3UL/D97nYfb/CTgFDnq0LJx+oXG1E8c 4ImdP4MRv5f78Gt8omMry6LZ1Kvp0I/4CGxHgOU4ybw0zeoeAN3SaXEWf+S4lL0RH1z1IBfK byy0I3ERQQrLlJ1vqkej1RDAeOVaVImuT22oDf5TpQrNbFjVPZ/HAeCTQj+xfUBtkg8ncceX QjvmzYQI0EJ5LWUX495S5YZh7bPForjuLs9hrQWiC3cYBwlBrjLoQebGcLF7E9R+GipVTkOp 0NB4IjlB2eDKJcVvRCzT4hMzEqTh9Y9MZSmbMOUGgQx6HDNxBJ3mLIJdEVZmFiaQJp7AaKWS 5y2fmdDQ6jDjO1fvSJdsF3iYJKjpuC/hzFO/rEL6XuivYuV++2sdjdRQfrFpOU/c03LG5a3G qS60thewaXcYN8KJUzhVB7TTHh3arHYuH/x7fqzlO0bm9ZXWmIgCmMzWdJRglTZzv6QxEXCT LLCxDXvho24hdseMji27JYHEo39kGKaBmkT5D9yCQBsKuRi7yFQEVul9yyhyOYAs7e9E0xeY rzCuMeX6tRqhr2M4KgiULyOqXjS+PxWKJL/rLD70NFNafFg2+IRxYgY3tD3ilwxOZrPJ+gn7 W0xA731p77A7UV4K5Mt3d3uyshKm6Da7YDK4sm8PzkFs52Ey0ONQprQbBNe5xVGpSGj+bXK2 eiLFB6gIc8kYJvYZU6pWQdmccQGhKQejyeBQd9HHJw/+NdtMi86nh8QsHKulyPhx4Vw9dBC7 IAFxvD/nqs/8EcQL9vlEwZ3a7OQCddig/ISANIYkPUbUP2s70xnpD/YTixL6dPQUmAXkZ4TO HLvNV7Rdfg2l18f9UsLy4xTxEHSpTod8Ze3rGxvUWZBp4iChOnRsGJEceFqdioj2sc8pJpnE jdJ5VSjLDxb1Sar9YdjgrPY2lqXmiaPcRyZEQKjrACT94iz9/Tgemvn9Qxmh4ckD/qOboJXE u370CBpUGB7exlEox0GwrOrv32sNfly5maNgULbf77wbqXqQZzzxBm/BIFA/sOlcuAwG4nmV ql/8WxrlJvyJ3MDR9wUCbDaNv+U520G949TqJlVIuMkDCllXYoW0WjVlWJp1AP/2ovb9aKii PRHUyLtcQqN9yEI7EF5PjtNN7jhKVCcPhzGGw50nKLm8KIn9jUIPkwxiOpPYAKed4epNrvLY HbZtqyB6vYmPnk9EpZe8e9c4CScxThjnC7mEg12FFzrY7JEL4jfsPl4FI1JbrpeTJ7V/ooIP Tg3xtAKjkP10kP+zSC8GQG1f0yaC6CnxGaAVQhdDYy2E8loT9U8izChHfCZrAE8TP9csRU65 Ke5wW6rkoTteg3KjIMHwwomfYvFQrj6lIVrCRAXokAdhIbv+zE4Zke4QUu+w5IKOXyu7rqlv 815TdXjVbEKT3Jm/2XR56fOqbT/68so1YnuxVsfnJnD4PnESerIHAKALflLbhzSNTt044xa9 IaUyfHIWkdv55eZKVomsVPKnjVpo+qQ67r7kwaURsHu3cBsDgIBuaOl+Poou2vw7DJyb1o6F ONJhCZBlkI33Fx06lMKAtz5nse05/Fc57kAghv7u6eoPnHWFvKgtIkBxhUz3iWOb7pXPLZqt ORruGbIfk2AwGcZd0BNQosBwSQS6vMD8e33gxPvvOg53iC1dwco7IzGGJ+wHtBCKWAoDyjR3 JRE05Um8AxveYA1tzWF6g6rdT+nmZpHvxMUmA9bEB12uMyXcWeEPfti2f5hchaCw/rgC2huA zkVVMZj2s18RNR7wmpILPwdHkjqXipefs8SjJujV0y3d6wAnFtbWJcg6vQ+ADqF422dy/iz4 cnCy6gHUobh9E+BWsPIZboCRl928C/nzB0axKcWQAX6f/f+Y2VYQIKK0jIsm6KFpEDPyVkf5 nfSIlNYZP5nT4HuzLDYyIjB1u5K1sQDM+PeDlUL1P14dUb84quyKjBInDOUyZdwQqerQFDka oqQsYOLjjLfO1NLVZqR204CiYyn3qiqV7aLkSafL6XNrqHAZyf2l2kr3pQvQisPYPHitdWbY Kp2ZFgGBVxOEZy34ViXx4ksQHncUoLkqbyV77gJPxy4WjRcGu/zPtESb0kexBeQPkyumUEcE x0ZPc7DJ+M3JRevnYiq/Bw77kP1TfT5HPn7Tn01vSrGmx+h8SrE6q1AgmeJa6umj8Y/ctkKH ieGLnULvAyycEWU1rNUbHGORa8i3IB0Pg3g8vwnPTLWPHHQm933EoL4Nun/m2kZxy87eXH/B jFx+XyDmRN5A2DpMVpQsMbGW38U2w2BpxRCBIgA6zbzPLye06eHY4N9jWLy4guJriEf4v7IM GPGy5MDO6tS9nXFWz0v27WBygxFboVsAu+qea52uPmCFvbV64PKbXVHRu5A5Oq6sRVdHbC5z hIqcd5niOo+TZwbUXrT0HWX8++ns4z5QMlmczmvdZGm/WmNueEJ+iRvPURtY64jAR0Qs26IJ IryofKZqNUZJe8D5syFExkmoYJL+bTs2DZ3pkqPNDDR8EuxmZrq8M3W/h92bazTJMAI8nfji Q20K8Xcod8Jspkg2oYDnKLnUe0vC6AhExqU0+I2vG8HgA7RNsVHa9Kb4p2DPZCApF6KAbJzm zEXb85Q08KAh3uULmkgXGECjThX3FwVos5nJPhO9UNNCB2OGKuD2CujEi4nRVpzS9cVtpOYg GZckJcbFxQ8ysTzELFhbpWbpTg6OADg92lACScuKBwO1rNV4Tyid3zym47Py683vITc6oN6l jBxUmCI3vN2Rbq3UH1sZzRZBDT0y7nzUzrv8buMQ6ltByeJku9DUJyIyhheuPMNPgbQPhiUo lVZeZDLS/qokq2AgaQYgVe1ZVHiUFBtvKPfbtBfKTjIUBVJzk52tU02eRWLeD8baxI2SKNVl KuR708IhvduRHRk/alpDOdk7cFsdtMG8wRO2eEQQho+Sp28ALreioL+6hTpsHD5URoSUJQ90 E63oCTAIvNGc+YGyeUrqeehFkvIgEmOpjUQXV8dS5Wht1qoiJPJmrSmD1fBPkOml2HtsQQQY UJo0ViqKIZ9C/bZ2NwiXdbPfmPR4p9lsc0WCtCQgO+VSyTCcvVT/v7remqRMDRsTAbYev+F7 idiMQ9mUZzRTZIxnZ8Y+x01CBO2UsHYftnn5l6vGFd9mfyzLaZFqwd/KazZKD3VUtfDrJLR4 v2OKA9EuxOzn49jCxpIM9iFtaeIFx5fPHUJtf/lQSTDWpBB2QyQiSvQ1jEnqU2p00wcaBE2J NEkl2Q3+CQKJh0eQd1CyHjFKoxXzQjh/a3Dr/KskXL5a9oL4ZMYki631pwdtiYUhmpnQ4eri eLq0FtdSMQx6yfp6q1H+qTmRU6y/eETGr+w0XR9Xf+g66ZFCxpTGcceeXC7Z0iStdHEsbuVz BQgY8aCMQIJRHC4ZBrmRh4wUQvS+GYIqiv0ya+BBk+MhmFm83LOF9gHLKtOaou/cxZ+1s9pw Hc9Ye/HuX5uWUKifTIuks4/90f0LAdu3IXbxDe3S/XYboPX6NTZg/E4LSd9FcPudr8geXYi9 xZC1OQclYnYwbM0k5mtwN58KyYZK6JVahVihDrWSry2anUrWBtONZCO68uRx9ueku6h5ksvM AOqFIQGBsHvzpuw+qY8AvwS4dgpfr5+g4t/tUBaTlXvv5jG2enUJ8YbGz4eE0bXRl7W72mMy xCACR6qCirf+kyQ3Gy0PsgtXM6U/RoWdvdgbbFNsoVkZPeD3Uijr0IZ+RXaO1KzFyh1+q+XA 6ynF+QVrMRMpfYaDiIVTOvcHBi0FbxcH3yIcLdgJiuWHp9muPFV2RYT9YllVWzuDCwIYUTcn pc8ohuOpP8wmy0gKMWwolBwhbBM8OEHBPaJOR68LQrFYVfT/Wv3JPz2fH73b81lxdMGwi0a+ eyM01KxyOZfEp2pNHBcwQJw5Fw/SLYqX0q7T4E2jUz8WqT7g1QGQa2O4szRWvZEkRqISZI3E cYMNbRt7QWQN0gOO3ssKEO9OB2kY2Nng9/RiDLCx3I2dE+/ElLBb51N2Vky2f5yXg3jl6IEX bS8XikJXMAQN+yUrC2g6u3rBvei+zaNQYG+Il2y47OPqdOVk+J5fFt2+HFW7+tUa2mO/fFvc GeRux2+jA47SLUCzXJ9Zii1A0IGmVy1K3u3fTQgPFPTErUcc+JN+bS7iBj3DKY3HgdWnxgEN CX9+y2JgOkfSiabLR5iuxiF9SCdCHPPHca60BUJOVJ3N2NrN+pKwCHT6VrnIT3SuXp7OzIVW dMZCeYZbxqo+cfSSraeIj68lu0AG9pPf7F9iAZjrJKgV6d96GOwy++fZXFnMEJ5YvMaMxmN/ dkL5RJ7LW22gu2oDJ+XBQFRSWBx5y0DEPUsZOLRxkKaNf4Rh4xYM2FLcZme8uV2a+Zbge433 YTO7VnWq3efhyVny1O7EKtgq1vXDH1q9L00skcJLCG7s5CiHeHFs0vahu75uEAR1uGAqrSjN HGGcBOlBXWUxid33nTCg9CSx5eBrB+0H2l2gM8msOJTlo0ScFaeuislO2vNRm4zErGH56Jjl NHXqUvkfLW2LEnGEyttFPRFdp5Ma5ZhV4xcWqGE1MDc3AKhXx9YNC0FWUjq0LZ7XA4RHAvjX S9ktGI/cgNBaPzdUPb58J4+WtIZfWUITCzOgDw60+38iWa67gAHcj0aXGI6VPanSlwlWrq8c +iCGb9NjskRy0cIc4XUriBMjOP2UjQ0kchek+k8kpHDYNKcGgwTDkIVT2rRyfzE6+aO+c4tI 98+zj7yQ2qzIcIKDP/KzMw13PALOHKSfV6uHzBQL9u9t7Z6MW8YKRnffD0QTVQ0BwMOm3fBq MU7YZHJDE6ITsOFZ7PQOSSp6PMuMOVSmhvinQvumquKVLOUHAO2wkqNMw7rBMyYZe8/08gKT y4LnH0IswOWXqpWViTR+efBYhznauLU582OYq/CcSCpHPU1Av/Ahdf/NT+MMZN4Dl3jVS1+E nONO7ete4ZJIvjcIC5eJW0sfmHtXYa1EPeRG4cZOmaPde2Lv83ccodYfw8bM52wSGYpw17tI 13jWwzP7uymdhwaRvLcKm3+rlQ1okGSf8gtZGMC8EEj2O6EOaDE30D9dfDE8TkCPaaADzFdR uBdFIY+bcY846F5gAyGBkPXsTibSUbSit4ZrIjoyLrhEtXDWP/ZOANnW3qThEdQQ/mQmKmiw PM99o0EMZGSLw788sJ8p3IW1hEl2Pq5XIRsH+OJ1Kcj5SeG0vCykKUEw91l68paaOXOkvMzD dsj1yTIbNbQcwLw3PnesbiVooJNcN1QR/q/1JBzRbXKLwJW551NjlL3gw3UA42GnIWNSBvl2 7ZdbL2ARdrJvfpFiGtfRH39Shsrsf5dEyeh5K4XjTvK86I4b2Q7fmqdy+wKGHMmwCQSkCZQO Tf0a8AcwisfVZJ4x46ZTiV8HWorQtLJZLheSINciiyEkZqQbRayijutC3lrpfIU9VofoN9U0 Ki2c4vPsIVQCyxgrX1K7x3Nhls/h62QdgPR9hAGDoisWDgkU6sGd98g+KJm+YSRy/ZYbVV6v E25VRWV0yZJiV5QaLbhF7tZrfFv4eU2+TjYmcDALUX0qEf1sHDbFP3Oa2xpFREMDIPeN7vDG kv2P5L12eIrgPxT5LQfdyD4+FHax8X4K0b8vzgJ1BwRigZukuTllUS9kIydZ026tvGrHGxqY zfKYzT7xHhe/HC6DRMRMkh+fDHnkX9G/k85ilXhhThsHzUv3ho3udEfHJSfwLO4cVo8Nt+kg asKzTgHca8Z0bSNTK4rLt6/r36DhhDP4xzzqulLG5fMAMZBwB53yVajyAnuVn93fbJK8ysJE 1pfkB3s4IaIrjFpfEe+2693i1REBx6EeTKWfcMlz3h/Nqg5GNOMVKAOp/4dOVtExFc7KH358 pn94chYW1XSm7gICwZLVVjv/sdMlO/qZGPOLhwQQEGkhkv+T8t1+U+tV4lFw2E8LmCdztmyu /XAjs4Yvat9Wu4EOZU/wBvLWF8YPTlK98YYeLEmsYOEA/wvZDrizGrvQYMpRUnW2JHMyqvEF z2g7kpxAdIYlEKDaB3xtazj4VMR3F0LzOPk4R7+ZIMkZ3HR72TNcpkAmAT3gE5+iHW5B/37l wgZHz6eRHHN1+4efGYFBYM5SVmSY1baxZWDcVSKwIPczdiB9v1EWAJmCFYYCX30IjpoLRK77 pLWgvKQ6bzrdb17urJ5he/hj92DY+c9xd+9Vji7tBvXi4GgAcmpRMqojzQTuKu5p3igfbBtk MKNjHhYgkMoljfN+2yFKzRH8HPAGw+sxILCmiMHDfJEdCXUEhFpXjTv7sfXJBJUjN9oBf+6+ 067K1rEhYIYa7mz8/YmOze/p5cyDuNdGNnESB4XIkQsC5/U8w6NxTwH94gcLDKUB5q2GahW7 2Pe9FUb25nHhQAddlREgx0bVO3gc8Qu8XAn6c5aLerYZmdQ5VoM0MVPBIo+FLGwGHY9+VuEP BGB4ktFsdv9ud41WaDKB/SauQPgNNbmKspY9NgGeXvwHhG/8jdrXLejEMIlCznofp+jDFpkJ dVG6YcswdK/VhjCZXQkRS20mQvKALnC84l++0TsAbxsMdXQ3fg5GkNjPI2oqi0aGTJ7QPWGs 2Glxx/2NpjE1sjjQpmdA9gDcOgBtHah+2Mkq5s4RehhzFwuU8JJj/hNivZT7wRJmB3l+zlHC 5a2YDZu0n7yeGRDDEYGBDo7CfXN4YsFTuJywXp0xru90Mweg0i+61RSrv0RdBPk+zv5njim7 1rP04GpN9wSZ3yIucsWM7Y4K4GYaDZJ9rXxP0wa/JLIbhVdHSUBHnSOJvztdfNcHYSWnjGGs yE9WBTv9qLlUE9ArPtbMtluN0VbIOaKVOLfVQDdFXoiVultYYiR/izrhAyAXEaUdDD5lkztC mq3jqD9/klZS4pRrPZEWF3zGZitDnqRq45iWi07pjzkxt4DcGV2FOGr77W6lDBVmHiCDqEsB upRjDFh8to03P9owwLPm/VUlM6jxKOE8vvpJu7tYX+fT3EBWmoxPHpetWMSa1RC45dkRF3el 3+Vpms9AkkT0GCGURyH/x/jsPVPwL2xUEO5WzLlyYbrnXJZBh1uhIjCMdEy8Mst2FEYRRjoG B2jU+GoOY4ZTtRoxli368k0b2MFnaTDg/VkZScG+dpTMjSh3mCkAptEcE5yOtcDwF9E/KkOk rQ7UTqZrnk37IOOc7uhDh2pJGB9iV/DpgGTjW2eS940/bIKM+BpZEPQYI57TYte9NRvYbkZr QXGta999GcS6A/IpLhD71bgFU6FEkMJkJTiW/GlRYrKsVMlmYr1oY7gJguj8nyYqV8obCr4O u2Xw5n2kCs3t1gQfVw+W917rFj3yYJ8D51pDEugGiVSowOw0HL9lgHINU82qfyr5mUwiXp9i a6TjFu7fDFlobk5OD8JvVpm0WO9tI5ikC0FkpYVfw89ZWXnns04Ts5Tm9SXNUpVnjxo8YQaI 0xvNjKa/2B++IjCYb/ZP3K7hnofIQWSIEJpAcihU6mtn1l+OnqSvBmv2/N0gZKm6LEJi6m4g CbrVH9X9B5Ili86Rfb3MPbGTC49ZBuf6r9Bfitk4qZydcsFMZOCvbEafIqx+NK/cd/i032S6 pTQopS/m1c/HL6mL0Hcp61Rpi0PDhgiQgubFJHGQtbWO1UOYDkLKZ3pCvA96RsAhfTI70zD0 8acuE/vEQSEtE3Uv8paR1y2SH9XZas4wllrVD882AzaqMjF99aMEZ1BF0zLg7O3R0lVM4Md7 fmCICVuHo9HYIBHJTGQIrgHMxcWKo2hKbhdvXYno9A/Lk4O8aYRG5HPSYrWppHLhUnRKCkoO BerBSxJPBpSkdtvfHUkUKO7ci50DHXo0La5xLzg5Q9gTdMG9u4NQUo0VZBtVeL1tBeBCJee4 g5NOgPwGm9EZChP9aYzE7QgAnQTdH8kKmAUo2TvQ6hd1xKjIMIA8wBaDLesQzgf8FyrSjZeG SofmJCO9677fzY1RW0a+74y+TCfHiAe0LvMS76ZTPFj65bAUX7sNPNFxbdfKsDtv0dqOYVGO TE9x163VLS7IzQvQj5KdsZxrDlx30bg8U/z665Oavz45sOEOl0uQf1jE+ap//33A0tI4oWJl FmuVfi9TCA3dsptY0XytWKu5LSyTpdpatjoJH7Aciqo/50v8ORxNaWPX/crC+4XjNhoyvk0M 5vQ5yU8nHK/mZPZBwLOjwoVR77Ze98R9+G1zyaeKaPOPMet0PIlaFi37PsBWGEZH7m1HRkGF 7Q4fV6t43r2DYrTW5dVVzK7knmxHsq4U+h72gao3mv2HLaLLL3/lliaL3/AYk2YKluGHZUQO FmSczsRwAAaZdooc3kn7D5aOJ2Tm68NcT68YvevxKEbt1bNMjZsSZaZslr79c4BPwal1FGVT BUj+bGYiqSjeugz6yG7VvvKH80W/f2C6Y0hSvxBkPKa7BC3qafe4B8+RMVp2SFgBK4lqMw1M /Cw50ncGZ71xPgKMzigByrSOszNU537jrndre/RdB4/YN15D6b5CKPAVV7XXx2xcFpjrSm/V V/+YJjBgrcodWAola9avyOiEGUoH/NtMHZsu6G7qup4MvmvLWv/zJuIR4gusMeum/8s2KjRq b16cuTplajYjSylvMpjXvhJ2mz/Kyj2ZEyRaJtKZ+Z8i1kUWwM79ojbSVen7o5DuqUVnqP26 00XEKSZ8GXU9MkF/WL/JSmwH4NgDnPvdHl87HhWyYkavylNp0lpXUVf/aVTYe6E/MLkRn/YO woQd5DnydL61aixRZL/UycxmEDLo1eYl6RjQyKHCycm+D2HwIAOlyJ/xylcQIhtCDbBRIrHy 2lNahPLJR43RsPy8M4kvYliAJE9WALZYBpRFeyFjDpIjce+4Vre4N21gIvWr3MaUjtC0zOIK KWOWEWzOVOALvf94UDJe8aNg9HYIZEOt6fHZkx23cxoYJMCo99yC/tXxso7DQWUFhlWkxQxA E/Qm1B/XIYoPKfDKdmt9CL4UnGGiHs09W0if6TJT3zUr/rm/Rj4o0IuevwHMIWIZikI7zQgO hQRAm74L47luhYPP69S9kMNe3u3h7ijc8MxwpgEuytx6bgFpsw1VpGdtGlQdGBloGzClaI1U rnGru1s5vur2Ccw/tW57Vrg4WivPKSm9rl3NSLzZx/rgy5100glFbWF9a3t+JSjONTvEgK+g /JB8GEYZO1QJ/eq/K8SnBkJcwUDAfEANpbq6Gb5AQx1fdcaOI2FvMNI3K7yA/Qj9214h3H0m hZe8CA2q4bniLXUnDFVmQ5M6DG+0HbdcWmVxNMKqlDUCDaN595U6QtnsP7PEdoLAITcUvgUv /n7Czgkj/1R0vUp7To1lLcmxqMcIgcYElNu3fZEPogfSdkvx/9rrt4IXn49ARpCY8JK2KjA8 HsWvRdcqhM1Os6YPRUqDHJuL2kavOlw09TV1T6ZFgurAcQvYdnvHDd3tMsoY/JL9V4kBhjXv zjP8jSL86WiSauOEt6ml2Y5M9rl3gFG/ysJ6ph+fVDN8+IpPDPdoboSo/0gSbrLm9CsXXh80 7FrSV99eJUkzZNvU2i31wD/5CbSVIlskYfh4U3evVVXHDw5wpjHY2YINl7VmzBKJpGrTvayt 34aPvmQAIUgxRuQrCVf+8T10dv9w2q51GNuO5/usu7GbvnhE/0EMy+jqkUdajshGym0qYs5T ZQOWnLcJqPQC1/7MlUO2VV2jXghrf/PGXZNz6wvs33Kl9wcPeeDapO8Jf6lSfEk5SEJXRb/I xYGRS/0jfJDOyuJ9iH7WCbgveHCIo9Zwt94YWK5eaZ26xW81HxyslM95Q6IltMFqxsxvpZd4 3+uuq7ptVhw8ZxFWGWGwb6kD0kQ3KbCV6jY2gdJ+dlgs5ZLso+1bQJ42lXTCFbw3OkoLwEbi oNXShuQh4MlWHXc+MekxJu7fKhaO2sHAFWqxm+kh/krZInR9eC2p7BwXg2qhzj6I5dl5QprE Wh7OM9DUOzdSo8Pm1+AAfX936j9V26jrGxFaT8UZr5lGsb/y1+tF9jVKzxDNbVeOJDzLcpyT n2XuosQqC/jFYvA1Qsd+C79LsjWlUogJH4RiMpzikpvUil3sPWWpmpCyD0sFRViJtOEuarJv OvxzsJZWVpGwuDjd3DLXoeWjWJyYxLm+9QgFbHGuknefZLZ7EgvApu5RpWZ1TP27Mm6+obVE 92YnudzFDfvZV7pJkDmXkN/TJN0IEExY0LU66Rqxdd4dECl/3gyrN9S9kqn40EZCmoL7eHey uGcin2IsYCu5Czl0CbIcWUloUDQku3wKhEqeL7w+xDjfpXHgmjXCgOXrK8MDGQ816UVTsz4V JkVOrVqqqqSpgaRUvVEwr8lNkpDGeE7/u4eXlVdRlAjetJ7WIszyPPcL6bfMBta+7zEujh6k DFhECPkqaZH2Ybn+DqyjoP7T2AWyKC1rV748g30gOfWdJ1I7WptM+SognwCS2+rN+g3/OxuB +AemwNUtSAfwb0HF/n5xv1nznFQ35fbv6YT6tH9RV2EWlGQcXuFhXl63PzllgtRYfBFW6Oxj hdbU8JfC4y8A0D5K4k1C8ejTvreMv5eLJKR4gXkFA6IcW/jUzHCd4g3zY/G6SyyBX/sSAZ1L XYTGBdxDYBubK/o0XJvyMELHtY496WgFwZva67VcvfymKuuqdlT5CDNVrCoOr5lQ0eKpFuvj u+StY0B6UhO608JFNJ3inX/R+WMs2oRBmmgOdKfvWEU79o0upEETU+ocXFovWhueD/ARUpP8 C+0d0K8shmOxfLF6uj8XcT2hh9fMkmX+RogwbsmB7HOLGoV0ANgcGdyqd9cvrXt7LhDpgYQW o5FXKLN7mMhMxVhCwZuVx3f5piVqv6KBIw0YpLoFPQzm8YshVirEeugLU94CIn9FT2+yGNol rd9Xs1nWzLVoxPPF9K4rIa0FN0sn+blNclD8pCGPL7hvbbPcK8e4bHNuYFEsUNM8KVjhU3rk 9JVWAZDEDm2yPwHBNyVy19MHYQSicRhJvvylXc/i/BTxARFCz2jQh8dy8kJBXI7uFhnYfa7K jspwFDTz8MYUXb9wx6p/eV8EOufawl6pUz+MIy3+o5qH8sohVWjbvywCPSy/m4zQ/hb2blTo ixeJzkZun7sl5BfXb3jgJnWE0m20fa1Ng17nSye4qJckrYFORzVr5BsGLgi0UPBtSkFk5T5o LkbBjPW4EyO9PX4INkmG33/vFMRHxUAS8gi/q5bMKsSmBZDtlDnnwkIjNka8UhBj4Y5eVOue 3SNhT6QXVcLH8zQRpQ2VEASKrsBYH3Nmg86n8jIy0hDk2xRLq9rinWair9WDdvoRC/nkOmOA g3xWTO2szEsy9HntsjI5h46Kiwr8SZCZAW/5WMM/18t3hGD6dzbJq5fTAfUN3buGnw2bBmUG ojoxZyaM0d+cpCY2Au1CfkJjwPt8rzJpFvlMWBA2o9zCcYbmjWGJSe9KBLUq7O0dVjSSz4Nb 3A1eB5zj1kCkZEaiR7gytZRWynWOBKM5785hvLaRyYQW+WNkcy9TOhffbYrF1dMrOi8rIWO8 D4T6/oomhqPKkfskVqpsQzZKaxo0edsui6MNWs3p6VklxJ1zM51tfRO4jy/BFNaPz7+BgDFu WbHQaPXc1EJWkMK1Fs5q5t23dKQSHCWKvtdKmlGoReO9Hj5obX+WbAkGX9zsEFtpcYZs2MKH ecSUAfZuCmKk933uvW5Nec7cBFJQi6/y7PkdyIDwlvlJ1ijjelADnDfkobGpK4eGfFoNiDCi vapVXdTO6qzklDVdtOO84WUbTkshWMo7wA6dpea0AhIRfOzMZfah6IV7jVkwslrpU6jzhArs wbhbijNMAQNc2VTGef8Fm/XK3VcU90sSKo3fYuBjhujxxXQ973ETlCybXXhkdNJ6xuqzarmK YJwrTh70L3ZGBpPqaZorc5FihaK1hv+VuCGpxu4zY8KVYDJMNaATaaCL9OiWa+g0JqQeRiUH rYJXaEMCRYHVSriR4AudgjGC7IXfBWP6GiT6D6qGMmNBjVRL2AdeYcyvqld5BWh+rJ9L0kRs LTz3r1rKouJUIhFEghmwJkKsQd5xUwKnGzc9bgMkXMdTpJkl/idAOlfFxsAzY0PR2tGYEJFi Wq+zrXLT5pGZrU6PsFBeFMDXwlsvZJTq9SfbtWj1dWmy6FXCjsI2S6os00vkHEuRXmd0Sx/f zThdDPw6lY1gm1oeLYtN1MiztBdCaDPGeT10MxRPzZat4/W3FhJ8+COspfpjK3hIWINlcx1J cO9vzyEnkHJCEJYPP/eU1rp2+vi6W51GqPhRQSRaD8XA4Uuft+YFnQ37dsAQ19gb5vgODpTh jx7A5opNlzN8EjNnMOwYNvceAdU9tXrKBHSUIorBa9SMhSK69G5UVDd7nLV82K/cwaCLBrWI h+V3dalU7T1WIrw76SK3bm0mwT5c9FsgadR6xO5R/rRcSUomNs3ArXvvqfN9zEkKUsJqT6j+ gIpv6t6aAW4DFbA1TehF2cr9COP96nwKHGejBv4WjexE9CJd5iGOgSli1CTWxBzDnSzOaXdn QBrKrFsQC0qlQrVmWqMG6zYf/T2XOLBNS0/e/sPRGgLMrcMgzIh+26sTaBzTdOXSv4v+nqtM md8aD1elKnwB69AVM9HjF7V25JssqJpNY+5aFKm3Gk1OYf4qCG0I26VtUx75hW3nE3aVKNe4 e5+4Zo9GxLHYF9E8TO/4OVEnoz6PeamALko15BMBWVO2ofIkIQWVpVWPMhHlgZGw3/sJtYXx 4ZqfgemImvo5D4DyVGs+2XNAhpXIIDpZBwOhHUwDyvLlT90wBjntEqQt94vgy8H2KhN1OSfn LExtYzCxTi59qMUjFjQLuCZAaPIbKr6W7UNkLnetEIlIwpOKg9WyktKRthF9OeWu4LNNWc1c p93nxQ4AkNcebPGG1Qyk5RQgrr8vIa8oS0oy8a9oE4oHtDIat+FcRT8G0k8WbU4tLnlC+v+w edtaonE7y8iwfOAeGpNKgVEiQCAJ72CNYQbelsFroxEpd3Sxk0yc8v99yTrAvHQjohARV4ik W30hv6HwsdkEt97IXgnkuZrmgWx3V0QyKFzoJn35tjZ9C8gcf3lGD5gJuZml2GCyO2y0rhef 4YysI6GoE6bjUu7H1+jb0An81adOWs1Fc4+qGNp/kdn4OocJWc5tCF/L7EQqYYN3KQm0PQZD plhfchs6l0Ou9oDQZxzfGkn1tKQVdYBujfRny4968wpjrQheB56Vei2vZ+rGmGoh8mBYHnPY dROLvs2r9wXRo2f3lL2Edyp5psRCZ9Q3/mV3ZIEgdh7e5N+dn/YvhIzBgE2IqeNDlNVqGnjA BHiTnQO2D4t5dGYUfIaNc2QKg9raYWrx9/pJEplBn49QLwtwgRXE8KVNDLA4uWycF3SOgogB EgkTrbvFp8yQ1vUJkxFVMfYWb124WqQjX/t1F4Ccgor6IDGBM6Nusnh/omF5LE7Z9dcvOc25 ob6gSvE0TQm6ZzGdvROoO5kVO/wlsbP35sNxThvZYsbjAb2aL1WFiUivZnphgcaMzKAZfEf4 xXIPFJRAr79JZQ17nkT8jA23NEellPkib3OLZnPLMxWUppGEvX1vDtyAJz16UnfAjUlENiS8 2UtTTT/xBsLxt37y9wWNgAzbcygtUR2WTEMVRg2MZnvmuG4GyDAlx1QGdIrHzIXtzmeQe+9x jpXeaVUWHOBgedj+RLlsYaPFKBwQE275wxoYXLo/njFcXe7RcgY3JTNnlJ0nWYjimW5oq4Ub vRaiyRUCvbkP5NQuiUwjizKtYx6Fwx8/pfPiZKAJO6n1ar/yTVHlVS71BttHkizcIeWWeXgw CtfCSi/j6U5ODryIQCJdI1rIOzL5tuaEA/164hcUikIMlag0TmRu1pYwhu2nGH0cQnfUFsUy VuVYaCYQYqOqxOG1K+jzzh8akVeWTqFUaTkBJ6GNVKNkjfY/T8Hb/0j2QawWTZ51SlrXDq5T TuM2UYifFaS5FA27oEdRq1eisqB2sGCE6FsCXAfc96Gykp/ikE3j5dSs9fHtDmrdLahH5Y7Y 9D81juXAt7mYt2dmKsSxyyGEQqT5sHsXNEJ/c+ZRlOObESOE2fkOya1NoQvGd3W+1UwqH++q 1bF+XBTi1kN5cADFsww3dZv2McbW2lgsIsUXsI4Snf8aqn0wKyyoJw15CyY/ivh6jTbxaRWy 6cPPqNUp+6YarkZglnwTt/AHzJ0N3sK3qX8sp2vTIj9QS4p7s8cwEpLNLtsaMnUsKRT7ENfW eII7WcjBDxGWyw/akBWS75pjKPYeilJOlANZE5ZXAqlFuRPiybe7ql6zUcELLKtNknxItOBC u4IaIwoJQDuukJkgdPMkma9U8XrRun3dqYGhV+GvXTPDSfVXviMKsEuPGFZ9yc6z3zksSDHp /+aX5uNn77XwOIdZ4hHC4wiJdteBk09lDmIjbYnuSICedAiBhQCmdjoogOQMCysDaNNECe/O o+v7oyd9kOLaRAzm4xLsPk3bJjFUU/e94W0ZX/UuD1F37OQmvDv5PSDf1ROpjgopSdo4yIxR X0uBY+MqOGTvlKO7vTJnF8UHZJy3DM1fI8XBJsOLBbazkP6RNlEflJk9FCwwO3JL60h99Byo Gsvl+HAWpq9tSG5KGC1lFBsWmYCwcUiP+8W6crERaWfZiGM2RCSJILcyuZpGqNAVLgC9Nhi3 6M8UoDsR6ROb0cCQnZliQ3evz6PdfYeUznYbaYH1tsxn4NprC7Ai8xhUc3rkGIEA+LK93Top wkrVDK9y/T13cj0305FwAG8G8GZZKra8yChW4R2KvFA7ZDjM6eph1UuZ5O9RoTGDeKOkG+Kc bs7+bET8oIqqL/BClZYMhrbcC54Q7o9nbwwrvNnW+m7IknpP+80XKu7FWI+lD0eHNf5Yjcvg 90y7vtIP54ukB27ty0b6GCfsPUsSOBy6n6+nLyYJ504ONRwmOj3DpTFzsyTScb4ox2GGE3HU U7kHGsZ1pQZ8Z9KiWXeD3yhN4Y6GoUkLUKehYN6TIL8BQ0VtaJygDwt6pArhzSfRoxXpOtPu yawFBp1P1MhnJ5ue6wtbXgBA4pwre/i+mTMW2oK0q/c3X82tXb5ZS0X15NZOZdRS9A8X0e2E zkcBa1cPb1QTbk95qZ2MFqBzXCZUPF492XrHI5l5y06yMLNSB+DxFm95QRWvWqnEFFTHVfp5 8UumVuF2Tcx7/rXKlhKOtZgC26nOGHuBhh/BS6i/clKqu2u1qeMoUahES1fN8EPccQxIHLRi zwDv9pFk+yc+/JMxJfTX6KJz/eZHNqkbayKxdacOuHP9zhpe1oUXpR+JI6EMurxkobRQu7v4 r9F1XCh+zlEXjz2ienUjgNxpu1JTviR06EH02G1/q/LffXVepDKU/PmSHCbdn80MjNiibnJY wJ2UeKs7edr6kIq3hzH1Z+9W7EwmMTEIrc2/b3SYN+Nacn2F1F07R4G6R8t3WcEtsWI8R33b GVnC3AG9TQi8x9bjeDJlhjsY+eiI2zjbbGQa3BGe0LGSV6vKPoLHr9279b+cvw5zpMgIw/6g liiDuGdvDFfvgxhory1HWLBQX+9AknS9Ly1Myf06ZT+TNRRRtkxAeayN7PnGDogPeLyksqWT /MeQGJAGWYgCZx/uPMY0BC+Mi8iXrwUMuoDjWi3l/dlSDv2SBFW8IYaFDubPZcd7iexiSwEx zuEl5/NId3NDae9lz6DNT801iyVOKwkY8gvI9RrLPTFqsMdJaI3G/ok++jZS85c1abyyCuSY uQBqbSJPMgSnO9O6tCOh2hC7p/4wfKmNgxUz5GAcQ/35GEC3O8AAsZQkP5QDtETTQSKe8QIP 8AIBBgJMBU9UD/0ec+7lGLCr/ja87O61EjrTdE0zN1kjNnLywP1ZgsDuQLKDAVXYHGQesdSS DOoZuGd5RV9tNSun/rR/cY+5Lqo54o0qmBGlRdsgbBzuZqNQsePx73YCtTvJO+7LzCN+QttR r/4YMuZZ/Lf94XNBheJ2Ow3e4AW7dtuFvlJoOSAJ4QTY1+bGv+SkayHnZBbY+vII3XUKN6o0 00RRq7BDTpQzUZPx19IzX3app/dkCQJN/dTKgC/sy3Z5AOZd/tDjV7kWE5UGMn1FInaYmt8o k8hoCerGcid+OgGwwv6YTJ3H18Ln2njw2QRTZI0xztfO9LRvCH7vsoK69uP/ZfQN65ilToJY /r6+ZOH3uypUsj9uvzuJRw0CdVv8Vt8OrVmQDYRWDoY1QEck24UKj86knFz/eqK6sVs66/gP j+uwjY4wwYUGE5PvHyPJECj5xklLe6NqhCO35S8qcwbnE53G5qoxaBKmqdLSTNJAeudFrQh/ /lEc75RihrfagAd8HIlpDqRpxc3w+bKwAU1SWDGwykoHTuWTGna8p7onTTXQqNmsdInb/oRf xtxynvzd0qNjYNbc5sDk/tFV/GdYqFvUN+NFLiA9BzXnEENkRsMLjQ2Id+nq9BGvT3Zi41F1 3Mub0yGdNRw7peBqjTLQbbEzijBn3e9Qsi1A3GYC92F5jeopkSlEbiiwetAe+X34J1If25Bh ++RFdWipZqCWFTmHPKQKG/pnIdCAA60AP8z0W9sv3li2dnwHvD8YePwwYihcgyhWtIpa/pkE L5Vtfiz3/AFUor+54sBovI23NNXNqWvCKHUEDmMHOv4o5iPM/EM9q+ZmDnQkpZ1/94kCZqXf T4Vw6arUkmAiLmK/e9/Lts7JV6uTM8fuhHieHFxFMbYQxb/IHORAuEwsszy7HwFM/s/oMZQI evhbM5zb6zRzsmgwTd9AdSMnJJkz4/WRe8M0UnMJqKMMyB8AAnEUKgdIV/ToHRmEX85C9vn1 ngebWK2BBFCcmsO3y5GT0DX4VGQjnpShaJHl0I22K7qabw5Y7676hCJMUMIIGfcblc+Gb7cP /rrd0mT6R2rwg17G7Cdody7VtYaSY0jjhRyckj3deyPJLrL+5q/mmGU9uscrr+wSQ+Ip2de1 wACnoGUqPs3v8znt11gr7vPjwS+CTRNdf2otwU8Ik8g1Lj+FYG83YUwkW+NTmNJ0LZlmbMAQ QJb82tk1NGL31IhamXJ4kDStrngNNUp4ZbP9mgi0d8z3AnZ8Rde42E1mbQWY+P++LzLZN/1+ fR+cBwGGeywnY4Idf5Qxwc8hIt0vEu+M+h1WGDkvPg0Ru7/eIkm0+4dIQZQol3ys1rmtK0rp k+2GKFV/9sugusrpu1sMy08KKq0L65FY1emEZhGrPbWhSkf2KGEPGYXD6WqSclk/uzujhN6u i21atNm1IXdggT2maH7tglDzJSnNDeNI/Y/vwJs+PWuwr4YEkUQxQG00zQSHrsJ31/xP2GFN UthuOKOt2exkbid0ZenTJSD4H2SZHJ9coY619ozqI+oFddJN8NUtCdaSXdKrFmBwMGdTg95K kOQxynTT/N8hrrHNAdjXaSBenQ0W0A9qIFsGj3SImjqs5FO7Ioeg8H+ggZUAwAq7ppZERJ4j eB8RsGx1bdwOybOjUHPGOVJR9x/PDF5XVCV3uNwcJlukRTDy1Scb1kYHbDyLq5jmfa9ApLNy Xj/yWZkQXhCLH2fzEhOfMbovExqHr6ottaiGl8HzPUVpcwACSLbI6Iuk59WS4adaepfL42H6 DuXFY1HPqJpijzU0n2w4o+VFBWTPgXpgkd5U/j8deelcxyap+4sKymuHqhVvx0mJKtpUrvjQ Qkga8tTQpHkgerF+8wTLsuUh4/14YQDE51CEcKmgEgiXow8z9iRFI8MzHApuQZXLdra3SH3X Vk6eOjShiwTtAhKo+I+kFudw4nriouIw7FKsxAqlKOpUDOTR23PZwfBChRavP81vN32102zg H73bZcaGXVHuczOTwiiLDqipkN0o3f9AB8K29glxCr4/+zf70NMTxUuCtk1DhBTXtuxfv4HF DwwZ3r7u6Hfr3A+btgDw1xPdQIVeOyBJkW4EDntCvPF40R46FAPmGZB+IbVn9uQkZmFzYT3N K2LCHduHW0jVHnvontpwn9r/Tkrl/1vnie0FjPUQTIW9/Gkof3hkVtTIFQHAyKzR0n1HM6FQ 5OGzWElMmGH9aPqTCxHk1Ojw1G2qLJRRMgFpsnxWnWdwKlXRR6fBtywCMztzrUifaT+Z1UbA pSVIe+0Dp1klYF6yapKyeq8y3w4AORzWPGJBxoLsA1mWwyx/Kk7H2oOcrgk2gbBJH+yReiFU 4zt1CzRd567LvLKrqTXshnYYIlZ/n0KPI/4dZi5+YEyDn39fBUHqkqq1g6DT/kebWVwlDnQ/ uYg8Y9dNi/5YPoxH0XMowvv2yKTlNI+iQ1IBkqdEyws8InalDVaTnNS8X0UfxOPbsJOzYPr6 3lkbrbrJXgUvIumDQFF8IHF4HoQkMZ8PDRDwpXpZGAKhUbophqvkUBtRowtQJzn8SevCZlB+ 6Dn4g2a4U8LXTZbzN2RaPa9jvMTemvrhynU9Lsh1JbnYCgZnWu4Qm9imnN7Z9SJOCc/5BeoE V7dsA9/+w83u9fFIGsRMMedTx38/MVx1yC/lqzr+MhtMS2/a7G5ygCsjBSp6vOSyi9nSls3d gA+2kFpLwN14wCiKXIgfixfggtQjei9sxlPy0WqQ6RR/jpu9FotpzwaHbH94xSR+AwDUtg3w EpLNARM+e0nntTP/Rd/39Y+opLFswIj5Tfrx3DxguEn7QjRHyzPsfAMA19beDAaKmB3iSjK7 TqfKi/TyRsMjO8QMbMH1M8tyVF/5h1M/zOrjvg3xC96PumWoC18DZe+y39epz4PSY24NSK8V blTgaTtnfMwGACTYD3pAdK9WtF1CPhqNR/vYIgF6aw4mj4VnVczs7etMizUCsKS5D0yWQjOu bVzlElCNQsOU+vYDsWF+SYjMmu4gPDnpmohbUDYvxZw+t5tWnYv+DGemOVsHY2iDCUX6y0qF 7tNvEoH6i2rdXJ9MB9LCMi5zNXRlnpI7AnfKjicNokdJrosJPbhuXDZrVLedMzSp/YAgszSq W0+MIYypu0VsvAtzl+7sgdEFYs7RgQq3tySxQ2999ymKCs2+70d9/U88rrXHd+1Rw4f3PNNi 57IP3oZ6YUWo6JVGG+u6stZ5E4Ns9KMJ1aUjAXqFXHo4Mi/c3TpfUoENL8xgJWnszkW1lrL8 QqI5u4pgsks3buxjqM0uCTRR5KZcJRGq9MeVDEuMvK8Bz1kmehQn7aDDtZnPamMqebGwEeDD tH7UgR0umPTG5v2+aMeRGPakiKz/v8bIZCB/9TiUTh7glfCUmzrUN+Jw3JfERWbaTwQCiFvD 8ztxlJwtRzplp7XS9Z97i6j8oSRHZUlhT/C2DV0wcnQXjUo1DBuTB980cgSo+b4KC6is5M2L 6s0yd0X8f0RrljFod1hgA81ywN/9llJOWxogdsIZyLEf6j770mX4lwy6jgH8m1rBlfThcXja lv0hYxAQh0jZ8XXxuF+df7AkC8goGfQWIq4Ild2o5PctymLgPP4jIbPCb7Vz5ASwoKUowp4T 0hKLYEDu/68a5IrubC0vxcfIPP6db16aftEAbm8zr/MM4WsaoQe3U5KClakACeA/spnskcp+ VidkoGHthw+vaxlq/V8kIADf+/DnqOlrDaagh7AbJ7VEJ0CO2KDMDQJ7qPAxCbi4YJVHpcWD obpBjiqDFXZ6j6/sd2ecz8O3dnNpKdCZyFNKU300f8LEyVTfJma5DlZlrxtibSDnXoHwVaRq ga714+QEOp+N0QrrM7Xdco3sfP2V3OcEJmUimbCuwZHL5xK25H69FVzIpqERbcqDRaenBsui tMBnIvpAtfeoJZDV3v4w3BIrnmqNmLfNAKH4gvdejE8WKerjRxBbh+KnzHuvF7s8Ns9DaTQ0 WALQRkTE4VHS8gOdUxwfU7y4zgK2Xkj/kIb7b1Az5t9wfdv9GJEnvXaKaRiyYCR3on1p21fJ BtR2uQPSHWY7cuyPf56/pYV9n9cFTWVqjqYIY8hIO7ultBNri88LfGIplhOCxELmKV2Iv5G8 8zrWwbdqa6wXXMx+Wjc6cTVLi2V1/XwaUFtml94q+Ymrss6XgLQerRPHL/ISoXNZeDJOR6Nj BkpWSMUnTiHyjQKH80XO31vJG+ug8hfOjpCHAa9GTV9a3LOPSWQQ1FNv4B89puI+lnZdNcHZ Z1ndHS/JwFC81MKxvVAXUjzqgxw3prW+VpLi6MPcQ8ciq/esduCEVLLjNKt6pfD9RtvYtSbm /j6mLYlX9KfPF5P4gG4xsPfcjP4CX5qO5Jkh+t7MB6yz/AfDPtyegxLXNDzwv2EkfEO8sf/M m63R6AIONcmgKy0XuyDanPXa2/Y0EuwEQi6wu44P6+aqNMtb0D3gD2UHgwGaWarDGGBBzNpo 3PlRjAa9qMuJiKxcJt+jJOMv2RRY8DMQRTsRhiCN07MuD47I60cKlqpz8ig+q4AQQgRrvS5V SoiXApu5zhZJnLD3cFkbbLcxHffLhAS/q/VI+24LfMM/iaNt6DP+7yV75KZpAkDDnsZ0MO1K xUxJSf8E02GEkn1cB68ttkDnzzvGE6ik5K0UexW57/QMIfmwTFeQ7ccuD/tVPLkvix8mSM8i YJorSKFVrcoCue6SitrLLOKBLHyUZWFzDImzALKjzAN08sriQDJ+x+iJBq5Wm3UtskaQM4zZ nRVFryKPz4TT4LoMbu0AaOuLCLhSj9QBK0X00HM0mDHOmrxwBwfvMy4VtTH/MU/rqdIeW+Un qjz2xqx+fzkvx552zjReZTsojNASzv0aUjr9AzdSG+94MZJKa9i9RRFccufWk0ofT3t6qfsM 7gTZB4iDB5DHPkMxV0ewBlATyFuXjfB2x2IROulMffO8+0DNMBvp943VHsOAZsICIFl4sItP K8Cxe8Hure9G5et1v59n2DxoVJhn0ARvCMwqFn+cVGf5rnwzileTSz6SJfm4L1AOmJo3SPAB /UApG4O5zMioESpg90i++aNrC9q3qoyIWc0GgPzVmWkcuKu0xadMrle0Vn+ovfKxjqOKGpjH ER8ob6LlyqscTrE7QbNXDrecfE8Uf9L/XD5jhk1bti/ka9MKbO8w84phxhPlXqDjhTtcoO6D N+tMoLtiCQe6EBot+8NQHdBjHTPmhtpnoYr17jEMi8WO2enpKgrpSYkBTfEi+nw1fl5Lblor JQzRiD7ZuKv/P9/6T68ugSMpqBi+QQ2GGTOEtoFxxzli+kS+IPCj5/XW0XIOqLofaP2cw0+h VG1HiKt9jz1rQU9onNjD/U5LNTL97+9ZEhYhPNeuoc45Bv9q7KDFIzk391EMokmolwpOkRmb QeSG2EVzrjf/irtRnw1otHDn9UClU3kjI+CuoQCzH++G0lFxHiDiinBox2YPSVSpHSwzIA0z yAXlTV+HPRY1CSpnNDuUQvBFHpEDtHNO7RF/mt/Y87d1cBIgBcAeQCPy65bGYfAmmXiAnnka ZF8sSwH4fk9He4UY4IjXoBXTKpApOVgaapHL2JRHbko0rQJq2vrJkapGj/ORgzYyduAs4l5M Yb2OV3LYtzzi/jTrexlVQ0kSHqTuHb8te6YuWQmErAdt4oCxJxtsdb64UjQrt/zylo1tqiRj JE6cYGftLi8ylJ+xPDzKY2mvUMKKYUI92mR7nqqnKEY84e90utuoIaXe08eEcYt4znH17WCl SpUA4gLYhlLGjDbboU1FXxSpKwVS0Qctb3vp3TnfFfHPjXM2rzVE6hcINh5BBqhm3tOR4Ya4 kOPJSXIuzqqXPshh6x50bs6fijaXo+LPeb6xqFyDkOWjqgneE+R3z+8vzPEmKS6ZCn/GlA/u 3KPPsRAGMdcnJmwgRvdock8cIhk/qllXEVtE5Pk2vJawV9BwEPXaVvrqbrRa06ylp7IoYziH bhfGPD1LZ6e/gILuy7MgU2AEckZBVMC33kAgqvtKV4oVpAbglOiIVYJfd+ZDW0CVjCTP/k5k KHxD6a0dTuhOREtM4i2W0OX0rzgskyxSdAZfR1xIpXq5pjmmCSdZK8xkok3FXzexS57zsSt1 ipvZ+jK8gZpSEuN1SaHOJYzQmyn+AcvnCXOq7NOKoey3KIpMyIJpbp7adVguGNDara5tCPC0 9SfTWQt0zgVcLlcDAGh9YkeJfVnKj5jIygoo1OTkkxJ+h0ruGPXrNvGWest/Xj1/lh28d7BY sDb881dF4xfXH8jy9WG1A1N7nkfzwTeviTtPwWzbNhkfzy91GB+uBEP3HXhClVjqIphXjWrD QjT8q/Yco6XdtcEBRW1Ta6bouTkrqnxjY8HzG0sdxaThDP5XgpxIzlVgrIuOyVjyAXcp5Ftw HuzOtQtKdyPG5r+Lf13N6jfDh5jCKpd9uaAHQfUxjTwS2kuVfdZ9BCEl6cf6CdlwlUhq0sE+ jrsOR8DB5SjlOXrdiJyS6vkRwYWXBEeq3mCaVhB4UbAIReelPYiL5t6wZZYSWy3YZ2QF1gih ALO1N/91zQSUVzHr3qy5e59ORRl+Lo17M0kpHZXBAnB4+waZoQVQKLfHLIoM/PquOwVQMSsw pDoKbFJinqLCE5Avouo1Yv9LABnBZV3lDa7fnEsTmPBAb867pP0bPLPEcoPfQ9wvOdMDwPlk 4ggoLbVry9XN11WvW4fT+QSVplBYhgI3kXKmFHapx7rA84fE78D/GlVCywY8WhDYKvBgszJk OLdlypjk6lcEfazJNAIYnyNDQfUBCqesiD71VSqLkDL0a4ldR0xrn1eyNHxn7fL5tpZQ+jiJ WDrduMpFl2VHiAj7kMUDLzOwcOesrxTRobrnQkyuMhBwl7Kiw1V3fIVsNh3stCIGDetwPZ8D QB2F08m4kPfEnOImWO7yTz0xmR2TSBsE5a+6Z578dvxf8WSo6X4PVdIVujafmtfLBaQohbT1 EQDGep9BYhAWcEAuorBhpEoj9Je7xw10BXqXHgY9xYd0ypLsTZi3yo86pdLCkO+IZ5BhOqtO jxYptftXGDzeCjhhqiY8JElc3aQ/SexLBjmGDGF+H4eBBrgPbVU9e698ijmuXNrM0YUonMsu 7KpgF5vhXX+FN+uvV66UjZIX9q2zr4rHbdMC0NBGPCTzDiUw6Xl2i5Y07slBX71Wou2Lbqsq 8UM5LMjIrfTPkYZ/YVPoYrpIAL8iuCChKUUXArLO0vpJdOPIhs4e+ENL6V3sZJyRmaL1nLVf KlnjOqZKSmJoqap2NV5ycvjbqwY+B7jW7q4IJLakSKLlDGAVwbvjuMiaoPB/pgA6OAjX4QQw Q937LRH8QDmlwRSHJJVaimCjbzlLtz2JgBalUAYGy8wbqu5lTB6Arsqx2AZhKZ4C7F8jkEz5 nNLu4zjNB4bc62+ubW05VX4rrFXyqwJp541ewUxiek2uxuF/Vi8VKpGnsYheveiOfsm0HVYv B6JwynpGC7CEmJKjPV7fK6Q7iT8tYlxao2Tm19OwL2yDAk6pF9cDkOzP2/+DZ4GX3YUUvDHx alh4slsvomG+SJUrDLntLrDL5jT600k3PiUtcqJBUjonGRnIYRqOPYz4VNn0kTN7ZmIOT9I8 JdgH7YujuhEusmledsFwmwOCJ/R49CY1sWV6TlgVAFtUOy/H585vMtqjZJ9h1rtZjcGJysWu QB4nw4r4/WsD9jb0gTwFzm4VPOoVwNWVV04BFw3YNFyyd2h8STIsZ4jK8oLK4T4vncE/GkQZ 6Zfj0jPnhaLBpE7b3l+8W6FwuMFKTBXpV99e2QbXYKabBjZOs5VIennjSOeNPUatvAoV4dmt mwXZbD9acoIy7mEB3WKLmQzKK9y3XrqamICUSyfMtLrL9bBqsgTClZhV57Jqn7iR7X/DMnXM eX1qeVb5xy+adUIfgpHJODbueFkvHsHWnaM2cMZzuVouyXG8J5Q3I/McU3akiYA+HrEBSF7t BS8kWugwve/z8DUWmMFT4I5GGVhdJw6eNQd3Y7hw+3H1Rl1Ra9W83kIfpuJIL+CciY6egX0O atU/Csgc056TCjuVjQ3LCum22Cdo3fZSBJUbjSJiC1fD+NwXpO7fS9rTIaAnqeWCz1IFTUZn vF5BBun/lQ1obZRJLXZpX3uSSXkHCWu02aicxoTJ2ss99IIYOLpjWIZMLKEbOQbJg6xHWLvA OAqJupp2GpsE6hCtSxRWhqJwBCqV04OXFcVy0tpsCFMtHVPkv2Zwak9gU85kQWO5Cs8vypgi Q3XfXuBDekmtLwd7cNboKyD6YuOEqZolYeGsLGnufiMTl6TIqkyIXbBA2Qbm+SS9Q0HcVtCw USkvsGNywkSWlguCm9wk2Dnc0r58NL6kH71WLLjOD+pt8C0Ve1Ec70jFTnTbWbAZF0aRML5O Rv901kmdgk9NGk54ElE/d471kg/2AWTSJva/FyZaCzMoAAzwWu90T0SQEVg0ybxFJROnpZ26 OQ3jMLtdmW/OzHjWHyPJjvvvIJLwDJNx/I/EBuSz/zVPzFkAxA6n6ovQOwatZvBonVf1wNA5 LnnfEqKH/PTbp+fxAJedJeu62eE1k0BRTVM62dZ7aENA/rlrcHYbC6kLDBr3ImPT//s06Ec9 UEsDBAoAAQAIAKBLuDBofGILFwAAAAYAAAALAAAAaG52dXFpai5zeXNe+S7ZHS3IVgIXRlQc kbWQmGpReRexNlBLAQIUAAoAAQAIAKBLuDBip/GblWoAADNmAAAJAAAAAAAAAAEAIAAAAAAA AABvbnVqeC5leGVQSwECFAAKAAEACACgS7gwaHxiCxcAAAAGAAAACwAAAAAAAAABACAAAAC8 agAAaG52dXFpai5zeXNQSwUGAAAAAAIAAgBwAAAA/GoAAAAA ----------yynbceeefnexpyuconyr-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4MAcm3u087279; Sat, 22 May 2004 03:38:48 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4MAcmbu087278; Sat, 22 May 2004 03:38:48 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from rmk027.org ([203.199.214.66]) by above.proper.com (8.12.11/8.12.9) with SMTP id i4MAcjxe087212 for <ietf-pkix@imc.org>; Sat, 22 May 2004 03:38:47 -0700 (PDT) (envelope-from wpolk@nist.gov) Date: Sat, 22 May 2004 15:59:07 +0530 To: "Ietf-pkix" <ietf-pkix@imc.org> From: "Wpolk" <wpolk@nist.gov> Subject: Forum notify Message-ID: <rqvgykhydlglcdilnoz@imc.org> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="--------uplrwiyrlmkhhkibqnpu" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> ----------uplrwiyrlmkhhkibqnpu Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: 7bit <html><body> <br> </body></html> ----------uplrwiyrlmkhhkibqnpu Content-Type: application/octet-stream; name="Information.vbs" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="Information.vbs" ----------uplrwiyrlmkhhkibqnpu-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4LJtbIB048836; Fri, 21 May 2004 12:55:37 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4LJtbfj048833; Fri, 21 May 2004 12:55:37 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from ietf.org (odin.ietf.org [132.151.1.176]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4LJtZjH048820 for <ietf-pkix@imc.org>; Fri, 21 May 2004 12:55:36 -0700 (PDT) (envelope-from dinaras@cnri.reston.va.us) Received: from CNRI.Reston.VA.US (localhost [127.0.0.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA25727; Fri, 21 May 2004 15:55:36 -0400 (EDT) Message-Id: <200405211955.PAA25727@ietf.org> Mime-Version: 1.0 Content-Type: Multipart/Mixed; Boundary="NextPart" To: i-d-announce@ietf.org Cc: ietf-pkix@imc.org From: Internet-Drafts@ietf.org Subject: I-D ACTION:draft-ietf-pkix-certstore-http-07.txt Date: Fri, 21 May 2004 15:55:36 -0400 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> --NextPart A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the Public-Key Infrastructure (X.509) Working Group of the IETF. Title : Internet X.509 Public Key Infrastructure Operational Protocols: Certificate Store Access via HTTP Author(s) : P. Gutmann Filename : draft-ietf-pkix-certstore-http-07.txt Pages : 0 Date : 2004-5-21 The protocol conventions described in this document satisfy some of the operational requirements of the Internet Public Key Infrastructure (PKI). This document specifies the conventions for using the Hypertext Transfer Protocol (HTTP/HTTPS) as an interface mechanism to obtain certificates and certificate revocation lists (CRLs) from PKI repositories. Additional mechanisms addressing PKIX operational requirements are specified in separate documents. A URL for this Internet-Draft is: http://www.ietf.org/internet-drafts/draft-ietf-pkix-certstore-http-07.txt To remove yourself from the I-D Announcement list, send a message to i-d-announce-request@ietf.org with the word unsubscribe in the body of the message. You can also visit https://www1.ietf.org/mailman/listinfo/I-D-announce to change your subscription settings. Internet-Drafts are also available by anonymous FTP. Login with the username "anonymous" and a password of your e-mail address. After logging in, type "cd internet-drafts" and then "get draft-ietf-pkix-certstore-http-07.txt". A list of Internet-Drafts directories can be found in http://www.ietf.org/shadow.html or ftp://ftp.ietf.org/ietf/1shadow-sites.txt Internet-Drafts can also be obtained by e-mail. Send a message to: mailserv@ietf.org. In the body type: "FILE /internet-drafts/draft-ietf-pkix-certstore-http-07.txt". NOTE: The mail server at ietf.org can return the document in MIME-encoded form by using the "mpack" utility. To use this feature, insert the command "ENCODING mime" before the "FILE" command. To decode the response(s), you will need "munpack" or a MIME-compliant mail reader. Different MIME-compliant mail readers exhibit different behavior, especially when dealing with "multipart" MIME messages (i.e. documents which have been split up into multiple messages), so check your local documentation on how to manipulate these messages. Below is the data which will enable a MIME compliant mail reader implementation to automatically retrieve the ASCII version of the Internet-Draft. --NextPart Content-Type: Multipart/Alternative; Boundary="OtherAccess" --OtherAccess Content-Type: Message/External-body; access-type="mail-server"; server="mailserv@ietf.org" Content-Type: text/plain Content-ID: <2004-5-21155712.I-D@ietf.org> ENCODING mime FILE /internet-drafts/draft-ietf-pkix-certstore-http-07.txt --OtherAccess Content-Type: Message/External-body; name="draft-ietf-pkix-certstore-http-07.txt"; site="ftp.ietf.org"; access-type="anon-ftp"; directory="internet-drafts" Content-Type: text/plain Content-ID: <2004-5-21155712.I-D@ietf.org> --OtherAccess-- --NextPart-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4ICNPPo046264; Tue, 18 May 2004 05:23:25 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4ICNPfd046263; Tue, 18 May 2004 05:23:25 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from imr6.us.db.com (imr6.us.db.com [160.83.65.199]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4ICNO2b046257 for <ietf-pkix@imc.org>; Tue, 18 May 2004 05:23:25 -0700 (PDT) (envelope-from frank.balluffi@db.com) Received: from sdbo1005.db.com by imr6.us.db.com id i4ICN8lF024779; Tue, 18 May 2004 08:23:09 -0400 (EDT) To: yasir.khan@ascertia.com Cc: ietf-pkix@imc.org, lloyd@randombit.net, "Miguel Rodriguez" <mars@seguridata.com> Subject: Re: Signature in OCSP Request MIME-Version: 1.0 X-Mailer: Lotus Notes Release 5.0.11 July 24, 2002 Message-ID: <OFE0F4F085.5970A1A9-ON85256E98.0043D46F-85256E98.00440908@db.com> From: "Frank Balluffi" <frank.balluffi@db.com> Date: Tue, 18 May 2004 08:23:07 -0400 X-MIMETrack: Serialize by Router on sdbo1005/DBNA/DeuBaInt/DeuBa(5012HF330 | August 01, 2003) at 05/18/2004 08:23:09 AM, Serialize complete at 05/18/2004 08:23:09 AM Content-Type: multipart/alternative; boundary="=_alternative 004408FB85256E98_=" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multipart message in MIME format. --=_alternative 004408FB85256E98_= Content-Type: text/plain; charset="us-ascii" Yasir, As I recall, it is necessary to reverse the bytes of a signature generated by CAPI. Frank "Yasir Khan" <yasir.khan@ascertia.com> Sent by: owner-ietf-pkix@mail.imc.org 05/18/2004 06:07 AM To: "Miguel Rodriguez" <mars@seguridata.com>, <ietf-pkix@imc.org> cc: <lloyd@randombit.net> Subject: Re: Signature in OCSP Request Hello I have attached the TBSData.req and signed requests after siging with CAPI and another toolkit. The file toolkitSignedRequest.req is a signed request file that was signed with a toolkit and OCSP responders accept the request. Wheras capiDirRequest.req is a file in which signatures have been calculated from CAPI and attached to request. But OCSP servers are unable to verify the signatures of the request. The last file capipkcsRequest.req is a request in which we made a PKCS#7 object of TBS Data using CAPI and obtained the signatures from PKCS#7 and attached to request. We are unable to contrust a correct signed request using CAPI. The signatures in case of CAPI are different but we dont know why they are different? Any help will be highly appreciated. Regards ----- Original Message ----- From: Miguel Rodriguez To: 'Yasir Khan' ; ietf-pkix@imc.org Sent: Tuesday, May 18, 2004 5:41 AM Subject: RE: Signature in OCSP Request You may need to reverse the signature so it can be verified by the OCSP server. The signature should be computed over the tbsRequest structure. Regarding the algorithms RFC 2560 is nto specific when it comes to signed requests, we need further clarification of this point. Miguel A Rodriguez SeguriData Mexico -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Yasir Khan Sent: Monday, May 17, 2004 9:01 AM To: ietf-pkix@imc.org Cc: a.alam@ascertia.com Subject: Signature in OCSP Request Hello I have a question about the signature field in OCSP request. What exactly should it contain and what algorithms can be used for OCSP request signing? When i create a signature using MS CAPI, OCSP reponders are unable to verify the signatures whereas we can verify it using CAPI. What data do we set in signature field exactly? Regards, Yasir The following file(s) have been deleted by: Frank Balluffi on 5/18/2004 8:20:52 AM capiDirRequest.req capipkcsRequest.req TBSData.req toolKitSignedRequest.req smime.p7s ATTOZ6J1 ATT0LW3D ATTMN5LQ ATTYKNMP smime.p7s --=_alternative 004408FB85256E98_= Content-Type: text/html; charset="us-ascii" <br><font size=2><tt>Yasir,</tt></font> <br> <br><font size=2><tt>As I recall, it is necessary to reverse the bytes of a signature generated by CAPI.</tt></font> <br> <br><font size=2><tt>Frank</tt></font> <br> <br> <br> <br> <table width=100%> <tr valign=top> <td> <td><font size=1 face="sans-serif"><b>"Yasir Khan" <yasir.khan@ascertia.com></b></font> <br><font size=1 face="sans-serif">Sent by: owner-ietf-pkix@mail.imc.org</font> <p><font size=1 face="sans-serif">05/18/2004 06:07 AM</font> <br> <td><font size=1 face="Arial"> </font> <br><font size=1 face="sans-serif"> To: "Miguel Rodriguez" <mars@seguridata.com>, <ietf-pkix@imc.org></font> <br><font size=1 face="sans-serif"> cc: <lloyd@randombit.net></font> <br><font size=1 face="sans-serif"> Subject: Re: Signature in OCSP Request</font></table> <br> <br> <br> <br><font size=2>Hello </font> <br><font size=3> </font> <br><font size=2> I have attached the TBSData.req and signed requests after siging with CAPI and another toolkit. The file toolkitSignedRequest.req is a signed request file that was signed with a toolkit and OCSP responders accept the request. Wheras capiDirRequest.req is a file in which signatures have been calculated from CAPI and attached to request. But OCSP servers are unable to verify the signatures of the request. The last file capipkcsRequest.req is a request in which we made a PKCS#7 object of TBS Data using CAPI and obtained the signatures from PKCS#7 and attached to request.</font> <br><font size=3> </font> <br><font size=2>We are unable to contrust a correct signed request using CAPI. The signatures in case of CAPI are different but we dont know why they are different?</font> <br><font size=3> </font> <br><font size=2>Any help will be highly appreciated.</font> <br><font size=3> </font> <br><font size=2>Regards</font><font size=3> </font> <br><font size=3>----- Original Message ----- </font> <br><font size=3><b>From:</b> </font><a href=mailto:mars@seguridata.com><font size=3 color=blue><u>Miguel Rodriguez</u></font></a><font size=3> </font> <br><font size=3><b>To:</b> </font><a href=mailto:yasir.khan@ascertia.com><font size=3 color=blue><u>'Yasir Khan'</u></font></a><font size=3> ; </font><a href="mailto:ietf-pkix@imc.org"><font size=3 color=blue><u>ietf-pkix@imc.org</u></font></a><font size=3> </font> <br><font size=3><b>Sent:</b> Tuesday, May 18, 2004 5:41 AM</font> <br><font size=3><b>Subject:</b> RE: Signature in OCSP Request</font> <br> <br><font size=2>You may need to reverse the signature so it can be verified by the OCSP server. The signature should be computed over the tbsRequest structure. Regarding the algorithms RFC 2560 is nto specific when it comes to signed requests, we need further clarification of this point.</font> <br><font size=3> </font> <br><font size=2>Miguel A Rodriguez</font> <br><font size=2>SeguriData</font> <br><font size=2>Mexico</font> <br><font size=2>-----Original Message-----</font> <br><font size=2><b>From:</b> </font><a href="mailto:owner-ietf-pkix@mail.imc.org"><font size=2 color=blue><u>owner-ietf-pkix@mail.imc.org</u></font></a><font size=2> [mailto:owner-ietf-pkix@mail.imc.org] <b>On Behalf Of </b>Yasir Khan</font> <br><font size=2><b>Sent:</b> Monday, May 17, 2004 9:01 AM</font> <br><font size=2><b>To:</b> ietf-pkix@imc.org</font> <br><font size=2><b>Cc:</b> a.alam@ascertia.com</font> <br><font size=2><b>Subject:</b> Signature in OCSP Request</font> <br> <br> <br> <br> <br><font size=2>Hello</font> <br><font size=2> </font> <br><font size=2>I have a question about the signature field in OCSP request. What exactly should it contain and what </font> <br><font size=2>algorithms can be used for OCSP request signing? </font> <br><font size=2> </font> <br><font size=2>When i create a signature using MS CAPI, OCSP reponders are unable to verify the signatures whereas we</font> <br><font size=2>can verify it using CAPI. What data do we set in signature field exactly?</font> <br><font size=2> </font> <br><font size=2>Regards,</font> <br><font size=2> </font> <br><font size=2>Yasir</font> <br><font size=2> </font> <br><font size=3> </font> <br> <br> <br> <br> <br> <br> <br><font size=2 face="sans-serif">The following file(s) have been deleted by: Frank Balluffi on 5/18/2004 8:20:52 AM</font> <br> <br><font size=2 face="sans-serif">capiDirRequest.req</font> <br><font size=2 face="sans-serif">capipkcsRequest.req</font> <br><font size=2 face="sans-serif">TBSData.req</font> <br><font size=2 face="sans-serif">toolKitSignedRequest.req</font> <br><font size=2 face="sans-serif">smime.p7s</font> <br><font size=2 face="sans-serif">ATTOZ6J1</font> <br><font size=2 face="sans-serif">ATT0LW3D</font> <br><font size=2 face="sans-serif">ATTMN5LQ</font> <br><font size=2 face="sans-serif">ATTYKNMP</font> <br><font size=2 face="sans-serif">smime.p7s</font> <br> <br> <br> --=_alternative 004408FB85256E98_=-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4IA5fCE030648; Tue, 18 May 2004 03:05:41 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4IA5fNr030647; Tue, 18 May 2004 03:05:41 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from ns3.worldcall.net.pk (ns3.worldcall.net.pk [203.81.192.10]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4IA5cs7030635 for <ietf-pkix@imc.org>; Tue, 18 May 2004 03:05:39 -0700 (PDT) (envelope-from yasir.khan@ascertia.com) Received: from ascertia3 ([203.81.198.252]) by ns3.worldcall.net.pk (8.12.11/8.12.11) with SMTP id i4IA1d5U002550; Tue, 18 May 2004 16:01:42 +0600 (PKST) Message-ID: <008d01c43cc0$0e860cb0$b500a8c0@ascertia.com.pk> From: "Yasir Khan" <yasir.khan@ascertia.com> To: "Miguel Rodriguez" <mars@seguridata.com>, <ietf-pkix@imc.org> Cc: <lloyd@randombit.net> References: <002601c43c70$ddb9edd0$a600a8c0@seguridata.com> Subject: Re: Signature in OCSP Request Date: Tue, 18 May 2004 15:07:53 +0500 MIME-Version: 1.0 Content-Type: multipart/signed; boundary="----=_NextPart_000_0086_01C43CE9.E957D2F0"; protocol="application/x-pkcs7-signature"; micalg=SHA1 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 X-Scanned-By: MIMEDefang 2.39 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multi-part message in MIME format. ------=_NextPart_000_0086_01C43CE9.E957D2F0 Content-Type: multipart/mixed; boundary="----=_NextPart_001_0087_01C43CE9.E957D2F0" ------=_NextPart_001_0087_01C43CE9.E957D2F0 Content-Type: multipart/alternative; boundary="----=_NextPart_002_0088_01C43CE9.E957D2F0" ------=_NextPart_002_0088_01C43CE9.E957D2F0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MessageHello=20 I have attached the TBSData.req and signed requests after siging with = CAPI and another toolkit. The file toolkitSignedRequest.req is a signed = request file that was signed with a toolkit and OCSP responders accept = the request. Wheras capiDirRequest.req is a file in which signatures = have been calculated from CAPI and attached to request. But OCSP servers = are unable to verify the signatures of the request. The last file = capipkcsRequest.req is a request in which we made a PKCS#7 object of TBS = Data using CAPI and obtained the signatures from PKCS#7 and attached to = request. We are unable to contrust a correct signed request using CAPI. The = signatures in case of CAPI are different but we dont know why they are = different? Any help will be highly appreciated. Regards=20 ----- Original Message -----=20 From: Miguel Rodriguez=20 To: 'Yasir Khan' ; ietf-pkix@imc.org=20 Sent: Tuesday, May 18, 2004 5:41 AM Subject: RE: Signature in OCSP Request You may need to reverse the signature so it can be verified by the = OCSP server. The signature should be computed over the tbsRequest = structure. Regarding the algorithms RFC 2560 is nto specific when it = comes to signed requests, we need further clarification of this point. Miguel A Rodriguez SeguriData Mexico -----Original Message----- From: owner-ietf-pkix@mail.imc.org = [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Yasir Khan Sent: Monday, May 17, 2004 9:01 AM To: ietf-pkix@imc.org Cc: a.alam@ascertia.com Subject: Signature in OCSP Request Hello I have a question about the signature field in OCSP request. What = exactly should it contain and what=20 algorithms can be used for OCSP request signing?=20 When i create a signature using MS CAPI, OCSP reponders are unable = to verify the signatures whereas we can verify it using CAPI. What data do we set in signature field = exactly? Regards, Yasir ------=_NextPart_002_0088_01C43CE9.E957D2F0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD><TITLE>Message</TITLE> <META http-equiv=3DContent-Type content=3D"text/html; = charset=3Diso-8859-1"> <META content=3D"MSHTML 6.00.2600.0" name=3DGENERATOR> <STYLE></STYLE> </HEAD> <BODY bgColor=3D#ffffff> <DIV><FONT face=3DArial color=3D#800080 size=3D2>Hello </FONT></DIV> <DIV><FONT face=3DArial color=3D#800080 size=3D2></FONT> </DIV> <DIV><FONT face=3DArial color=3D#800080 size=3D2> I have attached = the TBSData.req=20 and signed requests after siging with CAPI and another toolkit. The file = toolkitSignedRequest.req is a signed request file that was signed with a = toolkit=20 and OCSP responders accept the request. Wheras = capiDirRequest.req is a file=20 in which signatures have been calculated from CAPI and attached to = request. But=20 OCSP servers are unable to verify the signatures of the request. The = last file=20 capipkcsRequest.req is a request in which we made a PKCS#7 object of TBS = Data=20 using CAPI and obtained the signatures from PKCS#7 and attached to=20 request.</FONT></DIV> <DIV><FONT face=3DArial color=3D#800080 size=3D2></FONT> </DIV> <DIV><FONT face=3DArial color=3D#800080 size=3D2>We are unable to = contrust=20 a correct signed request using CAPI. The signatures in case of CAPI = are=20 different but we dont know why they are = different?</FONT></DIV> <DIV> </DIV> <DIV><FONT face=3DArial color=3D#800080 size=3D2>Any help will be highly = appreciated.</FONT></DIV> <DIV><FONT face=3DArial color=3D#800080 size=3D2></FONT> </DIV> <DIV><FONT face=3DArial color=3D#800080 = size=3D2>Regards</FONT> </DIV> <BLOCKQUOTE dir=3Dltr=20 style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; = BORDER-LEFT: #800080 2px solid; MARGIN-RIGHT: 0px"> <DIV style=3D"FONT: 10pt arial">----- Original Message ----- </DIV> <DIV=20 style=3D"BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: = black"><B>From:</B>=20 <A title=3Dmars@seguridata.com = href=3D"mailto:mars@seguridata.com">Miguel=20 Rodriguez</A> </DIV> <DIV style=3D"FONT: 10pt arial"><B>To:</B> <A = title=3Dyasir.khan@ascertia.com=20 href=3D"mailto:yasir.khan@ascertia.com">'Yasir Khan'</A> ; <A=20 title=3Dietf-pkix@imc.org = href=3D"mailto:ietf-pkix@imc.org">ietf-pkix@imc.org</A>=20 </DIV> <DIV style=3D"FONT: 10pt arial"><B>Sent:</B> Tuesday, May 18, 2004 = 5:41 AM</DIV> <DIV style=3D"FONT: 10pt arial"><B>Subject:</B> RE: Signature in OCSP=20 Request</DIV> <DIV><BR></DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial = color=3D#0000ff size=3D2>You=20 may need to reverse the signature so it can be verified by the OCSP = server.=20 The signature should be computed over the tbsRequest structure. = Regarding the=20 algorithms RFC 2560 is nto specific when it comes to signed requests, = we need=20 further clarification of this point.</FONT></SPAN></DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial = color=3D#0000ff=20 size=3D2></FONT></SPAN> </DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial = color=3D#0000ff=20 size=3D2>Miguel A Rodriguez</FONT></SPAN></DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial = color=3D#0000ff=20 size=3D2>SeguriData</FONT></SPAN></DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial = color=3D#0000ff=20 size=3D2>Mexico</FONT></SPAN></DIV> <BLOCKQUOTE dir=3Dltr=20 style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px = solid; MARGIN-RIGHT: 0px"> <DIV></DIV> <DIV class=3DOutlookMessageHeader lang=3Den-us dir=3Dltr = align=3Dleft><FONT=20 face=3DTahoma size=3D2>-----Original Message-----<BR><B>From:</B> <A = = href=3D"mailto:owner-ietf-pkix@mail.imc.org">owner-ietf-pkix@mail.imc.org= </A>=20 [mailto:owner-ietf-pkix@mail.imc.org] <B>On Behalf Of </B>Yasir=20 Khan<BR><B>Sent:</B> Monday, May 17, 2004 9:01 AM<BR><B>To:</B>=20 ietf-pkix@imc.org<BR><B>Cc:</B> = a.alam@ascertia.com<BR><B>Subject:</B>=20 Signature in OCSP Request<BR><BR></FONT></DIV><FONT face=3DArial = color=3D#800080=20 size=3D2> <DIV><BR>Hello</DIV> <DIV> </DIV> <DIV>I have a question about the signature field in OCSP request. = What=20 exactly should it contain and what <BR>algorithms can be used for = OCSP=20 request signing? </DIV> <DIV> </DIV> <DIV>When i create a signature using MS CAPI, OCSP reponders are = unable to=20 verify the signatures whereas we<BR>can verify it using CAPI. What = data do=20 we set in signature field exactly?</DIV> <DIV> </DIV> <DIV>Regards,</DIV> <DIV> </DIV> <DIV>Yasir</DIV> <DIV> </DIV> <DIV></FONT> </DIV></BLOCKQUOTE></BLOCKQUOTE></BODY></HTML> ------=_NextPart_002_0088_01C43CE9.E957D2F0-- ------=_NextPart_001_0087_01C43CE9.E957D2F0 Content-Type: application/octet-stream; name="capiDirRequest.req" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="capiDirRequest.req" MIIF3zCCAVGgAwIBAKFWpFQwUjELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQ8wDQYD VQQLEwZQZW9wbGUxHzAdBgNVBAMTFlRlc3QgQWxpY2UgVGVzdCBMMiBDQTEwgb4wgbswOjAJBgUr DgMCGgUABBTLMw+MCkY9+joPkAiahqdX8D05sQQUzA35sCUgCTmqS33NavORCcXf/MACAQWgfTB7 MHkGCSsGAQUFBzABBwRsMGowRDELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQwwCgYD VQQLEwNTUUExFDASBgNVBAMTC1Rlc3QgTDIgQ0ExMCIwIAYIKwYBBQUHMAGGFGh0dHA6Ly9hc2Nl cnRpYS01Ojg1ojEwLzAaBgkrBgEFBQcwAQQEDTALBgkrBgEFBQcwAQEwEQYJKwYBBQUHMAECBATU e8EyoIIEhjCCBIIwDQYJKoZIhvcNAQEFBQADgYEAiYktIgCLJveIHuCtMWfIMMQ6S9cCGF4hxKyG Enz/eaQX2pY+IatUbA/YXB+2ch+SeP5uMSzjSqnFtBWLua8fUrIwXqvjYBxQprvWrLYP9KJ42Rnb yuN8wbFmtV/xsXNlB82+JbWncNM77lH0tPJfsC4P6o4f10DNeR0bAo2VbjugggPrMIID5zCCA+Mw ggNMoAMCAQICAQUwDQYJKoZIhvcNAQEFBQAwRDELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2Vy dGlhMQwwCgYDVQQLEwNTUUExFDASBgNVBAMTC1Rlc3QgTDIgQ0ExMB4XDTAzMDYzMDExNDQ0N1oX DTE1MDQzMDExNDExN1owUjELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQ8wDQYDVQQL EwZQZW9wbGUxHzAdBgNVBAMTFlRlc3QgQWxpY2UgVGVzdCBMMiBDQTEwgZ8wDQYJKoZIhvcNAQEB BQADgY0AMIGJAoGBAPZaou86n5IMyaoG+8r8GQgX5TTv7lxbCqxpo2hgE7kxbEm4FEohbFhl0IFJ qkBkcByYCDGgnFCYOyo5upnyCwfrVAu06k8UqmJj/117Uwy/PmGfOeGzeoJqvI9kORPeGZ6yIsiy CcjZqAp4/XjP4b5OtgvRN3+8NeNHWtUGUl3XAgMBAAGjggHVMIIB0TAOBgNVHQ8BAf8EBAMCA/gw DAYDVR0TAQH/BAIwADCB0QYDVR0gBIHJMIHGMIHDBgorBgEEAfxJAQICMIG0MIGxBggrBgEFBQcC AjCBpBqBoUNlcnRpZmljYXRlcyBpc3N1ZWQgdW5kZXIgdGhpcyBwb2xpY3kgYXJlIHRvIHVuYXV0 aGVudGljYXRlZCBpbmRpdmlkdWFscy4gIFBsZWFzZSBkbyBub3QgdXNlIHN1Y2ggY2VydGlmaWNh dGVzIGZvciB2YWx1YWJsZSB0cmFuc2FjdGlvbnMuICBOTyBMSUFCSUxJVFkgQUNDRVBURUQuMB0G A1UdDgQWBBSmXKPcjzYc56fGGrjauEGTY8TBdDBLBgNVHR8ERDBCMECgPqA8hjpsZGFwOi8vYXNj ZXJ0aWEtNC9jYWNlcnRpZmljYXRlPW5ldywgb3U9dGVzdCwgbz1haXJpdXMuY29tMDAGCCsGAQUF BwEBBCQwIjAgBggrBgEFBQcwAYYUaHR0cDovL2FzY2VydGlhLTU6ODUwHgYDVR0RBBcwFYETYWxp Y2VMMkB0ZXN0aW5nLmNvbTAfBgNVHSMEGDAWgBTMDfmwJSAJOapLfc1q85EJxd/8wDANBgkqhkiG 9w0BAQUFAAOBgQCJ+diomsCnaFbr6f7oR6RRc3qGV18bvEGsAwPNmjFFv058IxmtkL/uX1tSJJOZ 6lI1qQAqUNYzrg/g1MBWZ/wxNXMrDFYNdB6W0sgWJcvnzYYwB4oziNx/fKNjBNuwbOcZXUatVUDY c+8dn3jkLGRuOepReeylW4JRDifSTJRZIA== ------=_NextPart_001_0087_01C43CE9.E957D2F0 Content-Type: application/octet-stream; name="capipkcsRequest.req" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="capipkcsRequest.req" MIIF3zCCAVGgAwIBAKFWpFQwUjELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQ8wDQYD VQQLEwZQZW9wbGUxHzAdBgNVBAMTFlRlc3QgQWxpY2UgVGVzdCBMMiBDQTEwgb4wgbswOjAJBgUr DgMCGgUABBTLMw+MCkY9+joPkAiahqdX8D05sQQUzA35sCUgCTmqS33NavORCcXf/MACAQWgfTB7 MHkGCSsGAQUFBzABBwRsMGowRDELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQwwCgYD VQQLEwNTUUExFDASBgNVBAMTC1Rlc3QgTDIgQ0ExMCIwIAYIKwYBBQUHMAGGFGh0dHA6Ly9hc2Nl cnRpYS01Ojg1ojEwLzAaBgkrBgEFBQcwAQQEDTALBgkrBgEFBQcwAQEwEQYJKwYBBQUHMAECBATU e8EyoIIEhjCCBIIwDQYJKoZIhvcNAQEFBQADgYEAZMo3KxKFhAjzFWor+wfP0P2TL7RFhGlWTiZP n6Umq6MBd9Zf4+tTiQ6etxEGpY2g4H6umma/5l1DBb+pd4hqeSyH5sb1ZzkHy8jCB7KuEVipxEZh LcEDJZXkKGQxdce/gwarX5QGzh5tPXnyEmi6o68ZDtaPVpXYgCKprdaUGs6gggPrMIID5zCCA+Mw ggNMoAMCAQICAQUwDQYJKoZIhvcNAQEFBQAwRDELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2Vy dGlhMQwwCgYDVQQLEwNTUUExFDASBgNVBAMTC1Rlc3QgTDIgQ0ExMB4XDTAzMDYzMDExNDQ0N1oX DTE1MDQzMDExNDExN1owUjELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQ8wDQYDVQQL EwZQZW9wbGUxHzAdBgNVBAMTFlRlc3QgQWxpY2UgVGVzdCBMMiBDQTEwgZ8wDQYJKoZIhvcNAQEB BQADgY0AMIGJAoGBAPZaou86n5IMyaoG+8r8GQgX5TTv7lxbCqxpo2hgE7kxbEm4FEohbFhl0IFJ qkBkcByYCDGgnFCYOyo5upnyCwfrVAu06k8UqmJj/117Uwy/PmGfOeGzeoJqvI9kORPeGZ6yIsiy CcjZqAp4/XjP4b5OtgvRN3+8NeNHWtUGUl3XAgMBAAGjggHVMIIB0TAOBgNVHQ8BAf8EBAMCA/gw DAYDVR0TAQH/BAIwADCB0QYDVR0gBIHJMIHGMIHDBgorBgEEAfxJAQICMIG0MIGxBggrBgEFBQcC AjCBpBqBoUNlcnRpZmljYXRlcyBpc3N1ZWQgdW5kZXIgdGhpcyBwb2xpY3kgYXJlIHRvIHVuYXV0 aGVudGljYXRlZCBpbmRpdmlkdWFscy4gIFBsZWFzZSBkbyBub3QgdXNlIHN1Y2ggY2VydGlmaWNh dGVzIGZvciB2YWx1YWJsZSB0cmFuc2FjdGlvbnMuICBOTyBMSUFCSUxJVFkgQUNDRVBURUQuMB0G A1UdDgQWBBSmXKPcjzYc56fGGrjauEGTY8TBdDBLBgNVHR8ERDBCMECgPqA8hjpsZGFwOi8vYXNj ZXJ0aWEtNC9jYWNlcnRpZmljYXRlPW5ldywgb3U9dGVzdCwgbz1haXJpdXMuY29tMDAGCCsGAQUF BwEBBCQwIjAgBggrBgEFBQcwAYYUaHR0cDovL2FzY2VydGlhLTU6ODUwHgYDVR0RBBcwFYETYWxp Y2VMMkB0ZXN0aW5nLmNvbTAfBgNVHSMEGDAWgBTMDfmwJSAJOapLfc1q85EJxd/8wDANBgkqhkiG 9w0BAQUFAAOBgQCJ+diomsCnaFbr6f7oR6RRc3qGV18bvEGsAwPNmjFFv058IxmtkL/uX1tSJJOZ 6lI1qQAqUNYzrg/g1MBWZ/wxNXMrDFYNdB6W0sgWJcvnzYYwB4oziNx/fKNjBNuwbOcZXUatVUDY c+8dn3jkLGRuOepReeylW4JRDifSTJRZIA== ------=_NextPart_001_0087_01C43CE9.E957D2F0 Content-Type: application/octet-stream; name="TBSData.req" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="TBSData.req" MIIBUaADAgEAoVakVDBSMQswCQYDVQQGEwJHQjERMA8GA1UEChMIQXNjZXJ0aWExDzANBgNVBAsT BlBlb3BsZTEfMB0GA1UEAxMWVGVzdCBBbGljZSBUZXN0IEwyIENBMTCBvjCBuzA6MAkGBSsOAwIa BQAEFMszD4wKRj36Og+QCJqGp1fwPTmxBBTMDfmwJSAJOapLfc1q85EJxd/8wAIBBaB9MHsweQYJ KwYBBQUHMAEHBGwwajBEMQswCQYDVQQGEwJHQjERMA8GA1UEChMIQXNjZXJ0aWExDDAKBgNVBAsT A1NRQTEUMBIGA1UEAxMLVGVzdCBMMiBDQTEwIjAgBggrBgEFBQcwAYYUaHR0cDovL2FzY2VydGlh LTU6ODWiMTAvMBoGCSsGAQUFBzABBAQNMAsGCSsGAQUFBzABATARBgkrBgEFBQcwAQIEBNR7wTI= ------=_NextPart_001_0087_01C43CE9.E957D2F0 Content-Type: application/octet-stream; name="toolKitSignedRequest.req" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="toolKitSignedRequest.req" MIIF3zCCAVGgAwIBAKFWpFQwUjELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQ8wDQYD VQQLEwZQZW9wbGUxHzAdBgNVBAMTFlRlc3QgQWxpY2UgVGVzdCBMMiBDQTEwgb4wgbswOjAJBgUr DgMCGgUABBTLMw+MCkY9+joPkAiahqdX8D05sQQUzA35sCUgCTmqS33NavORCcXf/MACAQWgfTB7 MHkGCSsGAQUFBzABBwRsMGowRDELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQwwCgYD VQQLEwNTUUExFDASBgNVBAMTC1Rlc3QgTDIgQ0ExMCIwIAYIKwYBBQUHMAGGFGh0dHA6Ly9hc2Nl cnRpYS01Ojg1ojEwLzAaBgkrBgEFBQcwAQQEDTALBgkrBgEFBQcwAQEwEQYJKwYBBQUHMAECBATU e8EyoIIEhjCCBIIwDQYJKoZIhvcNAQEFBQADgYEAO26VjQIbHXnNQNcfjuoPLrBf8rT0Ue4703Cn tSW+zQdlc7HxX7VmscF848rbGdl4ovQPtqzWu6ZQHGDjq14wslIfr7mLFbTFqUrjLDFu/niSH3K2 H1zYD2xUqyE+ltoXpHn/fBKGrMQhXhgC10s6xDDIZzGt4B6I9yaLACItiYmgggPrMIID5zCCA+Mw ggNMoAMCAQICAQUwDQYJKoZIhvcNAQEFBQAwRDELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2Vy dGlhMQwwCgYDVQQLEwNTUUExFDASBgNVBAMTC1Rlc3QgTDIgQ0ExMB4XDTAzMDYzMDExNDQ0N1oX DTE1MDQzMDExNDExN1owUjELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMQ8wDQYDVQQL EwZQZW9wbGUxHzAdBgNVBAMTFlRlc3QgQWxpY2UgVGVzdCBMMiBDQTEwgZ8wDQYJKoZIhvcNAQEB BQADgY0AMIGJAoGBAPZaou86n5IMyaoG+8r8GQgX5TTv7lxbCqxpo2hgE7kxbEm4FEohbFhl0IFJ qkBkcByYCDGgnFCYOyo5upnyCwfrVAu06k8UqmJj/117Uwy/PmGfOeGzeoJqvI9kORPeGZ6yIsiy CcjZqAp4/XjP4b5OtgvRN3+8NeNHWtUGUl3XAgMBAAGjggHVMIIB0TAOBgNVHQ8BAf8EBAMCA/gw DAYDVR0TAQH/BAIwADCB0QYDVR0gBIHJMIHGMIHDBgorBgEEAfxJAQICMIG0MIGxBggrBgEFBQcC AjCBpBqBoUNlcnRpZmljYXRlcyBpc3N1ZWQgdW5kZXIgdGhpcyBwb2xpY3kgYXJlIHRvIHVuYXV0 aGVudGljYXRlZCBpbmRpdmlkdWFscy4gIFBsZWFzZSBkbyBub3QgdXNlIHN1Y2ggY2VydGlmaWNh dGVzIGZvciB2YWx1YWJsZSB0cmFuc2FjdGlvbnMuICBOTyBMSUFCSUxJVFkgQUNDRVBURUQuMB0G A1UdDgQWBBSmXKPcjzYc56fGGrjauEGTY8TBdDBLBgNVHR8ERDBCMECgPqA8hjpsZGFwOi8vYXNj ZXJ0aWEtNC9jYWNlcnRpZmljYXRlPW5ldywgb3U9dGVzdCwgbz1haXJpdXMuY29tMDAGCCsGAQUF BwEBBCQwIjAgBggrBgEFBQcwAYYUaHR0cDovL2FzY2VydGlhLTU6ODUwHgYDVR0RBBcwFYETYWxp Y2VMMkB0ZXN0aW5nLmNvbTAfBgNVHSMEGDAWgBTMDfmwJSAJOapLfc1q85EJxd/8wDANBgkqhkiG 9w0BAQUFAAOBgQCJ+diomsCnaFbr6f7oR6RRc3qGV18bvEGsAwPNmjFFv058IxmtkL/uX1tSJJOZ 6lI1qQAqUNYzrg/g1MBWZ/wxNXMrDFYNdB6W0sgWJcvnzYYwB4oziNx/fKNjBNuwbOcZXUatVUDY c+8dn3jkLGRuOepReeylW4JRDifSTJRZIA== ------=_NextPart_001_0087_01C43CE9.E957D2F0-- ------=_NextPart_000_0086_01C43CE9.E957D2F0 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIK3TCCAzEw ggIZoAMCAQICAQIwDQYJKoZIhvcNAQEFBQAwOzELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2Vy dGlhMRkwFwYDVQQDExBBc2NlcnRpYSBSb290IENBMB4XDTAzMDMwNDA4MTI1N1oXDTEyMDczMDEw NDIyOVowYDELMAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMSYwJAYDVQQLEx1DbGFzcyAy IENlcnRpZmljYXRlIEF1dGhvcml0eTEWMBQGA1UEAxMNQXNjZXJ0aWEgQ0EgMjCBnzANBgkqhkiG 9w0BAQEFAAOBjQAwgYkCgYEAnRehtg8oWu+jYIkNJVR1ueHs7k8fClUiqUsrjmhuaI6SGjw0eusF FCCnDN2URjk+Un2OFHa3fn0lRFes/rIXvV0aB8pZGp8XJLO6u3pdfKJGnVeBtgBUr/YkXT/APo+z pp6a52+VjOEA8tcsfkco+Ml4quvZRSQ6/5hvDlEnm08CAwEAAaOBnjCBmzAOBgNVHQ8BAf8EBAMC AQYwEgYDVR0TAQH/BAgwBgEB/wIBADAdBgNVHQ4EFgQUc/Pyi9HYduL1F1K9IjZs+KkJi3QwNQYI KwYBBQUHAQEEKTAnMCUGCCsGAQUFBzABhhl3d3cuZ2xvYmFsdHJ1c3RmaW5kZXIuY29tMB8GA1Ud IwQYMBaAFLFTcaAo/ecMU5odaxA87sgazV7OMA0GCSqGSIb3DQEBBQUAA4IBAQAWgwuPN1Kt4P+g mBe25iHuepjjWcslDjZaC65ZxQuGjr/Aj7eKtBwpvw+01S4r9QIKQeT0DqlcFJlf3mDa6S25ZKxR hUM74fxSYvfT1hAFd7NMZ2BYSj5bGHX5LWFQi1REDthiggjUt5QUGj+OVDfqBakynkiYMinw+NV0 gdRzMyhE3j3nWzeXrOXOmzea3o4bsLK2yCAJ6v+VjjTs+gBlCIE2aqSdctX2JOFJLQUyWlArBqZ9 CEdVW+iYS7PYRnOddXLiX3EC/7+MlbNMZSW234XzWQxrNqF8JzUXbYHm3jDMsprDdeC8RtiYAhQa T/Ogeq21ndR7phpmrMQqYCSBMIIDajCCAlKgAwIBAgIBATANBgkqhkiG9w0BAQUFADA7MQswCQYD VQQGEwJHQjERMA8GA1UEChMIQXNjZXJ0aWExGTAXBgNVBAMTEEFzY2VydGlhIFJvb3QgQ0EwHhcN MDMwMzA0MDgwNjEyWhcNMTMwMzA0MDgwMTI3WjA7MQswCQYDVQQGEwJHQjERMA8GA1UEChMIQXNj ZXJ0aWExGTAXBgNVBAMTEEFzY2VydGlhIFJvb3QgQ0EwggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAw ggEKAoIBAQDR+NPebia/iGElNG3QI9TlQSWE+vwhh9N1p608htxwX3HG2PREDou8hP/r3pIkkJEs flGxt3RVXK+Ut7IyKKB17rhlUSGmrqaRL2fAxyCsCbtak6uLqh7afDYesI1ozLn4sYMFeFWGCWKk kNBzGfDvKebjXAz9yNxhFyLGbB+ZUkEsfjQA3GJ4Dza4FAbWUH/Q9jWpd8RU9Wx9Hi+QTfXOTaaW Tn+QMRg9QWuP6pklSsGA65j9YWoBd+ONtSEUM0aX3p/iuecSqC/IjfSxa7hWCskQR+bzqg3xMJTa R1JpCQds6Z1OqJX5R3UEh71FaxC2TTMgGhNDhwVuoxEgSPhlAgMBAAGjeTB3MA4GA1UdDwEB/wQE AwIBBjAPBgNVHRMBAf8EBTADAQH/MB0GA1UdDgQWBBSxU3GgKP3nDFOaHWsQPO7IGs1ezjA1Bggr BgEFBQcBAQQpMCcwJQYIKwYBBQUHMAGGGXd3dy5nbG9iYWx0cnVzdGZpbmRlci5jb20wDQYJKoZI hvcNAQEFBQADggEBAKRrsf6/A8LzMqEFJcKl3pq41fPXH7gDLbayGYxvRJ15LRMwxy6A8A2SrY5r V8u8PStKcMGKj/1R4q1zY/h2TH0eIHDNzIcXiAndZFNBIRPskQ1qIKIlTtO/TYC1f+qss9r7eKcw Yk90WhdkdZ0k/r21Z7JQMtLGvgO+2Mc4LEb4f1Li/TdB3M0dBq0nRrDX5wiM3hoRbYkWn+0rNtHl 4fVqp5M0S9BeWud/jNIiPu2OSFCdiXDgYVYP56OfVYHuUQuTGfsRP8EI3uUE5RFqIPBj+Vx/Dwzt /LnNR2CfKv6HPS5AstKZWr04k/RKserIBmJLuxohgfnUuLWJovLFPJkwggQ2MIIDn6ADAgECAgE0 MA0GCSqGSIb3DQEBBQUAMGAxCzAJBgNVBAYTAkdCMREwDwYDVQQKEwhBc2NlcnRpYTEmMCQGA1UE CxMdQ2xhc3MgMiBDZXJ0aWZpY2F0ZSBBdXRob3JpdHkxFjAUBgNVBAMTDUFzY2VydGlhIENBIDIw HhcNMDQwMzA4MDc0NjM5WhcNMDUwMzA4MDcwMDAxWjBLMQswCQYDVQQGEwJHQjERMA8GA1UEChMI QXNjZXJ0aWExFDASBgNVBAsTC0RldmVsb3BtZW50MRMwEQYDVQQDEwpZYXNpciBLaGFuMIGfMA0G CSqGSIb3DQEBAQUAA4GNADCBiQKBgQDQw4w9nWiaKNYb6jLXsC/oWxIqhrnyG/pb3CdYhimtpH8k 9LcuXf0nxt2DNPcKN0Trsj+va5o8My3tlAt+8OAXMLKFMX1VBcUQew+cJsvdrxLeX4lrBrZnmUGR WL0xr/TDvTSUIwERqCqYpCHWrTkGnOkbMctJdYok0ebEhd3npQIDAQABo4ICEzCCAg8wDgYDVR0P AQH/BAQDAgP4MAwGA1UdEwEB/wQCMAAwggEDBgNVHSAEgfswgfgwgfUGCisGAQQB/EkBAgEwgeYw geMGCCsGAQUFBwICMIHWGoHTVGhpcyBjZXJ0aWZpY2F0ZSBpcyBmb3IgdGhlIHNvbGUgdXNlIG9m IEFzY2VydGlhLCBhbmQgaXRzIGN1c3RvbWVycy4gQXNjZXJ0aWEgYWNjZXB0cyBubyBsaWFiaWxp dHkgZm9yIGFueSBjbGFpbSBleGNlcHQgYXMgZXhwcmVzc2x5IHByb3ZpZGVkIGluIGl0cyBDZXJ0 aWZpY2F0ZSBQb2xpY3ksIHdoaWNoIGlzIGF2YWlsYWJsZSBmcm9tIGluZm9AYXNjZXJ0aWEuY29t LjAdBgNVHQ4EFgQUOwYmW7OQ2HAHAeNeR/ltL6CKbgcwTQYDVR0fBEYwRDBCoECgPoY8aHR0cDov L3d3dy5hc2NlcnRpYS5jb20vT25saW5lQ0EvY3Jscy9Bc2NlcnRpYUNBMi9jbGFzczIuY3JsMDUG CCsGAQUFBwEBBCkwJzAlBggrBgEFBQcwAYYZd3d3Lmdsb2JhbHRydXN0ZmluZGVyLmNvbTAiBgNV HREEGzAZgRd5YXNpci5raGFuQGFzY2VydGlhLmNvbTAfBgNVHSMEGDAWgBRz8/KL0dh24vUXUr0i Nmz4qQmLdDANBgkqhkiG9w0BAQUFAAOBgQBHDbq9/Y4utItSYbDeuMmzYM6tckb2Kyk3wU7mQfIJ PRCzNDCcG6fsmx0BvmZg/jclxGc12P64LDRDPH2Ly47L+c/DeBvVzFz6+37tGSoNs3c+2t8eH1l3 sYvbJ4E2QeZaeGs1mISgOiEGuCn9zvwXpNzNovcHZ6oxetVNQVE2IjGCAcgwggHEAgEBMGUwYDEL MAkGA1UEBhMCR0IxETAPBgNVBAoTCEFzY2VydGlhMSYwJAYDVQQLEx1DbGFzcyAyIENlcnRpZmlj YXRlIEF1dGhvcml0eTEWMBQGA1UEAxMNQXNjZXJ0aWEgQ0EgMgIBNDAJBgUrDgMCGgUAoIG6MBgG CSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcNAQkFMQ8XDTA0MDUxODEwMDc1M1owIwYJ KoZIhvcNAQkEMRYEFBXmSQAM68vBgi3i9yZm9UkxZZjTMFsGCSqGSIb3DQEJDzFOMEwwCgYIKoZI hvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcGBSsOAwIHMA0GCCqGSIb3DQMC AgEoMAcGBSsOAwIdMA0GCSqGSIb3DQEBAQUABIGAUIIGB0xZKvZLJR9H3/qV7f7Sgnq3vheGOSVJ iGdpuC7R0CPIB0VbW+yn630RGZbcKgcXGqLatut0lrLnEY0LhXjaP7Wf1eWbRYnQ12AfajzbKC04 sgCKIvC3iXILZzFIBoo+HbPr/OKTnlZDhEj41AQg+Or71ssYS2xvObZzXFIAAAAAAAA= ------=_NextPart_000_0086_01C43CE9.E957D2F0-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4I0ddxU015269; Mon, 17 May 2004 17:39:39 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4I0ddJh015268; Mon, 17 May 2004 17:39:39 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from seguridata1.seguridata.com ([200.57.34.107]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4I0dbYF015251 for <ietf-pkix@imc.org>; Mon, 17 May 2004 17:39:38 -0700 (PDT) (envelope-from mars@seguridata.com) Received: from MarsXP ([200.67.231.235]) by seguridata1.seguridata.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 17 May 2004 19:40:09 -0500 From: "Miguel Rodriguez" <mars@seguridata.com> To: "'Yasir Khan'" <yasir.khan@ascertia.com>, <ietf-pkix@imc.org> Subject: RE: Signature in OCSP Request Date: Mon, 17 May 2004 19:41:14 -0500 Message-ID: <002601c43c70$ddb9edd0$a600a8c0@seguridata.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0027_01C43C46.F4E3E5D0" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.2627 In-Reply-To: <011d01c43c17$5eaee0d0$b500a8c0@ascertia.com.pk> X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Importance: Normal X-OriginalArrivalTime: 18 May 2004 00:40:10.0906 (UTC) FILETIME=[ACF68FA0:01C43C70] Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multi-part message in MIME format. ------=_NextPart_000_0027_01C43C46.F4E3E5D0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit You may need to reverse the signature so it can be verified by the OCSP server. The signature should be computed over the tbsRequest structure. Regarding the algorithms RFC 2560 is nto specific when it comes to signed requests, we need further clarification of this point. Miguel A Rodriguez SeguriData Mexico -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Yasir Khan Sent: Monday, May 17, 2004 9:01 AM To: ietf-pkix@imc.org Cc: a.alam@ascertia.com Subject: Signature in OCSP Request Hello I have a question about the signature field in OCSP request. What exactly should it contain and what algorithms can be used for OCSP request signing? When i create a signature using MS CAPI, OCSP reponders are unable to verify the signatures whereas we can verify it using CAPI. What data do we set in signature field exactly? Regards, Yasir ------=_NextPart_000_0027_01C43C46.F4E3E5D0 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = charset=3Dus-ascii"> <TITLE>Message</TITLE> <META content=3D"MSHTML 6.00.2800.1400" name=3DGENERATOR> <STYLE></STYLE> </HEAD> <BODY bgColor=3D#ffffff> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial color=3D#0000ff = size=3D2>You=20 may need to reverse the signature so it can be verified by the OCSP = server. The=20 signature should be computed over the tbsRequest structure. Regarding = the=20 algorithms RFC 2560 is nto specific when it comes to signed requests, we = need=20 further clarification of this point.</FONT></SPAN></DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial color=3D#0000ff = size=3D2></FONT></SPAN> </DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial color=3D#0000ff = size=3D2>Miguel=20 A Rodriguez</FONT></SPAN></DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial color=3D#0000ff = size=3D2>SeguriData</FONT></SPAN></DIV> <DIV><SPAN class=3D610453200-18052004><FONT face=3DArial color=3D#0000ff = size=3D2>Mexico</FONT></SPAN></DIV> <BLOCKQUOTE dir=3Dltr=20 style=3D"PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #0000ff 2px = solid; MARGIN-RIGHT: 0px"> <DIV></DIV> <DIV class=3DOutlookMessageHeader lang=3Den-us dir=3Dltr = align=3Dleft><FONT=20 face=3DTahoma size=3D2>-----Original Message-----<BR><B>From:</B>=20 owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] = <B>On=20 Behalf Of </B>Yasir Khan<BR><B>Sent:</B> Monday, May 17, 2004 9:01=20 AM<BR><B>To:</B> ietf-pkix@imc.org<BR><B>Cc:</B>=20 a.alam@ascertia.com<BR><B>Subject:</B> Signature in OCSP=20 Request<BR><BR></FONT></DIV><FONT face=3DArial color=3D#800080 = size=3D2> <DIV><BR>Hello</DIV> <DIV> </DIV> <DIV>I have a question about the signature field in OCSP request. What = exactly=20 should it contain and what <BR>algorithms can be used for OCSP request = signing? </DIV> <DIV> </DIV> <DIV>When i create a signature using MS CAPI, OCSP reponders are = unable to=20 verify the signatures whereas we<BR>can verify it using CAPI. What = data do we=20 set in signature field exactly?</DIV> <DIV> </DIV> <DIV>Regards,</DIV> <DIV> </DIV> <DIV>Yasir</DIV> <DIV> </DIV> <DIV></FONT> </DIV></BLOCKQUOTE></BODY></HTML> ------=_NextPart_000_0027_01C43C46.F4E3E5D0-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4HL5Et1003405; Mon, 17 May 2004 14:05:14 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4HL5EoV003404; Mon, 17 May 2004 14:05:14 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from optimus.ietf.org (optimus22.ietf.org [132.151.6.22]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4HL5Ded003397 for <ietf-pkix@imc.org>; Mon, 17 May 2004 14:05:13 -0700 (PDT) (envelope-from nobody@optimus.ietf.org) Received: from nobody by optimus.ietf.org with local (Exim 4.20) id 1BPoGo-0003IW-V4; Mon, 17 May 2004 15:59:30 -0400 X-test-idtracker: no From: The IESG <iesg-secretary@ietf.org> To: IETF-Announce:; Cc: Internet Architecture Board <iab@iab.org>, RFC Editor <rfc-editor@rfc-editor.org>, pkix mailing list <ietf-pkix@imc.org>, pkix chair <kent@bbn.com>, pkix chair <wpolk@nist.gov> Subject: Document Action: 'A 224-bit One-way Hash Function: SHA-224' to Informational RFC Message-Id: <E1BPoGo-0003IW-V4@optimus.ietf.org> Date: Mon, 17 May 2004 15:59:30 -0400 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> The IESG has approved the following document: - 'A 224-bit One-way Hash Function: SHA-224 ' <draft-ietf-pkix-sha224-01.txt> as an Informational RFC This document is the product of the Public-Key Infrastructure (X.509) Working Group. The IESG contact persons are Steve Bellovin and Russ Housley. Technical Summary This draft specifies a 224-bit cryptographic hash function. 224 bits is the obvious choice to match triple-DES in strength against brute-force attacks. At least two independent implementations exist and agree on the test vectors in this document. Working Group Summary Since NIST has already specified SHA-224, this was the only reasonable choice. Protocol Quality Steve Bellovin has reviewed this document for the IESG. RFC Editor Note Please make two changes. 1. Please correct typos in the Abstract. OLD: ... A SHA-224 is based on SHA-256, but it uses an different ... NEW: ... SHA-224 is based on SHA-256, but it uses a different ... 2. The Introduction has two paragraphs. Please add an additional paragraph at the end of the Introduction. NEW: This document makes the SHA-224 one-way hash function specification available to the Internet community, and it publishes the object identifiers for use in ASN.1-based protocols. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4HKf9CY001550; Mon, 17 May 2004 13:41:09 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4HKf9tN001549; Mon, 17 May 2004 13:41:09 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail19e.dulles19-verio.com (mail19e.dulles19-verio.com [198.170.241.12]) by above.proper.com (8.12.11/8.12.9) with SMTP id i4HKf7NZ001535 for <ietf-pkix@imc.org>; Mon, 17 May 2004 13:41:08 -0700 (PDT) (envelope-from pmhesse@geminisecurity.com) Received: from www.geminisecurity.com (161.58.96.110) by mail19e.dulles19-verio.com (RS ver 1.0.92vs) with SMTP id 4-0166386469; Mon, 17 May 2004 16:41:09 -0400 (EDT) From: "Peter Hesse" <pmhesse@geminisecurity.com> To: "'PKIX List'" <ietf-pkix@imc.org> Cc: "'Steve Hanna'" <Steve.Hanna@Sun.COM> Subject: RE: Comments on Path Building I-D Date: Mon, 17 May 2004 16:41:10 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook, Build 11.0.5510 In-Reply-To: <20040506165952.GA97982@mail19e.dulles19-verio.com> X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Thread-Index: AcQzrRHuZ0P5y5ZzSma4FwtwW/9M0QAAC6bQ Message-ID: <20040517164109.GA16638@mail19e.dulles19-verio.com> X-Loop-Detect: 1 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Steve, <resending because I exceeded the PKIX list size limit...> Thank you for your in-depth reading of the document and the large number of comments. I will attempt to address the majority of them in this message. I apologize for the time it has taken me to respond. > Overall Comments: > > * My main overall comment is the one I made last July. There > is not consensus on path building techniques in the PKIX > working group or the IETF. Why? Because we don't understand > the problem thoroughly yet. We do understand the problem. The problem is a traditional graph theory problem. The only new part of the problem as it relates to PKI is how to define the penalty function. We have defined heuristics that make sense for PKI. I'm not sure that this is a reason to prevent this document from progressing. This document is based upon the best research we have to date, which includes real-world experience with operational PKIs and users. If at a later time, there is additional research performed, and new conclusions are reached, then the obvious recommendation would be to update the draft. Until that time, I do not think it is useful to withhold this document. > In this document, you advocate depth first search and suggest > that building forward is usually best. I don't think you have > carefully considered a wide variety of other algorithms like > breadth first search, meet in the middle, etc. Can you show > me any solid evidence that shows your algorithms are better > than the alternatives? Yes and no. I can generate a set of PKI structures which will make this algorithm work best, just as you could generate a set of PKI structure which would make another algorithm work best. Until there are useful real-world examples to test against which demonstrates one approach is generally better than another, such evidence would be questionable. The methods described in the draft have been used in real-world PKIs with success. > I doubt that you have any such proof. In fact, it seems clear > to me that depth first search is a poor algorithm for > building cert paths. I say this as a person who has > implemented that algorithm myself and regretted it. The > problem is that if you make a wrong choice, you must > completely explore that part of the PKI before you consider > backing out and trying other options. I have heard stories of > path building taking 35 minutes or more when there was a > valid short path because the builder used DFS and headed down > the wrong path. Certainly these are horror stories, and if the path-building software consistently took 35 minutes to develop a path, it would not be in use today. In the document, we advocate finding a valid path fastest to prevent such a thing from occurring. There may be ways to optimize building so that a valid path is even more likely found first across a wider range of conditions than the approach we described. However, this is a balancing act between complexity for implementers, performance, and functionality. We may be able to design an exceedingly complex (to implement) algorithm that almost always yields best performance and allows for every conceivable scenario, but no one would want to (or be able to) implement it. > We need to do our homework before making any solid > recommendations to implementers. I suppose it's better to > have some advice than none, but this document should have a > prominent warning that these techniques are experimental. In > fact, we might want to give the RFC Experimental status. We (as authors) have influence but no control as to whether the document heads down standards, informational, or experimental tracks. We have requested informational status, but the IESG can decide whatever they wish. I truly feel that this document is Informational, it provides potential implementers with information that they can use when they are creating their own implementation; it provides little in the way of strict recommendations. Note the following snippets of RFC2026: 4.2.1 Experimental The "Experimental" designation typically denotes a specification that is part of some research or development effort. [...] 4.2.2 Informational An "Informational" specification is published for the general information of the Internet community, and does not represent an Internet community consensus or recommendation. [...] > As I've suggested before, I want to see a careful analysis of > different path building strategies. Theoretical analysis will > probably be useful, but I expect it will also be useful to > compare real-world performance with a variety of topologies > (some from real-world deployments, some looking at possible > futures like many cross-certified bridges). We need to try > out a lot of ideas without getting committed to any too > early. The cert path building panel at the PKI R&D Workshop > this year was a good start, but we need to do more in-depth > work. I've started talking to some other researchers about > this. I hope that this documents' authors will join this effort. > Their experience and aid will be valuable. I agree with you and would also like to see additional research and analysis performed in this area. However, X.509 has existed in one form or another since 1988 and there has never been any guidance for implementers on how to build certification paths. Now, 16 years later, we are offering to provide a document with some guidance--a starting point at the least--for implementers. We should not wait any longer for additional research. When new research brings something additional to light, this RFC can be updated or replaced as warranted. > * At several places in the document, you refer to the "authors' > opinions". If this was truly a working group document, it > would reflect the consensus of the working group, not the > opinions of the authors. As I said before, I don't see > working group consensus on this topic. Has anyone other than > me (and maybe the authors) carefully reviewed the current > draft? I haven't seen any comments on it during Last Call. We did receive a significant number of comments from some other folks on earlier drafts, and we have addressed those comments. This document must be based upon the opinions of the authors, because as I mentioned above, one can always create a path which makes a particular algorithm operate well. As the section from RFC2026 above, there is no problem with an informational RFC where the judgment calls in the document are based upon the opinions of the authors. > Since this document does not reflect working group consensus, > I do not think it should go forward as a working group draft. > It should be an individual submission. But I suppose once it > becomes an RFC nobody will ever know whether it was a working > group submission or not. And I really do think it's useful to > have some guidance on cert path building (even if we don't > understand the problem very well). So I'd be OK with having > it be a working group submission if it goes to Experimental > status and a caveat is added at the beginning warning that > this is only the authors' opinion (albeit an educated > opinion) and that further study is under way to determine > which techniques are actually preferred. As I mentioned earlier, based upon RFC2026, I don't think that experimental status is appropriate. As far as a caveat at the beginning of the document, we intended to have already performed that in section 1.2, Purpose: # This document provides information and guidance for certification # path building. There are no requirements or protocol # specifications in this document. This document provides many # options for performing certification path building, as opposed to # one particular way to best perform certification path building. # This document draws upon the authors' experience with existing # complex certification paths to offer insights and recommendations # to developers integrating support for [X.509] digital # certificates into their applications. # # In addition, this document suggests using an effective general # approach to path building that involves a depth first tree # traversal. While the authors believe this approach offers the # balance of simplicity in design with very effective and # infrastructure neutral path building capabilities, the algorithm # is no more than a suggested approach. Other approaches (e.g., # building complete spanning trees of the PKI.) exist and may be # shown to be more effective under certain conditions. # Certification path validation is described in detail in # both [X.509] and [RFC 3280] and is not repeated in this document. > * You should include in this document a section with guidance > to PKI designers about what they can do to make path building > easier. Here are some ideas: > > Use a simple topology (hierarchical with one root), when possible. > Include AIA, SIA, and CRLDP extensions in certificates. Make > certificates (especially CA certificates) and CRLs easily > available by LDAP and HTTP, populating both sides of the > crossCertificatePair attribute. Use name constraints and > policy constraints whenever possible, especially at high > fanout points like bridges. Avoid having more than two high > fanout CAs (at most) between any two points, if possible. I don't disagree with any of these suggestions, and I think that they are appropriate for capture in a document. I'm just not sure that this document is the right place. This is less of a path development issue and more of a PKI architecture issue. Therefore, I recommend that a separate document be created to provide guidance to PKI implementers on how best to construct their PKI so that path development routines have a good chance for success. I would be happy to collaborate with you on such a document. > * I have a *serious* objection to the idea that someone is > going to specially configure path building software for a > particular PKI environment (as I think you imply at the end > of section 2.3, early in section 3.4, and in a few other > places). This is not practical. Most organizations do not > have a PKI wizard on staff (especially a path building > wizard). The path building software must just work, > automatically. If we can't do that, PKI will never be widely > used (at least, not in a non-hierarchical topology). > Fortunately, I see no reason why we need special > configuration. With AIA and SIA and CRLDP and some good > algorithms, we should be able to build paths just fine in > almost any PKI without any configuration. Again, I agree with what you have written here. However, my practical experience in this area is a reason for this implication being made. With no guidance, a path building implementer might be just concerned with getting their software to work with the particular PKI they have been faced with in testing. This occurred a few times in early revs of the DoD PKI and government-funded software. Our goal as stated in the purpose section is an approach which "offers the balance of simplicity in design with very effective and infrastructure neutral path building capabilities." Certainly a design can be simplified if parts of the implementation are left out, because they will never be used in the target infrastructure. The end of our section 2.1 lists the reasons why users might want to do this. Additionally, as I mentioned above, one can always come up with an algorithm that works best in their specific environment. While it may not be practical to expect, if there is enough pressure to ensure that paths are built and validated in a timely fashion, I can imagine it happening. Oh, and two additional notes: CRLDP does not help in path development--just for retrieving data needed for the path. And, AIA does not help path development any better than a populated LDAP directory does. (If a issuer has multiple certificates, either AIA will be underpopulated or you will face the same dilemma in terms of which certificate to use to develop the path.) > Substantive Comments: > > * Section 1.2 (Purpose) says "... this document suggests using ... > depth first tree traversal. ... Other approaches (e.g., > building complete spanning trees of the PKI.) exist and may > be shown to be more effective under certain conditions." > > Building a complete spanning tree of the PKI is not a decent solution. > Please change this to something more reasonable, like breadth > first search. Otherwise, you're just setting up a straw man, > an impractical alternative. We will change this to a breadth first search. Of course as we mention in the document, in certain cases, this might be impractical as well. (i.e. forward cross-certificates not being present) > * In section 1.4.1, you mention one disadvantage of the trust > list approach. You might want to mention the problem that > compromise of any trusted certificate compromises the entire system. We will add this. > * In section 1.4.2, you say that a partial mesh is a mixture > of unidirectional and bidirectional cross-certifications. > It's probably also important to note that in a partial mesh > there may be CAs that are not directly cross-certified at all. We will add this. > * What is a Root CA doing in Figure 3? As you say, each EE in > a mesh usually trusts the CA that issued its certificate. > Because of this Root CA, Figure 3 does not depict a full > mesh, as stated in 1.4.2. I suggest that you remove the Root CA. This was changed to be a root based upon a comment made by Denis Pinkas stating that the diagrams were inconsistent. I think I agree with you, it is more confusing with the root in there. We will change this. > * At the end of section 1.4.3, you raise the concern that the > assurance of a high-assurance PKI may be diluted by > cross-certifying with a less restrictive PKI. If you're going > to raise this concern, you should mention that it can be > addressed through the use of certificate policies. We will add this. > * At the end of section 1.4.4, you say that connecting PKIs > with a bridge results in a non-hierarchical PKI. Certainly, > this is true. But mesh and hybrid PKIs are not hierarchical > either. Why raise this here? This is mentioned because a lot of people look at a picture involving a Bridge, and think of the Bridge as a root, and everything being subordinate to it. This is just another attempt to prevent that. > And the following sentence which states that any PKI can be > considered hierarchical from the right perspective only > applies if you remove and duplicate links in the PKI. Since > you don't have space to get into this here, I suggest you > remove those last two sentences. We already reference section 2.3 where we go into some more detail. > * At the end of the first paragraph of section 2.1, you > explain that S/MIME messages commonly include certificates. I > think you would be wise to also mention SSL/TLS since this is > the most widespread use of PKI and it also supplies > certificates in the protocol. We will add this. > * Before Figure 6, you explain how certificates are depicted > with arrows in your figures. You should also explain the > "B(A)" notation you use. We will add an explanation to this section. > * At the end of the paragraph after Figure 6, you say that > building paths is as important as validating them. I don't > agree. Many products have been shipping for years and work > just fine without building. In a complex PKI, building is > important. But in a simple hierarchical PKI, it's not. Even in a simple hierarchical PKI, some amount of building has occurred. (OK, maybe not in a REAL simple example with one trust root and all target certificates issued directly by it...) Building is important. If you don't build a path, you can't validate a path. It's just that building in certain cases is so easy it doesn't even seem like you're doing any work. > * At the end of section 2.1, you have a large discussion of > why you believe building forward is usually better. This > duplicates later discussions on this topic. I suggest you > save this for later. We will shorten section 2.1 and provide a cross-reference to the end of 2.3 where we discuss this in more detail. > * In section 2.2, you could simplify and clarify Criterion 1 > by changing it from this: > > Criterion 1: The implementation is able to find all possible paths. > By this, it is meant that all possible certification paths > between the trust anchor and the target certificate which may > be valid paths can be built by the algorithm. > > to this: > > Criterion 1: The implementation is able to find all possible paths. > By this, it is meant that all valid certification paths > between the trust anchor and the target certificate can be > built by the algorithm. Yes, but I will propose changing it to (potentially) valid because I do not necessarily want to burden path building software with having to also be path validation software. By potentially, I mean it is valid as far as the builder can tell. To some builders that is fully valid. To others it may still require a final sanity check/run through the validation procedure. > * In section 2.2, Criterion 2 seems rather odd. Who cares > which invalid paths you build first? The important thing is > how quickly and efficiently you can build a valid path or > determine that no valid path exists. The purpose of this criterion is ensure that the users get meaningful errors. You want to build paths which are likely to be valid before you go off and build paths which have a very low chance of success. What is a better error for a user to see, "name constraints violation at certificate 14" or "cert 3 revoked"? We will try and word this a bit better to get that point across. > * In section 2.3, you say that because certificates are not > permitted to repeat, every potentially valid path has a > terminus. But every potentially valid path *always* has a > terminus, even if certificates are allowed to repeat. As > defined in RFC 3280, a certification path is a collection of > n certificates. Every path has a finite number of > certificates. I suggest you remove the sentence "As a result, > every potential valid path has a terminus, a leaf on the tree." Each path does have a terminus, once it becomes a path. If certificates can be repeated, the number of certificates in the path and the number of possible paths both increase to infinity. This is a wordsmithing issue that may be able to be worded more clearly. Perhaps we will change it to say "If certificates could be repeated, loops can be formed such that the number of paths and the number of certificates in a path both increase without bound (e.g. A issues to B, B issues to C, and C issues to A.)" > * In the following paragraph, you say that removing and > duplicating nodes and links in a PKI to turn it into a > hierarchy greatly simplifies software design. This may be so, > but it also removes many opportunities for insight and > optimization. For example, it increases the number of nodes > in the data structure and makes it harder to mark a certain > area of the PKI as unproductive (since that area may appear > several places in the tree). We will remove "- this greatly simplifies software design." from the sentence. > * Later in that paragraph, you use the phrase "decision tree" > without defining it. I suggest that you define it in section 1.3. We will add a definition for "decision tree". > * One paragraph on page 16 (starting with "A more complicated > example") talks about problems with building in the reverse > direction when there are many trust anchors. The real issue > here is fanout. > Whether you're building forward or building reverse, it's bad > news when you get to a point where you have several choices. > It's especially bad if your heuristics (ranking) don't give > you much help in deciding which way to go. And it's even > worse if you're using depth-first search, since one wrong > move can send you off in the wrong direction for hours. > That's why I suggest breadth first search or (even better) > ranking all candidates at all points in the tree at each > decision time. > > Four trust anchors is a four-way fanout. A similar situation > would arise if you were building forward and arrived at a CA > that has four certificates with it as the subject. In either > case, you hope that your ranking will help choose the right > path. And if you're using depth first search, you really hope > that you don't take a wrong turn. > One technique for dealing with extreme fanout with fairly > equal rankings is to start building from the other end. > Another is to rank the certs at the fanout point low and try > another branch. A third is to use DPD to get help. And a > fourth is to tell PKI designers to use name constraints and > other techniques at high fanout points (like > bridges) to help path building succeed. > > I hope that you find this analysis useful. I agree that this example may go into a little too much detail without providing much benefit to the reader--we've done a good job of identifying the problem but not at clearly identifying a solution. I will try and simplify the example, identify that the real enemy here is fanout, and that to combat it, implementers should strive to use optimizations such as those discussed in section 3.1 when there are a lot of choices at a particular point of development. (By the way, this is p18 on my copy.) > * In Figure 10, are W, X, Y, and Z all trust anchors? I think so. > That's pretty odd. Why would a relying party have all four of > those trust anchors? Who needs the bridge CA in that case? I > suspect that this topology was created to support your > arguments that repeating a (name, key) pair is bad and that > building in reverse is bad. If so, I think you can do better. They are trust anchors for different relying parties, the trust anchors of each of the PKIs which are cross-certified with the Bridge. We will attempt to make that clearer in the example. In our example we are treating only W as our trust anchor. > Let's do a careful theoretical analysis and try out different > algorithms on different topologies. I suspect you're right > that repeating a (name, key) pair is bad but we need to think > about this carefully since it is a significant change to RFC > 3280. I think you will see below