Re: [pkix] Amendment to CABF Baseline Requirements

[Jeremy Rowley <[e-mail]>]

Hi Russ - Considering the CA/Browser Forum rules only applies to browser user agents, are you aware of any actual interoperability issues?

 

From: pkix [mailto:[e-mail]] On Behalf Of Russ Housley
Sent: Saturday, April 8, 2017 11:47 AM
To: Erik Andersen <[e-mail]>
Cc: IETF PKIX <[e-mail]>
Subject: Re: [pkix] Amendment to CABF Baseline Requirements

 

Erik:

 

The CA/Browser Forum seems to be finding the upper bounds in the X.500 5th edition and RFC 5280.  I’m not sure how to relax them without introducing interoperability concerns.

 

Russ

 

 

On Apr 8, 2017, at 4:20 AM, Erik Andersen <[e-mail] <mailto:[e-mail]> > wrote:

 

The change was made in the sixth edition of X.520 (2008). It was primarily driven by Hoyt Kesterson and by some X.500/LDAP vendors. Hoyt was generally against such limitations, as the future is difficult to predict. X.500/LDAP vendors were concerned about interworking between X.500 and LDAP (that does not have upper bounds). You may have noticed that ASN.1 information objects now also contains LDAP specific fields allowing also new attribute types to be usable also in an LDAP environment without the need for additional specifications.

 

I insisted that the upper bounds should be kept in an informative annex to be used by those having a requirement to limit the sizes.

 

I could, if you feel it relevant, try to suggest a Defect Report to X.509 saying that it is strongly recommend to observe the upper bounds specified in X.520. I will also try to get such a statement into any smart grid security standard that references X.509.

 

Erik

Fra: pkix [ <mailto:[e-mail]> mailto:[e-mail]] På vegne af Russ Housley
Sendt: 07 April 2017 22:43
Til: IETF PKIX < <mailto:[e-mail]> [e-mail]>
Emne: Re: [pkix] Amendment to CABF Baseline Requirements

 

The ASN.1 modules that were posted by Erik confirm that the upper bounds have been removed for all of the naming attributes in the SelectedAttributeTypes module.

 

UnboundedDirectoryString ::= CHOICE {

  teletexString    TeletexString(SIZE (1..MAX)),

  printableString  PrintableString(SIZE (1..MAX)),

  bmpString        BMPString(SIZE (1..MAX)),

  universalString  UniversalString(SIZE (1..MAX)),

  uTF8String       UTF8String(SIZE (1..MAX)) }

 

Russ

 

 

On Apr 7, 2017, at 3:15 AM, Erwann Abalea < <mailto:[e-mail]> [e-mail]> wrote:

 

Bonjour,

 

2017-04-06 22:39 GMT+02:00 Peter Bowen < <mailto:[e-mail]> [e-mail]>:

 

On Thu, Apr 6, 2017 at 12:24 PM Russ Housley < <mailto:[e-mail]> [e-mail]> wrote:

The comment in the UpperBounds ASN.1 module (the 8th edition) says:

-- EXPORTS All
-- The types and values defined in this module are exported for use in the other ASN.1
-- modules contained within these Directory Specifications, and for the use of other
-- applications which will use them to access Directory services. Other applications
-- may use them for their own purposes, but this will not constrain extensions and
-- modifications needed to maintain or improve the Directory service.

X.509 is part of the Directory Specifications, so they are not advisory.

It looks like ITU-T increased the length of the organizational unit name in the most recent edition.

RFC 5280 says:

ub-organization-name-length INTEGER ::= 64
ub-organizational-unit-name-length INTEGER ::= 32

The UpperBounds ASN.1 module (the 8th edition) says:

ub-organization-name                       INTEGER ::= 64
ub-organizational-unit-name                INTEGER ::= 64

So, we may already be in a place where implementations conforming to X.509 will produce a certificate that cannot be decoded by an implementation that conforms to RFC 5280.

I wish we gad gotten a heads-up …

 

It is even worse. 7th and 8th (and maybe prior releases) removed the usage of ub- from the schema. The schema itself no longer bounds DirectoryStrings and X.509 explicitly says they are unbounded. 

 

X.520 2005 defines the attributes using the DirectoryString{} parameterized type, while the 2008 edition uses the UnboundedDirectoryString type.

 

RFC5912, which proposes ASN.1 modules for certificates and other related things, still uses the DirectoryString{} variant.

 

-- 

Erwann.

 

_______________________________________________
pkix mailing list
 <mailto:[e-mail]> [e-mail]
 <https://www.ietf.org/mailman/listinfo/pkix> https://www.ietf.org/mailman/listinfo/pkix