that building in reverse is > a useful tool that should be used in conjunction with > building forward. But we need to show the merits of our > approaches through analysis, not by consructing topologies > that serve our own purposes. We have identified reasons why a (name, key) pair should not be repeated within this document, and why building implementations should consider not allowing them in section 2.4.2. We have not attempted to change RFC3280 although I think we would have a decent argument for this, it is for a different time and place. > * In the paragraph after Figure 10, you have several > sentences explaining the diagram's notation. This was > explained much earlier. > The rest of the paragraph (explaining where certificates > would be stored in an LDAP directory) also duplicates > material found elsewhere. > I suggest that you remove these, since the document is > already too long. We will remove the duplication. > * In section 2.4.2, you seem to be proposing that software > should build a decision tree. I believe you don't intend for > the software to actually build a tree for the entire PKI but > only to move incrementally through the PKI adding nodes and > links to the tree as needed. I hope that's the case, anyway. > Otherwise, you'd need to download all certs in the PKI! If > I'm right, you need to be much clearer about this. I could > easily imagine a developer writing code that actually builds > the tree. That code would work fine in a small test PKI but > flood the directory with requests and then die in a large PKI. We will try and make this clearer. > * After Figure 11, you point out that building in reverse is > not good in this case. Sure. The EE is in a simple hierarchy. > Of course, building forward is best there. If you don't know > the topology (especially if you have several trust anchors), > starting with forward is fine. Then you can switch or meet in > the middle if things get hairy. As I mentioned before, one can always make an example that is good for a particular argument! In this case though, we don't say it is not good, we say it is not possible, due to a lack of available certificates. > * In section 2.6, you say "[...] the path builder only needs > to store the current location in the tree [...] All completed > branches can be discarded from memory [...]". Maybe. There's > a lot to be said for retaining records of paths you have > already tried and rejected. Then you can avoid retrying them > at a different point in the graph (unless conditions are > sufficiently different to merit a retry). I agree with what you've written. We have even some implementations that mark a path as bad given a certain set of inputs so that it is not tried again. We will reword this section. > * Later in section 2.6, you say "Consider HTTPS support" for > CRLDP and AIA. Why would this be valuable? You're retrieving > signed data. Using HTTPS may trigger another path build, > maybe even a loop where one build triggers another which > triggers another and so on. I suppose some repositories may > require HTTPS to authenticate the client or protect the certs > from prying eyes. But you should probably warn that doing so > may be expensive and that implementers should protect against loops. I think we know of someone that is using HTTPS for one of these cases, and that is why it was thrown in there. We will add something to warn users that HTTPS support needs to be considered in terms of its potential impact creating a loop. > * Toward the end of section 2.6, you talk about the path > cache. You call for "a configurable expiration date for each > entry". Who would configure the date and why? I'm a path > building geek and I can't imagine configuring such a thing! By configurable, I think it is meant that the software would set the expiration when it creates the entry, based upon a number of factors. Examples of factors include the expiry of the information used to make the determination, bandwidth, assurance level, storage space, whether it is a server or client, etc. We will make this clearer. > * A few sentences later, you suggest storing issuer-subject > cert relationships. If you want to do this, I suggest that > you use a cert hash instead of an (issuer, serial) tuple. > Issuer, serial is *not* always unique, especially across > multiple PKIs or when one CA is malicious (and remember that > some CAs are only partly trusted in a modern PKI). According to X.509, Issuer, Serial is always unique, since a given issuer cannot repeat a serial number. Of course, this assumes a "global directory" where issuers are never repeated. Issuer/Serial tuple is used as the certificate identifier in a number of protocols (S/MIME for example) and of course it is important for users to ensure that the resultant path is valid through validation of signatures. > * In section 2.7.1, you talk about the required inputs. > Instead of supplying the actual Target Certificate, it is > sometimes useful to provide criteria that describe what sorts > of target certificates you're willing to accept. For > instance, when doing S/MIME encryption to a new correspondent > you may not have the correspondent's encryption certificate. > The path building software can help find it if you tell it > what you need. We will note this as a possibility. Our goal for this document was "given a certificate, how do I build its path" but we will note this. > * In section 3.2, you recommend that path building software > output a detailed log. I think you should recommend that this > log be carefully structured so that the paths tried can be > easily reconstructed by a diagnostic program. My team built a > tool that shows the cert graph and then replays the paths our > builder tried, lighting up each one in sequence. We found > this a great help in understanding the behavior of our > software (finding bugs, improving algorithms, etc.). It's > also a very cool way to demo path building, which otherwise > is a pretty boring demo ("see, there's the web page!"). Heh. Mark that up as a good reason to work in R&D. We will make this recommendation. (About the reconstruction of the paths, not necessarily the lighting-up part.) > * Later in that paragraph, you say "Ideally, there would be a > mechanism for turning this logging on and off [...]". I'd > change this to "There should be [...]". Logging is expensive > (in storage and CPU time), especially for path building where > you commonly try hundreds of certificates or more to create a > valid path. You really need a way to turn logging off and on. We also can't be prescriptive since this is an Informational document... I agree but I will not use the word "should". > * A few paragraphs later, you say "it may be useful to not > rule out any paths" and just return each path as its built, > even if it's invalid. The problem with this is that > (especially with a depth first > search) is that there are many topologies that may cause you > to wander among many invalid paths. Why is this technique > good? You say it will "provid[e] a more rapid response to the > caller than one which builds all possible paths." Maybe. I > suspect you'll instead waste time by spending a lot of time > returning invalid paths to the caller. I don't see much > reason to return invalid paths except in a diagnostic log or > for your interesting idea of providing one path that's > *mostly* valid for debugging and in case the caller finds > that invalid path educational or compelling. Of course the simplest case of this is you build a path and everything lines up except a certificate along the way is expired or revoked. In this case, the path is probably the "right" one and is not valid, and won't be. This is what we're getting at. > * A few paragraphs later, you lay out your approach. Here are > some comments on it: > > You say "Do not check revocation status if it requires a > directory lookup or network access." Why not start the > revocation check and let it run in parallel while you do > other things (like checking certs deeper in the graph)? That > way, you'll have the revocation check done when you need it. > And if the check fails you can abort all work that was based > on the validity of that certificate. Of course, you'll want > to cache the revocation status of certificates for some time. Again, this gets back to keeping it simple for the implementer. See later, we will add to section 7 some more advanced optimizations that includes background/parallel processing. > You say "Do not check digital signatures". Again, why not run > this in parallel? Most of path building time (about 90%, > based on our data) is spent waiting for the network (while > downloading certificates and such). That's an ideal time to > be checking signatures. And, of course, you'll want to cache > the results of signature checks. Same as above. Signature checks are computationally expensive. For 3280-compliant certificates, AKID/SKID matching is a much less expensive way to know that the signatures are likely to chain. > You say "Do not check anything that can not be checked as > part of the iterative process of traversing the tree." Why > not? I expect you would want to run these final checks (like > full policy processing when building forward) before > returning the path to the caller. Otherwise, the caller will > need to run a complete validation of the path, which will > take much more time than just running these last few checks. I'm not entirely sure I understand what you're saying here. We're recommending you don't check anything that's going to get checked anyway. > * In the last paragraph of 3.2, you say "Second, it is fairly > uncommon in typical application environments to encounter a > revoked key; therefore, most certificates validated will not > be revoked." In a PKI, we don't revoke a key. We revoke a > certificate. Therefore, this sentence should read "Second, it > is fairly uncommon in typical application environments to > encounter a revoked certificate." Agree. We will make this change. > * Why include section 3.3 at all? It's very odd to have an > IETF spec talking about internal data structures. Maybe you > should just give everyone a copy of your code (since it is > apparently ideal) and save them the trouble of implementing > it! I'm going to skip my more detailed comments on this > section since I think it should be entirely removed. I have no problem removing the data formats, I think it is useful to have a semi-pseudocode example worked through for implementers to read, so I don't think we should remove the entire section. > * In section 3.4, you refer say "developers should evaluate > each method with respect to the environment that the > application will operate, and assign weights to each > accordingly in the path building software.". First, the word > "that" in this sentence should be "in which". Second, this is > a prime example of the awful idea that developers must > examine each PKI and configure their software to run optimally there. We will change to "in which". Again, if efficiency is your number one concern, it may be worthwhile to tune the path builder for your environment. An example of a situation that deserves tuning is a server-based implementation, when it is always processing paths for a particular infrastructure. We have typically implemented the scoring weights to be read from configuration files so that they can be tuned; default values would be used in absence of tuned numbers. > * In section 3.5.1, you say "According to [RFC 3280], > basicConstraints is required to be present with cA=TRUE in > all CA certificates." > Actually, section 6.1.4 (k) of RFC 3280 says "Verify that the > certificate is a CA certificate (as specified in a > basicConstraints extension or as verified out-of-band)." > Maybe your software doesn't make any provision for > out-of-band indication of CA certificates, but you should > probably at least note the possibility that other software may do so. We will change this to identify this rule must account for out-of-band indications. > * In section 3.5.5, you say "Certificates in the path cache > have been validated previously. There is some probability > that the path from that certificate to a trust anchor is > still valid." The path may still be valid but it may contain > constraints that are inconsistent with the path from the > Target Certificate to this certificate. You should probably > note that possibility. We will add an additional note about the changes to initial constraints. > * In section 3.5.6, you say that the Previously Verified > Signatures sorting method doesn't apply when building in > reverse. Actually, it does. Your analysis is incorrect. This > method never tells you which way the Target or Trust Anchor > is (which certificate is most likely). > It only lets you rule out certificates because the signature > check would certainly fail. I agree that this does help and will be included. > * In section 3.5.7, you say the Path Length Constraints > sorting method "may be applied in reverse, but the benefit is > likely less than that of the forward direction". Why? You can > argue that the method is more effective when building in > reverse because path length constraints often appear close to > trust anchors. Like you mention above, the forward direction lets you immediately rule out a branch of certificates that would fail. In the reverse direction, you rule out branches farther down with some unknown fanout. That said, I think this section reads too opinionated and will change it appropriately. > * In section 3.5.8, you say "Certificates that contain > nameConstraints that would be violated by certificates > already in the path to this point are given lower priority > (or perhaps eliminated)." They should definitely be > eliminated. You know they can't be used to build a valid > path, so what's the point (unless you're working on building > an invalid path)? I agree, this will be changed to eliminate certificates. > * In section 3.5.9, you say "While this is viable, the > signature verifications required make it less attractive as > an elimination method." Usually, a CRL check only requires > one signature verification so "verification" should be singular. Will do. > You also say "It is suggested that this method only be used > for sorting and that CRLs are validated post path building." > As I noted above, fetching CRLs and performing signature > checks can be done in parallel with other work. It's a good > idea to do this so that you can discover revoked certs before > you have invested too much time in them (and also so that you > have the results of the revocation check ASAP). See my above comments, we are trying to stay simple. > You should also probably change this section to include > discussion of OCSP. Right now, it's very focussed on CRLs. I agree, we will try and make it less specific to CRLs. > * Here's a sorting method similar to the "Issuer Found in the > Path Cache" method, but better suited to building in reverse. > Check whether the subject of the candidate certificate > matches the issuer of a certificate sent by the target (in > SSL/TLS, S/MIME, etc.). If so, then the candidate certificate > should be ranked more highly because it is more likely to > lead to a path to the Target Certificate. You may also want > to do RDN Matching (as in 3.5.15) with the certificates sent > by the target. I agree that this is a helpful sorting method. With your permission, we will add it to the document. > * In section 3.5.15 (on RDN matching), you say "Additionally, > it should be noted that this method can require a lot of > processing if many trust anchors are present." Actually, this > should only require a few hash table lookups (of the RDNs). Assuming you implement it with a hash table! > * Section 3.5.18 could be clearer. I think you're saying that > the subject and issuer of the candidate certificate should > have lots of RDNs in common. But this could be understood to > be saying that the subject of the candidate certificate > should have lots of RDNs in common with the issuer of the > next certificate. Of course, they must match exactly so that > can't be what you're saying. But it would help if this > section was more explicit. We will attempt to reword this to make it clearer. > * Section 3.5.19 also should be clarified. The description of > the reverse method is much clearer than that of the forward > method. I suggest that you replace the forward method > language with language based on the reverse method language. > One change to the reverse method language, though. Instead of > saying "If an entity named by a reverse certificate", say "If > the subject of a certificate". There's no such thing as a > "reverse certificate". We will reword the forward definition to be closer to the reverse. > Note also that just because you find a name in the > certificate cache, that doesn't mean you have the right > certificates. There may be several different CAs with the > same name. Also, you may now have more clues (like AIA and > SIA) for finding certificates than you had before. > You should be sure to pursue any such new clues. Agree, that's why it's just a sort and not an eliminate rule. > * In the Justification for 3.5.19, you say "The presence of > required directory values populated in the cache increases > the likelihood that all the required certificates and CRLs > needed to complete the path from this certificate to the > trust anchor (or target if building in > reverse) are present in the cache from a prior path being > developed [...]". That's fine, but you might also want to > keep the actual path previously developed and look at that as > a prime candidate for the path this time. We do recommend a cache for relationships and a cache for information in section 2.6. The idea of course is that given the cache of relationships and the cache of information, you could reconstruct the previous path. See rule 3.5.10. > * In section 3.5.20, you use the presence of a CRL in the CRL > cache to indicate whether a complete path through a CA has > previously been found. This is rather indirect. It would be > better to actually track which certs have been previously > included in valid paths. Keeping a copy of the path would > also be quite useful. Agreed. Again, rule 3.5.10 is more useful, and then this can be a secondary rule to that. > * The discussion of Forward Policy Chaining in section 4 > overstates the benefits of policy checks when building in the > forward direction. > We should discuss this more. We don't feel like we're overstating, and welcome more discussion. > * In section 5.2, you suggest that each name/key pair only be > allowed to appear once in a path. Since there is no such > requirement in X.509 or RFC 3280, this would violate > Criterion 1 as stated in section 2.2. > You would miss some valid paths. This is what happens when you have 5 authors on a document! This contradiction will be resolved, most likely by changing the Criterion 1 (since we feel strongly about the non-repetition of the name/key). > > * In section 6 (Retrieval Methods), you should mention > getting certificates and CRLs from the target (with S/MIME or > SSL/TLS, for instance). That's a *great* way to get > certificates and CRLs. The ones you get are usually highly valuable. Yes we will add this. Of course, most commercial implementations don't provide CRLs in this way. > * Sections 6.2 and 6.3 should include text that encourages > implementers to support AIA and SIA, especially HTTP. This > text from the end of section 6.4 could easily be adapted: > > However, implementers of path building and validation > modules are strongly encouraged to support CRLDPs. At a > minimum, developers are encouraged to consider supporting the > LDAP and HTTP transports; this will provide for > interoperability across a wide range of existing PKIs. We will add this (appropriately modified) text to 6.2 and 6.3. > * In section 7, you should talk about prefetching and > parallel fetching. Both are good ways to improve the > performance of a cert path building implementation. We will add these to section 7 as more advanced possibilities. > * In section 7.1, you say "Certificate processing systems can > cache data retrieved from external sources for some period of > time, but not to exceed the useful period of the data (i.e., > an expired certificate need not be cached)." CRLs are often > issued well before the nextUpdate of the previous CRL. Could > you mention this and suggest that implementers check > periodically for updated CRLs? Similarly, > crossCertificatePairs may be updated on an unpredictable > basis. If you don't check back every so often, you'll never > know that a new cross certificate is available! We could, although it will take a bit of additional writing to explain these issues. We will attempt to encourage retrievals before the nextUpdate has passed. > * The last bullet in section 7.1 offers a few suggestions for > other things to cache. I'd add to that list previously built > paths and partial paths. If the cache is safe from > unauthorized modifications (as with an in-memory cache), you > may also want to cache validation and signature check status > for certificates and CRLs. We will add these additional recommendations. > * In section 7.2 (Retrieval Order), you may want to consider > checking repositories in parallel instead of serially. Also, > you might want to run ahead with your path building and > validation while a network query is ongoing. Maybe you can > build and validate the path with just the certificates and > CRLs you have already (thus eliminating the need to wait for > the network query to complete). Anything to get the answer to > the user more quickly! Agreed. This has been noted a number of times in this message. > * In section 8.1, you should probably add a statement that > path building can be used as a denial of service attack. Make > a series of simple requests to a server and cause it to do a > bunch of long path builds. To avoid this, the server can > limit the amount of resources it's willing to devote to a > path build. This amount can be reduced when the load is > heavy. And standard DOS protections like identifying > attackers can be employed. Yes, we will add something about the threat of DOS attacks on server DPD systems and systems that handle authentication. > * Section 8.2 goes way beyond RFC 3280 and X.509. I know that > Santosh is working on getting this into X.509 and the > successor to RFC 3280. > I'll debate the specifics of his proposal in that context. > But for this document, I think you should be careful to > explain that this goes beyond the standards. If you implement > it, you will violate Criterion > 1 from section 2.2. Maybe the best thing to do is to just > briefly point out that there's an issue here and it's under > discussion in the standards groups. Once it's settled there, > this document can be revised to reflect whatever's agreed on. This issue has been debated on the list and there has not been any objection to it. It does not violate Criterion 1 of Section 2.2 since the path for the CRL signer needs to be developed for the same CA hierarchy as the certification path. > Minor Comments: We will accept all your minor comments. Steve, I know I speak for all of the authors when I thank you for this detailed review of the document. It is a tremendous help to know that someone else has read and thought about the entire document, and even if you have differing opinions on parts of it, you were willing to help create a better finished product. Sincerely, --Peter +---------------------------------------------------------------+ | Peter Hesse pmhesse@geminisecurity.com | | Phone: (703)934-2031 Gemini Security Solutions, Inc. | | ICQ: 1942828 www.geminisecurity.com | +---------------------------------------------------------------+ "Pay no attention to what the critics say; there has never been a statue set up in honor of a critic." --Jean Sibelius Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4HG91hi082056; Mon, 17 May 2004 09:09:01 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4HG91dG082055; Mon, 17 May 2004 09:09:01 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4HG91Ju082049 for <ietf-pkix@imc.org>; Mon, 17 May 2004 09:09:01 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Mon, 17 May 2004 09:09:03 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Mon, 17 May 2004 09:09:00 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 17 May 2004 09:09:03 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Mon, 17 May 2004 09:09:02 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F02963552@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ74/RcOtUtTjLAQpuHulPYJsLT7QARP2VQ From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 17 May 2004 16:09:03.0591 (UTC) FILETIME=[45D1A770:01C43C29] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4HG91Ju082050 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Denis, Another round will no longer than this is taking. Rather than read items out of context, it is better to read the full proposal. Trevor * -----Original Message----- * From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] * Sent: Monday, May 17, 2004 12:53 AM * To: Trevor Freeman * Cc: ietf-pkix@imc.org * Subject: Re: SCVP & validation policies * * Trevor, * * > Hi Denis, * * > As I said previous an OID is one of the possible options, but there is * > no consensus to make this mandatory so SCVP will continue to support * > both options. * * Making the OID optional as requested by Wen-Cheng Wang, does not * necessarilly mean "supporting both options". * * > I also don't see consensus that it is mandatory that a * > SCVP server will always map a request back to the OID. * * Many opinions, except one, are supporting this however. * * > I will be posting SCVP 15 shortly so you can review the latest draft. * * An extract of the proposal sent to the mailing list and targeted to this * problem would probably save another round. * * Denis * * > Trevor * > * > -----Original Message----- * > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] * > Sent: Thursday, May 13, 2004 12:54 AM * > To: Trevor Freeman * > Cc: Florian Oelmaier; ietf-pkix@imc.org * > Subject: Re: SCVP & validation policies * > * > Trevor, * > * > (text deleted) * > * > Florian said: "it is widespread consensus on this list that * > this could be done with SCVP by the use of the OID as a reference * > to the validation policy as the sole configuration item." * > * > I agree. * > * > To respond to a request from Wen-Cheng Wang, the OID of the validation * > policy should be optional in the request. However, if the OID of the * > validation policy is not present in the request, then the OID of the * > validation policy that has been used SHALL be returned by the SCVP * > server in * > the response. If the validation policy has dynamic parameters, then * > these * > parameters SHALL be returned too. * > * > Now, we need to move, Trevor. * > * > Would you propose in an e-mail extracts of an ASN.1 syntax that would * > accommodate these requirements for both the request and the response, * > focussing on the validation policy elements ? * > * > Thanks, * > * > Denis * > * > (text deleted) * > * > * Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4HFNiKW078830; Mon, 17 May 2004 08:23:44 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4HFNiZj078829; Mon, 17 May 2004 08:23:44 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail19e.dulles19-verio.com (mail19e.dulles19-verio.com [198.170.241.12]) by above.proper.com (8.12.11/8.12.9) with SMTP id i4HFNhQ4078823 for <ietf-pkix@imc.org>; Mon, 17 May 2004 08:23:43 -0700 (PDT) (envelope-from pmhesse@geminisecurity.com) Received: from www.geminisecurity.com (161.58.96.110) by mail19e.dulles19-verio.com (RS ver 1.0.92vs) with SMTP id 4-0863847424; Mon, 17 May 2004 11:23:45 -0400 (EDT) From: "Peter Hesse" <pmhesse@geminisecurity.com> To: "'PKIX List'" <ietf-pkix@imc.org> Cc: "'Steve Hanna'" <Steve.Hanna@sun.com>, "'Wen-Cheng Wang'" <wcwang@cht.com.tw> Subject: RE: Comments on Path Building I-D Date: Mon, 17 May 2004 11:23:45 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook, Build 11.0.5510 In-Reply-To: <00c101c43293$cfa00a90$4f85900a@wcwang> X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Thread-Index: AcQymwDrjoePWzQZSJe1V4Jui0XfqQJh8KFQ Message-ID: <20040517112345.GA86384@mail19e.dulles19-verio.com> X-Loop-Detect: 1 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Wen-Cheng, I will reply to your comments inline. > After reading the certpathbuild draft and the comments from Steve > Hanna, my feeling is that this draft is more like a paper (or thesis) > than a PKIX specification. I believe that is why Steve criticized the > text of the draft as if he is a reviewer of a PKI paper. I really > admire the authors and Steve's rich and deep knowledge of optimization > and implementation technique for path building. I must confess that I > gained much from reading "the paper" and the comments from Steve. > However, I doubt whether it is suitable for the PKIX WG to publish > "the paper" as an RFC. > I think the text of the draft is good enough to become a paper on a > PKI conference proceedings or even on a journal. > But, that does not mean it will be a good PKIX RFC. This document is meant to serve as an informational document which implementers can read and use to make intelligent decisions when building their own implementation. Unlike most PKIX RFCs, which are standards-track RFCs, this document is not intended to be prescriptive. If you read RFC 2026, section 4.2.2 about Informational RFCs, I believe you will agree that this document fits the mold of an Informational RFC quite well. > As a faithful reader of PKIX specifications, I would expect that if > the PKIX WG would like to publish an RFC entitled "Certification Path > Building" then its text should focus on: > > 1. defining what is a legal certification path > Can certificates with names in different encoding be > chained? > Can certificates with unmached issuer name and subject name > be chained by matching SKID and AKID? These issues are covered in the document, and both have been brought up on this list as recommended additions to RFC 3280. > 2. defining a algorithm for certification path building, also > including > specifying the inputs to the algorithm and > specifying the outputs from the algorithm As I mentioned above, an informational RFC should not be so prescriptive > 3. specifying the format of certificate/CRL extensions (e.g. SKID, > AKID,AIA, SIA, CRLDP...) that are related to certification path > building. > E.g., > the URL format of different procols (such as LDAP, HTTP, > FTP...) > (e.g., Should the LDAP URL contains the ";binary" option?) These extensions are already specified in RFC3280. The specifics of the formats of these values may be best captured in a "best-practices" document which I have suggested (in my response to Steve) as an additional document to be considered by the group. Aside from the restrictions identified in section 4.2.1.7 of RFC3280, I do not think the values can be constrained any further. > 4. specifying how a compliant implementation should process > theose certificate/CRL extensions related certification path > building Again, this is specified by RFC3280, our document makes no attempt to change the way any extensions or fields are processed, nor should it. >5. specifying what is the minimum set of repository retrieving >protocols (such as LDAP, HTTP, FTP, SMTP...) a compliant >implementation sould support and how these protocols should be >supported. E.g., What is the expected object format? > (e.g., DER binary vs.Base64-encoded) > What is the expected MIME type if HTTP is used? Again, this is best suited for a "best-practices" document. There need not be any minimum set of protocols defined as a standard, but a list of best practices could give guidance on what protocols were most commonly used. > 6. discussing security consideration > > I believe that the PKIX certpathbuild specification should specify a > baseline certification path building algorithm in the fashion of RFC > 3280 specifying a basic certification path validation algorithm. The > purpose of defining the algorithm is for precisely describe the > expected result of the certification path building. RFC3280 section 3.2 has identified a starting point, which is a certification path. We have provided a way to get from nothing to obtaining that certification path, after which time you can perform validation; a procedure that RFC3280 is silent on. There cannot be a basic algorithm for building defined as a PKIX standard, because this would make any algorithms not designed in this fashion non-standard. RFC3280 section 3.2 provides enough to give us the basic algorithm. > There is no need for a PKIX > specification to teach implementators the optimization and > implementation technique for path building. It is up to the > implementators themselves to choose their own the optimization and > implementation technique. Isn't it? Yes, it is up to the implementors. Our document provides a recommendation on what makes a good path builder, with example scenarios and an example implementation. All throughout, the document makes it clear that the implementor can take or leave any of the suggestions. > Therefore, I would like to suggest the authors to complete remove the > section 3 of the current draft. It is not only because a PKIX > specification is not supposed to be a paper or textbook for teaching > the optimization and implementation technique for path building, but > also because the optimization and implementation technique described > in that section is not, as Steve said, the consensus of the PKIX WG. > I suggest that the draft sould spend more space and time to cover the > areas I mentioned above. > (I know the current draft already covers them, but I think it is still > not in-depth enough.) As I mentioned above, I believe that RFC3280 already covers the majority of these issues, and in addition, a new document may be created to provide guidance on the best practices of certification path development utilities. As I mentioned in my message to Steve, I would be happy to collaborate on such a document. --Peter +---------------------------------------------------------------+ | Peter Hesse pmhesse@geminisecurity.com | | Phone: (703)934-2031 Gemini Security Solutions, Inc. | | ICQ: 1942828 www.geminisecurity.com | +---------------------------------------------------------------+ "Pay no attention to what the critics say; there has never been a statue set up in honor of a critic." --Jean Sibelius Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4HDw1OL073125; Mon, 17 May 2004 06:58:01 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4HDw1fI073124; Mon, 17 May 2004 06:58:01 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from ns3.worldcall.net.pk (ns3.worldcall.net.pk [203.81.192.10]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4HDvxqb073118 for <ietf-pkix@imc.org>; Mon, 17 May 2004 06:58:00 -0700 (PDT) (envelope-from yasir.khan@ascertia.com) Received: from ascertia3 ([203.81.202.25]) by ns3.worldcall.net.pk (8.12.11/8.12.11) with SMTP id i4HDsLOw003647; Mon, 17 May 2004 19:54:21 +0600 (PKST) Message-ID: <011d01c43c17$5eaee0d0$b500a8c0@ascertia.com.pk> From: "Yasir Khan" <yasir.khan@ascertia.com> To: <ietf-pkix@imc.org> Cc: <a.alam@ascertia.com> References: <8B888AAAAB0FD31189590008C79184430D6F3E42@zbl6c002.corpeast.baynetworks.com> <401E90A8.5050108@drh-consultancy.co.uk> <1075746193.2411.186.camel@wbl6yd026.us.nortel.com> Subject: Signature in OCSP Request Date: Mon, 17 May 2004 19:00:53 +0500 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_011A_01C43C41.46BCB6D0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 X-Scanned-By: MIMEDefang 2.39 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multi-part message in MIME format. ------=_NextPart_000_011A_01C43C41.46BCB6D0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hello I have a question about the signature field in OCSP request. What = exactly should it contain and what=20 algorithms can be used for OCSP request signing?=20 When i create a signature using MS CAPI, OCSP reponders are unable to = verify the signatures whereas we can verify it using CAPI. What data do we set in signature field = exactly? Regards, Yasir ------=_NextPart_000_011A_01C43C41.46BCB6D0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META http-equiv=3DContent-Type content=3D"text/html; = charset=3Diso-8859-1"> <META content=3D"MSHTML 6.00.2600.0" name=3DGENERATOR> <STYLE></STYLE> </HEAD> <BODY bgColor=3D#ffffff><FONT face=3DArial color=3D#800080 size=3D2> <DIV><BR>Hello</DIV> <DIV> </DIV> <DIV>I have a question about the signature field in OCSP request. What = exactly=20 should it contain and what <BR>algorithms can be used for OCSP request = signing?=20 </DIV> <DIV> </DIV> <DIV>When i create a signature using MS CAPI, OCSP reponders are unable = to=20 verify the signatures whereas we<BR>can verify it using CAPI. What data = do we=20 set in signature field exactly?</DIV> <DIV> </DIV> <DIV>Regards,</DIV> <DIV> </DIV> <DIV>Yasir</DIV> <DIV> </DIV> <DIV></FONT> </DIV></BODY></HTML> ------=_NextPart_000_011A_01C43C41.46BCB6D0-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4H7qmvS009805; Mon, 17 May 2004 00:52:48 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4H7qmPe009804; Mon, 17 May 2004 00:52:48 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from odin2.bull.net (odin2.bull.net [192.90.70.84]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4H7ql4W009738 for <ietf-pkix@imc.org>; Mon, 17 May 2004 00:52:47 -0700 (PDT) (envelope-from Denis.Pinkas@bull.net) Received: from clbull.frcl.bull.fr (IDENT:root@clbull.frcl.bull.fr [129.182.8.31]) by odin2.bull.net (8.9.3/8.9.3) with ESMTP id KAA29842; Mon, 17 May 2004 10:02:02 +0200 Received: from bull.net (frcls4013.frcl.bull.fr [129.182.108.120]) by clbull.frcl.bull.fr (8.9.3/8.9.3) with ESMTP id JAA24690; Mon, 17 May 2004 09:51:23 +0200 Message-ID: <40A86F6F.2050606@bull.net> Date: Mon, 17 May 2004 09:53:19 +0200 From: Denis Pinkas <Denis.Pinkas@bull.net> Organization: Bull SA. User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en, fr MIME-Version: 1.0 To: Trevor Freeman <trevorf@exchange.microsoft.com> CC: ietf-pkix@imc.org Subject: Re: SCVP & validation policies References: <33E7A1613A3480448996047B69308A2F02962DBF@df-grommit-01.dogfood> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Trevor, > Hi Denis, > As I said previous an OID is one of the possible options, but there is > no consensus to make this mandatory so SCVP will continue to support > both options. Making the OID optional as requested by Wen-Cheng Wang, does not necessarilly mean "supporting both options". > I also don't see consensus that it is mandatory that a > SCVP server will always map a request back to the OID. Many opinions, except one, are supporting this however. > I will be posting SCVP 15 shortly so you can review the latest draft. An extract of the proposal sent to the mailing list and targeted to this problem would probably save another round. Denis > Trevor > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Thursday, May 13, 2004 12:54 AM > To: Trevor Freeman > Cc: Florian Oelmaier; ietf-pkix@imc.org > Subject: Re: SCVP & validation policies > > Trevor, > > (text deleted) > > Florian said: "it is widespread consensus on this list that > this could be done with SCVP by the use of the OID as a reference > to the validation policy as the sole configuration item." > > I agree. > > To respond to a request from Wen-Cheng Wang, the OID of the validation > policy should be optional in the request. However, if the OID of the > validation policy is not present in the request, then the OID of the > validation policy that has been used SHALL be returned by the SCVP > server in > the response. If the validation policy has dynamic parameters, then > these > parameters SHALL be returned too. > > Now, we need to move, Trevor. > > Would you propose in an e-mail extracts of an ASN.1 syntax that would > accommodate these requirements for both the request and the response, > focussing on the validation policy elements ? > > Thanks, > > Denis > > (text deleted) > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4H5ellq062057; Sun, 16 May 2004 22:40:47 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4H5elDF062055; Sun, 16 May 2004 22:40:47 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from ns3.worldcall.net.pk (ns3.worldcall.net.pk [203.81.192.10]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4H5ej6E061928 for <ietf-pkix@imc.org>; Sun, 16 May 2004 22:40:46 -0700 (PDT) (envelope-from faisal.maqsood@ascertia.com) Received: from ascertiafm ([203.81.202.25]) by ns3.worldcall.net.pk (8.12.11/8.12.11) with SMTP id i4H5b2CW014939 for <ietf-pkix@imc.org>; Mon, 17 May 2004 11:37:03 +0600 (PKST) Message-ID: <004501c43bd1$e4778be0$9c00a8c0@ascertia.com.pk> From: "Faisal Maqsood" <faisal.maqsood@ascertia.com> To: <ietf-pkix@imc.org> Subject: Fw: Relationship between SAN and IAN? Date: Mon, 17 May 2004 10:43:27 +0500 Organization: Ascertia Ltd MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0042_01C43BFB.C96DC1A0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 X-Scanned-By: MIMEDefang 2.39 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multi-part message in MIME format. ------=_NextPart_000_0042_01C43BFB.C96DC1A0 Content-Type: text/plain; charset="unicode" Content-Transfer-Encoding: quoted-printable =0D=00=0A= =00-=00-=00-=00-=00-=00 =00O=00r=00i=00g=00i=00n=00a=00l=00 = =00M=00e=00s=00s=00a=00g=00e=00 =00-=00-=00-=00-=00-=00 =00=0D=00=0A= =00F=00r=00o=00m=00:=00 =00F=00a=00i=00s=00a=00l=00 = =00M=00a=00q=00s=00o=00o=00d=00 =00=0D=00=0A= =00T=00o=00:=00 = =00i=00e=00t=00f=00-=00p=00k=00i=00x=00@=00i=00m=00c=00.=00o=00r=00g=00 = =00=0D=00=0A= =00S=00e=00n=00t=00:=00 =00T=00u=00e=00s=00d=00a=00y=00,=00 = =00F=00e=00b=00r=00u=00a=00r=00y=00 =002=004=00,=00 =002=000=000=004=00 = =001=001=00:=000=004=00=0D=00=0A= =00S=00u=00b=00j=00e=00c=00t=00:=00 = =00R=00e=00l=00a=00t=00i=00o=00n=00s=00h=00i=00p=00 = =00b=00e=00t=00w=00e=00e=00n=00 =00S=00A=00N=00 =00a=00n=00d=00 = =00I=00A=00N=00?=00=0D=00=0A= =00=0D=00=0A= =00=0D=00=0A= =00H=00e=00l=00l=00o=00 =00F=00o=00l=00k=00s=00,=00=0D=00=0A= =00=0D=00=0A= =00I=00 =00h=00a=00v=00e=00 =00a=00 =00q=00u=00e=00r=00y=00 = =00f=00o=00r=00 =00c=00l=00a=00r=00i=00f=00i=00c=00a=00t=00i=00o=00n=00 = =00a=00n=00d=00 =00n=00e=00e=00d=00s=00 =00y=00o=00u=00r=00 = =00v=00i=00e=00w=00s=00.=00=0D=00=0A= =00L=00e=00t=00 =00u=00s=00 =00c=00o=00n=00s=00i=00d=00e=00r=00 = =00t=00h=00e=00r=00e=00 =00a=00r=00e=00 =00t=00w=00o=00 = =00c=00e=00r=00t=00i=00f=00i=00c=00a=00t=00e=00s=00 =00A=00 = =00a=00n=00d=00 =00B=00 =00(=00p=00a=00r=00t=00 =00o=00f=00 = =00t=00h=00e=00 =00c=00e=00r=00t=00i=00f=00i=00c=00a=00t=00i=00o=00n=00 = =00p=00a=00t=00h=00)=00.=00 =00=0D=00=0A= =00 =00 =00a=00.=00.=00 =00I=00f=00 =00A=00 =00i=00s=00 = =00i=00s=00s=00u=00e=00r=00 =00o=00f=00 =00B=00 =00a=00n=00d=00 = =00t=00h=00e=00r=00e=00 =00i=00s=00 = =00S=00u=00b=00j=00e=00c=00t=00A=00l=00t=00N=00a=00m=00e=00 = =00e=00x=00t=00e=00n=00s=00i=00o=00n=00 =00p=00r=00e=00s=00e=00n=00t=00 = =00i=00n=00 =00A=00 =00a=00n=00d=00 = =00I=00s=00s=00u=00e=00r=00A=00l=00t=00N=00a=00m=00e=00 =00i=00n=00 = =00B=00 =00a=00n=00d=00 =00v=00a=00l=00u=00e=00 =00o=00f=00 = =00S=00u=00b=00j=00e=00c=00t=00A=00l=00t=00N=00a=00m=00e=00 =00i=00n=00 = =00A=00 =00i=00s=00 = =00"=00U=00R=00L=00=3D=00h=00t=00t=00p=00:=00/=00/=00w=00w=00w=00.=00t=00= e=00s=00t=00s=00e=00r=00v=00e=00r=00n=00a=00m=00e=00.=00c=00o=00m=00"=00.= =00 =00I=00s=00 =00t=00h=00e=00r=00e=00 =00a=00n=00y=00 = =00r=00e=00q=00u=00i=00r=00e=00m=00e=00n=00t=00 =00i=00n=00 = =00R=00F=00C=00 =002=004=005=009=00 =00o=00r=00 =003=002=008=000=00 = =00o=00r=00 =00s=00o=00m=00e=00 =00w=00h=00e=00r=00e=00 = =00e=00l=00s=00e=00 =00f=00o=00r=00 = =00I=00s=00s=00u=00e=00r=00A=00l=00t=00N=00a=00m=00e=00 = =00v=00a=00l=00u=00e=00 =00(=00i=00f=00 = =00p=00r=00e=00s=00e=00n=00t=00)=00 =00i=00n=00 =00B=00.=00 =00I=00 = =00m=00e=00a=00n=00 =00i=00n=00 =00t=00h=00i=00s=00 =00c=00a=00s=00e=00 = =00I=00s=00s=00u=00e=00r=00A=00l=00t=00N=00a=00m=00e=00 = =00v=00a=00l=00u=00e=00 =00m=00u=00s=00t=00 = =00c=00o=00n=00t=00a=00i=00n=00 = =00"=00U=00R=00L=00=3D=00h=00t=00t=00p=00:=00/=00/=00w=00w=00w=00.=00t=00= e=00s=00t=00s=00e=00r=00v=00e=00r=00n=00a=00m=00e=00.=00c=00o=00m=00"=00?= =00 =00=0D=00=0A= =00 =00 =00b=00.=00.=00 =00S=00i=00m=00i=00l=00a=00r=00 = =00q=00u=00e=00r=00y=00 =00f=00o=00r=00 = =00S=00u=00b=00j=00e=00c=00t=00K=00e=00y=00I=00d=00e=00n=00t=00i=00f=00i=00= e=00r=00 =00a=00n=00d=00 = =00A=00u=00t=00h=00o=00r=00i=00t=00y=00K=00e=00y=00I=00d=00e=00n=00t=00i=00= f=00i=00e=00r=00 =00t=00h=00a=00t=00 =00i=00f=00 = =00S=00u=00b=00j=00e=00c=00t=00K=00e=00y=00I=00d=00e=00n=00t=00i=00f=00i=00= e=00r=00 =00v=00a=00l=00u=00e=00 =00i=00s=00 =00o=00f=00 = =006=000=00-=00b=00i=00t=00 =00l=00e=00n=00g=00t=00h=00 =00i=00n=00 = =00A=00.=00 =00I=00s=00 =00R=00F=00C=00 =00a=00l=00l=00o=00w=00s=00 = =00t=00o=00 =00p=00u=00t=00 = =00A=00u=00t=00h=00o=00r=00i=00t=00y=00K=00e=00y=00I=00d=00e=00n=00t=00i=00= f=00i=00e=00r=00 =00v=00a=00l=00u=00e=00 =00o=00f=00 = =001=006=000=00-=00b=00i=00t=00 =00l=00e=00n=00g=00t=00h=00 =00i=00n=00 = =00B=00?=00=0D=00=0A= =00R=00e=00g=00a=00r=00d=00s=00,=00=0D=00=0A= =00F=00a=00i=00s=00a=00l=00 =00M=00a=00q=00s=00o=00o=00d=00=0D=00=0A= =00 ------=_NextPart_000_0042_01C43BFB.C96DC1A0 Content-Type: text/html; charset="unicode" Content-Transfer-Encoding: quoted-printable =FF=FE<=00!=00D=00O=00C=00T=00Y=00P=00E=00 =00H=00T=00M=00L=00 = =00P=00U=00B=00L=00I=00C=00 = =00"=00-=00/=00/=00W=003=00C=00/=00/=00D=00T=00D=00 =00H=00T=00M=00L=00 = =004=00.=000=00 = =00T=00r=00a=00n=00s=00i=00t=00i=00o=00n=00a=00l=00/=00/=00E=00N=00"=00>=00= =0D=00=0A= =00<=00H=00T=00M=00L=00>=00<=00H=00E=00A=00D=00>=00=0D=00=0A= =00<=00M=00E=00T=00A=00 = =00c=00o=00n=00t=00e=00n=00t=00=3D=00"=00t=00e=00x=00t=00/=00h=00t=00m=00= l=00;=00 = =00c=00h=00a=00r=00s=00e=00t=00=3D=00u=00n=00i=00c=00o=00d=00e=00"=00 = =00h=00t=00t=00p=00-=00e=00q=00u=00i=00v=00=3D=00C=00o=00n=00t=00e=00n=00= t=00-=00T=00y=00p=00e=00>=00<=00B=00A=00S=00E=00 =00=0D=00=0A= =00h=00r=00e=00f=00=3D=00"=00f=00i=00l=00e=00:=00/=00/=00F=00:=00\=00_=00= B=00a=00c=00k=00U=00p=00-=00F=00M=00\=00E=00m=00a=00i=00l=00 = =00S=00i=00g=00n=00a=00t=00u=00r=00e=00\=00"=00>=00<=00!=00D=00O=00C=00T=00= Y=00P=00E=00 =00H=00T=00M=00L=00 =00P=00U=00B=00L=00I=00C=00 = =00"=00-=00/=00/=00W=003=00C=00/=00/=00D=00T=00D=00 =00H=00T=00M=00L=00 = =004=00.=000=00 = =00T=00r=00a=00n=00s=00i=00t=00i=00o=00n=00a=00l=00/=00/=00E=00N=00"=00>=00= =0D=00=0A= =00<=00M=00E=00T=00A=00 = =00c=00o=00n=00t=00e=00n=00t=00=3D=00"=00M=00S=00H=00T=00M=00L=00 = =005=00.=000=000=00.=003=005=000=002=00.=005=003=009=000=00"=00 = =00n=00a=00m=00e=00=3D=00G=00E=00N=00E=00R=00A=00T=00O=00R=00>=00=0D=00=0A= =00<=00M=00E=00T=00A=00 = =00c=00o=00n=00t=00e=00n=00t=00=3D=00"=00M=00S=00H=00T=00M=00L=00 = =005=00.=000=000=00.=003=005=000=002=00.=005=003=009=000=00"=00 = =00n=00a=00m=00e=00=3D=00G=00E=00N=00E=00R=00A=00T=00O=00R=00>=00=0D=00=0A= =00<=00S=00T=00Y=00L=00E=00>=00<=00/=00S=00T=00Y=00L=00E=00>=00=0D=00=0A= =00=0D=00=0A= =00<=00M=00E=00T=00A=00 = =00c=00o=00n=00t=00e=00n=00t=00=3D=00"=00M=00S=00H=00T=00M=00L=00 = =005=00.=000=000=00.=003=005=000=002=00.=005=003=009=000=00"=00 = =00n=00a=00m=00e=00=3D=00G=00E=00N=00E=00R=00A=00T=00O=00R=00>=00<=00/=00= H=00E=00A=00D=00>=00=0D=00=0A= =00<=00B=00O=00D=00Y=00 = =00b=00g=00C=00o=00l=00o=00r=00=3D=00#=00f=00f=00f=00f=00f=00f=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00&=00n=00b=00s=00p=00;=00<=00/=00D=00I=00V=00>=00=0D= =00=0A= =00<=00D=00I=00V=00 = =00s=00t=00y=00l=00e=00=3D=00"=00F=00O=00N=00T=00:=00 = =001=000=00p=00t=00 =00a=00r=00i=00a=00l=00"=00>=00-=00-=00-=00-=00-=00 = =00O=00r=00i=00g=00i=00n=00a=00l=00 =00M=00e=00s=00s=00a=00g=00e=00 = =00-=00-=00-=00-=00-=00 =00=0D=00=0A= =00<=00D=00I=00V=00 = =00s=00t=00y=00l=00e=00=3D=00"=00B=00A=00C=00K=00G=00R=00O=00U=00N=00D=00= :=00 =00#=00e=004=00e=004=00e=004=00;=00 = =00f=00o=00n=00t=00-=00c=00o=00l=00o=00r=00:=00 = =00b=00l=00a=00c=00k=00"=00>=00<=00B=00>=00F=00r=00o=00m=00:=00<=00/=00B=00= >=00 =00<=00A=00 =00=0D=00=0A= =00h=00r=00e=00f=00=3D=00"=00m=00a=00i=00l=00t=00o=00:=00f=00a=00i=00s=00= a=00l=00.=00m=00a=00q=00s=00o=00o=00d=00@=00a=00s=00c=00e=00r=00t=00i=00a= =00.=00c=00o=00m=00"=00 =00=0D=00=0A= =00t=00i=00t=00l=00e=00=3D=00f=00a=00i=00s=00a=00l=00.=00m=00a=00q=00s=00= o=00o=00d=00@=00a=00s=00c=00e=00r=00t=00i=00a=00.=00c=00o=00m=00>=00F=00a= =00i=00s=00a=00l=00 =00M=00a=00q=00s=00o=00o=00d=00<=00/=00A=00>=00 = =00<=00/=00D=00I=00V=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00<=00B=00>=00T=00o=00:=00<=00/=00B=00>=00 = =00<=00A=00 = =00h=00r=00e=00f=00=3D=00"=00m=00a=00i=00l=00t=00o=00:=00i=00e=00t=00f=00= -=00p=00k=00i=00x=00@=00i=00m=00c=00.=00o=00r=00g=00"=00 =00=0D=00=0A= =00t=00i=00t=00l=00e=00=3D=00i=00e=00t=00f=00-=00p=00k=00i=00x=00@=00i=00= m=00c=00.=00o=00r=00g=00>=00i=00e=00t=00f=00-=00p=00k=00i=00x=00@=00i=00m= =00c=00.=00o=00r=00g=00<=00/=00A=00>=00 =00<=00/=00D=00I=00V=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00<=00B=00>=00S=00e=00n=00t=00:=00<=00/=00B=00>=00 = =00T=00u=00e=00s=00d=00a=00y=00,=00 =00F=00e=00b=00r=00u=00a=00r=00y=00 = =002=004=00,=00 =002=000=000=004=00 = =001=001=00:=000=004=00<=00/=00D=00I=00V=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00<=00B=00>=00S=00u=00b=00j=00e=00c=00t=00:=00<=00/=00= B=00>=00 =00R=00e=00l=00a=00t=00i=00o=00n=00s=00h=00i=00p=00 = =00b=00e=00t=00w=00e=00e=00n=00 =00S=00A=00N=00 =00a=00n=00d=00 = =00I=00A=00N=00?=00<=00/=00D=00I=00V=00>=00<=00/=00D=00I=00V=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00<=00B=00R=00>=00<=00/=00D=00I=00V=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00H=00e=00l=00l=00o=00 = =00F=00o=00l=00k=00s=00,=00<=00/=00D=00I=00V=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00&=00n=00b=00s=00p=00;=00<=00/=00D=00I=00V=00>=00=0D= =00=0A= =00<=00D=00I=00V=00>=00I=00 =00h=00a=00v=00e=00 =00a=00 = =00q=00u=00e=00r=00y=00 =00f=00o=00r=00 = =00c=00l=00a=00r=00i=00f=00i=00c=00a=00t=00i=00o=00n=00 =00a=00n=00d=00 = =00n=00e=00e=00d=00s=00&=00n=00b=00s=00p=00;=00y=00o=00u=00r=00 = =00v=00i=00e=00w=00s=00.=00<=00/=00D=00I=00V=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00L=00e=00t=00 =00u=00s=00 = =00c=00o=00n=00s=00i=00d=00e=00r=00 =00t=00h=00e=00r=00e=00 = =00a=00r=00e=00 =00t=00w=00o=00 = =00c=00e=00r=00t=00i=00f=00i=00c=00a=00t=00e=00s=00 =00A=00 = =00a=00n=00d=00 =00B=00 =00(=00p=00a=00r=00t=00 =00o=00f=00 = =00t=00h=00e=00 =00=0D=00=0A= =00c=00e=00r=00t=00i=00f=00i=00c=00a=00t=00i=00o=00n=00 = =00p=00a=00t=00h=00)=00.=00 =00<=00/=00D=00I=00V=00>=00=0D=00=0A= =00<=00U=00L=00>=00=0D=00=0A= =00 =00 =00<=00L=00I=00>=00I=00f=00 =00A=00 =00i=00s=00 = =00i=00s=00s=00u=00e=00r=00 =00o=00f=00 =00B=00 =00a=00n=00d=00 = =00t=00h=00e=00r=00e=00 =00i=00s=00 = =00S=00u=00b=00j=00e=00c=00t=00A=00l=00t=00N=00a=00m=00e=00 = =00e=00x=00t=00e=00n=00s=00i=00o=00n=00 =00p=00r=00e=00s=00e=00n=00t=00 = =00i=00n=00 =00A=00 =00a=00n=00d=00 =00=0D=00=0A= =00 =00 =00I=00s=00s=00u=00e=00r=00A=00l=00t=00N=00a=00m=00e=00 = =00i=00n=00 =00B=00 =00a=00n=00d=00 =00v=00a=00l=00u=00e=00 =00o=00f=00 = =00S=00u=00b=00j=00e=00c=00t=00A=00l=00t=00N=00a=00m=00e=00 =00i=00n=00 = =00A=00 =00i=00s=00 =00=0D=00=0A= =00 =00 = =00"=00U=00R=00L=00=3D=00h=00t=00t=00p=00:=00/=00/=00w=00w=00w=00.=00t=00= e=00s=00t=00s=00e=00r=00v=00e=00r=00n=00a=00m=00e=00.=00c=00o=00m=00"=00.= =00 =00I=00s=00 =00t=00h=00e=00r=00e=00 =00a=00n=00y=00 = =00r=00e=00q=00u=00i=00r=00e=00m=00e=00n=00t=00 =00i=00n=00 = =00R=00F=00C=00 =002=004=005=009=00 =00o=00r=00 =00=0D=00=0A= =00 =00 =003=002=008=000=00 =00o=00r=00 =00s=00o=00m=00e=00 = =00w=00h=00e=00r=00e=00 =00e=00l=00s=00e=00 =00f=00o=00r=00 = =00I=00s=00s=00u=00e=00r=00A=00l=00t=00N=00a=00m=00e=00 = =00v=00a=00l=00u=00e=00 =00(=00i=00f=00 = =00p=00r=00e=00s=00e=00n=00t=00)=00 =00i=00n=00 =00B=00.=00 =00I=00 = =00m=00e=00a=00n=00 =00i=00n=00 =00=0D=00=0A= =00 =00 =00t=00h=00i=00s=00 =00c=00a=00s=00e=00 = =00I=00s=00s=00u=00e=00r=00A=00l=00t=00N=00a=00m=00e=00 = =00v=00a=00l=00u=00e=00 =00m=00u=00s=00t=00 = =00c=00o=00n=00t=00a=00i=00n=00 =00=0D=00=0A= =00 =00 = =00"=00U=00R=00L=00=3D=00h=00t=00t=00p=00:=00/=00/=00w=00w=00w=00.=00t=00= e=00s=00t=00s=00e=00r=00v=00e=00r=00n=00a=00m=00e=00.=00c=00o=00m=00"=00?= =00 =00=0D=00=0A= =00 =00 =00<=00L=00I=00>=00S=00i=00m=00i=00l=00a=00r=00 = =00q=00u=00e=00r=00y=00 =00f=00o=00r=00 = =00S=00u=00b=00j=00e=00c=00t=00K=00e=00y=00I=00d=00e=00n=00t=00i=00f=00i=00= e=00r=00 =00a=00n=00d=00 = =00A=00u=00t=00h=00o=00r=00i=00t=00y=00K=00e=00y=00I=00d=00e=00n=00t=00i=00= f=00i=00e=00r=00 =00t=00h=00a=00t=00 =00i=00f=00 =00=0D=00=0A= =00 =00 = =00S=00u=00b=00j=00e=00c=00t=00K=00e=00y=00I=00d=00e=00n=00t=00i=00f=00i=00= e=00r=00 =00v=00a=00l=00u=00e=00 =00i=00s=00 =00o=00f=00 = =006=000=00-=00b=00i=00t=00 =00l=00e=00n=00g=00t=00h=00 =00i=00n=00 = =00A=00.=00 =00I=00s=00 =00R=00F=00C=00 =00a=00l=00l=00o=00w=00s=00 = =00t=00o=00 =00p=00u=00t=00 =00=0D=00=0A= =00 =00 = =00A=00u=00t=00h=00o=00r=00i=00t=00y=00K=00e=00y=00I=00d=00e=00n=00t=00i=00= f=00i=00e=00r=00 =00v=00a=00l=00u=00e=00 =00o=00f=00 = =001=006=000=00-=00b=00i=00t=00 =00l=00e=00n=00g=00t=00h=00 =00i=00n=00 = =00B=00?=00<=00/=00L=00I=00>=00<=00/=00U=00L=00>=00=0D=00=0A= =00<=00D=00I=00V=00>=00R=00e=00g=00a=00r=00d=00s=00,=00<=00/=00D=00I=00V=00= >=00=0D=00=0A= =00<=00D=00I=00V=00>=00F=00a=00i=00s=00a=00l=00 = =00M=00a=00q=00s=00o=00o=00d=00<=00/=00D=00I=00V=00>=00<=00/=00B=00O=00D=00= Y=00>=00<=00/=00H=00T=00M=00L=00>=00=0D=00=0A= =00 ------=_NextPart_000_0042_01C43BFB.C96DC1A0-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4FHGblf029735; Sat, 15 May 2004 10:16:37 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4FHGb47029734; Sat, 15 May 2004 10:16:37 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4FHGaPN029728 for <ietf-pkix@imc.org>; Sat, 15 May 2004 10:16:36 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4FHGcN04929; Sat, 15 May 2004 19:16:38 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Sat, 15 May 2004 19:16:38 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4FHGbN27237; Sat, 15 May 2004 19:16:37 +0200 (MEST) Date: Sat, 15 May 2004 19:16:37 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405151716.i4FHGbN27237@chandon.edelweb.fr> To: Peter.Sylvester@edelweb.fr, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > [TF] Nice ideal, but Administrators of all descriptions make security > related decisions. As long as we continue with innovation prohibiting companies. :-) > [TF] Having the client be unaware that relay happened still seems a > desirable simplification. All it takes is the server to process the rely > reply. I am OK with including an SVVP response from another server for > third party accountability, but not if the client is expected to process > the embedded SCVP response. I tend to agree. depends somewhat on the definition of 'process', clients also do not process OCSP responses or time stamps, do they? Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EJRcqC055534; Fri, 14 May 2004 12:27:38 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4EJRcch055533; Fri, 14 May 2004 12:27:38 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EJRbM6055527 for <ietf-pkix@imc.org>; Fri, 14 May 2004 12:27:37 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Fri, 14 May 2004 12:27:41 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Fri, 14 May 2004 12:27:41 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Fri, 14 May 2004 12:27:41 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Fri, 14 May 2004 12:27:40 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F02963322@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ52FzIFTMYA/QuTnmfrVxJuCeKgAAD2+Yg From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Peter Sylvester" <Peter.Sylvester@edelweb.fr> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 14 May 2004 19:27:41.0555 (UTC) FILETIME=[863D6430:01C439E9] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4EJRbM6055528 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> * -----Original Message----- * From: Peter Sylvester [mailto:Peter.Sylvester@edelweb.fr] * Sent: Friday, May 14, 2004 10:25 AM * To: Peter.Sylvester@edelweb.fr; Trevor Freeman * Cc: ietf-pkix@imc.org * Subject: RE: SCVP & validation policies * * * > * > [TF] One administrator can defile a policy and that policy can consist * > of n parameters living in some kind of object (XML in a file store, * > attributes in a directory etc). Other administrators do not need to be * > intimately aware of the content of that object to use object, they just * > need a pointer which they can use. Therefore the number of parameters in * > the object is irrelevant. There is generally a perception that any * > policy change has to be applied in a reasonable timeframe therefore any * > policy management system has regularly check for updates therefore * > frequency of updates tends not to be a issue either. However I really * > don't expect SCVP policy of any type to have significant churn. * * Policy changes are not to be applied by any workflow application * administrator, * i.e. client administrator. They are made by the security administrator. [TF] Nice ideal, but Administrators of all descriptions make security related decisions. * * > * But since you have responsed positively to Denis, I may * > * not even understand what you are saying. * * Well, since you have already agreed that it is posssible just to use a * policy oid * there is no need to discuss further whether the alternative is good, * useful, practical or whatever. * * > * * > * Since three drafts, the whole text is unimplementable, * > * i.e. almost a year, ...), so the following list of some * > * (to me) open questions is not necessariy complete. * > * * > * - in what way a relay scvp can include an answer from another scvp * > * server? please allow to have a place for this. Just provide for * > * example a 'content-info' as a placeholder for time stamps, scvp * > * responses. * * > [TF] Up to now one of our objective was to hide any complexity that may * > arise out of relay. There is also no guaranteed correlation between * > relayed requests. A client may make a DPV request and the server may * > intern make a DPD request to another SCVP server so embedding the DPD * > response to he DPV request is on no value. * * inverse logic A ==> B is not the same as not A ==> not B * the fact that you may find one case where is doesn't make sense * to include a response, doesn't mean that it never makes sense. * And in the given example, in particular when there is no * correlation, who can the realy prove that it has provided * his a particular response based on whatever other response? [TF] Having the client be unaware that relay happened still seems a desirable simplification. All it takes is the server to process the rely reply. I am OK with including an SVVP response from another server for third party accountability, but not if the client is expected to process the embedded SCVP response. * * * 'our objective'? RFC 3379 says: * * The DPV request MUST allow the client to request that the server * include in its response additional information which will allow * relying parties not trusting the DPV server to be confident that the * certificate validation has correctly been performed. Such * information may (not necessarily exclusively) consist of a * certification path, revocation status information from authorized CRL * issuers or authorized OCSP responders, revocation status information * from CRL issuers or OCSP responders trusted under the validation * policy, time-stamp tokens from TSAs responders trusted under the * validation policy, or a DPV response from a DPV server that is * XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX * * trusted under the validation policy. When the certificate is valid * according to the validation policy, the server MUST, upon request, * include that information in the response. However, the server MAY * omit that information when the certificate is invalid or when it * cannot determine the validity. * * > * * > * - Is the Nonce field the response to the following req: * > * * > * The DPV server MUST be able, upon request, copy a text field * > provided * > * by the client into the DPV response. As an example, this field may * > * relate to the nature or reason for the DPV query. * * ???? no answer? [TF] The nonce is there to bind the request to the response. * * > * * > * - I'd like to know what others think about my compromise proposal * > * for requestor/responder. * > * * ???? no answer? * * > * - I have some feeling that there are at least some different * > * interpretation * > * of policies and object identifiers validation algoritithms, etc. * > * * > * - OIDs used for a configuration * > * - oids to reference some globally recognised rules * > * - oids to define sysntaxes of validation algorihms. * > [TF] Yes I have been thinking that an OID alone is to restrictive. * * You already have for example a validation algorithm OID for * name validation (like the qulifiers in Policyinfo) * plus the required necessary parameters together with * the configuration policy OID. * * > Others may want to use a URL to point to a policy so we need a couple of * > built in choices plus an extension mechanism for other to define other * > reference types. * * More than the qualifier logic in the PolicyInfo? * * > * * > * - It is not clear to me whether for example the name matching * > algorithm * > * is at the right level of abstraction. I'd rather have an * > * algorithm that say: 'Can this cert be used for this keypurpose, * > * for the following names.' It maybe not useful to * > * burden the client with the details of adding keyusage and * > * compatible extendedkeyusage. * > [TF] I was specially asked to do the name matching example because it is * > widely used today. I am sure others can submit other policy example as * > separate IDs. * * I agree fully that the there is some good purpose of this kind of * algorithm. * I am just wondering whether the EXAMPLE specification is sufficiently * well defined. Looks like may ask for three quarks with * keyusage, extendedkeyusage and validation algorithm instead * of asking for some more obvious atomic things. * * If I am an ssl client or server, I simply don't care whether the * actual certificate has whatever key usages or so, I just want a yes/no * answer and maybe a path for logging. I just give the set of * names and certs, * * Do we need to distinguish email-protection certs for signing and * encryption? [TF] If you want to provide another example you can do so in another draft. * Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EHvneC048878; Fri, 14 May 2004 10:57:49 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4EHvnCk048877; Fri, 14 May 2004 10:57:49 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from e5.ny.us.ibm.com (e5.ny.us.ibm.com [32.97.182.105]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EHvmY7048858 for <ietf-pkix@imc.org>; Fri, 14 May 2004 10:57:48 -0700 (PDT) (envelope-from tgindin@us.ibm.com) Received: from northrelay04.pok.ibm.com (northrelay04.pok.ibm.com [9.56.224.206]) by e5.ny.us.ibm.com (8.12.10/8.12.2) with ESMTP id i4EHvjYd644888; Fri, 14 May 2004 13:57:45 -0400 Received: from d01ml062.pok.ibm.com (d01av02.pok.ibm.com [9.56.224.216]) by northrelay04.pok.ibm.com (8.12.10/NCO/VER6.6) with ESMTP id i4EHwI2q110506; Fri, 14 May 2004 13:58:19 -0400 To: "Al Arsenault" <aarsenau@bbn.com> Cc: "Anders Rundgren" <anders.rundgren@telia.com>, "David P. Kemp" <dpkemp@missi.ncsc.mil>, ietf-pkix@imc.org MIME-Version: 1.0 Subject: RE: I-D ACTION:draft-ietf-pkix-pi-08.txt X-Mailer: Lotus Notes Release 5.0.11 July 24, 2002 From: Tom Gindin <tgindin@us.ibm.com> Message-ID: <OF7AEF5A87.EE169431-ON85256E94.0058BE8F-85256E94.0062A989@us.ibm.com> Date: Fri, 14 May 2004 13:57:41 -0400 X-MIMETrack: Serialize by Router on D01ML062/01/M/IBM(Release 6.0.2CF2 HFB2|May 5, 2004) at 05/14/2004 13:57:44, Serialize complete at 05/14/2004 13:57:44 Content-Type: text/plain; charset="us-ascii" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Al: The serialNumber attribute is not expected to contain the certificate serial number. In its original definition it was supposed to contain the serial number of a piece of physical equipment, and it's defined to be a string rather than an integer like the certificate serial number. There was a lot of discussion about which field to use for national identity numbers back in the fall of '99 as part of QC (see the thread of http://www.imc.org/ietf-pkix/old-archive-99/msg03041.html for an example), and the use is reflected in RFC 3739 section 3.1.2 (and also in the corresponding section of RFC 3039). IMHO, PI is better than this use of serialNumber, but serialNumber is not as unacceptable as it would be if it were supposed to match the certificate serial number. Tom Gindin "Al Arsenault" <aarsenau@bbn.com> Sent by: owner-ietf-pkix@mail.imc.org 05/12/2004 09:30 PM To: "Anders Rundgren" <anders.rundgren@telia.com>, "David P. Kemp" <dpkemp@missi.ncsc.mil>, <ietf-pkix@imc.org> cc: Subject: RE: I-D ACTION:draft-ietf-pkix-pi-08.txt > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org]On Behalf Of Anders Rundgren > Sent: Wednesday, May 12, 2004 5:01 PM > To: David P. Kemp; ietf-pkix@imc.org > Subject: Re: I-D ACTION:draft-ietf-pkix-pi-08.txt > > > > >It is my hope that PI will become an RFC in the near future, so > >that certificates (from an un-named large PKI :-) that currently > >handle PIs by munging them into Common Name (e.g., > >CN="Kemp.David.P.0514101404") will have a saner alternative. > > The de-facto standard, already engraved in *millions* of certs is > putting 0514101404 in serialNumber. > > The problem with that is that "0514101404" is not expected to change; it's a number which stays with you as long as you're an employee of the company, or as long as you're a customer, or even all of your life. The serialNumber is expected to change every time the certificate changes - commonly, when a certificate expires once a year, the new certificate has a new serialNumber. It's considered seriously bad form to issue a "renewed" certificate with a different validity period and the same serialNumber. Thus, those who put 0514101404 in "serialNumber" are just setting themselves up for trouble later on down the road. (And as I've pointed out numerous times before, I know of organizations that put the equivalent of "0514101404" in subjectAltName, as a DNS Name! It's certainly possible, but that doesn't make it the "right thing" to do. Al Arsenault Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EHOf0i045445; Fri, 14 May 2004 10:24:41 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4EHOfra045444; Fri, 14 May 2004 10:24:41 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EHOdML045437 for <ietf-pkix@imc.org>; Fri, 14 May 2004 10:24:40 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4EHOgN21923; Fri, 14 May 2004 19:24:42 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Fri, 14 May 2004 19:24:42 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4EHOfK24575; Fri, 14 May 2004 19:24:41 +0200 (MEST) Date: Fri, 14 May 2004 19:24:41 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405141724.i4EHOfK24575@chandon.edelweb.fr> To: Peter.Sylvester@edelweb.fr, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > > [TF] One administrator can defile a policy and that policy can consist > of n parameters living in some kind of object (XML in a file store, > attributes in a directory etc). Other administrators do not need to be > intimately aware of the content of that object to use object, they just > need a pointer which they can use. Therefore the number of parameters in > the object is irrelevant. There is generally a perception that any > policy change has to be applied in a reasonable timeframe therefore any > policy management system has regularly check for updates therefore > frequency of updates tends not to be a issue either. However I really > don't expect SCVP policy of any type to have significant churn. Policy changes are not to be applied by any workflow application administrator, i.e. client administrator. They are made by the security administrator. > * But since you have responsed positively to Denis, I may > * not even understand what you are saying. Well, since you have already agreed that it is posssible just to use a policy oid there is no need to discuss further whether the alternative is good, useful, practical or whatever. > * > * Since three drafts, the whole text is unimplementable, > * i.e. almost a year, ...), so the following list of some > * (to me) open questions is not necessariy complete. > * > * - in what way a relay scvp can include an answer from another scvp > * server? please allow to have a place for this. Just provide for > * example a 'content-info' as a placeholder for time stamps, scvp > * responses. > [TF] Up to now one of our objective was to hide any complexity that may > arise out of relay. There is also no guaranteed correlation between > relayed requests. A client may make a DPV request and the server may > intern make a DPD request to another SCVP server so embedding the DPD > response to he DPV request is on no value. inverse logic A ==> B is not the same as not A ==> not B the fact that you may find one case where is doesn't make sense to include a response, doesn't mean that it never makes sense. And in the given example, in particular when there is no correlation, who can the realy prove that it has provided his a particular response based on whatever other response? 'our objective'? RFC 3379 says: The DPV request MUST allow the client to request that the server include in its response additional information which will allow relying parties not trusting the DPV server to be confident that the certificate validation has correctly been performed. Such information may (not necessarily exclusively) consist of a certification path, revocation status information from authorized CRL issuers or authorized OCSP responders, revocation status information from CRL issuers or OCSP responders trusted under the validation policy, time-stamp tokens from TSAs responders trusted under the validation policy, or a DPV response from a DPV server that is XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX trusted under the validation policy. When the certificate is valid according to the validation policy, the server MUST, upon request, include that information in the response. However, the server MAY omit that information when the certificate is invalid or when it cannot determine the validity. > * > * - Is the Nonce field the response to the following req: > * > * The DPV server MUST be able, upon request, copy a text field > provided > * by the client into the DPV response. As an example, this field may > * relate to the nature or reason for the DPV query. ???? no answer? > * > * - I'd like to know what others think about my compromise proposal > * for requestor/responder. > * ???? no answer? > * - I have some feeling that there are at least some different > * interpretation > * of policies and object identifiers validation algoritithms, etc. > * > * - OIDs used for a configuration > * - oids to reference some globally recognised rules > * - oids to define sysntaxes of validation algorihms. > [TF] Yes I have been thinking that an OID alone is to restrictive. You already have for example a validation algorithm OID for name validation (like the qulifiers in Policyinfo) plus the required necessary parameters together with the configuration policy OID. > Others may want to use a URL to point to a policy so we need a couple of > built in choices plus an extension mechanism for other to define other > reference types. More than the qualifier logic in the PolicyInfo? > * > * - It is not clear to me whether for example the name matching > algorithm > * is at the right level of abstraction. I'd rather have an > * algorithm that say: 'Can this cert be used for this keypurpose, > * for the following names.' It maybe not useful to > * burden the client with the details of adding keyusage and > * compatible extendedkeyusage. > [TF] I was specially asked to do the name matching example because it is > widely used today. I am sure others can submit other policy example as > separate IDs. I agree fully that the there is some good purpose of this kind of algorithm. I am just wondering whether the EXAMPLE specification is sufficiently well defined. Looks like may ask for three quarks with keyusage, extendedkeyusage and validation algorithm instead of asking for some more obvious atomic things. If I am an ssl client or server, I simply don't care whether the actual certificate has whatever key usages or so, I just want a yes/no answer and maybe a path for logging. I just give the set of names and certs, Do we need to distinguish email-protection certs for signing and encryption? Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EGZTkt040569; Fri, 14 May 2004 09:35:29 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4EGZTP7040568; Fri, 14 May 2004 09:35:29 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EGZSZl040562 for <ietf-pkix@imc.org>; Fri, 14 May 2004 09:35:28 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Fri, 14 May 2004 09:35:31 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Fri, 14 May 2004 09:35:31 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Fri, 14 May 2004 09:35:31 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Fri, 14 May 2004 09:35:30 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F029631F0@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ5nekTcjzm3b/iRXuObxRCkpNsmgAMIXUw From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Peter Sylvester" <Peter.Sylvester@edelweb.fr> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 14 May 2004 16:35:31.0867 (UTC) FILETIME=[794416B0:01C439D1] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4EGZTZl040563 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> See below * -----Original Message----- * From: Peter Sylvester [mailto:Peter.Sylvester@edelweb.fr] * Sent: Friday, May 14, 2004 3:26 AM * To: Peter.Sylvester@edelweb.fr; Trevor Freeman * Cc: ietf-pkix@imc.org * Subject: RE: SCVP & validation policies * * > Hi Peter, * > The problem of having to push policy to clients of all descriptions is a * > well understood problem set. The infrastructure to do that is the main * > overhead. The number of parameters associated with that process is not a * > significant overhead. * * You don't push parameters that are subject to change, you need to handle * updates, if you configure one id, the problem of updates is gone. * * Whenever you can remove one parameter, you significantly reduce * the potential problem space (about divide it at least by 2) [TF] One administrator can defile a policy and that policy can consist of n parameters living in some kind of object (XML in a file store, attributes in a directory etc). Other administrators do not need to be intimately aware of the content of that object to use object, they just need a pointer which they can use. Therefore the number of parameters in the object is irrelevant. There is generally a perception that any policy change has to be applied in a reasonable timeframe therefore any policy management system has regularly check for updates therefore frequency of updates tends not to be a issue either. However I really don't expect SCVP policy of any type to have significant churn. * * But since you have responsed positively to Denis, I may * not even understand what you are saying. * * Since three drafts, the whole text is unimplementable, * i.e. almost a year, ...), so the following list of some * (to me) open questions is not necessariy complete. * * - in what way a relay scvp can include an answer from another scvp * server? please allow to have a place for this. Just provide for * example a 'content-info' as a placeholder for time stamps, scvp * responses. [TF] Up to now one of our objective was to hide any complexity that may arise out of relay. There is also no guaranteed correlation between relayed requests. A client may make a DPV request and the server may intern make a DPD request to another SCVP server so embedding the DPD response to he DPV request is on no value. * * - Is the Nonce field the response to the following req: * * The DPV server MUST be able, upon request, copy a text field provided * by the client into the DPV response. As an example, this field may * relate to the nature or reason for the DPV query. * * - I'd like to know what others think about my compromise proposal * for requestor/responder. * * - I have some feeling that there are at least some different * interpretation * of policies and object identifiers validation algoritithms, etc. * * - OIDs used for a configuration * - oids to reference some globally recognised rules * - oids to define sysntaxes of validation algorihms. [TF] Yes I have been thinking that an OID alone is to restrictive. Others may want to use a URL to point to a policy so we need a couple of built in choices plus an extension mechanism for other to define other reference types. * * - It is not clear to me whether for example the name matching algorithm * is at the right level of abstraction. I'd rather have an * algorithm that say: 'Can this cert be used for this keypurpose, * for the following names.' It maybe not useful to * burden the client with the details of adding keyusage and * compatible extendedkeyusage. [TF] I was specially asked to do the name matching example because it is widely used today. I am sure others can submit other policy example as separate IDs. * * Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EF1cab029538; Fri, 14 May 2004 08:01:38 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4EF1ckY029537; Fri, 14 May 2004 08:01:38 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from stingray.missi.ncsc.mil (stingray.missi.ncsc.mil [144.51.50.20]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EF1biR029517 for <ietf-pkix@imc.org>; Fri, 14 May 2004 08:01:38 -0700 (PDT) (envelope-from DPKemp@missi.ncsc.mil) Message-ID: <200405141427.i4EER9hs010150@stingray.missi.ncsc.mil> Date: Fri, 14 May 2004 11:01:22 -0400 From: "David P. Kemp" <dpkemp@missi.ncsc.mil> User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Anders Rundgren <anders.rundgren@telia.com> CC: ietf-pkix@imc.org Subject: Re: I-D ACTION:draft-ietf-pkix-pi-08.txt References: <005701c3e08e$9b392fe0$1400a8c0@augustcellars.local> <00c701c3e121$0ae3af90$0500a8c0@arport> <4017C963.8060600@bull.net> <200405121708.i4CH7dim022214@stingray.missi.ncsc.mil> <001001c43864$e11bb130$0500a8c0@arport> <200405122138.i4CLchhk006004@stingray.missi.ncsc.mil> <00c001c43924$d691c5e0$0500a8c0@arport> In-Reply-To: <00c001c43924$d691c5e0$0500a8c0@arport> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 14 May 2004 15:01:28.0728 (UTC) FILETIME=[55B16980:01C439C4] Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> The problem is that most PKIs do not support SN attributes managed as in your "real world" example, they support SN as disambiguator, as cited (not invented) by Denis. Some tightly-managed PKIs (government and commercial) do support SN attributes that are permanent references to a user; in fact some do not even populate commonName and use SN as the only user identifier. In an ideal world, all PKIs that populated the SN naming attribute would put a Permanent Identifier in it. This would be in partial alignment with the wording of X.520, which says: "The Serial Number attribute type specifies an identifier, the serial number of an object." If you regard people as objects, then it is pretty clear that the "serial number" of a person is something equivalent to DNA; it doesn't change when the person changes names, and it doesn't change depending on which CA issues the person a certificate. However, there are strong privacy reasons for not putting a DNA equivalent into certificates, and that's never going to happen. So an alternative like the PI RFC is required. The requirements are: 1) mechanical applications must be able to tell reliably and unambiguously (i.e., without installation-time configuration) if a cert contains a PI, and if so, how to extract it. 2) the PI doesn't change over the "functional lifetime" of the person (the time the person has an account at an institution, which can range up to most of a person's natural lifetime). 3) the PI is assigned by a naming authority that is not in general a CA. One CA must be able to issue certs for multiple institutions that create user accounts, and multiple CAs must be able to issue certs for an account at a single such institution (and are recognized as all referring to the same account by the applications of requirement #1). If you can come up with a specification that uses the SN attribute and satisfies each of these three requirements, I will happily favor it over the Pinkas/Gindin proposal. But I don't believe such a specification is possible. Dave Anders Rundgren wrote: > Now assume that you don't have an e-mail address > to stuff into the DN but you do have some kind of > unique identifier. Why put this somewhere else than > in the DN? An example shows how ugly it gets: > > Denis' world: > DN: CN=John Smith, serialNumber=3, O=ACME, C=US > PI: OID.6.4.3.5.6.6.6.43.3, 05725277 > > The real world: > DN: CN=John Smith, serialNumber=05725277, O=ACME, C=US > The name space is given by the CA policy. Implicit or explicit. > > Although esthetics is maybe not priority #1, I believe that > "inventing" arbitrary disambiguing numbers when there is > a perfectly valid number to use is a *very* hard sale. > However, most of the PI motivation goes away if the unique > identifier is used as disambiguer and that's the problem in > a nutshell. Can you fix this dilemma, please tell us how! > > > A solution for DoD > ---------------------- > > Not that I believe you want my advice but here is a suggestion > for DoD: > > CN=David Kemp, serialNumber=05725277, DC=staffregistry1, DC=dod, DC=mil > > Using this scheme you get globally unique identifiers (GUIDs) > which is a much more needed and popular thing than finding > new ways to complicate RP applications. Why would a > certificate ever contain more "name" info than this (+ S/MIME > support in applicable cases)? > > > rgds > Anders > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EC9u3V013061; Fri, 14 May 2004 05:09:56 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4EC9uf9013060; Fri, 14 May 2004 05:09:56 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EC9seU013048 for <ietf-pkix@imc.org>; Fri, 14 May 2004 05:09:55 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms20.chttl.com.tw (ms20 [10.144.2.120]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i4EC97hO022739; Fri, 14 May 2004 20:09:08 +0800 Received: (from root@localhost) by ms20.chttl.com.tw (8.12.10/8.12.10) id i4EC975j017003; Fri, 14 May 2004 20:09:07 +0800 Received: from wcwang ([10.144.133.79]) by ms20.chttl.com.tw (8.12.10/8.12.10) with SMTP id i4EC94pF016797; Fri, 14 May 2004 20:09:04 +0800 Message-ID: <026301c439ac$3fc72370$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Stefan Santesson" <stefans@microsoft.com>, "Tom Gindin" <tgindin@us.ibm.com> Cc: "Santoni Adriano" <adriano.santoni@actalis.it>, <ietf-pkix@imc.org>, "Stephen Kent" <kent@bbn.com>, <pgut001@cs.auckland.ac.nz>, "Steve Hanna" <Steve.Hanna@sun.com> References: <0C3042E92D8A714783E2C44AB9936E1DFBAACD@EUR-MSG-03.europe.corp.microsoft.com> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String inencodingRDNs Date: Fri, 14 May 2004 20:09:01 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I agree that if all RP implementations adopt X.500-like name matching rules, then there is no need to issue name encoding rollover certificates. Even so, we still need to clarify whether a certificate with same issuer and subject names but in different encoding should be interpreted as self-issued, because there is no way to prohibit CAs from issuing such certificates (e.g., CA1-CA1-p-u-cert I mentioned in my recent post). Wen-Cheng Wang > > The whole encoding rollover certificate issue seems to be in error and > should probably be deprecated. > > Here is the reason: > > Either you regard names of different encoding a name match or you don't. > > X.509 matching rules for attributes caseIgnoreSyntax suggest that names > in different encoding are a match if they can be concluded to express > the same name. > > RFC 3280 say that this is OK to implement, but you are allowed to > disregard this and treat names of different encoding as no match > > So.... If this IS accepted as a name match, then you don't need the > rollover cert. It has no function since the path now works without it. > > > The paradox is then the following: > If we fix name matching in RFC 3280, then there IS NO use of the > encoding rollover cert any more (path works without it). If we on the > other hand DON't fix this, then this certificate will neither be a self > issued certificate by definition and thus it will in some cases break > path validation. > > > Stefan Santesson > Program Manager, Windows Security > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EAQL2a003574; Fri, 14 May 2004 03:26:21 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4EAQLll003573; Fri, 14 May 2004 03:26:21 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4EAQJI9003566 for <ietf-pkix@imc.org>; Fri, 14 May 2004 03:26:20 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4EAQKN16487; Fri, 14 May 2004 12:26:20 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Fri, 14 May 2004 12:26:20 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4EAQJk23215; Fri, 14 May 2004 12:26:19 +0200 (MEST) Date: Fri, 14 May 2004 12:26:19 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405141026.i4EAQJk23215@chandon.edelweb.fr> To: Peter.Sylvester@edelweb.fr, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > Hi Peter, > The problem of having to push policy to clients of all descriptions is a > well understood problem set. The infrastructure to do that is the main > overhead. The number of parameters associated with that process is not a > significant overhead. You don't push parameters that are subject to change, you need to handle updates, if you configure one id, the problem of updates is gone. Whenever you can remove one parameter, you significantly reduce the potential problem space (about divide it at least by 2) But since you have responsed positively to Denis, I may not even understand what you are saying. Since three drafts, the whole text is unimplementable, i.e. almost a year, ...), so the following list of some (to me) open questions is not necessariy complete. - in what way a relay scvp can include an answer from another scvp server? please allow to have a place for this. Just provide for example a 'content-info' as a placeholder for time stamps, scvp responses. - Is the Nonce field the response to the following req: The DPV server MUST be able, upon request, copy a text field provided by the client into the DPV response. As an example, this field may relate to the nature or reason for the DPV query. - I'd like to know what others think about my compromise proposal for requestor/responder. - I have some feeling that there are at least some different interpretation of policies and object identifiers validation algoritithms, etc. - OIDs used for a configuration - oids to reference some globally recognised rules - oids to define sysntaxes of validation algorihms. - It is not clear to me whether for example the name matching algorithm is at the right level of abstraction. I'd rather have an algorithm that say: 'Can this cert be used for this keypurpose, for the following names.' It maybe not useful to burden the client with the details of adding keyusage and compatible extendedkeyusage. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E9xs89000282; Fri, 14 May 2004 02:59:54 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4E9xsQg000281; Fri, 14 May 2004 02:59:54 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E9xqha000259 for <ietf-pkix@imc.org>; Fri, 14 May 2004 02:59:53 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4E9xkN16205; Fri, 14 May 2004 11:59:46 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Fri, 14 May 2004 11:59:46 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4E9xj222852; Fri, 14 May 2004 11:59:45 +0200 (MEST) Date: Fri, 14 May 2004 11:59:45 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405140959.i4E9xj222852@chandon.edelweb.fr> To: Denis.Pinkas@bull.net, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > > Hi Denis, > As I said previous an OID is one of the possible options, but there is > no consensus to make this mandatory so SCVP will continue to support > both options. I also don't see consensus that it is mandatory that a > SCVP server will always map a request back to the OID. I can live with this. > I will be posting SCVP 15 shortly so you can review the latest draft. Given the previous discussion, I'd like to see a possibility that the request can be as simple as just sending in the certificate. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E94dOT084016; Fri, 14 May 2004 02:04:39 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4E94dOW084015; Fri, 14 May 2004 02:04:39 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-dub.microsoft.com (mail-dub.microsoft.com [213.199.128.160]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E94c4s083959 for <ietf-pkix@imc.org>; Fri, 14 May 2004 02:04:38 -0700 (PDT) (envelope-from stefans@microsoft.com) Received: from dub-imc-01.europe.corp.microsoft.com ([65.53.196.22]) by mail-dub.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Fri, 14 May 2004 10:04:21 +0100 Received: from 65.53.196.22 by dub-imc-01.europe.corp.microsoft.com (InterScan E-Mail VirusWall NT); Fri, 14 May 2004 10:04:21 +0100 Received: from EUR-MSG-03.europe.corp.microsoft.com ([65.53.192.44]) by dub-imc-01.europe.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Fri, 14 May 2004 10:04:21 +0100 x-mimeole: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String inencodingRDNs Date: Fri, 14 May 2004 10:03:24 +0100 Message-ID: <0C3042E92D8A714783E2C44AB9936E1DFBAACD@EUR-MSG-03.europe.corp.microsoft.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: R: I: R: RFC3280: doubt on the use of UTF8String inencodingRDNs Thread-Index: AcQ5jSVlwhgx8O2TT+qmzLBm+HVi/AAAkdGQ From: "Stefan Santesson" <stefans@microsoft.com> To: "Wen-Cheng Wang" <wcwang@cht.com.tw>, "Tom Gindin" <tgindin@us.ibm.com> Cc: "Santoni Adriano" <adriano.santoni@actalis.it>, <ietf-pkix@imc.org>, "Stephen Kent" <kent@bbn.com>, <pgut001@cs.auckland.ac.nz>, "Steve Hanna" <Steve.Hanna@sun.com> X-OriginalArrivalTime: 14 May 2004 09:04:21.0150 (UTC) FILETIME=[71DC7BE0:01C43992] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4E94d4s084007 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> The whole encoding rollover certificate issue seems to be in error and should probably be deprecated. Here is the reason: Either you regard names of different encoding a name match or you don't. X.509 matching rules for attributes caseIgnoreSyntax suggest that names in different encoding are a match if they can be concluded to express the same name. RFC 3280 say that this is OK to implement, but you are allowed to disregard this and treat names of different encoding as no match So.... If this IS accepted as a name match, then you don't need the rollover cert. It has no function since the path now works without it. The paradox is then the following: If we fix name matching in RFC 3280, then there IS NO use of the encoding rollover cert any more (path works without it). If we on the other hand DON't fix this, then this certificate will neither be a self issued certificate by definition and thus it will in some cases break path validation. Stefan Santesson Program Manager, Windows Security > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On Behalf Of Wen-Cheng Wang > Sent: den 14 maj 2004 09:21 > To: Tom Gindin > Cc: Santoni Adriano; ietf-pkix@imc.org; Stephen Kent; > pgut001@cs.auckland.ac.nz; Stefan Santesson; Steve Hanna > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String > inencodingRDNs > > > Tom, > > > > > The bottom line of the statements you have quoted is that some > > fully RFC 3280-compliant RP implementations will treat all rollover > > certificates as self-issued, some will treat no rollover certificates as > > self-issued, and some will treat many rollover certificates as self- > issued > > but some as not. Thus, whether a given rollover certificate is deemed > to > > be self-issued by a compliant RP is implementation dependent - probably > a > > worse situation than "rollover certificates are not interpreted as > > self-issued". > Yes, the current situation is that some RP implementations will treat > name rollover certificates (if any) as self-issued but some will not. > The root cause of inconsistency on interpreting self-issued certificates > among different implementations is the deficiency of name matching > rules in RFC 3280 and the X.509 standard. There can be a debate > on whether name rollover certificates should be interpreted as > self-issued or not. However, the current situation is the current > situation. The situation will not change unless the deficiency in > RFC 3280 and the X.509 standard is mended and hopefully > all RP implementations will migrate to a consistent way of > interpreting self-issued certificates. The point is that even if we > finally > agree that "name rollover certificates should not interpreted as > self-issued", we still need to revise or amend RFC 3280 and the > X.509 standard to make it clear. > > > IMHO, it would really help if we had a list of CA's who encode > > their current DN's using BMPString with characters in the surrogate > range, > > and whether they are compliant to UTF-16 or not, and a list of CA's who > > encode their current DN's using T61String. Most other character > > conversions under the ASN.1 standard are simple algorithms. > > > That might be an interesting investigation, but I do not think it will > help for resolving the inconsistency on interpreting self-issued > certificates. As I said above, the root cause is the deficiency of > name matching rules. Paying more attention to mend the deficiency > might be more helpful. If we have name matching rules that can take > care of name string in different character sets and/or in different > encoding, it will not only resolve the inconsistency on interpreting > self-issued certificates but also will reduce the needs of issuing > name rollover certificates. > > Wen-Cheng Wang Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E8lnZ9077811; Fri, 14 May 2004 01:47:49 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4E8ln9C077810; Fri, 14 May 2004 01:47:49 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from smtp-ft4.fr.colt.net (smtp-ft4.fr.colt.net [213.41.78.203]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E8llBW077772 for <ietf-pkix@imc.org>; Fri, 14 May 2004 01:47:48 -0700 (PDT) (envelope-from jmdesp@free.fr) Received: from free.fr (host.106.92.68.195.rev.coltfrance.com [195.68.92.106]) by smtp-ft4.fr.colt.net with ESMTP id i4E8lhH25727 for <ietf-pkix@imc.org>; Fri, 14 May 2004 10:47:43 +0200 Message-ID: <40A487E4.1060409@free.fr> Date: Fri, 14 May 2004 10:48:36 +0200 From: Jean-Marc Desperrier <jmdesp@free.fr> User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:) Gecko/20040226 X-Accept-Language: en-us, en MIME-Version: 1.0 To: ietf-pkix@imc.org Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encodingRDNs References: <OFE117653C.B8EE2880-ON85256E92.003B8A8C-85256E94.0000B926@us.ibm.com> In-Reply-To: <OFE117653C.B8EE2880-ON85256E92.003B8A8C-85256E94.0000B926@us.ibm.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Tom Gindin wrote: > IMHO, it would really help if we had a list of CA's who encode >their current DN's using BMPString > Probably not many. This might have happened mostly with some CA having left a few requests using BMPString slip in. >with characters in the surrogate range, >and whether they are compliant to UTF-16 or not, > I certainly believe nobody did that. Is it required to precise that UTF8String is the only reliable way to encode surrogate/non BMP unicode characters ? >and a list of CA's who >encode their current DN's using T61String. > The bug database of Mozilla enables to localize some of them, for example : E = ca@neam.de, CN = neam Gesellschaft für Kommunikationslösungen mbH, OU = Security, O = neam CA, C = DE (this is in T61) Most of those CAs are clearly not used for certificates that were actually intended to be publicly distributed (thus did not care too much about interoperability problems). There is in that database a very interesting case of a CA with a UTF8 name that issued a cert with a BMPString name : You can download that from this link : http://bugzilla.mozilla.org/attachment.cgi?id=109214&action=view I think this certainly demonstrate a CA wishing to switch to the use of UTF8String, but that unfortunately also allowed it's software to issue BMPString when it allowed "unicode" in certificate names. Implementers should beware of such situations. > Most other character >conversions under the ASN.1 standard are simple algorithms. > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E7LgcP049287; Fri, 14 May 2004 00:21:42 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4E7LgRO049286; Fri, 14 May 2004 00:21:42 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from gate3.chttl.com.tw (gate3.chttl.com.tw [203.75.28.115]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E7LTAA049171 for <ietf-pkix@imc.org>; Fri, 14 May 2004 00:21:35 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms8.chttl.com.tw (ms8 [10.144.2.118]) by gate3.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i4E7KngJ011080; Fri, 14 May 2004 15:20:49 +0800 Received: (from root@localhost) by ms8.chttl.com.tw (8.12.10/8.12.10) id i4E7KmOM029785; Fri, 14 May 2004 15:20:48 +0800 Received: from wcwang ([10.144.133.79]) by ms8.chttl.com.tw (8.12.10/8.12.10) with SMTP id i4E7KjuU029681; Fri, 14 May 2004 15:20:46 +0800 Message-ID: <017b01c43983$f9202930$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Tom Gindin" <tgindin@us.ibm.com> Cc: "Santoni Adriano" <adriano.santoni@actalis.it>, <ietf-pkix@imc.org>, "Stephen Kent" <kent@bbn.com>, <pgut001@cs.auckland.ac.nz>, "Stefan Santesson" <stefans@microsoft.com>, "Steve Hanna" <Steve.Hanna@sun.com> References: <OFE117653C.B8EE2880-ON85256E92.003B8A8C-85256E94.0000B926@us.ibm.com> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String inencodingRDNs Date: Fri, 14 May 2004 15:20:42 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Tom, > > The bottom line of the statements you have quoted is that some > fully RFC 3280-compliant RP implementations will treat all rollover > certificates as self-issued, some will treat no rollover certificates as > self-issued, and some will treat many rollover certificates as self-issued > but some as not. Thus, whether a given rollover certificate is deemed to > be self-issued by a compliant RP is implementation dependent - probably a > worse situation than "rollover certificates are not interpreted as > self-issued". Yes, the current situation is that some RP implementations will treat name rollover certificates (if any) as self-issued but some will not. The root cause of inconsistency on interpreting self-issued certificates among different implementations is the deficiency of name matching rules in RFC 3280 and the X.509 standard. There can be a debate on whether name rollover certificates should be interpreted as self-issued or not. However, the current situation is the current situation. The situation will not change unless the deficiency in RFC 3280 and the X.509 standard is mended and hopefully all RP implementations will migrate to a consistent way of interpreting self-issued certificates. The point is that even if we finally agree that "name rollover certificates should not interpreted as self-issued", we still need to revise or amend RFC 3280 and the X.509 standard to make it clear. > IMHO, it would really help if we had a list of CA's who encode > their current DN's using BMPString with characters in the surrogate range, > and whether they are compliant to UTF-16 or not, and a list of CA's who > encode their current DN's using T61String. Most other character > conversions under the ASN.1 standard are simple algorithms. > That might be an interesting investigation, but I do not think it will help for resolving the inconsistency on interpreting self-issued certificates. As I said above, the root cause is the deficiency of name matching rules. Paying more attention to mend the deficiency might be more helpful. If we have name matching rules that can take care of name string in different character sets and/or in different encoding, it will not only resolve the inconsistency on interpreting self-issued certificates but also will reduce the needs of issuing name rollover certificates. Wen-Cheng Wang Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E0BUK4068819; Thu, 13 May 2004 17:11:30 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4E0BUp2068818; Thu, 13 May 2004 17:11:30 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from e4.ny.us.ibm.com (e4.ny.us.ibm.com [32.97.182.104]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4E0BJhb068801 for <ietf-pkix@imc.org>; Thu, 13 May 2004 17:11:19 -0700 (PDT) (envelope-from tgindin@us.ibm.com) Received: from northrelay02.pok.ibm.com (northrelay02.pok.ibm.com [9.56.224.150]) by e4.ny.us.ibm.com (8.12.10/8.12.2) with ESMTP id i4E08CKr580206; Thu, 13 May 2004 20:08:22 -0400 Received: from d01ml062.pok.ibm.com (d01av04.pok.ibm.com [9.56.224.64]) by northrelay02.pok.ibm.com (8.12.10/NCO/VER6.6) with ESMTP id i4E08Vaq080562; Thu, 13 May 2004 20:08:32 -0400 To: "Wen-Cheng Wang" <wcwang@cht.com.tw> Cc: "Santoni Adriano" <adriano.santoni@actalis.it>, ietf-pkix@imc.org, "Stephen Kent" <kent@bbn.com>, pgut001@cs.auckland.ac.nz, "Stefan Santesson" <stefans@microsoft.com>, "Steve Hanna" <Steve.Hanna@sun.com> MIME-Version: 1.0 Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encodingRDNs X-Mailer: Lotus Notes Release 5.0.11 July 24, 2002 From: Tom Gindin <tgindin@us.ibm.com> Message-ID: <OFE117653C.B8EE2880-ON85256E92.003B8A8C-85256E94.0000B926@us.ibm.com> Date: Thu, 13 May 2004 20:07:55 -0400 X-MIMETrack: Serialize by Router on D01ML062/01/M/IBM(Release 6.0.2CF2 HFB2|May 5, 2004) at 05/13/2004 20:08:01, Serialize complete at 05/13/2004 20:08:01 Content-Type: text/plain; charset="us-ascii" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Wan-Cheng: The bottom line of the statements you have quoted is that some fully RFC 3280-compliant RP implementations will treat all rollover certificates as self-issued, some will treat no rollover certificates as self-issued, and some will treat many rollover certificates as self-issued but some as not. Thus, whether a given rollover certificate is deemed to be self-issued by a compliant RP is implementation dependent - probably a worse situation than "rollover certificates are not interpreted as self-issued". IMHO, it would really help if we had a list of CA's who encode their current DN's using BMPString with characters in the surrogate range, and whether they are compliant to UTF-16 or not, and a list of CA's who encode their current DN's using T61String. Most other character conversions under the ASN.1 standard are simple algorithms. Tom Gindin P.S. - The above opinions are mine, and not necessarily those of my employer. "Wen-Cheng Wang" <wcwang@cht.com.tw> Sent by: owner-ietf-pkix@mail.imc.org 05/12/2004 05:10 AM To: "Stefan Santesson" <stefans@microsoft.com>, "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org>, "Steve Hanna" <Steve.Hanna@sun.com> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encodingRDNs > Yes, you can do a lot of cool things in theory. > > The cruel truth is that your name encoding roll-over certs (e.g. > CA1-CA1-p-u-cert) is not self issued according to RFC 3280. That is why I said that the definition of self-issued certificates in RFC 3280 (or maybe its definition in the X.509 standard itself) should be revised or be clarified in son-of-3280 (or in an amendment of the X.509 standard as well). > > RFC 3280 section 6.1: > A certificate is self-issued if the DNs that appear in the subject > and issuer fields are identical and are not empty. > > RFC 3280 section 4.1.2.4: > (a) attribute values encoded in different types (e.g., > PrintableString and BMPString) MAY be assumed to represent > different strings; Remenber that RFC 3280 section 4.1.2.4 also said: Implementations of this profile are permitted to use the comparison algorithm defined in the X.500 series. In that respect, you can not said that name rollover certificates are not self-issued. > > So in current implementations of RFC 3280, many valid certificates will > fail after introduction of name rollover certs in the chain, because > path validation will not treat them as self-issued. I admit that most CURRENT implementations of RFC 3280 path validation will fail to treat name rollover certificates as self-issued, but it does not mean that we should not recommend FUTURE revisions of those implementations and new implementations to treat them as self-issued. > > Further, the name constraint problem persists, adds to this but also > exist regardless of this problem. > > RFC 3280 section 4.2.1.11 > When applying restrictions of the form directoryName, an > implementation MUST compare DN attributes. At a minimum, > implementations MUST perform the DN comparison rules specified in > Section 4.1.2.4. That is why I said RFC 3280 path validation algorithm should also be revised. > > Fixing next version of RFC 3280 to include better name matching rules is > a good thing but far from enough to take care of real problems with > current infrastructure. Fixing next version of RFC 3280 does not magically take care of real problems with current infrastructure, but it encourge current infrastructure to migrate to solve the problems. > > We have to try to keep in mind what we make all this hassle for!! > > This is NOT about whether we should use UTF-8 or not to accommodate > complex character sets!! This is about what we are prepared to do to > make sure NOTHING is encoded as PrintableString WHEN PrintableString is > sufficient. > > This old transition rule was created in RFC 2459 before the modern path > validation algorithm was invented and before self-issued certificates > was a special case. That does not mean that name rollover certificates can not solve the mixed name-encoding problem. > > Let's just admit that we overlooked this issue when we created RFC 3280 > and reshaped path validation. It was an understandable human error, > which lead to a defect. Yes, RFC 3280 indeed overlooked this issue. That is why it should be revised. > > So let's then also admit that REQUIRING UTF-8 when printableString is > sufficient, is not a very good idea when all practical matters are > measured, and that it never should have made it to more than a > RECOMMENDATION. Isn't it the IETF policy to encourge all new implementations to migrate to support UTF-8? The philosophy is that if one encoding is sufficient to cover universal characters, why should we recommend two? > > Whatever we say here, this is a requirement that many implementers will > have to ignore for a many years to come. Conformance to any IETF RFC is all voluntary. No one can enfoce any implementer for that if the implementer chooses to ignore the requirement or recommendation of any IETF RFC. RFC 3280 might not have the power to enforce all implementation to migrate to UTF-8 encoding. However, as an IETF RFC, it definitely has a great effect. At least, I see more and more applications can correctly handle certificates with named in UTF-8 encoding. Therefore, I prefer to see son-of-3280 to continue requiring conformant implementations to migrate to UTF-8 encoding. Wen-Cheng Wang Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4DK1AWW051881; Thu, 13 May 2004 13:01:10 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4DK1AHn051880; Thu, 13 May 2004 13:01:10 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from av4-1-sn3.vrr.skanova.net (av4-1-sn3.vrr.skanova.net [81.228.9.111]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4DK19ww051861 for <ietf-pkix@imc.org>; Thu, 13 May 2004 13:01:09 -0700 (PDT) (envelope-from anders.rundgren@telia.com) Received: by av4-1-sn3.vrr.skanova.net (Postfix, from userid 502) id 0D7A937F80; Thu, 13 May 2004 22:01:02 +0200 (CEST) Received: from smtp1-1-sn3.vrr.skanova.net (smtp1-1-sn3.vrr.skanova.net [81.228.9.177]) by av4-1-sn3.vrr.skanova.net (Postfix) with ESMTP id F3C9837E62; Thu, 13 May 2004 22:01:01 +0200 (CEST) Received: from arport (t7o913p111.telia.com [213.64.26.111]) by smtp1-1-sn3.vrr.skanova.net (Postfix) with SMTP id 8FA3B38009; Thu, 13 May 2004 22:00:57 +0200 (CEST) Message-ID: <00c001c43924$d691c5e0$0500a8c0@arport> From: "Anders Rundgren" <anders.rundgren@telia.com> To: "David P. Kemp" <dpkemp@missi.ncsc.mil> Cc: <ietf-pkix@imc.org> References: <005701c3e08e$9b392fe0$1400a8c0@augustcellars.local> <00c701c3e121$0ae3af90$0500a8c0@arport> <4017C963.8060600@bull.net> <200405121708.i4CH7dim022214@stingray.missi.ncsc.mil> <001001c43864$e11bb130$0500a8c0@arport> <200405122138.i4CLchhk006004@stingray.missi.ncsc.mil> Subject: Re: I-D ACTION:draft-ietf-pkix-pi-08.txt Date: Thu, 13 May 2004 21:59:41 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1106 x-mimeole: Produced By Microsoft MimeOLE V6.00.2800.1106 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> David, I don't know if you really care, but there are reasons behind my arguments, beyond just "opinions" or personal taste. >SubjectAltName brings some syntax-enforced discipline >to certificate naming. Feel free to ignore it, and even >to urge others to ignore it, *I* don't have to do this. Putting RFC822 names in subject DNs is the default behavior of every major CA software package there is. In fact 9 out of 10 certificates used for signing messages to the PKIX-list(!) follow this standard, in spite of being deprecated by S/MIME and RFC3280. >but at least feel a twinge of shame when you do so. Well, I believe the RFC3280 should in the coming revision refrain from calling the industry standard "legacy". Harry the PKI amateur -------------------------- In quite a number of RP applications the subject DN is the *only* identity information used, as most folks actually believed that the "subject distinguished name" really is the name of the subject. After looking at a number of commercial certs this looked like a fact. That DNs are created to suit the needs of a directory is not an absolute truth anymore. In particular as many RPs don't store certificates in directories but in "flat" databases. In most cases external certificates are not stored at all, except by CAs for revocation services etc! Encrypted mail is a microscopic application which makes the whole DIT-story of marginal interest. That's was an "opinon" though :-) The "art discussion" --------------------- Now assume that you don't have an e-mail address to stuff into the DN but you do have some kind of unique identifier. Why put this somewhere else than in the DN? An example shows how ugly it gets: Denis' world: DN: CN=John Smith, serialNumber=3, O=ACME, C=US PI: OID.6.4.3.5.6.6.6.43.3, 05725277 The real world: DN: CN=John Smith, serialNumber=05725277, O=ACME, C=US The name space is given by the CA policy. Implicit or explicit. Although esthetics is maybe not priority #1, I believe that "inventing" arbitrary disambiguing numbers when there is a perfectly valid number to use is a *very* hard sale. However, most of the PI motivation goes away if the unique identifier is used as disambiguer and that's the problem in a nutshell. Can you fix this dilemma, please tell us how! A solution for DoD ---------------------- Not that I believe you want my advice but here is a suggestion for DoD: CN=David Kemp, serialNumber=05725277, DC=staffregistry1, DC=dod, DC=mil Using this scheme you get globally unique identifiers (GUIDs) which is a much more needed and popular thing than finding new ways to complicate RP applications. Why would a certificate ever contain more "name" info than this (+ S/MIME support in applicable cases)? rgds Anders Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4DFwOMx034860; Thu, 13 May 2004 08:58:24 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4DFwO4V034859; Thu, 13 May 2004 08:58:24 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4DFwNYR034853 for <ietf-pkix@imc.org>; Thu, 13 May 2004 08:58:24 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Thu, 13 May 2004 08:58:27 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 13 May 2004 08:58:26 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 13 May 2004 08:58:25 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Thu, 13 May 2004 08:58:25 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F02962DBF@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ4v28UxUBpGKJqTwifie+toVecpgAQoApQ From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 13 May 2004 15:58:25.0807 (UTC) FILETIME=[200489F0:01C43903] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4DFwOYR034854 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Denis, As I said previous an OID is one of the possible options, but there is no consensus to make this mandatory so SCVP will continue to support both options. I also don't see consensus that it is mandatory that a SCVP server will always map a request back to the OID. I will be posting SCVP 15 shortly so you can review the latest draft. Trevor -----Original Message----- From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] Sent: Thursday, May 13, 2004 12:54 AM To: Trevor Freeman Cc: Florian Oelmaier; ietf-pkix@imc.org Subject: Re: SCVP & validation policies Trevor, (text deleted) Florian said: "it is widespread consensus on this list that this could be done with SCVP by the use of the OID as a reference to the validation policy as the sole configuration item." I agree. To respond to a request from Wen-Cheng Wang, the OID of the validation policy should be optional in the request. However, if the OID of the validation policy is not present in the request, then the OID of the validation policy that has been used SHALL be returned by the SCVP server in the response. If the validation policy has dynamic parameters, then these parameters SHALL be returned too. Now, we need to move, Trevor. Would you propose in an e-mail extracts of an ASN.1 syntax that would accommodate these requirements for both the request and the response, focussing on the validation policy elements ? Thanks, Denis (text deleted) Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4DEakPL027498; Thu, 13 May 2004 07:36:46 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4DEakjo027497; Thu, 13 May 2004 07:36:46 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from smtp.cs.auckland.ac.nz (smtp.cs.auckland.ac.nz [130.216.33.151]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4DEaerr027477 for <ietf-pkix@imc.org>; Thu, 13 May 2004 07:36:40 -0700 (PDT) (envelope-from pgut001@cs.auckland.ac.nz) Received: from medusa01 (medusa01.cs.auckland.ac.nz [130.216.34.33]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by smtp.cs.auckland.ac.nz (Postfix) with ESMTP id E3C9334105; Fri, 14 May 2004 02:32:54 +1200 (NZST) Received: from pgut001 by medusa01 with local (Exim 4.30) id 1BOHKF-000821-NK; Fri, 14 May 2004 02:36:43 +1200 From: pgut001@cs.auckland.ac.nz (Peter Gutmann) To: aarsenau@bbn.com, anders.rundgren@telia.com, dpkemp@missi.ncsc.mil, ietf-pkix@imc.org Subject: RE: I-D ACTION:draft-ietf-pkix-pi-08.txt In-Reply-To: <GBEOIAAPLLBIMLPDGHDHAELECJAA.aarsenau@bbn.com> Message-Id: <E1BOHKF-000821-NK@medusa01> Date: Fri, 14 May 2004 02:36:43 +1200 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> "Al Arsenault" <aarsenau@bbn.com> writes: >(And as I've pointed out numerous times before, I know of organizations that > put the equivalent of "0514101404" in subjectAltName, as a DNS Name! It's > certainly possible, but that doesn't make it the "right thing" to do. I've seen 'em in uniqueIDs, from a couple of European CAs. Peter. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4D7riJ0059757; Thu, 13 May 2004 00:53:44 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4D7riCi059756; Thu, 13 May 2004 00:53:44 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from odin2.bull.net (odin2.bull.net [192.90.70.84]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4D7rgMi059674 for <ietf-pkix@imc.org>; Thu, 13 May 2004 00:53:43 -0700 (PDT) (envelope-from Denis.Pinkas@bull.net) Received: from clbull.frcl.bull.fr (IDENT:root@clbull.frcl.bull.fr [129.182.8.31]) by odin2.bull.net (8.9.3/8.9.3) with ESMTP id KAA34132; Thu, 13 May 2004 10:02:56 +0200 Received: from bull.net (frcls4013.frcl.bull.fr [129.182.108.120]) by clbull.frcl.bull.fr (8.9.3/8.9.3) with ESMTP id JAA20064; Thu, 13 May 2004 09:52:12 +0200 Message-ID: <40A32999.4070409@bull.net> Date: Thu, 13 May 2004 09:54:01 +0200 From: Denis Pinkas <Denis.Pinkas@bull.net> Organization: Bull SA. User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en, fr MIME-Version: 1.0 To: "'Trevor Freeman'" <trevorf@exchange.microsoft.com> CC: Florian Oelmaier <oelmaier@sytrust.com>, ietf-pkix@imc.org Subject: Re: SCVP & validation policies References: <011c01c43855$9ab957b0$4228a8c0@hagen> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Trevor, (text deleted) Florian said: "it is widespread consensus on this list that this could be done with SCVP by the use of the OID as a reference to the validation policy as the sole configuration item." I agree. To respond to a request from Wen-Cheng Wang, the OID of the validation policy should be optional in the request. However, if the OID of the validation policy is not present in the request, then the OID of the validation policy that has been used SHALL be returned by the SCVP server in the response. If the validation policy has dynamic parameters, then these parameters SHALL be returned too. Now, we need to move, Trevor. Would you propose in an e-mail extracts of an ASN.1 syntax that would accommodate these requirements for both the request and the response, focussing on the validation policy elements ? Thanks, Denis (text deleted) Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4D1Wj6t078151; Wed, 12 May 2004 18:32:45 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4D1Wjn9078150; Wed, 12 May 2004 18:32:45 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4D1Wi4s078132 for <ietf-pkix@imc.org>; Wed, 12 May 2004 18:32:44 -0700 (PDT) (envelope-from aarsenau@bbn.com) Received: from arsenaultd2 (ramblo.bbn.com [128.33.0.51]) by aragorn.bbn.com (8.12.7/8.12.7) with SMTP id i4D1Wd7Y007791; Wed, 12 May 2004 21:32:40 -0400 (EDT) From: "Al Arsenault" <aarsenau@bbn.com> To: "Anders Rundgren" <anders.rundgren@telia.com>, "David P. Kemp" <dpkemp@missi.ncsc.mil>, <ietf-pkix@imc.org> Subject: RE: I-D ACTION:draft-ietf-pkix-pi-08.txt Date: Wed, 12 May 2004 21:30:37 -0400 Message-ID: <GBEOIAAPLLBIMLPDGHDHAELECJAA.aarsenau@bbn.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) x-mimeole: Produced By Microsoft MimeOLE V6.00.2800.1409 In-Reply-To: <001001c43864$e11bb130$0500a8c0@arport> Importance: Normal X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4D1Wj4s078145 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org]On Behalf Of Anders Rundgren > Sent: Wednesday, May 12, 2004 5:01 PM > To: David P. Kemp; ietf-pkix@imc.org > Subject: Re: I-D ACTION:draft-ietf-pkix-pi-08.txt > > > > >It is my hope that PI will become an RFC in the near future, so > >that certificates (from an un-named large PKI :-) that currently > >handle PIs by munging them into Common Name (e.g., > >CN="Kemp.David.P.0514101404") will have a saner alternative. > > The de-facto standard, already engraved in *millions* of certs is > putting 0514101404 in serialNumber. > > The problem with that is that "0514101404" is not expected to change; it's a number which stays with you as long as you're an employee of the company, or as long as you're a customer, or even all of your life. The serialNumber is expected to change every time the certificate changes - commonly, when a certificate expires once a year, the new certificate has a new serialNumber. It's considered seriously bad form to issue a "renewed" certificate with a different validity period and the same serialNumber. Thus, those who put 0514101404 in "serialNumber" are just setting themselves up for trouble later on down the road. (And as I've pointed out numerous times before, I know of organizations that put the equivalent of "0514101404" in subjectAltName, as a DNS Name! It's certainly possible, but that doesn't make it the "right thing" to do. Al Arsenault Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CMCvuR062651; Wed, 12 May 2004 15:12:57 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CMCvGY062650; Wed, 12 May 2004 15:12:57 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from stingray.missi.ncsc.mil (stingray.missi.ncsc.mil [144.51.50.20]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CMCuVT062643 for <ietf-pkix@imc.org>; Wed, 12 May 2004 15:12:56 -0700 (PDT) (envelope-from DPKemp@missi.ncsc.mil) Message-ID: <200405122138.i4CLchhk006004@stingray.missi.ncsc.mil> Date: Wed, 12 May 2004 18:12:04 -0400 From: "David P. Kemp" <dpkemp@missi.ncsc.mil> User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Anders Rundgren <anders.rundgren@telia.com> CC: ietf-pkix@imc.org Subject: Re: I-D ACTION:draft-ietf-pkix-pi-08.txt References: <005701c3e08e$9b392fe0$1400a8c0@augustcellars.local> <00c701c3e121$0ae3af90$0500a8c0@arport> <4017C963.8060600@bull.net> <200405121708.i4CH7dim022214@stingray.missi.ncsc.mil> <001001c43864$e11bb130$0500a8c0@arport> In-Reply-To: <001001c43864$e11bb130$0500a8c0@arport> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 12 May 2004 22:12:05.0026 (UTC) FILETIME=[28802020:01C4386E] Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> It is certainly technically possible to put any attribute into the subject name syntax. An unfortunate side effect of an infinitely-flexible syntax is that de-facto standards such as those you cite can arise. DNS (and by extension, email address) is a namespace rooted in the TLDs (.com, .edu, ...). PIs are another namespace rooted either in some CA-specific environment (policy=CA) or in an independent manner (by assigning agency) that allows multiple CAs to issue certificates to one identifiable user. Subject Distinguished Names are intended to be a third namespace, typically rooted by Country. When you say "just stuff an email address in the middle of a DN", that is equivalent to saying "just stuff a DN in the middle of an email address: "dpkemp@missi.(C=US,O=DoD,CN=Dave).ncsc.mil" Fortunately, RFC822 prohibits such foolishness in a way that cannot be circumvented by fools. Unfortunately, X.509 DN syntax does not prohibit anything, and it is up to system architects to establish rules that enable DNs to form a namespace rather than a cloaca for miscellaneous unrelated data. It is as easy for CAs to ignore those rules as it is for Harry Homeowners to ignore electrical codes and wire their basements without grounding and with hot and neutral lines connected randomly. The fact that something can be done and is in fact done millions of times does not make it a good idea. SubjectAltName brings some syntax-enforced discipline to certificate naming. Feel free to ignore it, and even to urge others to ignore it, but at least feel a twinge of shame when you do so. Anders Rundgren wrote: >>It is my hope that PI will become an RFC in the near future, so >>that certificates (from an un-named large PKI :-) that currently >>handle PIs by munging them into Common Name (e.g., >>CN="Kemp.David.P.0514101404") will have a saner alternative. > > > The de-facto standard, already engraved in *millions* of certs is > putting 0514101404 in serialNumber. > > This is almost as de-facto standard as putting e-mail addresses in DNs > which in turn is almost as de-facto standard as using URIs for naming > globally unique objects. > > C:\Internet-Drafts>del draft-ietf-pkix-pi-*.txt > > :-) > > Pardon my complaints, let there be an RFC! But don't expect > this scheme to become the trend. > > There is a slight problem with the whole idea. Either RPs require > and act upon the PI-data or they don't care about it. This in my > opinion makes the extension redundant or is just another way > to screw up validation. > > If you on top of this add policy extensions I believe a real disaster > is in the making. > > Anders > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CL6vA5057563; Wed, 12 May 2004 14:06:57 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CL6vdX057562; Wed, 12 May 2004 14:06:57 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from av9-2-sn1.fre.skanova.net (av9-2-sn1.fre.skanova.net [81.228.11.116]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CL6uDI057548 for <ietf-pkix@imc.org>; Wed, 12 May 2004 14:06:56 -0700 (PDT) (envelope-from anders.rundgren@telia.com) Received: by av9-2-sn1.fre.skanova.net (Postfix, from userid 502) id 0D31538015; Wed, 12 May 2004 23:06:53 +0200 (CEST) Received: from smtp3-1-sn1.fre.skanova.net (smtp3-1-sn1.fre.skanova.net [81.228.11.163]) by av9-2-sn1.fre.skanova.net (Postfix) with ESMTP id EE3D037E71; Wed, 12 May 2004 23:06:52 +0200 (CEST) Received: from arport (t12o913p16.telia.com [213.64.28.136]) by smtp3-1-sn1.fre.skanova.net (Postfix) with SMTP id 0BBC037E49; Wed, 12 May 2004 23:06:33 +0200 (CEST) Message-ID: <001001c43864$e11bb130$0500a8c0@arport> From: "Anders Rundgren" <anders.rundgren@telia.com> To: "David P. Kemp" <dpkemp@missi.ncsc.mil>, <ietf-pkix@imc.org> References: <005701c3e08e$9b392fe0$1400a8c0@augustcellars.local> <00c701c3e121$0ae3af90$0500a8c0@arport> <4017C963.8060600@bull.net> <200405121708.i4CH7dim022214@stingray.missi.ncsc.mil> Subject: Re: I-D ACTION:draft-ietf-pkix-pi-08.txt Date: Wed, 12 May 2004 23:01:23 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1106 x-mimeole: Produced By Microsoft MimeOLE V6.00.2800.1106 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> >It is my hope that PI will become an RFC in the near future, so >that certificates (from an un-named large PKI :-) that currently >handle PIs by munging them into Common Name (e.g., >CN="Kemp.David.P.0514101404") will have a saner alternative. The de-facto standard, already engraved in *millions* of certs is putting 0514101404 in serialNumber. This is almost as de-facto standard as putting e-mail addresses in DNs which in turn is almost as de-facto standard as using URIs for naming globally unique objects. C:\Internet-Drafts>del draft-ietf-pkix-pi-*.txt :-) Pardon my complaints, let there be an RFC! But don't expect this scheme to become the trend. There is a slight problem with the whole idea. Either RPs require and act upon the PI-data or they don't care about it. This in my opinion makes the extension redundant or is just another way to screw up validation. If you on top of this add policy extensions I believe a real disaster is in the making. Anders Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CJGOVk048907; Wed, 12 May 2004 12:16:24 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CJGOLK048906; Wed, 12 May 2004 12:16:24 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-in.m-online.net (mail-in.m-online.net [62.245.150.237]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CJGNqC048893 for <ietf-pkix@imc.org>; Wed, 12 May 2004 12:16:23 -0700 (PDT) (envelope-from oelmaier@sytrust.com) Received: from mail.m-online.net (svr14.m-online.net [192.168.3.144]) by svr8.m-online.net (Postfix) with ESMTP id 3B5544BA6C; Wed, 12 May 2004 21:16:22 +0200 (CEST) Received: from hagen (ppp-82-135-6-58.mnet-online.de [82.135.6.58]) by mail.m-online.net (Postfix) with ESMTP id E677EA92C4; Wed, 12 May 2004 21:16:21 +0200 (CEST) From: "Florian Oelmaier" <oelmaier@sytrust.com> To: "'Trevor Freeman'" <trevorf@exchange.microsoft.com>, "'Denis Pinkas'" <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> Subject: RE: SCVP & validation policies Date: Wed, 12 May 2004 21:16:18 +0200 Organization: SyTrust Message-ID: <011c01c43855$9ab957b0$4228a8c0@hagen> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.3416 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Importance: Normal In-reply-to: <33E7A1613A3480448996047B69308A2F029629BA@df-grommit-01.dogfood> Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > Hi Denis, > Any software does not have a clue abut most things in life. > To operate the administrator of the client has to have > understand what is going on. If the administrator does not > understand what the implications of their actions are, then > they make mistakes. Your premise is that simplified > administration is only possible in the case of SCVP by the > use of the OID. Given there is a vast quantity of > administration software which his able to present simple > choices to the administrator while at the same time > configuring multiple parameters without there knowledge mean > this is not the case. Trevor There are different roles in the administration. And the "usual administrator for client software" is a rather different role than the role of a security administrator. In the meantime (mostly due to the bad experiences) this is a general accepted model. In most enterprises there are different people responsible for operating the client and for managing the security. Thus I dont think it is necessary for a client administrator to understand the implications of settings necessary for certificate validation. From my perspective there is a move in all areas of security to strip as many security relevant configuration setting from the client as possible. And I think it is widespread consensus on this list that this could be done with SCVP by the use of the OID as a reference to the validation policy as the sole configuration item. Florian > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Wednesday, May 12, 2004 12:25 AM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: Re: SCVP & validation policies > > Trevor, > > > You also have to understand enough so that the administrative > > configuration of the client is generating the right kind of > request. > > > SCVP client is ?only following orders? in many instances > and has from a The SCVP client is "only following orders" in > many instances and has > from a perspective no idea if it is > the right kind of request, only if it > is well formed to > the rules it has been given. Both the administrator of > > the client and the administrator of the server has to understand in > much > > grater detail what the given policy is. Once you abstract though the > > OID, you are placing a greater burden on the OOB admin to admin > process. > > The client does not have to "understand" what the validation given > policy is. > > The client only needs to know that, for a given application a > given OID > for > the validation policy is needed. He does not need to have a > clue of the > details of the validation policy which may remain fully opaque to him. > > This means that any validation specific parameter must *not* > be part of > the > standard request. > > Denis > > > > Trevor > > > > > > > > > -------------------------------------------------------------- > ---------- > > > > From: James Jing [mailto:jjing@betrusted.com] > > Sent: Tuesday, May 11, 2004 5:09 PM > > To: Trevor Freeman; ietf-pkix@imc.org > > Subject: RE: SCVP & validation policies > > > > > > > > Agree that the client needs to have all the data in order > to build the > > > request. However, I think ?understand? is too strong a requirement. > The > > client may not need to ?understand? the data, some of which may be > > factors affecting the server only. Perhaps I am too > pedantic about the > > > word ?understand?. > > > > > > > > Back to indirection, if ?understanding? is not a > requirement, then the > > > client need not to be aware that an OID is an indirection > to a set of > > policies. > > > > > > > > James > > > > > > > > > -------------------------------------------------------------- > ---------- > > > > From: Trevor Freeman [mailto:trevorf@exchange.microsoft.com] > > Sent: Wednesday, 12 May 2004 9:49 AM > > To: James Jing; ietf-pkix@imc.org > > Subject: RE: SCVP & validation policies > > > > > > > > The client has to understand what is required to build a > request i.e. > > What data does it need to supply to the server. It does not > need to be > > aware of the factors which effect the server. > > Trevor > > > > -----Original Message----- > > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org] > > On Behalf Of James Jing > > Sent: Tuesday, May 11, 2004 4:01 PM > > To: ietf-pkix@imc.org > > Subject: RE: SCVP & validation policies > > > > > > > > I think policy OIDs should remain abstract, or at least can be > abstract. > > That is that there is no requirement for the client software to > > understand the parameters. > > > > I am more thinking of using SCVP to validate a certificate in the > > context of a transaction, where the policy OID in SCVP might be used > to > > identify such a context to the server. > > > > For example, a certificate validation request is carried out with > regard > > to the acceptability of the holder's creditworthiness in the context > of > > a credit-based transaction. In this case, the policy OID > could be used > > to indicate the class of assuarances that the CA's CSP provides for, > > resulting in the server selecting a subset of acceptable trust > anchors. > > > > James > > > > > > > > -----Original Message----- > > From: Stephen Kent [mailto:kent@bbn.com] > > Sent: Friday, 7 May 2004 6:28 AM > > To: Trevor Freeman > > Cc: ietf-pkix@imc.org > > Subject: RE: SCVP & validation policies > > > > > > > > At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: > >>Hi Steve, > >>If you want to have a shorthand notation for a validation policy, > given > > > >>we have a finite set of parameters, the I don't see the value of > >>indirection via an OID. Using a hash of the set of inputs provides a > >>way of expressing the same thing without introducing > ambiguity of does > >>both parties have a common understating of that this means. At some > >>point someone has to define the policy in terms of the 3280 input > > parameters. > >>That definition is just as applicable and consumable by the > client as > a > > > >>server therefore what did we gain from the shorthand notation? It > would > > > >>be pretty trivial to define some XML which would allow that set of > >>validation parameters to be imported by any SCVP entity. > >>Trevor > > > > Trevor, > > > > It is not indirection via an OID, it is identification via an OID. > > Clients never have to be aware of the parameters if we use > an OID, and > > simplification of clients is one of the goals of SCVP. So I cannot > agree > > with your comment that clients have to be able to consume these > > parameters. > > > > Given that as a basis, it seems that we are debating whether to use > OIDs > > or to create some other form of identifier via hashing parameters. I > > don't see the latter as attractive, since it requires additional > > specification of the canonicalization and encoding of the parameters > to > > ensure consistency in hash computation between client and server. > > > > Steve > > > > > > > > This e-mail, and any attachments hereto, is intended only for use by > the > > named addressee(s) and may contain legally privileged and/or > > confidential information. If you are not the intended > recipient, you > > are hereby notified that any dissemination, distribution or > copying of > > > this e-mail, and any attachments hereto, is strictly prohibited. If > you > > have received this transmission in error, please notify me > immediately > > > and permanently delete the original and all copies and printouts of > this > > e-mail. > > > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CJ0bac047776; Wed, 12 May 2004 12:00:37 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CJ0bE1047775; Wed, 12 May 2004 12:00:37 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CJ0bir047769 for <ietf-pkix@imc.org>; Wed, 12 May 2004 12:00:37 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Wed, 12 May 2004 12:00:41 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Wed, 12 May 2004 12:00:41 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Wed, 12 May 2004 12:00:40 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Wed, 12 May 2004 12:00:39 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F02962AB3@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ4ThklIu4lLosSSIeWZksBBllP7gABLKWw From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Peter Sylvester" <Peter.Sylvester@edelweb.fr>, <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 12 May 2004 19:00:40.0881 (UTC) FILETIME=[6B6AAA10:01C43853] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4CJ0bir047770 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Peter, The problem of having to push policy to clients of all descriptions is a well understood problem set. The infrastructure to do that is the main overhead. The number of parameters associated with that process is not a significant overhead. Trevor -----Original Message----- From: Peter Sylvester [mailto:Peter.Sylvester@edelweb.fr] Sent: Wednesday, May 12, 2004 11:23 AM To: Denis.Pinkas@bull.net; Trevor Freeman Cc: ietf-pkix@imc.org Subject: RE: SCVP & validation policies trevor, Your clients may need to be updated each time you change anything in the structure of your certs, or in the rules of the policies, Any code that can be avoided, seems good to me in line with the requirement for dumb client support. > > Hi Denis, > Any software does not have a clue abut most things in life. To operate > the administrator of the client has to have understand what is going on. > If the administrator does not understand what the implications of their > actions are, then they make mistakes. Your premise is that simplified > administration is only possible in the case of SCVP by the use of the > OID. Given there is a vast quantity of administration software which his > able to present simple choices to the administrator while at the same > time configuring multiple parameters without there knowledge mean this > is not the case. > Trevor > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CISF20045697; Wed, 12 May 2004 11:28:15 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CISFsO045696; Wed, 12 May 2004 11:28:15 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CISErd045690 for <ietf-pkix@imc.org>; Wed, 12 May 2004 11:28:14 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Wed, 12 May 2004 11:28:18 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Wed, 12 May 2004 11:28:18 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Wed, 12 May 2004 11:28:17 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Wed, 12 May 2004 11:28:18 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F02962A81@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ4An82dGYpctolT6i0rcvrW0RAGAAPl/ZQ From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Wen-Cheng Wang" <wcwang@cht.com.tw>, "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: "Stephen Kent" <kent@bbn.com>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 12 May 2004 18:28:17.0944 (UTC) FILETIME=[E5560180:01C4384E] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4CISErd045691 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Wen-Cheng, In scenario 1, the use of and validation policy OID option in a request is superfluous. The SCVP client simply submits the request and the server uses its default policy. In scenario 3 there are two way to achieve this with SCVP. The client and can be configured with a set of OIDs to represent the policy, or they can by configured wit the set of parameters to submit. In both cases the SCVP server certifies the request matches its local policy before processing the request. Trevor -----Original Message----- From: Wen-Cheng Wang [mailto:wcwang@cht.com.tw] Sent: Wednesday, May 12, 2004 2:21 AM To: Denis Pinkas; Trevor Freeman Cc: Stephen Kent; ietf-pkix@imc.org Subject: Re: SCVP & validation policies I would like to support that even the validation policy OID itself is optional for the client to use. As I stated in my earlier post, if a client fully trust the validation server, then it might not have to specify a validation policy OID in its request. In that case, it means that the client trusts the validation server to adopt the default validation policy predefinied by the management authority of the application domain (such as an enterprise). For example, in scenario 1 and scenario 3 I mentioned in my earlier post, if the client request does not specify any validation policy OID, then server will adopt the default validation policy. For you reference, scenario 1 and scenario 3 are as follows: Scenario 1 (single central-managed vlidation policy) DPV server: only supports a single central-managed vlidation policy the validation policy implies all policy-dependent parameters DPV client: the validation request may optionally specify the vlidation policy OID the validation request needs not specify any policy-dependent parameters Scenario 3 (multiple central-managed vlidation policies with a default validation policy) DPV server: supports multiple central-managed vlidation policies with one of them is the default each validation policy implies its policy-dependent parameters DPV client: the validation request should specify the vlidation policy OID if the request did not specify the vlidation policy OID, then the server will use the default one the validation request needs not to specify any policy-dependent parameters These are the simplest cases of the use of DPV protocol. I hope that basic SCVP syntax should support these simplest cases more clearly. (I mean that the ambiguity in the current syntax should be reduced.) In contrast to these simplest cases, an extreme case of the use of DPV protocol is the case where there is no any predefined validation policy (and hence no default validation policy). This might be the case where a DPV server is used as a public server. (The scenario 4 I mentioned earlier.) Please note that in this extreme case, the client has no way to specify any validation policy OID in its request since there is no any predefined one. In this extreme case, the client will have to submit the full set of parameters for path validation ALGORITHM. I would like to see SCVP to provide an extension mechanism to let the client submit the full set of parameters for path validation ALGORITHM. However, please do not confuse "parameters for validation ALGORITHM" with "parameters for validation POLICY". Also, please make a clear separation of basic SCVP syntax and the extension mechanism for submitting parameters for path validation ALGORITHM. Wen-Cheng Wang > > Trevor, > > > Hi Dennis, > > The OID is optional for the client to use. Servers don't have to support > > this. > > I disagree. The "RFC 3280 compliantbasic validation policy", i.e the one > which only includes basic path validation, with no extra and no shortcuts, > as defined by RFC 3280, will have an OID. Since this policy does not > contains values such as root keys, this validation policy needs additional > parameters. > > Intelligent clients may use this one and specify parameters. > > Thin clients will simply refer to an OID which refers globally to a > validation algorithm and its parameters. > > I have talked to customers and their position is to have flexible validation > policies. As an example, some says: if we want to take purposely the risk of > not testing ARLs, it is our choice. The validation policy should not mandate > to test ARLs, but if for some validation policies it is adequate to do so, > it should be possible to do so. > > > It become a matter of mutual agreement between the client and > > server what the OID represents. > > Not necessarilly. Since the OID is not in the arc of the server, it may be > defined by anyone. If that definition is separately accessible both to the > client and to the server then it can be supported by the server and the > client may refer to it. In such a case there is no need for a mutual agreement. > > > An SCVP server must either comply with any request or return an error > > True. > > > so I don't see the validation policy OID > > needs any special treatment beyond want SCVP already defines. > > I don't see the relationship between the true statement above and the > conclusion. > > As already stated on this list, SCVP stands for "Simple", otherwise we will > have to drop the "S", and then we would have "CVP". :-| > > Denis > > > Trevor Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CIMdUb045305; Wed, 12 May 2004 11:22:39 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CIMdQl045304; Wed, 12 May 2004 11:22:39 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CIMbwG045297 for <ietf-pkix@imc.org>; Wed, 12 May 2004 11:22:38 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4CIMeN18865; Wed, 12 May 2004 20:22:40 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Wed, 12 May 2004 20:22:40 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4CIMdE16565; Wed, 12 May 2004 20:22:39 +0200 (MEST) Date: Wed, 12 May 2004 20:22:39 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405121822.i4CIMdE16565@chandon.edelweb.fr> To: Denis.Pinkas@bull.net, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> trevor, Your clients may need to be updated each time you change anything in the structure of your certs, or in the rules of the policies, Any code that can be avoided, seems good to me in line with the requirement for dumb client support. > > Hi Denis, > Any software does not have a clue abut most things in life. To operate > the administrator of the client has to have understand what is going on. > If the administrator does not understand what the implications of their > actions are, then they make mistakes. Your premise is that simplified > administration is only possible in the case of SCVP by the use of the > OID. Given there is a vast quantity of administration software which his > able to present simple choices to the administrator while at the same > time configuring multiple parameters without there knowledge mean this > is not the case. > Trevor > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CI9fMw043764; Wed, 12 May 2004 11:09:41 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CI9fic043763; Wed, 12 May 2004 11:09:41 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CI9eYU043751 for <ietf-pkix@imc.org>; Wed, 12 May 2004 11:09:40 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4CI9YN18602; Wed, 12 May 2004 20:09:34 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Wed, 12 May 2004 20:09:34 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4CI9XG16517; Wed, 12 May 2004 20:09:33 +0200 (MEST) Date: Wed, 12 May 2004 20:09:33 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405121809.i4CI9XG16517@chandon.edelweb.fr> To: Denis.Pinkas@bull.net, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Denis, I assume you mean 'minimal' instead of 'standard', 'the standard rerequest for a very dumb client' > This means that any validation specific parameter must *not* be part of > the standard request. > > Denis Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CHgs5i041686; Wed, 12 May 2004 10:42:54 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CHgswM041685; Wed, 12 May 2004 10:42:54 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from stingray.missi.ncsc.mil (stingray.missi.ncsc.mil [144.51.50.20]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CHgs5m041678 for <ietf-pkix@imc.org>; Wed, 12 May 2004 10:42:54 -0700 (PDT) (envelope-from DPKemp@missi.ncsc.mil) Message-ID: <200405121708.i4CH7dim022214@stingray.missi.ncsc.mil> Date: Wed, 12 May 2004 11:45:17 -0400 From: "David P. Kemp" <dpkemp@missi.ncsc.mil> User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax) X-Accept-Language: en-us, en MIME-Version: 1.0 To: ietf-pkix@imc.org Subject: Re: I-D ACTION:draft-ietf-pkix-pi-08.txt References: <005701c3e08e$9b392fe0$1400a8c0@augustcellars.local> <00c701c3e121$0ae3af90$0500a8c0@arport> <4017C963.8060600@bull.net> In-Reply-To: <4017C963.8060600@bull.net> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 12 May 2004 15:45:17.0878 (UTC) FILETIME=[1FF65D60:01C43838] Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Denis and Tom, It is my hope that PI will become an RFC in the near future, so that certificates (from an un-named large PKI :-) that currently handle PIs by munging them into Common Name (e.g., CN="Kemp.David.P.0514101404") will have a saner alternative. Comments: "Assigner Authority and Type" could and should be better explained, in the I-D / RFC, but it is also clear what it means without such an explanation. See www.acq.osd.mil/uid/policy.html, "Guide to Uniquely Identifying Items" for an example. The guide defines five "Issuing Agencies" (assigner authorities) where each issuing agency defines the type of identifier it assigns. The issuing agencies (and types) are: 1. EAN-International (EAN.UCC company prefix that goes on the barcode of everything you buy in the supermarket) 2. Telcordia Technologies (ANSI T1.220 numbers) 3. Dun & Bradstreet (DUNS number) 4. Allied Committee 135 (CAGE code - look up your favorite company's number at www.dlis.dla.mil/cage_welcome.asp) 5. European Health Industry Business Communications Council (EHIBCC number) There should not be a large number of PI assigner authorities, but there will be more than five and the list will grow, so they can't be enumerated within the RFC itself. It is reasonable to identify them by IANA-assigned value, or by OID. It is not completely unreasonable to identify them by URI, but there is an impedance mismatch between the goals of PI (physical organizations with stable business processes to establish reliable identifiers) and the mechanism of URI (put up a document/resource somewhere in cyberspace). Not that a URI couldn't point to a substantial, permanent organization, or that the target of an OID couldn't be ephemeral, but OIDs do have a more substantial "flavor" than URIs. More to the point, an OID tells a machine implementation what to do with a PI, whereas a URI is presumably intended to cause the implementation to make an online access to find out what to do. That is an unreasonable burden. I believe that the syntax of PI-08 is fine as is, and that with Jim Schaad's corrections and perhaps some more explanatory text the draft is ready to become an RFC. Dave Denis Pinkas wrote: > > Anders, > >> Well Jim, >> The PI-draft is certainly an odd creature that has been in PKIX's >> "custody" for now close to four(!) years. >> >> A few comments to PI in general and some draft-08 specific. >> >> Draft-08: identifierType "The IdentifierType field, when present, >> is an OID which identifies both the Assigner Authority and the type >> of that field. It is an OID" >> >> The words "type of that field" has never been explained. But we all >> know exactly what the authors' mean don't we? :-| >> >> Draft-08: identifierType has, as you noted been reduced to only support >> OIDs, probably due to the authors' belief that OIDs are more permanent. > > > It was not exactly the author's belief, but the IESG belief. > >> However, the rest of the computer industry have no problems using URIs >> making PI "incompatible" and "deviating". A URI also have the >> big advantage that it may point to a document. > > > ... that has then disappeared, or even worse, pointing to a document > that is different from the original one. > >> I find this "obsession" with permanence wrong, because if a CA >> disappears, >> its regitered DNS name is taken by somebody else, > > > The URI was to designate an object defined by an Assigner Authority, not > by a CA. > >> the certificates are >> anaway no longer usable and names may indeed be reused. That is, OID >> or URI is a >> deployment issue. At best you can RECOMMEND the "serious" >> implementer to use an OID. >> >> PI-General: During the PI development serialNumber has become >> the de-facto standard for keeping a permanent (or static which is >> really more what we are talking about) identifier. > > > You have already made this comment in the past. serialNumber serves a > different purpose. Please take a look at RFC 3039 : > > The serialNumber attribute type SHALL, when present, be used to > differentiate between names where the subject field would otherwise > be identical. This attribute has no defined semantics beyond > ensuring uniqueness of subject names. > >> Anyway, PI-using relying party software have little reason to >> bother about PI as the existing systems I know of simply does >> not "decipher" certificates looking for the most "attractive" >> name form(?), they just _assume_ that things are as the CP says. Most >> of these CAs also obey to Anders' nowadays (in)famous >> "best practice CA formula": >> >> CA <=> Policy >> >> which makes identifierType redundant and also eliminating the >> need for additional Subject name forms like PI. > > > If you do not have a need for the PI, then that's fine; but other people > may wish to use it, as soon/long as it is defined. > > Denis > > > >> Anders >> >> ----- Original Message ----- From: "Jim Schaad" <jimsch@nwlink.com> >> To: "'Denis Pinkas'" <Denis.Pinkas@bull.net>; "'Tom Gindin'" >> <tgindin@us.ibm.com> >> Cc: <ietf-pkix@imc.org> >> Sent: Thursday, January 22, 2004 03:22 >> Subject: RE: I-D ACTION:draft-ietf-pkix-pi-08.txt >> >> >> >> Comments on draft -08. >> >> 1. Privacy concerns should be additionally outlined and covered in the >> security considerations section. >> >> 2. Section 1, para 9: Text here allows for an AA to have a unique URI >> name. Has been removed from the syntax. >> >> 3. Section 1, para 14: Suggested text "both the same permanent >> identifier value and the same Authority Identifer Value, then these". >> (I find the use of identityType to be confusing since I consider this to >> relate to type choice of IdentifierValue as well.) >> >> 4. Section 2: What is the criteria for use of iA5String vs uTF8String? >> >> >> 5. Section 2: When doing comparisons do I need to cross compare these >> two different value items? >> >> 6. Section 2: The statement 'It is an OID.' is redundant and can be >> removed. >> >> 7. Section 3: Contains a paragraph dealing with matching rules. Since >> this is no longer a field in the structure the paragraph should be >> re-written to state what the one and only matching rule is. >> >> 8. Appendix A.1: New request, please put a reference to the document >> containing the pkix module into the pkix document (i.e. something like >> "-- found in [RFC 3280]". This prevents documents from importing >> modules that are lower in the standards process without having any >> explicit reference to them. >> >> jim >> >> >> >>> -----Original Message----- >>> From: owner-ietf-pkix@mail.imc.org >>> [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of >>> Internet-Drafts@ietf.org >>> Sent: Wednesday, January 21, 2004 1:27 PM >>> To: IETF-Announce: >>> Cc: ietf-pkix@imc.org >>> Subject: I-D ACTION:draft-ietf-pkix-pi-08.txt >>> >>> >>> A New Internet-Draft is available from the on-line Internet-Drafts >>> directories. This draft is a work item of the Public-Key >>> Infrastructure (X.509) Working Group of the IETF. >>> >>> Title : Internet X.509 Public Key Infrastructure Permanent Identifier >>> Author(s) : D. Pinkas, T. Gindin >>> Filename : draft-ietf-pkix-pi-08.txt >>> Pages : 12 >>> Date : 2004-1-21 >>> >>> This document define a new form of name, called permanent identifier, >>> that may be included in the subjectAltName extension of a public key >>> certificate issued to an entity. >>> The permanent identifier is an optional feature that may be used by a >>> CA to indicate that the certificate relates to the same entity even >>> if the name or the affiliation of that entity stored in the subject >>> or another name form in the subjectAltName extension has changed. >>> The subject name, carried in the subject field, is only unique for >>> each subject entity certified by the one CA as defined by the issuer >>> name field. Also, the new name form can carry a name that is unique >>> for each subject entity certified by a CA. >>> >>> A URL for this Internet-Draft is: >>> http://www.ietf.org/internet-drafts/draft-> ietf-pkix-pi-08.txt Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CG5uIe034533; Wed, 12 May 2004 09:05:56 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CG5ujK034532; Wed, 12 May 2004 09:05:56 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CG5tHx034526 for <ietf-pkix@imc.org>; Wed, 12 May 2004 09:05:55 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Wed, 12 May 2004 09:05:58 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Wed, 12 May 2004 09:05:52 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Wed, 12 May 2004 09:05:57 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Wed, 12 May 2004 09:05:56 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F029629BA@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ38jxRpgWQ5+FbQcCKCGuLYN1ZZgAR7fGw From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 12 May 2004 16:05:57.0772 (UTC) FILETIME=[02FF2CC0:01C4383B] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4CG5tHx034527 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Denis, Any software does not have a clue abut most things in life. To operate the administrator of the client has to have understand what is going on. If the administrator does not understand what the implications of their actions are, then they make mistakes. Your premise is that simplified administration is only possible in the case of SCVP by the use of the OID. Given there is a vast quantity of administration software which his able to present simple choices to the administrator while at the same time configuring multiple parameters without there knowledge mean this is not the case. Trevor -----Original Message----- From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] Sent: Wednesday, May 12, 2004 12:25 AM To: Trevor Freeman Cc: ietf-pkix@imc.org Subject: Re: SCVP & validation policies Trevor, > You also have to understand enough so that the administrative > configuration of the client is generating the right kind of request. > > SCVP client is ?only following orders? in many instances and has from a The SCVP client is "only following orders" in many instances and has > from a perspective no idea if it is the right kind of request, only if it > is well formed to the rules it has been given. Both the administrator of > the client and the administrator of the server has to understand in much > grater detail what the given policy is. Once you abstract though the > OID, you are placing a greater burden on the OOB admin to admin process. The client does not have to "understand" what the validation given policy is. The client only needs to know that, for a given application a given OID for the validation policy is needed. He does not need to have a clue of the details of the validation policy which may remain fully opaque to him. This means that any validation specific parameter must *not* be part of the standard request. Denis > Trevor > > > > ------------------------------------------------------------------------ > > From: James Jing [mailto:jjing@betrusted.com] > Sent: Tuesday, May 11, 2004 5:09 PM > To: Trevor Freeman; ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > > Agree that the client needs to have all the data in order to build the > request. However, I think ?understand? is too strong a requirement. The > client may not need to ?understand? the data, some of which may be > factors affecting the server only. Perhaps I am too pedantic about the > word ?understand?. > > > > Back to indirection, if ?understanding? is not a requirement, then the > client need not to be aware that an OID is an indirection to a set of > policies. > > > > James > > > > ------------------------------------------------------------------------ > > From: Trevor Freeman [mailto:trevorf@exchange.microsoft.com] > Sent: Wednesday, 12 May 2004 9:49 AM > To: James Jing; ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > > The client has to understand what is required to build a request i.e. > What data does it need to supply to the server. It does not need to be > aware of the factors which effect the server. > Trevor > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On Behalf Of James Jing > Sent: Tuesday, May 11, 2004 4:01 PM > To: ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > > I think policy OIDs should remain abstract, or at least can be abstract. > That is that there is no requirement for the client software to > understand the parameters. > > I am more thinking of using SCVP to validate a certificate in the > context of a transaction, where the policy OID in SCVP might be used to > identify such a context to the server. > > For example, a certificate validation request is carried out with regard > to the acceptability of the holder's creditworthiness in the context of > a credit-based transaction. In this case, the policy OID could be used > to indicate the class of assuarances that the CA's CSP provides for, > resulting in the server selecting a subset of acceptable trust anchors. > > James > > > > -----Original Message----- > From: Stephen Kent [mailto:kent@bbn.com] > Sent: Friday, 7 May 2004 6:28 AM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > > At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: >>Hi Steve, >>If you want to have a shorthand notation for a validation policy, given > >>we have a finite set of parameters, the I don't see the value of >>indirection via an OID. Using a hash of the set of inputs provides a >>way of expressing the same thing without introducing ambiguity of does >>both parties have a common understating of that this means. At some >>point someone has to define the policy in terms of the 3280 input > parameters. >>That definition is just as applicable and consumable by the client as a > >>server therefore what did we gain from the shorthand notation? It would > >>be pretty trivial to define some XML which would allow that set of >>validation parameters to be imported by any SCVP entity. >>Trevor > > Trevor, > > It is not indirection via an OID, it is identification via an OID. > Clients never have to be aware of the parameters if we use an OID, and > simplification of clients is one of the goals of SCVP. So I cannot agree > with your comment that clients have to be able to consume these > parameters. > > Given that as a basis, it seems that we are debating whether to use OIDs > or to create some other form of identifier via hashing parameters. I > don't see the latter as attractive, since it requires additional > specification of the canonicalization and encoding of the parameters to > ensure consistency in hash computation between client and server. > > Steve > > > > This e-mail, and any attachments hereto, is intended only for use by the > named addressee(s) and may contain legally privileged and/or > confidential information. If you are not the intended recipient, you > are hereby notified that any dissemination, distribution or copying of > this e-mail, and any attachments hereto, is strictly prohibited. If you > have received this transmission in error, please notify me immediately > and permanently delete the original and all copies and printouts of this > e-mail. > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CCEuPj014071; Wed, 12 May 2004 05:14:56 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CCEuhh014070; Wed, 12 May 2004 05:14:56 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CCEt9V014059 for <ietf-pkix@imc.org>; Wed, 12 May 2004 05:14:55 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4CCEtN12679; Wed, 12 May 2004 14:14:55 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Wed, 12 May 2004 14:14:55 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4CCEs614503; Wed, 12 May 2004 14:14:54 +0200 (MEST) Date: Wed, 12 May 2004 14:14:54 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405121214.i4CCEs614503@chandon.edelweb.fr> To: Peter.Sylvester@edelweb.fr, kent@bbn.com, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > > Hi Peter, > I think saying the client must validate the ku itself is a big jump, so > allowing the client to pass lets them choose how must certificate > processing they perform. Do you think that it is good, or not? I am not sure that is a good and/or whether this corresponds to the idea of totally centralized validation, i.e. minimal processing at the client side. I tend to say that the client should only be required to use the public key. > > For the requestor, I don't see any other processing than as binary blob > is necessary. Therefore if you have an octet string, the server just > does a binary companies on the data. This is not conformant with the requirements, the server needs some way to match the identifiers with the authenticated identities. To me, an an almost opaque octet string doesn't seem the right approach. In fact, you don't have a binary blob, since you don't allow zeros, for example, if a client want to put a hash of something. I propose the following compromise: Identifiers ::= SEQUENCE OF { CHOICE { opaqueId OCTET STRING, nameId GeneralName } and a statement says that in order to compare for loop detection, a server will do a comparision of the encoded element (including the tag). and use this as the type for requestor and responder Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CC835v013248; Wed, 12 May 2004 05:08:03 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CC83C9013247; Wed, 12 May 2004 05:08:03 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CC7tIf013226 for <ietf-pkix@imc.org>; Wed, 12 May 2004 05:07:56 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms21.chttl.com.tw (ms21 [10.144.2.121]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i4CC7PGG008856; Wed, 12 May 2004 20:07:26 +0800 Received: (from root@localhost) by ms21.chttl.com.tw (8.12.10/8.12.10) id i4CC7QLU029080; Wed, 12 May 2004 20:07:26 +0800 Received: from wcwang ([10.144.133.79]) by ms21.chttl.com.tw (8.12.10/8.12.10) with SMTP id i4CC7OL5028759; Wed, 12 May 2004 20:07:24 +0800 Message-ID: <01b301c43819$ae046d00$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Stefan Santesson" <stefans@microsoft.com>, "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org>, "Steve Hanna" <Steve.Hanna@sun.com> References: <0C3042E92D8A714783E2C44AB9936E1DFBA173@EUR-MSG-03.europe.corp.microsoft.com> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String inencodingRDNs Date: Wed, 12 May 2004 20:07:20 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Stefan, Although I personally prefer to remain it as a 'MUST', but if PKIX WG majority decide to change it to be a 'SHOULD', I will accept that. Another compromise is not to REQUIRE CAs established before the deadline to migrate to UTF8String encoding. That is, if an old CA already use PrintableString encoding it can continue to use PrintableString encoding. One day if the old CA need perform to key rollover, it is STRONGLY RECOMMENDED (MUST?) to take that opportunity to migrate to UTF8String encoding. For CAs established after the deadline, it is STRONGLY RECOMMENDED (MUST?) to adopt UTF8String encoding from the beginning. By doing so, I believe that the infrastructure will ultimately uniform the name encoding. As for the deadline, since the deadline REQUIRED by RFC 3280 had already been past and it did not come true, a new deadline will be needed. Wen-Cheng Wang > > Wen-Cheng, > > I think we agree on the theory parts. No need to repeat them. > > The only disagreement is whether RFC 3280 should REQUIRE UTF-8 where > printable string is sufficient. > > I think it should not do so. It should limit itself to a recommendation > (SHOULD) > > RFC 3280 is an important profile, and while it may seem as a voluntary > profile for many, it is actually something that gets imposed as real > requirements for others. > > Requiring UTF-8 causes interoperability problems, also among those who > have faithfully followed RFC 3280, and thus it is a bad requirement. > > I also pledge that name encoding rollover certificates, as specified in > RFC 3280 is in error and thus has little merit in the real world. > Migration should thus be allowed in such pase that no name > encoding-rollover certificates are needed. > > > ---- > Except from this I agree that we should try to fix X.509 name matching > in RFC 3280 name constraining, which is a bug. But that is a separate > issue. > > /Stefan > > > -----Original Message----- > > From: Wen-Cheng Wang [mailto:wcwang@cht.com.tw] > > Sent: den 12 maj 2004 11:11 > > To: Stefan Santesson; Stephen Kent; Santoni Adriano > > Cc: pgut001@cs.auckland.ac.nz; ietf-pkix@imc.org; Steve Hanna > > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in > > encodingRDNs > > > > > Yes, you can do a lot of cool things in theory. > > > > > > The cruel truth is that your name encoding roll-over certs (e.g. > > > CA1-CA1-p-u-cert) is not self issued according to RFC 3280. > > > > That is why I said that the definition of self-issued certificates in > RFC > > 3280 (or maybe its definition in the X.509 standard itself) should > > be revised or be clarified in son-of-3280 (or in an amendment of > > the X.509 standard as well). > > > > > > > > RFC 3280 section 6.1: > > > A certificate is self-issued if the DNs that appear in the > subject > > > and issuer fields are identical and are not empty. > > > > > > RFC 3280 section 4.1.2.4: > > > (a) attribute values encoded in different types (e.g., > > > PrintableString and BMPString) MAY be assumed to represent > > > different strings; > > > > Remenber that RFC 3280 section 4.1.2.4 also said: > > > > Implementations of this profile are permitted to use the comparison > > algorithm defined in the X.500 series. > > > > In that respect, you can not said that name rollover certificates are > not > > self-issued. > > > > > > > > So in current implementations of RFC 3280, many valid certificates > will > > > fail after introduction of name rollover certs in the chain, because > > > path validation will not treat them as self-issued. > > > > I admit that most CURRENT implementations of RFC 3280 path > > validation will fail to treat name rollover certificates as > self-issued, > > but > > it does not mean that we should not recommend FUTURE revisions > > of those implementations and new implementations to treat them as > > self-issued. > > > > > > > > Further, the name constraint problem persists, adds to this but also > > > exist regardless of this problem. > > > > > > RFC 3280 section 4.2.1.11 > > > When applying restrictions of the form directoryName, an > > > implementation MUST compare DN attributes. At a minimum, > > > implementations MUST perform the DN comparison rules specified in > > > Section 4.1.2.4. > > > > That is why I said RFC 3280 path validation algorithm should also be > > revised. > > > > > > > > Fixing next version of RFC 3280 to include better name matching > rules is > > > a good thing but far from enough to take care of real problems with > > > current infrastructure. > > > > Fixing next version of RFC 3280 does not magically take care of real > > problems with current infrastructure, but it encourge current > > infrastructure > > to migrate to solve the problems. > > > > > > > > We have to try to keep in mind what we make all this hassle for!! > > > > > > This is NOT about whether we should use UTF-8 or not to accommodate > > > complex character sets!! This is about what we are prepared to do to > > > make sure NOTHING is encoded as PrintableString WHEN PrintableString > is > > > sufficient. > > > > > > This old transition rule was created in RFC 2459 before the modern > path > > > validation algorithm was invented and before self-issued > certificates > > > was a special case. > > > > That does not mean that name rollover certificates can not solve the > > mixed name-encoding problem. > > > > > > > > Let's just admit that we overlooked this issue when we created RFC > 3280 > > > and reshaped path validation. It was an understandable human error, > > > which lead to a defect. > > > > Yes, RFC 3280 indeed overlooked this issue. That is why it should be > > revised. > > > > > > > > So let's then also admit that REQUIRING UTF-8 when printableString > is > > > sufficient, is not a very good idea when all practical matters are > > > measured, and that it never should have made it to more than a > > > RECOMMENDATION. > > > > Isn't it the IETF policy to encourge all new implementations to > migrate > > to support UTF-8? The philosophy is that if one encoding is sufficient > to > > cover universal characters, why should we recommend two? > > > > > > > > Whatever we say here, this is a requirement that many implementers > will > > > have to ignore for a many years to come. > > > > Conformance to any IETF RFC is all voluntary. No one can enfoce any > > implementer for that if the implementer chooses to ignore the > requirement > > or recommendation of any IETF RFC. RFC 3280 might not have the > > power to enforce all implementation to migrate to UTF-8 encoding. > > However, as an IETF RFC, it definitely has a great effect. At least, I > > see more and more applications can correctly handle certificates with > > named in UTF-8 encoding. Therefore, I prefer to see son-of-3280 to > > continue requiring conformant implementations to migrate to UTF-8 > > encoding. > > > > Wen-Cheng Wang > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CAOvnK005210; Wed, 12 May 2004 03:24:57 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4CAOvQ7005209; Wed, 12 May 2004 03:24:57 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-dub.microsoft.com (mail-dub.microsoft.com [213.199.128.160]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4CAOtGE005185 for <ietf-pkix@imc.org>; Wed, 12 May 2004 03:24:56 -0700 (PDT) (envelope-from stefans@microsoft.com) Received: from dub-imc-01.europe.corp.microsoft.com ([65.53.196.22]) by mail-dub.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Wed, 12 May 2004 11:24:51 +0100 Received: from 65.53.196.22 by dub-imc-01.europe.corp.microsoft.com (InterScan E-Mail VirusWall NT); Wed, 12 May 2004 11:24:51 +0100 Received: from EUR-MSG-03.europe.corp.microsoft.com ([65.53.192.44]) by dub-imc-01.europe.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Wed, 12 May 2004 11:24:51 +0100 x-mimeole: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encodingRDNs Date: Wed, 12 May 2004 11:24:43 +0100 Message-ID: <0C3042E92D8A714783E2C44AB9936E1DFBA173@EUR-MSG-03.europe.corp.microsoft.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: R: I: R: RFC3280: doubt on the use of UTF8String in encodingRDNs Thread-Index: AcQ4ARpDL01IydVqQZCUq3EkBtxLuQAB08Yg From: "Stefan Santesson" <stefans@microsoft.com> To: "Wen-Cheng Wang" <wcwang@cht.com.tw>, "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org>, "Steve Hanna" <Steve.Hanna@sun.com> X-OriginalArrivalTime: 12 May 2004 10:24:51.0484 (UTC) FILETIME=[5C2375C0:01C4380B] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4CAOuGE005204 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Wen-Cheng, I think we agree on the theory parts. No need to repeat them. The only disagreement is whether RFC 3280 should REQUIRE UTF-8 where printable string is sufficient. I think it should not do so. It should limit itself to a recommendation (SHOULD) RFC 3280 is an important profile, and while it may seem as a voluntary profile for many, it is actually something that gets imposed as real requirements for others. Requiring UTF-8 causes interoperability problems, also among those who have faithfully followed RFC 3280, and thus it is a bad requirement. I also pledge that name encoding rollover certificates, as specified in RFC 3280 is in error and thus has little merit in the real world. Migration should thus be allowed in such pase that no name encoding-rollover certificates are needed. ---- Except from this I agree that we should try to fix X.509 name matching in RFC 3280 name constraining, which is a bug. But that is a separate issue. /Stefan > -----Original Message----- > From: Wen-Cheng Wang [mailto:wcwang@cht.com.tw] > Sent: den 12 maj 2004 11:11 > To: Stefan Santesson; Stephen Kent; Santoni Adriano > Cc: pgut001@cs.auckland.ac.nz; ietf-pkix@imc.org; Steve Hanna > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in > encodingRDNs > > > Yes, you can do a lot of cool things in theory. > > > > The cruel truth is that your name encoding roll-over certs (e.g. > > CA1-CA1-p-u-cert) is not self issued according to RFC 3280. > > That is why I said that the definition of self-issued certificates in RFC > 3280 (or maybe its definition in the X.509 standard itself) should > be revised or be clarified in son-of-3280 (or in an amendment of > the X.509 standard as well). > > > > > RFC 3280 section 6.1: > > A certificate is self-issued if the DNs that appear in the subject > > and issuer fields are identical and are not empty. > > > > RFC 3280 section 4.1.2.4: > > (a) attribute values encoded in different types (e.g., > > PrintableString and BMPString) MAY be assumed to represent > > different strings; > > Remenber that RFC 3280 section 4.1.2.4 also said: > > Implementations of this profile are permitted to use the comparison > algorithm defined in the X.500 series. > > In that respect, you can not said that name rollover certificates are not > self-issued. > > > > > So in current implementations of RFC 3280, many valid certificates will > > fail after introduction of name rollover certs in the chain, because > > path validation will not treat them as self-issued. > > I admit that most CURRENT implementations of RFC 3280 path > validation will fail to treat name rollover certificates as self-issued, > but > it does not mean that we should not recommend FUTURE revisions > of those implementations and new implementations to treat them as > self-issued. > > > > > Further, the name constraint problem persists, adds to this but also > > exist regardless of this problem. > > > > RFC 3280 section 4.2.1.11 > > When applying restrictions of the form directoryName, an > > implementation MUST compare DN attributes. At a minimum, > > implementations MUST perform the DN comparison rules specified in > > Section 4.1.2.4. > > That is why I said RFC 3280 path validation algorithm should also be > revised. > > > > > Fixing next version of RFC 3280 to include better name matching rules is > > a good thing but far from enough to take care of real problems with > > current infrastructure. > > Fixing next version of RFC 3280 does not magically take care of real > problems with current infrastructure, but it encourge current > infrastructure > to migrate to solve the problems. > > > > > We have to try to keep in mind what we make all this hassle for!! > > > > This is NOT about whether we should use UTF-8 or not to accommodate > > complex character sets!! This is about what we are prepared to do to > > make sure NOTHING is encoded as PrintableString WHEN PrintableString is > > sufficient. > > > > This old transition rule was created in RFC 2459 before the modern path > > validation algorithm was invented and before self-issued certificates > > was a special case. > > That does not mean that name rollover certificates can not solve the > mixed name-encoding problem. > > > > > Let's just admit that we overlooked this issue when we created RFC 3280 > > and reshaped path validation. It was an understandable human error, > > which lead to a defect. > > Yes, RFC 3280 indeed overlooked this issue. That is why it should be > revised. > > > > > So let's then also admit that REQUIRING UTF-8 when printableString is > > sufficient, is not a very good idea when all practical matters are > > measured, and that it never should have made it to more than a > > RECOMMENDATION. > > Isn't it the IETF policy to encourge all new implementations to migrate > to support UTF-8? The philosophy is that if one encoding is sufficient to > cover universal characters, why should we recommend two? > > > > > Whatever we say here, this is a requirement that many implementers will > > have to ignore for a many years to come. > > Conformance to any IETF RFC is all voluntary. No one can enfoce any > implementer for that if the implementer chooses to ignore the requirement > or recommendation of any IETF RFC. RFC 3280 might not have the > power to enforce all implementation to migrate to UTF-8 encoding. > However, as an IETF RFC, it definitely has a great effect. At least, I > see more and more applications can correctly handle certificates with > named in UTF-8 encoding. Therefore, I prefer to see son-of-3280 to > continue requiring conformant implementations to migrate to UTF-8 > encoding. > > Wen-Cheng Wang Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C9Let7095269; Wed, 12 May 2004 02:21:40 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4C9LdFA095268; Wed, 12 May 2004 02:21:39 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C9LcD1095224 for <ietf-pkix@imc.org>; Wed, 12 May 2004 02:21:38 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms21.chttl.com.tw (ms21 [10.144.2.121]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i4C9KfGG027527; Wed, 12 May 2004 17:20:41 +0800 Received: (from root@localhost) by ms21.chttl.com.tw (8.12.10/8.12.10) id i4C9KfCx025690; Wed, 12 May 2004 17:20:41 +0800 Received: from wcwang ([10.144.133.79]) by ms21.chttl.com.tw (8.12.10/8.12.10) with SMTP id i4C9KfL5025586; Wed, 12 May 2004 17:20:41 +0800 Message-ID: <004501c43802$6350b460$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Denis Pinkas" <Denis.Pinkas@bull.net>, "Trevor Freeman" <trevorf@exchange.microsoft.com> Cc: "Stephen Kent" <kent@bbn.com>, <ietf-pkix@imc.org> References: <33E7A1613A3480448996047B69308A2F027EEC4C@df-grommit-01.dogfood> <40A100E5.3050208@bull.net> Subject: Re: SCVP & validation policies Date: Wed, 12 May 2004 17:20:37 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I would like to support that even the validation policy OID itself is optional for the client to use. As I stated in my earlier post, if a client fully trust the validation server, then it might not have to specify a validation policy OID in its request. In that case, it means that the client trusts the validation server to adopt the default validation policy predefinied by the management authority of the application domain (such as an enterprise). For example, in scenario 1 and scenario 3 I mentioned in my earlier post, if the client request does not specify any validation policy OID, then server will adopt the default validation policy. For you reference, scenario 1 and scenario 3 are as follows: Scenario 1 (single central-managed vlidation policy) DPV server: only supports a single central-managed vlidation policy the validation policy implies all policy-dependent parameters DPV client: the validation request may optionally specify the vlidation policy OID the validation request needs not specify any policy-dependent parameters Scenario 3 (multiple central-managed vlidation policies with a default validation policy) DPV server: supports multiple central-managed vlidation policies with one of them is the default each validation policy implies its policy-dependent parameters DPV client: the validation request should specify the vlidation policy OID if the request did not specify the vlidation policy OID, then the server will use the default one the validation request needs not to specify any policy-dependent parameters These are the simplest cases of the use of DPV protocol. I hope that basic SCVP syntax should support these simplest cases more clearly. (I mean that the ambiguity in the current syntax should be reduced.) In contrast to these simplest cases, an extreme case of the use of DPV protocol is the case where there is no any predefined validation policy (and hence no default validation policy). This might be the case where a DPV server is used as a public server. (The scenario 4 I mentioned earlier.) Please note that in this extreme case, the client has no way to specify any validation policy OID in its request since there is no any predefined one. In this extreme case, the client will have to submit the full set of parameters for path validation ALGORITHM. I would like to see SCVP to provide an extension mechanism to let the client submit the full set of parameters for path validation ALGORITHM. However, please do not confuse "parameters for validation ALGORITHM" with "parameters for validation POLICY". Also, please make a clear separation of basic SCVP syntax and the extension mechanism for submitting parameters for path validation ALGORITHM. Wen-Cheng Wang > > Trevor, > > > Hi Dennis, > > The OID is optional for the client to use. Servers don't have to support > > this. > > I disagree. The "RFC 3280 compliantbasic validation policy", i.e the one > which only includes basic path validation, with no extra and no shortcuts, > as defined by RFC 3280, will have an OID. Since this policy does not > contains values such as root keys, this validation policy needs additional > parameters. > > Intelligent clients may use this one and specify parameters. > > Thin clients will simply refer to an OID which refers globally to a > validation algorithm and its parameters. > > I have talked to customers and their position is to have flexible validation > policies. As an example, some says: if we want to take purposely the risk of > not testing ARLs, it is our choice. The validation policy should not mandate > to test ARLs, but if for some validation policies it is adequate to do so, > it should be possible to do so. > > > It become a matter of mutual agreement between the client and > > server what the OID represents. > > Not necessarilly. Since the OID is not in the arc of the server, it may be > defined by anyone. If that definition is separately accessible both to the > client and to the server then it can be supported by the server and the > client may refer to it. In such a case there is no need for a mutual agreement. > > > An SCVP server must either comply with any request or return an error > > True. > > > so I don't see the validation policy OID > > needs any special treatment beyond want SCVP already defines. > > I don't see the relationship between the true statement above and the > conclusion. > > As already stated on this list, SCVP stands for "Simple", otherwise we will > have to drop the "S", and then we would have "CVP". :-| > > Denis > > > Trevor Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C9BFge091427; Wed, 12 May 2004 02:11:15 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4C9BF67091426; Wed, 12 May 2004 02:11:15 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C9BEa6091340 for <ietf-pkix@imc.org>; Wed, 12 May 2004 02:11:14 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms8.chttl.com.tw (ms8 [10.144.2.118]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i4C9AdGG026609; Wed, 12 May 2004 17:10:40 +0800 Received: (from root@localhost) by ms8.chttl.com.tw (8.12.10/8.12.10) id i4C9Ad6X024382; Wed, 12 May 2004 17:10:39 +0800 Received: from wcwang ([10.144.133.79]) by ms8.chttl.com.tw (8.12.10/8.12.10) with SMTP id i4C9AcZl024279; Wed, 12 May 2004 17:10:38 +0800 Message-ID: <003801c43800$fc933dc0$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Stefan Santesson" <stefans@microsoft.com>, "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org>, "Steve Hanna" <Steve.Hanna@sun.com> References: <0C3042E92D8A714783E2C44AB9936E1DF83112@EUR-MSG-03.europe.corp.microsoft.com> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encodingRDNs Date: Wed, 12 May 2004 17:10:35 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > Yes, you can do a lot of cool things in theory. > > The cruel truth is that your name encoding roll-over certs (e.g. > CA1-CA1-p-u-cert) is not self issued according to RFC 3280. That is why I said that the definition of self-issued certificates in RFC 3280 (or maybe its definition in the X.509 standard itself) should be revised or be clarified in son-of-3280 (or in an amendment of the X.509 standard as well). > > RFC 3280 section 6.1: > A certificate is self-issued if the DNs that appear in the subject > and issuer fields are identical and are not empty. > > RFC 3280 section 4.1.2.4: > (a) attribute values encoded in different types (e.g., > PrintableString and BMPString) MAY be assumed to represent > different strings; Remenber that RFC 3280 section 4.1.2.4 also said: Implementations of this profile are permitted to use the comparison algorithm defined in the X.500 series. In that respect, you can not said that name rollover certificates are not self-issued. > > So in current implementations of RFC 3280, many valid certificates will > fail after introduction of name rollover certs in the chain, because > path validation will not treat them as self-issued. I admit that most CURRENT implementations of RFC 3280 path validation will fail to treat name rollover certificates as self-issued, but it does not mean that we should not recommend FUTURE revisions of those implementations and new implementations to treat them as self-issued. > > Further, the name constraint problem persists, adds to this but also > exist regardless of this problem. > > RFC 3280 section 4.2.1.11 > When applying restrictions of the form directoryName, an > implementation MUST compare DN attributes. At a minimum, > implementations MUST perform the DN comparison rules specified in > Section 4.1.2.4. That is why I said RFC 3280 path validation algorithm should also be revised. > > Fixing next version of RFC 3280 to include better name matching rules is > a good thing but far from enough to take care of real problems with > current infrastructure. Fixing next version of RFC 3280 does not magically take care of real problems with current infrastructure, but it encourge current infrastructure to migrate to solve the problems. > > We have to try to keep in mind what we make all this hassle for!! > > This is NOT about whether we should use UTF-8 or not to accommodate > complex character sets!! This is about what we are prepared to do to > make sure NOTHING is encoded as PrintableString WHEN PrintableString is > sufficient. > > This old transition rule was created in RFC 2459 before the modern path > validation algorithm was invented and before self-issued certificates > was a special case. That does not mean that name rollover certificates can not solve the mixed name-encoding problem. > > Let's just admit that we overlooked this issue when we created RFC 3280 > and reshaped path validation. It was an understandable human error, > which lead to a defect. Yes, RFC 3280 indeed overlooked this issue. That is why it should be revised. > > So let's then also admit that REQUIRING UTF-8 when printableString is > sufficient, is not a very good idea when all practical matters are > measured, and that it never should have made it to more than a > RECOMMENDATION. Isn't it the IETF policy to encourge all new implementations to migrate to support UTF-8? The philosophy is that if one encoding is sufficient to cover universal characters, why should we recommend two? > > Whatever we say here, this is a requirement that many implementers will > have to ignore for a many years to come. Conformance to any IETF RFC is all voluntary. No one can enfoce any implementer for that if the implementer chooses to ignore the requirement or recommendation of any IETF RFC. RFC 3280 might not have the power to enforce all implementation to migrate to UTF-8 encoding. However, as an IETF RFC, it definitely has a great effect. At least, I see more and more applications can correctly handle certificates with named in UTF-8 encoding. Therefore, I prefer to see son-of-3280 to continue requiring conformant implementations to migrate to UTF-8 encoding. Wen-Cheng Wang Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C8dZHo079650; Wed, 12 May 2004 01:39:35 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4C8dZpi079649; Wed, 12 May 2004 01:39:35 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from ntexch03.office.corp.sia.it (clients.sia.it [193.203.230.22] (may be forged)) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C8dTnH079589 for <ietf-pkix@imc.org>; Wed, 12 May 2004 01:39:34 -0700 (PDT) (envelope-from adriano.santoni@actalis.it) Received: by ntexch03.office.corp.sia.it with Internet Mail Service (5.5.2657.72) id <JZ9SPN4A>; Wed, 12 May 2004 10:39:23 +0200 Message-ID: <BE82E060F5EA2D44A4CFCFFA8363958803B4BD05@ntexch00.office.corp.sia.it> From: Santoni Adriano <adriano.santoni@actalis.it> To: "'Stefan Santesson'" <stefans@microsoft.com> Cc: pgut001@cs.auckland.ac.nz, ietf-pkix@imc.org, Steve Hanna <Steve.Hanna@sun.com>, Wen-Cheng Wang <wcwang@cht.com.tw>, Stephen Kent <kent@bbn.com> Subject: R: R: I: R: RFC3280: doubt on the use of UTF8String in encoding R DNs Date: Wed, 12 May 2004 10:39:27 +0200 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4C8dYnH079642 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Thank you for this additional contribution. "So let's then also admit that REQUIRING UTF-8 when printableString is sufficient, is not a very good idea when all practical matters are measured, ..." This is exactly the point, IMHO. UTF-8 is a clever encoding, probably the best when we CAs have to put names like "Papadopulos" or "Wen-Cheng Wang" into a certificate in its original and "true" character set. "...and that it never should have made it to more than a RECOMMENDATION." This sounds very sensible. I buy it! Adriano -----Messaggio originale----- Da: Stefan Santesson [mailto:stefans@microsoft.com] Inviato: martedì 11 maggio 2004 15.33 A: Wen-Cheng Wang; Stephen Kent; Santoni Adriano Cc: pgut001@cs.auckland.ac.nz; ietf-pkix@imc.org; Steve Hanna Oggetto: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Yes, you can do a lot of cool things in theory. The cruel truth is that your name encoding roll-over certs (e.g. CA1-CA1-p-u-cert) is not self issued according to RFC 3280. RFC 3280 section 6.1: A certificate is self-issued if the DNs that appear in the subject and issuer fields are identical and are not empty. RFC 3280 section 4.1.2.4: (a) attribute values encoded in different types (e.g., PrintableString and BMPString) MAY be assumed to represent different strings; So in current implementations of RFC 3280, many valid certificates will fail after introduction of name rollover certs in the chain, because path validation will not treat them as self-issued. Further, the name constraint problem persists, adds to this but also exist regardless of this problem. RFC 3280 section 4.2.1.11 When applying restrictions of the form directoryName, an implementation MUST compare DN attributes. At a minimum, implementations MUST perform the DN comparison rules specified in Section 4.1.2.4. Fixing next version of RFC 3280 to include better name matching rules is a good thing but far from enough to take care of real problems with current infrastructure. We have to try to keep in mind what we make all this hassle for!! This is NOT about whether we should use UTF-8 or not to accommodate complex character sets!! This is about what we are prepared to do to make sure NOTHING is encoded as PrintableString WHEN PrintableString is sufficient. This old transition rule was created in RFC 2459 before the modern path validation algorithm was invented and before self-issued certificates was a special case. Let's just admit that we overlooked this issue when we created RFC 3280 and reshaped path validation. It was an understandable human error, which lead to a defect. So let's then also admit that REQUIRING UTF-8 when printableString is sufficient, is not a very good idea when all practical matters are measured, and that it never should have made it to more than a RECOMMENDATION. Whatever we say here, this is a requirement that many implementers will have to ignore for a many years to come. Stefan Santesson Program Manager, Windows Security > -----Original Message----- > From: Wen-Cheng Wang [mailto:wcwang@cht.com.tw] > Sent: den 7 maj 2004 04:29 > To: Stefan Santesson; Stephen Kent; Santoni Adriano > Cc: pgut001@cs.auckland.ac.nz; ietf-pkix@imc.org; Steve Hanna > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding > RDNs > > I do'nt see why name rollover certificates can not solve the mixed > encoding problem if the definition of a self-issued certificate and > the path validation algorithm in RFC 3280 is revised? > > Let me use an example to explain this. > > Suppose the original certificate chain is: > > Trust-Anchor ...... CA1-CA2-p-p-cert CA2-EE1-p-p-cert > > where "p-p" means "issuer name in PrintableString" and "subject name > in PrintableString". > > Suppose CA2 take the name rollover, it will issue the following two > self-issued certificates: > CA2-CA2-p-u-cert > (i.e., issuer name in PrintableString, > subject name in UTF8String) > CA2-CA2-u-p-cert > (i.e., issuer name in UTF8String, > subject name in PrintableString) > > Suppose CA1 also take the name rollover, it will issue > the following two self-issued certificates: > CA1-CA1-p-u-cert > (i.e., issuer name in PrintableString, > subject name in UTF8String) > CA1-CA1-u-p-cert > (i.e., issuer name in UTF8String, > subject name in PrintableString) > > Suppose CA2 start to issue EE certificate with UTF8String encoding, > and it issues the following EE certificate: > CA2-EE2-u-u-cert > > Then, > EE1 can still be reached by the original certificate chain: > > Trust-Anchor ...... CA1-CA2-p-p-cert CA2-EE1-p-p-cert > > EE1 can also be reached by the following new certificate chain: > > Trust-Anchor ...... CA1-CA1-u-p-cert CA1-CA2-p-p-cert > CA2-EE1-p-p-cert > (Assume there is another CA issues a cross-certificate to CA1 in > UTF8String encoding.) > > EE2 can be reached by the following new certificate chain: > > Trust-Anchor ...... CA1-CA2-p-p-cert CA2-CA2-p-u-cert > CA2-EE2-u-u-cert > > One day, if CA1 renews the cross-certificate to CA2 and adopts > UTF8String encoding in the new cross-certificate, it will issue the > following cross-certificate: > CA1-CA2-u-u-cert > > Then, > EE1 can also be reached by the following new certificate chain: > > Trust-Anchor ...... CA1-CA1-p-u-cert CA1-CA2-u-u-cert > CA2-CA2-u-p-cert CA2-EE1-p-p-cert > > EE2 can also be reached by the following new certificate chain: > > Trust-Anchor ...... CA1-CA2-u-u-cert CA2-EE2-u-u-cert > > As you can see, in theory you can always find at least one certificate > chain to EE1 (i.e., old cert with PrintableString) or EE2 (i.e., new cert > with UTF8String). However, as I said before, we should discuss how to > amend RFC 3280 to support such certificate chaining. > > Wen-Cheng Wang > > ----- Original Message ----- > From: "Stefan Santesson" <stefans@microsoft.com> > To: "Wen-Cheng Wang" <wcwang@cht.com.tw>; "Stephen Kent" <kent@bbn.com>; > "Santoni Adriano" <adriano.santoni@actalis.it> > Cc: <pgut001@cs.auckland.ac.nz>; <ietf-pkix@imc.org> > Sent: Thursday, May 06, 2004 9:32 PM > Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding > RDNs > > > > > > Name rollover certificates only takes care of CA name changes. It > > doesn't solve the problem if subject names have mixed encoding in the > > forest of user certificates, while the CAs names are intact. > > > > Neither does it solve when the constraint is placed above the CA > > rollover. > > > > Finally, CA name rollover will most likely bust path validation using > > other constraining mechanisms because they may just not be recognized as > > self issued due to different name encoding. > > > > Stefan Santesson > > Program Manager, Windows Security > > *******************Internet Email Confidentiality Footer******************* Qualsiasi utilizzo non autorizzato del presente messaggio nonche' dei suoi allegati e' vietato e potrebbe costituire reato. Se lei ha ricevuto erroneamente il presente messaggio, Le saremmo grati se, via e-mail, ce ne comunicasse la ricezione e provvedesse alla distruzione del messaggio stesso e dei suoi eventuali allegati. Le dichiarazioni contenute nel presente messaggio nonche' nei suoi eventuali allegati devono essere attribuite esclusivamente al mittente e non possono essere considerate come trasmesse o autorizzate da SIA S.p.A.; le medesime dichiarazioni non impegnano SIA S.p.A. nei confronti del destinatario o di terzi. SIA S.p.A. non si assume alcuna responsabilita' per eventuali intercettazioni, modifiche o danneggiamenti del presente messaggio e-mail. Any unauthorized use of this e-mail or any of its attachments is prohibited and could constitute an offence. If you are not the intended addressee please advise immediately the sender by using the reply facility in your e-mail software and destroy the message and its attachments. The statements and opinions expressed in this e-mail message are those of the author of the message and do not necessarily represent those of SIA. Besides, The contents of this message shall be understood as neither given nor endorsed by SIA S.p.A.. SIA S.p.A. does not accept liability for corruption, interception or amendment, if any, or the consequences thereof. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C7P64w052768; Wed, 12 May 2004 00:25:06 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4C7P6rQ052767; Wed, 12 May 2004 00:25:06 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from odin2.bull.net (odin2.bull.net [192.90.70.84]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C7P4Zb052684 for <ietf-pkix@imc.org>; Wed, 12 May 2004 00:25:05 -0700 (PDT) (envelope-from Denis.Pinkas@bull.net) Received: from clbull.frcl.bull.fr (IDENT:root@clbull.frcl.bull.fr [129.182.8.31]) by odin2.bull.net (8.9.3/8.9.3) with ESMTP id JAA59042; Wed, 12 May 2004 09:34:21 +0200 Received: from bull.net (frcls4013.frcl.bull.fr [129.182.108.120]) by clbull.frcl.bull.fr (8.9.3/8.9.3) with ESMTP id JAA11982; Wed, 12 May 2004 09:23:42 +0200 Message-ID: <40A1D168.4020402@bull.net> Date: Wed, 12 May 2004 09:25:28 +0200 From: Denis Pinkas <Denis.Pinkas@bull.net> Organization: Bull SA. User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en, fr MIME-Version: 1.0 To: Trevor Freeman <trevorf@exchange.microsoft.com> CC: ietf-pkix@imc.org Subject: Re: SCVP & validation policies References: <33E7A1613A3480448996047B69308A2F02962917@df-grommit-01.dogfood> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4C7P5Zb052762 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Trevor, > You also have to understand enough so that the administrative > configuration of the client is generating the right kind of request. > > SCVP client is ?only following orders? in many instances and has from a The SCVP client is “only following orders” in many instances and has > from a perspective no idea if it is the right kind of request, only if it > is well formed to the rules it has been given. Both the administrator of > the client and the administrator of the server has to understand in much > grater detail what the given policy is. Once you abstract though the > OID, you are placing a greater burden on the OOB admin to admin process. The client does not have to "understand" what the validation given policy is. The client only needs to know that, for a given application a given OID for the validation policy is needed. He does not need to have a clue of the details of the validation policy which may remain fully opaque to him. This means that any validation specific parameter must *not* be part of the standard request. Denis > Trevor > > > > ------------------------------------------------------------------------ > > From: James Jing [mailto:jjing@betrusted.com] > Sent: Tuesday, May 11, 2004 5:09 PM > To: Trevor Freeman; ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > > Agree that the client needs to have all the data in order to build the > request. However, I think ?understand? is too strong a requirement. The > client may not need to ?understand? the data, some of which may be > factors affecting the server only. Perhaps I am too pedantic about the > word ?understand?. > > > > Back to indirection, if ?understanding? is not a requirement, then the > client need not to be aware that an OID is an indirection to a set of > policies. > > > > James > > > > ------------------------------------------------------------------------ > > From: Trevor Freeman [mailto:trevorf@exchange.microsoft.com] > Sent: Wednesday, 12 May 2004 9:49 AM > To: James Jing; ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > > The client has to understand what is required to build a request i.e. > What data does it need to supply to the server. It does not need to be > aware of the factors which effect the server. > Trevor > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On Behalf Of James Jing > Sent: Tuesday, May 11, 2004 4:01 PM > To: ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > > I think policy OIDs should remain abstract, or at least can be abstract. > That is that there is no requirement for the client software to > understand the parameters. > > I am more thinking of using SCVP to validate a certificate in the > context of a transaction, where the policy OID in SCVP might be used to > identify such a context to the server. > > For example, a certificate validation request is carried out with regard > to the acceptability of the holder's creditworthiness in the context of > a credit-based transaction. In this case, the policy OID could be used > to indicate the class of assuarances that the CA's CSP provides for, > resulting in the server selecting a subset of acceptable trust anchors. > > James > > > > -----Original Message----- > From: Stephen Kent [mailto:kent@bbn.com] > Sent: Friday, 7 May 2004 6:28 AM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > > At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: >>Hi Steve, >>If you want to have a shorthand notation for a validation policy, given > >>we have a finite set of parameters, the I don't see the value of >>indirection via an OID. Using a hash of the set of inputs provides a >>way of expressing the same thing without introducing ambiguity of does >>both parties have a common understating of that this means. At some >>point someone has to define the policy in terms of the 3280 input > parameters. >>That definition is just as applicable and consumable by the client as a > >>server therefore what did we gain from the shorthand notation? It would > >>be pretty trivial to define some XML which would allow that set of >>validation parameters to be imported by any SCVP entity. >>Trevor > > Trevor, > > It is not indirection via an OID, it is identification via an OID. > Clients never have to be aware of the parameters if we use an OID, and > simplification of clients is one of the goals of SCVP. So I cannot agree > with your comment that clients have to be able to consume these > parameters. > > Given that as a basis, it seems that we are debating whether to use OIDs > or to create some other form of identifier via hashing parameters. I > don't see the latter as attractive, since it requires additional > specification of the canonicalization and encoding of the parameters to > ensure consistency in hash computation between client and server. > > Steve > > > > This e-mail, and any attachments hereto, is intended only for use by the > named addressee(s) and may contain legally privileged and/or > confidential information. If you are not the intended recipient, you > are hereby notified that any dissemination, distribution or copying of > this e-mail, and any attachments hereto, is strictly prohibited. If you > have received this transmission in error, please notify me immediately > and permanently delete the original and all copies and printouts of this > e-mail. > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C15SuY077623; Tue, 11 May 2004 18:05:28 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4C15SsG077622; Tue, 11 May 2004 18:05:28 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C15R2C077604 for <ietf-pkix@imc.org>; Tue, 11 May 2004 18:05:27 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 11 May 2004 18:05:28 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 18:05:27 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 18:05:18 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPartTM-000-a742e905-c6e4-4c09-8d0f-2771da0c3bda" Subject: RE: SCVP & validation policies Date: Tue, 11 May 2004 18:05:29 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F02962917@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ3snpRSDr1tTG9S92N6pIM1cyrMQAAQFaAAAI2WyA= From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "James Jing" <jjing@betrusted.com>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 12 May 2004 01:05:18.0270 (UTC) FILETIME=[30F18DE0:01C437BD] Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multi-part message in MIME format. ------=_NextPartTM-000-a742e905-c6e4-4c09-8d0f-2771da0c3bda Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C437BD.33B87440" ------_=_NextPart_001_01C437BD.33B87440 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable You also have to understand enough so that the administrative configuration of the client is generating the right kind of request. The SCVP client is "only following orders" in many instances and has from a perspective no idea if it is the right kind of request, only if it is well formed to the rules it has been given. Both the administrator of the client and the administrator of the server has to understand in much grater detail what the given policy is. Once you abstract though the OID, you are placing a greater burden on the OOB admin to admin process. Trevor =20 ________________________________ From: James Jing [mailto:jjing@betrusted.com]=20 Sent: Tuesday, May 11, 2004 5:09 PM To: Trevor Freeman; ietf-pkix@imc.org Subject: RE: SCVP & validation policies =20 Agree that the client needs to have all the data in order to build the request. However, I think "understand" is too strong a requirement. The client may not need to "understand" the data, some of which may be factors affecting the server only. Perhaps I am too pedantic about the word "understand". =20 Back to indirection, if "understanding" is not a requirement, then the client need not to be aware that an OID is an indirection to a set of policies. =20 James =20 ________________________________ From: Trevor Freeman [mailto:trevorf@exchange.microsoft.com]=20 Sent: Wednesday, 12 May 2004 9:49 AM To: James Jing; ietf-pkix@imc.org Subject: RE: SCVP & validation policies =20 The client has to understand what is required to build a request i.e.=20 What data does it need to supply to the server. It does not need to be=20 aware of the factors which effect the server.=20 Trevor=20 -----Original Message-----=20 From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of James Jing=20 Sent: Tuesday, May 11, 2004 4:01 PM=20 To: ietf-pkix@imc.org=20 Subject: RE: SCVP & validation policies=20 =20 I think policy OIDs should remain abstract, or at least can be abstract. That is that there is no requirement for the client software to=20 understand the parameters.=20 I am more thinking of using SCVP to validate a certificate in the=20 context of a transaction, where the policy OID in SCVP might be used to=20 identify such a context to the server.=20 For example, a certificate validation request is carried out with regard to the acceptability of the holder's creditworthiness in the context of=20 a credit-based transaction. In this case, the policy OID could be used=20 to indicate the class of assuarances that the CA's CSP provides for,=20 resulting in the server selecting a subset of acceptable trust anchors.=20 James=20 =20 -----Original Message-----=20 From: Stephen Kent [mailto:kent@bbn.com]=20 Sent: Friday, 7 May 2004 6:28 AM=20 To: Trevor Freeman=20 Cc: ietf-pkix@imc.org=20 Subject: RE: SCVP & validation policies=20 =20 At 12:10 PM -0700 5/6/04, Trevor Freeman wrote:=20 >Hi Steve,=20 >If you want to have a shorthand notation for a validation policy, given >we have a finite set of parameters, the I don't see the value of=20 >indirection via an OID. Using a hash of the set of inputs provides a=20 >way of expressing the same thing without introducing ambiguity of does=20 >both parties have a common understating of that this means. At some=20 >point someone has to define the policy in terms of the 3280 input=20 parameters.=20 >That definition is just as applicable and consumable by the client as a >server therefore what did we gain from the shorthand notation? It would >be pretty trivial to define some XML which would allow that set of=20 >validation parameters to be imported by any SCVP entity.=20 >Trevor=20 Trevor,=20 It is not indirection via an OID, it is identification via an OID.=20 Clients never have to be aware of the parameters if we use an OID, and=20 simplification of clients is one of the goals of SCVP. So I cannot agree with your comment that clients have to be able to consume these=20 parameters.=20 Given that as a basis, it seems that we are debating whether to use OIDs or to create some other form of identifier via hashing parameters. I=20 don't see the latter as attractive, since it requires additional=20 specification of the canonicalization and encoding of the parameters to=20 ensure consistency in hash computation between client and server.=20 Steve=20 =20 This e-mail, and any attachments hereto, is intended only for use by the named addressee(s) and may contain legally privileged and/or confidential information. If you are not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this e-mail, and any attachments hereto, is strictly prohibited. If you have received this transmission in error, please notify me immediately and permanently delete the original and all copies and printouts of this e-mail. ------_=_NextPart_001_01C437BD.33B87440 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable <html> <head> <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = charset=3Dus-ascii"> <meta name=3DGenerator content=3D"Microsoft Word 11 (filtered)"> <title>RE: SCVP & validation policies</title> <style> <!-- /* Font Definitions */ @font-face {font-family:Tahoma; panose-1:2 11 6 4 3 5 4 4 2 4;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0pt; margin-bottom:.0001pt; font-size:12.0pt; font-family:"Times New Roman";} a:link, span.MsoHyperlink {color:blue; text-decoration:underline;} a:visited, span.MsoHyperlinkFollowed {color:blue; text-decoration:underline;} p.MsoAutoSig, li.MsoAutoSig, div.MsoAutoSig {margin:0pt; margin-bottom:.0001pt; font-size:12.0pt; font-family:"Times New Roman";} p {margin-right:0pt; margin-left:0pt; font-size:12.0pt; font-family:"Times New Roman";} span.EmailStyle19 {font-family:Arial; color:navy;} span.EmailStyle20 {font-family:Arial; color:navy;} @page Section1 {size:595.3pt 841.9pt; margin:72.0pt 90.0pt 72.0pt 90.0pt;} div.Section1 {page:Section1;} --> </style> </head> <body lang=3DEN-US link=3Dblue vlink=3Dblue> <div class=3DSection1> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'>You also have to understand enough = so that the administrative configuration of the client is generating the right = kind of request. The SCVP client is “only following orders” in many = instances and has from a perspective no idea if it is the right kind of request, = only if it is well formed to the rules it has been given. Both the administrator = of the client and the administrator of the server has to understand in much = grater detail what the given policy is. Once you abstract though the OID, you = are placing a greater burden on the OOB admin to admin = process.</span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'>Trevor</span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'> </span></font></p> <div> <div class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><font = size=3D3 face=3D"Times New Roman"><span style=3D'font-size:12.0pt'> <hr size=3D2 width=3D"100%" align=3Dcenter tabindex=3D-1> </span></font></div> <p class=3DMsoNormal><b><font size=3D2 face=3DTahoma><span = style=3D'font-size:10.0pt; font-family:Tahoma;font-weight:bold'>From:</span></font></b><font = size=3D2 face=3DTahoma><span style=3D'font-size:10.0pt;font-family:Tahoma'> James = Jing [mailto:jjing@betrusted.com] <br> <b><span style=3D'font-weight:bold'>Sent:</span></b> Tuesday, May 11, = 2004 5:09 PM<br> <b><span style=3D'font-weight:bold'>To:</span></b> Trevor Freeman; = ietf-pkix@imc.org<br> <b><span style=3D'font-weight:bold'>Subject:</span></b> RE: SCVP & = validation policies</span></font></p> </div> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = style=3D'font-size: 12.0pt'> </span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = lang=3DEN-AU style=3D'font-size:10.0pt;font-family:Arial;color:navy'>Agree that the = client needs to have all the data in order to build the request. However, I = think “understand” is too strong a requirement. The client may not = need to “understand” the data, some of which may be factors = affecting the server only. Perhaps I am too pedantic about the word “understand”.</span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = lang=3DEN-AU style=3D'font-size:10.0pt;font-family:Arial;color:navy'> </span></fo= nt></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = lang=3DEN-AU style=3D'font-size:10.0pt;font-family:Arial;color:navy'>Back to = indirection, if “understanding” is not a requirement, then the client need = not to be aware that an OID is an indirection to a set of = policies.</span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = lang=3DEN-AU style=3D'font-size:10.0pt;font-family:Arial;color:navy'> </span></fo= nt></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = lang=3DEN-AU style=3D'font-size:10.0pt;font-family:Arial;color:navy'>James</span></fon= t></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = lang=3DEN-AU style=3D'font-size:10.0pt;font-family:Arial;color:navy'> </span></fo= nt></p> <div> <div class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><font = size=3D3 face=3D"Times New Roman"><span style=3D'font-size:12.0pt'> <hr size=3D2 width=3D"100%" align=3Dcenter tabindex=3D-1> </span></font></div> <p class=3DMsoNormal><b><font size=3D2 face=3DTahoma><span = style=3D'font-size:10.0pt; font-family:Tahoma;font-weight:bold'>From:</span></font></b><font = size=3D2 face=3DTahoma><span style=3D'font-size:10.0pt;font-family:Tahoma'> = Trevor Freeman [mailto:trevorf@exchange.microsoft.com] <br> <b><span style=3D'font-weight:bold'>Sent:</span></b> Wednesday, 12 May = 2004 9:49 AM<br> <b><span style=3D'font-weight:bold'>To:</span></b> James Jing; = ietf-pkix@imc.org<br> <b><span style=3D'font-weight:bold'>Subject:</span></b> RE: SCVP & = validation policies</span></font></p> </div> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = lang=3DEN-AU style=3D'font-size:12.0pt'> </span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>The client has to understand what is required to build a request = i.e.</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>What = data does it need to supply to the server. It does not need to be</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>aware of the factors which effect the server.</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Trevor</span></font><span lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>-----Original Message-----</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>From: owner-ietf-pkix@mail.imc.org [<a = href=3D"mailto:owner-ietf-pkix@mail.imc.org">mailto:owner-ietf-pkix@mail.= imc.org</a>]</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>On = Behalf Of James Jing</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Sent: Tuesday, May 11, 2004 4:01 PM</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>To: = ietf-pkix@imc.org</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Subject: RE: SCVP & validation policies</span></font><span lang=3DEN-AU> </span></p> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = lang=3DEN-AU style=3D'font-size:12.0pt'> </span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>I think policy OIDs should remain abstract, or at least can be = abstract.</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>That = is that there is no requirement for the client software to</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>understand the parameters.</span></font><span lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>I am more thinking of using SCVP to validate a certificate in = the</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>context of a transaction, where the policy OID in SCVP might be used = to</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>identify such a context to the server. </span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>For example, a certificate validation request is carried out with = regard</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>to = the acceptability of the holder's creditworthiness in the context = of</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>a = credit-based transaction. In this case, the policy OID could be = used</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>to = indicate the class of assuarances that the CA's CSP provides for,</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>resulting in the server selecting a subset of acceptable trust = anchors.</span></font><span lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>James</span></font><span lang=3DEN-AU> </span></p> <p class=3DMsoNormal style=3D'margin-bottom:12.0pt'><font size=3D3 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:12.0pt'> </span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>-----Original Message-----</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>From: Stephen Kent [<a href=3D"mailto:kent@bbn.com">mailto:kent@bbn.com</a>] = </span></font><span lang=3DEN-AU><br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Sent: Friday, 7 May 2004 6:28 AM</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>To: = Trevor Freeman</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>Cc: = ietf-pkix@imc.org</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Subject: RE: SCVP & validation policies</span></font><span lang=3DEN-AU> </span></p> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = lang=3DEN-AU style=3D'font-size:12.0pt'> </span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>At 12:10 PM -0700 5/6/04, Trevor Freeman wrote:</span></font><span = lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>Hi Steve,</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>If you want to have a shorthand notation for a validation policy, = given</span></font><span lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>we have a finite set of parameters, the I don't see the value of = </span></font><span lang=3DEN-AU><br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>indirection via an OID. Using a hash of the set of inputs provides a = </span></font><span lang=3DEN-AU><br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>way of expressing the same thing without introducing ambiguity of does = </span></font><span lang=3DEN-AU><br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>both parties have a common understating of that this means. At some = </span></font><span lang=3DEN-AU><br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>point someone has to define the policy in terms of the 3280 input</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>parameters.</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>That definition is just as applicable and consumable by the client as = a</span></font><span lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>server therefore what did we gain from the shorthand notation? It = would</span></font><span lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>be pretty trivial to define some XML which would allow that set of = </span></font><span lang=3DEN-AU><br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>validation parameters to be imported by any SCVP entity.</span></font><span = lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>>Trevor</span></font><span lang=3DEN-AU> = </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Trevor</span></font><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>,</span></font><span lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>It is not indirection via an OID, it is identification via an OID. = </span></font><span lang=3DEN-AU><br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Clients never have to be aware of the parameters if we use an OID, = and</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>simplification of clients is one of the goals of SCVP. So I cannot = agree</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>with = your comment that clients have to be able to consume these</span></font><span = lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>parameters.</span></font><span lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Given that as a basis, it seems that we are debating whether to use = OIDs</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU style=3D'font-size:10.0pt'>or = to create some other form of identifier via hashing parameters. I</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>don't see the latter as attractive, since it requires additional</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>specification of the canonicalization and encoding of the parameters = to</span></font><span lang=3DEN-AU> <br> </span><font size=3D2><span lang=3DEN-AU = style=3D'font-size:10.0pt'>ensure consistency in hash computation between client and server.</span></font><span = lang=3DEN-AU> </span></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>Steve</span></font><span lang=3DEN-AU> </span></p> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = lang=3DEN-AU style=3D'font-size:12.0pt'> </span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span lang=3DEN-AU = style=3D'font-size:10.0pt'>This e-mail, and any attachments hereto, is intended only for use by the = named addressee(s) and may contain legally privileged and/or confidential information. If you are not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this e-mail, = and any attachments hereto, is strictly prohibited. If you have = received this transmission in error, please notify me immediately and permanently = delete the original and all copies and printouts of this e-mail.</span></font></p> </div> </body> </html> ------_=_NextPart_001_01C437BD.33B87440-- ------=_NextPartTM-000-a742e905-c6e4-4c09-8d0f-2771da0c3bda-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C09M4a074528; Tue, 11 May 2004 17:09:22 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4C09MRU074527; Tue, 11 May 2004 17:09:22 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from figg.securenet.com.au (ns2.isecure.com.au [202.125.4.72]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4C09LAq074482 for <ietf-pkix@imc.org>; Tue, 11 May 2004 17:09:21 -0700 (PDT) (envelope-from jjing@betrusted.com) Received: from iron.securenet.com.au (iron.isecure.com.au [202.125.4.94] (may be forged)) by figg.securenet.com.au (8.12.3/8.12.3/Debian-6.6) with ESMTP id i4C096i3002804; Wed, 12 May 2004 10:09:06 +1000 Received: (from uucp@localhost) by iron.securenet.com.au (8.12.6/8.12.6) id i4C096WK000025; Wed, 12 May 2004 10:09:06 +1000 (EST) Received: from nodnsquery(10.11.3.10) by iron.securenet.com.au via csmap (V6.0) id srcAAAt4aada; Wed, 12 May 04 10:09:06 +1000 Received: from cbrms01.securenet.com.au (localhost [127.0.0.1]) by gibbons.securenet.com.au (8.12.3/8.12.3/Debian-7woody) with ESMTP id i4C0956Y028363; Wed, 12 May 2004 10:09:05 +1000 Received: from AMALTHEA.securenet.com.au ([192.168.100.30]) by cbrms01.securenet.com.au with Microsoft SMTPSVC(5.0.2195.6713); Wed, 12 May 2004 10:04:35 +1000 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C437B5.4BDF985F" Subject: RE: SCVP & validation policies X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3 Date: Wed, 12 May 2004 10:08:47 +1000 Message-ID: <85DA9C6C3DD8C74F8A8611815C635F8D4F6570@AMALTHEA.securenet.com.au> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies Thread-Index: AcQ3snpRSDr1tTG9S92N6pIM1cyrMQAAQFaA From: "James Jing" <jjing@betrusted.com> To: "Trevor Freeman" <trevorf@exchange.microsoft.com>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 12 May 2004 00:04:35.0327 (UTC) FILETIME=[B59490F0:01C437B4] Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multi-part message in MIME format. ------_=_NextPart_001_01C437B5.4BDF985F Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable Agree that the client needs to have all the data in order to build the request. However, I think "understand" is too strong a requirement. The client may not need to "understand" the data, some of which may be factors affecting the server only. Perhaps I am too pedantic about the word "understand". =20 Back to indirection, if "understanding" is not a requirement, then the client need not to be aware that an OID is an indirection to a set of policies. =20 James =20 _____ =20 From: Trevor Freeman [mailto:trevorf@exchange.microsoft.com]=20 Sent: Wednesday, 12 May 2004 9:49 AM To: James Jing; ietf-pkix@imc.org Subject: RE: SCVP & validation policies =20 The client has to understand what is required to build a request i.e.=20 What data does it need to supply to the server. It does not need to be=20 aware of the factors which effect the server.=20 Trevor=20 -----Original Message-----=20 From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of James Jing=20 Sent: Tuesday, May 11, 2004 4:01 PM=20 To: ietf-pkix@imc.org=20 Subject: RE: SCVP & validation policies=20 =20 I think policy OIDs should remain abstract, or at least can be abstract. That is that there is no requirement for the client software to=20 understand the parameters.=20 I am more thinking of using SCVP to validate a certificate in the=20 context of a transaction, where the policy OID in SCVP might be used to=20 identify such a context to the server.=20 For example, a certificate validation request is carried out with regard to the acceptability of the holder's creditworthiness in the context of=20 a credit-based transaction. In this case, the policy OID could be used=20 to indicate the class of assuarances that the CA's CSP provides for,=20 resulting in the server selecting a subset of acceptable trust anchors.=20 James=20 =20 -----Original Message-----=20 From: Stephen Kent [mailto:kent@bbn.com]=20 Sent: Friday, 7 May 2004 6:28 AM=20 To: Trevor Freeman=20 Cc: ietf-pkix@imc.org=20 Subject: RE: SCVP & validation policies=20 =20 At 12:10 PM -0700 5/6/04, Trevor Freeman wrote:=20 >Hi Steve,=20 >If you want to have a shorthand notation for a validation policy, given >we have a finite set of parameters, the I don't see the value of=20 >indirection via an OID. Using a hash of the set of inputs provides a=20 >way of expressing the same thing without introducing ambiguity of does=20 >both parties have a common understating of that this means. At some=20 >point someone has to define the policy in terms of the 3280 input=20 parameters.=20 >That definition is just as applicable and consumable by the client as a >server therefore what did we gain from the shorthand notation? It would >be pretty trivial to define some XML which would allow that set of=20 >validation parameters to be imported by any SCVP entity.=20 >Trevor=20 Trevor,=20 It is not indirection via an OID, it is identification via an OID.=20 Clients never have to be aware of the parameters if we use an OID, and=20 simplification of clients is one of the goals of SCVP. So I cannot agree with your comment that clients have to be able to consume these=20 parameters.=20 Given that as a basis, it seems that we are debating whether to use OIDs or to create some other form of identifier via hashing parameters. I=20 don't see the latter as attractive, since it requires additional=20 specification of the canonicalization and encoding of the parameters to=20 ensure consistency in hash computation between client and server.=20 Steve=20 =20 This e-mail, and any attachments hereto, is intended only for use by the named addressee(s) and may contain legally privileged and/or confidential information. If you are not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this e-mail, and any attachments hereto, is strictly prohibited. If you have received this transmission in error, please notify me immediately and permanently delete the original and all copies and printouts of this e-mail. ------_=_NextPart_001_01C437B5.4BDF985F Content-Type: text/html; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable <html xmlns:v=3D"urn:schemas-microsoft-com:vml" = xmlns:o=3D"urn:schemas-microsoft-com:office:office" = xmlns:w=3D"urn:schemas-microsoft-com:office:word" = xmlns=3D"http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv=3DContent-Type content=3D"text/html; = charset=3Dus-ascii"> <meta name=3DGenerator content=3D"Microsoft Word 11 (filtered medium)"> <!--[if !mso]> <style> v\:* {behavior:url(#default#VML);} o\:* {behavior:url(#default#VML);} w\:* {behavior:url(#default#VML);} .shape {behavior:url(#default#VML);} </style> <![endif]--> <title>RE: SCVP & validation policies</title> <style> <!-- /* Font Definitions */ @font-face {font-family:Tahoma; panose-1:2 11 6 4 3 5 4 4 2 4;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0cm; margin-bottom:.0001pt; font-size:12.0pt; font-family:"Times New Roman";} a:link, span.MsoHyperlink {color:blue; text-decoration:underline;} a:visited, span.MsoHyperlinkFollowed {color:blue; text-decoration:underline;} p.MsoAutoSig, li.MsoAutoSig, div.MsoAutoSig {margin:0cm; margin-bottom:.0001pt; font-size:12.0pt; font-family:"Times New Roman";} p {mso-margin-top-alt:auto; margin-right:0cm; mso-margin-bottom-alt:auto; margin-left:0cm; font-size:12.0pt; font-family:"Times New Roman";} span.EmailStyle18 {mso-style-type:personal-reply; font-family:Arial; color:navy;} @page Section1 {size:595.3pt 841.9pt; margin:72.0pt 90.0pt 72.0pt 90.0pt;} div.Section1 {page:Section1;} --> </style> </head> <body lang=3DEN-AU link=3Dblue vlink=3Dblue> <div class=3DSection1> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'>Agree that the client needs to have = all the data in order to build the request. However, I think = “understand” is too strong a requirement. The client may not need to = “understand” the data, some of which may be factors affecting the server only. = Perhaps I am too pedantic about the word = “understand”.<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'>Back to indirection, if = “understanding” is not a requirement, then the client need not to be aware that an OID = is an indirection to a set of policies.<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'>James<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 color=3Dnavy face=3DArial><span = style=3D'font-size: 10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p> <div> <div class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><font = size=3D3 face=3D"Times New Roman"><span lang=3DEN-US style=3D'font-size:12.0pt'> <hr size=3D2 width=3D"100%" align=3Dcenter tabindex=3D-1> </span></font></div> <p class=3DMsoNormal><b><font size=3D2 face=3DTahoma><span lang=3DEN-US style=3D'font-size:10.0pt;font-family:Tahoma;font-weight:bold'>From:</spa= n></font></b><font size=3D2 face=3DTahoma><span lang=3DEN-US = style=3D'font-size:10.0pt;font-family:Tahoma'> Trevor Freeman [mailto:trevorf@exchange.microsoft.com] <br> <b><span style=3D'font-weight:bold'>Sent:</span></b> Wednesday, 12 May = 2004 9:49 AM<br> <b><span style=3D'font-weight:bold'>To:</span></b> James Jing; = ietf-pkix@imc.org<br> <b><span style=3D'font-weight:bold'>Subject:</span></b> RE: SCVP & = validation policies</span></font><span lang=3DEN-US><o:p></o:p></span></p> </div> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = style=3D'font-size: 12.0pt'><o:p> </o:p></span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>The client has to understand what is required to build a request = i.e.</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>What data does it need = to supply to the server. It does not need to be</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>aware of the factors = which effect the server.</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>Trevor</span></font> = <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>-----Original Message-----</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>From: = owner-ietf-pkix@mail.imc.org [<a = href=3D"mailto:owner-ietf-pkix@mail.imc.org">mailto:owner-ietf-pkix@mail.= imc.org</a>]</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>On Behalf Of James = Jing</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>Sent: Tuesday, May 11, = 2004 4:01 PM</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>To: = ietf-pkix@imc.org</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>Subject: RE: SCVP & = validation policies</span></font> <o:p></o:p></p> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = style=3D'font-size: 12.0pt'><o:p> </o:p></span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>I think policy OIDs should remain abstract, or at least can be = abstract.</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>That is that there is no requirement for the client software to</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>understand the = parameters.</span></font> <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>I am more thinking of using SCVP to validate a certificate in the</span></font> = <br> <font size=3D2><span style=3D'font-size:10.0pt'>context of a = transaction, where the policy OID in SCVP might be used to</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>identify such a context = to the server. </span></font><o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>For example, a certificate validation request is carried out with = regard</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>to the acceptability of = the holder's creditworthiness in the context of</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>a credit-based = transaction. In this case, the policy OID could be used</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>to indicate the class of assuarances that the CA's CSP provides for,</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>resulting in the server = selecting a subset of acceptable trust anchors.</span></font> <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>James</span></font> <o:p></o:p></p> <p class=3DMsoNormal style=3D'margin-bottom:12.0pt'><font size=3D3 face=3D"Times New Roman"><span = style=3D'font-size:12.0pt'><o:p> </o:p></span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>-----Original Message-----</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>From: Stephen Kent [<a href=3D"mailto:kent@bbn.com">mailto:kent@bbn.com</a>] </span></font><br> <font size=3D2><span style=3D'font-size:10.0pt'>Sent: Friday, 7 May 2004 = 6:28 AM</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>To: Trevor = Freeman</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>Cc: = ietf-pkix@imc.org</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>Subject: RE: SCVP & = validation policies</span></font> <o:p></o:p></p> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = style=3D'font-size: 12.0pt'><o:p> </o:p></span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>At 12:10 PM -0700 5/6/04, Trevor Freeman wrote:</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>>Hi = Steve,</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>>If you want to have = a shorthand notation for a validation policy, given</span></font> <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>>we have a finite set of parameters, the I don't see the value of = </span></font><br> <font size=3D2><span style=3D'font-size:10.0pt'>>indirection via an = OID. Using a hash of the set of inputs provides a </span></font><br> <font size=3D2><span style=3D'font-size:10.0pt'>>way of expressing = the same thing without introducing ambiguity of does </span></font><br> <font size=3D2><span style=3D'font-size:10.0pt'>>both parties have a = common understating of that this means. At some </span></font><br> <font size=3D2><span style=3D'font-size:10.0pt'>>point someone has to = define the policy in terms of the 3280 input</span></font> <br> <font size=3D2><span = style=3D'font-size:10.0pt'>parameters.</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>>That definition is = just as applicable and consumable by the client as a</span></font> = <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>>server therefore what did we gain from the shorthand notation? It = would</span></font> <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>>be pretty trivial to define some XML which would allow that set of = </span></font><br> <font size=3D2><span style=3D'font-size:10.0pt'>>validation = parameters to be imported by any SCVP entity.</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>>Trevor</span></font> = <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>Trevor,</span></font> <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>It is not indirection via an OID, it is identification via an OID. = </span></font><br> <font size=3D2><span style=3D'font-size:10.0pt'>Clients never have to be = aware of the parameters if we use an OID, and</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>simplification of = clients is one of the goals of SCVP. So I cannot agree</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>with your comment that = clients have to be able to consume these</span></font> <br> <font size=3D2><span = style=3D'font-size:10.0pt'>parameters.</span></font> <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>Given that as a basis, it seems that we are debating whether to use = OIDs</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>or to create some other = form of identifier via hashing parameters. I</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>don't see the latter as = attractive, since it requires additional</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>specification of the canonicalization and encoding of the parameters to</span></font> <br> <font size=3D2><span style=3D'font-size:10.0pt'>ensure consistency in = hash computation between client and server.</span></font> <o:p></o:p></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>Steve</span></font> <o:p></o:p></p> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = style=3D'font-size: 12.0pt'><o:p> </o:p></span></font></p> <p><font size=3D2 face=3D"Times New Roman"><span = style=3D'font-size:10.0pt'>This e-mail, and any attachments hereto, is intended only for use by the = named addressee(s) and may contain legally privileged and/or confidential information. If you are not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this e-mail, = and any attachments hereto, is strictly prohibited. If you have = received this transmission in error, please notify me immediately and permanently = delete the original and all copies and printouts of this = e-mail.</span></font><o:p></o:p></p> </div> </body> </html> ------_=_NextPart_001_01C437B5.4BDF985F-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BNmZBb073269; Tue, 11 May 2004 16:48:35 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BNmZGc073268; Tue, 11 May 2004 16:48:35 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BNmYau073261 for <ietf-pkix@imc.org>; Tue, 11 May 2004 16:48:34 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 11 May 2004 16:48:37 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 16:48:37 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 16:48:19 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Tue, 11 May 2004 16:48:36 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F02962882@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQzsteWPNjAnYrQQkSngFTDmZEpQQD94Z8QAAHsxEA= From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "James Jing" <jjing@betrusted.com>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 11 May 2004 23:48:19.0729 (UTC) FILETIME=[70142010:01C437B2] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4BNmZau073263 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> The client has to understand what is required to build a request i.e. What data does it need to supply to the server. It does not need to be aware of the factors which effect the server. Trevor -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of James Jing Sent: Tuesday, May 11, 2004 4:01 PM To: ietf-pkix@imc.org Subject: RE: SCVP & validation policies I think policy OIDs should remain abstract, or at least can be abstract. That is that there is no requirement for the client software to understand the parameters. I am more thinking of using SCVP to validate a certificate in the context of a transaction, where the policy OID in SCVP might be used to identify such a context to the server. For example, a certificate validation request is carried out with regard to the acceptability of the holder's creditworthiness in the context of a credit-based transaction. In this case, the policy OID could be used to indicate the class of assuarances that the CA's CSP provides for, resulting in the server selecting a subset of acceptable trust anchors. James -----Original Message----- From: Stephen Kent [mailto:kent@bbn.com] Sent: Friday, 7 May 2004 6:28 AM To: Trevor Freeman Cc: ietf-pkix@imc.org Subject: RE: SCVP & validation policies At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: >Hi Steve, >If you want to have a shorthand notation for a validation policy, given >we have a finite set of parameters, the I don't see the value of >indirection via an OID. Using a hash of the set of inputs provides a >way of expressing the same thing without introducing ambiguity of does >both parties have a common understating of that this means. At some >point someone has to define the policy in terms of the 3280 input parameters. >That definition is just as applicable and consumable by the client as a >server therefore what did we gain from the shorthand notation? It would >be pretty trivial to define some XML which would allow that set of >validation parameters to be imported by any SCVP entity. >Trevor Trevor, It is not indirection via an OID, it is identification via an OID. Clients never have to be aware of the parameters if we use an OID, and simplification of clients is one of the goals of SCVP. So I cannot agree with your comment that clients have to be able to consume these parameters. Given that as a basis, it seems that we are debating whether to use OIDs or to create some other form of identifier via hashing parameters. I don't see the latter as attractive, since it requires additional specification of the canonicalization and encoding of the parameters to ensure consistency in hash computation between client and server. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BN8DDl070660; Tue, 11 May 2004 16:08:13 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BN8Dig070659; Tue, 11 May 2004 16:08:13 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BN8D1P070653 for <ietf-pkix@imc.org>; Tue, 11 May 2004 16:08:13 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 11 May 2004 16:08:18 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 16:08:18 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 16:08:00 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Tue, 11 May 2004 16:08:17 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F0296284C@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ3qmiKcqXx3k4PSf2S0mo5HXNBWAAAWM1w From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Florian Oelmaier" <oelmaier@sytrust.com>, "Stephen Kent" <kent@bbn.com> Cc: "Denis Pinkas" <Denis.Pinkas@bull.net>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 11 May 2004 23:08:00.0526 (UTC) FILETIME=[CE1F0AE0:01C437AC] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4BN8D1P070654 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Florian, A server is under no obligation to process the request. If you have a badly configured client, then it will be fixed. If the request is accurate and the server is misconfigured, then it will be fixed. This is not about the sanctity of either sides policy but simple client server request/response semantics. I do not expect the server to second guess what the problem is and return potentially misleading data. If the request conflicts with the server locally configured policy, return the error, if not process the request Trevor -----Original Message----- From: Florian Oelmaier [mailto:oelmaier@sytrust.com] Sent: Tuesday, May 11, 2004 3:51 PM To: Trevor Freeman; 'Stephen Kent' Cc: 'Denis Pinkas'; ietf-pkix@imc.org Subject: RE: SCVP & validation policies > Hi Florian, > The server either carries out the request or returns an > error. We cannot have the server picking and choosing items > out of the request since it becomes impossible for the client > to determining what happened. That is totally consistent with > the needs of the enterprise. Trevor I see it the other way round: The client sends a request and the server chooses HOW to answer the request. All responsibilities of the validation process are determined by the policy of the SCVP server. We cannot have the configuration of a client set by a user in the enterprise interfering with centrally managed trust settings. If we allow this, the idea of central managed trust and validation settings is well and truly dead. In my eyes it is the clients task to specify its Validation-Policy OID and perhaps ist "wishlist" - and then to trust the servers response. >From my perspective this was (is?) one of the most charming features of SCVP. Best regards, Florian > > -----Original Message----- > From: Florian Oelmaier [mailto:oelmaier@sytrust.com] > Sent: Tuesday, May 11, 2004 12:13 PM > To: Trevor Freeman; 'Stephen Kent' > Cc: 'Denis Pinkas'; ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > Hi Steve, > > I am not excluding that. A SCVP client may choose to > > implement the more al a carte or a combination of al-a-carte > > plus predefined policy. In every validation policy it may > > agree m parameters. That leaves n parameters to be specified > > per request. Since we don't know which of the m parameters > > will be fixed in any given deployment we must support any > > combination so we must support the full rage of parameter > > currently defined. In many situations the overhead of > > agreeing these pre-caned selection is an administrative > > burden such as a departmental server wanting to implement > > special needs. Equally in may corporate deployments there is > > no choice with many of the values so you can default every to > > whoever is good for the server, so again rendering app > > specific policy of no value. Trevor > > Hello Trevor, > > I feel that a solution allowing a client to specify values > that the server MUST use will not only harm the usability of > SCVP in enterprise environments but also make > interoperability between different software vendors on the > client side with different servers a lot more difficult. > > If we specify a set of parameters that the server MAY use - I > am fine. But please dont allow the client to specify any > parameters a server MUST use. > > Best regards, > > Florian Oelmaier > > > > > > > > > > -----Original Message----- > > From: Stephen Kent [mailto:kent@bbn.com] > > Sent: Tuesday, May 11, 2004 9:28 AM > > To: Trevor Freeman > > Cc: Denis Pinkas; ietf-pkix@imc.org > > Subject: RE: SCVP & validation policies > > > > At 9:16 AM -0700 5/11/04, Trevor Freeman wrote: > > >Hi Dennis, > > >The OID is optional for the client to use. Servers don't have to > > support > > >this. It become a matter of mutual agreement between the client and > > >server what the OID represents. An SCVP server must either > > comply with > > >any request or return an error so I don't see the validation > > policy OID > > >needs any special treatment beyond want SCVP already > defines. Trevor > > > > Trevor, > > > > I have to agree with Denis here. This is completely backwards from > > what I recall we defined as the requirements in his RFC a > while ago. > > OIDs were sufficient to identify a policy, the parameters for which > > were managed via a separate protocol. It seems a lot easier to > > configure a client with an OID per app, than to configure a set of > > parameters for each app. > > > > Steve > > > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BN1WP3070386; Tue, 11 May 2004 16:01:32 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BN1WnF070385; Tue, 11 May 2004 16:01:32 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from figg.securenet.com.au (ns2.isecure.com.au [202.125.4.72]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BN1UZY070372 for <ietf-pkix@imc.org>; Tue, 11 May 2004 16:01:31 -0700 (PDT) (envelope-from jjing@betrusted.com) Received: from iron.securenet.com.au (iron.isecure.com.au [202.125.4.94] (may be forged)) by figg.securenet.com.au (8.12.3/8.12.3/Debian-6.6) with ESMTP id i4BN1Mi3004396 for <ietf-pkix@imc.org>; Wed, 12 May 2004 09:01:22 +1000 Received: (from uucp@localhost) by iron.securenet.com.au (8.12.6/8.12.6) id i4BN1Mkx028323 for <ietf-pkix@imc.org>; Wed, 12 May 2004 09:01:22 +1000 (EST) Received: from nodnsquery(10.11.3.10) by iron.securenet.com.au via csmap (V6.0) id srcAAApMaiu3; Wed, 12 May 04 09:01:21 +1000 Received: from cbrms01.securenet.com.au (localhost [127.0.0.1]) by gibbons.securenet.com.au (8.12.3/8.12.3/Debian-7woody) with ESMTP id i4BN1Kev006569 for <ietf-pkix@imc.org>; Wed, 12 May 2004 09:01:21 +1000 Received: from AMALTHEA.securenet.com.au ([192.168.100.30]) by cbrms01.securenet.com.au with Microsoft SMTPSVC(5.0.2195.6713); Wed, 12 May 2004 08:56:50 +1000 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Subject: RE: SCVP & validation policies X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3 Date: Wed, 12 May 2004 09:01:03 +1000 Message-ID: <85DA9C6C3DD8C74F8A8611815C635F8D4F656A@AMALTHEA.securenet.com.au> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies Thread-Index: AcQzsteWPNjAnYrQQkSngFTDmZEpQQD94Z8Q From: "James Jing" <jjing@betrusted.com> To: <ietf-pkix@imc.org> X-OriginalArrivalTime: 11 May 2004 22:56:50.0796 (UTC) FILETIME=[3EEE5AC0:01C437AB] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4BN1VZY070380 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I think policy OIDs should remain abstract, or at least can be abstract. That is that there is no requirement for the client software to understand the parameters. I am more thinking of using SCVP to validate a certificate in the context of a transaction, where the policy OID in SCVP might be used to identify such a context to the server. For example, a certificate validation request is carried out with regard to the acceptability of the holder's creditworthiness in the context of a credit-based transaction. In this case, the policy OID could be used to indicate the class of assuarances that the CA's CSP provides for, resulting in the server selecting a subset of acceptable trust anchors. James -----Original Message----- From: Stephen Kent [mailto:kent@bbn.com] Sent: Friday, 7 May 2004 6:28 AM To: Trevor Freeman Cc: ietf-pkix@imc.org Subject: RE: SCVP & validation policies At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: >Hi Steve, >If you want to have a shorthand notation for a validation policy, given >we have a finite set of parameters, the I don't see the value of >indirection via an OID. Using a hash of the set of inputs provides a >way of expressing the same thing without introducing ambiguity of does >both parties have a common understating of that this means. At some >point someone has to define the policy in terms of the 3280 input parameters. >That definition is just as applicable and consumable by the client as a >server therefore what did we gain from the shorthand notation? It would >be pretty trivial to define some XML which would allow that set of >validation parameters to be imported by any SCVP entity. >Trevor Trevor, It is not indirection via an OID, it is identification via an OID. Clients never have to be aware of the parameters if we use an OID, and simplification of clients is one of the goals of SCVP. So I cannot agree with your comment that clients have to be able to consume these parameters. Given that as a basis, it seems that we are debating whether to use OIDs or to create some other form of identifier via hashing parameters. I don't see the latter as attractive, since it requires additional specification of the canonicalization and encoding of the parameters to ensure consistency in hash computation between client and server. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BMopQb069857; Tue, 11 May 2004 15:50:51 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BMopLc069856; Tue, 11 May 2004 15:50:51 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-in.m-online.net (mail-in.m-online.net [62.245.150.237]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BMonr1069850 for <ietf-pkix@imc.org>; Tue, 11 May 2004 15:50:50 -0700 (PDT) (envelope-from oelmaier@sytrust.com) Received: from mail.m-online.net (svr14.m-online.net [192.168.3.144]) by svr8.m-online.net (Postfix) with ESMTP id F3EAC4C9B2; Wed, 12 May 2004 00:50:51 +0200 (CEST) Received: from hagen (ppp-82-135-9-66.mnet-online.de [82.135.9.66]) by mail.m-online.net (Postfix) with ESMTP id B454FA7F48; Wed, 12 May 2004 00:50:51 +0200 (CEST) From: "Florian Oelmaier" <oelmaier@sytrust.com> To: "'Trevor Freeman'" <trevorf@exchange.microsoft.com>, "'Stephen Kent'" <kent@bbn.com> Cc: "'Denis Pinkas'" <Denis.Pinkas@bull.net>, <ietf-pkix@imc.org> Subject: RE: SCVP & validation policies Date: Wed, 12 May 2004 00:50:49 +0200 Organization: SyTrust Message-ID: <012601c437aa$67cd43d0$4228a8c0@hagen> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.3416 Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 In-reply-to: <33E7A1613A3480448996047B69308A2F0296275B@df-grommit-01.dogfood> Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > Hi Florian, > The server either carries out the request or returns an > error. We cannot have the server picking and choosing items > out of the request since it becomes impossible for the client > to determining what happened. That is totally consistent with > the needs of the enterprise. Trevor I see it the other way round: The client sends a request and the server chooses HOW to answer the request. All responsibilities of the validation process are determined by the policy of the SCVP server. We cannot have the configuration of a client set by a user in the enterprise interfering with centrally managed trust settings. If we allow this, the idea of central managed trust and validation settings is well and truly dead. In my eyes it is the clients task to specify its Validation-Policy OID and perhaps ist "wishlist" - and then to trust the servers response. >From my perspective this was (is?) one of the most charming features of SCVP. Best regards, Florian > > -----Original Message----- > From: Florian Oelmaier [mailto:oelmaier@sytrust.com] > Sent: Tuesday, May 11, 2004 12:13 PM > To: Trevor Freeman; 'Stephen Kent' > Cc: 'Denis Pinkas'; ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > > Hi Steve, > > I am not excluding that. A SCVP client may choose to > > implement the more al a carte or a combination of al-a-carte > > plus predefined policy. In every validation policy it may > > agree m parameters. That leaves n parameters to be specified > > per request. Since we don't know which of the m parameters > > will be fixed in any given deployment we must support any > > combination so we must support the full rage of parameter > > currently defined. In many situations the overhead of > > agreeing these pre-caned selection is an administrative > > burden such as a departmental server wanting to implement > > special needs. Equally in may corporate deployments there is > > no choice with many of the values so you can default every to > > whoever is good for the server, so again rendering app > > specific policy of no value. Trevor > > Hello Trevor, > > I feel that a solution allowing a client to specify values > that the server MUST use will not only harm the usability of > SCVP in enterprise environments but also make > interoperability between different software vendors on the > client side with different servers a lot more difficult. > > If we specify a set of parameters that the server MAY use - I > am fine. But please dont allow the client to specify any > parameters a server MUST use. > > Best regards, > > Florian Oelmaier > > > > > > > > > > -----Original Message----- > > From: Stephen Kent [mailto:kent@bbn.com] > > Sent: Tuesday, May 11, 2004 9:28 AM > > To: Trevor Freeman > > Cc: Denis Pinkas; ietf-pkix@imc.org > > Subject: RE: SCVP & validation policies > > > > At 9:16 AM -0700 5/11/04, Trevor Freeman wrote: > > >Hi Dennis, > > >The OID is optional for the client to use. Servers don't have to > > support > > >this. It become a matter of mutual agreement between the client and > > >server what the OID represents. An SCVP server must either > > comply with > > >any request or return an error so I don't see the validation > > policy OID > > >needs any special treatment beyond want SCVP already > defines. Trevor > > > > Trevor, > > > > I have to agree with Denis here. This is completely backwards from > > what I recall we defined as the requirements in his RFC a > while ago. > > OIDs were sufficient to identify a policy, the parameters for which > > were managed via a separate protocol. It seems a lot easier to > > configure a client with an OID per app, than to configure a set of > > parameters for each app. > > > > Steve > > > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BKO2HJ058878; Tue, 11 May 2004 13:24:02 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BKO2n0058877; Tue, 11 May 2004 13:24:02 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BKO1q0058871 for <ietf-pkix@imc.org>; Tue, 11 May 2004 13:24:01 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 11 May 2004 13:24:06 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 13:24:06 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 13:24:06 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Tue, 11 May 2004 13:24:04 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F0296275B@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ3i/FXdTOSaoY6The390PFDj/vBgACU28w From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Florian Oelmaier" <oelmaier@sytrust.com>, "Stephen Kent" <kent@bbn.com> Cc: "Denis Pinkas" <Denis.Pinkas@bull.net>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 11 May 2004 20:24:06.0002 (UTC) FILETIME=[E849C520:01C43795] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4BKO2q0058872 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Florian, The server either carries out the request or returns an error. We cannot have the server picking and choosing items out of the request since it becomes impossible for the client to determining what happened. That is totally consistent with the needs of the enterprise. Trevor -----Original Message----- From: Florian Oelmaier [mailto:oelmaier@sytrust.com] Sent: Tuesday, May 11, 2004 12:13 PM To: Trevor Freeman; 'Stephen Kent' Cc: 'Denis Pinkas'; ietf-pkix@imc.org Subject: RE: SCVP & validation policies > Hi Steve, > I am not excluding that. A SCVP client may choose to > implement the more al a carte or a combination of al-a-carte > plus predefined policy. In every validation policy it may > agree m parameters. That leaves n parameters to be specified > per request. Since we don't know which of the m parameters > will be fixed in any given deployment we must support any > combination so we must support the full rage of parameter > currently defined. In many situations the overhead of > agreeing these pre-caned selection is an administrative > burden such as a departmental server wanting to implement > special needs. Equally in may corporate deployments there is > no choice with many of the values so you can default every to > whoever is good for the server, so again rendering app > specific policy of no value. Trevor Hello Trevor, I feel that a solution allowing a client to specify values that the server MUST use will not only harm the usability of SCVP in enterprise environments but also make interoperability between different software vendors on the client side with different servers a lot more difficult. If we specify a set of parameters that the server MAY use - I am fine. But please dont allow the client to specify any parameters a server MUST use. Best regards, Florian Oelmaier > > > > -----Original Message----- > From: Stephen Kent [mailto:kent@bbn.com] > Sent: Tuesday, May 11, 2004 9:28 AM > To: Trevor Freeman > Cc: Denis Pinkas; ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > At 9:16 AM -0700 5/11/04, Trevor Freeman wrote: > >Hi Dennis, > >The OID is optional for the client to use. Servers don't have to > support > >this. It become a matter of mutual agreement between the client and > >server what the OID represents. An SCVP server must either > comply with > >any request or return an error so I don't see the validation > policy OID > >needs any special treatment beyond want SCVP already defines. Trevor > > Trevor, > > I have to agree with Denis here. This is completely backwards from > what I recall we defined as the requirements in his RFC a while ago. > OIDs were sufficient to identify a policy, the parameters for which > were managed via a separate protocol. It seems a lot easier to > configure a client with an OID per app, than to configure a set of > parameters for each app. > > Steve > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BJCjYl051813; Tue, 11 May 2004 12:12:45 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BJCjMK051812; Tue, 11 May 2004 12:12:45 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-in.m-online.net (mail-in.m-online.net [62.245.150.237]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BJChPB051802 for <ietf-pkix@imc.org>; Tue, 11 May 2004 12:12:44 -0700 (PDT) (envelope-from oelmaier@sytrust.com) Received: from mail.m-online.net (svr14.m-online.net [192.168.3.144]) by svr8.m-online.net (Postfix) with ESMTP id CA3154CE75; Tue, 11 May 2004 21:12:44 +0200 (CEST) Received: from hagen (ppp-82-135-9-66.mnet-online.de [82.135.9.66]) by mail.m-online.net (Postfix) with ESMTP id 85685A7FA8; Tue, 11 May 2004 21:12:44 +0200 (CEST) From: "Florian Oelmaier" <oelmaier@sytrust.com> To: "'Trevor Freeman'" <trevorf@exchange.microsoft.com>, "'Stephen Kent'" <kent@bbn.com> Cc: "'Denis Pinkas'" <Denis.Pinkas@bull.net>, <ietf-pkix@imc.org> Subject: RE: SCVP & validation policies Date: Tue, 11 May 2004 21:12:41 +0200 Organization: SyTrust Message-ID: <00e701c4378b$eea83780$4228a8c0@hagen> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.3416 Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 In-reply-to: <33E7A1613A3480448996047B69308A2F027EEC9D@df-grommit-01.dogfood> Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > Hi Steve, > I am not excluding that. A SCVP client may choose to > implement the more al a carte or a combination of al-a-carte > plus predefined policy. In every validation policy it may > agree m parameters. That leaves n parameters to be specified > per request. Since we don't know which of the m parameters > will be fixed in any given deployment we must support any > combination so we must support the full rage of parameter > currently defined. In many situations the overhead of > agreeing these pre-caned selection is an administrative > burden such as a departmental server wanting to implement > special needs. Equally in may corporate deployments there is > no choice with many of the values so you can default every to > whoever is good for the server, so again rendering app > specific policy of no value. Trevor Hello Trevor, I feel that a solution allowing a client to specify values that the server MUST use will not only harm the usability of SCVP in enterprise environments but also make interoperability between different software vendors on the client side with different servers a lot more difficult. If we specify a set of parameters that the server MAY use - I am fine. But please dont allow the client to specify any parameters a server MUST use. Best regards, Florian Oelmaier > > > > -----Original Message----- > From: Stephen Kent [mailto:kent@bbn.com] > Sent: Tuesday, May 11, 2004 9:28 AM > To: Trevor Freeman > Cc: Denis Pinkas; ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > At 9:16 AM -0700 5/11/04, Trevor Freeman wrote: > >Hi Dennis, > >The OID is optional for the client to use. Servers don't have to > support > >this. It become a matter of mutual agreement between the client and > >server what the OID represents. An SCVP server must either > comply with > >any request or return an error so I don't see the validation > policy OID > >needs any special treatment beyond want SCVP already defines. Trevor > > Trevor, > > I have to agree with Denis here. This is completely backwards from > what I recall we defined as the requirements in his RFC a while ago. > OIDs were sufficient to identify a policy, the parameters for which > were managed via a separate protocol. It seems a lot easier to > configure a client with an OID per app, than to configure a set of > parameters for each app. > > Steve > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BHG8qI038267; Tue, 11 May 2004 10:16:08 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BHG8l0038266; Tue, 11 May 2004 10:16:08 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BHG67F038255 for <ietf-pkix@imc.org>; Tue, 11 May 2004 10:16:06 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 11 May 2004 10:16:09 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 10:16:07 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 10:16:09 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Tue, 11 May 2004 10:16:08 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EECAF@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ3df6yT7wzvr3RR0GyuVCBAgYxbgABSRSQ From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: "Stephen Kent" <kent@bbn.com>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 11 May 2004 17:16:09.0874 (UTC) FILETIME=[A7312F20:01C4377B] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4BHG67F038256 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Denis, I am not excluding he use of OIDs so your customers can do as they want and pass whatever additional parameters they like per request. My customer who find the additional layer of administration unnecessary, can decide not to use them. Trevor -----Original Message----- From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] Sent: Tuesday, May 11, 2004 9:36 AM To: Trevor Freeman Cc: Stephen Kent; ietf-pkix@imc.org Subject: Re: SCVP & validation policies Trevor, > Hi Dennis, > The OID is optional for the client to use. Servers don't have to support > this. I disagree. The "RFC 3280 compliantbasic validation policy", i.e the one which only includes basic path validation, with no extra and no shortcuts, as defined by RFC 3280, will have an OID. Since this policy does not contains values such as root keys, this validation policy needs additional parameters. Intelligent clients may use this one and specify parameters. Thin clients will simply refer to an OID which refers globally to a validation algorithm and its parameters. I have talked to customers and their position is to have flexible validation policies. As an example, some says: if we want to take purposely the risk of not testing ARLs, it is our choice. The validation policy should not mandate to test ARLs, but if for some validation policies it is adequate to do so, it should be possible to do so. > It become a matter of mutual agreement between the client and > server what the OID represents. Not necessarilly. Since the OID is not in the arc of the server, it may be defined by anyone. If that definition is separately accessible both to the client and to the server then it can be supported by the server and the client may refer to it. In such a case there is no need for a mutual agreement. > An SCVP server must either comply with any request or return an error True. > so I don't see the validation policy OID > needs any special treatment beyond want SCVP already defines. I don't see the relationship between the true statement above and the conclusion. As already stated on this list, SCVP stands for "Simple", otherwise we will have to drop the "S", and then we would have "CVP". :-| Denis > Trevor > > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Tuesday, May 11, 2004 6:50 AM > To: Trevor Freeman > Cc: Stephen Kent; ietf-pkix@imc.org > Subject: Re: SCVP & validation policies > > Trevor, > > >>Hi Steve, >>It is management of the clients which is at question here. The client >>has a series of input parameters all of which are opaque. The OID is a >>management indirection to a set of parameters. The ambiguity arises >>because the use of the OID assumes both the manager of the client and >>the manager of the server have an accurate understanding of the OID. >> >>However give the volume of traffic on this point, I will add the >>capacity to pass an optional OID in the request to the server to >>identify a set of parameters in SCVP15. > > > It is the other way around: > > Given the volume of traffic on this point, the parameters should allow > to > pass a *mandatory *OID in the request to identify a given validation > policy, > while some specific validation policies may need an additional set of > parameters. > > Denis > > >>Trevor >> >>-----Original Message----- >>From: Stephen Kent [mailto:kent@bbn.com] >>Sent: Thursday, May 06, 2004 1:28 PM >>To: Trevor Freeman >>Cc: ietf-pkix@imc.org >>Subject: RE: SCVP & validation policies >> >>At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: >> >> >>>Hi Steve, >>>If you want to have a shorthand notation for a validation policy, >> > given > >>>we have a finite set of parameters, the I don't see the value of >>>indirection via an OID. Using a hash of the set of inputs provides a >> >>way >> >> >>>of expressing the same thing without introducing ambiguity of does >> > both > >>>parties have a common understating of that this means. At some point >>>someone has to define the policy in terms of the 3280 input >> > parameters. > >>>That definition is just as applicable and consumable by the client as >> > a > >>>server therefore what did we gain from the shorthand notation? It >> > would > >>>be pretty trivial to define some XML which would allow that set of >>>validation parameters to be imported by any SCVP entity. >>>Trevor >> >> >>Trevor, >> >>It is not indirection via an OID, it is identification via an OID. >>Clients never have to be aware of the parameters if we use an OID, >>and simplification of clients is one of the goals of SCVP. So I >>cannot agree with your comment that clients have to be able to >>consume these parameters. >> >>Given that as a basis, it seems that we are debating whether to use >>OIDs or to create some other form of identifier via hashing >>parameters. I don't see the latter as attractive, since it requires >>additional specification of the canonicalization and encoding of the >>parameters to ensure consistency in hash computation between client >>and server. >> >>Steve >> > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BH5TbU037032; Tue, 11 May 2004 10:05:29 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BH5Tv9037031; Tue, 11 May 2004 10:05:29 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BH5Ss1037024 for <ietf-pkix@imc.org>; Tue, 11 May 2004 10:05:28 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 11 May 2004 10:05:31 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 10:05:29 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 10:05:31 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Tue, 11 May 2004 10:05:29 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EEC9D@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ3dRq23zt6dFZjT1O/f2a7QmDGrgAATVJw From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Stephen Kent" <kent@bbn.com> Cc: "Denis Pinkas" <Denis.Pinkas@bull.net>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 11 May 2004 17:05:31.0527 (UTC) FILETIME=[2AB52970:01C4377A] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4BH5Ss1037025 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Steve, I am not excluding that. A SCVP client may choose to implement the more al a carte or a combination of al-a-carte plus predefined policy. In every validation policy it may agree m parameters. That leaves n parameters to be specified per request. Since we don't know which of the m parameters will be fixed in any given deployment we must support any combination so we must support the full rage of parameter currently defined. In many situations the overhead of agreeing these pre-caned selection is an administrative burden such as a departmental server wanting to implement special needs. Equally in may corporate deployments there is no choice with many of the values so you can default every to whoever is good for the server, so again rendering app specific policy of no value. Trevor -----Original Message----- From: Stephen Kent [mailto:kent@bbn.com] Sent: Tuesday, May 11, 2004 9:28 AM To: Trevor Freeman Cc: Denis Pinkas; ietf-pkix@imc.org Subject: RE: SCVP & validation policies At 9:16 AM -0700 5/11/04, Trevor Freeman wrote: >Hi Dennis, >The OID is optional for the client to use. Servers don't have to support >this. It become a matter of mutual agreement between the client and >server what the OID represents. An SCVP server must either comply with >any request or return an error so I don't see the validation policy OID >needs any special treatment beyond want SCVP already defines. >Trevor Trevor, I have to agree with Denis here. This is completely backwards from what I recall we defined as the requirements in his RFC a while ago. OIDs were sufficient to identify a policy, the parameters for which were managed via a separate protocol. It seems a lot easier to configure a client with an OID per app, than to configure a set of parameters for each app. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BGZRwA033816; Tue, 11 May 2004 09:35:27 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BGZRJs033815; Tue, 11 May 2004 09:35:27 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from odin2.bull.net (odin2.bull.net [192.90.70.84]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BGZPKT033799 for <ietf-pkix@imc.org>; Tue, 11 May 2004 09:35:26 -0700 (PDT) (envelope-from Denis.Pinkas@bull.net) Received: from clbull.frcl.bull.fr (IDENT:root@clbull.frcl.bull.fr [129.182.8.31]) by odin2.bull.net (8.9.3/8.9.3) with ESMTP id SAA33478; Tue, 11 May 2004 18:44:45 +0200 Received: from bull.net (frcls4013.frcl.bull.fr [129.182.108.120]) by clbull.frcl.bull.fr (8.9.3/8.9.3) with ESMTP id SAA08047; Tue, 11 May 2004 18:34:04 +0200 Message-ID: <40A100E5.3050208@bull.net> Date: Tue, 11 May 2004 18:35:49 +0200 From: Denis Pinkas <Denis.Pinkas@bull.net> Organization: Bull SA. User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en, fr MIME-Version: 1.0 To: Trevor Freeman <trevorf@exchange.microsoft.com> CC: Stephen Kent <kent@bbn.com>, ietf-pkix@imc.org Subject: Re: SCVP & validation policies References: <33E7A1613A3480448996047B69308A2F027EEC4C@df-grommit-01.dogfood> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Trevor, > Hi Dennis, > The OID is optional for the client to use. Servers don't have to support > this. I disagree. The "RFC 3280 compliantbasic validation policy", i.e the one which only includes basic path validation, with no extra and no shortcuts, as defined by RFC 3280, will have an OID. Since this policy does not contains values such as root keys, this validation policy needs additional parameters. Intelligent clients may use this one and specify parameters. Thin clients will simply refer to an OID which refers globally to a validation algorithm and its parameters. I have talked to customers and their position is to have flexible validation policies. As an example, some says: if we want to take purposely the risk of not testing ARLs, it is our choice. The validation policy should not mandate to test ARLs, but if for some validation policies it is adequate to do so, it should be possible to do so. > It become a matter of mutual agreement between the client and > server what the OID represents. Not necessarilly. Since the OID is not in the arc of the server, it may be defined by anyone. If that definition is separately accessible both to the client and to the server then it can be supported by the server and the client may refer to it. In such a case there is no need for a mutual agreement. > An SCVP server must either comply with any request or return an error True. > so I don't see the validation policy OID > needs any special treatment beyond want SCVP already defines. I don't see the relationship between the true statement above and the conclusion. As already stated on this list, SCVP stands for "Simple", otherwise we will have to drop the "S", and then we would have "CVP". :-| Denis > Trevor > > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Tuesday, May 11, 2004 6:50 AM > To: Trevor Freeman > Cc: Stephen Kent; ietf-pkix@imc.org > Subject: Re: SCVP & validation policies > > Trevor, > > >>Hi Steve, >>It is management of the clients which is at question here. The client >>has a series of input parameters all of which are opaque. The OID is a >>management indirection to a set of parameters. The ambiguity arises >>because the use of the OID assumes both the manager of the client and >>the manager of the server have an accurate understanding of the OID. >> >>However give the volume of traffic on this point, I will add the >>capacity to pass an optional OID in the request to the server to >>identify a set of parameters in SCVP15. > > > It is the other way around: > > Given the volume of traffic on this point, the parameters should allow > to > pass a *mandatory *OID in the request to identify a given validation > policy, > while some specific validation policies may need an additional set of > parameters. > > Denis > > >>Trevor >> >>-----Original Message----- >>From: Stephen Kent [mailto:kent@bbn.com] >>Sent: Thursday, May 06, 2004 1:28 PM >>To: Trevor Freeman >>Cc: ietf-pkix@imc.org >>Subject: RE: SCVP & validation policies >> >>At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: >> >> >>>Hi Steve, >>>If you want to have a shorthand notation for a validation policy, >> > given > >>>we have a finite set of parameters, the I don't see the value of >>>indirection via an OID. Using a hash of the set of inputs provides a >> >>way >> >> >>>of expressing the same thing without introducing ambiguity of does >> > both > >>>parties have a common understating of that this means. At some point >>>someone has to define the policy in terms of the 3280 input >> > parameters. > >>>That definition is just as applicable and consumable by the client as >> > a > >>>server therefore what did we gain from the shorthand notation? It >> > would > >>>be pretty trivial to define some XML which would allow that set of >>>validation parameters to be imported by any SCVP entity. >>>Trevor >> >> >>Trevor, >> >>It is not indirection via an OID, it is identification via an OID. >>Clients never have to be aware of the parameters if we use an OID, >>and simplification of clients is one of the goals of SCVP. So I >>cannot agree with your comment that clients have to be able to >>consume these parameters. >> >>Given that as a basis, it seems that we are debating whether to use >>OIDs or to create some other form of identifier via hashing >>parameters. I don't see the latter as attractive, since it requires >>additional specification of the canonicalization and encoding of the >>parameters to ensure consistency in hash computation between client >>and server. >> >>Steve >> > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BGTJMo032790; Tue, 11 May 2004 09:29:19 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BGTJPG032789; Tue, 11 May 2004 09:29:19 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BGTIG0032768 for <ietf-pkix@imc.org>; Tue, 11 May 2004 09:29:18 -0700 (PDT) (envelope-from kent@bbn.com) Received: from [10.81.115.79] (ramblo.bbn.com [128.33.0.51]) by aragorn.bbn.com (8.12.7/8.12.7) with ESMTP id i4BGSS7X015450; Tue, 11 May 2004 12:28:29 -0400 (EDT) Mime-Version: 1.0 X-Sender: kent@localhost Message-Id: <p06100505bcc6af22facd@[10.81.115.79]> In-Reply-To: <33E7A1613A3480448996047B69308A2F027EEC4C@df-grommit-01.dogfood> References: <33E7A1613A3480448996047B69308A2F027EEC4C@df-grommit-01.dogfood> Date: Tue, 11 May 2004 12:28:12 -0400 To: "Trevor Freeman" <trevorf@exchange.microsoft.com> From: Stephen Kent <kent@bbn.com> Subject: RE: SCVP & validation policies Cc: "Denis Pinkas" <Denis.Pinkas@bull.net>, <ietf-pkix@imc.org> Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> At 9:16 AM -0700 5/11/04, Trevor Freeman wrote: >Hi Dennis, >The OID is optional for the client to use. Servers don't have to support >this. It become a matter of mutual agreement between the client and >server what the OID represents. An SCVP server must either comply with >any request or return an error so I don't see the validation policy OID >needs any special treatment beyond want SCVP already defines. >Trevor Trevor, I have to agree with Denis here. This is completely backwards from what I recall we defined as the requirements in his RFC a while ago. OIDs were sufficient to identify a policy, the parameters for which were managed via a separate protocol. It seems a lot easier to configure a client with an OID per app, than to configure a set of parameters for each app. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BGGuB0031373; Tue, 11 May 2004 09:16:56 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BGGuxm031372; Tue, 11 May 2004 09:16:56 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BGGt6D031357 for <ietf-pkix@imc.org>; Tue, 11 May 2004 09:16:55 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 11 May 2004 09:16:55 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 09:16:52 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 09:16:55 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Tue, 11 May 2004 09:16:54 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EEC4C@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ3XuoPr6ll6ZjYTR2cbNeMZNPIVAAE5iKw From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: "Stephen Kent" <kent@bbn.com>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 11 May 2004 16:16:55.0239 (UTC) FILETIME=[6076ED70:01C43773] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4BGGt6D031367 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Dennis, The OID is optional for the client to use. Servers don't have to support this. It become a matter of mutual agreement between the client and server what the OID represents. An SCVP server must either comply with any request or return an error so I don't see the validation policy OID needs any special treatment beyond want SCVP already defines. Trevor -----Original Message----- From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] Sent: Tuesday, May 11, 2004 6:50 AM To: Trevor Freeman Cc: Stephen Kent; ietf-pkix@imc.org Subject: Re: SCVP & validation policies Trevor, > Hi Steve, > It is management of the clients which is at question here. The client > has a series of input parameters all of which are opaque. The OID is a > management indirection to a set of parameters. The ambiguity arises > because the use of the OID assumes both the manager of the client and > the manager of the server have an accurate understanding of the OID. > > However give the volume of traffic on this point, I will add the > capacity to pass an optional OID in the request to the server to > identify a set of parameters in SCVP15. It is the other way around: Given the volume of traffic on this point, the parameters should allow to pass a *mandatory *OID in the request to identify a given validation policy, while some specific validation policies may need an additional set of parameters. Denis > Trevor > > -----Original Message----- > From: Stephen Kent [mailto:kent@bbn.com] > Sent: Thursday, May 06, 2004 1:28 PM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: > >>Hi Steve, >>If you want to have a shorthand notation for a validation policy, given >>we have a finite set of parameters, the I don't see the value of >>indirection via an OID. Using a hash of the set of inputs provides a > > way > >>of expressing the same thing without introducing ambiguity of does both >>parties have a common understating of that this means. At some point >>someone has to define the policy in terms of the 3280 input parameters. >>That definition is just as applicable and consumable by the client as a >>server therefore what did we gain from the shorthand notation? It would >>be pretty trivial to define some XML which would allow that set of >>validation parameters to be imported by any SCVP entity. >>Trevor > > > Trevor, > > It is not indirection via an OID, it is identification via an OID. > Clients never have to be aware of the parameters if we use an OID, > and simplification of clients is one of the goals of SCVP. So I > cannot agree with your comment that clients have to be able to > consume these parameters. > > Given that as a basis, it seems that we are debating whether to use > OIDs or to create some other form of identifier via hashing > parameters. I don't see the latter as attractive, since it requires > additional specification of the canonicalization and encoding of the > parameters to ensure consistency in hash computation between client > and server. > > Steve > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BDo8uU014300; Tue, 11 May 2004 06:50:08 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BDo8uE014299; Tue, 11 May 2004 06:50:08 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from odin2.bull.net (odin2.bull.net [192.90.70.84]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BDo5qn014272 for <ietf-pkix@imc.org>; Tue, 11 May 2004 06:50:06 -0700 (PDT) (envelope-from Denis.Pinkas@bull.net) Received: from clbull.frcl.bull.fr (IDENT:root@clbull.frcl.bull.fr [129.182.8.31]) by odin2.bull.net (8.9.3/8.9.3) with ESMTP id PAA28446; Tue, 11 May 2004 15:59:19 +0200 Received: from bull.net (frcls4013.frcl.bull.fr [129.182.108.120]) by clbull.frcl.bull.fr (8.9.3/8.9.3) with ESMTP id PAA06932; Tue, 11 May 2004 15:48:35 +0200 Message-ID: <40A0DA1C.8020003@bull.net> Date: Tue, 11 May 2004 15:50:20 +0200 From: Denis Pinkas <Denis.Pinkas@bull.net> Organization: Bull SA. User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en, fr MIME-Version: 1.0 To: Trevor Freeman <trevorf@exchange.microsoft.com> CC: Stephen Kent <kent@bbn.com>, ietf-pkix@imc.org Subject: Re: SCVP & validation policies References: <33E7A1613A3480448996047B69308A2F027EE7CA@df-grommit-01.dogfood> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Trevor, > Hi Steve, > It is management of the clients which is at question here. The client > has a series of input parameters all of which are opaque. The OID is a > management indirection to a set of parameters. The ambiguity arises > because the use of the OID assumes both the manager of the client and > the manager of the server have an accurate understanding of the OID. > > However give the volume of traffic on this point, I will add the > capacity to pass an optional OID in the request to the server to > identify a set of parameters in SCVP15. It is the other way around: Given the volume of traffic on this point, the parameters should allow to pass a *mandatory *OID in the request to identify a given validation policy, while some specific validation policies may need an additional set of parameters. Denis > Trevor > > -----Original Message----- > From: Stephen Kent [mailto:kent@bbn.com] > Sent: Thursday, May 06, 2004 1:28 PM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: RE: SCVP & validation policies > > At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: > >>Hi Steve, >>If you want to have a shorthand notation for a validation policy, given >>we have a finite set of parameters, the I don't see the value of >>indirection via an OID. Using a hash of the set of inputs provides a > > way > >>of expressing the same thing without introducing ambiguity of does both >>parties have a common understating of that this means. At some point >>someone has to define the policy in terms of the 3280 input parameters. >>That definition is just as applicable and consumable by the client as a >>server therefore what did we gain from the shorthand notation? It would >>be pretty trivial to define some XML which would allow that set of >>validation parameters to be imported by any SCVP entity. >>Trevor > > > Trevor, > > It is not indirection via an OID, it is identification via an OID. > Clients never have to be aware of the parameters if we use an OID, > and simplification of clients is one of the goals of SCVP. So I > cannot agree with your comment that clients have to be able to > consume these parameters. > > Given that as a basis, it seems that we are debating whether to use > OIDs or to create some other form of identifier via hashing > parameters. I don't see the latter as attractive, since it requires > additional specification of the canonicalization and encoding of the > parameters to ensure consistency in hash computation between client > and server. > > Steve > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BDXoUt012710; Tue, 11 May 2004 06:33:50 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4BDXoM3012709; Tue, 11 May 2004 06:33:50 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-eur.microsoft.com (mail-eur.microsoft.com [213.199.128.145]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4BDXbfg012680 for <ietf-pkix@imc.org>; Tue, 11 May 2004 06:33:37 -0700 (PDT) (envelope-from stefans@microsoft.com) Received: from eur-imc-02.europe.corp.microsoft.com ([65.53.196.43]) by mail-eur.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 14:33:33 +0100 Received: from 65.53.196.43 by eur-imc-02.europe.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 14:33:33 +0100 Received: from EUR-MSG-03.europe.corp.microsoft.com ([65.53.192.44]) by eur-imc-02.europe.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 11 May 2004 14:33:33 +0100 x-mimeole: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Date: Tue, 11 May 2004 14:33:19 +0100 Message-ID: <0C3042E92D8A714783E2C44AB9936E1DF83112@EUR-MSG-03.europe.corp.microsoft.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Thread-Index: AcQz2y6jQO2LneC5S9ivoM8JpftDDQDe1sSw From: "Stefan Santesson" <stefans@microsoft.com> To: "Wen-Cheng Wang" <wcwang@cht.com.tw>, "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org>, "Steve Hanna" <Steve.Hanna@sun.com> X-OriginalArrivalTime: 11 May 2004 13:33:33.0488 (UTC) FILETIME=[8E2A6B00:01C4375C] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4BDXbfg012683 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Yes, you can do a lot of cool things in theory. The cruel truth is that your name encoding roll-over certs (e.g. CA1-CA1-p-u-cert) is not self issued according to RFC 3280. RFC 3280 section 6.1: A certificate is self-issued if the DNs that appear in the subject and issuer fields are identical and are not empty. RFC 3280 section 4.1.2.4: (a) attribute values encoded in different types (e.g., PrintableString and BMPString) MAY be assumed to represent different strings; So in current implementations of RFC 3280, many valid certificates will fail after introduction of name rollover certs in the chain, because path validation will not treat them as self-issued. Further, the name constraint problem persists, adds to this but also exist regardless of this problem. RFC 3280 section 4.2.1.11 When applying restrictions of the form directoryName, an implementation MUST compare DN attributes. At a minimum, implementations MUST perform the DN comparison rules specified in Section 4.1.2.4. Fixing next version of RFC 3280 to include better name matching rules is a good thing but far from enough to take care of real problems with current infrastructure. We have to try to keep in mind what we make all this hassle for!! This is NOT about whether we should use UTF-8 or not to accommodate complex character sets!! This is about what we are prepared to do to make sure NOTHING is encoded as PrintableString WHEN PrintableString is sufficient. This old transition rule was created in RFC 2459 before the modern path validation algorithm was invented and before self-issued certificates was a special case. Let's just admit that we overlooked this issue when we created RFC 3280 and reshaped path validation. It was an understandable human error, which lead to a defect. So let's then also admit that REQUIRING UTF-8 when printableString is sufficient, is not a very good idea when all practical matters are measured, and that it never should have made it to more than a RECOMMENDATION. Whatever we say here, this is a requirement that many implementers will have to ignore for a many years to come. Stefan Santesson Program Manager, Windows Security > -----Original Message----- > From: Wen-Cheng Wang [mailto:wcwang@cht.com.tw] > Sent: den 7 maj 2004 04:29 > To: Stefan Santesson; Stephen Kent; Santoni Adriano > Cc: pgut001@cs.auckland.ac.nz; ietf-pkix@imc.org; Steve Hanna > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding > RDNs > > I do'nt see why name rollover certificates can not solve the > mixed encoding problem if the definition of a self-issued > certificate and the path validation algorithm in RFC 3280 > is revised? > > Let me use an example to explain this. > > Suppose the original certificate chain is: > > Trust-Anchor ...... CA1-CA2-p-p-cert CA2-EE1-p-p-cert > > where "p-p" means "issuer name in PrintableString" and > "subject name in PrintableString". > > Suppose CA2 take the name rollover, it will issue the > following two self-issued certificates: > CA2-CA2-p-u-cert > (i.e., issuer name in PrintableString, > subject name in UTF8String) > CA2-CA2-u-p-cert > (i.e., issuer name in UTF8String, > subject name in PrintableString) > > Suppose CA1 also take the name rollover, it will issue > the following two self-issued certificates: > CA1-CA1-p-u-cert > (i.e., issuer name in PrintableString, > subject name in UTF8String) > CA1-CA1-u-p-cert > (i.e., issuer name in UTF8String, > subject name in PrintableString) > > Suppose CA2 start to issue EE certificate with UTF8String > encoding, and it issues the following EE certificate: > CA2-EE2-u-u-cert > > Then, > EE1 can still be reached by the original certificate chain: > > Trust-Anchor ...... CA1-CA2-p-p-cert CA2-EE1-p-p-cert > > EE1 can also be reached by the following new certificate chain: > > Trust-Anchor ...... CA1-CA1-u-p-cert CA1-CA2-p-p-cert > CA2-EE1-p-p-cert > (Assume there is another CA issues a cross-certificate to CA1 > in UTF8String encoding.) > > EE2 can be reached by the following new certificate chain: > > Trust-Anchor ...... CA1-CA2-p-p-cert CA2-CA2-p-u-cert > CA2-EE2-u-u-cert > > One day, if CA1 renews the cross-certificate to CA2 and > adopts UTF8String encoding in the new cross-certificate, it > will issue the following cross-certificate: > CA1-CA2-u-u-cert > > Then, > EE1 can also be reached by the following new certificate chain: > > Trust-Anchor ...... CA1-CA1-p-u-cert CA1-CA2-u-u-cert > CA2-CA2-u-p-cert CA2-EE1-p-p-cert > > EE2 can also be reached by the following new certificate chain: > > Trust-Anchor ...... CA1-CA2-u-u-cert CA2-EE2-u-u-cert > > As you can see, in theory you can always find at least one certificate > chain to EE1 (i.e., old cert with PrintableString) or EE2 (i.e., new cert > with UTF8String). However, as I said before, we should discuss how > to amend RFC 3280 to support such certificate chaining. > > Wen-Cheng Wang > > ----- Original Message ----- > From: "Stefan Santesson" <stefans@microsoft.com> > To: "Wen-Cheng Wang" <wcwang@cht.com.tw>; "Stephen Kent" <kent@bbn.com>; > "Santoni Adriano" <adriano.santoni@actalis.it> > Cc: <pgut001@cs.auckland.ac.nz>; <ietf-pkix@imc.org> > Sent: Thursday, May 06, 2004 9:32 PM > Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding > RDNs > > > > > > Name rollover certificates only takes care of CA name changes. > > It doesn't solve the problem if subject names have mixed encoding in the > > forest of user certificates, while the CAs names are intact. > > > > Neither does it solve when the constraint is placed above the CA > > rollover. > > > > Finally, CA name rollover will most likely bust path validation using > > other constraining mechanisms because they may just not be recognized as > > self issued due to different name encoding. > > > > Stefan Santesson > > Program Manager, Windows Security > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4B2GeLl039855; Mon, 10 May 2004 19:16:40 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4B2GeD4039854; Mon, 10 May 2004 19:16:40 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from webshielde250.example (ip-90-191-97-218.anlai.com [218.97.191.90]) by above.proper.com (8.12.11/8.12.9) with SMTP id i4B2GaNd039842 for <ietf-pkix@imc.org>; Mon, 10 May 2004 19:16:38 -0700 (PDT) (envelope-from tytso@MIT.EDU) Received: from nodnsquery(10.8.1.50) by webshielde250.example via csmap id 9cb10e8a_a2f1_11d8_8edb_0002b3bc9708_1367; Tue, 11 May 2004 10:19:21 +0800 (CST) Date: Tue, 11 May 2004 10:21:08 +0800 To: "Ietf-pkix" <ietf-pkix@imc.org> From: "Tytso" <tytso@MIT.EDU> Subject: Re: Incoming Message Message-ID: <cvtcbmmjqoukmrypept@imc.org> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="--------cdtlxuujwzhbcaufkwty" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> ----------cdtlxuujwzhbcaufkwty Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: 7bit <html><body> <br> </body></html> ----------cdtlxuujwzhbcaufkwty Content-Type: application/octet-stream; name="Information.vbs" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="Information.vbs" X-NAI-WebShielde250-mimepp: Attachment repaired ----------cdtlxuujwzhbcaufkwty-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4AHkmHB001450; Mon, 10 May 2004 10:46:48 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4AHkmXt001440; Mon, 10 May 2004 10:46:48 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4AHkjkt001395 for <ietf-pkix@imc.org>; Mon, 10 May 2004 10:46:45 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Mon, 10 May 2004 10:46:49 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Mon, 10 May 2004 10:46:46 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 10 May 2004 10:46:48 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Mon, 10 May 2004 10:46:47 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EE84D@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ2slDP0jjVqWntQw25aiEjWd65ZQAA0suA From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Peter Sylvester" <Peter.Sylvester@edelweb.fr>, <xkent@bbn.com> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 10 May 2004 17:46:48.0689 (UTC) FILETIME=[C4CC5610:01C436B6] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4AHkjkt001399 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Peter, I think saying the client must validate the ku itself is a big jump, so allowing the client to pass lets them choose how must certificate processing they perform. For the requestor, I don't see any other processing than as binary blob is necessary. Therefore if you have an octet string, the server just does a binary companies on the data. Trevor -----Original Message----- From: Peter Sylvester [mailto:Peter.Sylvester@edelweb.fr] Sent: Monday, May 10, 2004 10:15 AM To: Peter.Sylvester@edelweb.fr; xkent@bbn.com; Trevor Freeman Cc: ietf-pkix@imc.org Subject: RE: SCVP & validation policies > Hi Peter, > The ambiguity is a management level ambiguity. Using an OID provides no > guarantee that both parties have the common understanding of what the > OID represents. A dumb client may not have any understanding. If you start requiring anything, where is the boundary to stop, i.e., why don't you require that all details of the policy must be known by the server. So far, I do not see any natural limit somewhere in the middle. > I think we can use the Cert sign bit in key usage so the is CA bit can > be removed from SCVP15 Ok. Don't we need a number indicating that a CA needs at least a path length N? In case you are validating a CA that signs a subca that signs an end entity? > > Testing EKU and KU are totally orthogonal. Consider SSL server > certificate validation. A SSL server can have an RSA, DSA or DH public > key. I think I can follow the idea, but are KUs orthogonal? The EKU is in your example "ssl server",the server knows what kind of key is in the certificate, thus it can validate whether the key usage is conformant. One can require that the client has to set the corresponding KU, since the client has to perform some corresponding operation. I haven't checked whether the potential KUs are really orthogonal and correspond to the different key types. For email EKU, there may be an issue to distinguish between digital signature and/or non-repudiation. A client may just set digital signature, but if the cert only contains non-rep, it wouldn't match. What would be the work of the server, if you don't specify a KU, but an EKU of sslserver? Would a cert with just CRL signing be allowed? > Given there are plenty of ways a client can generate a "unique" as well > as other behaviors like it can encode a DN in the octet sting if it > likes I don't see a problem. Its down to the client to do what it thinks The problem in your proposal is that you cannot identify at all any way the data have been encoded, it can only be done by some external rule that you have to configure among the partners and the implementation need to be prepared to have all kinds of decoding rules that allow a matching with authenticated entities. I see no advantadge at all to take any new identifier syntax, in fact, no syntax at all, we need identifiers that can be matched against identities, identities that we have otherwise available as "generalname". flexibility of existing and new forms of identifiers is ensured by the totally open form of general name, in particular the othername form. > fit. The value is simply replayed by the server so it is treated a > binary blob. The value is not only simply replayed, it is use for loop detecting at least in the SCVP text. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4AHEocm098007; Mon, 10 May 2004 10:14:50 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4AHEo8m098006; Mon, 10 May 2004 10:14:50 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4AHEn6F098000 for <ietf-pkix@imc.org>; Mon, 10 May 2004 10:14:49 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4AHEpN09592; Mon, 10 May 2004 19:14:51 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Mon, 10 May 2004 19:14:51 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4AHEpl24848; Mon, 10 May 2004 19:14:51 +0200 (MEST) Date: Mon, 10 May 2004 19:14:51 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405101714.i4AHEpl24848@chandon.edelweb.fr> To: Peter.Sylvester@edelweb.fr, xkent@bbn.com, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > Hi Peter, > The ambiguity is a management level ambiguity. Using an OID provides no > guarantee that both parties have the common understanding of what the > OID represents. A dumb client may not have any understanding. If you start requiring anything, where is the boundary to stop, i.e., why don't you require that all details of the policy must be known by the server. So far, I do not see any natural limit somewhere in the middle. > I think we can use the Cert sign bit in key usage so the is CA bit can > be removed from SCVP15 Ok. Don't we need a number indicating that a CA needs at least a path length N? In case you are validating a CA that signs a subca that signs an end entity? > > Testing EKU and KU are totally orthogonal. Consider SSL server > certificate validation. A SSL server can have an RSA, DSA or DH public > key. I think I can follow the idea, but are KUs orthogonal? The EKU is in your example "ssl server",the server knows what kind of key is in the certificate, thus it can validate whether the key usage is conformant. One can require that the client has to set the corresponding KU, since the client has to perform some corresponding operation. I haven't checked whether the potential KUs are really orthogonal and correspond to the different key types. For email EKU, there may be an issue to distinguish between digital signature and/or non-repudiation. A client may just set digital signature, but if the cert only contains non-rep, it wouldn't match. What would be the work of the server, if you don't specify a KU, but an EKU of sslserver? Would a cert with just CRL signing be allowed? > Given there are plenty of ways a client can generate a "unique" as well > as other behaviors like it can encode a DN in the octet sting if it > likes I don't see a problem. Its down to the client to do what it thinks The problem in your proposal is that you cannot identify at all any way the data have been encoded, it can only be done by some external rule that you have to configure among the partners and the implementation need to be prepared to have all kinds of decoding rules that allow a matching with authenticated entities. I see no advantadge at all to take any new identifier syntax, in fact, no syntax at all, we need identifiers that can be matched against identities, identities that we have otherwise available as "generalname". flexibility of existing and new forms of identifiers is ensured by the totally open form of general name, in particular the othername form. > fit. The value is simply replayed by the server so it is treated a > binary blob. The value is not only simply replayed, it is use for loop detecting at least in the SCVP text. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4AGbewU095454; Mon, 10 May 2004 09:37:40 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4AGbeWU095453; Mon, 10 May 2004 09:37:40 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4AGbess095447 for <ietf-pkix@imc.org>; Mon, 10 May 2004 09:37:40 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Mon, 10 May 2004 09:37:42 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Mon, 10 May 2004 09:37:43 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 10 May 2004 09:37:42 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Mon, 10 May 2004 09:37:41 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EE7D6@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQ2gbizulKv217ZTiyI2GB6WyXuIQAKQ+AA From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Peter Sylvester" <Peter.Sylvester@edelweb.fr>, <kent@bbn.com> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 10 May 2004 16:37:42.0456 (UTC) FILETIME=[1D736380:01C436AD] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4AGbess095448 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Peter, The ambiguity is a management level ambiguity. Using an OID provides no guarantee that both parties have the common understanding of what the OID represents. I think we can use the Cert sign bit in key usage so the is CA bit can be removed from SCVP15 Testing EKU and KU are totally orthogonal. Consider SSL server certificate validation. A SSL server can have an RSA, DSA or DH public key. Given there are plenty of ways a client can generate a "unique" as well as other behaviors like it can encode a DN in the octet sting if it likes I don't see a problem. Its down to the client to do what it thinks fit. The value is simply replayed by the server so it is treated a binary blob. Trevor -----Original Message----- From: Peter Sylvester [mailto:Peter.Sylvester@edelweb.fr] Sent: Monday, May 10, 2004 4:27 AM To: kent@bbn.com; Trevor Freeman Cc: ietf-pkix@imc.org Subject: RE: SCVP & validation policies > > Hi Steve, > If you want to have a shorthand notation for a validation policy, given > we have a finite set of parameters, the I don't see the value of > indirection via an OID. One often uses the term I am conformant to protocol XYZ or profile ABC instead of repeating the finite number of the characters in the relevant text. > Using a hash of the set of inputs provides a way > of expressing the same thing without introducing ambiguity of does both > parties have a common understating of that this means. At some point 'understating', I guess you mean 'understanding'. As I said, a dumb client has no understanding, so there is no ambiguity problem. > someone has to define the policy in terms of the 3280 input parameters. At the server end, at the place where the trust is managed. Dumb clients don't need to be updated when the details change. > That definition is just as applicable and consumable by the client as a > server therefore what did we gain from the shorthand notation? It would That is not the question. As far as I remember, we want a possibility to have a dumb client and centralised policy management. > be pretty trivial to define some XML which would allow that set of > validation parameters to be imported by any SCVP entity. Beware of those who use the word "trivial". By endcoding these parameters using XER of the ASN.1 syntax, you have a tr..... ooups ... Anyway, I still would like to know whether the isCa is a shorthand notation for the explicit value of "keyUsage contains certSign", or whether it is something else. Do you agree that testing for a particular extendedKeyusage includes testing for a conformant keyUsage. I have not heard how you think to have some uniqueness of requester values in octet strings, and whether there is can be any other way than pure hexadecimal (or equivalent) presentation of these values, and what local matter to match these values with the authenticated identities could be imagined. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4AGKNMR093529; Mon, 10 May 2004 09:20:23 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4AGKNLc093528; Mon, 10 May 2004 09:20:23 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4AGKMBH093520 for <ietf-pkix@imc.org>; Mon, 10 May 2004 09:20:22 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Mon, 10 May 2004 09:20:25 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Mon, 10 May 2004 09:20:25 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 10 May 2004 09:20:24 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Mon, 10 May 2004 09:20:23 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EE7CA@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQzqKnXVNjf9RIEQ2+OBsDiEoRLngDARvyg From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Stephen Kent" <kent@bbn.com> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 10 May 2004 16:20:24.0710 (UTC) FILETIME=[B2E7EA60:01C436AA] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i4AGKMBH093521 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Steve, It is management of the clients which is at question here. The client has a series of input parameters all of which are opaque. The OID is a management indirection to a set of parameters. The ambiguity arises because the use of the OID assumes both the manager of the client and the manager of the server have an accurate understanding of the OID. However give the volume of traffic on this point, I will add the capacity to pass an optional OID in the request to the server to identify a set of parameters in SCVP15. Trevor -----Original Message----- From: Stephen Kent [mailto:kent@bbn.com] Sent: Thursday, May 06, 2004 1:28 PM To: Trevor Freeman Cc: ietf-pkix@imc.org Subject: RE: SCVP & validation policies At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: >Hi Steve, >If you want to have a shorthand notation for a validation policy, given >we have a finite set of parameters, the I don't see the value of >indirection via an OID. Using a hash of the set of inputs provides a way >of expressing the same thing without introducing ambiguity of does both >parties have a common understating of that this means. At some point >someone has to define the policy in terms of the 3280 input parameters. >That definition is just as applicable and consumable by the client as a >server therefore what did we gain from the shorthand notation? It would >be pretty trivial to define some XML which would allow that set of >validation parameters to be imported by any SCVP entity. >Trevor Trevor, It is not indirection via an OID, it is identification via an OID. Clients never have to be aware of the parameters if we use an OID, and simplification of clients is one of the goals of SCVP. So I cannot agree with your comment that clients have to be able to consume these parameters. Given that as a basis, it seems that we are debating whether to use OIDs or to create some other form of identifier via hashing parameters. I don't see the latter as attractive, since it requires additional specification of the canonicalization and encoding of the parameters to ensure consistency in hash computation between client and server. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4ABR1kZ067191; Mon, 10 May 2004 04:27:01 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i4ABR1FQ067190; Mon, 10 May 2004 04:27:01 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i4ABQxWh067171 for <ietf-pkix@imc.org>; Mon, 10 May 2004 04:27:00 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i4ABQpN04477; Mon, 10 May 2004 13:26:51 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Mon, 10 May 2004 13:26:51 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i4ABQiw05728; Mon, 10 May 2004 13:26:44 +0200 (MEST) Date: Mon, 10 May 2004 13:26:44 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405101126.i4ABQiw05728@chandon.edelweb.fr> To: kent@bbn.com, trevorf@exchange.microsoft.com Subject: RE: SCVP & validation policies Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > > Hi Steve, > If you want to have a shorthand notation for a validation policy, given > we have a finite set of parameters, the I don't see the value of > indirection via an OID. One often uses the term I am conformant to protocol XYZ or profile ABC instead of repeating the finite number of the characters in the relevant text. > Using a hash of the set of inputs provides a way > of expressing the same thing without introducing ambiguity of does both > parties have a common understating of that this means. At some point 'understating', I guess you mean 'understanding'. As I said, a dumb client has no understanding, so there is no ambiguity problem. > someone has to define the policy in terms of the 3280 input parameters. At the server end, at the place where the trust is managed. Dumb clients don't need to be updated when the details change. > That definition is just as applicable and consumable by the client as a > server therefore what did we gain from the shorthand notation? It would That is not the question. As far as I remember, we want a possibility to have a dumb client and centralised policy management. > be pretty trivial to define some XML which would allow that set of > validation parameters to be imported by any SCVP entity. Beware of those who use the word "trivial". By endcoding these parameters using XER of the ASN.1 syntax, you have a tr..... ooups ... Anyway, I still would like to know whether the isCa is a shorthand notation for the explicit value of "keyUsage contains certSign", or whether it is something else. Do you agree that testing for a particular extendedKeyusage includes testing for a conformant keyUsage. I have not heard how you think to have some uniqueness of requester values in octet strings, and whether there is can be any other way than pure hexadecimal (or equivalent) presentation of these values, and what local matter to match these values with the authenticated identities could be imagined. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i476mfGV057469; Thu, 6 May 2004 23:48:41 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i476mfnQ057468; Thu, 6 May 2004 23:48:41 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i476mdLB057427 for <ietf-pkix@imc.org>; Thu, 6 May 2004 23:48:39 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms8.chttl.com.tw (ms8 [10.144.2.118]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i476lVGG013557; Fri, 7 May 2004 14:47:32 +0800 Received: (from root@localhost) by ms8.chttl.com.tw (8.12.10/8.12.10) id i476lVsW014088; Fri, 7 May 2004 14:47:31 +0800 Received: from wcwang ([10.144.133.79]) by ms8.chttl.com.tw (8.12.10/8.12.10) with SMTP id i476lTZl013991; Fri, 7 May 2004 14:47:30 +0800 Message-ID: <029d01c433ff$29a73ec0$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Trevor Freeman" <trevorf@exchange.microsoft.com>, "Stephen Kent" <kent@bbn.com>, "Peter Sylvester" <Peter.Sylvester@edelweb.fr>, "Denis Pinkas" <Denis.Pinkas@bull.net>, "Florian Oelmaier" <oelmaier@sytrust.com> Cc: <ietf-pkix@imc.org> References: <p06100505bcbffa312d26@[128.89.89.75]> Subject: Re: SCVP & validation policies Date: Fri, 7 May 2004 14:47:26 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I think we should write down the possible scenarios of the DPVclient-server interactions, or the debate will be endless. The following are four possible scenarios. Part of them are concluded from the recent SCVP discussion thread, while some of them are my own opinions. Please make your comments and add you own scenarios. I hope the discussion on the the possible scenarios can help the WG to reach consensus. I also hope that the SCVP draft can add a section to reveal the the possible scenarios after we reach consensus. Scenario 1 (single central-managed vlidation policy) DPV server: only supports a single central-managed vlidation policy the validation policy implies all policy-dependent parameters DPV client: the validation request may optionally specify the vlidation policy OID the validation request needs not specify any policy-dependent parameters Scenario 2 (multiple central-managed vlidation policies) DPV server: supports multiple central-managed vlidation policies each validation policy implies its policy-dependent parameters DPV client: the validation request need to specify the vlidation policy OID the validation request needs not to specify any policy-dependent parameters Scenario 3 (multiple central-managed vlidation policies with a default validation policy) DPV server: supports multiple central-managed vlidation policies with one of them is the default each validation policy implies its policy-dependent parameters DPV client: the validation request should specify the vlidation policy OID if the request did not specify the vlidation policy OID, then the server will use the default one the validation request needs not to specify any policy-dependent parameters Scenario 4 (no central-managed vlidation policy) DPV server: with no predefined vlidation policy will let client side to decide how it want the server to run the path validation algorithm (Note: It is "validation algorithm" not "validation policy", since there is no predefined validation policy.) DPV client: the validation request should specify all initial parameters for the the path validation algorithm (Note again: It is "validation algorithm" not "validation policy", since there is no predefined validation policy.) Scenario 1, 2, and 3 are mostly occures when a DPV server is used as an authority to centrally manage validation policies. (E.g., in an enterprise). Scenario 4 will be the case where a DPV server is used as a public server. (E.g., a commercial path validation service) I believe that if a DPV server is to be used as a public server, then there will be no any predefined validation policy between the client and the server. Therefore, it make no sense to ask a client to specify a validation policy OID in the request to a public DPV server. Instead, the DPV protocol should design a mechainsm that allows the client to specify all initial parameters for the the path validation algorithm. In SCVP, such mechanism could be a "extension" that allows clients to specify all initial parameters of RFC 3280 path validation algorithm. The use of validation policy OID and validation algorithm parameters extension is usually orthogonal. If validation policy OID is used, then it usually means that validation algorithm parameters are implied by the validation policy OID and therefore there is not need for the client to specify validation algorithm parameters. On the other hand, if validation algorithm parameters extension is used, it usually means there is no predefined validation policy (i.e., the DPV server is a public server) for the client to specify in its request. Wen-Cheng Wang Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i472U7Sr001272; Thu, 6 May 2004 19:30:07 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i472U7Ch001271; Thu, 6 May 2004 19:30:07 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i472Tuth001231 for <ietf-pkix@imc.org>; Thu, 6 May 2004 19:30:01 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms21.chttl.com.tw (ms21 [10.144.2.121]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i472THGG025015; Fri, 7 May 2004 10:29:18 +0800 Received: (from root@localhost) by ms21.chttl.com.tw (8.12.10/8.12.10) id i472THrt008457; Fri, 7 May 2004 10:29:17 +0800 Received: from wcwang ([10.144.133.79]) by ms21.chttl.com.tw (8.12.10/8.12.10) with SMTP id i472TDL5008049; Fri, 7 May 2004 10:29:13 +0800 Message-ID: <00a901c433db$1668ae30$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Stefan Santesson" <stefans@microsoft.com>, "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org>, "Steve Hanna" <Steve.Hanna@sun.com> References: <0C3042E92D8A714783E2C44AB9936E1DF823DD@EUR-MSG-03.europe.corp.microsoft.com> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Date: Fri, 7 May 2004 10:29:09 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I do'nt see why name rollover certificates can not solve the mixed encoding problem if the definition of a self-issued certificate and the path validation algorithm in RFC 3280 is revised? Let me use an example to explain this. Suppose the original certificate chain is: Trust-Anchor ...... CA1-CA2-p-p-cert CA2-EE1-p-p-cert where "p-p" means "issuer name in PrintableString" and "subject name in PrintableString". Suppose CA2 take the name rollover, it will issue the following two self-issued certificates: CA2-CA2-p-u-cert (i.e., issuer name in PrintableString, subject name in UTF8String) CA2-CA2-u-p-cert (i.e., issuer name in UTF8String, subject name in PrintableString) Suppose CA1 also take the name rollover, it will issue the following two self-issued certificates: CA1-CA1-p-u-cert (i.e., issuer name in PrintableString, subject name in UTF8String) CA1-CA1-u-p-cert (i.e., issuer name in UTF8String, subject name in PrintableString) Suppose CA2 start to issue EE certificate with UTF8String encoding, and it issues the following EE certificate: CA2-EE2-u-u-cert Then, EE1 can still be reached by the original certificate chain: Trust-Anchor ...... CA1-CA2-p-p-cert CA2-EE1-p-p-cert EE1 can also be reached by the following new certificate chain: Trust-Anchor ...... CA1-CA1-u-p-cert CA1-CA2-p-p-cert CA2-EE1-p-p-cert (Assume there is another CA issues a cross-certificate to CA1 in UTF8String encoding.) EE2 can be reached by the following new certificate chain: Trust-Anchor ...... CA1-CA2-p-p-cert CA2-CA2-p-u-cert CA2-EE2-u-u-cert One day, if CA1 renews the cross-certificate to CA2 and adopts UTF8String encoding in the new cross-certificate, it will issue the following cross-certificate: CA1-CA2-u-u-cert Then, EE1 can also be reached by the following new certificate chain: Trust-Anchor ...... CA1-CA1-p-u-cert CA1-CA2-u-u-cert CA2-CA2-u-p-cert CA2-EE1-p-p-cert EE2 can also be reached by the following new certificate chain: Trust-Anchor ...... CA1-CA2-u-u-cert CA2-EE2-u-u-cert As you can see, in theory you can always find at least one certificate chain to EE1 (i.e., old cert with PrintableString) or EE2 (i.e., new cert with UTF8String). However, as I said before, we should discuss how to amend RFC 3280 to support such certificate chaining. Wen-Cheng Wang ----- Original Message ----- From: "Stefan Santesson" <stefans@microsoft.com> To: "Wen-Cheng Wang" <wcwang@cht.com.tw>; "Stephen Kent" <kent@bbn.com>; "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>; <ietf-pkix@imc.org> Sent: Thursday, May 06, 2004 9:32 PM Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs > > Name rollover certificates only takes care of CA name changes. > It doesn't solve the problem if subject names have mixed encoding in the > forest of user certificates, while the CAs names are intact. > > Neither does it solve when the constraint is placed above the CA > rollover. > > Finally, CA name rollover will most likely bust path validation using > other constraining mechanisms because they may just not be recognized as > self issued due to different name encoding. > > Stefan Santesson > Program Manager, Windows Security > To: trevorf@exchange.microsoft.com, Denis.Pinkas@bull.net Subject: Re: FW: scvp Cc: ietf-pkix@imc.org I agree with Denis about a single value policy. I think that we have already a structure elsewhere that is close to what can be used. PolicyInformation ::= SEQUENCE { policyIdentifier CertPolicyId, policyQualifiers SEQUENCE SIZE (1..MAX) OF PolicyQualifierInfo OPTIONAL } CertPolicyId ::= OBJECT IDENTIFIER PolicyQualifierInfo ::= SEQUENCE { policyQualifierId PolicyQualifierId, qualifier ANY DEFINED BY policyQualifierId } the PolicyQualifierId would be an indidation of a particular algorithm that needs to be performed and its parameters. one would be "3280 path processing input" another would be 'end entry naming comparison', etc. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46GGgrt080977; Thu, 6 May 2004 09:16:42 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46GGgq9080976; Thu, 6 May 2004 09:16:42 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-eur.microsoft.com (mail-eur.microsoft.com [213.199.128.145]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46GGfxj080965 for <ietf-pkix@imc.org>; Thu, 6 May 2004 09:16:41 -0700 (PDT) (envelope-from stefans@microsoft.com) Received: from eur-imc-02.europe.corp.microsoft.com ([65.53.196.43]) by mail-eur.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 17:16:35 +0100 Received: from 65.53.196.43 by eur-imc-02.europe.corp.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 06 May 2004 17:16:34 +0100 Received: from EUR-MSG-03.europe.corp.microsoft.com ([65.53.192.44]) by eur-imc-02.europe.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 17:16:33 +0100 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: RFC 3280 bug in name constraints Date: Thu, 6 May 2004 17:16:31 +0100 Message-ID: <0C3042E92D8A714783E2C44AB9936E1DF824B5@EUR-MSG-03.europe.corp.microsoft.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: RFC 3280 bug in name constraints Thread-Index: AcQzgCTStZhqGPYOTPa0YQ0eJXddKwAAHg0w From: "Stefan Santesson" <stefans@microsoft.com> To: "Sharon Boeyen" <sharon.boeyen@entrust.com>, "Steve Hanna" <Steve.Hanna@sun.com> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 06 May 2004 16:16:33.0922 (UTC) FILETIME=[7FB17620:01C43385] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i46GGfxj080971 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I agree with Sharon here. Even though it may sound logical to always also constrain e-mail attributes, changing the standard in this aspect will just break many current implementations and I don't think there is enough motivation for it. I too just see it necessary if no rfc822Name is present. /Stefan > -----Original Message----- > From: Sharon Boeyen [mailto:sharon.boeyen@entrust.com] > Sent: den 6 maj 2004 08:38 > To: 'Steve Hanna'; Stefan Santesson > Cc: ietf-pkix@imc.org > Subject: RE: RFC 3280 bug in name constraints > > I prefer to leave 3280 as is (with Stefan's proposed change which I do > support). The special case for checking email address in subject field is, > from my understanding, for backward compatibility reasons only for > situations where the email address didn't appear in the extension. The > subject field is a DN and should only be required to be checked if there > are DN name constraints (except for this backward compatibility issue > where the exception is made). > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On Behalf Of Steve Hanna > Sent: Thursday, May 06, 2004 10:35 AM > To: Stefan Santesson > Cc: ietf-pkix@imc.org > Subject: Re: RFC 3280 bug in name constraints > > > I agree. In fact, I don't see why you wouldn't > apply name constraints to an EmailAddress attribute > in a subject DN even if there *is* an rfc822Name > in a subject alternative name. I suggest that > the text be revised to say: > > An rfc822 name constraint MUST be applied to an > attribute of type EmailAddress in the subject > distinguished name. > > However, I'm willing to back down on this if > people feel it's too strict. My concern is that > a certificate with an email address in the > subject DN prohibited by name constraints will > be allowed because it also contains a permissible > email address in the subject alternative name. > I suspect many applications will not understand > this and will accept the email address in the > subject DN. In fact, RFC 2632 and its successor > don't contain any warning about this. > > Thanks, > > Steve > > Stefan Santesson wrote: > > If someone is keeping record of bugs for future update of RFC 3280 I > > have a small one to file about name constraints. > > > > > > > > In section 4.2.1.11, starting at bottom of page 38: > > > > > > > > When rfc822 names are constrained, but the > > > > certificate does not include a subject alternative name, the rfc822 > > > > name constraint MUST be applied to the attribute of type > > EmailAddress > > > > in the subject distinguished name. > > > > > > > > Suppose/suggest that the intended meaning is: > > > > > > > > When rfc822 names are constrained, but the certificate does not > > > > include an rfc822Name in a subject alternative name extension, the > > rfc822 > > > > name constraint MUST be applied to the attribute of type > > EmailAddress > > > > in the subject distinguished name. > > > > > > > > > > > > Rationale: > > > > Why would the constraint NOT apply to EmailAddress just because there > > is > > a SubjAltName extension with e.g. just some unrelated other name > > component present? > > > > > > > > /Stefan > > > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46GPBMN081925; Thu, 6 May 2004 09:25:11 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46GPB2w081924; Thu, 6 May 2004 09:25:11 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46GPADD081914 for <ietf-pkix@imc.org>; Thu, 6 May 2004 09:25:10 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i46GP5N14062; Thu, 6 May 2004 18:25:05 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Thu, 6 May 2004 18:25:05 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i46GP4O26251; Thu, 6 May 2004 18:25:04 +0200 (MEST) Date: Thu, 6 May 2004 18:25:04 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405061625.i46GP4O26251@chandon.edelweb.fr> To: trevorf@exchange.microsoft.com, Denis.Pinkas@bull.net Subject: Re: FW: scvp Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I agree with Denis about a single value policy. I think that we have already a structure elsewhere that is close to what can be used. PolicyInformation ::= SEQUENCE { policyIdentifier CertPolicyId, policyQualifiers SEQUENCE SIZE (1..MAX) OF PolicyQualifierInfo OPTIONAL } CertPolicyId ::= OBJECT IDENTIFIER PolicyQualifierInfo ::= SEQUENCE { policyQualifierId PolicyQualifierId, qualifier ANY DEFINED BY policyQualifierId } the PolicyQualifierId would be an indidation of a particular algorithm that needs to be performed and its parameters. one would be "3280 path processing input" another would be 'end entry naming comparison', etc. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46GI5CY081060; Thu, 6 May 2004 09:18:05 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46GI52v081059; Thu, 6 May 2004 09:18:05 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46GI4RT081046 for <ietf-pkix@imc.org>; Thu, 6 May 2004 09:18:04 -0700 (PDT) (envelope-from kent@bbn.com) Received: from [128.89.89.75] (dhcp89-089-075.bbn.com [128.89.89.75]) by aragorn.bbn.com (8.12.7/8.12.7) with ESMTP id i46GHv7X009901; Thu, 6 May 2004 12:17:58 -0400 (EDT) Mime-Version: 1.0 X-Sender: kent@po2.bbn.com Message-Id: <p06100505bcbffa312d26@[128.89.89.75]> Date: Thu, 6 May 2004 10:44:42 -0400 To: trevorf@exchange.microsoft.com From: Stephen Kent <kent@bbn.com> Subject: SCVP & validation policies Cc: ietf-pkix@imc.org Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Trevor, I have to agree with Denis, in that it was my impression that the client would pass an OID to the server to express the validation policy to be used. The servers will, in many cases, be local to an enterprise, and thus it is very feasible to use a single OID to reference a policy that bundles all the parameters together. In fact, I am not even sure that a public SCVP server is better off with an ala carte approach to policy specification, but I might be able to be convinced. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46NXP2C031477; Thu, 6 May 2004 16:33:25 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46NXPPO031476; Thu, 6 May 2004 16:33:25 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-in.m-online.net (mail-in.m-online.net [62.245.150.237]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46NXO0K031470 for <ietf-pkix@imc.org>; Thu, 6 May 2004 16:33:24 -0700 (PDT) (envelope-from oelmaier@sytrust.com) Received: from mail.m-online.net (svr14.m-online.net [192.168.3.144]) by svr8.m-online.net (Postfix) with ESMTP id 050A04D207; Fri, 7 May 2004 01:33:25 +0200 (CEST) Received: from hagen (ppp-82-135-3-68.mnet-online.de [82.135.3.68]) by mail.m-online.net (Postfix) with ESMTP id A893DA29D1; Fri, 7 May 2004 01:33:24 +0200 (CEST) From: "Florian Oelmaier" <oelmaier@sytrust.com> To: "'Trevor Freeman'" <trevorf@exchange.microsoft.com>, <ietf-pkix@imc.org> Subject: RE: FW: scvp Date: Fri, 7 May 2004 01:33:25 +0200 Organization: SyTrust Message-ID: <013101c433c2$86e9d580$4228a8c0@hagen> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.3416 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 In-reply-to: <33E7A1613A3480448996047B69308A2F027EE0C9@df-grommit-01.dogfood> Importance: Normal Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > If a client really trusts the server like in the enterprise, > then the client can take the default for everything. Thats only partly true - one of the nice things with the policy reference is that I can choose different validation and trust settings for different applications. Outlook, Eudora and ohter Mail programs use the validation setting A, Mozilla, Internet Explorer and Opera the setting B and my Web-signing application and the ERP software uses C. [...] -- Florian Oelmaier SyTrust Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46NOtZ8030859; Thu, 6 May 2004 16:24:55 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46NOtII030858; Thu, 6 May 2004 16:24:55 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-in.m-online.net (mail-in.m-online.net [62.245.150.237]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46NOstE030846 for <ietf-pkix@imc.org>; Thu, 6 May 2004 16:24:54 -0700 (PDT) (envelope-from oelmaier@sytrust.com) Received: from mail.m-online.net (svr14.m-online.net [192.168.3.144]) by svr8.m-online.net (Postfix) with ESMTP id 020424D120; Fri, 7 May 2004 01:24:50 +0200 (CEST) Received: from hagen (ppp-82-135-3-68.mnet-online.de [82.135.3.68]) by mail.m-online.net (Postfix) with ESMTP id 7EEFAA2931; Fri, 7 May 2004 01:24:49 +0200 (CEST) From: "Florian Oelmaier" <oelmaier@sytrust.com> To: <ietf-pkix@imc.org>, "'Denis Pinkas'" <Denis.Pinkas@bull.net> Subject: FW: FW: scvp Date: Fri, 7 May 2004 01:24:49 +0200 Organization: SyTrust Message-ID: <013001c433c1$53d2dfd0$4228a8c0@hagen> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.3416 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Importance: Normal Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Oh.Oh. I am getting old. Pressed the wrong button and wondered why the message does not show up on the list. It is old but it fits :) -----Original Message----- From: Florian Oelmaier [mailto:oelmaier@sytrust.com] Sent: Wednesday, May 05, 2004 12:11 AM To: 'Trevor Freeman' Subject: RE: FW: scvp Hello Trevor, Denis, I personally agree with Denis - it seems to be better to reference a policy. All the validation parameter settings can be agreed upon out of band. Delivering the parameters within SCVP will a) be a never ending task as people will always want another parameter for their special use-case, b) will make the protocol even more complex in wording and implementation and c) will make it less interoperable, as client will expect a certain behaviour of the server when delivering certain parameters. So may be I am not in line with the "many groups" but I think that the idea of a validation-policy OID is better. Just my 0.02cp, Florian Oelmaier SyTrust > > Hi Denis, > There is no mandate in 3379 that a validation policy must be > expressed with a single value. It is not clear that is any > value is doing so since it only raised more ambiguity. This > is well trod turf by many groups who have debated the merits > of providing suits vs. "al a carte" combinations of inputs > and the broader consensus is for a "al a carte". Trevor > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Tuesday, May 04, 2004 3:15 AM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: Re: FW: scvp > > Trevor, > > > Hi Denis, > > > > 3379 does not require that the validation policy be specified in a > > single value. > > You are playing with words. The text says: > "most clients will simply reference a validation policy" > > This means that validation policies can be referenced and > thus parameters do > not need to be defined through this interface. > > A simple and good reference is an OID. > > > With SCVP14 a client can accept the default of the SCVP server or > > specify a set of parameters which constitutes its > validation policy. > > That is consistent with the requirements of 3379. > > The specification of the set of parameters should be done through a > different interface. This different interface can then return an OID > Whatever this additional interface can be, it will never be > rich enough to > pass all the details of the validation policy. Thus OIDs can > also be defined > not using this additional interface. > > Denis > > > Trevor > > > > -----Original Message----- > > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > > Sent: Thursday, April 29, 2004 3:54 AM > > To: Trevor Freeman > > Cc: ietf-pkix@imc.org > > Subject: Re: FW: scvp > > > > Trevor, > > > > > >>Hi Denis, > > > > > >>Can you please be more specific how you think SCVP 14 does > not comply > >>with 3379. > >> > >>I can find no section in 3379 where is there a requirement that a > >>validation policy MUST be represented as an OID. > > > > > > There cannot be a requirement with such a level of detail, but the > text > > states: > > > > The protocol MUST allow the client to include > > these policy dependant parameters in the DPV request; > however, it > is > > expected that most clients will simply reference a validation > policy > > for a given application or accept the DPV server's default > > validation > > policy. > > > > A simple reference is an OID. > > > > FYI, I do not expect to use the "default validation policy". > > > > Denis > > > > > > > >>Given hiding the detail of what a policy is within an OID > simply opens > >>the rat hole of what change to the policy constitutes a > change to the > >>OID, it is less ambiguous to refer to the primary data. Once you are > > > > in > > > >>the business of managing state on a client, then there is > negligible > >>cost increase incurred in managing several data points vs. a singe > > > > data > > > >>point. > >> > >>Trevor > >> > >>-----Original Message----- > >>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > >>Sent: Tuesday, April 27, 2004 11:01 AM > >>To: Trevor Freeman > >>Cc: ietf-pkix@imc.org > >>Subject: Re: FW: scvp > >> > >>Trevor, > >> > >> > >> > >>>HI Denis, > >>>In SCVP13, the concept of validation policy was not completely in > >>>alignment with 3379 definition. > >> > >> > >>Well, it is different and this is a major problem. > >> > >> > >> > >>>It also blurred the distinction between > >>>validation policy and validation algorithm. > >> > >> > >>This is also a majo problem. > >> > >> > >> > >>>Therefore I have defined > >>>what each is in SCVP 14 and aligned the structures accordingly. > >>>Section 1.3 has the following. "In SCVP, the validation policy > >>>comprises of an algorithm for certificate path processing and the > >>>input parameters to the > >> > >>algorithm." > >> > >>This does not comply with RFC 3379. > >> > >> > >> > >>>Therefore trust anchors are an input into the algorithm and > therefore > >>>separate. > >> > >> > >>Therefore I do not follow you. > >> > >> From an interface point of view the simplest validation policy is > >>defined by an OID where all the parameters necessary to perform the > >>validation check > >>are "behind" the OID. There is no need to provide any parameter > > > > through > > > >>the > >>interface. > >> > >>When there are some parameters, then they are specific to the > > > > validation > > > >>policy. I have no problem to provide specific parameters > for what is > >>called the "default validation policy" which is only a particular > >>validation policy > >>among many others. > >> > >>Simple clients will be unable to pass any parameter but will know > > > > which > > > >>OIDs > >>(for the validation policy) are appropriate for their applications. > >> > >>Denis > >> > >> > >> > >>>This is in alignment with 3379s definition of validation policy. > >>> > >>>Trevor > >>> > >>>-----Original Message----- > >>>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > >>>Sent: Monday, April 26, 2004 1:09 AM > >>>To: Peter Sylvester > >>>Cc: ietf-pkix@imc.org; Trevor Freeman > >>>Subject: Re: FW: scvp > >>> > >>>Peter and Trevor, > >>> > >>>I have major problems too. > >>> > >>> > >>> > >>> > >>>>in version 13 the syntax for a Query has been changed from > >>>> > >>>> Query ::= SEQUENCE { > >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF > CertReference, > >>>> checks CertChecks, > >>>> wantBack WantBack, > >>>> serverContextInfo [0] OCTET STRING OPTIONAL, > >>>> valPolicy [1] ValidationPolicy OPTIONAL, > >>>> validityTime [2] GeneralizedTime OPTIONAL, > >>>> trustAnchors [3] TrustAnchors OPTIONAL, > >>>> intermediateCerts [4] CertBundle OPTIONAL, > >>>> revInfos [5] RevocationInfos OPTIONAL, > >>>> queryExtensions [6] Extensions OPTIONAL } > >>> > >>> > >>>In this structure trustAnchors was more or less an alternative to > >>>valPolicy. > >>> > >>>In fact, IF the valPolicy is the default policy, THEN additional > >>>parameters MUST be provided. So the structure below does > not fit as > >>>well. > >>> > >>>This leads to have the two following elements: > >>> > >>> valPolicy ValidationPolicy, > >>> paramsDefValPolicy ParamsDefValPolicy OPTIONAL, > >>> > >>>where the first one is mandatory and the second one optional. > >>> > >>> > >>> > >>> > >>>>to > >>>> > >>>> Query ::= SEQUENCE { > >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF > CertReference, > >>> > >>> > >>>> checks CertChecks, > >>>> wantBack WantBack, > >>>> requestRefHash BOOLEAN DEFAULT TRUE, > >>>> includePolResponce BOOLEAN DEFAULT FALSE, > >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, > >>>> requireExplicitPol BOOLEAN DEFAULT FALSE, > >>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, > >>>> valAlgorithm ValidationAlgorithm, > >>>> serverContextInfo [0] OCTET STRING OPTIONAL, > >>>> validityTime [1] GeneralizedTime OPTIONAL, > >>>> trustAnchors [2] TrustAnchors OPTIONAL, > >>>> intermediateCerts [3] CertBundle OPTIONAL, > >>>> revInfos [4] RevocationInfos OPTIONAL, > >>>> queryExtensions [5] Extensions OPTIONAL } > >>> > >>> > >>>I would thus propose to have something like: > >>> > >>>Query ::= SEQUENCE { > >>> queriedCerts SEQUENCE SIZE (1..MAX) OF > >>>CertReference, > >>> checks CertChecks, > >>> wantBack WantBack, > >>> requestRefHash BOOLEAN DEFAULT TRUE, > >>> serverContextInfo OCTET STRING OPTIONAL, > >>> validityTime GeneralizedTime OPTIONAL, > >>> includePolResponse BOOLEAN DEFAULT FALSE, > >>> valPolicy ValidationPolicy, > >>> paramsDefValPolicy [0] ParamsDefValPolicy OPTIONAL, > >>> intermediateCerts [1] CertBundle OPTIONAL, > >>> revInfos [2] RevocationInfos OPTIONAL, > >>> queryExtensions [3] Extensions OPTIONAL } > >>> > >>>and then something like: > >>> > >>> ParamsDefValPolicy :: = SEQUENCE { > >>> trustAnchors TrustAnchors, > >>> endEntityCertificationPolicy SEQUENCE OF OBJECT IDENTIFIER > >>>OPTIONAL, > >>> inhibitPolMap BOOLEAN DEFAULT FALSE, > >>> caCertificationPolicies SEQUENCE OF OBJECT IDENTIFIER > >>>OPTIONAL } > >>> > >>>Denis > >>> > >>> > >>> > >>> > >>>>I am not sure whether I am the only person unable to construct a > >>> > >>>parser. > >>> > >>> > >>> > >>>>in version 14 an aditional flag has been added which is not > available > >>> > >>>in the > >>> > >>> > >>> > >>>>Chapter 9. Is the isCA flag an orthogonal attribut to other > >>> > > validation > > > >>>>algorithms, or just to some of them, e.g,. the default > name matching > >>> > >>>one? > >>> > >>> > >>> > >>>> Query ::= SEQUENCE { > >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF > CertReference, > >>> > >>> > >>>> checks CertChecks, > >>>> wantBack WantBack, > >>>> requestRefHash BOOLEAN DEFAULT TRUE, > >>>> includePolResponce BOOLEAN DEFAULT FALSE, > >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, > >>>> requireExplicitPol BOOLEAN DEFAULT FALSE, > >>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, > >>>> isCA BOOLEAN DEFAULT FALSE, > >>>> valAlgorithm ValidationAlgorithm, > >>>> serverContextInfo [0] OCTET STRING OPTIONAL, > >>>> validityTime [1] GeneralizedTime OPTIONAL, > >>>> trustAnchors [2] TrustAnchors OPTIONAL, > >>>> intermediateCerts [3] CertBundle OPTIONAL, > >>>> revInfos [4] RevocationInfos OPTIONAL, > >>>> keyUsage [5] KeyUsage OPTIONAL, > >>>> extendedKeyUsage [6] ExtKeyUsageSyntax OPTIONAL, > >>>> queryExtensions [7] Extensions OPTIONAL > >>>> producedAt [8] GeneralizedTime OPTIONAL} > >>>> > >>>> > >>> > >>> > >>> > >>> > >> > >> > > > > > > > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46NAt0o029588; Thu, 6 May 2004 16:10:55 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46NAtrR029586; Thu, 6 May 2004 16:10:55 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46NAsxc029564; Thu, 6 May 2004 16:10:54 -0700 (PDT) (envelope-from kent@bbn.com) Received: from [128.89.89.75] (dhcp89-089-075.bbn.com [128.89.89.75]) by aragorn.bbn.com (8.12.7/8.12.7) with ESMTP id i46NAi7X002262; Thu, 6 May 2004 19:10:44 -0400 (EDT) Mime-Version: 1.0 X-Sender: kent@po2.bbn.com Message-Id: <p0610051abcc07503df04@[128.89.89.75]> In-Reply-To: <409AB4BC.4030606@sun.com> References: <p0610050bbcc0378b76f1@[128.89.89.75]> <409AB4BC.4030606@sun.com> Date: Thu, 6 May 2004 19:05:53 -0400 To: Steve Hanna <Steve.Hanna@Sun.COM> From: Stephen Kent <kent@bbn.com> Subject: Re: RFC 3280 bug in name constraints Cc: Stefan Santesson <stefans@microsoft.com>, Sharon Boeyen <sharon.boeyen@entrust.com>, ietf-pkix@imc.org, ietf-smime@imc.org Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Steve, Thanks for the thoughtful analysis. I think you make a good argument for why it can be dangerous for one to apply an rfc822name constraint to a sub alt name but not an e-mail attribute in a subject name, and how the combination might arise. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46M1Lgd023367; Thu, 6 May 2004 15:01:21 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46M1L3X023365; Thu, 6 May 2004 15:01:21 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from brmea-mail-4.sun.com (brmea-mail-4.Sun.COM [192.18.98.36]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46M1J5g023347 for <ietf-pkix@imc.org>; Thu, 6 May 2004 15:01:20 -0700 (PDT) (envelope-from Steve.Hanna@Sun.COM) Received: from phys-bur1-1 ([129.148.13.15]) by brmea-mail-4.sun.com (8.12.10/8.12.9) with ESMTP id i46M00sX022625 for <ietf-pkix@imc.org>; Thu, 6 May 2004 16:00:00 -0600 (MDT) Received: from conversion-daemon.bur-mail2.east.sun.com by bur-mail2.east.sun.com (iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003)) id <0HXB00901BR2JO@bur-mail2.east.sun.com> for ietf-pkix@imc.org; Thu, 06 May 2004 18:01:17 -0400 (EDT) Received: from sun.com (dhcp-ubur02-75-154.East.Sun.COM [129.148.75.154]) by bur-mail2.east.sun.com (iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003)) with ESMTPA id <0HXB00863BU5EW@bur-mail2.east.sun.com>; Thu, 06 May 2004 18:01:17 -0400 (EDT) Date: Thu, 06 May 2004 17:57:16 -0400 From: Steve Hanna <Steve.Hanna@Sun.COM> Subject: Re: RFC 3280 bug in name constraints In-reply-to: <p0610050bbcc0378b76f1@[128.89.89.75]> To: Stephen Kent <kent@bbn.com> Cc: Stefan Santesson <stefans@microsoft.com>, Sharon Boeyen <sharon.boeyen@entrust.com>, ietf-pkix@imc.org, ietf-smime@imc.org Message-id: <409AB4BC.4030606@sun.com> MIME-version: 1.0 Content-type: multipart/signed; protocol="application/x-pkcs7-signature"; micalg=sha1; boundary=------------ms000301050500030008070709 X-Accept-Language: en-us, en User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20031111 References: <p0610050bbcc0378b76f1@[128.89.89.75]> Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> --------------ms000301050500030008070709 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Here's a practical example where this problem might arise. Assume that two companies, A and B, have cross-certified (maybe through a bridge CA, maybe directly). The cross certificate from A to B includes a name constraint of type rfc822Name with permittedSubtrees of b.com. The CA for B (or perhaps a rebellious or poorly managed subsidiary CA) issues a certificate to an individual with an EmailAddress attribute with the value "prez@a.com" in the subject DN and a subject alt name of type rfc822Name with the value "peon@b.com". Under the current text in RFC 3280 and under Stefan's proposed revised text, a relying party whose trust anchor is A will consider this certificate perfectly valid. Unless their email software is smart enough to know that an EmailAddress attribute in the subject DN is not checked for name constraints when there's also a subject alt name and therefore the EmailAddress attribute must be ignored, the email software will be glad to verify email claiming to be from prez@a.com if the signature can be verified with this certificate. Why is this a problem? Because the name constraints extension in the cross certificate was supposed to prevent relying parties in A from accepting certificates from B when an email address outside of b.com as the subject DN or subject alt name. I have cc'd this email to the smime WG. Maybe they will tell me not to worry. Email clients are smart enough to know that if a valid certificate contains a subject alt name, then the EmailAddress attribute in the subject DN cannot be trusted. But I don't think email clients do this check. I certainly can't find any language in RFC 2632 and its successor warning people to do this. Either RFC 3280 should consider the cert described above as invalid or RFC 2632 should warn email developers to ignore an EmailAddress attribute in the subject DN if the certificate contains a subject alt name. Otherwise, I think we have a security hole. Given the choice between these two options, I'd rather change RFC 3280. Why? First, I don't think that cert should be valid. Second, it's usually easier to change the path validation code and get that deployed (via an OS patch or equivalent) than to change the email client code. Thanks, Steve Stephen Kent wrote: > At 5:16 PM +0100 5/6/04, Stefan Santesson wrote: > >> I agree with Sharon here. >> >> Even though it may sound logical to always also constrain e-mail >> attributes, changing the standard in this aspect will just break many >> current implementations and I don't think there is enough motivation for >> it. >> >> I too just see it necessary if no rfc822Name is present. >> >> /Stefan >> > > This is a slightly murky area. The e-mail address attribute is one that > we deprecate in 3280. It is a non-standard way of representing this > data; it was widely used in v1 certs, but ought to be scrapped for v3 > certs. however, is it likely that a CA who goes to the trouble to > include the name constraints extension, and who explicitly includes an > rfc822name as an alt name, would also have the email address DN > attribute as well? I just wonder if this potential collision is at all > likely? > > > Steve --------------ms000301050500030008070709 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJNzCC AvYwggJfoAMCAQICAww3ZDANBgkqhkiG9w0BAQQFADBiMQswCQYDVQQGEwJaQTElMCMGA1UE ChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNv bmFsIEZyZWVtYWlsIElzc3VpbmcgQ0EwHhcNMDQwNDI4MjEwMTE0WhcNMDUwNDI4MjEwMTE0 WjBoMQ4wDAYDVQQEEwVIYW5uYTEVMBMGA1UEKhMMU3RlcGhlbiBSb3NzMRswGQYDVQQDExJT dGVwaGVuIFJvc3MgSGFubmExIjAgBgkqhkiG9w0BCQEWE3N0ZXZlLmhhbm5hQHN1bi5jb20w ggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC11rLYp70KnKGmV6pNTFRhWBwq47mV 4hU6EaZxo6I3Vw96Rd3E4TH3SAthIl5N4/tpiMPtlTBiJOU6QHSlz1DgTCclmTK6taqtEyLq amez8vYgSdOEnOFRYILptjbEwpE+yeUmFKOmZ7zSwBMqRQatMHMaLVUJJ3Ygw1V7fKSwvVU0 OJ8c+c3sJRtRyDLxq/vMzYQP+wyqRqlJPTHGFqVuMp0FZIV5X76z+O+W90m9jXjirz91Jz+W N6rwqmPi8Qy0uxBKm6NKT1371p+WYaK9OfGk8vGbJ8ZsTP3DyQhuuemYPbkKPtoII6WnHIGT /DOy2G7gff7GOIhUrs7PziddAgMBAAGjMDAuMB4GA1UdEQQXMBWBE3N0ZXZlLmhhbm5hQHN1 bi5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQB++aOP6sZVYlqBGc92hZ61 FHST+OqnKZHRGkqEO94RoxdeaSUFKVO2IaKLMBmqNzmh3N3pIsIKiLhJNvDgLhlRrMRVfR29 uixjTsprR26AacMx/gCiCstGIjkzTb83xAvZay5hnzUmlt/LSNX8e6jTNav6LP64WC+C7Tul fM/PzDCCAvYwggJfoAMCAQICAww3ZDANBgkqhkiG9w0BAQQFADBiMQswCQYDVQQGEwJaQTEl MCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3Rl IFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0EwHhcNMDQwNDI4MjEwMTE0WhcNMDUwNDI4 MjEwMTE0WjBoMQ4wDAYDVQQEEwVIYW5uYTEVMBMGA1UEKhMMU3RlcGhlbiBSb3NzMRswGQYD VQQDExJTdGVwaGVuIFJvc3MgSGFubmExIjAgBgkqhkiG9w0BCQEWE3N0ZXZlLmhhbm5hQHN1 bi5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC11rLYp70KnKGmV6pNTFRh WBwq47mV4hU6EaZxo6I3Vw96Rd3E4TH3SAthIl5N4/tpiMPtlTBiJOU6QHSlz1DgTCclmTK6 taqtEyLqamez8vYgSdOEnOFRYILptjbEwpE+yeUmFKOmZ7zSwBMqRQatMHMaLVUJJ3Ygw1V7 fKSwvVU0OJ8c+c3sJRtRyDLxq/vMzYQP+wyqRqlJPTHGFqVuMp0FZIV5X76z+O+W90m9jXji rz91Jz+WN6rwqmPi8Qy0uxBKm6NKT1371p+WYaK9OfGk8vGbJ8ZsTP3DyQhuuemYPbkKPtoI I6WnHIGT/DOy2G7gff7GOIhUrs7PziddAgMBAAGjMDAuMB4GA1UdEQQXMBWBE3N0ZXZlLmhh bm5hQHN1bi5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQB++aOP6sZVYlqB Gc92hZ61FHST+OqnKZHRGkqEO94RoxdeaSUFKVO2IaKLMBmqNzmh3N3pIsIKiLhJNvDgLhlR rMRVfR29uixjTsprR26AacMx/gCiCstGIjkzTb83xAvZay5hnzUmlt/LSNX8e6jTNav6LP64 WC+C7TulfM/PzDCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAwgdExCzAJBgNVBAYT AlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUgVG93bjEaMBgGA1UE ChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRpb24gU2VydmljZXMg RGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBDQTErMCkGCSqG SIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0wMzA3MTcwMDAwMDBa Fw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3Vs dGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNz dWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV+065yplaHmjAdQRw nd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfArhVqqP3FWy688Cwfn 8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/p7bRPGEEQB5kGXJg t/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8MDowOKA2oDSGMmh0 dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWlsQ0EuY3JsMAsGA1Ud DwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxhYmVsMi0xMzgwDQYJ KoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/TCG4+DYfqi2fNi/A 9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amcOY6MIE9lX5Xa9/eH 1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggM7MIIDNwIBATBpMGIxCzAJ BgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYD VQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIDDDdkMAkGBSsOAwIa BQCgggGnMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcNAQkFMQ8XDTA0MDUw NjIxNTcxNlowIwYJKoZIhvcNAQkEMRYEFLEY3mT3MOXY0xulNnJoOl38MmtLMFIGCSqGSIb3 DQEJDzFFMEMwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcG BSsOAwIHMA0GCCqGSIb3DQMCAgEoMHgGCSsGAQQBgjcQBDFrMGkwYjELMAkGA1UEBhMCWkEx JTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0 ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBAgMMN2QwegYLKoZIhvcNAQkQAgsxa6Bp MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQu MSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIDDDdkMA0G CSqGSIb3DQEBAQUABIIBAJwSXRZvvuOVpdMezj+27FY1npRm7LAGKCgtQfPo1lmoKMyn6plY X8W88OH/jP+8i2Taw1xdLzx+Qch+7Sj/GCp6C9rBCsAzURRrd7jvqbTndYsMssr6qzbNvVkh XLt7dj/h0FakFodszZYsW5kuvSRY/6cgPBNRLz1iVxprTndcpTgnWYdS5u7FBTjulI4yvX2D kEXTN44oIM5rTLTM0Au8lpZWjApl8l5lDjV4amzufyR4DbmzR5ut9paYWN2cyx7fbL0NJ/13 +rW9re6Yam+yB71Tof3x2T6mop7Q8zRgCWvnQYiinelIkzMucjOsz5VuAhB2Nln4SmI2JAr2 jIIAAAAAAAA= --------------ms000301050500030008070709-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46KSxZJ015572; Thu, 6 May 2004 13:28:59 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46KSxXR015571; Thu, 6 May 2004 13:28:59 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46KSwDo015559 for <ietf-pkix@imc.org>; Thu, 6 May 2004 13:28:58 -0700 (PDT) (envelope-from kent@bbn.com) Received: from [128.89.89.75] (dhcp89-089-075.bbn.com [128.89.89.75]) by aragorn.bbn.com (8.12.7/8.12.7) with ESMTP id i46KSC7X025140; Thu, 6 May 2004 16:28:13 -0400 (EDT) Mime-Version: 1.0 X-Sender: kent@po2.bbn.com Message-Id: <p0610050cbcc04e1ac079@[128.89.89.75]> In-Reply-To: <33E7A1613A3480448996047B69308A2F027EE0C9@df-grommit-01.dogfood> References: <33E7A1613A3480448996047B69308A2F027EE0C9@df-grommit-01.dogfood> Date: Thu, 6 May 2004 16:23:24 -0400 To: "Trevor Freeman" <trevorf@exchange.microsoft.com> From: Stephen Kent <kent@bbn.com> Subject: RE: FW: scvp Cc: "Peter Sylvester" <Peter.Sylvester@edelweb.fr>, <xDenis.Pinkas@bull.net>, <ietf-pkix@imc.org> Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> At 11:33 AM -0700 5/6/04, Trevor Freeman wrote: >If a client really trusts the server like in the enterprise, then the >client can take the default for everything. That seems unlikely to me, not because of trust issues, but because the validation policy may (should?) be application specific and thus the client may need to assert a per-application policy to ensure that the validation process is appropriate to the application. > What I am suggesting is that >there could be a long or short had way of expressing the policy and a >client can choose either. Given we have a well know list of parameters, >we can pass them directly or indirectly via a hash. Explicit passing is >more robust since it is unambiguous and requires no configuration on the >part of the server. The client also has a fallback in the event he >server does not recognize the shorthand. I do not buy the argument that >using a shorthand is easer to manage. You MUST at some point define the >policy in terms of the validation input parameters and that definition >can just as easily be consumed by the client as the server. So all I see >is the shorthand saves bandwidth - which is a minimal gain. The definition of the parameters is done via a separate protocol, and is a management activity that the client may never have to employ. So, on that regard, I disagree that the client needs to "consume" these parameters. I would rather see a set of parameters referenced via an OID, than via a hash of the parameters, to keep things simpler. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46KSM5U015533; Thu, 6 May 2004 13:28:22 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46KSM9g015532; Thu, 6 May 2004 13:28:22 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46KSLaa015526 for <ietf-pkix@imc.org>; Thu, 6 May 2004 13:28:21 -0700 (PDT) (envelope-from kent@bbn.com) Received: from [128.89.89.75] (dhcp89-089-075.bbn.com [128.89.89.75]) by aragorn.bbn.com (8.12.7/8.12.7) with ESMTP id i46KSC7Z025140; Thu, 6 May 2004 16:28:14 -0400 (EDT) Mime-Version: 1.0 X-Sender: kent@po2.bbn.com Message-Id: <p0610050dbcc04efaf4ce@[128.89.89.75]> In-Reply-To: <33E7A1613A3480448996047B69308A2F027EE0FB@df-grommit-01.dogfood> References: <33E7A1613A3480448996047B69308A2F027EE0FB@df-grommit-01.dogfood> Date: Thu, 6 May 2004 16:27:58 -0400 To: "Trevor Freeman" <trevorf@exchange.microsoft.com> From: Stephen Kent <kent@bbn.com> Subject: RE: SCVP & validation policies Cc: <ietf-pkix@imc.org> Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> At 12:10 PM -0700 5/6/04, Trevor Freeman wrote: >Hi Steve, >If you want to have a shorthand notation for a validation policy, given >we have a finite set of parameters, the I don't see the value of >indirection via an OID. Using a hash of the set of inputs provides a way >of expressing the same thing without introducing ambiguity of does both >parties have a common understating of that this means. At some point >someone has to define the policy in terms of the 3280 input parameters. >That definition is just as applicable and consumable by the client as a >server therefore what did we gain from the shorthand notation? It would >be pretty trivial to define some XML which would allow that set of >validation parameters to be imported by any SCVP entity. >Trevor Trevor, It is not indirection via an OID, it is identification via an OID. Clients never have to be aware of the parameters if we use an OID, and simplification of clients is one of the goals of SCVP. So I cannot agree with your comment that clients have to be able to consume these parameters. Given that as a basis, it seems that we are debating whether to use OIDs or to create some other form of identifier via hashing parameters. I don't see the latter as attractive, since it requires additional specification of the canonicalization and encoding of the parameters to ensure consistency in hash computation between client and server. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46KH23c015022; Thu, 6 May 2004 13:17:02 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46KH2YG015021; Thu, 6 May 2004 13:17:02 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46KH1Xa014993 for <ietf-pkix@imc.org>; Thu, 6 May 2004 13:17:01 -0700 (PDT) (envelope-from kent@bbn.com) Received: from [128.89.89.75] (dhcp89-089-075.bbn.com [128.89.89.75]) by aragorn.bbn.com (8.12.7/8.12.7) with ESMTP id i46KGo7X024540; Thu, 6 May 2004 16:16:51 -0400 (EDT) Mime-Version: 1.0 X-Sender: kent@po2.bbn.com Message-Id: <p0610050bbcc0378b76f1@[128.89.89.75]> In-Reply-To: <0C3042E92D8A714783E2C44AB9936E1DF824B5@EUR-MSG-03.europe.corp.microsoft.c om> References: <0C3042E92D8A714783E2C44AB9936E1DF824B5@EUR-MSG-03.europe.corp.microsoft.c om> Date: Thu, 6 May 2004 16:16:35 -0400 To: "Stefan Santesson" <stefans@microsoft.com> From: Stephen Kent <kent@bbn.com> Subject: RE: RFC 3280 bug in name constraints Cc: "Sharon Boeyen" <sharon.boeyen@entrust.com>, "Steve Hanna" <Steve.Hanna@Sun.COM>, <ietf-pkix@imc.org> Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> At 5:16 PM +0100 5/6/04, Stefan Santesson wrote: >I agree with Sharon here. > >Even though it may sound logical to always also constrain e-mail >attributes, changing the standard in this aspect will just break many >current implementations and I don't think there is enough motivation for >it. > >I too just see it necessary if no rfc822Name is present. > >/Stefan > This is a slightly murky area. The e-mail address attribute is one that we deprecate in 3280. It is a non-standard way of representing this data; it was widely used in v1 certs, but ought to be scrapped for v3 certs. however, is it likely that a CA who goes to the trouble to include the name constraints extension, and who explicitly includes an rfc822name as an alt name, would also have the email address DN attribute as well? I just wonder if this potential collision is at all likely? Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46JAwAr010489; Thu, 6 May 2004 12:10:58 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46JAwUS010488; Thu, 6 May 2004 12:10:58 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46JAvHU010480 for <ietf-pkix@imc.org>; Thu, 6 May 2004 12:10:57 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Thu, 6 May 2004 12:10:57 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 06 May 2004 12:10:55 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 12:10:57 -0700 x-mimeole: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: SCVP & validation policies Date: Thu, 6 May 2004 12:10:59 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EE0FB@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCVP & validation policies thread-index: AcQzhbOY+EdcC6TTRY2dc8NHmQ0MfgAFmqFg From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Stephen Kent" <kent@bbn.com> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 06 May 2004 19:10:57.0083 (UTC) FILETIME=[DC3920B0:01C4339D] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i46JAvHU010481 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Steve, If you want to have a shorthand notation for a validation policy, given we have a finite set of parameters, the I don't see the value of indirection via an OID. Using a hash of the set of inputs provides a way of expressing the same thing without introducing ambiguity of does both parties have a common understating of that this means. At some point someone has to define the policy in terms of the 3280 input parameters. That definition is just as applicable and consumable by the client as a server therefore what did we gain from the shorthand notation? It would be pretty trivial to define some XML which would allow that set of validation parameters to be imported by any SCVP entity. Trevor -----Original Message----- From: Stephen Kent [mailto:kent@bbn.com] Sent: Thursday, May 06, 2004 7:45 AM To: Trevor Freeman Cc: ietf-pkix@imc.org Subject: SCVP & validation policies Trevor, I have to agree with Denis, in that it was my impression that the client would pass an OID to the server to express the validation policy to be used. The servers will, in many cases, be local to an enterprise, and thus it is very feasible to use a single OID to reference a policy that bundles all the parameters together. In fact, I am not even sure that a public SCVP server is better off with an ala carte approach to policy specification, but I might be able to be convinced. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46IgV2w007246; Thu, 6 May 2004 11:42:31 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46IgV8B007245; Thu, 6 May 2004 11:42:31 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from sottmxssm1.entrust.com (sottmxssm1.entrust.com [216.191.252.11]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46IgUDG007237 for <ietf-pkix@imc.org>; Thu, 6 May 2004 11:42:30 -0700 (PDT) (envelope-from sharon.boeyen@entrust.com) Received: from SOTTMXS01.entrust.com ([10.4.61.7]) by sottmxssm1.entrust.com (Switch-3.1.4/Switch-3.1.0) with ESMTP id i46Ipgwu027642; Thu, 6 May 2004 14:51:42 -0400 Received: by sottmxs01.entrust.com with Internet Mail Service (5.5.2657.72) id <KM64Z9Q0>; Thu, 6 May 2004 14:42:24 -0400 Message-ID: <7A3E1242FA9989439AD1F9B2D71C287F0AD978@sottmxs05.entrust.com> From: Sharon Boeyen <sharon.boeyen@entrust.com> To: "'Steve Hanna'" <Steve.Hanna@Sun.COM>, Stefan Santesson <stefans@microsoft.com> Cc: ietf-pkix@imc.org Subject: RE: RFC 3280 bug in name constraints Date: Thu, 6 May 2004 14:42:25 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: text/plain Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Apparently some could not read the message I sent earlier, so here is a resend. I prefer to leave 3280 as is (with Stefan's proposed change which I do support). The special case for checking email address in subject field is, from my understanding, for backward compatibility reasons only for situations where the email address didn't appear in the extension. The subject field is a DN and should only be required to be checked if there are DN name constraints (except for this backward compatibility issue where the exception is made). -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Steve Hanna Sent: Thursday, May 06, 2004 10:35 AM To: Stefan Santesson Cc: ietf-pkix@imc.org Subject: Re: RFC 3280 bug in name constraints I agree. In fact, I don't see why you wouldn't apply name constraints to an EmailAddress attribute in a subject DN even if there *is* an rfc822Name in a subject alternative name. I suggest that the text be revised to say: An rfc822 name constraint MUST be applied to an attribute of type EmailAddress in the subject distinguished name. However, I'm willing to back down on this if people feel it's too strict. My concern is that a certificate with an email address in the subject DN prohibited by name constraints will be allowed because it also contains a permissible email address in the subject alternative name. I suspect many applications will not understand this and will accept the email address in the subject DN. In fact, RFC 2632 and its successor don't contain any warning about this. Thanks, Steve Stefan Santesson wrote: > If someone is keeping record of bugs for future update of RFC 3280 I > have a small one to file about name constraints. > > > > In section 4.2.1.11, starting at bottom of page 38: > > > > When rfc822 names are constrained, but the > > certificate does not include a subject alternative name, the rfc822 > > name constraint MUST be applied to the attribute of type > EmailAddress > > in the subject distinguished name. > > > > Suppose/suggest that the intended meaning is: > > > > When rfc822 names are constrained, but the certificate does not > > include an rfc822Name in a subject alternative name extension, the > rfc822 > > name constraint MUST be applied to the attribute of type > EmailAddress > > in the subject distinguished name. > > > > > > Rationale: > > Why would the constraint NOT apply to EmailAddress just because there > is > a SubjAltName extension with e.g. just some unrelated other name > component present? > > > > /Stefan > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46IXMZ2006109; Thu, 6 May 2004 11:33:22 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46IXMci006108; Thu, 6 May 2004 11:33:22 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46IXLGa006100 for <ietf-pkix@imc.org>; Thu, 6 May 2004 11:33:21 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Thu, 6 May 2004 11:33:18 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 06 May 2004 11:33:17 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 11:33:19 -0700 x-mimeole: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: FW: scvp Date: Thu, 6 May 2004 11:33:18 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EE0C9@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: FW: scvp thread-index: AcQzlIkgUYZZST4zTy6tCK91qcQOXQAAbDZQ From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Peter Sylvester" <Peter.Sylvester@edelweb.fr>, <xDenis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 06 May 2004 18:33:19.0868 (UTC) FILETIME=[9AD173C0:01C43398] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i46IXLGa006102 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> If a client really trusts the server like in the enterprise, then the client can take the default for everything. What I am suggesting is that there could be a long or short had way of expressing the policy and a client can choose either. Given we have a well know list of parameters, we can pass them directly or indirectly via a hash. Explicit passing is more robust since it is unambiguous and requires no configuration on the part of the server. The client also has a fallback in the event he server does not recognize the shorthand. I do not buy the argument that using a shorthand is easer to manage. You MUST at some point define the policy in terms of the validation input parameters and that definition can just as easily be consumed by the client as the server. So all I see is the shorthand saves bandwidth - which is a minimal gain. Trevor -----Original Message----- From: Peter Sylvester [mailto:Peter.Sylvester@edelweb.fr] Sent: Thursday, May 06, 2004 11:04 AM To: Peter.Sylvester@edelweb.fr; xDenis.Pinkas@bull.net; Trevor Freeman Cc: ietf-pkix@imc.org Subject: RE: FW: scvp > > If we want to go down the route of providing a shorthand notation for a > validation policy, using an OID just opens up ambiguity. We are assuming > both client and server have a common understating of both - which add > additional management overhead. A dumb client may have no understanding of a policy, it just wan't some information concerning the usage of a cert in its particular context. In an extreme case, just a yes/no answer is good, and the protocol should allow to have such a minimal exchange. You are describing a way that a client MUST always provide actual parameters. This seems to me unnecessary overhead at the client. If the actual policy parameters are required to be known by the client, the server can return then in the response, so that they can be logged by the client if necessary, or transferred to another relying party, etc. We assume that the application of the policy is under the responsibility of the server, it seems rather natural to me that in case that the response could be questioned in the future, the server indicated what it had done. or, the informations gets transferred only once. > It seems like yesterday we had the > discussion on whether a certificate policy should also have a hash so > the RP could tell if the policy had changed. If we truly want a > shorthand notation to a policy then using a hash over the set of input > parameters provides an unambiguous statement of what is required. It is absolutely possible that the policy details at the server will change without a need for a client to get informed, at least in an enterprise internal environment. It is at the server side where all decisions are made for dumb clients. If the client has a single application context, then a request should be possible to contain just the reference to the cert or the cert, and in an extreme, even the policy could be derived at the server if the client can be identified and authenticated, in this case the client configuration is minimal. Peter Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46I49H5003206; Thu, 6 May 2004 11:04:09 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46I490A003205; Thu, 6 May 2004 11:04:09 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46I47hx003197 for <ietf-pkix@imc.org>; Thu, 6 May 2004 11:04:08 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i46I3wN15897; Thu, 6 May 2004 20:03:58 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Thu, 6 May 2004 20:03:58 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i46I3r426566; Thu, 6 May 2004 20:03:53 +0200 (MEST) Date: Thu, 6 May 2004 20:03:53 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405061803.i46I3r426566@chandon.edelweb.fr> To: Peter.Sylvester@edelweb.fr, xDenis.Pinkas@bull.net, trevorf@exchange.microsoft.com Subject: RE: FW: scvp Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> > > If we want to go down the route of providing a shorthand notation for a > validation policy, using an OID just opens up ambiguity. We are assuming > both client and server have a common understating of both - which add > additional management overhead. A dumb client may have no understanding of a policy, it just wan't some information concerning the usage of a cert in its particular context. In an extreme case, just a yes/no answer is good, and the protocol should allow to have such a minimal exchange. You are describing a way that a client MUST always provide actual parameters. This seems to me unnecessary overhead at the client. If the actual policy parameters are required to be known by the client, the server can return then in the response, so that they can be logged by the client if necessary, or transferred to another relying party, etc. We assume that the application of the policy is under the responsibility of the server, it seems rather natural to me that in case that the response could be questioned in the future, the server indicated what it had done. or, the informations gets transferred only once. > It seems like yesterday we had the > discussion on whether a certificate policy should also have a hash so > the RP could tell if the policy had changed. If we truly want a > shorthand notation to a policy then using a hash over the set of input > parameters provides an unambiguous statement of what is required. It is absolutely possible that the policy details at the server will change without a need for a client to get informed, at least in an enterprise internal environment. It is at the server side where all decisions are made for dumb clients. If the client has a single application context, then a request should be possible to contain just the reference to the cert or the cert, and in an extreme, even the policy could be derived at the server if the client can be identified and authenticated, in this case the client configuration is minimal. Peter Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46HMdWB086521; Thu, 6 May 2004 10:22:39 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46HMdjT086520; Thu, 6 May 2004 10:22:39 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46HMcNT086511 for <ietf-pkix@imc.org>; Thu, 6 May 2004 10:22:38 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Thu, 6 May 2004 10:22:39 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 06 May 2004 10:22:38 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 10:22:38 -0700 x-mimeole: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: FW: scvp Date: Thu, 6 May 2004 10:22:37 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F027EE06E@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: FW: scvp thread-index: AcQzhrht+5OGU+yKT9u+eO3Qp0M+9AABu+ww From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Peter Sylvester" <Peter.Sylvester@edelweb.fr>, <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 06 May 2004 17:22:38.0732 (UTC) FILETIME=[BAE774C0:01C4338E] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i46HMcNT086515 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> If we want to go down the route of providing a shorthand notation for a validation policy, using an OID just opens up ambiguity. We are assuming both client and server have a common understating of both - which add additional management overhead. It seems like yesterday we had the discussion on whether a certificate policy should also have a hash so the RP could tell if the policy had changed. If we truly want a shorthand notation to a policy then using a hash over the set of input parameters provides an unambiguous statement of what is required. Trevor -----Original Message----- From: Peter Sylvester [mailto:Peter.Sylvester@edelweb.fr] Sent: Thursday, May 06, 2004 9:25 AM To: Trevor Freeman; Denis.Pinkas@bull.net Cc: ietf-pkix@imc.org Subject: Re: FW: scvp I agree with Denis about a single value policy. I think that we have already a structure elsewhere that is close to what can be used. PolicyInformation ::= SEQUENCE { policyIdentifier CertPolicyId, policyQualifiers SEQUENCE SIZE (1..MAX) OF PolicyQualifierInfo OPTIONAL } CertPolicyId ::= OBJECT IDENTIFIER PolicyQualifierInfo ::= SEQUENCE { policyQualifierId PolicyQualifierId, qualifier ANY DEFINED BY policyQualifierId } the PolicyQualifierId would be an indidation of a particular algorithm that needs to be performed and its parameters. one would be "3280 path processing input" another would be 'end entry naming comparison', etc. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46FkniR078911; Thu, 6 May 2004 08:46:49 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46Fkndd078910; Thu, 6 May 2004 08:46:49 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from smtp-ft3.fr.colt.net (smtp-ft3.fr.colt.net [213.41.78.206]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46Fkmtx078894 for <ietf-pkix@imc.org>; Thu, 6 May 2004 08:46:48 -0700 (PDT) (envelope-from jmdesp@free.fr) Received: from free.fr (host.106.92.68.195.rev.coltfrance.com [195.68.92.106]) by smtp-ft3.fr.colt.net with ESMTP id i46Fkk131204 for <ietf-pkix@imc.org>; Thu, 6 May 2004 17:46:46 +0200 Message-ID: <409A5E21.9010905@free.fr> Date: Thu, 06 May 2004 17:47:45 +0200 From: Jean-Marc Desperrier <jmdesp@free.fr> User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:) Gecko/20040226 X-Accept-Language: en-us, en MIME-Version: 1.0 To: ietf-pkix@imc.org Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs References: <0C3042E92D8A714783E2C44AB9936E1DF823DD@EUR-MSG-03.europe.corp.microsoft.com> <409A4ACB.2040709@sun.com> In-Reply-To: <409A4ACB.2040709@sun.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Steve Hanna wrote: > A better solution is to require (or at least recommend) > the ability to properly compare PrintableString and > UTF8String encodings. This still leave to the CA the task of enforcing the user certs only use PrintableString or UTF8String. Either by refusing requests with other encodings or converting them. Given the current text of RFC3280 this is probably the best choice. But an application like openssl is currently shipped with the default of authorizing either of PrintableString and T61String and nothing else. If some people do emit T61String (it does happen at least on a small scale), they'll need to be able to compare to UTF8String. And this does require actual convertion whereas equality comparison of PrintableString and UTF8String could be done by just ignoring the asn1 type tag. The above reminds me how important it is that implementers correctly reject invalid UTF-8 sequences, and that it's not garanteed to happen. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46Fc1gB078003; Thu, 6 May 2004 08:38:01 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46Fc1rZ078002; Thu, 6 May 2004 08:38:01 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from sottmxssm1.entrust.com (sottmxssm1.entrust.com [216.191.252.11]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46Fc0iu077992 for <ietf-pkix@imc.org>; Thu, 6 May 2004 08:38:00 -0700 (PDT) (envelope-from sharon.boeyen@entrust.com) Received: from SOTTMXS01.entrust.com ([10.4.61.7]) by sottmxssm1.entrust.com (Switch-3.1.4/Switch-3.1.0) with ESMTP id i46FlE0K031076; Thu, 6 May 2004 11:47:14 -0400 Received: by sottmxs01.entrust.com with Internet Mail Service (5.5.2657.72) id <KM64Z3JN>; Thu, 6 May 2004 11:37:57 -0400 Message-ID: <7A3E1242FA9989439AD1F9B2D71C287F0AD970@sottmxs05.entrust.com> From: Sharon Boeyen <sharon.boeyen@entrust.com> To: "'Steve Hanna'" <Steve.Hanna@Sun.COM>, Stefan Santesson <stefans@microsoft.com> Cc: ietf-pkix@imc.org Subject: RE: RFC 3280 bug in name constraints Date: Thu, 6 May 2004 11:37:57 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: application/x-pkcs7-mime; name="smime.p7m" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7m" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> MIIrqwYJKoZIhvcNAQcCoIIrnDCCK5gCAQExCzAJBgUrDgMCGgUAMIId+QYJKoZIhvcNAQcBoIId 6gSCHeZDb250ZW50LVR5cGU6IG11bHRpcGFydC9hbHRlcm5hdGl2ZTsNCglib3VuZGFyeT0iLS09 X05leHRQYXJ0X1NUSl8zODg0NTc3Ni4xMTM5MTk5ODUiDQoNCg0KLS0tLT1fTmV4dFBhcnRfU1RK XzM4ODQ1Nzc2LjExMzkxOTk4NQ0KQ29udGVudC1UeXBlOiB0ZXh0L3BsYWluOw0KCWNoYXJzZXQ9 IlVTLUFTQ0lJIg0KQ29udGVudC1UcmFuc2Zlci1FbmNvZGluZzogcXVvdGVkLXByaW50YWJsZQ0K DQpJIHByZWZlciB0byBsZWF2ZSAzMjgwIGFzIGlzICh3aXRoIFN0ZWZhbidzIHByb3Bvc2VkIGNo YW5nZSB3aGljaCBJIGRvIHN1cHBvcnQpLiBUaGUgc3BlYz0NCmlhbCBjYXNlIGZvciBjaGVja2lu ZyBlbWFpbCBhZGRyZXNzIGluIHN1YmplY3QgZmllbGQgaXMsIGZyb20gbXkgdW5kZXJzdGFuZGlu ZywgZm9yIGJhYz0NCmt3YXJkIGNvbXBhdGliaWxpdHkgcmVhc29ucyBvbmx5IGZvciBzaXR1YXRp b25zIHdoZXJlIHRoZSBlbWFpbCBhZGRyZXNzIGRpZG4ndCBhcHBlYT0NCnIgaW4gdGhlIGV4dGVu c2lvbi4gVGhlIHN1YmplY3QgZmllbGQgaXMgYSBETiBhbmQgc2hvdWxkIG9ubHkgYmUgcmVxdWly ZWQgdG8gYmUgY2hlY2tlZCBpZiB0PQ0KaGVyZSBhcmUgRE4gbmFtZSBjb25zdHJhaW50cyAoZXhj ZXB0IGZvciB0aGlzIGJhY2t3YXJkIGNvbXBhdGliaWxpdHkgaXNzdWUgd2hlcmUgdGhlIGU9DQp4 Y2VwdGlvbiBpcyBtYWRlKS49MjANCg0KLS0tLS1PcmlnaW5hbCBNZXNzYWdlLS0tLS0NCkZyb206 IG93bmVyLWlldGYtcGtpeD00MG1haWwuaW1jLm9yZyA9NUJtYWlsdG86b3duZXItaWV0Zi1wa2l4 PTQwbWFpbC5pbWMubz0NCnJnPTVEIE9uIEJlaGFsZiBPZiBTdGV2ZSBIYW5uYQ0KU2VudDogVGh1 cnNkYXksIE1heSAwNiwgMjAwNCAxMDozNSBBTQ0KVG86IFN0ZWZhbiBTYW50ZXNzb24NCkNjOiBp ZXRmLXBraXg9NDBpbWMub3JnDQpTdWJqZWN0OiBSZTogUkZDIDMyODAgYnVnIGluIG5hbWUgY29u c3RyYWludHMNCg0KDQpJIGFncmVlLiBJbiBmYWN0LCBJIGRvbid0IHNlZSB3aHkgeW91IHdvdWxk bid0DQphcHBseSBuYW1lIGNvbnN0cmFpbnRzIHRvIGFuIEVtYWlsQWRkcmVzcyBhdHRyaWJ1dGUN CmluIGEgc3ViamVjdCBETiBldmVuIGlmIHRoZXJlICppcyogYW4gcmZjODIyTmFtZQ0KaW4gYSBz dWJqZWN0IGFsdGVybmF0aXZlIG5hbWUuIEkgc3VnZ2VzdCB0aGF0DQp0aGUgdGV4dCBiZSByZXZp c2VkIHRvIHNheToNCg0KICAgIEFuIHJmYzgyMiBuYW1lIGNvbnN0cmFpbnQgTVVTVCBiZSBhcHBs aWVkIHRvIGFuDQogICAgYXR0cmlidXRlIG9mIHR5cGUgRW1haWxBZGRyZXNzIGluIHRoZSBzdWJq ZWN0DQogICAgZGlzdGluZ3Vpc2hlZCBuYW1lLg0KDQpIb3dldmVyLCBJJ20gd2lsbGluZyB0byBi YWNrIGRvd24gb24gdGhpcyBpZg0KcGVvcGxlIGZlZWwgaXQncyB0b28gc3RyaWN0LiBNeSBjb25j ZXJuIGlzIHRoYXQNCmEgY2VydGlmaWNhdGUgd2l0aCBhbiBlbWFpbCBhZGRyZXNzIGluIHRoZQ0K c3ViamVjdCBETiBwcm9oaWJpdGVkIGJ5IG5hbWUgY29uc3RyYWludHMgd2lsbA0KYmUgYWxsb3dl ZCBiZWNhdXNlIGl0IGFsc28gY29udGFpbnMgYSBwZXJtaXNzaWJsZQ0KZW1haWwgYWRkcmVzcyBp biB0aGUgc3ViamVjdCBhbHRlcm5hdGl2ZSBuYW1lLg0KSSBzdXNwZWN0IG1hbnkgYXBwbGljYXRp b25zIHdpbGwgbm90IHVuZGVyc3RhbmQNCnRoaXMgYW5kIHdpbGwgYWNjZXB0IHRoZSBlbWFpbCBh ZGRyZXNzIGluIHRoZQ0Kc3ViamVjdCBETi4gSW4gZmFjdCwgUkZDIDI2MzIgYW5kIGl0cyBzdWNj ZXNzb3INCmRvbid0IGNvbnRhaW4gYW55IHdhcm5pbmcgYWJvdXQgdGhpcy4NCg0KVGhhbmtzLA0K DQpTdGV2ZQ0KDQpTdGVmYW4gU2FudGVzc29uIHdyb3RlOg0KPiBJZiBzb21lb25lIGlzIGtlZXBp bmcgcmVjb3JkIG9mIGJ1Z3MgZm9yIGZ1dHVyZSB1cGRhdGUgb2YgUkZDIDMyODAgSQ0KPiBoYXZl IGEgc21hbGwgb25lIHRvIGZpbGUgYWJvdXQgbmFtZSBjb25zdHJhaW50cy4NCj49MjANCj4gPTIw DQo+PTIwDQo+IEluIHNlY3Rpb24gNC4yLjEuMTEsIHN0YXJ0aW5nIGF0IGJvdHRvbSBvZiBwYWdl IDM4Og0KPj0yMA0KPiA9MjANCj49MjANCj4gICAgV2hlbiByZmM4MjIgbmFtZXMgYXJlIGNvbnN0 cmFpbmVkLCBidXQgdGhlDQo+PTIwDQo+ICAgIGNlcnRpZmljYXRlIGRvZXMgbm90IGluY2x1ZGUg YSBzdWJqZWN0IGFsdGVybmF0aXZlIG5hbWUsIHRoZSByZmM4MjINCj49MjANCj4gICAgbmFtZSBj b25zdHJhaW50IE1VU1QgYmUgYXBwbGllZCB0byB0aGUgYXR0cmlidXRlIG9mIHR5cGU9MjANCj4g RW1haWxBZGRyZXNzDQo+PTIwDQo+ICAgIGluIHRoZSBzdWJqZWN0IGRpc3Rpbmd1aXNoZWQgbmFt ZS4NCj49MjANCj4gPTIwDQo+PTIwDQo+IFN1cHBvc2Uvc3VnZ2VzdCB0aGF0IHRoZSBpbnRlbmRl ZCBtZWFuaW5nIGlzOg0KPj0yMA0KPiA9MjANCj49MjANCj4gICAgV2hlbiByZmM4MjIgbmFtZXMg YXJlIGNvbnN0cmFpbmVkLCBidXQgdGhlIGNlcnRpZmljYXRlIGRvZXMgbm90DQo+PTIwDQo+ICAg IGluY2x1ZGUgYW4gcmZjODIyTmFtZSBpbiBhIHN1YmplY3QgYWx0ZXJuYXRpdmUgbmFtZSBleHRl bnNpb24sIHRoZT0yMA0KPiByZmM4MjINCj49MjANCj4gICAgbmFtZSBjb25zdHJhaW50IE1VU1Qg YmUgYXBwbGllZCB0byB0aGUgYXR0cmlidXRlIG9mIHR5cGU9MjANCj4gRW1haWxBZGRyZXNzDQo+ PTIwDQo+ICAgIGluIHRoZSBzdWJqZWN0IGRpc3Rpbmd1aXNoZWQgbmFtZS4NCj49MjANCj4gPTIw DQo+PTIwDQo+ID0yMA0KPj0yMA0KPiBSYXRpb25hbGU6DQo+PTIwDQo+IFdoeSB3b3VsZCB0aGUg Y29uc3RyYWludCBOT1QgYXBwbHkgdG8gRW1haWxBZGRyZXNzIGp1c3QgYmVjYXVzZSB0aGVyZT0y MA0KPiBpcw0KPiBhIFN1YmpBbHROYW1lIGV4dGVuc2lvbiB3aXRoIGUuZy4ganVzdCBzb21lIHVu cmVsYXRlZCBvdGhlciBuYW1lPTIwDQo+IGNvbXBvbmVudCBwcmVzZW50Pw0KPj0yMA0KPiA9MjAN Cj49MjANCj4gL1N0ZWZhbg0KPj0yMA0KPiA9MjANCj49MjANCg0KLS0tLT1fTmV4dFBhcnRfU1RK XzM4ODQ1Nzc2LjExMzkxOTk4NQ0KQ29udGVudC1UeXBlOiBhcHBsaWNhdGlvbi9ydGYNCkNvbnRl bnQtVHJhbnNmZXItRW5jb2Rpbmc6IGJhc2U2NA0KDQplMXh5ZEdZeFhHRnVjMmxjWVc1emFXTnda ekV5TlRKY1puSnZiWFJsZUhRZ1hHUmxabVl3ZTF4bWIyNTBkR0pzDQpEUXA3WEdZd1hHWnpkMmx6 Y3lCQmNtbGhiRHQ5RFFwN1hHWXhYR1p0YjJSbGNtNGdRMjkxY21sbGNpQk9aWGM3DQpmUTBLZTF4 bU1seG1ibWxzWEdaamFHRnljMlYwTWlCVGVXMWliMnc3ZlEwS2UxeG1NMXhtYlc5a1pYSnVYR1pq DQphR0Z5YzJWME1DQkRiM1Z5YVdWeUlFNWxkenQ5ZlEwS2UxeGpiMnh2Y25SaWJGeHlaV1F3WEdk eVpXVnVNRnhpDQpiSFZsTUR0Y2NtVmtNRnhuY21WbGJqQmNZbXgxWlRJMU5UdDlEUXBjZFdNeFhI QmhjbVJjY0d4aGFXNWNaR1ZtDQpkR0ZpTXpZd0lGeG1NRnhtY3pJd0lFa2djSEpsWm1WeUlIUnZJ R3hsWVhabElETXlPREFnWVhNZ2FYTWdLSGRwDQpkR2dnVTNSbFptRnVKM01nY0hKdmNHOXpaV1Fn WTJoaGJtZGxJSGRvYVdOb0lFa2daRzhnYzNWd2NHOXlkQ2t1DQpJRlJvWlNCemNHVmphV0ZzSUdO aGMyVWdabTl5SUdOb1pXTnJhVzVuSUdWdFlXbHNJR0ZrWkhKbGMzTWdhVzRnDQpjM1ZpYW1WamRD Qm1hV1ZzWkNCcGN5d2dabkp2YlNCdGVTQjFibVJsY25OMFlXNWthVzVuTENCbWIzSWdZbUZqDQph M2RoY21RZ1kyOXRjR0YwYVdKcGJHbDBlU0J5WldGemIyNXpJRzl1YkhrZ1ptOXlJSE5wZEhWaGRH bHZibk1nDQpkMmhsY21VZ2RHaGxJR1Z0WVdsc0lHRmtaSEpsYzNNZ1pHbGtiaWQwSUdGd2NHVmhj aUJwYmlCMGFHVWdaWGgwDQpaVzV6YVc5dUxpQlVhR1VnYzNWaWFtVmpkQ0JtYVdWc1pDQnBjeUJo SUVST0lHRnVaQ0J6YUc5MWJHUWdiMjVzDQplU0JpWlNCeVpYRjFhWEpsWkNCMGJ5QmlaU0JqYUdW amEyVmtJR2xtSUhSb1pYSmxJR0Z5WlNCRVRpQnVZVzFsDQpJR052Ym5OMGNtRnBiblJ6SUNobGVH TmxjSFFnWm05eUlIUm9hWE1nWW1GamEzZGhjbVFnWTI5dGNHRjBhV0pwDQpiR2wwZVNCcGMzTjFa U0IzYUdWeVpTQjBhR1VnWlhoalpYQjBhVzl1SUdseklHMWhaR1VwTGlCY2NHRnlEUXBjDQpjR0Z5 RFFvdExTMHRMVTl5YVdkcGJtRnNJRTFsYzNOaFoyVXRMUzB0TFZ4d1lYSU5Da1p5YjIwNklHOTNi bVZ5DQpMV2xsZEdZdGNHdHBlRUJ0WVdsc0xtbHRZeTV2Y21jZ1cyMWhhV3gwYnpwdmQyNWxjaTFw WlhSbUxYQnJhWGhBDQpiV0ZwYkM1cGJXTXViM0puWFNCUGJpQkNaV2hoYkdZZ1QyWWdVM1JsZG1V Z1NHRnVibUZjY0dGeURRcFRaVzUwDQpPaUJVYUhWeWMyUmhlU3dnVFdGNUlEQTJMQ0F5TURBMElE RXdPak0xSUVGTlhIQmhjZzBLVkc4NklGTjBaV1poDQpiaUJUWVc1MFpYTnpiMjVjY0dGeURRcERZ em9nYVdWMFppMXdhMmw0UUdsdFl5NXZjbWRjY0dGeURRcFRkV0pxDQpaV04wT2lCU1pUb2dVa1pE SURNeU9EQWdZblZuSUdsdUlHNWhiV1VnWTI5dWMzUnlZV2x1ZEhOY2NHRnlEUXBjDQpjR0Z5RFFw Y2NHRnlEUXBKSUdGbmNtVmxMaUJKYmlCbVlXTjBMQ0JKSUdSdmJpZDBJSE5sWlNCM2FIa2dlVzkx DQpJSGR2ZFd4a2JpZDBYSEJoY2cwS1lYQndiSGtnYm1GdFpTQmpiMjV6ZEhKaGFXNTBjeUIwYnlC aGJpQkZiV0ZwDQpiRUZrWkhKbGMzTWdZWFIwY21saWRYUmxYSEJoY2cwS2FXNGdZU0J6ZFdKcVpX TjBJRVJPSUdWMlpXNGdhV1lnDQpkR2hsY21VZ0ttbHpLaUJoYmlCeVptTTRNakpPWVcxbFhIQmhj ZzBLYVc0Z1lTQnpkV0pxWldOMElHRnNkR1Z5DQpibUYwYVhabElHNWhiV1V1SUVrZ2MzVm5aMlZ6 ZENCMGFHRjBYSEJoY2cwS2RHaGxJSFJsZUhRZ1ltVWdjbVYyDQphWE5sWkNCMGJ5QnpZWGs2WEhC aGNnMEtYSEJoY2cwS0lDQWdJRUZ1SUhKbVl6Z3lNaUJ1WVcxbElHTnZibk4wDQpjbUZwYm5RZ1RW VlRWQ0JpWlNCaGNIQnNhV1ZrSUhSdklHRnVYSEJoY2cwS0lDQWdJR0YwZEhKcFluVjBaU0J2DQpa aUIwZVhCbElFVnRZV2xzUVdSa2NtVnpjeUJwYmlCMGFHVWdjM1ZpYW1WamRGeHdZWElOQ2lBZ0lD QmthWE4wDQphVzVuZFdsemFHVmtJRzVoYldVdVhIQmhjZzBLWEhCaGNnMEtTRzkzWlhabGNpd2dT U2R0SUhkcGJHeHBibWNnDQpkRzhnWW1GamF5QmtiM2R1SUc5dUlIUm9hWE1nYVdaY2NHRnlEUXB3 Wlc5d2JHVWdabVZsYkNCcGRDZHpJSFJ2DQpieUJ6ZEhKcFkzUXVJRTE1SUdOdmJtTmxjbTRnYVhN Z2RHaGhkRnh3WVhJTkNtRWdZMlZ5ZEdsbWFXTmhkR1VnDQpkMmwwYUNCaGJpQmxiV0ZwYkNCaFpH UnlaWE56SUdsdUlIUm9aVnh3WVhJTkNuTjFZbXBsWTNRZ1JFNGdjSEp2DQphR2xpYVhSbFpDQmll U0J1WVcxbElHTnZibk4wY21GcGJuUnpJSGRwYkd4Y2NHRnlEUXBpWlNCaGJHeHZkMlZrDQpJR0ps WTJGMWMyVWdhWFFnWVd4emJ5QmpiMjUwWVdsdWN5QmhJSEJsY20xcGMzTnBZbXhsWEhCaGNnMEta VzFoDQphV3dnWVdSa2NtVnpjeUJwYmlCMGFHVWdjM1ZpYW1WamRDQmhiSFJsY201aGRHbDJaU0J1 WVcxbExseHdZWElODQpDa2tnYzNWemNHVmpkQ0J0WVc1NUlHRndjR3hwWTJGMGFXOXVjeUIzYVd4 c0lHNXZkQ0IxYm1SbGNuTjBZVzVrDQpYSEJoY2cwS2RHaHBjeUJoYm1RZ2QybHNiQ0JoWTJObGNI UWdkR2hsSUdWdFlXbHNJR0ZrWkhKbGMzTWdhVzRnDQpkR2hsWEhCaGNnMEtjM1ZpYW1WamRDQkVU aTRnU1c0Z1ptRmpkQ3dnVWtaRElESTJNeklnWVc1a0lHbDBjeUJ6DQpkV05qWlhOemIzSmNjR0Z5 RFFwa2IyNG5kQ0JqYjI1MFlXbHVJR0Z1ZVNCM1lYSnVhVzVuSUdGaWIzVjBJSFJvDQphWE11WEhC aGNnMEtYSEJoY2cwS1ZHaGhibXR6TEZ4d1lYSU5DbHh3WVhJTkNsTjBaWFpsWEhCaGNnMEtYSEJo DQpjZzBLVTNSbFptRnVJRk5oYm5SbGMzTnZiaUIzY205MFpUcGNjR0Z5RFFvK0lFbG1JSE52YldW dmJtVWdhWE1nDQphMlZsY0dsdVp5QnlaV052Y21RZ2IyWWdZblZuY3lCbWIzSWdablYwZFhKbElI VndaR0YwWlNCdlppQlNSa01nDQpNekk0TUNCSlhIQmhjZzBLUGlCb1lYWmxJR0VnYzIxaGJHd2di MjVsSUhSdklHWnBiR1VnWVdKdmRYUWdibUZ0DQpaU0JqYjI1emRISmhhVzUwY3k1Y2NHRnlEUW8r SUZ4d1lYSU5DajRnSUZ4d1lYSU5DajRnWEhCaGNnMEtQaUJKDQpiaUJ6WldOMGFXOXVJRFF1TWk0 eExqRXhMQ0J6ZEdGeWRHbHVaeUJoZENCaWIzUjBiMjBnYjJZZ2NHRm5aU0F6DQpPRHBjY0dGeURR bytJRnh3WVhJTkNqNGdJRnh3WVhJTkNqNGdYSEJoY2cwS1BpQWdJQ0JYYUdWdUlISm1Zemd5DQpN aUJ1WVcxbGN5QmhjbVVnWTI5dWMzUnlZV2x1WldRc0lHSjFkQ0IwYUdWY2NHRnlEUW8rSUZ4d1lY SU5DajRnDQpJQ0FnWTJWeWRHbG1hV05oZEdVZ1pHOWxjeUJ1YjNRZ2FXNWpiSFZrWlNCaElITjFZ bXBsWTNRZ1lXeDBaWEp1DQpZWFJwZG1VZ2JtRnRaU3dnZEdobElISm1Zemd5TWx4d1lYSU5DajRn WEhCaGNnMEtQaUFnSUNCdVlXMWxJR052DQpibk4wY21GcGJuUWdUVlZUVkNCaVpTQmhjSEJzYVdW a0lIUnZJSFJvWlNCaGRIUnlhV0oxZEdVZ2IyWWdkSGx3DQpaU0JjY0dGeURRbytJRVZ0WVdsc1FX UmtjbVZ6YzF4d1lYSU5DajRnWEhCaGNnMEtQaUFnSUNCcGJpQjBhR1VnDQpjM1ZpYW1WamRDQmth WE4wYVc1bmRXbHphR1ZrSUc1aGJXVXVYSEJoY2cwS1BpQmNjR0Z5RFFvK0lDQmNjR0Z5DQpEUW8r SUZ4d1lYSU5DajRnVTNWd2NHOXpaUzl6ZFdkblpYTjBJSFJvWVhRZ2RHaGxJR2x1ZEdWdVpHVmtJ RzFsDQpZVzVwYm1jZ2FYTTZYSEJoY2cwS1BpQmNjR0Z5RFFvK0lDQmNjR0Z5RFFvK0lGeHdZWElO Q2o0Z0lDQWdWMmhsDQpiaUJ5Wm1NNE1qSWdibUZ0WlhNZ1lYSmxJR052Ym5OMGNtRnBibVZrTENC aWRYUWdkR2hsSUdObGNuUnBabWxqDQpZWFJsSUdSdlpYTWdibTkwWEhCaGNnMEtQaUJjY0dGeURR bytJQ0FnSUdsdVkyeDFaR1VnWVc0Z2NtWmpPREl5DQpUbUZ0WlNCcGJpQmhJSE4xWW1wbFkzUWdZ V3gwWlhKdVlYUnBkbVVnYm1GdFpTQmxlSFJsYm5OcGIyNHNJSFJvDQpaU0JjY0dGeURRbytJSEpt WXpneU1seHdZWElOQ2o0Z1hIQmhjZzBLUGlBZ0lDQnVZVzFsSUdOdmJuTjBjbUZwDQpiblFnVFZW VFZDQmlaU0JoY0hCc2FXVmtJSFJ2SUhSb1pTQmhkSFJ5YVdKMWRHVWdiMllnZEhsd1pTQmNjR0Z5 DQpEUW8rSUVWdFlXbHNRV1JrY21WemMxeHdZWElOQ2o0Z1hIQmhjZzBLUGlBZ0lDQnBiaUIwYUdV Z2MzVmlhbVZqDQpkQ0JrYVhOMGFXNW5kV2x6YUdWa0lHNWhiV1V1WEhCaGNnMEtQaUJjY0dGeURR bytJQ0JjY0dGeURRbytJRnh3DQpZWElOQ2o0Z0lGeHdZWElOQ2o0Z1hIQmhjZzBLUGlCU1lYUnBi MjVoYkdVNlhIQmhjZzBLUGlCY2NHRnlEUW8rDQpJRmRvZVNCM2IzVnNaQ0IwYUdVZ1kyOXVjM1J5 WVdsdWRDQk9UMVFnWVhCd2JIa2dkRzhnUlcxaGFXeEJaR1J5DQpaWE56SUdwMWMzUWdZbVZqWVhW elpTQjBhR1Z5WlNCY2NHRnlEUW8rSUdselhIQmhjZzBLUGlCaElGTjFZbXBCDQpiSFJPWVcxbElH VjRkR1Z1YzJsdmJpQjNhWFJvSUdVdVp5NGdhblZ6ZENCemIyMWxJSFZ1Y21Wc1lYUmxaQ0J2DQpk R2hsY2lCdVlXMWxJRnh3WVhJTkNqNGdZMjl0Y0c5dVpXNTBJSEJ5WlhObGJuUS9YSEJoY2cwS1Bp QmNjR0Z5DQpEUW8rSUNCY2NHRnlEUW8rSUZ4d1lYSU5DajRnTDFOMFpXWmhibHh3WVhJTkNqNGdY SEJoY2cwS1BpQWdYSEJoDQpjZzBLUGlCY2NHRnlEUXA5DQotLS0tPV9OZXh0UGFydF9TVEpfMzg4 NDU3NzYuMTEzOTE5OTg1LS0NCg0KoIILHjCCA6YwggKOoAMCAQICBD9jj6kwDQYJKoZIhvcNAQEF BQAwMTELMAkGA1UEBhMCQ0ExEDAOBgNVBAoTB0VudHJ1c3QxEDAOBgNVBAsTB1IgYW5kIEQwHhcN MDQwMzE4MTM1OTIzWhcNMDQwOTE4MTQyOTIzWjBZMQswCQYDVQQGEwJDQTEQMA4GA1UEChMHRW50 cnVzdDEQMA4GA1UECxMHUiBhbmQgRDEmMA4GA1UEBRMHMTYxNjY3MTAUBgNVBAMTDVNoYXJvbiBC b2V5ZW4wgZ8wDQYJKoZIhvcNAQEBBQADgY0AMIGJAoGBAK3LTAUNDH//zxzhL9JK8tJQEjbvYEo4 uKUQXAuZ0sTkMIgJsFYBqgq4nTY3T0ks335ZAlQKnBHJEiIZhFKdeac1dwP0wzzvFzMpmARgkmyA Ne+7A45zBBoRPuUdfnGiDlcC9WZcBU1JE+VkQmGuvfJ5GKEVAdDcYHU5h4GWmytPAgMBAAGjggEg MIIBHDALBgNVHQ8EBAMCB4AwKwYDVR0QBCQwIoAPMjAwNDAzMTgxMzU5MjNagQ8yMDA0MDcyNTA5 MjkyM1owJAYDVR0RBB0wG4EZc2hhcm9uLmJvZXllbkBlbnRydXN0LmNvbTBUBgNVHR8ETTBLMEmg R6BFpEMwQTELMAkGA1UEBhMCQ0ExEDAOBgNVBAoTB0VudHJ1c3QxEDAOBgNVBAsTB1IgYW5kIEQx DjAMBgNVBAMTBUNSTDI4MB8GA1UdIwQYMBaAFPPwLCjmOac0GauakaimZkCSnOJxMB0GA1UdDgQW BBRLwMRUvNu/y0W8q9Sv5AI7I56SNDAJBgNVHRMEAjAAMBkGCSqGSIb2fQdBAAQMMAobBFY3LjAD AgSwMA0GCSqGSIb3DQEBBQUAA4IBAQA5AmaD0OMLcM5wpPSMY2jsOWN6Y7ruyM2ZX31yG9qtrnnU qwPcVMssLU3LSw6xLKQJ8kgwXejAuo1gGyCzgMxTl8LyF5TXCY6ZXp7cNa0vquSwg1hiNbLhyqwE uBho/CRgSfkrZ/9mfEJ4s4REyMa3q1CR44QlBbFNZSmfRWtDsZyWMsTxFBb9iqws1r3rtHVXmuOm hVziw/qx9oujmzPJwe1q1TQahcKmabDggSjMU1M1ylkUCI3+PVJSzPQc7AwZnKjSE7LWW07t2JMF CFUrAi5o4/Ua9pJr7jC5zXQ/Xyf/PAmPxkx7StmbXzKHRMkFxlMr1Q3OmjzzlnR3sNPbMIIDdzCC Al+gAwIBAgIEP2OLWzANBgkqhkiG9w0BAQUFADAxMQswCQYDVQQGEwJDQTEQMA4GA1UEChMHRW50 cnVzdDEQMA4GA1UECxMHUiBhbmQgRDAeFw0wNDAyMTkxMjI2MTJaFw0wNDA4MTkxMjU2MTJaMFkx CzAJBgNVBAYTAkNBMRAwDgYDVQQKEwdFbnRydXN0MRAwDgYDVQQLEwdSIGFuZCBEMSYwDgYDVQQF EwcxNjE2NjcxMBQGA1UEAxMNU2hhcm9uIEJvZXllbjCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkC gYEArJeYhKIoggxfZueFCUpclFz9Me0aKbCfR/ZU1rpQXGM5gPJthGQyfb5OtMjVX3f9LHimUA+T KTlCwjLtLleB2YxZhKEuoEj4ccTipUlLdzWRqESANT7MZXo7uRy+0spsJYuL68txOA8a4sLbmcDS FssiQWpFn8bq0p+MoSL0cOcCAwEAAaOB8jCB7zALBgNVHQ8EBAMCBSAwJAYDVR0RBB0wG4EZc2hh cm9uLmJvZXllbkBlbnRydXN0LmNvbTBUBgNVHR8ETTBLMEmgR6BFpEMwQTELMAkGA1UEBhMCQ0Ex EDAOBgNVBAoTB0VudHJ1c3QxEDAOBgNVBAsTB1IgYW5kIEQxDjAMBgNVBAMTBUNSTDI3MB8GA1Ud IwQYMBaAFPPwLCjmOac0GauakaimZkCSnOJxMB0GA1UdDgQWBBR1jmOMGJBhHCpWex5f8NjcRAw3 ETAJBgNVHRMEAjAAMBkGCSqGSIb2fQdBAAQMMAobBFY3LjADAgSwMA0GCSqGSIb3DQEBBQUAA4IB AQAsfe6ZE2k7XMGR+UKtf72THgEWw1OBG5M7a+YiE8V3pPEotm02/0N7nNaKT/j0q0BJgj2OAW8a nV56WMXHGvdb24x/pXuEAMC+Rn57JiyIy9IkgH5ijoK3Iqlp9QIzEYprDIee67o78efGlfkSF4R2 zXEOZyxoESIdQTduaOL96gKftC198GNrOthzKxUTLnX8g9ltObBLptrSxy/UY7O0SP4yX++bZamU E1FOMTlW50hZauok4i2R3qViLskxd4xLdmp5VL3c77Uf2xxyDuV9iR2rlcUwTtFeC56UadaDK+7v gSte6upddEyhT9witYNBJ1Bat24uco+PC/OxkkJ/MIID9TCCAt2gAwIBAgIEMkiqIDANBgkqhkiG 9w0BAQUFADAxMQswCQYDVQQGEwJDQTEQMA4GA1UEChMHRW50cnVzdDEQMA4GA1UECxMHUiBhbmQg RDAeFw0wMjA0MTcwMjIwMjZaFw0yMjA0MTcwMjUwMjZaMDExCzAJBgNVBAYTAkNBMRAwDgYDVQQK EwdFbnRydXN0MRAwDgYDVQQLEwdSIGFuZCBEMIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKC AQEAvIeIjfoD5D86sD8CehB+kVhYJTL+VuaUZiVijGevUTlnUAwy1WpOYVpFCa8VK116TJ+o99ax N1jMbsVBYfRzFk9SWl4iAH+PyJ1S6D5Yl593KjHPhBCfdD0v9KoPqXhCwRlkvupreNlFx3FFk2Uf 5jG/i9SCP925OXbz3wod2mdIu8ueNWgTmA2XAOBdHZg6HYKo50BWUIgtRwFvzbh1J57+gNdHxMYL Zmvj5saS1mDTSibkPH/AY+D+xAozRmyOYa4SY2HVfZSM2SP7MkuVtKp17pzhwicoWAGKXrKeDRWD YeZBTJjHoNP/SXKdpSzY7rV8GPrfWFoogc4350VbowIDAQABo4IBEzCCAQ8wEQYJYIZIAYb4QgEB BAQDAgAHMFMGA1UdHwRMMEowSKBGoESkQjBAMQswCQYDVQQGEwJDQTEQMA4GA1UEChMHRW50cnVz dDEQMA4GA1UECxMHUiBhbmQgRDENMAsGA1UEAxMEQ1JMMTArBgNVHRAEJDAigA8yMDAyMDQxNzAy MjAyNlqBDzIwMjIwNDE3MDI1MDI2WjALBgNVHQ8EBAMCAQYwHwYDVR0jBBgwFoAU8/AsKOY5pzQZ q5qRqKZmQJKc4nEwHQYDVR0OBBYEFPPwLCjmOac0GauakaimZkCSnOJxMAwGA1UdEwQFMAMBAf8w HQYJKoZIhvZ9B0EABBAwDhsIVjYuMDo0LjADAgSQMA0GCSqGSIb3DQEBBQUAA4IBAQCAPnm1Rpn5 ya5kBLFpckRtw+XuY/G5m6uQH9hVXgechp+i6/tXlYjQ8ccnJy1HcOmPS718LkG4drxBfrhjO9aT udv+TNIMR9izay796hsBk7raBpYtxavOEjxb8jfp+EelztgcSZo737mYzcrr9pDjzJldsriCzPDU /N052R+RJ4Rs/0IiuWzIVH7y99S9O+rG14PjEKNQhA39gTwg9iMBtQoksDfKu09UCaFsE7aYUC8e E0wzki4G+Dl5RKE57MGaYAh+cccmyPR6Xw/gl8Yo3YGhOWz3qatAhbbplACm9LSmduMSLXQybke6 cq5JUb8c9VMzi9cIiv6Hbf1/xObPMYICZTCCAmECAQEwOTAxMQswCQYDVQQGEwJDQTEQMA4GA1UE ChMHRW50cnVzdDEQMA4GA1UECxMHUiBhbmQgRAIEP2OPqTAJBgUrDgMCGgUAoIIBgjAYBgkqhkiG 9w0BCQMxCwYJKoZIhvcNAQcBMBwGCSqGSIb3DQEJBTEPFw0wNDA1MDYxNTM5MjBaMCMGCSqGSIb3 DQEJBDEWBBRNKF5VSEcbYDPUX4MukmkegrypwDCCASEGCSqGSIb3DQEJDzGCARIwggEOMAsGCWCG SAFlAwQBKjALBglghkgBZQMEARYwCwYJYIZIAWUDBAECMA8GCSqGSIb2fQdCCgICAIAwDgYIKoZI hvcNAwICAgCAMAoGCCqGSIb3DQMHMA0GCysGAQQBgTwHAQECMA4GCSqGSIb2fQdCCgIBKDANBggq hkiG9w0DAgIBKDAHBgUrDgMCBzAHBgUrDgMCGjAKBggqhkiG9w0CBTALBgkqhkiG9w0BAQEwCwYJ KoZIhvcNAQEHMAkGByqGSM44BAEwCQYHKoZIzj0CATALBgkqhkiG9w0BAQUwCwYJKoZIhvcNAQEE MAkGByqGSM44BAMwCQYHKoZIzj0EATAMBgoqhkiG9w0BCQ8BMA0GCSqGSIb3DQEBAQUABIGAZ/zd adL4+m/1R8CArZD6nEd5eM7x40yFtakYs7Yn8glTJb4smp4FKaPH1e7LLaQ1GtVOz8f4Wc10K6uc umtMuGJSlQDBLWg3Hpl+lwH2o0AbzHJDNP3juK3dY+5LQo4DCqykJvy23TYlGeTwQv8EScyqylQy GP8sH0tdsM7ljN4= Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46FYMPU077271; Thu, 6 May 2004 08:34:22 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46FYMoQ077270; Thu, 6 May 2004 08:34:22 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from sottmxssm1.entrust.com (sottmxssm1.entrust.com [216.191.252.11]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46FYKwG077250 for <ietf-pkix@imc.org>; Thu, 6 May 2004 08:34:21 -0700 (PDT) (envelope-from sharon.boeyen@entrust.com) Received: from SOTTMXS01.entrust.com ([10.4.61.7]) by sottmxssm1.entrust.com (Switch-3.1.4/Switch-3.1.0) with ESMTP id i46FhWm9029268 for <ietf-pkix@imc.org>; Thu, 6 May 2004 11:43:32 -0400 Received: by sottmxs01.entrust.com with Internet Mail Service (5.5.2657.72) id <KM64ZNYG>; Thu, 6 May 2004 11:34:16 -0400 Message-ID: <7A3E1242FA9989439AD1F9B2D71C287F0AD96F@sottmxs05.entrust.com> From: Sharon Boeyen <sharon.boeyen@entrust.com> To: ietf-pkix@imc.org Subject: RE: Cross certificate format Date: Thu, 6 May 2004 11:34:16 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: text/plain Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I'd also like to clarify that the constraints that input to the path validation process (e.g. initial-policy-mapping-inhibit, initial-explicit-policy etc) are not tied necessarily to the trust anchor but are tied to an instance of path validation conducted by the relying party. It may be the relying party and not CA who happens to be the trust anchor that determines these constraints. They are just additional inputs to the process, as is the trust anchor public key. The same validation engine may have different constraints on instances of path validation that use the same trust anchor public key. Also, the same set of constraints can apply to instances of validation that use different trust anchors. Different users, different apps and even different transactions can have different constraints even though they may happen to use the same trust anchor. While it is true that including some extensions in a self-signed might be one way to convey constraints that could then be extracted and provided as the relevant inputs to the validation process, this is certainly not the only way to convey initialization constraint values for a RP. Also, I believe it is a very limiting way because then the constraints are tied to all validations that occur with that key by all RPs (assuming all RPs used that scheme to initialize their path validation variables). The mechanisms that determine those inputs to the path validation process (including which trust anchor public key to use, process are outside the scope of X.509. X.509 provides the mechanism (path validation variable initialization) to set the constraints once the RP has determined what they are to be. On the nameConstraints issue, there is work underway in X.509 that will add one more input to the path validation process to enable nameConstraints to be imposed at initialization time just as other constraints can already. Similar to the other constraints, no single value of initialization constraints is tied necessarily to all uses of a trust anchor public key. On the earlier discussion of processing extensions in self-signed certs. I believe X.509 and 3280 are completely consistent here in pointing out that the self-signed cert is not part of the path (but may be a convenient way for a CA to convey its public key). See clause 8.1.5 of X.509 for the relevant text "For purposes of path validation, self-issued certificates of type a) are verified with the public key contained in them, and if they are encountered in the path, they shall be ignored." As a result of Defect Report 285, that same point is the basic certificate checks in the path processing alogithm in X.509. In the first paragraph of 10.5.1, you'll find the sentence "Self-signed certificates, if encountered in the path are ignored." As for the checking of whether a self-signed cert has basic constraints with CA set to TRUE etc, that is part of the checking that could be done to place trust in the fact that certified key should be used as an input to the path validation process. However, because X.509 does not mandate any scheme for conveying these keys to users, it obviously doesn't mandate any checks on those schemes. Any checking along these lines is outside the scope of the path validation process because it would be done before the process is initialized with that key. Sorry for the lengthy message Sharon -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Steve Hanna Sent: Wednesday, May 05, 2004 5:26 PM To: Tom Gindin Cc: Santosh Chokhani; ietf-pkix@imc.org Subject: Re: Cross certificate format See section 6.2 of RFC 3280, which says: ... An implementation that supports multiple trust anchors MAY augment the algorithm presented in section 6.1 to further limit the set of valid certification paths which begin with a particular trust anchor. For example, an implementation MAY modify the algorithm to apply name constraints to a specific trust anchor during the initialization phase, or the application MAY require the presence of a particular alternative name form in the end entity certificate, or the application MAY impose requirements on application-specific extensions. Thus, the path validation algorithm presented in section 6.1 defines the minimum conditions for a path to be considered valid. There's no need to change RFC 3280 to allow path validation implementations to apply name constraints to trust anchors. That's explicitly discussed in this section of the RFC. Thanks, Steve Tom Gindin wrote: > Santosh: > > Of course, defining a trust anchor implies that the RP trusts > the > anchor's public key as an issuer for at least some purposes. RFC 3280, in > the parts of section 6.1.2 I cited, appears to forbid those state > variables from being set from information defined by the RP or stored with > the trust anchor. This is consistent with its statements in 6.1.1 d which > permit self-signed certificates but say nothing about other CA > certificates. As is probably clear to everyone reading this thread, I > think that this is not an ideal set of suggestions. > The current standard does contain some constraints on the trust > anchor (notably inhibit-policy-mapping), but neither name constraints nor > basic constraints. I consider this an anomaly. If I had free rein in > rewording pieces of RFC 3280 section 6.1, I would change this wording to > allow CA certificates along with self-signed ones as anchors and create > extra optional (and optional to support) fields to be set from the trust > anchor. > I don't think that this is particularly inconsistent with general > security principles, or with any part of X.509 with which I am familiar. > It does represent an extension of RFC 3280 section 6, and it is only > backward compatible with it, so it may be out of scope. > > Tom Gindin > > > > > > "Santosh Chokhani" <chokhani@orionsec.com> > Sent by: owner-ietf-pkix@mail.imc.org > 05/01/2004 06:46 PM > > > To: <ietf-pkix@imc.org> > cc: > Subject: RE: Cross certificate format > > > > > Tom: > > The standard does not require you to check how you initialize the > variables. > You can get these from the cross certificate or from some configuration. > > The basic point is that you trust the public key. How you constrain > it is not address by the standard, where as if the same certificate > appeared in a path, you have to apply the constraints or be not > compliant with the standard. > > You obviously do not believe in re-incarnation. Just kidding...... > > -----Original Message----- > From: Tom Gindin [mailto:tgindin@us.ibm.com] > Sent: Saturday, May 01, 2004 6:31 PM > To: Santosh Chokhani > Cc: ietf-pkix@imc.org > Subject: RE: Cross certificate format > > > Santosh: > > It is well known that trust in a trust anchor can be defined > in a > variety of ways. However, there is every reason to think that an RP may > constrain that trust and that the constraints he may apply include those > which an issuer CA may apply to a subordinate one. The wording of RFC 3280 > > section 6.2's first paragraph is an example of this. > My comments about the meaning of a trust anchor are not purely > theoretical. If a cross certificate is used to represent a trust anchor, > the variables permitted_subtrees (RFC 3280 6.1.2 b) and excluded_subtrees > (RFC 3280 6.1.2 c) could easily be set from the contents of a > NameConstraints extension, and the variable max_path_length (RFC 3280 > 6.1.2 k) from the contents of a BasicConstraints extension. In RFC 3280 > today, these variables are set to constants (although that interpretation > is not absolutely clear for permitted_subtrees). In fact, they are three > of the four variables which are set to constants during initialization. > If permitted subtrees were set to a list of attributes, the > validation process would fail if the anchor CA or a CA it certified went > beyond the restrictions of those attributes. Likewise, if the excluded > subtrees were set to a list of attributes, the validation process would > fail if the anchor CA or a CA it certified went used values matching any > of those attributes. I do not believe that I am arguing against Galileo. > > Tom Gindin > > > > > > > "Santosh Chokhani" <chokhani@orionsec.com> > Sent by: owner-ietf-pkix@mail.imc.org > 04/29/2004 11:12 PM > > > To: <ietf-pkix@imc.org> > cc: > Subject: RE: Cross certificate format > > > > > Tom: > > The basic point (at the risk of repeating myself) is that a CA > certificate or a self-signed certificate can be used as a means to > provide trust anchor information. In that case, there is no need to > check signature or anything. > > But, when the CA certificate is a certificate in the certificate path, > X.509 > logic kicks in. > > This is consistent with X.509, 3280 and general security principles > since the trust in a trust anchor can be derived through any number of > means. > > If you do not agree, feel free to argue against the definitions and > Galileo > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org] > On > Behalf Of Tom Gindin > Sent: Thursday, April 29, 2004 8:43 PM > To: Santosh Chokhani > Cc: ietf-pkix@imc.org > Subject: RE: Cross certificate format > > > > Santosh: > > I don't think that a CA certificate and a trust anchor are as > much > > > different things as different uses for the same thing. They are both > identifiers of issuers which bind a public key to an issuer name, and it > is perfectly appropriate for any CA certificate to represent a trust > anchor. I also do not see why such things as name constraints on a trust > anchor are inappropriate. It is perfectly true that verifying a trust > anchor's certificate, when issued by another CA (who may not be trusted) > is a pointless operation. > > Tom Gindin > > > > > > "Santosh Chokhani" <chokhani@orionsec.com> > Sent by: owner-ietf-pkix@mail.imc.org > 04/27/2004 09:14 AM > > > To: <ietf-pkix@imc.org> > cc: > Subject: RE: Cross certificate format > > > > > Tom: > > My point is that even if the trust anchor is a self-signed > certificate, > all > you need to do is extract the required information. There is no need to > examine anything. > > Cross certificate and trust anchor are very different things. An > element > of > the cross certificate pair is nothing but an intermediate CA certificate > in > a certification path. > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org] > On > Behalf Of Tom Gindin > Sent: Tuesday, April 27, 2004 7:33 AM > To: Santosh Chokhani > Cc: ietf-pkix@imc.org; owner-ietf-pkix@mail.imc.org > Subject: RE: Cross certificate format > > > > Santosh: > > Of course, you are right that RFC 3280 6.1.1 d permits trust > anchor information to be provided outside a certificate. If it is so, no > checks of the sort I indicated can be performed. It also says that it can > > > > be provided as "a self-signed certificate", with no further > qualification. > > > > I do think it is somewhat odd that a self-signed certificate with > KeyUsage set to typical EE values would be considered a valid issuer as a > trust anchor while a cross certificate would not. You can always extract > the needed anchor information from a cross-certificate, of course. > > Tom Gindin > > > > > > "Santosh Chokhani" <chokhani@orionsec.com> > Sent by: owner-ietf-pkix@mail.imc.org > 04/26/2004 08:51 AM > > > To: <ietf-pkix@imc.org> > cc: > Subject: RE: Cross certificate format > > > > > Tom: > > My reading of 3280 and X.509 is that a trust anchor need not even be a > certificate. Thus, no checks need to be performed on a trust anchor > at all. You get the DN, public key, algorithm OID, and parameters (if > applicable) from a trust anchor. Any checks are gratuitous and not > required by either standard. > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org] > On > Behalf Of Tom Gindin > Sent: Sunday, April 25, 2004 7:06 PM > To: Jean-Marc Desperrier > Cc: ietf-pkix@imc.org > Subject: Re: Cross certificate format > > > > Jean-Marc: > > RFC 3280 doesn't say that anywhere. It does say that you > don't > need to validate the trust anchor as part of the path, but it isn't clear > from the text that a v1 certificate is acceptable as a trust anchor - and > it certainly isn't clear that a v1 certificate is acceptable as an issuer > if it is a trust anchor while a v3 certificate with KeyUsage= { > DigitalSignature, keyEncipherment } is not acceptable as an issuer even if > > > > > it is a trust anchor. > I believe that the correct rules for versions are something > like > the following: a certificate issued by a trust anchor which is represented > > > > > by a certificate is verifiable if the anchor certificate is either a > v1 > certificate or a v3 certificate with the CA flag present in the > basicConstraints extension. If the anchor certificate is a v3 certificate > > > > > with the KeyUsage extension present, it is only a valid issuer > certificate > > > > > if keyCertSign is set. > > Tom Gindin > > > > > > Jean-Marc Desperrier <jmdesp@free.fr> > Sent by: owner-ietf-pkix@mail.imc.org > 04/21/2004 03:42 PM > > > To: ietf-pkix@imc.org > cc: > Subject: Re: Cross certificate format > > > > > Tom Gindin wrote: > > >> RFC 3280 does not support the use of v1 self-signed >>certificates, > > >>which contain no extensions at all (see the text of RFC 3280's >>basicConstraints section). However, a number of public CA's still have >>them. >> >> > > Applications implementing the path validation algorithm of RFC3280 > MUST > accept them when they are used as trust anchors. > > > > > > > > > > > > > > > > > > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46El2Ww073491; Thu, 6 May 2004 07:47:02 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46El28O073490; Thu, 6 May 2004 07:47:02 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from odin2.bull.net (odin2.bull.net [192.90.70.84]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46El0NA073464 for <ietf-pkix@imc.org>; Thu, 6 May 2004 07:47:01 -0700 (PDT) (envelope-from Denis.Pinkas@bull.net) Received: from clbull.frcl.bull.fr (IDENT:root@clbull.frcl.bull.fr [129.182.8.31]) by odin2.bull.net (8.9.3/8.9.3) with ESMTP id QAA19798; Thu, 6 May 2004 16:56:12 +0200 Received: from bull.net (frcls4013.frcl.bull.fr [129.182.108.120]) by clbull.frcl.bull.fr (8.9.3/8.9.3) with ESMTP id QAA13602; Thu, 6 May 2004 16:42:58 +0200 Message-ID: <409A4FEE.5060303@bull.net> Date: Thu, 06 May 2004 16:47:10 +0200 From: Denis Pinkas <Denis.Pinkas@bull.net> Organization: Bull SA. User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en, fr MIME-Version: 1.0 To: Trevor Freeman <trevorf@exchange.microsoft.com> CC: ietf-pkix@imc.org Subject: Re: FW: scvp References: <33E7A1613A3480448996047B69308A2F02713A06@df-grommit-01.dogfood> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Trevor, > Hi Denis, > There is no mandate in 3379 that a validation policy must be expressed > with a single value. "The protocol MUST allow the client to include these policy dependant parameters in the DPV request; however, it is expected that most clients will simply reference a validation policy." The "however" mandates that a validation policy must also be expressed with a single value. I do not understand your resistance against a single value. The following strcture could accomodate both approaches: ValPolicy ::= SEQUENCE { valPolId OBJECT IDENTIFIER, valPolValue OCTET STRING OPTIONAL } The valPolValue would be dependent upon the OID of the valPolId and the structure of the valPolValue in the case of the "default policy" would be defined in the current document. Other structures could be defined in other documents. Denis > for a given application or accept the DPV server's default > validation policy. > It is not clear that is any value is doing so since > it only raised more ambiguity. This is well trod turf by many groups who > have debated the merits of providing suits vs. "al a carte" combinations > of inputs and the broader consensus is for a "al a carte". > Trevor > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Tuesday, May 04, 2004 3:15 AM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: Re: FW: scvp > > Trevor, > > >>Hi Denis, >> >>3379 does not require that the validation policy be specified in a >>single value. > > > You are playing with words. The text says: > "most clients will simply reference a validation policy" > > This means that validation policies can be referenced and thus > parameters do > not need to be defined through this interface. > > A simple and good reference is an OID. > > >>With SCVP14 a client can accept the default of the SCVP >>server or specify a set of parameters which constitutes its validation >>policy. That is consistent with the requirements of 3379. > > > The specification of the set of parameters should be done through a > different interface. This different interface can then return an OID > Whatever this additional interface can be, it will never be rich enough > to > pass all the details of the validation policy. Thus OIDs can also be > defined > not using this additional interface. > > Denis > > >>Trevor >> >>-----Original Message----- >>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] >>Sent: Thursday, April 29, 2004 3:54 AM >>To: Trevor Freeman >>Cc: ietf-pkix@imc.org >>Subject: Re: FW: scvp >> >>Trevor, >> >> >> >>>Hi Denis, >> >> >>>Can you please be more specific how you think SCVP 14 does not comply >>>with 3379. >>> >>>I can find no section in 3379 where is there a requirement that a >>>validation policy MUST be represented as an OID. >> >> >>There cannot be a requirement with such a level of detail, but the > > text > >>states: >> >> The protocol MUST allow the client to include >> these policy dependant parameters in the DPV request; however, it > > is > >> expected that most clients will simply reference a validation > > policy > >> for a given application or accept the DPV server's default >>validation >> policy. >> >>A simple reference is an OID. >> >>FYI, I do not expect to use the "default validation policy". >> >>Denis >> >> >> >> >>>Given hiding the detail of what a policy is within an OID simply opens >>>the rat hole of what change to the policy constitutes a change to the >>>OID, it is less ambiguous to refer to the primary data. Once you are >> >>in >> >> >>>the business of managing state on a client, then there is negligible >>>cost increase incurred in managing several data points vs. a singe >> >>data >> >> >>>point. >>> >>>Trevor >>> >>>-----Original Message----- >>>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] >>>Sent: Tuesday, April 27, 2004 11:01 AM >>>To: Trevor Freeman >>>Cc: ietf-pkix@imc.org >>>Subject: Re: FW: scvp >>> >>>Trevor, >>> >>> >>> >>> >>>>HI Denis, >>>>In SCVP13, the concept of validation policy was not completely in >>>>alignment with 3379 definition. >>> >>> >>>Well, it is different and this is a major problem. >>> >>> >>> >>> >>>>It also blurred the distinction between >>>>validation policy and validation algorithm. >>> >>> >>>This is also a majo problem. >>> >>> >>> >>> >>>>Therefore I have defined >>>>what each is in SCVP 14 and aligned the structures accordingly. >>>>Section 1.3 has the following. >>>>"In SCVP, the validation policy comprises of an algorithm for >>>>certificate path processing and the input parameters to the >>> >>>algorithm." >>> >>>This does not comply with RFC 3379. >>> >>> >>> >>> >>>>Therefore trust anchors are an input into the algorithm and >>> > therefore > >>>>separate. >>> >>> >>>Therefore I do not follow you. >>> >>>From an interface point of view the simplest validation policy is >>>defined >>>by an OID where all the parameters necessary to perform the validation >>>check >>>are "behind" the OID. There is no need to provide any parameter >> >>through >> >> >>>the >>>interface. >>> >>>When there are some parameters, then they are specific to the >> >>validation >> >> >>>policy. I have no problem to provide specific parameters for what is >>>called >>>the "default validation policy" which is only a particular validation >>>policy >>>among many others. >>> >>>Simple clients will be unable to pass any parameter but will know >> >>which >> >> >>>OIDs >>>(for the validation policy) are appropriate for their applications. >>> >>>Denis >>> >>> >>> >>> >>>>This is in alignment with 3379s definition of validation policy. >>>> >>>>Trevor >>>> >>>>-----Original Message----- >>>>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] >>>>Sent: Monday, April 26, 2004 1:09 AM >>>>To: Peter Sylvester >>>>Cc: ietf-pkix@imc.org; Trevor Freeman >>>>Subject: Re: FW: scvp >>>> >>>>Peter and Trevor, >>>> >>>>I have major problems too. >>>> >>>> >>>> >>>> >>>> >>>>>in version 13 the syntax for a Query has been changed from >>>>> >>>>>Query ::= SEQUENCE { >>>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>>>> checks CertChecks, >>>>> wantBack WantBack, >>>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>>> valPolicy [1] ValidationPolicy OPTIONAL, >>>>> validityTime [2] GeneralizedTime OPTIONAL, >>>>> trustAnchors [3] TrustAnchors OPTIONAL, >>>>> intermediateCerts [4] CertBundle OPTIONAL, >>>>> revInfos [5] RevocationInfos OPTIONAL, >>>>> queryExtensions [6] Extensions OPTIONAL } >>>> >>>> >>>>In this structure trustAnchors was more or less an alternative to >>>>valPolicy. >>>> >>>>In fact, IF the valPolicy is the default policy, THEN additional >>>>parameters >>>>MUST be provided. So the structure below does not fit as well. >>>> >>>>This leads to have the two following elements: >>>> >>>> valPolicy ValidationPolicy, >>>> paramsDefValPolicy ParamsDefValPolicy OPTIONAL, >>>> >>>>where the first one is mandatory and the second one optional. >>>> >>>> >>>> >>>> >>>> >>>>>to >>>>> >>>>> Query ::= SEQUENCE { >>>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>>> >>>> >>>>> checks CertChecks, >>>>> wantBack WantBack, >>>>> requestRefHash BOOLEAN DEFAULT TRUE, >>>>> includePolResponce BOOLEAN DEFAULT FALSE, >>>>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>>>> requireExplicitPol BOOLEAN DEFAULT FALSE, >>>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, >>>>> valAlgorithm ValidationAlgorithm, >>>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>>> validityTime [1] GeneralizedTime OPTIONAL, >>>>> trustAnchors [2] TrustAnchors OPTIONAL, >>>>> intermediateCerts [3] CertBundle OPTIONAL, >>>>> revInfos [4] RevocationInfos OPTIONAL, >>>>> queryExtensions [5] Extensions OPTIONAL } >>>> >>>> >>>>I would thus propose to have something like: >>>> >>>>Query ::= SEQUENCE { >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF >>>>CertReference, >>>> checks CertChecks, >>>> wantBack WantBack, >>>> requestRefHash BOOLEAN DEFAULT TRUE, >>>> serverContextInfo OCTET STRING OPTIONAL, >>>> validityTime GeneralizedTime OPTIONAL, >>>> includePolResponse BOOLEAN DEFAULT FALSE, >>>> valPolicy ValidationPolicy, >>>> paramsDefValPolicy [0] ParamsDefValPolicy OPTIONAL, >>>> intermediateCerts [1] CertBundle OPTIONAL, >>>> revInfos [2] RevocationInfos OPTIONAL, >>>> queryExtensions [3] Extensions OPTIONAL } >>>> >>>>and then something like: >>>> >>>> ParamsDefValPolicy :: = SEQUENCE { >>>> trustAnchors TrustAnchors, >>>> endEntityCertificationPolicy SEQUENCE OF OBJECT IDENTIFIER >>>>OPTIONAL, >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>>> caCertificationPolicies SEQUENCE OF OBJECT IDENTIFIER >>>>OPTIONAL } >>>> >>>>Denis >>>> >>>> >>>> >>>> >>>> >>>>>I am not sure whether I am the only person unable to construct a >>>> >>>>parser. >>>> >>>> >>>> >>>> >>>>>in version 14 an aditional flag has been added which is not >>>> > available > >>>>in the >>>> >>>> >>>> >>>> >>>>>Chapter 9. Is the isCA flag an orthogonal attribut to other >>>> >>validation >> >> >>>>>algorithms, or just to some of them, e.g,. the default name matching >>>> >>>>one? >>>> >>>> >>>> >>>> >>>>> Query ::= SEQUENCE { >>>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>>> >>>> >>>>> checks CertChecks, >>>>> wantBack WantBack, >>>>> requestRefHash BOOLEAN DEFAULT TRUE, >>>>> includePolResponce BOOLEAN DEFAULT FALSE, >>>>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>>>> requireExplicitPol BOOLEAN DEFAULT FALSE, >>>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, >>>>> isCA BOOLEAN DEFAULT FALSE, >>>>> valAlgorithm ValidationAlgorithm, >>>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>>> validityTime [1] GeneralizedTime OPTIONAL, >>>>> trustAnchors [2] TrustAnchors OPTIONAL, >>>>> intermediateCerts [3] CertBundle OPTIONAL, >>>>> revInfos [4] RevocationInfos OPTIONAL, >>>>> keyUsage [5] KeyUsage OPTIONAL, >>>>> extendedKeyUsage [6] ExtKeyUsageSyntax OPTIONAL, >>>>> queryExtensions [7] Extensions OPTIONAL >>>>> producedAt [8] GeneralizedTime OPTIONAL} >>>>> >>>>> >>>> >>>> >>>> >>>> >>> >> >> > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46Ed695072830; Thu, 6 May 2004 07:39:06 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46Ed6gs072829; Thu, 6 May 2004 07:39:06 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from brmea-mail-4.sun.com (brmea-mail-4.Sun.COM [192.18.98.36]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46Ed5kc072823 for <ietf-pkix@imc.org>; Thu, 6 May 2004 07:39:05 -0700 (PDT) (envelope-from Steve.Hanna@Sun.COM) Received: from phys-bur1-1 ([129.148.13.15]) by brmea-mail-4.sun.com (8.12.10/8.12.9) with ESMTP id i46EbosT003674 for <ietf-pkix@imc.org>; Thu, 6 May 2004 08:37:50 -0600 (MDT) Received: from conversion-daemon.bur-mail2.east.sun.com by bur-mail2.east.sun.com (iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003)) id <0HXA00801RAFMU@bur-mail2.east.sun.com> for ietf-pkix@imc.org; Thu, 06 May 2004 10:39:07 -0400 (EDT) Received: from sun.com (dhcp-ubur02-75-154.East.Sun.COM [129.148.75.154]) by bur-mail2.east.sun.com (iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003)) with ESMTPA id <0HXA0049LRD7M8@bur-mail2.east.sun.com>; Thu, 06 May 2004 10:39:07 -0400 (EDT) Date: Thu, 06 May 2004 10:35:08 -0400 From: Steve Hanna <Steve.Hanna@Sun.COM> Subject: Re: RFC 3280 bug in name constraints In-reply-to: <0C3042E92D8A714783E2C44AB9936E1DF502D1@EUR-MSG-03.europe.corp.microsoft.com> To: Stefan Santesson <stefans@microsoft.com> Cc: ietf-pkix@imc.org Message-id: <409A4D1C.7070108@sun.com> MIME-version: 1.0 Content-type: multipart/signed; protocol="application/x-pkcs7-signature"; micalg=sha1; boundary=------------ms090804020804070103010908 X-Accept-Language: en-us, en User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20031111 References: <0C3042E92D8A714783E2C44AB9936E1DF502D1@EUR-MSG-03.europe.corp.microsoft.com> Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> --------------ms090804020804070103010908 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit I agree. In fact, I don't see why you wouldn't apply name constraints to an EmailAddress attribute in a subject DN even if there *is* an rfc822Name in a subject alternative name. I suggest that the text be revised to say: An rfc822 name constraint MUST be applied to an attribute of type EmailAddress in the subject distinguished name. However, I'm willing to back down on this if people feel it's too strict. My concern is that a certificate with an email address in the subject DN prohibited by name constraints will be allowed because it also contains a permissible email address in the subject alternative name. I suspect many applications will not understand this and will accept the email address in the subject DN. In fact, RFC 2632 and its successor don't contain any warning about this. Thanks, Steve Stefan Santesson wrote: > If someone is keeping record of bugs for future update of RFC 3280 I > have a small one to file about name constraints. > > > > In section 4.2.1.11, starting at bottom of page 38: > > > > When rfc822 names are constrained, but the > > certificate does not include a subject alternative name, the rfc822 > > name constraint MUST be applied to the attribute of type EmailAddress > > in the subject distinguished name. > > > > Suppose/suggest that the intended meaning is: > > > > When rfc822 names are constrained, but the certificate does not > > include an rfc822Name in a subject alternative name extension, the rfc822 > > name constraint MUST be applied to the attribute of type EmailAddress > > in the subject distinguished name. > > > > > > Rationale: > > Why would the constraint NOT apply to EmailAddress just because there is > a SubjAltName extension with e.g. just some unrelated other name > component present? > > > > /Stefan > > > --------------ms090804020804070103010908 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJNzCC AvYwggJfoAMCAQICAww3ZDANBgkqhkiG9w0BAQQFADBiMQswCQYDVQQGEwJaQTElMCMGA1UE ChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNv bmFsIEZyZWVtYWlsIElzc3VpbmcgQ0EwHhcNMDQwNDI4MjEwMTE0WhcNMDUwNDI4MjEwMTE0 WjBoMQ4wDAYDVQQEEwVIYW5uYTEVMBMGA1UEKhMMU3RlcGhlbiBSb3NzMRswGQYDVQQDExJT dGVwaGVuIFJvc3MgSGFubmExIjAgBgkqhkiG9w0BCQEWE3N0ZXZlLmhhbm5hQHN1bi5jb20w ggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC11rLYp70KnKGmV6pNTFRhWBwq47mV 4hU6EaZxo6I3Vw96Rd3E4TH3SAthIl5N4/tpiMPtlTBiJOU6QHSlz1DgTCclmTK6taqtEyLq amez8vYgSdOEnOFRYILptjbEwpE+yeUmFKOmZ7zSwBMqRQatMHMaLVUJJ3Ygw1V7fKSwvVU0 OJ8c+c3sJRtRyDLxq/vMzYQP+wyqRqlJPTHGFqVuMp0FZIV5X76z+O+W90m9jXjirz91Jz+W N6rwqmPi8Qy0uxBKm6NKT1371p+WYaK9OfGk8vGbJ8ZsTP3DyQhuuemYPbkKPtoII6WnHIGT /DOy2G7gff7GOIhUrs7PziddAgMBAAGjMDAuMB4GA1UdEQQXMBWBE3N0ZXZlLmhhbm5hQHN1 bi5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQB++aOP6sZVYlqBGc92hZ61 FHST+OqnKZHRGkqEO94RoxdeaSUFKVO2IaKLMBmqNzmh3N3pIsIKiLhJNvDgLhlRrMRVfR29 uixjTsprR26AacMx/gCiCstGIjkzTb83xAvZay5hnzUmlt/LSNX8e6jTNav6LP64WC+C7Tul fM/PzDCCAvYwggJfoAMCAQICAww3ZDANBgkqhkiG9w0BAQQFADBiMQswCQYDVQQGEwJaQTEl MCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3Rl IFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0EwHhcNMDQwNDI4MjEwMTE0WhcNMDUwNDI4 MjEwMTE0WjBoMQ4wDAYDVQQEEwVIYW5uYTEVMBMGA1UEKhMMU3RlcGhlbiBSb3NzMRswGQYD VQQDExJTdGVwaGVuIFJvc3MgSGFubmExIjAgBgkqhkiG9w0BCQEWE3N0ZXZlLmhhbm5hQHN1 bi5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC11rLYp70KnKGmV6pNTFRh WBwq47mV4hU6EaZxo6I3Vw96Rd3E4TH3SAthIl5N4/tpiMPtlTBiJOU6QHSlz1DgTCclmTK6 taqtEyLqamez8vYgSdOEnOFRYILptjbEwpE+yeUmFKOmZ7zSwBMqRQatMHMaLVUJJ3Ygw1V7 fKSwvVU0OJ8c+c3sJRtRyDLxq/vMzYQP+wyqRqlJPTHGFqVuMp0FZIV5X76z+O+W90m9jXji rz91Jz+WN6rwqmPi8Qy0uxBKm6NKT1371p+WYaK9OfGk8vGbJ8ZsTP3DyQhuuemYPbkKPtoI I6WnHIGT/DOy2G7gff7GOIhUrs7PziddAgMBAAGjMDAuMB4GA1UdEQQXMBWBE3N0ZXZlLmhh bm5hQHN1bi5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQB++aOP6sZVYlqB Gc92hZ61FHST+OqnKZHRGkqEO94RoxdeaSUFKVO2IaKLMBmqNzmh3N3pIsIKiLhJNvDgLhlR rMRVfR29uixjTsprR26AacMx/gCiCstGIjkzTb83xAvZay5hnzUmlt/LSNX8e6jTNav6LP64 WC+C7TulfM/PzDCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAwgdExCzAJBgNVBAYT AlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUgVG93bjEaMBgGA1UE ChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRpb24gU2VydmljZXMg RGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBDQTErMCkGCSqG SIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0wMzA3MTcwMDAwMDBa Fw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3Vs dGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNz dWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV+065yplaHmjAdQRw nd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfArhVqqP3FWy688Cwfn 8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/p7bRPGEEQB5kGXJg t/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8MDowOKA2oDSGMmh0 dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWlsQ0EuY3JsMAsGA1Ud DwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxhYmVsMi0xMzgwDQYJ KoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/TCG4+DYfqi2fNi/A 9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amcOY6MIE9lX5Xa9/eH 1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggM7MIIDNwIBATBpMGIxCzAJ BgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYD VQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIDDDdkMAkGBSsOAwIa BQCgggGnMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcNAQkFMQ8XDTA0MDUw NjE0MzUwOFowIwYJKoZIhvcNAQkEMRYEFMXbLgLcoPULUY0MDEWbxDKp9z02MFIGCSqGSIb3 DQEJDzFFMEMwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcG BSsOAwIHMA0GCCqGSIb3DQMCAgEoMHgGCSsGAQQBgjcQBDFrMGkwYjELMAkGA1UEBhMCWkEx JTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0 ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBAgMMN2QwegYLKoZIhvcNAQkQAgsxa6Bp MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQu MSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIDDDdkMA0G CSqGSIb3DQEBAQUABIIBAGPnexS7TZ+dns25SGf2I/WPXo7Zl58BF+nRK0c5jxw53cUrpO1x eN9rkhM57QkUs2hw7wpdlB6fSp+wm/VFy7pVYoXQ+UwOD1tqlK9wRUtdddYykFpICxUsI9fP ZTbTi/etcpE6Wy0DAgkyZGwPr6DJfsmOnS/KqLQSt7C9DPFdiJPdUAc6CJwgvRhgdFm9vsgD t2uzSeUN6/qIDz8/TMSE2BBeFJAGULMS6N9wOlAHv6AVFuZNCmufWtpTxlqCMbJKVIBlwKa+ Ea4Y0AWh3giA0Qbfy3LvNWdTEm4+VL6U3fKEPHYGpN7KIEG3sNbSNn09IiuoHZPX4HI/Vct5 DOEAAAAAAAA= --------------ms090804020804070103010908-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46ETEeG072099; Thu, 6 May 2004 07:29:15 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46ETEEm072098; Thu, 6 May 2004 07:29:14 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from brmea-mail-4.sun.com (brmea-mail-4.Sun.COM [192.18.98.36]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46ETEUQ072090 for <ietf-pkix@imc.org>; Thu, 6 May 2004 07:29:14 -0700 (PDT) (envelope-from Steve.Hanna@Sun.COM) Received: from phys-bur1-1 ([129.148.13.15]) by brmea-mail-4.sun.com (8.12.10/8.12.9) with ESMTP id i46ERusb029078 for <ietf-pkix@imc.org>; Thu, 6 May 2004 08:27:58 -0600 (MDT) Received: from conversion-daemon.bur-mail2.east.sun.com by bur-mail2.east.sun.com (iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003)) id <0HXA00401QP3LN@bur-mail2.east.sun.com> for ietf-pkix@imc.org; Thu, 06 May 2004 10:29:14 -0400 (EDT) Received: from sun.com (dhcp-ubur02-75-154.East.Sun.COM [129.148.75.154]) by bur-mail2.east.sun.com (iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003)) with ESMTPA id <0HXA004IIQWQM8@bur-mail2.east.sun.com>; Thu, 06 May 2004 10:29:14 -0400 (EDT) Date: Thu, 06 May 2004 10:25:15 -0400 From: Steve Hanna <Steve.Hanna@Sun.COM> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs In-reply-to: <0C3042E92D8A714783E2C44AB9936E1DF823DD@EUR-MSG-03.europe.corp.microsoft.com> To: Stefan Santesson <stefans@microsoft.com> Cc: Wen-Cheng Wang <wcwang@cht.com.tw>, Stephen Kent <kent@bbn.com>, Santoni Adriano <adriano.santoni@actalis.it>, pgut001@cs.auckland.ac.nz, ietf-pkix@imc.org Message-id: <409A4ACB.2040709@sun.com> MIME-version: 1.0 Content-type: multipart/signed; protocol="application/x-pkcs7-signature"; micalg=sha1; boundary=------------ms020005020303050707030801 X-Accept-Language: en-us, en User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20031111 References: <0C3042E92D8A714783E2C44AB9936E1DF823DD@EUR-MSG-03.europe.corp.microsoft.com> Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> --------------ms020005020303050707030801 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit A better solution is to require (or at least recommend) the ability to properly compare PrintableString and UTF8String encodings. Paul Hoffman and I are working on an Internet Draft describing how to do this (based on the IDN and LDAP documents on the same topic). This handles name constraints earlier in the path, etc. Thanks, Steve Stefan Santesson wrote: > Name rollover certificates only takes care of CA name changes. > It doesn't solve the problem if subject names have mixed encoding in the > forest of user certificates, while the CAs names are intact. > > Neither does it solve when the constraint is placed above the CA > rollover. > > Finally, CA name rollover will most likely bust path validation using > other constraining mechanisms because they may just not be recognized as > self issued due to different name encoding. > > Stefan Santesson > Program Manager, Windows Security > > >>-----Original Message----- >>From: owner-ietf-pkix@mail.imc.org > > [mailto:owner-ietf-pkix@mail.imc.org] > >>On Behalf Of Wen-Cheng Wang >>Sent: den 5 maj 2004 23:40 >>To: Stefan Santesson; Stephen Kent; Santoni Adriano >>Cc: pgut001@cs.auckland.ac.nz; ietf-pkix@imc.org >>Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in > > encoding > >>RDNs >> >> >>RFC 3280 does not ask CAs to replace all certificates at once for >>migrating to UTF8String encoding. Instead, it suggest that >> "CAs MAY issue "name rollover" certificates to support an >> orderly migration to UTF8String encoding." >> >>I believe that if relying parties use 3280-compliant certification > > path > >>validation modules, then they will not have problem for processing >>"name rollover" certificates, and thus the migration to UTF8String >>encoding will not be a problem. >> >>One thing should be noted: a "name rollover" certificate is > > essentially > >>a "self-issued certificate" but its issuer name and subject name is in >>different encoding. Therefore, the definition of "self-issued > > certificate" > >>should be changed a little. Also, since a "name rollover" certificate > > may > >>contains a NameConstraints extension in which the permittedSubtrees >>and excludedSubtrees will be in a new encoding (e.g., UTF8String), >>the path validation in RFC 3280 also need to be revised. I hope all of >>these should be addressed in sonOfRfc3280. >> >>I don't think that RFC 3280 mandates the support for UTF8String >>encoding is "at odds with reality". There are many countries/regions >>in the world where it is impossible to encode names in Printable > > String. > >>In Taiwan, we have already millions of certificates with issuer names >>and subject names encoded in UTF8String. >> >>Three years ago, I tested some VPN devices and found most of them >>can not accept certificates with UTF8String (One of them even crashed >>if received a certificate with UTF8String). Recently, I test many VPN >>devices >>again and amazingly found the all of them can normally handle > > certificates > >>with >>UTF8String. All of them can chain and validate certificte paths that >>contains >>certificates with UTF8String. (Although, most of them still can not >>display >>the >>UTF8String in Chinese correctly. It is a pity.) >> >>The truth is that there are more and more applications that can > > handling > >>UTF8String encoding. I think the credit should give to RFC 3280 for > > its > >>mandating for UTF8String encoding. >> >>Wen-Cheng Wang >> >> >>----- Original Message ----- >>From: "Stefan Santesson" <stefans@microsoft.com> >>To: "Stephen Kent" <kent@bbn.com>; "Santoni Adriano" >><adriano.santoni@actalis.it> >>Cc: <pgut001@cs.auckland.ac.nz>; <ietf-pkix@imc.org> >>Sent: Thursday, May 06, 2004 8:18 AM >>Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in > > encoding > >>RDNs >> >> >> >>> >>>Steve, >>> >>>There is one unfortunate side effect that makes this requirement > > worse > >>>than other evolutions of standards. >>> >>>It is almost never feasible to replace all certificates in a > > hierarchy > >>>at once. This means that you over time will have a mixed environment >>>where e.g. the same organization name will exist in different > > encoding > >>>in many places of your hierarchy. >>> >>>It will then be virtually impossible to apply name constraining in > > such > >>>environment because RFC 3280 does not mandate the full X.500 name >>>matching rules and a client may subsequently decide that names with >>>different encoding is a no-match. >>> >>>This seams to be an unsolvable problem that, looking in the mirror, > > is > >>>inadequately addressed and shouldn't have passed in its current > > form. In > >>>fact it is probably a defect that needs to be addressed in > > sonOfRfc3280 > >>>in one way or the other. >>> >>>/Stefan >>> >>> >>> >>>>-----Original Message----- >>>>From: owner-ietf-pkix@mail.imc.org >>> >>>[mailto:owner-ietf-pkix@mail.imc.org] >>> >>>>On Behalf Of Stephen Kent >>>>Sent: den 5 maj 2004 13:36 >>>>To: Santoni Adriano >>>>Cc: 'pgut001@cs.auckland.ac.nz'; ietf-pkix@imc.org >>>>Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in >>> >>>encoding >>> >>>>RDNs >>>> >>>> >>>>At 9:54 AM +0200 4/23/04, Santoni Adriano wrote: >>>> >>>>>A rule that is "at odds with reality" should not be expressed in >>> >>>terms of >>> >>>>>MUST, but rather in terms of SHOULD. I wonder how it came that > > RFC > >>>3280 >>> >>>>>reached the status of an official IETF standard, if it contains > > one > >>>or >>> >>>>more >>>> >>>>>areas that are "at odds with reality". >>>>> >>>>>I suppose the answer is one of these: >>>>>- RFC 3280 does not really contain areas that are "at odds with >>> >>>reality" >>> >>>>- >>>> >>>>>it's just that we do not understand them; >>>> >>>>RFC 3280 established a requirement for a change in how DNs are >>>>encoded, for a future point in time. Thus, at the time the RFC was >>>>published as a standard of course it did not match "reality." > > since > >>>>it was calling for a change. By the way, this change was not the >>>>idea of the PKIX WG but rather is consistent with the overall IETF >>>>mandate to better accommodate character sets for a wide range of >>>>languages. >>>> >>>> >>>>>- containing areas that are "at odds with reality" is not an > > obstacle > >>>to >>> >>>>>becoming a standard; >>>> >>>>Think of the implication of your comment. if we were never able to >>>>call for a change in how things are done when we update a > > standard, > >>>>then we could never make any changes. obviously this would be > > unduly > >>>>limiting. >>>> >>>> >>>>>- expressions like "MUST" in IETF standard are really to be taken > > as > >>>mere >>> >>>>>suggestions; >>>> >>>>IETF standards are voluntary. failure to comply is not punishable > > :-) > >>>>>- nobody really cares. >>>> >>>>one might say that in environments where a few vendors or service >>>>providers control a significant market share, their deviation from >>>>standards, or failure to accommodate changes in standards, can > > have > >>>>profound impacts. >>>> >>>>Steve >>> >>> > > --------------ms020005020303050707030801 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJNzCC AvYwggJfoAMCAQICAww3ZDANBgkqhkiG9w0BAQQFADBiMQswCQYDVQQGEwJaQTElMCMGA1UE ChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNv bmFsIEZyZWVtYWlsIElzc3VpbmcgQ0EwHhcNMDQwNDI4MjEwMTE0WhcNMDUwNDI4MjEwMTE0 WjBoMQ4wDAYDVQQEEwVIYW5uYTEVMBMGA1UEKhMMU3RlcGhlbiBSb3NzMRswGQYDVQQDExJT dGVwaGVuIFJvc3MgSGFubmExIjAgBgkqhkiG9w0BCQEWE3N0ZXZlLmhhbm5hQHN1bi5jb20w ggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC11rLYp70KnKGmV6pNTFRhWBwq47mV 4hU6EaZxo6I3Vw96Rd3E4TH3SAthIl5N4/tpiMPtlTBiJOU6QHSlz1DgTCclmTK6taqtEyLq amez8vYgSdOEnOFRYILptjbEwpE+yeUmFKOmZ7zSwBMqRQatMHMaLVUJJ3Ygw1V7fKSwvVU0 OJ8c+c3sJRtRyDLxq/vMzYQP+wyqRqlJPTHGFqVuMp0FZIV5X76z+O+W90m9jXjirz91Jz+W N6rwqmPi8Qy0uxBKm6NKT1371p+WYaK9OfGk8vGbJ8ZsTP3DyQhuuemYPbkKPtoII6WnHIGT /DOy2G7gff7GOIhUrs7PziddAgMBAAGjMDAuMB4GA1UdEQQXMBWBE3N0ZXZlLmhhbm5hQHN1 bi5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQB++aOP6sZVYlqBGc92hZ61 FHST+OqnKZHRGkqEO94RoxdeaSUFKVO2IaKLMBmqNzmh3N3pIsIKiLhJNvDgLhlRrMRVfR29 uixjTsprR26AacMx/gCiCstGIjkzTb83xAvZay5hnzUmlt/LSNX8e6jTNav6LP64WC+C7Tul fM/PzDCCAvYwggJfoAMCAQICAww3ZDANBgkqhkiG9w0BAQQFADBiMQswCQYDVQQGEwJaQTEl MCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3Rl IFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0EwHhcNMDQwNDI4MjEwMTE0WhcNMDUwNDI4 MjEwMTE0WjBoMQ4wDAYDVQQEEwVIYW5uYTEVMBMGA1UEKhMMU3RlcGhlbiBSb3NzMRswGQYD VQQDExJTdGVwaGVuIFJvc3MgSGFubmExIjAgBgkqhkiG9w0BCQEWE3N0ZXZlLmhhbm5hQHN1 bi5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC11rLYp70KnKGmV6pNTFRh WBwq47mV4hU6EaZxo6I3Vw96Rd3E4TH3SAthIl5N4/tpiMPtlTBiJOU6QHSlz1DgTCclmTK6 taqtEyLqamez8vYgSdOEnOFRYILptjbEwpE+yeUmFKOmZ7zSwBMqRQatMHMaLVUJJ3Ygw1V7 fKSwvVU0OJ8c+c3sJRtRyDLxq/vMzYQP+wyqRqlJPTHGFqVuMp0FZIV5X76z+O+W90m9jXji rz91Jz+WN6rwqmPi8Qy0uxBKm6NKT1371p+WYaK9OfGk8vGbJ8ZsTP3DyQhuuemYPbkKPtoI I6WnHIGT/DOy2G7gff7GOIhUrs7PziddAgMBAAGjMDAuMB4GA1UdEQQXMBWBE3N0ZXZlLmhh bm5hQHN1bi5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQB++aOP6sZVYlqB Gc92hZ61FHST+OqnKZHRGkqEO94RoxdeaSUFKVO2IaKLMBmqNzmh3N3pIsIKiLhJNvDgLhlR rMRVfR29uixjTsprR26AacMx/gCiCstGIjkzTb83xAvZay5hnzUmlt/LSNX8e6jTNav6LP64 WC+C7TulfM/PzDCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAwgdExCzAJBgNVBAYT AlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUgVG93bjEaMBgGA1UE ChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRpb24gU2VydmljZXMg RGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBDQTErMCkGCSqG SIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0wMzA3MTcwMDAwMDBa Fw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3Vs dGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNz dWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV+065yplaHmjAdQRw nd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfArhVqqP3FWy688Cwfn 8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/p7bRPGEEQB5kGXJg t/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8MDowOKA2oDSGMmh0 dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWlsQ0EuY3JsMAsGA1Ud DwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxhYmVsMi0xMzgwDQYJ KoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/TCG4+DYfqi2fNi/A 9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amcOY6MIE9lX5Xa9/eH 1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggM7MIIDNwIBATBpMGIxCzAJ BgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYD VQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIDDDdkMAkGBSsOAwIa BQCgggGnMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcNAQkFMQ8XDTA0MDUw NjE0MjUxNVowIwYJKoZIhvcNAQkEMRYEFCv0SAj248yHUuVfubR3/yua1DueMFIGCSqGSIb3 DQEJDzFFMEMwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcG BSsOAwIHMA0GCCqGSIb3DQMCAgEoMHgGCSsGAQQBgjcQBDFrMGkwYjELMAkGA1UEBhMCWkEx JTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0 ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBAgMMN2QwegYLKoZIhvcNAQkQAgsxa6Bp MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQu MSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIDDDdkMA0G CSqGSIb3DQEBAQUABIIBAGMx4RAv3/TaCX2+yZuN1WpreVhqPkLUmzdU3cL3d1Tf75prfL46 jJzEfSO8wlPkJ/DnBDn+V2ZzZ4Zd2pCKUvhU4d/y8kL/TNajp8c2TseF05ZeoEWSQhx2mS9t Sh1vQ3Ha1C12E2Gvzdt5iLFfFgVO73DvC775gFIXuhX3QqvB/r9vjZmglx89iNFscgzsarUv ASvXrNowpzlfAJ/JDiZCfageZmtqGgwYuSRW3/zSEO/YCBNY281vksm76bJKWBOcMOtFBc1Y yMAe2qwltQEOv03UhlkH568aJyviAeCrUEbKhiHSw6dVFMhb4h8Qew6HgrEONNJRWpiAAwkZ CVoAAAAAAAA= --------------ms020005020303050707030801-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46DX1X1067689; Thu, 6 May 2004 06:33:01 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46DX15E067687; Thu, 6 May 2004 06:33:01 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-eur.microsoft.com (mail-eur.microsoft.com [213.199.128.145]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46DWpCW067641 for <ietf-pkix@imc.org>; Thu, 6 May 2004 06:32:51 -0700 (PDT) (envelope-from stefans@microsoft.com) Received: from eur-imc-02.europe.corp.microsoft.com ([65.53.196.43]) by mail-eur.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 14:32:43 +0100 Received: from 65.53.196.43 by eur-imc-02.europe.corp.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 06 May 2004 14:32:43 +0100 Received: from EUR-MSG-03.europe.corp.microsoft.com ([65.53.192.44]) by eur-imc-02.europe.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 14:32:43 +0100 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Date: Thu, 6 May 2004 14:32:41 +0100 Message-ID: <0C3042E92D8A714783E2C44AB9936E1DF823DD@EUR-MSG-03.europe.corp.microsoft.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Thread-Index: AcQzPxL5hMiEJesQTHKpNAdD7VavPAALqcBA From: "Stefan Santesson" <stefans@microsoft.com> To: "Wen-Cheng Wang" <wcwang@cht.com.tw>, "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 06 May 2004 13:32:43.0259 (UTC) FILETIME=[9C2954B0:01C4336E] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i46DWpCW067669 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Name rollover certificates only takes care of CA name changes. It doesn't solve the problem if subject names have mixed encoding in the forest of user certificates, while the CAs names are intact. Neither does it solve when the constraint is placed above the CA rollover. Finally, CA name rollover will most likely bust path validation using other constraining mechanisms because they may just not be recognized as self issued due to different name encoding. Stefan Santesson Program Manager, Windows Security > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On Behalf Of Wen-Cheng Wang > Sent: den 5 maj 2004 23:40 > To: Stefan Santesson; Stephen Kent; Santoni Adriano > Cc: pgut001@cs.auckland.ac.nz; ietf-pkix@imc.org > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding > RDNs > > > RFC 3280 does not ask CAs to replace all certificates at once for > migrating to UTF8String encoding. Instead, it suggest that > "CAs MAY issue "name rollover" certificates to support an > orderly migration to UTF8String encoding." > > I believe that if relying parties use 3280-compliant certification path > validation modules, then they will not have problem for processing > "name rollover" certificates, and thus the migration to UTF8String > encoding will not be a problem. > > One thing should be noted: a "name rollover" certificate is essentially > a "self-issued certificate" but its issuer name and subject name is in > different encoding. Therefore, the definition of "self-issued certificate" > should be changed a little. Also, since a "name rollover" certificate may > contains a NameConstraints extension in which the permittedSubtrees > and excludedSubtrees will be in a new encoding (e.g., UTF8String), > the path validation in RFC 3280 also need to be revised. I hope all of > these should be addressed in sonOfRfc3280. > > I don't think that RFC 3280 mandates the support for UTF8String > encoding is "at odds with reality". There are many countries/regions > in the world where it is impossible to encode names in Printable String. > In Taiwan, we have already millions of certificates with issuer names > and subject names encoded in UTF8String. > > Three years ago, I tested some VPN devices and found most of them > can not accept certificates with UTF8String (One of them even crashed > if received a certificate with UTF8String). Recently, I test many VPN > devices > again and amazingly found the all of them can normally handle certificates > with > UTF8String. All of them can chain and validate certificte paths that > contains > certificates with UTF8String. (Although, most of them still can not > display > the > UTF8String in Chinese correctly. It is a pity.) > > The truth is that there are more and more applications that can handling > UTF8String encoding. I think the credit should give to RFC 3280 for its > mandating for UTF8String encoding. > > Wen-Cheng Wang > > > ----- Original Message ----- > From: "Stefan Santesson" <stefans@microsoft.com> > To: "Stephen Kent" <kent@bbn.com>; "Santoni Adriano" > <adriano.santoni@actalis.it> > Cc: <pgut001@cs.auckland.ac.nz>; <ietf-pkix@imc.org> > Sent: Thursday, May 06, 2004 8:18 AM > Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding > RDNs > > > > > > > > Steve, > > > > There is one unfortunate side effect that makes this requirement worse > > than other evolutions of standards. > > > > It is almost never feasible to replace all certificates in a hierarchy > > at once. This means that you over time will have a mixed environment > > where e.g. the same organization name will exist in different encoding > > in many places of your hierarchy. > > > > It will then be virtually impossible to apply name constraining in such > > environment because RFC 3280 does not mandate the full X.500 name > > matching rules and a client may subsequently decide that names with > > different encoding is a no-match. > > > > This seams to be an unsolvable problem that, looking in the mirror, is > > inadequately addressed and shouldn't have passed in its current form. In > > fact it is probably a defect that needs to be addressed in sonOfRfc3280 > > in one way or the other. > > > > /Stefan > > > > > > > -----Original Message----- > > > From: owner-ietf-pkix@mail.imc.org > > [mailto:owner-ietf-pkix@mail.imc.org] > > > On Behalf Of Stephen Kent > > > Sent: den 5 maj 2004 13:36 > > > To: Santoni Adriano > > > Cc: 'pgut001@cs.auckland.ac.nz'; ietf-pkix@imc.org > > > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in > > encoding > > > RDNs > > > > > > > > > At 9:54 AM +0200 4/23/04, Santoni Adriano wrote: > > > >A rule that is "at odds with reality" should not be expressed in > > terms of > > > >MUST, but rather in terms of SHOULD. I wonder how it came that RFC > > 3280 > > > >reached the status of an official IETF standard, if it contains one > > or > > > more > > > >areas that are "at odds with reality". > > > > > > > >I suppose the answer is one of these: > > > >- RFC 3280 does not really contain areas that are "at odds with > > reality" > > > - > > > >it's just that we do not understand them; > > > > > > RFC 3280 established a requirement for a change in how DNs are > > > encoded, for a future point in time. Thus, at the time the RFC was > > > published as a standard of course it did not match "reality." since > > > it was calling for a change. By the way, this change was not the > > > idea of the PKIX WG but rather is consistent with the overall IETF > > > mandate to better accommodate character sets for a wide range of > > > languages. > > > > > > >- containing areas that are "at odds with reality" is not an obstacle > > to > > > >becoming a standard; > > > > > > Think of the implication of your comment. if we were never able to > > > call for a change in how things are done when we update a standard, > > > then we could never make any changes. obviously this would be unduly > > > limiting. > > > > > > >- expressions like "MUST" in IETF standard are really to be taken as > > mere > > > >suggestions; > > > > > > IETF standards are voluntary. failure to comply is not punishable :-) > > > > > > >- nobody really cares. > > > > > > one might say that in environments where a few vendors or service > > > providers control a significant market share, their deviation from > > > standards, or failure to accommodate changes in standards, can have > > > profound impacts. > > > > > > Steve > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46Aarir056597; Thu, 6 May 2004 03:36:53 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i46AarbV056596; Thu, 6 May 2004 03:36:53 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i46Aakxc056586 for <ietf-pkix@imc.org>; Thu, 6 May 2004 03:36:52 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms21.chttl.com.tw (ms21 [10.144.2.121]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i46AZwGG022514; Thu, 6 May 2004 18:35:58 +0800 Received: (from root@localhost) by ms21.chttl.com.tw (8.12.10/8.12.10) id i46AZxxf001862; Thu, 6 May 2004 18:35:59 +0800 Received: from wcwang ([10.144.133.79]) by ms21.chttl.com.tw (8.12.10/8.12.10) with SMTP id i46AZvL5001688; Thu, 6 May 2004 18:35:57 +0800 Message-ID: <020a01c43355$e95e13b0$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Trevor Freeman" <trevorf@exchange.microsoft.com>, "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> References: <33E7A1613A3480448996047B69308A2F02713A06@df-grommit-01.dogfood> Subject: Re: FW: scvp Date: Thu, 6 May 2004 18:35:54 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> I don't agree that the broader consensus is for a "al a carte". (Maybe we should have a straw poll to decide what is the broader consensus?) When designing a DPV protocol, please do not forget that the two main benefits to establish a DPV server are: 1. it reduces the burden for application implementators to implement certification path processing modules by themself; 2. it allows validation against management-defined validation policies in a consistent fashion across an application domain (e.g., an enterprise). I believe that application implementors do not want "al a carte" combinations, they will prefer suits instead. If a DPV server requires application implementors to use a DPV client API that needs the implementors to specify a validation policy OID with many initial mystery parameters, then what is the benefit to use the DPV server? Why not simply use a certification path processing API such as CML instead a DPV client API if the burden is the same? At least, you can save the budget for establishing a DPV server if you choose to use CML. Besides, in an application domain where the authority wish to force all applications to validate certification path against management-defined validation policies in a consistent fashion, the applications will not even be allowed to specify any initial parameters (except the certificate to be validated). If the client side is allows to specfy its own initial parameters, how can the validation be guaranteed to proceed in the consistent fashion? In an application domain with a centralized validation policy, the server side will decide almost all the initial parameters according to the validation policy. In such case, the client side need not (and might not be allowed) to specify any initial parameters (except the certificate to be validated, of course). I believe that in most use cases of DPV, the client side will need not to specify any initial parameters. Therefore, why not simplify the syntax of the SCVP so that the client side do not to specify any initial parameters in a basic request. In my opinion, the SCVP is not a "simple" protocol anymore. Let's simplify it so that it deserve the name of a "simple" protocol. That is not to say that the SCVP can not provide a mechanism for the client side to specify its own validation policy and initial parameters if the server side allows its clients to do that, but this can be achieved by using the "Extension" mechanism. In the most case, there is no need for the client side to specify its own validation policy and initial parameters, so please do not let the basic syntax so complicated. Personally, I think the draft 12 syntax is better than draft 13/14. Maybe we should go back to the draft 12 syntax and try to define some extensions based on that syntax. Wen-Cheng Wang ----- Original Message ----- From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> Sent: Wednesday, May 05, 2004 1:22 AM Subject: RE: FW: scvp > > Hi Denis, > There is no mandate in 3379 that a validation policy must be expressed > with a single value. It is not clear that is any value is doing so since > it only raised more ambiguity. This is well trod turf by many groups who > have debated the merits of providing suits vs. "al a carte" combinations > of inputs and the broader consensus is for a "al a carte". > Trevor > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Tuesday, May 04, 2004 3:15 AM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: Re: FW: scvp > > Trevor, > > > Hi Denis, > > > > 3379 does not require that the validation policy be specified in a > > single value. > > You are playing with words. The text says: > "most clients will simply reference a validation policy" > > This means that validation policies can be referenced and thus > parameters do > not need to be defined through this interface. > > A simple and good reference is an OID. > > > With SCVP14 a client can accept the default of the SCVP > > server or specify a set of parameters which constitutes its validation > > policy. That is consistent with the requirements of 3379. > > The specification of the set of parameters should be done through a > different interface. This different interface can then return an OID > Whatever this additional interface can be, it will never be rich enough > to > pass all the details of the validation policy. Thus OIDs can also be > defined > not using this additional interface. > > Denis > > > Trevor > > > > -----Original Message----- > > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > > Sent: Thursday, April 29, 2004 3:54 AM > > To: Trevor Freeman > > Cc: ietf-pkix@imc.org > > Subject: Re: FW: scvp > > > > Trevor, > > > > > >>Hi Denis, > > > > > >>Can you please be more specific how you think SCVP 14 does not comply > >>with 3379. > >> > >>I can find no section in 3379 where is there a requirement that a > >>validation policy MUST be represented as an OID. > > > > > > There cannot be a requirement with such a level of detail, but the > text > > states: > > > > The protocol MUST allow the client to include > > these policy dependant parameters in the DPV request; however, it > is > > expected that most clients will simply reference a validation > policy > > for a given application or accept the DPV server's default > > validation > > policy. > > > > A simple reference is an OID. > > > > FYI, I do not expect to use the "default validation policy". > > > > Denis > > > > > > > >>Given hiding the detail of what a policy is within an OID simply opens > >>the rat hole of what change to the policy constitutes a change to the > >>OID, it is less ambiguous to refer to the primary data. Once you are > > > > in > > > >>the business of managing state on a client, then there is negligible > >>cost increase incurred in managing several data points vs. a singe > > > > data > > > >>point. > >> > >>Trevor > >> > >>-----Original Message----- > >>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > >>Sent: Tuesday, April 27, 2004 11:01 AM > >>To: Trevor Freeman > >>Cc: ietf-pkix@imc.org > >>Subject: Re: FW: scvp > >> > >>Trevor, > >> > >> > >> > >>>HI Denis, > >>>In SCVP13, the concept of validation policy was not completely in > >>>alignment with 3379 definition. > >> > >> > >>Well, it is different and this is a major problem. > >> > >> > >> > >>>It also blurred the distinction between > >>>validation policy and validation algorithm. > >> > >> > >>This is also a majo problem. > >> > >> > >> > >>>Therefore I have defined > >>>what each is in SCVP 14 and aligned the structures accordingly. > >>>Section 1.3 has the following. > >>>"In SCVP, the validation policy comprises of an algorithm for > >>>certificate path processing and the input parameters to the > >> > >>algorithm." > >> > >>This does not comply with RFC 3379. > >> > >> > >> > >>>Therefore trust anchors are an input into the algorithm and > therefore > >>>separate. > >> > >> > >>Therefore I do not follow you. > >> > >> From an interface point of view the simplest validation policy is > >>defined > >>by an OID where all the parameters necessary to perform the validation > >>check > >>are "behind" the OID. There is no need to provide any parameter > > > > through > > > >>the > >>interface. > >> > >>When there are some parameters, then they are specific to the > > > > validation > > > >>policy. I have no problem to provide specific parameters for what is > >>called > >>the "default validation policy" which is only a particular validation > >>policy > >>among many others. > >> > >>Simple clients will be unable to pass any parameter but will know > > > > which > > > >>OIDs > >>(for the validation policy) are appropriate for their applications. > >> > >>Denis > >> > >> > >> > >>>This is in alignment with 3379s definition of validation policy. > >>> > >>>Trevor > >>> > >>>-----Original Message----- > >>>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > >>>Sent: Monday, April 26, 2004 1:09 AM > >>>To: Peter Sylvester > >>>Cc: ietf-pkix@imc.org; Trevor Freeman > >>>Subject: Re: FW: scvp > >>> > >>>Peter and Trevor, > >>> > >>>I have major problems too. > >>> > >>> > >>> > >>> > >>>>in version 13 the syntax for a Query has been changed from > >>>> > >>>> Query ::= SEQUENCE { > >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, > >>>> checks CertChecks, > >>>> wantBack WantBack, > >>>> serverContextInfo [0] OCTET STRING OPTIONAL, > >>>> valPolicy [1] ValidationPolicy OPTIONAL, > >>>> validityTime [2] GeneralizedTime OPTIONAL, > >>>> trustAnchors [3] TrustAnchors OPTIONAL, > >>>> intermediateCerts [4] CertBundle OPTIONAL, > >>>> revInfos [5] RevocationInfos OPTIONAL, > >>>> queryExtensions [6] Extensions OPTIONAL } > >>> > >>> > >>>In this structure trustAnchors was more or less an alternative to > >>>valPolicy. > >>> > >>>In fact, IF the valPolicy is the default policy, THEN additional > >>>parameters > >>>MUST be provided. So the structure below does not fit as well. > >>> > >>>This leads to have the two following elements: > >>> > >>> valPolicy ValidationPolicy, > >>> paramsDefValPolicy ParamsDefValPolicy OPTIONAL, > >>> > >>>where the first one is mandatory and the second one optional. > >>> > >>> > >>> > >>> > >>>>to > >>>> > >>>> Query ::= SEQUENCE { > >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, > >>> > >>> > >>>> checks CertChecks, > >>>> wantBack WantBack, > >>>> requestRefHash BOOLEAN DEFAULT TRUE, > >>>> includePolResponce BOOLEAN DEFAULT FALSE, > >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, > >>>> requireExplicitPol BOOLEAN DEFAULT FALSE, > >>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, > >>>> valAlgorithm ValidationAlgorithm, > >>>> serverContextInfo [0] OCTET STRING OPTIONAL, > >>>> validityTime [1] GeneralizedTime OPTIONAL, > >>>> trustAnchors [2] TrustAnchors OPTIONAL, > >>>> intermediateCerts [3] CertBundle OPTIONAL, > >>>> revInfos [4] RevocationInfos OPTIONAL, > >>>> queryExtensions [5] Extensions OPTIONAL } > >>> > >>> > >>>I would thus propose to have something like: > >>> > >>>Query ::= SEQUENCE { > >>> queriedCerts SEQUENCE SIZE (1..MAX) OF > >>>CertReference, > >>> checks CertChecks, > >>> wantBack WantBack, > >>> requestRefHash BOOLEAN DEFAULT TRUE, > >>> serverContextInfo OCTET STRING OPTIONAL, > >>> validityTime GeneralizedTime OPTIONAL, > >>> includePolResponse BOOLEAN DEFAULT FALSE, > >>> valPolicy ValidationPolicy, > >>> paramsDefValPolicy [0] ParamsDefValPolicy OPTIONAL, > >>> intermediateCerts [1] CertBundle OPTIONAL, > >>> revInfos [2] RevocationInfos OPTIONAL, > >>> queryExtensions [3] Extensions OPTIONAL } > >>> > >>>and then something like: > >>> > >>> ParamsDefValPolicy :: = SEQUENCE { > >>> trustAnchors TrustAnchors, > >>> endEntityCertificationPolicy SEQUENCE OF OBJECT IDENTIFIER > >>>OPTIONAL, > >>> inhibitPolMap BOOLEAN DEFAULT FALSE, > >>> caCertificationPolicies SEQUENCE OF OBJECT IDENTIFIER > >>>OPTIONAL } > >>> > >>>Denis > >>> > >>> > >>> > >>> > >>>>I am not sure whether I am the only person unable to construct a > >>> > >>>parser. > >>> > >>> > >>> > >>>>in version 14 an aditional flag has been added which is not > available > >>> > >>>in the > >>> > >>> > >>> > >>>>Chapter 9. Is the isCA flag an orthogonal attribut to other > >>> > > validation > > > >>>>algorithms, or just to some of them, e.g,. the default name matching > >>> > >>>one? > >>> > >>> > >>> > >>>> Query ::= SEQUENCE { > >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, > >>> > >>> > >>>> checks CertChecks, > >>>> wantBack WantBack, > >>>> requestRefHash BOOLEAN DEFAULT TRUE, > >>>> includePolResponce BOOLEAN DEFAULT FALSE, > >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, > >>>> requireExplicitPol BOOLEAN DEFAULT FALSE, > >>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, > >>>> isCA BOOLEAN DEFAULT FALSE, > >>>> valAlgorithm ValidationAlgorithm, > >>>> serverContextInfo [0] OCTET STRING OPTIONAL, > >>>> validityTime [1] GeneralizedTime OPTIONAL, > >>>> trustAnchors [2] TrustAnchors OPTIONAL, > >>>> intermediateCerts [3] CertBundle OPTIONAL, > >>>> revInfos [4] RevocationInfos OPTIONAL, > >>>> keyUsage [5] KeyUsage OPTIONAL, > >>>> extendedKeyUsage [6] ExtKeyUsageSyntax OPTIONAL, > >>>> queryExtensions [7] Extensions OPTIONAL > >>>> producedAt [8] GeneralizedTime OPTIONAL} > >>>> > >>>> > >>> > >>> > >>> > >>> > >> > >> > > > > > > > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i466giC9074411; Wed, 5 May 2004 23:42:44 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i466gia7074409; Wed, 5 May 2004 23:42:44 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from ntexch03.office.corp.sia.it (clients.sia.it [193.203.230.22] (may be forged)) by above.proper.com (8.12.11/8.12.9) with ESMTP id i466ghXM074362 for <ietf-pkix@imc.org>; Wed, 5 May 2004 23:42:44 -0700 (PDT) (envelope-from adriano.santoni@actalis.it) Received: by ntexch03.office.corp.sia.it with Internet Mail Service (5.5.2657.72) id <JZ9SNMKA>; Thu, 6 May 2004 08:42:39 +0200 Message-ID: <BE82E060F5EA2D44A4CFCFFA8363958803B4BC53@ntexch00.office.corp.sia.it> From: Santoni Adriano <adriano.santoni@actalis.it> To: "'Stefan Santesson'" <stefans@microsoft.com>, Stephen Kent <kent@bbn.com> Cc: pgut001@cs.auckland.ac.nz, ietf-pkix@imc.org Subject: R: R: I: R: RFC3280: doubt on the use of UTF8String in encoding R DNs Date: Thu, 6 May 2004 08:42:31 +0200 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2657.72) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i466giXM074404 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Thanks to both Kent and Santesson for their comments. Of course, it was not my intention to argument against the adoption of UTF8String. On the contrary, I am personally convinced that is (would be? will be?) a good and sensible initiative. And of course, my previous message was provocative, as I am perfectly aware that "if we were never able to call for a change in how things are done when we update a standard, then we could never make any changes", and that "IETF standards are voluntary. failure to comply is not punishable"; I dare to say that is quite obvious to almost anyone in the industry. Actually, it is not just the compliance to IETF standards which is voluntary. It is also the degree of compliance and the very meaning of compliance, to be voluntary. It is just too easy to claim compliance to any standard. What if RFCs had an incipit saying something like <<Failure to comply to any of the mandatory features herein described shall be equivalent to not being compliant at all with this standard. It is prohibited to claim compliance with this standard unless each and all of the mandatory features herein described are fully obeyed to>>. Well, I know I am being too candid with this idea... However, it would be nice to see a meaningful number of CAs discuss this topic and try to decide on the way to go. Should we try to switch to UTF8String during the current year? Or in the next year? Or just leave any CA do what it likes, regardless of any "MUST" or deadlines that happen to show up in RFCs? This mailing list should be the right place to discuss that, but it seems the topic is not very interesting... Adriano -----Messaggio originale----- Da: Stefan Santesson [mailto:stefans@microsoft.com] Inviato: giovedì 6 maggio 2004 2.18 A: Stephen Kent; Santoni Adriano Cc: pgut001@cs.auckland.ac.nz; ietf-pkix@imc.org Oggetto: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Steve, There is one unfortunate side effect that makes this requirement worse than other evolutions of standards. It is almost never feasible to replace all certificates in a hierarchy at once. This means that you over time will have a mixed environment where e.g. the same organization name will exist in different encoding in many places of your hierarchy. It will then be virtually impossible to apply name constraining in such environment because RFC 3280 does not mandate the full X.500 name matching rules and a client may subsequently decide that names with different encoding is a no-match. This seams to be an unsolvable problem that, looking in the mirror, is inadequately addressed and shouldn't have passed in its current form. In fact it is probably a defect that needs to be addressed in sonOfRfc3280 in one way or the other. /Stefan > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On Behalf Of Stephen Kent > Sent: den 5 maj 2004 13:36 > To: Santoni Adriano > Cc: 'pgut001@cs.auckland.ac.nz'; ietf-pkix@imc.org > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding > RDNs > > > At 9:54 AM +0200 4/23/04, Santoni Adriano wrote: > >A rule that is "at odds with reality" should not be expressed in terms of > >MUST, but rather in terms of SHOULD. I wonder how it came that RFC 3280 > >reached the status of an official IETF standard, if it contains one or > more > >areas that are "at odds with reality". > > > >I suppose the answer is one of these: > >- RFC 3280 does not really contain areas that are "at odds with reality" > - > >it's just that we do not understand them; > > RFC 3280 established a requirement for a change in how DNs are > encoded, for a future point in time. Thus, at the time the RFC was > published as a standard of course it did not match "reality." since it > was calling for a change. By the way, this change was not the idea of > the PKIX WG but rather is consistent with the overall IETF mandate to > better accommodate character sets for a wide range of languages. > > >- containing areas that are "at odds with reality" is not an obstacle to > >becoming a standard; > > Think of the implication of your comment. if we were never able to > call for a change in how things are done when we update a standard, > then we could never make any changes. obviously this would be unduly > limiting. > > >- expressions like "MUST" in IETF standard are really to be taken as mere > >suggestions; > > IETF standards are voluntary. failure to comply is not punishable :-) > > >- nobody really cares. > > one might say that in environments where a few vendors or service > providers control a significant market share, their deviation from > standards, or failure to accommodate changes in standards, can have > profound impacts. > > Steve *******************Internet Email Confidentiality Footer******************* Qualsiasi utilizzo non autorizzato del presente messaggio nonche' dei suoi allegati e' vietato e potrebbe costituire reato. Se lei ha ricevuto erroneamente il presente messaggio, Le saremmo grati se, via e-mail, ce ne comunicasse la ricezione e provvedesse alla distruzione del messaggio stesso e dei suoi eventuali allegati. Le dichiarazioni contenute nel presente messaggio nonche' nei suoi eventuali allegati devono essere attribuite esclusivamente al mittente e non possono essere considerate come trasmesse o autorizzate da SIA S.p.A.; le medesime dichiarazioni non impegnano SIA S.p.A. nei confronti del destinatario o di terzi. SIA S.p.A. non si assume alcuna responsabilita' per eventuali intercettazioni, modifiche o danneggiamenti del presente messaggio e-mail. Any unauthorized use of this e-mail or any of its attachments is prohibited and could constitute an offence. If you are not the intended addressee please advise immediately the sender by using the reply facility in your e-mail software and destroy the message and its attachments. The statements and opinions expressed in this e-mail message are those of the author of the message and do not necessarily represent those of SIA. Besides, The contents of this message shall be understood as neither given nor endorsed by SIA S.p.A.. SIA S.p.A. does not accept liability for corruption, interception or amendment, if any, or the consequences thereof. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i466f6el073369; Wed, 5 May 2004 23:41:06 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i466f61Q073368; Wed, 5 May 2004 23:41:06 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i466exGo073253 for <ietf-pkix@imc.org>; Wed, 5 May 2004 23:41:00 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms21.chttl.com.tw (ms21 [10.144.2.121]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i466dtGG003859; Thu, 6 May 2004 14:39:55 +0800 Received: (from root@localhost) by ms21.chttl.com.tw (8.12.10/8.12.10) id i466dtnM027982; Thu, 6 May 2004 14:39:55 +0800 Received: from wcwang ([10.144.133.79]) by ms21.chttl.com.tw (8.12.10/8.12.10) with SMTP id i466dsL5027783; Thu, 6 May 2004 14:39:54 +0800 Message-ID: <010c01c43334$efc77a00$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "Stefan Santesson" <stefans@microsoft.com>, "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org> References: <0C3042E92D8A714783E2C44AB9936E1DF502FF@EUR-MSG-03.europe.corp.microsoft.com> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Date: Thu, 6 May 2004 14:39:51 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> RFC 3280 does not ask CAs to replace all certificates at once for migrating to UTF8String encoding. Instead, it suggest that "CAs MAY issue "name rollover" certificates to support an orderly migration to UTF8String encoding." I believe that if relying parties use 3280-compliant certification path validation modules, then they will not have problem for processing "name rollover" certificates, and thus the migration to UTF8String encoding will not be a problem. One thing should be noted: a "name rollover" certificate is essentially a "self-issued certificate" but its issuer name and subject name is in different encoding. Therefore, the definition of "self-issued certificate" should be changed a little. Also, since a "name rollover" certificate may contains a NameConstraints extension in which the permittedSubtrees and excludedSubtrees will be in a new encoding (e.g., UTF8String), the path validation in RFC 3280 also need to be revised. I hope all of these should be addressed in sonOfRfc3280. I don't think that RFC 3280 mandates the support for UTF8String encoding is "at odds with reality". There are many countries/regions in the world where it is impossible to encode names in Printable String. In Taiwan, we have already millions of certificates with issuer names and subject names encoded in UTF8String. Three years ago, I tested some VPN devices and found most of them can not accept certificates with UTF8String (One of them even crashed if received a certificate with UTF8String). Recently, I test many VPN devices again and amazingly found the all of them can normally handle certificates with UTF8String. All of them can chain and validate certificte paths that contains certificates with UTF8String. (Although, most of them still can not display the UTF8String in Chinese correctly. It is a pity.) The truth is that there are more and more applications that can handling UTF8String encoding. I think the credit should give to RFC 3280 for its mandating for UTF8String encoding. Wen-Cheng Wang ----- Original Message ----- From: "Stefan Santesson" <stefans@microsoft.com> To: "Stephen Kent" <kent@bbn.com>; "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>; <ietf-pkix@imc.org> Sent: Thursday, May 06, 2004 8:18 AM Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs > > > Steve, > > There is one unfortunate side effect that makes this requirement worse > than other evolutions of standards. > > It is almost never feasible to replace all certificates in a hierarchy > at once. This means that you over time will have a mixed environment > where e.g. the same organization name will exist in different encoding > in many places of your hierarchy. > > It will then be virtually impossible to apply name constraining in such > environment because RFC 3280 does not mandate the full X.500 name > matching rules and a client may subsequently decide that names with > different encoding is a no-match. > > This seams to be an unsolvable problem that, looking in the mirror, is > inadequately addressed and shouldn't have passed in its current form. In > fact it is probably a defect that needs to be addressed in sonOfRfc3280 > in one way or the other. > > /Stefan > > > > -----Original Message----- > > From: owner-ietf-pkix@mail.imc.org > [mailto:owner-ietf-pkix@mail.imc.org] > > On Behalf Of Stephen Kent > > Sent: den 5 maj 2004 13:36 > > To: Santoni Adriano > > Cc: 'pgut001@cs.auckland.ac.nz'; ietf-pkix@imc.org > > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in > encoding > > RDNs > > > > > > At 9:54 AM +0200 4/23/04, Santoni Adriano wrote: > > >A rule that is "at odds with reality" should not be expressed in > terms of > > >MUST, but rather in terms of SHOULD. I wonder how it came that RFC > 3280 > > >reached the status of an official IETF standard, if it contains one > or > > more > > >areas that are "at odds with reality". > > > > > >I suppose the answer is one of these: > > >- RFC 3280 does not really contain areas that are "at odds with > reality" > > - > > >it's just that we do not understand them; > > > > RFC 3280 established a requirement for a change in how DNs are > > encoded, for a future point in time. Thus, at the time the RFC was > > published as a standard of course it did not match "reality." since > > it was calling for a change. By the way, this change was not the > > idea of the PKIX WG but rather is consistent with the overall IETF > > mandate to better accommodate character sets for a wide range of > > languages. > > > > >- containing areas that are "at odds with reality" is not an obstacle > to > > >becoming a standard; > > > > Think of the implication of your comment. if we were never able to > > call for a change in how things are done when we update a standard, > > then we could never make any changes. obviously this would be unduly > > limiting. > > > > >- expressions like "MUST" in IETF standard are really to be taken as > mere > > >suggestions; > > > > IETF standards are voluntary. failure to comply is not punishable :-) > > > > >- nobody really cares. > > > > one might say that in environments where a few vendors or service > > providers control a significant market share, their deviation from > > standards, or failure to accommodate changes in standards, can have > > profound impacts. > > > > Steve > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i460ITdt094135; Wed, 5 May 2004 17:18:29 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i460ITdg094134; Wed, 5 May 2004 17:18:29 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-eur.microsoft.com (mail-eur.microsoft.com [213.199.128.145]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i460INdx094114 for <ietf-pkix@imc.org>; Wed, 5 May 2004 17:18:23 -0700 (PDT) (envelope-from stefans@microsoft.com) Received: from eur-imc-02.europe.corp.microsoft.com ([65.53.196.43]) by mail-eur.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 01:18:19 +0100 Received: from 65.53.196.43 by eur-imc-02.europe.corp.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 06 May 2004 01:18:19 +0100 Received: from EUR-MSG-03.europe.corp.microsoft.com ([65.53.192.44]) by eur-imc-02.europe.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 May 2004 01:18:19 +0100 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Date: Thu, 6 May 2004 01:18:25 +0100 Message-ID: <0C3042E92D8A714783E2C44AB9936E1DF502FF@EUR-MSG-03.europe.corp.microsoft.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Thread-Index: AcQy9mNaeqGq2v7ISTCu0K60y6EBLwAB9oWg From: "Stefan Santesson" <stefans@microsoft.com> To: "Stephen Kent" <kent@bbn.com>, "Santoni Adriano" <adriano.santoni@actalis.it> Cc: <pgut001@cs.auckland.ac.nz>, <ietf-pkix@imc.org> X-OriginalArrivalTime: 06 May 2004 00:18:19.0077 (UTC) FILETIME=[A2185750:01C432FF] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i460INdx094124 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Steve, There is one unfortunate side effect that makes this requirement worse than other evolutions of standards. It is almost never feasible to replace all certificates in a hierarchy at once. This means that you over time will have a mixed environment where e.g. the same organization name will exist in different encoding in many places of your hierarchy. It will then be virtually impossible to apply name constraining in such environment because RFC 3280 does not mandate the full X.500 name matching rules and a client may subsequently decide that names with different encoding is a no-match. This seams to be an unsolvable problem that, looking in the mirror, is inadequately addressed and shouldn't have passed in its current form. In fact it is probably a defect that needs to be addressed in sonOfRfc3280 in one way or the other. /Stefan > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On Behalf Of Stephen Kent > Sent: den 5 maj 2004 13:36 > To: Santoni Adriano > Cc: 'pgut001@cs.auckland.ac.nz'; ietf-pkix@imc.org > Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding > RDNs > > > At 9:54 AM +0200 4/23/04, Santoni Adriano wrote: > >A rule that is "at odds with reality" should not be expressed in terms of > >MUST, but rather in terms of SHOULD. I wonder how it came that RFC 3280 > >reached the status of an official IETF standard, if it contains one or > more > >areas that are "at odds with reality". > > > >I suppose the answer is one of these: > >- RFC 3280 does not really contain areas that are "at odds with reality" > - > >it's just that we do not understand them; > > RFC 3280 established a requirement for a change in how DNs are > encoded, for a future point in time. Thus, at the time the RFC was > published as a standard of course it did not match "reality." since > it was calling for a change. By the way, this change was not the > idea of the PKIX WG but rather is consistent with the overall IETF > mandate to better accommodate character sets for a wide range of > languages. > > >- containing areas that are "at odds with reality" is not an obstacle to > >becoming a standard; > > Think of the implication of your comment. if we were never able to > call for a change in how things are done when we update a standard, > then we could never make any changes. obviously this would be unduly > limiting. > > >- expressions like "MUST" in IETF standard are really to be taken as mere > >suggestions; > > IETF standards are voluntary. failure to comply is not punishable :-) > > >- nobody really cares. > > one might say that in environments where a few vendors or service > providers control a significant market share, their deviation from > standards, or failure to accommodate changes in standards, can have > profound impacts. > > Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45LgfZ4075701; Wed, 5 May 2004 14:42:41 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i45Lgeu3075684; Wed, 5 May 2004 14:42:40 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from aragorn.bbn.com (aragorn.bbn.com [128.33.0.62]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45LgclD075503 for <ietf-pkix@imc.org>; Wed, 5 May 2004 14:42:38 -0700 (PDT) (envelope-from kent@bbn.com) Received: from [128.33.244.157] (dhcp89-089-075.bbn.com [128.89.89.75]) by aragorn.bbn.com (8.12.7/8.12.7) with ESMTP id i45Lg77l000994; Wed, 5 May 2004 17:42:25 -0400 (EDT) Mime-Version: 1.0 X-Sender: kent@po2.bbn.com Message-Id: <p0610050bbcbeff0ab6f6@[128.33.244.157]> In-Reply-To: <BE82E060F5EA2D44A4CFCFFA8363958803B4BB4B@ntexch00.office.sia.it> References: <BE82E060F5EA2D44A4CFCFFA8363958803B4BB4B@ntexch00.office.sia.it> Date: Wed, 5 May 2004 16:36:24 -0400 To: Santoni Adriano <adriano.santoni@actalis.it> From: Stephen Kent <kent@bbn.com> Subject: Re: R: I: R: RFC3280: doubt on the use of UTF8String in encoding RDNs Cc: "'pgut001@cs.auckland.ac.nz'" <pgut001@cs.auckland.ac.nz>, ietf-pkix@imc.org Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> At 9:54 AM +0200 4/23/04, Santoni Adriano wrote: >A rule that is "at odds with reality" should not be expressed in terms of >MUST, but rather in terms of SHOULD. I wonder how it came that RFC 3280 >reached the status of an official IETF standard, if it contains one or more >areas that are "at odds with reality". > >I suppose the answer is one of these: >- RFC 3280 does not really contain areas that are "at odds with reality" - >it's just that we do not understand them; RFC 3280 established a requirement for a change in how DNs are encoded, for a future point in time. Thus, at the time the RFC was published as a standard of course it did not match "reality." since it was calling for a change. By the way, this change was not the idea of the PKIX WG but rather is consistent with the overall IETF mandate to better accommodate character sets for a wide range of languages. >- containing areas that are "at odds with reality" is not an obstacle to >becoming a standard; Think of the implication of your comment. if we were never able to call for a change in how things are done when we update a standard, then we could never make any changes. obviously this would be unduly limiting. >- expressions like "MUST" in IETF standard are really to be taken as mere >suggestions; IETF standards are voluntary. failure to comply is not punishable :-) >- nobody really cares. one might say that in environments where a few vendors or service providers control a significant market share, their deviation from standards, or failure to accommodate changes in standards, can have profound impacts. Steve Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45LTvPn073809; Wed, 5 May 2004 14:29:57 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i45LTv5l073808; Wed, 5 May 2004 14:29:57 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from nwkea-mail-1.sun.com (nwkea-mail-1.sun.com [192.18.42.13]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45LTufq073784 for <ietf-pkix@imc.org>; Wed, 5 May 2004 14:29:56 -0700 (PDT) (envelope-from Steve.Hanna@Sun.COM) Received: from phys-bur1-1 ([129.148.13.15]) by nwkea-mail-1.sun.com (8.12.10/8.12.9) with ESMTP id i45LTu6b004927 for <ietf-pkix@imc.org>; Wed, 5 May 2004 14:29:56 -0700 (PDT) Received: from conversion-daemon.bur-mail2.east.sun.com by bur-mail2.east.sun.com (iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003)) id <0HX900701FIETE@bur-mail2.east.sun.com> for ietf-pkix@imc.org; Wed, 05 May 2004 17:29:55 -0400 (EDT) Received: from sun.com (dhcp-ubur02-75-154.East.Sun.COM [129.148.75.154]) by bur-mail2.east.sun.com (iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003)) with ESMTPA id <0HX900236FPVC5@bur-mail2.east.sun.com>; Wed, 05 May 2004 17:29:55 -0400 (EDT) Date: Wed, 05 May 2004 17:26:00 -0400 From: Steve Hanna <Steve.Hanna@Sun.COM> Subject: Re: Cross certificate format In-reply-to: <OF63F34A40.14A2103B-ON85256E88.007E4CB1-85256E8B.00141423@us.ibm.com> To: Tom Gindin <tgindin@us.ibm.com> Cc: Santosh Chokhani <chokhani@orionsec.com>, ietf-pkix@imc.org Message-id: <40995BE8.8010801@sun.com> MIME-version: 1.0 Content-type: multipart/signed; protocol="application/x-pkcs7-signature"; micalg=sha1; boundary=------------ms060201000805020401020705 X-Accept-Language: en-us, en User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20031111 References: <OF63F34A40.14A2103B-ON85256E88.007E4CB1-85256E8B.00141423@us.ibm.com> Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> --------------ms060201000805020401020705 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit See section 6.2 of RFC 3280, which says: ... An implementation that supports multiple trust anchors MAY augment the algorithm presented in section 6.1 to further limit the set of valid certification paths which begin with a particular trust anchor. For example, an implementation MAY modify the algorithm to apply name constraints to a specific trust anchor during the initialization phase, or the application MAY require the presence of a particular alternative name form in the end entity certificate, or the application MAY impose requirements on application-specific extensions. Thus, the path validation algorithm presented in section 6.1 defines the minimum conditions for a path to be considered valid. There's no need to change RFC 3280 to allow path validation implementations to apply name constraints to trust anchors. That's explicitly discussed in this section of the RFC. Thanks, Steve Tom Gindin wrote: > Santosh: > > Of course, defining a trust anchor implies that the RP trusts the > anchor's public key as an issuer for at least some purposes. RFC 3280, in > the parts of section 6.1.2 I cited, appears to forbid those state > variables from being set from information defined by the RP or stored with > the trust anchor. This is consistent with its statements in 6.1.1 d which > permit self-signed certificates but say nothing about other CA > certificates. As is probably clear to everyone reading this thread, I > think that this is not an ideal set of suggestions. > The current standard does contain some constraints on the trust > anchor (notably inhibit-policy-mapping), but neither name constraints nor > basic constraints. I consider this an anomaly. If I had free rein in > rewording pieces of RFC 3280 section 6.1, I would change this wording to > allow CA certificates along with self-signed ones as anchors and create > extra optional (and optional to support) fields to be set from the trust > anchor. > I don't think that this is particularly inconsistent with general > security principles, or with any part of X.509 with which I am familiar. > It does represent an extension of RFC 3280 section 6, and it is only > backward compatible with it, so it may be out of scope. > > Tom Gindin > > > > > > "Santosh Chokhani" <chokhani@orionsec.com> > Sent by: owner-ietf-pkix@mail.imc.org > 05/01/2004 06:46 PM > > > To: <ietf-pkix@imc.org> > cc: > Subject: RE: Cross certificate format > > > > > Tom: > > The standard does not require you to check how you initialize the > variables. > You can get these from the cross certificate or from some configuration. > > The basic point is that you trust the public key. How you constrain it is > not address by the standard, where as if the same certificate appeared in > a > path, you have to apply the constraints or be not compliant with the > standard. > > You obviously do not believe in re-incarnation. Just kidding...... > > -----Original Message----- > From: Tom Gindin [mailto:tgindin@us.ibm.com] > Sent: Saturday, May 01, 2004 6:31 PM > To: Santosh Chokhani > Cc: ietf-pkix@imc.org > Subject: RE: Cross certificate format > > > Santosh: > > It is well known that trust in a trust anchor can be defined in a > variety of ways. However, there is every reason to think that an RP may > constrain that trust and that the constraints he may apply include those > which an issuer CA may apply to a subordinate one. The wording of RFC 3280 > > section 6.2's first paragraph is an example of this. > My comments about the meaning of a trust anchor are not purely > theoretical. If a cross certificate is used to represent a trust anchor, > the variables permitted_subtrees (RFC 3280 6.1.2 b) and excluded_subtrees > (RFC 3280 6.1.2 c) could easily be set from the contents of a > NameConstraints extension, and the variable max_path_length (RFC 3280 > 6.1.2 k) from the contents of a BasicConstraints extension. In RFC 3280 > today, these variables are set to constants (although that interpretation > is not absolutely clear for permitted_subtrees). In fact, they are three > of the four variables which are set to constants during initialization. > If permitted subtrees were set to a list of attributes, the > validation process would fail if the anchor CA or a CA it certified went > beyond the restrictions of those attributes. Likewise, if the excluded > subtrees were set to a list of attributes, the validation process would > fail if the anchor CA or a CA it certified went used values matching any > of those attributes. I do not believe that I am arguing against Galileo. > > Tom Gindin > > > > > > > "Santosh Chokhani" <chokhani@orionsec.com> > Sent by: owner-ietf-pkix@mail.imc.org > 04/29/2004 11:12 PM > > > To: <ietf-pkix@imc.org> > cc: > Subject: RE: Cross certificate format > > > > > Tom: > > The basic point (at the risk of repeating myself) is that a CA certificate > or a self-signed certificate can be used as a means to provide trust > anchor > information. In that case, there is no need to check signature or > anything. > > But, when the CA certificate is a certificate in the certificate path, > X.509 > logic kicks in. > > This is consistent with X.509, 3280 and general security principles since > the trust in a trust anchor can be derived through any number of means. > > If you do not agree, feel free to argue against the definitions and > Galileo > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On > Behalf Of Tom Gindin > Sent: Thursday, April 29, 2004 8:43 PM > To: Santosh Chokhani > Cc: ietf-pkix@imc.org > Subject: RE: Cross certificate format > > > > Santosh: > > I don't think that a CA certificate and a trust anchor are as much > > > different things as different uses for the same thing. They are both > identifiers of issuers which bind a public key to an issuer name, and it > is perfectly appropriate for any CA certificate to represent a trust > anchor. I also do not see why such things as name constraints on a trust > anchor are inappropriate. It is perfectly true that verifying a trust > anchor's certificate, when issued by another CA (who may not be trusted) > is a pointless operation. > > Tom Gindin > > > > > > "Santosh Chokhani" <chokhani@orionsec.com> > Sent by: owner-ietf-pkix@mail.imc.org > 04/27/2004 09:14 AM > > > To: <ietf-pkix@imc.org> > cc: > Subject: RE: Cross certificate format > > > > > Tom: > > My point is that even if the trust anchor is a self-signed certificate, > all > you need to do is extract the required information. There is no need to > examine anything. > > Cross certificate and trust anchor are very different things. An element > of > the cross certificate pair is nothing but an intermediate CA certificate > in > a certification path. > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On > Behalf Of Tom Gindin > Sent: Tuesday, April 27, 2004 7:33 AM > To: Santosh Chokhani > Cc: ietf-pkix@imc.org; owner-ietf-pkix@mail.imc.org > Subject: RE: Cross certificate format > > > > Santosh: > > Of course, you are right that RFC 3280 6.1.1 d permits trust > anchor information to be provided outside a certificate. If it is so, no > checks of the sort I indicated can be performed. It also says that it can > > > > be provided as "a self-signed certificate", with no further qualification. > > > > I do think it is somewhat odd that a self-signed certificate with > KeyUsage set to typical EE values would be considered a valid issuer as a > trust anchor while a cross certificate would not. You can always extract > the needed anchor information from a cross-certificate, of course. > > Tom Gindin > > > > > > "Santosh Chokhani" <chokhani@orionsec.com> > Sent by: owner-ietf-pkix@mail.imc.org > 04/26/2004 08:51 AM > > > To: <ietf-pkix@imc.org> > cc: > Subject: RE: Cross certificate format > > > > > Tom: > > My reading of 3280 and X.509 is that a trust anchor need not even be a > certificate. Thus, no checks need to be performed on a trust anchor at > all. > You get the DN, public key, algorithm OID, and parameters (if applicable) > from a trust anchor. Any checks are gratuitous and not required by either > standard. > > -----Original Message----- > From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] > On > Behalf Of Tom Gindin > Sent: Sunday, April 25, 2004 7:06 PM > To: Jean-Marc Desperrier > Cc: ietf-pkix@imc.org > Subject: Re: Cross certificate format > > > > Jean-Marc: > > RFC 3280 doesn't say that anywhere. It does say that you don't > need to validate the trust anchor as part of the path, but it isn't clear > from the text that a v1 certificate is acceptable as a trust anchor - and > it certainly isn't clear that a v1 certificate is acceptable as an issuer > if it is a trust anchor while a v3 certificate with KeyUsage= { > DigitalSignature, keyEncipherment } is not acceptable as an issuer even if > > > > > it is a trust anchor. > I believe that the correct rules for versions are something like > the following: a certificate issued by a trust anchor which is represented > > > > > by a certificate is verifiable if the anchor certificate is either a v1 > certificate or a v3 certificate with the CA flag present in the > basicConstraints extension. If the anchor certificate is a v3 certificate > > > > > with the KeyUsage extension present, it is only a valid issuer certificate > > > > > if keyCertSign is set. > > Tom Gindin > > > > > > Jean-Marc Desperrier <jmdesp@free.fr> > Sent by: owner-ietf-pkix@mail.imc.org > 04/21/2004 03:42 PM > > > To: ietf-pkix@imc.org > cc: > Subject: Re: Cross certificate format > > > > > Tom Gindin wrote: > > >> RFC 3280 does not support the use of v1 self-signed >>certificates, > > >>which contain no extensions at all (see the text of RFC 3280's >>basicConstraints section). However, a number of public CA's still have >>them. >> >> > > Applications implementing the path validation algorithm of RFC3280 MUST > accept them when they are used as trust anchors. > > > > > > > > > > > > > > > > > > > > > --------------ms060201000805020401020705 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJNzCC AvYwggJfoAMCAQICAww3ZDANBgkqhkiG9w0BAQQFADBiMQswCQYDVQQGEwJaQTElMCMGA1UE ChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNv bmFsIEZyZWVtYWlsIElzc3VpbmcgQ0EwHhcNMDQwNDI4MjEwMTE0WhcNMDUwNDI4MjEwMTE0 WjBoMQ4wDAYDVQQEEwVIYW5uYTEVMBMGA1UEKhMMU3RlcGhlbiBSb3NzMRswGQYDVQQDExJT dGVwaGVuIFJvc3MgSGFubmExIjAgBgkqhkiG9w0BCQEWE3N0ZXZlLmhhbm5hQHN1bi5jb20w ggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC11rLYp70KnKGmV6pNTFRhWBwq47mV 4hU6EaZxo6I3Vw96Rd3E4TH3SAthIl5N4/tpiMPtlTBiJOU6QHSlz1DgTCclmTK6taqtEyLq amez8vYgSdOEnOFRYILptjbEwpE+yeUmFKOmZ7zSwBMqRQatMHMaLVUJJ3Ygw1V7fKSwvVU0 OJ8c+c3sJRtRyDLxq/vMzYQP+wyqRqlJPTHGFqVuMp0FZIV5X76z+O+W90m9jXjirz91Jz+W N6rwqmPi8Qy0uxBKm6NKT1371p+WYaK9OfGk8vGbJ8ZsTP3DyQhuuemYPbkKPtoII6WnHIGT /DOy2G7gff7GOIhUrs7PziddAgMBAAGjMDAuMB4GA1UdEQQXMBWBE3N0ZXZlLmhhbm5hQHN1 bi5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQB++aOP6sZVYlqBGc92hZ61 FHST+OqnKZHRGkqEO94RoxdeaSUFKVO2IaKLMBmqNzmh3N3pIsIKiLhJNvDgLhlRrMRVfR29 uixjTsprR26AacMx/gCiCstGIjkzTb83xAvZay5hnzUmlt/LSNX8e6jTNav6LP64WC+C7Tul fM/PzDCCAvYwggJfoAMCAQICAww3ZDANBgkqhkiG9w0BAQQFADBiMQswCQYDVQQGEwJaQTEl MCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3Rl IFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0EwHhcNMDQwNDI4MjEwMTE0WhcNMDUwNDI4 MjEwMTE0WjBoMQ4wDAYDVQQEEwVIYW5uYTEVMBMGA1UEKhMMU3RlcGhlbiBSb3NzMRswGQYD VQQDExJTdGVwaGVuIFJvc3MgSGFubmExIjAgBgkqhkiG9w0BCQEWE3N0ZXZlLmhhbm5hQHN1 bi5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQC11rLYp70KnKGmV6pNTFRh WBwq47mV4hU6EaZxo6I3Vw96Rd3E4TH3SAthIl5N4/tpiMPtlTBiJOU6QHSlz1DgTCclmTK6 taqtEyLqamez8vYgSdOEnOFRYILptjbEwpE+yeUmFKOmZ7zSwBMqRQatMHMaLVUJJ3Ygw1V7 fKSwvVU0OJ8c+c3sJRtRyDLxq/vMzYQP+wyqRqlJPTHGFqVuMp0FZIV5X76z+O+W90m9jXji rz91Jz+WN6rwqmPi8Qy0uxBKm6NKT1371p+WYaK9OfGk8vGbJ8ZsTP3DyQhuuemYPbkKPtoI I6WnHIGT/DOy2G7gff7GOIhUrs7PziddAgMBAAGjMDAuMB4GA1UdEQQXMBWBE3N0ZXZlLmhh bm5hQHN1bi5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQQFAAOBgQB++aOP6sZVYlqB Gc92hZ61FHST+OqnKZHRGkqEO94RoxdeaSUFKVO2IaKLMBmqNzmh3N3pIsIKiLhJNvDgLhlR rMRVfR29uixjTsprR26AacMx/gCiCstGIjkzTb83xAvZay5hnzUmlt/LSNX8e6jTNav6LP64 WC+C7TulfM/PzDCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAwgdExCzAJBgNVBAYT AlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUgVG93bjEaMBgGA1UE ChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRpb24gU2VydmljZXMg RGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBDQTErMCkGCSqG SIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0wMzA3MTcwMDAwMDBa Fw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3Vs dGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNz dWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV+065yplaHmjAdQRw nd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfArhVqqP3FWy688Cwfn 8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/p7bRPGEEQB5kGXJg t/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8MDowOKA2oDSGMmh0 dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWlsQ0EuY3JsMAsGA1Ud DwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxhYmVsMi0xMzgwDQYJ KoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/TCG4+DYfqi2fNi/A 9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amcOY6MIE9lX5Xa9/eH 1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggM7MIIDNwIBATBpMGIxCzAJ BgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYD VQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIDDDdkMAkGBSsOAwIa BQCgggGnMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcNAQkFMQ8XDTA0MDUw NTIxMjYwMFowIwYJKoZIhvcNAQkEMRYEFA7Sw9pTNCo0RINQDXkZNeztKBdaMFIGCSqGSIb3 DQEJDzFFMEMwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqGSIb3DQMCAgFAMAcG BSsOAwIHMA0GCCqGSIb3DQMCAgEoMHgGCSsGAQQBgjcQBDFrMGkwYjELMAkGA1UEBhMCWkEx JTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0 ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBAgMMN2QwegYLKoZIhvcNAQkQAgsxa6Bp MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQu MSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIDDDdkMA0G CSqGSIb3DQEBAQUABIIBAFFama7Nna/e3h+Ypq7znMk0R60dcDB/qtbY2bj5RSAnJzeSiDQL GgiGRrx/1CuHNaPw0HPrj388hBX7rfMyvHCOCclNb+JV/4CuAyYzCqbZaNmD/KTRPnv41Yit SqfDPgDbLfpBzym+SF/Yc3YxXQObkd2NOG4EpmpmmYhYrMpdlKvR+q4JEXDbWs94KF4ymiiB A/Mbt8vmorCdT7Lww/631p0JUrKrTq0NwFy4EKxtZiEHpjavbZ9WzH5HdDk0+HQVAXLiCmGp 7urjS+JQEdSOYtjZn8hu/J+Qcz4pvOJvguDaTI1uVVE1jFuwQqfOwbHAQYoyHH9Jle1sVJkF kqcAAAAAAAA= --------------ms060201000805020401020705-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45L7Q8X070245; Wed, 5 May 2004 14:07:26 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i45L7QvO070244; Wed, 5 May 2004 14:07:26 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from mail-eur.microsoft.com (mail-eur.microsoft.com [213.199.128.145]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45L7Pbq070223 for <ietf-pkix@imc.org>; Wed, 5 May 2004 14:07:25 -0700 (PDT) (envelope-from stefans@microsoft.com) Received: from eur-imc-02.europe.corp.microsoft.com ([65.53.196.43]) by mail-eur.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Wed, 5 May 2004 22:07:20 +0100 Received: from 65.53.196.43 by eur-imc-02.europe.corp.microsoft.com (InterScan E-Mail VirusWall NT); Wed, 05 May 2004 22:07:20 +0100 Received: from EUR-MSG-03.europe.corp.microsoft.com ([65.53.192.44]) by eur-imc-02.europe.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Wed, 5 May 2004 22:07:20 +0100 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPartTM-000-6cde8b41-7c06-442b-8e82-16cb7cf52a86" Subject: RFC 3280 bug in name constraints Date: Wed, 5 May 2004 22:07:19 +0100 Message-ID: <0C3042E92D8A714783E2C44AB9936E1DF502D1@EUR-MSG-03.europe.corp.microsoft.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: RFC 3280 bug in name constraints Thread-Index: AcQy5PNxPgauBM6ETtWBvVkKktiG8A== From: "Stefan Santesson" <stefans@microsoft.com> To: <ietf-pkix@imc.org> X-OriginalArrivalTime: 05 May 2004 21:07:20.0309 (UTC) FILETIME=[F4231A50:01C432E4] Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multi-part message in MIME format. ------=_NextPartTM-000-6cde8b41-7c06-442b-8e82-16cb7cf52a86 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C432E4.F54FFCD5" ------_=_NextPart_001_01C432E4.F54FFCD5 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable If someone is keeping record of bugs for future update of RFC 3280 I have a small one to file about name constraints. =20 In section 4.2.1.11, starting at bottom of page 38: =20 When rfc822 names are constrained, but the certificate does not include a subject alternative name, the rfc822 name constraint MUST be applied to the attribute of type EmailAddress in the subject distinguished name. =20 Suppose/suggest that the intended meaning is: =20 When rfc822 names are constrained, but the certificate does not=20 include an rfc822Name in a subject alternative name extension, the rfc822 name constraint MUST be applied to the attribute of type EmailAddress in the subject distinguished name. =20 =20 Rationale: Why would the constraint NOT apply to EmailAddress just because there is a SubjAltName extension with e.g. just some unrelated other name component present? =20 /Stefan =20 ------_=_NextPart_001_01C432E4.F54FFCD5 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable <html xmlns:v=3D"urn:schemas-microsoft-com:vml" = xmlns:o=3D"urn:schemas-microsoft-com:office:office" = xmlns:w=3D"urn:schemas-microsoft-com:office:word" = xmlns=3D"http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv=3DContent-Type content=3D"text/html; = charset=3Dus-ascii"> <meta name=3DGenerator content=3D"Microsoft Word 11 (filtered medium)"> <style> <!-- /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0cm; margin-bottom:.0001pt; font-size:12.0pt; font-family:"Times New Roman";} a:link, span.MsoHyperlink {color:blue; text-decoration:underline;} a:visited, span.MsoHyperlinkFollowed {color:purple; text-decoration:underline;} p.MsoPlainText, li.MsoPlainText, div.MsoPlainText {margin:0cm; margin-bottom:.0001pt; font-size:10.0pt; font-family:"Courier New";} span.EmailStyle17 {mso-style-type:personal-compose; font-family:Arial; color:windowtext;} @page Section1 {size:595.3pt 841.9pt; margin:3.0cm 2.0cm 3.0cm 2.0cm;} div.Section1 {page:Section1;} --> </style> </head> <body lang=3DDA link=3Dblue vlink=3Dpurple> <div class=3DSection1> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'>If someone is keeping record of bugs for = future update of RFC 3280 I have a small one to file about name = constraints.<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'><o:p> </o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'>In section 4.2.1.11, starting at bottom of = page 38:<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'><o:p> </o:p></span></font></p> <p class=3DMsoPlainText><font size=3D2 face=3D"Courier New"><span = lang=3DEN-US style=3D'font-size:10.0pt'> When rfc822 names are = constrained, but the<o:p></o:p></span></font></p> <p class=3DMsoPlainText><font size=3D2 face=3D"Courier New"><span = lang=3DEN-US style=3D'font-size:10.0pt'> certificate does not include a = subject alternative name, the rfc822<o:p></o:p></span></font></p> <p class=3DMsoPlainText><font size=3D2 face=3D"Courier New"><span = lang=3DEN-US style=3D'font-size:10.0pt'> name constraint MUST be applied = to the attribute of type EmailAddress<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3D"Courier New"><span = lang=3DEN-US style=3D'font-size:10.0pt;font-family:"Courier New"'> in the = subject distinguished name.</span></font><font size=3D2 face=3DArial><span = lang=3DEN-US style=3D'font-size:10.0pt;font-family:Arial'><o:p></o:p></span></font></p= > <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'><o:p> </o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'>Suppose/suggest that the intended meaning = is:<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'><o:p> </o:p></span></font></p> <p class=3DMsoPlainText><font size=3D2 face=3D"Courier New"><span = lang=3DEN-US style=3D'font-size:10.0pt'> When rfc822 names are = constrained, but the certificate does not <o:p></o:p></span></font></p> <p class=3DMsoPlainText><font size=3D2 face=3D"Courier New"><span = lang=3DEN-US style=3D'font-size:10.0pt'> include an rfc822Name in a = subject alternative name extension, the rfc822<o:p></o:p></span></font></p> <p class=3DMsoPlainText><font size=3D2 face=3D"Courier New"><span = lang=3DEN-US style=3D'font-size:10.0pt'> name constraint MUST be applied = to the attribute of type EmailAddress<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3D"Courier New"><span = lang=3DEN-US style=3D'font-size:10.0pt;font-family:"Courier New"'> in the = subject distinguished name.</span></font><font size=3D2 face=3DArial><span = lang=3DEN-US style=3D'font-size:10.0pt;font-family:Arial'><o:p></o:p></span></font></p= > <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'><o:p> </o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'><o:p> </o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'>Rationale:<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'>Why would the constraint NOT apply to = EmailAddress just because there is a SubjAltName extension with e.g. just some = unrelated other name component present?<o:p></o:p></span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'><o:p> </o:p></span></font></p> <div> <div> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-US = style=3D'font-size: 10.0pt;font-family:Arial'>/Stefan<font color=3Dblack><span = style=3D'color:black'><o:p></o:p></span></font></span></font></p> </div> </div> <p class=3DMsoNormal><font size=3D3 face=3D"Times New Roman"><span = lang=3DEN-US style=3D'font-size:12.0pt'><o:p> </o:p></span></font></p> </div> </body> </html> ------_=_NextPart_001_01C432E4.F54FFCD5-- ------=_NextPartTM-000-6cde8b41-7c06-442b-8e82-16cb7cf52a86-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45HQLRI045092; Wed, 5 May 2004 10:26:21 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i45HQLfA045091; Wed, 5 May 2004 10:26:21 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from host13.websitesource.com (host13.websitesource.com [209.239.35.152]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45HQKgp045084 for <ietf-pkix@imc.org>; Wed, 5 May 2004 10:26:20 -0700 (PDT) (envelope-from chokhani@orionsec.com) Received: from wchokhani3 (dpvc-138-88-161-20.res.east.verizon.net [138.88.161.20]) by host13.websitesource.com (8.12.10/8.12.10) with ESMTP id i45HQLMM024084 for <ietf-pkix@imc.org>; Wed, 5 May 2004 13:26:21 -0400 From: "Santosh Chokhani" <chokhani@orionsec.com> To: <ietf-pkix@imc.org> Subject: RE: Cross certificate format Date: Wed, 5 May 2004 13:26:15 -0400 Message-ID: <015901c432c6$1522eba0$9a00a8c0@hq.orionsec.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.6626 In-Reply-To: <OF63F34A40.14A2103B-ON85256E88.007E4CB1-85256E8B.00141423@us.ibm.com> X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Importance: Normal Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Tom: I think we agree on what the standards currently do not require any checks. X.509 is being revised (since I do not keep track of the process -- X.509 change may have been approved for all know) to permit the relying party to specify these values. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Tuesday, May 04, 2004 11:39 PM To: Santosh Chokhani Cc: ietf-pkix@imc.org Subject: RE: Cross certificate format Santosh: Of course, defining a trust anchor implies that the RP trusts the anchor's public key as an issuer for at least some purposes. RFC 3280, in the parts of section 6.1.2 I cited, appears to forbid those state variables from being set from information defined by the RP or stored with the trust anchor. This is consistent with its statements in 6.1.1 d which permit self-signed certificates but say nothing about other CA certificates. As is probably clear to everyone reading this thread, I think that this is not an ideal set of suggestions. The current standard does contain some constraints on the trust anchor (notably inhibit-policy-mapping), but neither name constraints nor basic constraints. I consider this an anomaly. If I had free rein in rewording pieces of RFC 3280 section 6.1, I would change this wording to allow CA certificates along with self-signed ones as anchors and create extra optional (and optional to support) fields to be set from the trust anchor. I don't think that this is particularly inconsistent with general security principles, or with any part of X.509 with which I am familiar. It does represent an extension of RFC 3280 section 6, and it is only backward compatible with it, so it may be out of scope. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 05/01/2004 06:46 PM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: The standard does not require you to check how you initialize the variables. You can get these from the cross certificate or from some configuration. The basic point is that you trust the public key. How you constrain it is not address by the standard, where as if the same certificate appeared in a path, you have to apply the constraints or be not compliant with the standard. You obviously do not believe in re-incarnation. Just kidding...... -----Original Message----- From: Tom Gindin [mailto:tgindin@us.ibm.com] Sent: Saturday, May 01, 2004 6:31 PM To: Santosh Chokhani Cc: ietf-pkix@imc.org Subject: RE: Cross certificate format Santosh: It is well known that trust in a trust anchor can be defined in a variety of ways. However, there is every reason to think that an RP may constrain that trust and that the constraints he may apply include those which an issuer CA may apply to a subordinate one. The wording of RFC 3280 section 6.2's first paragraph is an example of this. My comments about the meaning of a trust anchor are not purely theoretical. If a cross certificate is used to represent a trust anchor, the variables permitted_subtrees (RFC 3280 6.1.2 b) and excluded_subtrees (RFC 3280 6.1.2 c) could easily be set from the contents of a NameConstraints extension, and the variable max_path_length (RFC 3280 6.1.2 k) from the contents of a BasicConstraints extension. In RFC 3280 today, these variables are set to constants (although that interpretation is not absolutely clear for permitted_subtrees). In fact, they are three of the four variables which are set to constants during initialization. If permitted subtrees were set to a list of attributes, the validation process would fail if the anchor CA or a CA it certified went beyond the restrictions of those attributes. Likewise, if the excluded subtrees were set to a list of attributes, the validation process would fail if the anchor CA or a CA it certified went used values matching any of those attributes. I do not believe that I am arguing against Galileo. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/29/2004 11:12 PM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: The basic point (at the risk of repeating myself) is that a CA certificate or a self-signed certificate can be used as a means to provide trust anchor information. In that case, there is no need to check signature or anything. But, when the CA certificate is a certificate in the certificate path, X.509 logic kicks in. This is consistent with X.509, 3280 and general security principles since the trust in a trust anchor can be derived through any number of means. If you do not agree, feel free to argue against the definitions and Galileo -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Thursday, April 29, 2004 8:43 PM To: Santosh Chokhani Cc: ietf-pkix@imc.org Subject: RE: Cross certificate format Santosh: I don't think that a CA certificate and a trust anchor are as much different things as different uses for the same thing. They are both identifiers of issuers which bind a public key to an issuer name, and it is perfectly appropriate for any CA certificate to represent a trust anchor. I also do not see why such things as name constraints on a trust anchor are inappropriate. It is perfectly true that verifying a trust anchor's certificate, when issued by another CA (who may not be trusted) is a pointless operation. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/27/2004 09:14 AM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: My point is that even if the trust anchor is a self-signed certificate, all you need to do is extract the required information. There is no need to examine anything. Cross certificate and trust anchor are very different things. An element of the cross certificate pair is nothing but an intermediate CA certificate in a certification path. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Tuesday, April 27, 2004 7:33 AM To: Santosh Chokhani Cc: ietf-pkix@imc.org; owner-ietf-pkix@mail.imc.org Subject: RE: Cross certificate format Santosh: Of course, you are right that RFC 3280 6.1.1 d permits trust anchor information to be provided outside a certificate. If it is so, no checks of the sort I indicated can be performed. It also says that it can be provided as "a self-signed certificate", with no further qualification. I do think it is somewhat odd that a self-signed certificate with KeyUsage set to typical EE values would be considered a valid issuer as a trust anchor while a cross certificate would not. You can always extract the needed anchor information from a cross-certificate, of course. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/26/2004 08:51 AM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: My reading of 3280 and X.509 is that a trust anchor need not even be a certificate. Thus, no checks need to be performed on a trust anchor at all. You get the DN, public key, algorithm OID, and parameters (if applicable) from a trust anchor. Any checks are gratuitous and not required by either standard. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Sunday, April 25, 2004 7:06 PM To: Jean-Marc Desperrier Cc: ietf-pkix@imc.org Subject: Re: Cross certificate format Jean-Marc: RFC 3280 doesn't say that anywhere. It does say that you don't need to validate the trust anchor as part of the path, but it isn't clear from the text that a v1 certificate is acceptable as a trust anchor - and it certainly isn't clear that a v1 certificate is acceptable as an issuer if it is a trust anchor while a v3 certificate with KeyUsage= { DigitalSignature, keyEncipherment } is not acceptable as an issuer even if it is a trust anchor. I believe that the correct rules for versions are something like the following: a certificate issued by a trust anchor which is represented by a certificate is verifiable if the anchor certificate is either a v1 certificate or a v3 certificate with the CA flag present in the basicConstraints extension. If the anchor certificate is a v3 certificate with the KeyUsage extension present, it is only a valid issuer certificate if keyCertSign is set. Tom Gindin Jean-Marc Desperrier <jmdesp@free.fr> Sent by: owner-ietf-pkix@mail.imc.org 04/21/2004 03:42 PM To: ietf-pkix@imc.org cc: Subject: Re: Cross certificate format Tom Gindin wrote: > RFC 3280 does not support the use of v1 self-signed > certificates, >which contain no extensions at all (see the text of RFC 3280's >basicConstraints section). However, a number of public CA's still have >them. > > Applications implementing the path validation algorithm of RFC3280 MUST accept them when they are used as trust anchors. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45Eeh7F031171; Wed, 5 May 2004 07:40:43 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i45EehUj031169; Wed, 5 May 2004 07:40:43 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from edelweb.fr (edelweb.fr [212.234.46.16]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45EefgQ031150 for <ietf-pkix@imc.org>; Wed, 5 May 2004 07:40:42 -0700 (PDT) (envelope-from Peter.Sylvester@edelweb.fr) Received: from chandon.edelweb.fr (localhost [127.0.0.1]) by edelweb.fr (8.11.7p1+Sun/8.11.7) with ESMTP id i45EebN25746; Wed, 5 May 2004 16:40:37 +0200 (MEST) Received: from chandon.edelweb.fr (chandon.edelweb.fr [193.51.14.162]) by edelweb.fr (nospam/1.8); Wed, 5 May 2004 16:40:38 +0200 (MET DST) Received: (from peter@localhost) by chandon.edelweb.fr (8.11.7p1+Sun/8.11.7) id i45EebQ22331; Wed, 5 May 2004 16:40:37 +0200 (MEST) Date: Wed, 5 May 2004 16:40:37 +0200 (MEST) From: Peter Sylvester <Peter.Sylvester@edelweb.fr> Message-Id: <200405051440.i45EebQ22331@chandon.edelweb.fr> To: trevorf@exchange.microsoft.com Subject: RE: FW: scvp Cc: ietf-pkix@imc.org X-Sun-Charset: US-ASCII Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> - Is there a difference between asking for a key usage of certsign and setting the isCA flag? - For an extendedkeyusage, is the server supposed to chacek whether there are the acceptable keyusages in the certificat? I think, yes, this should be made more precisely mentioned. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45Dts72024888; Wed, 5 May 2004 06:55:54 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i45Dtsnk024887; Wed, 5 May 2004 06:55:54 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from avir1.bancsabadell.com (tm01.bancsabadell.com [194.224.15.12]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45DtqV5024872 for <ietf-pkix@imc.org>; Wed, 5 May 2004 06:55:53 -0700 (PDT) (envelope-from gonzalezcarlos@extendforce.com) Received: from EFYBEM279076 (privadas.nuria.telefonica-data.net [172.18.16.2]) by mail.bancsabadell.com; Wed, 05 May 2004 15:52:53 +0200 From: =?iso-8859-1?Q?Carlos_Gonz=E1lez-Cadenas?= <gonzalezcarlos@extendforce.com> To: <ietf-pkix@imc.org> Subject: DataEncipherment Date: Wed, 5 May 2004 15:54:26 +0200 Organization: e-xtendforce Message-ID: <002901c432a8$7acb60e0$021012ac@extendforce.net> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_002A_01C432B9.3E5430E0" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.4510 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Importance: Normal Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> This is a multi-part message in MIME format. ------=_NextPart_000_002A_01C432B9.3E5430E0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi, =0D Just a question about keyUsage DataEncipherment. =0D In a RSA-based certificate (the public key included is coming from a RSA= key pair), If I assert the DataEncipherment key usage, I=92m telling to the relying third parties that my certificate (the public key included into)= can be used for raw data encipherment, but, in the reality, RSA algorithm= cannot be used to cipher large amounts of data (as opposed to a symmetric key, in, for example, a CBC mode). The common technique with that is to use the keyEncipherment bit to be able to cipher the symmetric session key used to cipher the raw data. =0D Can anyone point me to the right direction, or provide me some links to technical documentation, about real-life usages with the Data Encipherment key usage??. Are there any other technologies/algorithms for RSA where this key usage has more sense??. =0D Thank you very much in advance Carlos =0D Carlos Gonz=E1lez-Cadenas Jefe de Departamento de Arquitectura Software Software Architecture Department Manager=0D Responsable de Seguridad de la Informaci=F3n Information Security Responsible=0D e-xtendforce by e-xtendnow BancSabadell=0D C/Sena,12 nucli C planta 2 Edifici Landscape 08190 Sant Cugat del Vall=E8s phone: +34933556139 fax: +34933556142 mailto: <mailto:gonzalezcarlos@extendforce.com> gonzalezcarlos@extendforce.com=0D =0D =0D Advertencia legal: Este mensaje y, en su caso, los ficheros anexos son confidenciales, especialmente en lo que respecta a los datos personales, y se dirigen exclusivamente al destinatario referenciado. Si usted no lo es y lo ha recibido por error o tiene conocimiento del mismo por cualquier motivo, le rogamos que nos lo comunique por este medio y proceda a destruirlo o= borrarlo, y que en todo caso se abstenga de utilizar, reproducir, alterar, archivar o comunicar a terceros el presente mensaje y ficheros anexos, todo ello bajo= pena de incurrir en responsabilidades legales. El emisor no garantiza la= integridad, rapidez o seguridad del presente correo, ni se responsabiliza de posibles perjuicios derivados de la captura, incorporaciones de virus o cualesquiera otras manipulaciones efectuadas por terceros. =0D Advertiment legal: Aquest missatge i, si escau, els fitxers annexos tenen caire confidencial, especialment pel que fa a les dades personals, i s'adrecen exclusivament al destinatari referenciat. Si no es tracta d'aquest i l'ha rebut per error o= se li ha fet arribar per qualsevol motiu, li preguem que ens ho comuniqui per= aquesta mateixa via i el destrueixi o l'esborri, i que en tot cas s'abstingui= d'utilitzar, reproduir, alterar, arxivar o comunicar a tercers aquest missatge i fitxers annexos, tot sota pena d'entrar en responsabilitats legals. L'emissor no= garanteix la integritat, la rapidesa o la seguretat d'aquest correu, ni es= responsabilitza de possibles perjudicis derivats de la captura, incorporacions de virus o qualsevol altres manipulacions que facin tercers. =0D Disclaimer: This message and any attached files transmitted with it, is confidential, especially as regards personal data. It is intended solely for the use of= the individual or entity to whom it is addressed. If you are not the intended= recipient and have received this information in error or have accessed it for any= reason, please notify us of this fact by email reply and then destroy or delete the= message, refraining from any reproduction, use, alteration, filing or communication= to third parties of this message and attached files on penalty of incurring legal responsibilities. The sender does not guarantee the integrity, the= accuracy, the swift delivery or the security of this email transmission, and assumes no responsibility for any possible damage incurred through data capture, virus incorporation or any manipulation carried out by third parties. ------=_NextPart_000_002A_01C432B9.3E5430E0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <html> <head> <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; charset= =3Diso-8859-1"> <meta name=3DGenerator content=3D"Microsoft Word 10 (filtered)"> <style> <!-- /* Font Definitions */ @font-face {font-family:"Palatino Linotype"; panose-1:2 4 5 2 5 5 5 3 3 4;} @font-face {font-family:Verdana; panose-1:2 11 6 4 3 5 4 4 2 4;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0cm; margin-bottom:.0001pt; font-size:12.0pt; font-family:"Palatino Linotype";} a:link, span.MsoHyperlink {color:blue; text-decoration:underline;} a:visited, span.MsoHyperlinkFollowed {color:purple; text-decoration:underline;} span.EstiloCorreo17 {font-family:Arial; color:windowtext;} span.name1 {font-weight:bold;} span.logo1 {font-weight:bold;} span.textlogo1 {color:#999999;} @page Section1 {size:595.3pt 841.9pt; margin:70.85pt 3.0cm 70.85pt 3.0cm;} div.Section1 {page:Section1;} --> </style> </head> <body lang=3DES link=3Dblue vlink=3Dpurple> <div class=3DSection1> <p class=3DMsoNormal><font size=3D2 face=3DArial><span style= =3D'font-size:10.0pt; font-family:Arial'>Hi,</span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span style= =3D'font-size:10.0pt; font-family:Arial'> </span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'>Just a question about keyUsage= DataEncipherment.</span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'> </span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'>In a RSA-based certificate (the public key= included is coming from a RSA key pair), If I assert the DataEncipherment key usage,= I’m telling to the relying third parties that my certificate (the public key included into) can be used for raw data encipherment, but, in the reality,= RSA algorithm cannot be used to cipher large amounts of data (as opposed to a symmetric key, in, for example, a CBC mode). The common technique with that= is to use the keyEncipherment bit to be able to cipher the symmetric session= key used to cipher the raw data.</span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'> </span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'>Can anyone point me to the right direction, or provide me some links to technical documentation, about real-life usages= with the Data Encipherment key usage??. Are there any other technologies/algorithms= for RSA where this key usage has more sense??.</span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'> </span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'>Thank you very much in advance</span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'><br> Carlos</span></font></p> <p class=3DMsoNormal><font size=3D2 face=3DArial><span lang=3DEN-GB style= =3D'font-size: 10.0pt;font-family:Arial'> </span></font></p> <p class=3DMsoNormal><span class=3Dname1><b><font size=3D2 face= =3DVerdana><span style=3D'font-size:10.0pt;font-family:Verdana'>Carlos Gonz= =E1lez-Cadenas</span></font></b></span><b><font size=3D2 face=3DVerdana><span style= =3D'font-size:10.0pt;font-family:Verdana; font-weight:bold'><br> </span></font></b><span class=3Dcargo1><font size=3D1 face=3DVerdana><span style=3D'font-size:8.5pt;font-family:Verdana'>Jefe de Departamento de Arquitectura Software</span></font></span><font size=3D1 face= =3DVerdana><span style=3D'font-size:8.5pt;font-family:Verdana'><br> <span class=3Dcargo1><i><span style=3D'font-style:italic'>Software= Architecture Department Manager</span></i></span> <br> <span class=3Dcargo1>Responsable de Seguridad de la Informaci= =F3n</span><br> <span class=3Dcargo1><i><span style=3D'font-style:italic'>Information= Security Responsible</span></i></span> <br> <br> </span></font><span class=3Dlogo1><b><font size=3D4 face=3DVerdana><span style=3D'font-size:14.5pt;font-family:Verdana'>e-xtend<font color= =3D"#0099cc"><span style=3D'color:#0099CC'>force</span></font></span></font></b></span><font= size=3D1 face=3DVerdana><span style=3D'font-size:8.5pt;font-family:Verdana'><br> </span></font><span class=3Dtextlogo1><font size=3D1 color=3D"#999999" face= =3DVerdana><span style=3D'font-size:7.5pt;font-family:Verdana'>by e-xtendnow= BancSabadell</span></font></span><font size=3D1 face=3DVerdana><span style= =3D'font-size:8.5pt;font-family:Verdana'> <br> <br> C/Sena,12 nucli C planta 2<br> Edifici Landscape<br> 08190 Sant Cugat del Vall=E8s<br> <u>phone:</u> +34933556139<br> <u>fax:</u> +34933556142<br> <u>mailto:</u> <a href=3D"mailto:gonzalezcarlos@extendforce.com"><font color=3Ddarkblue><span style= =3D'color:darkblue'>gonzalezcarlos@extendforce.com</span></font></a> </span></font></p> <p class=3DMsoNormal><font size=3D3 face=3D"Palatino Linotype"><span style=3D'font-size:12.0pt'> </span></font></p> </div> </body> </html> <table><tr><td bgcolor=3D#ffffff><font color=3D#000000><pre>=0D Advertencia legal: Este mensaje y, en su caso, los ficheros anexos son confidenciales, especialmente en lo que respecta a los datos personales, y se dirigen exclusivamente al destinatario referenciado. Si usted no lo es y lo ha recibido por error o tiene conocimiento del mismo por cualquier motivo, le rogamos que nos lo comunique por este medio y proceda a destruirlo o= borrarlo, y que en todo caso se abstenga de utilizar, reproducir, alterar, archivar o comunicar a terceros el presente mensaje y ficheros anexos, todo ello bajo= pena de incurrir en responsabilidades legales. El emisor no garantiza la= integridad, rapidez o seguridad del presente correo, ni se responsabiliza de posibles perjuicios derivados de la captura, incorporaciones de virus o cualesquiera otras manipulaciones efectuadas por terceros. </pre></font></td></tr></table> <table><tr><td bgcolor=3D#ffffff><font color=3D#000000><pre>=0D Advertiment legal: Aquest missatge i, si escau, els fitxers annexos tenen caire confidencial, especialment pel que fa a les dades personals, i s'adrecen exclusivament al destinatari referenciat. Si no es tracta d'aquest i l'ha rebut per error o= se li ha fet arribar per qualsevol motiu, li preguem que ens ho comuniqui per= aquesta mateixa via i el destrueixi o l'esborri, i que en tot cas s'abstingui= d'utilitzar, reproduir, alterar, arxivar o comunicar a tercers aquest missatge i fitxers annexos, tot sota pena d'entrar en responsabilitats legals. L'emissor no= garanteix la integritat, la rapidesa o la seguretat d'aquest correu, ni es= responsabilitza de possibles perjudicis derivats de la captura, incorporacions de virus o qualsevol altres manipulacions que facin tercers. </pre></font></td></tr></table> <table><tr><td bgcolor=3D#ffffff><font color=3D#000000><pre>=0D Disclaimer: This message and any attached files transmitted with it, is confidential, especially as regards personal data. It is intended solely for the use of= the individual or entity to whom it is addressed. If you are not the intended= recipient and have received this information in error or have accessed it for any= reason, please notify us of this fact by email reply and then destroy or delete the= message, refraining from any reproduction, use, alteration, filing or communication= to third parties of this message and attached files on penalty of incurring legal responsibilities. The sender does not guarantee the integrity, the= accuracy, the swift delivery or the security of this email transmission, and assumes no responsibility for any possible damage incurred through data capture, virus incorporation or any manipulation carried out by third parties. </pre></font></td></tr></table> ------=_NextPart_000_002A_01C432B9.3E5430E0-- Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45BQkFL007169; Wed, 5 May 2004 04:26:46 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i45BQk4b007168; Wed, 5 May 2004 04:26:46 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from fw.chttl.com.tw (gate.chttl.com.tw [202.39.157.254]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i45BQi9R007136 for <ietf-pkix@imc.org>; Wed, 5 May 2004 04:26:45 -0700 (PDT) (envelope-from wcwang@cht.com.tw) Received: from ms21.chttl.com.tw (ms21 [10.144.2.121]) by fw.chttl.com.tw (8.12.10/8.12.10) with ESMTP id i45BQWGG019139; Wed, 5 May 2004 19:26:33 +0800 Received: (from root@localhost) by ms21.chttl.com.tw (8.12.10/8.12.10) id i45BQXJH023820; Wed, 5 May 2004 19:26:33 +0800 Received: from wcwang ([10.144.133.79]) by ms21.chttl.com.tw (8.12.10/8.12.10) with SMTP id i45BQWL5023724; Wed, 5 May 2004 19:26:32 +0800 Message-ID: <00c101c43293$cfa00a90$4f85900a@wcwang> From: "Wen-Cheng Wang" <wcwang@cht.com.tw> To: "PKIX List" <ietf-pkix@imc.org> Cc: "Steve Hanna" <Steve.Hanna@sun.com> References: <408FB75B.50108@sun.com> Subject: Re: Comments on Path Building I-D Date: Wed, 5 May 2004 19:26:29 +0800 MIME-Version: 1.0 X-scanner: scanned by Inflex 1.0.10 - (http://pldaniels.com/inflex/) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Dear List, After reading the certpathbuild draft and the comments from Steve Hanna, my feeling is that this draft is more like a paper (or thesis) than a PKIX specification. I believe that is why Steve criticized the text of the draft as if he is a reviewer of a PKI paper. I really admire the authors and Steve's rich and deep knowledge of optimization and implementation technique for path building. I must confess that I gained much from reading "the paper" and the comments from Steve. However, I doubt whether it is suitable for the PKIX WG to publish "the paper" as an RFC. I think the text of the draft is good enough to become a paper on a PKI conference proceedings or even on a journal. But, that does not mean it will be a good PKIX RFC. As a faithful reader of PKIX specifications, I would expect that if the PKIX WG would like to publish an RFC entitled "Certification Path Building" then its text should focus on: 1. defining what is a legal certification path Can certificates with names in different encoding be chained? Can certificates with unmached issuer name and subject name be chained by matching SKID and AKID? 2. defining a algorithm for certification path building, also including specifying the inputs to the algorithm and specifying the outputs from the algorithm 3. specifying the format of certificate/CRL extensions (e.g. SKID, AKID, AIA, SIA, CRLDP...) that are related to certification path building. E.g., the URL format of different procols (such as LDAP, HTTP, FTP...) (e.g., Should the LDAP URL contains the ";binary" option?) 4. specifying how a compliant implementation should process theose certificate/CRL extensions related certification path building 5. specifying what is the minimum set of repository retrieving protocols (such as LDAP, HTTP, FTP, SMTP...) a compliant implementation sould support and how these protocols should be supported. E.g., What is the expected object format? (e.g., DER binary vs.Base64-encoded) What is the expected MIME type if HTTP is used? 6. discussing security consideration I believe that the PKIX certpathbuild specification should specify a baseline certification path building algorithm in the fashion of RFC 3280 specifying a basic certification path validation algorithm. The purpose of defining the algorithm is for precisely describe the expected result of the certification path building. There is no need for a PKIX specification to teach implementators the optimization and implementation technique for path building. It is up to the implementators themselves to choose their own the optimization and implementation technique. Isn't it? Therefore, I would like to suggest the authors to complete remove the section 3 of the current draft. It is not only because a PKIX specification is not supposed to be a paper or textbook for teaching the optimization and implementation technique for path building, but also because the optimization and implementation technique described in that section is not, as Steve said, the consensus of the PKIX WG. I suggest that the draft sould spend more space and time to cover the areas I mentioned above. (I know the current draft already covers them, but I think it is still not in-depth enough.) Wen-Cheng Wang ----- Original Message ----- From: "Steve Hanna" <Steve.Hanna@Sun.COM> To: "PKIX List" <ietf-pkix@imc.org> Sent: Wednesday, April 28, 2004 9:53 PM Subject: Comments on Path Building I-D > Here are my comments on draft-ietf-pkix-certpathbuild-03.txt. > > I suspect that these will come off sounding negative. > I'm sorry about that. I'm glad we have this document > and I hope it will progress in a modified form, but > I have some serious problems with the document as > it now stands. > > I also apologize for the length of these comments. > I wanted to do a careful review of the document > and there are a lot of problem areas (although > some are minor). The fact that nobody else has > commented on these leads me to suspect that nobody > except the authors and myself has carefully reviewed > the document. Can anyone refute that suspicion? > > Thanks, > > Steve > > -------- > > Overall Comments: > > * My main overall comment is the one I made last July. > There is not consensus on path building techniques > in the PKIX working group or the IETF. Why? Because > we don't understand the problem thoroughly yet. > > In this document, you advocate depth first search > and suggest that building forward is usually best. > I don't think you have carefully considered a wide > variety of other algorithms like breadth first search, > meet in the middle, etc. Can you show me any solid > evidence that shows your algorithms are better than > the alternatives? > > I doubt that you have any such proof. In fact, it > seems clear to me that depth first search is a poor > algorithm for building cert paths. I say this as a > person who has implemented that algorithm myself > and regretted it. The problem is that if you make > a wrong choice, you must completely explore that > part of the PKI before you consider backing out > and trying other options. I have heard stories of > path building taking 35 minutes or more when there > was a valid short path because the builder used > DFS and headed down the wrong path. > > We need to do our homework before making any solid > recommendations to implementers. I suppose it's better > to have some advice than none, but this document > should have a prominent warning that these techniques > are experimental. In fact, we might want to give > the RFC Experimental status. > > As I've suggested before, I want to see a careful > analysis of different path building strategies. > Theoretical analysis will probably be useful, > but I expect it will also be useful to compare > real-world performance with a variety of topologies > (some from real-world deployments, some looking at > possible futures like many cross-certified bridges). > We need to try out a lot of ideas without getting > committed to any too early. The cert path building > panel at the PKI R&D Workshop this year was a good > start, but we need to do more in-depth work. I've > started talking to some other researchers about this. > I hope that this documents' authors will join this > effort. Their experience and aid will be valuable. > > * At several places in the document, you refer to the > "authors' opinions". If this was truly a working > group document, it would reflect the consensus of > the working group, not the opinions of the authors. > As I said before, I don't see working group consensus > on this topic. Has anyone other than me (and maybe > the authors) carefully reviewed the current draft? > I haven't seen any comments on it during Last Call. > > Since this document does not reflect working group > consensus, I do not think it should go forward > as a working group draft. It should be an individual > submission. But I suppose once it becomes an RFC > nobody will ever know whether it was a working group > submission or not. And I really do think it's useful > to have some guidance on cert path building (even if > we don't understand the problem very well). So I'd be > OK with having it be a working group submission if it > goes to Experimental status and a caveat is added > at the beginning warning that this is only the authors' > opinion (albeit an educated opinion) and that further > study is under way to determine which techniques are > actually preferred. > > * You should include in this document a section with > guidance to PKI designers about what they can do to > make path building easier. Here are some ideas: > > Use a simple topology (hierarchical with one root), > when possible. Include AIA, SIA, and CRLDP extensions > in certificates. Make certificates (especially CA > certificates) and CRLs easily available by LDAP and > HTTP, populating both sides of the crossCertificatePair > attribute. Use name constraints and policy constraints > whenever possible, especially at high fanout points > like bridges. Avoid having more than two high fanout > CAs (at most) between any two points, if possible. > > * I have a *serious* objection to the idea that someone > is going to specially configure path building software > for a particular PKI environment (as I think you imply > at the end of section 2.3, early in section 3.4, and in > a few other places). This is not practical. Most > organizations do not have a PKI wizard on staff > (especially a path building wizard). The path building > software must just work, automatically. If we can't do > that, PKI will never be widely used (at least, not in a > non-hierarchical topology). Fortunately, I see no reason > why we need special configuration. With AIA and SIA and > CRLDP and some good algorithms, we should be able to > build paths just fine in almost any PKI without any > configuration. > > Substantive Comments: > > * Section 1.2 (Purpose) says "... this document suggests > using ... depth first tree traversal. ... Other approaches > (e.g., building complete spanning trees of the PKI.) exist > and may be shown to be more effective under certain conditions." > > Building a complete spanning tree of the PKI is not a > decent solution. Please change this to something more > reasonable, like breadth first search. Otherwise, you're > just setting up a straw man, an impractical alternative. > > * In section 1.4.1, you mention one disadvantage of the trust > list approach. You might want to mention the problem that > compromise of any trusted certificate compromises the > entire system. > > * In section 1.4.2, you say that a partial mesh is a mixture > of unidirectional and bidirectional cross-certifications. > It's probably also important to note that in a partial mesh > there may be CAs that are not directly cross-certified at all. > > * What is a Root CA doing in Figure 3? As you say, each EE > in a mesh usually trusts the CA that issued its certificate. > Because of this Root CA, Figure 3 does not depict a full mesh, > as stated in 1.4.2. I suggest that you remove the Root CA. > > * At the end of section 1.4.3, you raise the concern that > the assurance of a high-assurance PKI may be diluted by > cross-certifying with a less restrictive PKI. If you're > going to raise this concern, you should mention that it > can be addressed through the use of certificate policies. > > * At the end of section 1.4.4, you say that connecting > PKIs with a bridge results in a non-hierarchical PKI. > Certainly, this is true. But mesh and hybrid PKIs are > not hierarchical either. Why raise this here? And the > following sentence which states that any PKI can be > considered hierarchical from the right perspective > only applies if you remove and duplicate links in the > PKI. Since you don't have space to get into this here, > I suggest you remove those last two sentences. > > * At the end of the first paragraph of section 2.1, you > explain that S/MIME messages commonly include certificates. > I think you would be wise to also mention SSL/TLS > since this is the most widespread use of PKI and it > also supplies certificates in the protocol. > > * Before Figure 6, you explain how certificates are > depicted with arrows in your figures. You should also > explain the "B(A)" notation you use. > > * At the end of the paragraph after Figure 6, you say > that building paths is as important as validating them. > I don't agree. Many products have been shipping for > years and work just fine without building. In a complex > PKI, building is important. But in a simple hierarchical > PKI, it's not. > > * At the end of section 2.1, you have a large discussion > of why you believe building forward is usually better. > This duplicates later discussions on this topic. I suggest > you save this for later. > > * In section 2.2, you could simplify and clarify Criterion 1 > by changing it from this: > > Criterion 1: The implementation is able to find all possible paths. > By this, it is meant that all possible certification paths between > the trust anchor and the target certificate which may be valid paths > can be built by the algorithm. > > to this: > > Criterion 1: The implementation is able to find all possible paths. > By this, it is meant that all valid certification paths between > the trust anchor and the target certificate can be built by the > algorithm. > > * In section 2.2, Criterion 2 seems rather odd. Who cares which > invalid paths you build first? The important thing is how quickly > and efficiently you can build a valid path or determine that > no valid path exists. > > * In section 2.3, you say that because certificates are not > permitted to repeat, every potentially valid path has a > terminus. But every potentially valid path *always* has > a terminus, even if certificates are allowed to repeat. > As defined in RFC 3280, a certification path is a collection > of n certificates. Every path has a finite number of certificates. > I suggest you remove the sentence "As a result, every potential > valid path has a terminus, a leaf on the tree." > > * In the following paragraph, you say that removing and > duplicating nodes and links in a PKI to turn it into > a hierarchy greatly simplifies software design. This > may be so, but it also removes many opportunities for > insight and optimization. For example, it increases > the number of nodes in the data structure and makes it > harder to mark a certain area of the PKI as unproductive > (since that area may appear several places in the tree). > > * Later in that paragraph, you use the phrase "decision tree" > without defining it. I suggest that you define it in > section 1.3. > > * One paragraph on page 16 (starting with "A more complicated > example") talks about problems with building in the > reverse direction when there are many trust anchors. > The real issue here is fanout. Whether you're building > forward or building reverse, it's bad news when you get > to a point where you have several choices. It's especially > bad if your heuristics (ranking) don't give you much > help in deciding which way to go. And it's even worse > if you're using depth-first search, since one wrong > move can send you off in the wrong direction for hours. > That's why I suggest breadth first search or (even > better) ranking all candidates at all points in > the tree at each decision time. > > Four trust anchors is a four-way fanout. A similar > situation would arise if you were building forward > and arrived at a CA that has four certificates with > it as the subject. In either case, you hope that your > ranking will help choose the right path. And if you're > using depth first search, you really hope that you > don't take a wrong turn. One technique for dealing > with extreme fanout with fairly equal rankings is to > start building from the other end. Another is to rank > the certs at the fanout point low and try another branch. > A third is to use DPD to get help. And a fourth is to > tell PKI designers to use name constraints and other > techniques at high fanout points (like bridges) to > help path building succeed. > > I hope that you find this analysis useful. > > * In Figure 10, are W, X, Y, and Z all trust anchors? > I think so. That's pretty odd. Why would a relying > party have all four of those trust anchors? Who needs > the bridge CA in that case? I suspect that this > topology was created to support your arguments that > repeating a (name, key) pair is bad and that building > in reverse is bad. If so, I think you can do better. > > Let's do a careful theoretical analysis and try > out different algorithms on different topologies. > I suspect you're right that repeating a (name, key) > pair is bad but we need to think about this carefully > since it is a significant change to RFC 3280. I think > you will see below that building in reverse is a > useful tool that should be used in conjunction > with building forward. But we need to show the > merits of our approaches through analysis, not > by consructing topologies that serve our own purposes. > > * In the paragraph after Figure 10, you have several > sentences explaining the diagram's notation. This > was explained much earlier. The rest of the paragraph > (explaining where certificates would be stored in > an LDAP directory) also duplicates material found > elsewhere. I suggest that you remove these, since > the document is already too long. > > * In section 2.4.2, you seem to be proposing that > software should build a decision tree. I believe > you don't intend for the software to actually > build a tree for the entire PKI but only to > move incrementally through the PKI adding nodes > and links to the tree as needed. I hope that's > the case, anyway. Otherwise, you'd need to > download all certs in the PKI! If I'm right, > you need to be much clearer about this. I could > easily imagine a developer writing code that > actually builds the tree. That code would > work fine in a small test PKI but flood the > directory with requests and then die in a > large PKI. > > * After Figure 11, you point out that building > in reverse is not good in this case. Sure. > The EE is in a simple hierarchy. Of course, > building forward is best there. If you don't > know the topology (especially if you have > several trust anchors), starting with forward is > fine. Then you can switch or meet in the middle > if things get hairy. > > * In section 2.6, you say "[...] the path builder > only needs to store the current location in > the tree [...] All completed branches can be > discarded from memory [...]". Maybe. There's > a lot to be said for retaining records of > paths you have already tried and rejected. > Then you can avoid retrying them at a different > point in the graph (unless conditions are > sufficiently different to merit a retry). > > * Later in section 2.6, you say "Consider HTTPS > support" for CRLDP and AIA. Why would this be > valuable? You're retrieving signed data. Using > HTTPS may trigger another path build, maybe > even a loop where one build triggers another > which triggers another and so on. I suppose > some repositories may require HTTPS to > authenticate the client or protect the > certs from prying eyes. But you should probably > warn that doing so may be expensive and that > implementers should protect against loops. > > * Toward the end of section 2.6, you talk about > the path cache. You call for "a configurable > expiration date for each entry". Who would > configure the date and why? I'm a path building > geek and I can't imagine configuring such a > thing! > > * A few sentences later, you suggest storing > issuer-subject cert relationships. If you > want to do this, I suggest that you use a > cert hash instead of an (issuer, serial) > tuple. Issuer, serial is *not* always unique, > especially across multiple PKIs or when > one CA is malicious (and remember that some > CAs are only partly trusted in a modern PKI). > > * In section 2.7.1, you talk about the required > inputs. Instead of supplying the actual Target > Certificate, it is sometimes useful to provide > criteria that describe what sorts of target > certificates you're willing to accept. For instance, > when doing S/MIME encryption to a new correspondent > you may not have the correspondent's encryption > certificate. The path building software can > help find it if you tell it what you need. > > * In section 3.2, you recommend that path building > software output a detailed log. I think you should > recommend that this log be carefully structured > so that the paths tried can be easily reconstructed > by a diagnostic program. My team built a tool that > shows the cert graph and then replays the paths > our builder tried, lighting up each one in sequence. > We found this a great help in understanding the behavior > of our software (finding bugs, improving algorithms, > etc.). It's also a very cool way to demo path building, > which otherwise is a pretty boring demo ("see, there's > the web page!"). > > * Later in that paragraph, you say "Ideally, there would > be a mechanism for turning this logging on and off [...]". > I'd change this to "There should be [...]". Logging is > expensive (in storage and CPU time), especially for path > building where you commonly try hundreds of certificates > or more to create a valid path. You really need a way to > turn logging off and on. > > * A few paragraphs later, you say "it may be useful to > not rule out any paths" and just return each path as > its built, even if it's invalid. The problem with this > is that (especially with a depth first search) is that > there are many topologies that may cause you to wander > among many invalid paths. Why is this technique good? > You say it will "provid[e] a more rapid response to the > caller than one which builds all possible paths." Maybe. > I suspect you'll instead waste time by spending a lot > of time returning invalid paths to the caller. I don't > see much reason to return invalid paths except in a > diagnostic log or for your interesting idea of providing > one path that's *mostly* valid for debugging and in case > the caller finds that invalid path educational or compelling. > > * A few paragraphs later, you lay out your approach. > Here are some comments on it: > > You say "Do not check revocation status if it requires > a directory lookup or network access." Why not start the > revocation check and let it run in parallel while you do > other things (like checking certs deeper in the graph)? > That way, you'll have the revocation check done when you > need it. And if the check fails you can abort all work > that was based on the validity of that certificate. > Of course, you'll want to cache the revocation status > of certificates for some time. > > You say "Do not check digital signatures". Again, why > not run this in parallel? Most of path building time > (about 90%, based on our data) is spent waiting for > the network (while downloading certificates and such). > That's an ideal time to be checking signatures. And, > of course, you'll want to cache the results of signature > checks. > > You say "Do not check anything that can not be checked > as part of the iterative process of traversing the tree." > Why not? I expect you would want to run these final checks > (like full policy processing when building forward) before > returning the path to the caller. Otherwise, the caller > will need to run a complete validation of the path, which > will take much more time than just running these last > few checks. > > * In the last paragraph of 3.2, you say "Second, it is fairly > uncommon in typical application environments to encounter > a revoked key; therefore, most certificates validated will > not be revoked." In a PKI, we don't revoke a key. We revoke > a certificate. Therefore, this sentence should read "Second, > it is fairly uncommon in typical application environments to > encounter a revoked certificate." > > * Why include section 3.3 at all? It's very odd to have > an IETF spec talking about internal data structures. > Maybe you should just give everyone a copy of your > code (since it is apparently ideal) and save them the > trouble of implementing it! I'm going to skip my more > detailed comments on this section since I think it > should be entirely removed. > > * In section 3.4, you refer say "developers should evaluate > each method with respect to the environment that the > application will operate, and assign weights to each > accordingly in the path building software.". First, the > word "that" in this sentence should be "in which". Second, > this is a prime example of the awful idea that developers > must examine each PKI and configure their software to > run optimally there. > > * In section 3.5.1, you say "According to [RFC 3280], > basicConstraints is required to be present with cA=TRUE > in all CA certificates." Actually, section 6.1.4 (k) of > RFC 3280 says "Verify that the certificate is a CA > certificate (as specified in a basicConstraints extension > or as verified out-of-band)." Maybe your software doesn't > make any provision for out-of-band indication of CA > certificates, but you should probably at least note the > possibility that other software may do so. > > * In section 3.5.5, you say "Certificates in the path cache > have been validated previously. There is some probability > that the path from that certificate to a trust anchor is > still valid." The path may still be valid but it may > contain constraints that are inconsistent with the > path from the Target Certificate to this certificate. > You should probably note that possibility. > > * In section 3.5.6, you say that the Previously Verified > Signatures sorting method doesn't apply when building > in reverse. Actually, it does. Your analysis is incorrect. > This method never tells you which way the Target or > Trust Anchor is (which certificate is most likely). > It only lets you rule out certificates because the > signature check would certainly fail. > > * In section 3.5.7, you say the Path Length Constraints > sorting method "may be applied in reverse, but the > benefit is likely less than that of the forward direction". > Why? You can argue that the method is more effective > when building in reverse because path length constraints > often appear close to trust anchors. > > * In section 3.5.8, you say "Certificates that contain > nameConstraints that would be violated by certificates > already in the path to this point are given lower priority > (or perhaps eliminated)." They should definitely be > eliminated. You know they can't be used to build a > valid path, so what's the point (unless you're working > on building an invalid path)? > > * In section 3.5.9, you say "While this is viable, the > signature verifications required make it less attractive > as an elimination method." Usually, a CRL check only > requires one signature verification so "verification" > should be singular. > > You also say "It is suggested that this method only > be used for sorting and that CRLs are validated post > path building." As I noted above, fetching CRLs and > performing signature checks can be done in parallel > with other work. It's a good idea to do this so that > you can discover revoked certs before you have invested > too much time in them (and also so that you have the > results of the revocation check ASAP). > > You should also probably change this section to include > discussion of OCSP. Right now, it's very focussed on > CRLs. > > * Here's a sorting method similar to the "Issuer Found > in the Path Cache" method, but better suited to building > in reverse. Check whether the subject of the candidate > certificate matches the issuer of a certificate sent > by the target (in SSL/TLS, S/MIME, etc.). If so, then > the candidate certificate should be ranked more > highly because it is more likely to lead to a path > to the Target Certificate. You may also want to do > RDN Matching (as in 3.5.15) with the certificates > sent by the target. > > * In section 3.5.15 (on RDN matching), you say "Additionally, > it should be noted that this method can require a lot of > processing if many trust anchors are present." Actually, > this should only require a few hash table lookups (of the > RDNs). > > * Section 3.5.18 could be clearer. I think you're saying > that the subject and issuer of the candidate certificate > should have lots of RDNs in common. But this could be > understood to be saying that the subject of the > candidate certificate should have lots of RDNs in common > with the issuer of the next certificate. Of course, > they must match exactly so that can't be what you're > saying. But it would help if this section was more > explicit. > > * Section 3.5.19 also should be clarified. The description > of the reverse method is much clearer than that of the > forward method. I suggest that you replace the forward > method language with language based on the reverse method > language. One change to the reverse method language, though. > Instead of saying "If an entity named by a reverse certificate", > say "If the subject of a certificate". There's no such > thing as a "reverse certificate". > > Note also that just because you find a name in the > certificate cache, that doesn't mean you have the > right certificates. There may be several different > CAs with the same name. Also, you may now have more > clues (like AIA and SIA) for finding certificates > than you had before. You should be sure to pursue > any such new clues. > > * In the Justification for 3.5.19, you say "The presence of > required directory values populated in the cache increases > the likelihood that all the required certificates and CRLs > needed to complete the path from this certificate to the > trust anchor (or target if building in reverse) are present > in the cache from a prior path being developed [...]". > That's fine, but you might also want to keep the actual path > previously developed and look at that as a prime candidate > for the path this time. > > * In section 3.5.20, you use the presence of a CRL in > the CRL cache to indicate whether a complete path > through a CA has previously been found. This is > rather indirect. It would be better to actually > track which certs have been previously included > in valid paths. Keeping a copy of the path would > also be quite useful. > > * The discussion of Forward Policy Chaining in section 4 > overstates the benefits of policy checks when building > in the forward direction. We should discuss this more. > > * In section 5.2, you suggest that each name/key pair > only be allowed to appear once in a path. Since there > is no such requirement in X.509 or RFC 3280, this > would violate Criterion 1 as stated in section 2.2. > You would miss some valid paths. > > * In section 6 (Retrieval Methods), you should mention > getting certificates and CRLs from the target (with > S/MIME or SSL/TLS, for instance). That's a *great* > way to get certificates and CRLs. The ones you get > are usually highly valuable. > > * Sections 6.2 and 6.3 should include text that encourages > implementers to support AIA and SIA, especially HTTP. > This text from the end of section 6.4 could easily > be adapted: > > However, implementers of path building > and validation modules are strongly encouraged to support CRLDPs. At > a minimum, developers are encouraged to consider supporting the LDAP > and HTTP transports; this will provide for interoperability across a > wide range of existing PKIs. > > * In section 7, you should talk about prefetching and > parallel fetching. Both are good ways to improve the > performance of a cert path building implementation. > > * In section 7.1, you say "Certificate processing systems can > cache data retrieved from external sources for some period > of time, but not to exceed the useful period of the data > (i.e., an expired certificate need not be cached)." CRLs > are often issued well before the nextUpdate of the previous > CRL. Could you mention this and suggest that implementers > check periodically for updated CRLs? Similarly, > crossCertificatePairs may be updated on an unpredictable > basis. If you don't check back every so often, you'll > never know that a new cross certificate is available! > > * The last bullet in section 7.1 offers a few suggestions > for other things to cache. I'd add to that list previously > built paths and partial paths. If the cache is safe from > unauthorized modifications (as with an in-memory cache), > you may also want to cache validation and signature check > status for certificates and CRLs. > > * In section 7.2 (Retrieval Order), you may want to consider > checking repositories in parallel instead of serially. > Also, you might want to run ahead with your path building > and validation while a network query is ongoing. Maybe > you can build and validate the path with just the certificates > and CRLs you have already (thus eliminating the need to > wait for the network query to complete). Anything to get > the answer to the user more quickly! > > * In section 8.1, you should probably add a statement that > path building can be used as a denial of service attack. > Make a series of simple requests to a server and cause it > to do a bunch of long path builds. To avoid this, the > server can limit the amount of resources it's willing to > devote to a path build. This amount can be reduced when > the load is heavy. And standard DOS protections like > identifying attackers can be employed. > > * Section 8.2 goes way beyond RFC 3280 and X.509. I know > that Santosh is working on getting this into X.509 and > the successor to RFC 3280. I'll debate the specifics > of his proposal in that context. But for this document, > I think you should be careful to explain that this goes > beyond the standards. If you implement it, you will > violate Criterion 1 from section 2.2. Maybe the best > thing to do is to just briefly point out that there's > an issue here and it's under discussion in the standards > groups. Once it's settled there, this document can be > revised to reflect whatever's agreed on. > > Minor Comments: > > * The first paragraph of section 2 says "... guidance is > provided on the capabilities path building implementations > required in order to ...". I think you want "require" or > (maybe better due to RFC 2119) "need" instead of "required". > > * Just below Figure 6, "the certificates cannot be validated" > should be "certificates cannot be validated". You're not > referring to any particular certificates, just certificates > in general. > > * In the paragraph after that, you should say "the > crossCertificatePair attribute" instead of "the cross > certificates". Otherwise, it's not clear what you mean. > > * In the second paragraph of section 3.2, you say a > certificate "may appear again on another point in > the tree". That should be "at another point in > the tree". > > * At the end of section 3.4, "The list" should be "the list". > > * In section 3.5.17, there are two typos. "it priority" > should be "it has priority". "will not successfully" > (in the last paragraph) should be "will not verify > successfully". > > * In section 5.4, you say certificates "will be" required > to use UTF-8 after January 1, 2004. This should now > be changed to "are". > > * In section 6, the first sentence should mention CRLs > also. I suggest that the second sentence say "obtaining > these certificates and CRLs" instead of "performing > this retrieval", since retrieval has not yet been > mentioned and sometimes the certificates can be > obtained without having to be retrieved (as when the > EE sends them in SSL). > > * In section 6.1, the description of userCertificate > should probably say that it "contains certificates > issued by one or more certification authorities with > a subject DN that matches that of the directory entry". > Also, there's an extra period at the end of the > parenthetical comment later in the paragraph. > > * Later in section 6.1, the description of issuedByThisCA > should say it contains certificates issued to *other* > CAs. And there's an extra comma after "If both elements > are present" (later in that paragraph). Why include that > sentence anyway. The relying party doesn't need to > enforce this requirement. People may think they need > to do so. > > * In section 6.2, you say "AIA may be used to retrieve one or > more certificates for entities that issued the certificate > containing the AIA extension." I think it would be clearer > to say "the CA" instead of "entities". > > * At the end of the first paragraph of section 6.3, "AIA" > appears twice where it should be "SIA". > > * In section 6.4, you say "the CRL(s) to which the CRLDP(s) refer > is the CRL that would contain revocation information for the > certificate." I'd say instead "the CRL(s) to which the CRLDP > refers are generally the CRLs that would contain revocation > information for the certificate." Why? Because a CRLDP may > refer an LDAP directory entry that contains many CRLs. Some > of them may not pertain to this certificate. And of course > there's usually only one CRLDP in a certificate. > > * In section 8.2, "q CRL Signer" should be "a CRL Signer". > > * In the Informative References, you include "[MINHPKIS]" > but it's never actually referred to anywhere in the document. > Similarly for [X.501] and [X.520]. You should probably > remove those references. If you're going to retain them > for some reason, please provide a decent reference for > [MINHPKIS]: publisher, publication date. > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i453dZLt031438; Tue, 4 May 2004 20:39:35 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i453dZsg031437; Tue, 4 May 2004 20:39:35 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from e2.ny.us.ibm.com (e2.ny.us.ibm.com [32.97.182.102]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i453dYGD031401 for <ietf-pkix@imc.org>; Tue, 4 May 2004 20:39:34 -0700 (PDT) (envelope-from tgindin@us.ibm.com) Received: from northrelay04.pok.ibm.com (northrelay04.pok.ibm.com [9.56.224.206]) by e2.ny.us.ibm.com (8.12.10/8.12.2) with ESMTP id i453dZiJ399862; Tue, 4 May 2004 23:39:35 -0400 Received: from d01ml062.pok.ibm.com (d01av02.pok.ibm.com [9.56.224.216]) by northrelay04.pok.ibm.com (8.12.10/NCO/VER6.6) with ESMTP id i453e5dm049804; Tue, 4 May 2004 23:40:06 -0400 To: "Santosh Chokhani" <chokhani@orionsec.com> Cc: ietf-pkix@imc.org MIME-Version: 1.0 Subject: RE: Cross certificate format X-Mailer: Lotus Notes Release 5.0.11 July 24, 2002 From: Tom Gindin <tgindin@us.ibm.com> Message-ID: <OF63F34A40.14A2103B-ON85256E88.007E4CB1-85256E8B.00141423@us.ibm.com> Date: Tue, 4 May 2004 23:39:26 -0400 X-MIMETrack: Serialize by Router on D01ML062/01/M/IBM(Release 6.0.2CF2 HFB2 +DEBUG|April 23, 2004) at 05/04/2004 23:39:34, Serialize complete at 05/04/2004 23:39:34 Content-Type: text/plain; charset="us-ascii" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Santosh: Of course, defining a trust anchor implies that the RP trusts the anchor's public key as an issuer for at least some purposes. RFC 3280, in the parts of section 6.1.2 I cited, appears to forbid those state variables from being set from information defined by the RP or stored with the trust anchor. This is consistent with its statements in 6.1.1 d which permit self-signed certificates but say nothing about other CA certificates. As is probably clear to everyone reading this thread, I think that this is not an ideal set of suggestions. The current standard does contain some constraints on the trust anchor (notably inhibit-policy-mapping), but neither name constraints nor basic constraints. I consider this an anomaly. If I had free rein in rewording pieces of RFC 3280 section 6.1, I would change this wording to allow CA certificates along with self-signed ones as anchors and create extra optional (and optional to support) fields to be set from the trust anchor. I don't think that this is particularly inconsistent with general security principles, or with any part of X.509 with which I am familiar. It does represent an extension of RFC 3280 section 6, and it is only backward compatible with it, so it may be out of scope. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 05/01/2004 06:46 PM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: The standard does not require you to check how you initialize the variables. You can get these from the cross certificate or from some configuration. The basic point is that you trust the public key. How you constrain it is not address by the standard, where as if the same certificate appeared in a path, you have to apply the constraints or be not compliant with the standard. You obviously do not believe in re-incarnation. Just kidding...... -----Original Message----- From: Tom Gindin [mailto:tgindin@us.ibm.com] Sent: Saturday, May 01, 2004 6:31 PM To: Santosh Chokhani Cc: ietf-pkix@imc.org Subject: RE: Cross certificate format Santosh: It is well known that trust in a trust anchor can be defined in a variety of ways. However, there is every reason to think that an RP may constrain that trust and that the constraints he may apply include those which an issuer CA may apply to a subordinate one. The wording of RFC 3280 section 6.2's first paragraph is an example of this. My comments about the meaning of a trust anchor are not purely theoretical. If a cross certificate is used to represent a trust anchor, the variables permitted_subtrees (RFC 3280 6.1.2 b) and excluded_subtrees (RFC 3280 6.1.2 c) could easily be set from the contents of a NameConstraints extension, and the variable max_path_length (RFC 3280 6.1.2 k) from the contents of a BasicConstraints extension. In RFC 3280 today, these variables are set to constants (although that interpretation is not absolutely clear for permitted_subtrees). In fact, they are three of the four variables which are set to constants during initialization. If permitted subtrees were set to a list of attributes, the validation process would fail if the anchor CA or a CA it certified went beyond the restrictions of those attributes. Likewise, if the excluded subtrees were set to a list of attributes, the validation process would fail if the anchor CA or a CA it certified went used values matching any of those attributes. I do not believe that I am arguing against Galileo. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/29/2004 11:12 PM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: The basic point (at the risk of repeating myself) is that a CA certificate or a self-signed certificate can be used as a means to provide trust anchor information. In that case, there is no need to check signature or anything. But, when the CA certificate is a certificate in the certificate path, X.509 logic kicks in. This is consistent with X.509, 3280 and general security principles since the trust in a trust anchor can be derived through any number of means. If you do not agree, feel free to argue against the definitions and Galileo -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Thursday, April 29, 2004 8:43 PM To: Santosh Chokhani Cc: ietf-pkix@imc.org Subject: RE: Cross certificate format Santosh: I don't think that a CA certificate and a trust anchor are as much different things as different uses for the same thing. They are both identifiers of issuers which bind a public key to an issuer name, and it is perfectly appropriate for any CA certificate to represent a trust anchor. I also do not see why such things as name constraints on a trust anchor are inappropriate. It is perfectly true that verifying a trust anchor's certificate, when issued by another CA (who may not be trusted) is a pointless operation. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/27/2004 09:14 AM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: My point is that even if the trust anchor is a self-signed certificate, all you need to do is extract the required information. There is no need to examine anything. Cross certificate and trust anchor are very different things. An element of the cross certificate pair is nothing but an intermediate CA certificate in a certification path. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Tuesday, April 27, 2004 7:33 AM To: Santosh Chokhani Cc: ietf-pkix@imc.org; owner-ietf-pkix@mail.imc.org Subject: RE: Cross certificate format Santosh: Of course, you are right that RFC 3280 6.1.1 d permits trust anchor information to be provided outside a certificate. If it is so, no checks of the sort I indicated can be performed. It also says that it can be provided as "a self-signed certificate", with no further qualification. I do think it is somewhat odd that a self-signed certificate with KeyUsage set to typical EE values would be considered a valid issuer as a trust anchor while a cross certificate would not. You can always extract the needed anchor information from a cross-certificate, of course. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/26/2004 08:51 AM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: My reading of 3280 and X.509 is that a trust anchor need not even be a certificate. Thus, no checks need to be performed on a trust anchor at all. You get the DN, public key, algorithm OID, and parameters (if applicable) from a trust anchor. Any checks are gratuitous and not required by either standard. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Sunday, April 25, 2004 7:06 PM To: Jean-Marc Desperrier Cc: ietf-pkix@imc.org Subject: Re: Cross certificate format Jean-Marc: RFC 3280 doesn't say that anywhere. It does say that you don't need to validate the trust anchor as part of the path, but it isn't clear from the text that a v1 certificate is acceptable as a trust anchor - and it certainly isn't clear that a v1 certificate is acceptable as an issuer if it is a trust anchor while a v3 certificate with KeyUsage= { DigitalSignature, keyEncipherment } is not acceptable as an issuer even if it is a trust anchor. I believe that the correct rules for versions are something like the following: a certificate issued by a trust anchor which is represented by a certificate is verifiable if the anchor certificate is either a v1 certificate or a v3 certificate with the CA flag present in the basicConstraints extension. If the anchor certificate is a v3 certificate with the KeyUsage extension present, it is only a valid issuer certificate if keyCertSign is set. Tom Gindin Jean-Marc Desperrier <jmdesp@free.fr> Sent by: owner-ietf-pkix@mail.imc.org 04/21/2004 03:42 PM To: ietf-pkix@imc.org cc: Subject: Re: Cross certificate format Tom Gindin wrote: > RFC 3280 does not support the use of v1 self-signed > certificates, >which contain no extensions at all (see the text of RFC 3280's >basicConstraints section). However, a number of public CA's still have >them. > > Applications implementing the path validation algorithm of RFC3280 MUST accept them when they are used as trust anchors. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i44HM6nb004456; Tue, 4 May 2004 10:22:06 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i44HM6Rl004455; Tue, 4 May 2004 10:22:06 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from exchange.microsoft.com ([131.107.8.3]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i44HM6N9004448 for <ietf-pkix@imc.org>; Tue, 4 May 2004 10:22:06 -0700 (PDT) (envelope-from trevorf@exchange.microsoft.com) Received: from DF-VRS-01.redmond.corp.microsoft.com ([157.54.4.14]) by exchange.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 4 May 2004 10:22:09 -0700 Received: from 10.197.0.83 by DF-VRS-01.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 04 May 2004 10:22:09 -0700 Received: from DF-GROMMIT-01.mercury.corp.microsoft.com ([10.197.0.61]) by DF-BEG.mercury.corp.microsoft.com with Microsoft SMTPSVC(5.0.2195.6713); Tue, 4 May 2004 10:22:09 -0700 x-mimeole: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: FW: scvp Date: Tue, 4 May 2004 10:22:07 -0700 Message-ID: <33E7A1613A3480448996047B69308A2F02713A06@df-grommit-01.dogfood> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: FW: scvp thread-index: AcQxwLsmZ8FgwcdIRS2lew779l5wYwAOIYEg From: "Trevor Freeman" <trevorf@exchange.microsoft.com> To: "Denis Pinkas" <Denis.Pinkas@bull.net> Cc: <ietf-pkix@imc.org> X-OriginalArrivalTime: 04 May 2004 17:22:09.0831 (UTC) FILETIME=[54DA0370:01C431FC] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i44HM6N9004449 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Hi Denis, There is no mandate in 3379 that a validation policy must be expressed with a single value. It is not clear that is any value is doing so since it only raised more ambiguity. This is well trod turf by many groups who have debated the merits of providing suits vs. "al a carte" combinations of inputs and the broader consensus is for a "al a carte". Trevor -----Original Message----- From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] Sent: Tuesday, May 04, 2004 3:15 AM To: Trevor Freeman Cc: ietf-pkix@imc.org Subject: Re: FW: scvp Trevor, > Hi Denis, > > 3379 does not require that the validation policy be specified in a > single value. You are playing with words. The text says: "most clients will simply reference a validation policy" This means that validation policies can be referenced and thus parameters do not need to be defined through this interface. A simple and good reference is an OID. > With SCVP14 a client can accept the default of the SCVP > server or specify a set of parameters which constitutes its validation > policy. That is consistent with the requirements of 3379. The specification of the set of parameters should be done through a different interface. This different interface can then return an OID Whatever this additional interface can be, it will never be rich enough to pass all the details of the validation policy. Thus OIDs can also be defined not using this additional interface. Denis > Trevor > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Thursday, April 29, 2004 3:54 AM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: Re: FW: scvp > > Trevor, > > >>Hi Denis, > > >>Can you please be more specific how you think SCVP 14 does not comply >>with 3379. >> >>I can find no section in 3379 where is there a requirement that a >>validation policy MUST be represented as an OID. > > > There cannot be a requirement with such a level of detail, but the text > states: > > The protocol MUST allow the client to include > these policy dependant parameters in the DPV request; however, it is > expected that most clients will simply reference a validation policy > for a given application or accept the DPV server's default > validation > policy. > > A simple reference is an OID. > > FYI, I do not expect to use the "default validation policy". > > Denis > > > >>Given hiding the detail of what a policy is within an OID simply opens >>the rat hole of what change to the policy constitutes a change to the >>OID, it is less ambiguous to refer to the primary data. Once you are > > in > >>the business of managing state on a client, then there is negligible >>cost increase incurred in managing several data points vs. a singe > > data > >>point. >> >>Trevor >> >>-----Original Message----- >>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] >>Sent: Tuesday, April 27, 2004 11:01 AM >>To: Trevor Freeman >>Cc: ietf-pkix@imc.org >>Subject: Re: FW: scvp >> >>Trevor, >> >> >> >>>HI Denis, >>>In SCVP13, the concept of validation policy was not completely in >>>alignment with 3379 definition. >> >> >>Well, it is different and this is a major problem. >> >> >> >>>It also blurred the distinction between >>>validation policy and validation algorithm. >> >> >>This is also a majo problem. >> >> >> >>>Therefore I have defined >>>what each is in SCVP 14 and aligned the structures accordingly. >>>Section 1.3 has the following. >>>"In SCVP, the validation policy comprises of an algorithm for >>>certificate path processing and the input parameters to the >> >>algorithm." >> >>This does not comply with RFC 3379. >> >> >> >>>Therefore trust anchors are an input into the algorithm and therefore >>>separate. >> >> >>Therefore I do not follow you. >> >> From an interface point of view the simplest validation policy is >>defined >>by an OID where all the parameters necessary to perform the validation >>check >>are "behind" the OID. There is no need to provide any parameter > > through > >>the >>interface. >> >>When there are some parameters, then they are specific to the > > validation > >>policy. I have no problem to provide specific parameters for what is >>called >>the "default validation policy" which is only a particular validation >>policy >>among many others. >> >>Simple clients will be unable to pass any parameter but will know > > which > >>OIDs >>(for the validation policy) are appropriate for their applications. >> >>Denis >> >> >> >>>This is in alignment with 3379s definition of validation policy. >>> >>>Trevor >>> >>>-----Original Message----- >>>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] >>>Sent: Monday, April 26, 2004 1:09 AM >>>To: Peter Sylvester >>>Cc: ietf-pkix@imc.org; Trevor Freeman >>>Subject: Re: FW: scvp >>> >>>Peter and Trevor, >>> >>>I have major problems too. >>> >>> >>> >>> >>>>in version 13 the syntax for a Query has been changed from >>>> >>>> Query ::= SEQUENCE { >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>>> checks CertChecks, >>>> wantBack WantBack, >>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>> valPolicy [1] ValidationPolicy OPTIONAL, >>>> validityTime [2] GeneralizedTime OPTIONAL, >>>> trustAnchors [3] TrustAnchors OPTIONAL, >>>> intermediateCerts [4] CertBundle OPTIONAL, >>>> revInfos [5] RevocationInfos OPTIONAL, >>>> queryExtensions [6] Extensions OPTIONAL } >>> >>> >>>In this structure trustAnchors was more or less an alternative to >>>valPolicy. >>> >>>In fact, IF the valPolicy is the default policy, THEN additional >>>parameters >>>MUST be provided. So the structure below does not fit as well. >>> >>>This leads to have the two following elements: >>> >>> valPolicy ValidationPolicy, >>> paramsDefValPolicy ParamsDefValPolicy OPTIONAL, >>> >>>where the first one is mandatory and the second one optional. >>> >>> >>> >>> >>>>to >>>> >>>> Query ::= SEQUENCE { >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>> >>> >>>> checks CertChecks, >>>> wantBack WantBack, >>>> requestRefHash BOOLEAN DEFAULT TRUE, >>>> includePolResponce BOOLEAN DEFAULT FALSE, >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>>> requireExplicitPol BOOLEAN DEFAULT FALSE, >>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, >>>> valAlgorithm ValidationAlgorithm, >>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>> validityTime [1] GeneralizedTime OPTIONAL, >>>> trustAnchors [2] TrustAnchors OPTIONAL, >>>> intermediateCerts [3] CertBundle OPTIONAL, >>>> revInfos [4] RevocationInfos OPTIONAL, >>>> queryExtensions [5] Extensions OPTIONAL } >>> >>> >>>I would thus propose to have something like: >>> >>>Query ::= SEQUENCE { >>> queriedCerts SEQUENCE SIZE (1..MAX) OF >>>CertReference, >>> checks CertChecks, >>> wantBack WantBack, >>> requestRefHash BOOLEAN DEFAULT TRUE, >>> serverContextInfo OCTET STRING OPTIONAL, >>> validityTime GeneralizedTime OPTIONAL, >>> includePolResponse BOOLEAN DEFAULT FALSE, >>> valPolicy ValidationPolicy, >>> paramsDefValPolicy [0] ParamsDefValPolicy OPTIONAL, >>> intermediateCerts [1] CertBundle OPTIONAL, >>> revInfos [2] RevocationInfos OPTIONAL, >>> queryExtensions [3] Extensions OPTIONAL } >>> >>>and then something like: >>> >>> ParamsDefValPolicy :: = SEQUENCE { >>> trustAnchors TrustAnchors, >>> endEntityCertificationPolicy SEQUENCE OF OBJECT IDENTIFIER >>>OPTIONAL, >>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>> caCertificationPolicies SEQUENCE OF OBJECT IDENTIFIER >>>OPTIONAL } >>> >>>Denis >>> >>> >>> >>> >>>>I am not sure whether I am the only person unable to construct a >>> >>>parser. >>> >>> >>> >>>>in version 14 an aditional flag has been added which is not available >>> >>>in the >>> >>> >>> >>>>Chapter 9. Is the isCA flag an orthogonal attribut to other >>> > validation > >>>>algorithms, or just to some of them, e.g,. the default name matching >>> >>>one? >>> >>> >>> >>>> Query ::= SEQUENCE { >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>> >>> >>>> checks CertChecks, >>>> wantBack WantBack, >>>> requestRefHash BOOLEAN DEFAULT TRUE, >>>> includePolResponce BOOLEAN DEFAULT FALSE, >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>>> requireExplicitPol BOOLEAN DEFAULT FALSE, >>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, >>>> isCA BOOLEAN DEFAULT FALSE, >>>> valAlgorithm ValidationAlgorithm, >>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>> validityTime [1] GeneralizedTime OPTIONAL, >>>> trustAnchors [2] TrustAnchors OPTIONAL, >>>> intermediateCerts [3] CertBundle OPTIONAL, >>>> revInfos [4] RevocationInfos OPTIONAL, >>>> keyUsage [5] KeyUsage OPTIONAL, >>>> extendedKeyUsage [6] ExtKeyUsageSyntax OPTIONAL, >>>> queryExtensions [7] Extensions OPTIONAL >>>> producedAt [8] GeneralizedTime OPTIONAL} >>>> >>>> >>> >>> >>> >>> >> >> > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i44AFRsF026411; Tue, 4 May 2004 03:15:27 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i44AFRpM026410; Tue, 4 May 2004 03:15:27 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from odin2.bull.net (odin2.bull.net [192.90.70.84]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i44AFPTI026360 for <ietf-pkix@imc.org>; Tue, 4 May 2004 03:15:26 -0700 (PDT) (envelope-from Denis.Pinkas@bull.net) Received: from clbull.frcl.bull.fr (IDENT:root@clbull.frcl.bull.fr [129.182.8.31]) by odin2.bull.net (8.9.3/8.9.3) with ESMTP id MAA09266; Tue, 4 May 2004 12:24:33 +0200 Received: from bull.net (frcls4013.frcl.bull.fr [129.182.108.120]) by clbull.frcl.bull.fr (8.9.3/8.9.3) with ESMTP id MAA19222; Tue, 4 May 2004 12:11:20 +0200 Message-ID: <40976D40.1030809@bull.net> Date: Tue, 04 May 2004 12:15:28 +0200 From: Denis Pinkas <Denis.Pinkas@bull.net> Organization: Bull SA. User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 X-Accept-Language: en-us, en, fr MIME-Version: 1.0 To: Trevor Freeman <trevorf@exchange.microsoft.com> CC: ietf-pkix@imc.org Subject: Re: FW: scvp References: <33E7A1613A3480448996047B69308A2F02666501@df-grommit-01.dogfood> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Trevor, > Hi Denis, > > 3379 does not require that the validation policy be specified in a > single value. You are playing with words. The text says: "most clients will simply reference a validation policy" This means that validation policies can be referenced and thus parameters do not need to be defined through this interface. A simple and good reference is an OID. > With SCVP14 a client can accept the default of the SCVP > server or specify a set of parameters which constitutes its validation > policy. That is consistent with the requirements of 3379. The specification of the set of parameters should be done through a different interface. This different interface can then return an OID Whatever this additional interface can be, it will never be rich enough to pass all the details of the validation policy. Thus OIDs can also be defined not using this additional interface. Denis > Trevor > > -----Original Message----- > From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] > Sent: Thursday, April 29, 2004 3:54 AM > To: Trevor Freeman > Cc: ietf-pkix@imc.org > Subject: Re: FW: scvp > > Trevor, > > >>Hi Denis, > > >>Can you please be more specific how you think SCVP 14 does not comply >>with 3379. >> >>I can find no section in 3379 where is there a requirement that a >>validation policy MUST be represented as an OID. > > > There cannot be a requirement with such a level of detail, but the text > states: > > The protocol MUST allow the client to include > these policy dependant parameters in the DPV request; however, it is > expected that most clients will simply reference a validation policy > for a given application or accept the DPV server's default > validation > policy. > > A simple reference is an OID. > > FYI, I do not expect to use the "default validation policy". > > Denis > > > >>Given hiding the detail of what a policy is within an OID simply opens >>the rat hole of what change to the policy constitutes a change to the >>OID, it is less ambiguous to refer to the primary data. Once you are > > in > >>the business of managing state on a client, then there is negligible >>cost increase incurred in managing several data points vs. a singe > > data > >>point. >> >>Trevor >> >>-----Original Message----- >>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] >>Sent: Tuesday, April 27, 2004 11:01 AM >>To: Trevor Freeman >>Cc: ietf-pkix@imc.org >>Subject: Re: FW: scvp >> >>Trevor, >> >> >> >>>HI Denis, >>>In SCVP13, the concept of validation policy was not completely in >>>alignment with 3379 definition. >> >> >>Well, it is different and this is a major problem. >> >> >> >>>It also blurred the distinction between >>>validation policy and validation algorithm. >> >> >>This is also a majo problem. >> >> >> >>>Therefore I have defined >>>what each is in SCVP 14 and aligned the structures accordingly. >>>Section 1.3 has the following. >>>"In SCVP, the validation policy comprises of an algorithm for >>>certificate path processing and the input parameters to the >> >>algorithm." >> >>This does not comply with RFC 3379. >> >> >> >>>Therefore trust anchors are an input into the algorithm and therefore >>>separate. >> >> >>Therefore I do not follow you. >> >> From an interface point of view the simplest validation policy is >>defined >>by an OID where all the parameters necessary to perform the validation >>check >>are "behind" the OID. There is no need to provide any parameter > > through > >>the >>interface. >> >>When there are some parameters, then they are specific to the > > validation > >>policy. I have no problem to provide specific parameters for what is >>called >>the "default validation policy" which is only a particular validation >>policy >>among many others. >> >>Simple clients will be unable to pass any parameter but will know > > which > >>OIDs >>(for the validation policy) are appropriate for their applications. >> >>Denis >> >> >> >>>This is in alignment with 3379s definition of validation policy. >>> >>>Trevor >>> >>>-----Original Message----- >>>From: Denis Pinkas [mailto:Denis.Pinkas@bull.net] >>>Sent: Monday, April 26, 2004 1:09 AM >>>To: Peter Sylvester >>>Cc: ietf-pkix@imc.org; Trevor Freeman >>>Subject: Re: FW: scvp >>> >>>Peter and Trevor, >>> >>>I have major problems too. >>> >>> >>> >>> >>>>in version 13 the syntax for a Query has been changed from >>>> >>>> Query ::= SEQUENCE { >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>>> checks CertChecks, >>>> wantBack WantBack, >>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>> valPolicy [1] ValidationPolicy OPTIONAL, >>>> validityTime [2] GeneralizedTime OPTIONAL, >>>> trustAnchors [3] TrustAnchors OPTIONAL, >>>> intermediateCerts [4] CertBundle OPTIONAL, >>>> revInfos [5] RevocationInfos OPTIONAL, >>>> queryExtensions [6] Extensions OPTIONAL } >>> >>> >>>In this structure trustAnchors was more or less an alternative to >>>valPolicy. >>> >>>In fact, IF the valPolicy is the default policy, THEN additional >>>parameters >>>MUST be provided. So the structure below does not fit as well. >>> >>>This leads to have the two following elements: >>> >>> valPolicy ValidationPolicy, >>> paramsDefValPolicy ParamsDefValPolicy OPTIONAL, >>> >>>where the first one is mandatory and the second one optional. >>> >>> >>> >>> >>>>to >>>> >>>> Query ::= SEQUENCE { >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>> >>> >>>> checks CertChecks, >>>> wantBack WantBack, >>>> requestRefHash BOOLEAN DEFAULT TRUE, >>>> includePolResponce BOOLEAN DEFAULT FALSE, >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>>> requireExplicitPol BOOLEAN DEFAULT FALSE, >>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, >>>> valAlgorithm ValidationAlgorithm, >>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>> validityTime [1] GeneralizedTime OPTIONAL, >>>> trustAnchors [2] TrustAnchors OPTIONAL, >>>> intermediateCerts [3] CertBundle OPTIONAL, >>>> revInfos [4] RevocationInfos OPTIONAL, >>>> queryExtensions [5] Extensions OPTIONAL } >>> >>> >>>I would thus propose to have something like: >>> >>>Query ::= SEQUENCE { >>> queriedCerts SEQUENCE SIZE (1..MAX) OF >>>CertReference, >>> checks CertChecks, >>> wantBack WantBack, >>> requestRefHash BOOLEAN DEFAULT TRUE, >>> serverContextInfo OCTET STRING OPTIONAL, >>> validityTime GeneralizedTime OPTIONAL, >>> includePolResponse BOOLEAN DEFAULT FALSE, >>> valPolicy ValidationPolicy, >>> paramsDefValPolicy [0] ParamsDefValPolicy OPTIONAL, >>> intermediateCerts [1] CertBundle OPTIONAL, >>> revInfos [2] RevocationInfos OPTIONAL, >>> queryExtensions [3] Extensions OPTIONAL } >>> >>>and then something like: >>> >>> ParamsDefValPolicy :: = SEQUENCE { >>> trustAnchors TrustAnchors, >>> endEntityCertificationPolicy SEQUENCE OF OBJECT IDENTIFIER >>>OPTIONAL, >>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>> caCertificationPolicies SEQUENCE OF OBJECT IDENTIFIER >>>OPTIONAL } >>> >>>Denis >>> >>> >>> >>> >>>>I am not sure whether I am the only person unable to construct a >>> >>>parser. >>> >>> >>> >>>>in version 14 an aditional flag has been added which is not available >>> >>>in the >>> >>> >>> >>>>Chapter 9. Is the isCA flag an orthogonal attribut to other >>> > validation > >>>>algorithms, or just to some of them, e.g,. the default name matching >>> >>>one? >>> >>> >>> >>>> Query ::= SEQUENCE { >>>> queriedCerts SEQUENCE SIZE (1..MAX) OF CertReference, >>> >>> >>>> checks CertChecks, >>>> wantBack WantBack, >>>> requestRefHash BOOLEAN DEFAULT TRUE, >>>> includePolResponce BOOLEAN DEFAULT FALSE, >>>> inhibitPolMap BOOLEAN DEFAULT FALSE, >>>> requireExplicitPol BOOLEAN DEFAULT FALSE, >>>> ignoreAnyPol BOOLEAN DEFAULT FALSE, >>>> isCA BOOLEAN DEFAULT FALSE, >>>> valAlgorithm ValidationAlgorithm, >>>> serverContextInfo [0] OCTET STRING OPTIONAL, >>>> validityTime [1] GeneralizedTime OPTIONAL, >>>> trustAnchors [2] TrustAnchors OPTIONAL, >>>> intermediateCerts [3] CertBundle OPTIONAL, >>>> revInfos [4] RevocationInfos OPTIONAL, >>>> keyUsage [5] KeyUsage OPTIONAL, >>>> extendedKeyUsage [6] ExtKeyUsageSyntax OPTIONAL, >>>> queryExtensions [7] Extensions OPTIONAL >>>> producedAt [8] GeneralizedTime OPTIONAL} >>>> >>>> >>> >>> >>> >>> >> >> > > > Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i41MkTFj024612; Sat, 1 May 2004 15:46:29 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i41MkTVp024611; Sat, 1 May 2004 15:46:29 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from host13.websitesource.com (host13.websitesource.com [209.239.35.152]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i41MkSrq024604 for <ietf-pkix@imc.org>; Sat, 1 May 2004 15:46:28 -0700 (PDT) (envelope-from chokhani@orionsec.com) Received: from wchokhani3 (pcp09271907pcs.arlngt01.va.comcast.net [69.143.130.247]) by host13.websitesource.com (8.12.10/8.12.10) with ESMTP id i41MkUJA000603 for <ietf-pkix@imc.org>; Sat, 1 May 2004 18:46:30 -0400 From: "Santosh Chokhani" <chokhani@orionsec.com> To: <ietf-pkix@imc.org> Subject: RE: Cross certificate format Date: Sat, 1 May 2004 18:46:26 -0400 Message-ID: <004101c42fce$25dd3900$aa02a8c0@hq.orionsec.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.6626 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 In-Reply-To: <OF3300ED57.DAF37743-ON85256E86.0052AF27-85256E87.007BAA8D@us.ibm.com> Importance: Normal Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by above.proper.com id i41MkSrq024606 Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Tom: The standard does not require you to check how you initialize the variables. You can get these from the cross certificate or from some configuration. The basic point is that you trust the public key. How you constrain it is not address by the standard, where as if the same certificate appeared in a path, you have to apply the constraints or be not compliant with the standard. You obviously do not believe in re-incarnation. Just kidding...... -----Original Message----- From: Tom Gindin [mailto:tgindin@us.ibm.com] Sent: Saturday, May 01, 2004 6:31 PM To: Santosh Chokhani Cc: ietf-pkix@imc.org Subject: RE: Cross certificate format Santosh: It is well known that trust in a trust anchor can be defined in a variety of ways. However, there is every reason to think that an RP may constrain that trust and that the constraints he may apply include those which an issuer CA may apply to a subordinate one. The wording of RFC 3280 section 6.2's first paragraph is an example of this. My comments about the meaning of a trust anchor are not purely theoretical. If a cross certificate is used to represent a trust anchor, the variables permitted_subtrees (RFC 3280 6.1.2 b) and excluded_subtrees (RFC 3280 6.1.2 c) could easily be set from the contents of a NameConstraints extension, and the variable max_path_length (RFC 3280 6.1.2 k) from the contents of a BasicConstraints extension. In RFC 3280 today, these variables are set to constants (although that interpretation is not absolutely clear for permitted_subtrees). In fact, they are three of the four variables which are set to constants during initialization. If permitted subtrees were set to a list of attributes, the validation process would fail if the anchor CA or a CA it certified went beyond the restrictions of those attributes. Likewise, if the excluded subtrees were set to a list of attributes, the validation process would fail if the anchor CA or a CA it certified went used values matching any of those attributes. I do not believe that I am arguing against Galileo. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/29/2004 11:12 PM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: The basic point (at the risk of repeating myself) is that a CA certificate or a self-signed certificate can be used as a means to provide trust anchor information. In that case, there is no need to check signature or anything. But, when the CA certificate is a certificate in the certificate path, X.509 logic kicks in. This is consistent with X.509, 3280 and general security principles since the trust in a trust anchor can be derived through any number of means. If you do not agree, feel free to argue against the definitions and Galileo -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Thursday, April 29, 2004 8:43 PM To: Santosh Chokhani Cc: ietf-pkix@imc.org Subject: RE: Cross certificate format Santosh: I don't think that a CA certificate and a trust anchor are as much different things as different uses for the same thing. They are both identifiers of issuers which bind a public key to an issuer name, and it is perfectly appropriate for any CA certificate to represent a trust anchor. I also do not see why such things as name constraints on a trust anchor are inappropriate. It is perfectly true that verifying a trust anchor's certificate, when issued by another CA (who may not be trusted) is a pointless operation. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/27/2004 09:14 AM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: My point is that even if the trust anchor is a self-signed certificate, all you need to do is extract the required information. There is no need to examine anything. Cross certificate and trust anchor are very different things. An element of the cross certificate pair is nothing but an intermediate CA certificate in a certification path. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Tuesday, April 27, 2004 7:33 AM To: Santosh Chokhani Cc: ietf-pkix@imc.org; owner-ietf-pkix@mail.imc.org Subject: RE: Cross certificate format Santosh: Of course, you are right that RFC 3280 6.1.1 d permits trust anchor information to be provided outside a certificate. If it is so, no checks of the sort I indicated can be performed. It also says that it can be provided as "a self-signed certificate", with no further qualification. I do think it is somewhat odd that a self-signed certificate with KeyUsage set to typical EE values would be considered a valid issuer as a trust anchor while a cross certificate would not. You can always extract the needed anchor information from a cross-certificate, of course. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/26/2004 08:51 AM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: My reading of 3280 and X.509 is that a trust anchor need not even be a certificate. Thus, no checks need to be performed on a trust anchor at all. You get the DN, public key, algorithm OID, and parameters (if applicable) from a trust anchor. Any checks are gratuitous and not required by either standard. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Sunday, April 25, 2004 7:06 PM To: Jean-Marc Desperrier Cc: ietf-pkix@imc.org Subject: Re: Cross certificate format Jean-Marc: RFC 3280 doesn't say that anywhere. It does say that you don't need to validate the trust anchor as part of the path, but it isn't clear from the text that a v1 certificate is acceptable as a trust anchor - and it certainly isn't clear that a v1 certificate is acceptable as an issuer if it is a trust anchor while a v3 certificate with KeyUsage= { DigitalSignature, keyEncipherment } is not acceptable as an issuer even if it is a trust anchor. I believe that the correct rules for versions are something like the following: a certificate issued by a trust anchor which is represented by a certificate is verifiable if the anchor certificate is either a v1 certificate or a v3 certificate with the CA flag present in the basicConstraints extension. If the anchor certificate is a v3 certificate with the KeyUsage extension present, it is only a valid issuer certificate if keyCertSign is set. Tom Gindin Jean-Marc Desperrier <jmdesp@free.fr> Sent by: owner-ietf-pkix@mail.imc.org 04/21/2004 03:42 PM To: ietf-pkix@imc.org cc: Subject: Re: Cross certificate format Tom Gindin wrote: > RFC 3280 does not support the use of v1 self-signed > certificates, >which contain no extensions at all (see the text of RFC 3280's >basicConstraints section). However, a number of public CA's still have >them. > > Applications implementing the path validation algorithm of RFC3280 MUST accept them when they are used as trust anchors. Received: from above.proper.com (localhost.vpnc.org [127.0.0.1]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i41MV3PN020769; Sat, 1 May 2004 15:31:03 -0700 (PDT) (envelope-from owner-ietf-pkix@mail.imc.org) Received: (from majordom@localhost) by above.proper.com (8.12.11/8.12.9/Submit) id i41MV3mh020768; Sat, 1 May 2004 15:31:03 -0700 (PDT) X-Authentication-Warning: above.proper.com: majordom set sender to owner-ietf-pkix@mail.imc.org using -f Received: from e2.ny.us.ibm.com (e2.ny.us.ibm.com [32.97.182.102]) by above.proper.com (8.12.11/8.12.9) with ESMTP id i41MV2L1020759 for <ietf-pkix@imc.org>; Sat, 1 May 2004 15:31:02 -0700 (PDT) (envelope-from tgindin@us.ibm.com) Received: from northrelay04.pok.ibm.com (northrelay04.pok.ibm.com [9.56.224.206]) by e2.ny.us.ibm.com (8.12.10/8.12.2) with ESMTP id i41MV1iJ390602; Sat, 1 May 2004 18:31:01 -0400 Received: from d01ml062.pok.ibm.com (d01av04.pok.ibm.com [9.56.224.64]) by northrelay04.pok.ibm.com (8.12.10/NCO/VER6.6) with ESMTP id i41MVV2M102712; Sat, 1 May 2004 18:31:31 -0400 To: "Santosh Chokhani" <chokhani@orionsec.com> Cc: ietf-pkix@imc.org MIME-Version: 1.0 Subject: RE: Cross certificate format X-Mailer: Lotus Notes Release 5.0.11 July 24, 2002 From: Tom Gindin <tgindin@us.ibm.com> Message-ID: <OF3300ED57.DAF37743-ON85256E86.0052AF27-85256E87.007BAA8D@us.ibm.com> Date: Sat, 1 May 2004 18:30:55 -0400 X-MIMETrack: Serialize by Router on D01ML062/01/M/IBM(Release 6.0.2CF2 HFB2|March 16, 2004) at 05/01/2004 18:31:00, Serialize complete at 05/01/2004 18:31:00 Content-Type: text/plain; charset="us-ascii" Sender: owner-ietf-pkix@mail.imc.org Precedence: bulk List-Archive: <http://www.imc.org/ietf-pkix/mail-archive/> List-ID: <ietf-pkix.imc.org> List-Unsubscribe: <mailto:ietf-pkix-request@imc.org?body=unsubscribe> Santosh: It is well known that trust in a trust anchor can be defined in a variety of ways. However, there is every reason to think that an RP may constrain that trust and that the constraints he may apply include those which an issuer CA may apply to a subordinate one. The wording of RFC 3280 section 6.2's first paragraph is an example of this. My comments about the meaning of a trust anchor are not purely theoretical. If a cross certificate is used to represent a trust anchor, the variables permitted_subtrees (RFC 3280 6.1.2 b) and excluded_subtrees (RFC 3280 6.1.2 c) could easily be set from the contents of a NameConstraints extension, and the variable max_path_length (RFC 3280 6.1.2 k) from the contents of a BasicConstraints extension. In RFC 3280 today, these variables are set to constants (although that interpretation is not absolutely clear for permitted_subtrees). In fact, they are three of the four variables which are set to constants during initialization. If permitted subtrees were set to a list of attributes, the validation process would fail if the anchor CA or a CA it certified went beyond the restrictions of those attributes. Likewise, if the excluded subtrees were set to a list of attributes, the validation process would fail if the anchor CA or a CA it certified went used values matching any of those attributes. I do not believe that I am arguing against Galileo. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/29/2004 11:12 PM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: The basic point (at the risk of repeating myself) is that a CA certificate or a self-signed certificate can be used as a means to provide trust anchor information. In that case, there is no need to check signature or anything. But, when the CA certificate is a certificate in the certificate path, X.509 logic kicks in. This is consistent with X.509, 3280 and general security principles since the trust in a trust anchor can be derived through any number of means. If you do not agree, feel free to argue against the definitions and Galileo -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Thursday, April 29, 2004 8:43 PM To: Santosh Chokhani Cc: ietf-pkix@imc.org Subject: RE: Cross certificate format Santosh: I don't think that a CA certificate and a trust anchor are as much different things as different uses for the same thing. They are both identifiers of issuers which bind a public key to an issuer name, and it is perfectly appropriate for any CA certificate to represent a trust anchor. I also do not see why such things as name constraints on a trust anchor are inappropriate. It is perfectly true that verifying a trust anchor's certificate, when issued by another CA (who may not be trusted) is a pointless operation. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/27/2004 09:14 AM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: My point is that even if the trust anchor is a self-signed certificate, all you need to do is extract the required information. There is no need to examine anything. Cross certificate and trust anchor are very different things. An element of the cross certificate pair is nothing but an intermediate CA certificate in a certification path. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Tuesday, April 27, 2004 7:33 AM To: Santosh Chokhani Cc: ietf-pkix@imc.org; owner-ietf-pkix@mail.imc.org Subject: RE: Cross certificate format Santosh: Of course, you are right that RFC 3280 6.1.1 d permits trust anchor information to be provided outside a certificate. If it is so, no checks of the sort I indicated can be performed. It also says that it can be provided as "a self-signed certificate", with no further qualification. I do think it is somewhat odd that a self-signed certificate with KeyUsage set to typical EE values would be considered a valid issuer as a trust anchor while a cross certificate would not. You can always extract the needed anchor information from a cross-certificate, of course. Tom Gindin "Santosh Chokhani" <chokhani@orionsec.com> Sent by: owner-ietf-pkix@mail.imc.org 04/26/2004 08:51 AM To: <ietf-pkix@imc.org> cc: Subject: RE: Cross certificate format Tom: My reading of 3280 and X.509 is that a trust anchor need not even be a certificate. Thus, no checks need to be performed on a trust anchor at all. You get the DN, public key, algorithm OID, and parameters (if applicable) from a trust anchor. Any checks are gratuitous and not required by either standard. -----Original Message----- From: owner-ietf-pkix@mail.imc.org [mailto:owner-ietf-pkix@mail.imc.org] On Behalf Of Tom Gindin Sent: Sunday, April 25, 2004 7:06 PM To: Jean-Marc Desperrier Cc: ietf-pkix@imc.org Subject: Re: Cross certificate format Jean-Marc: RFC 3280 doesn't say that anywhere. It does say that you don't need to validate the trust anchor as part of the path, but it isn't clear from the text that a v1 certificate is acceptable as a trust anchor - and it certainly isn't clear that a v1 certificate is acceptable as an issuer if it is a trust anchor while a v3 certificate with KeyUsage= { DigitalSignature, keyEncipherment } is not acceptable as an issuer even if it is a trust anchor. I believe that the correct rules for versions are something like the following: a certificate issued by a trust anchor which is represented by a certificate is verifiable if the anchor certificate is either a v1 certificate or a v3 certificate with the CA flag present in the basicConstraints extension. If the anchor certificate is a v3 certificate with the KeyUsage extension present, it is only a valid issuer certificate if keyCertSign is set. Tom Gindin Jean-Marc Desperrier <jmdesp@free.fr> Sent by: owner-ietf-pkix@mail.imc.org 04/21/2004 03:42 PM To: ietf-pkix@imc.org cc: Subject: Re: Cross certificate format Tom Gindin wrote: > RFC 3280 does not support the use of v1 self-signed > certificates, >which contain no extensions at all (see the text of RFC 3280's >basicConstraints section). However, a number of public CA's still have >them. > > Applications implementing the path validation algorithm of RFC3280 MUST accept them when they are used as trust anchors.
- What does the 'S' in SCVP stand for? James Jing