[dtn-security] Re: are offsets enough? --was: (dictionary or not) Re: [dtn-security] 00 version of the Bundle Security Protocol Spec.

Stephen Farrell <stephen.farrell@cs.tcd.ie> Thu, 26 May 2005 13:54 UTC

Received: from smtp3.tcd.ie (smtp3.tcd.ie []) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id j4QDseV06790 for <dtn-security@mailman.dtnrg.org>; Thu, 26 May 2005 06:54:40 -0700
Received: from Vams.smtp3 (smtp3.tcd.ie []) by smtp3.tcd.ie (Postfix) with SMTP id 69A9E14C069; Thu, 26 May 2005 14:54:34 +0100 (IST)
Received: from smtp3.tcd.ie ([]) by smtp3.tcd.ie ([]) with SMTP (gateway) id A0021D59B9D; Thu, 26 May 2005 14:54:34 +0100
Received: from [] (sfarrel2.dsg.cs.tcd.ie []) by smtp3.tcd.ie (Postfix) with ESMTP id 4141514C069; Thu, 26 May 2005 14:54:34 +0100 (IST)
Message-ID: <4295D611.40606@cs.tcd.ie>
Date: Thu, 26 May 2005 14:58:41 +0100
From: Stephen Farrell <stephen.farrell@cs.tcd.ie>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.8) Gecko/20050511
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: dtn-security@mailman.dtnrg.org
Cc: "'Michael Demmer'" <demmer@cs.berkeley.edu>
References: <200505251831.j4PIVRX27084@smtp-bedford.mitre.org>
In-Reply-To: <200505251831.j4PIVRX27084@smtp-bedford.mitre.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-AntiVirus-Status: Checked by TCD Vexira. (version=1.55.010 VDF=8.731)
X-AntiVirus-Status: NONE
X-AntiVirus-Status: Action Taken:
X-AntiVirus-Status: Host: smtp3.tcd.ie
X-AntiVirus-Status: MessageID = A1021D59B9D
Subject: [dtn-security] Re: are offsets enough? --was: (dictionary or not) Re: [dtn-security] 00 version of the Bundle Security Protocol Spec.
Sender: dtn-security-admin@mailman.dtnrg.org
Errors-To: dtn-security-admin@mailman.dtnrg.org
X-BeenThere: dtn-security@mailman.dtnrg.org
X-Mailman-Version: 2.0.13
Precedence: bulk
Reply-To: dtn-security@mailman.dtnrg.org
List-Unsubscribe: <http://mailman.dtnrg.org/mailman/listinfo/dtn-security>, <mailto:dtn-security-request@mailman.dtnrg.org?subject=unsubscribe>
List-Id: DTN Security Discussion <dtn-security.mailman.dtnrg.org>
List-Post: <mailto:dtn-security@mailman.dtnrg.org>
List-Help: <mailto:dtn-security-request@mailman.dtnrg.org?subject=help>
List-Subscribe: <http://mailman.dtnrg.org/mailman/listinfo/dtn-security>, <mailto:dtn-security-request@mailman.dtnrg.org?subject=subscribe>
List-Archive: <http://mailman.dtnrg.org/pipermail/dtn-security/>

Since we now have extensions and even, apparently, criticality,
some of those offsets might be in headers you don't know how
to parse so the dictionary encoding has to be self-contained,
other than whatever refers to the dictionary. Similarly, you
cannot move stuff about in the dictionary unless you understand
how to fix all of the headers in a bundle.


Susan F. Symington wrote:
>>Personally, I think I'd rather that dictionary entries contain
>>a length field (whatever encoding) since I've found lists of
>>null-terminated strings are a great source for implementation
> Mike: Yeah -- I just feel like it's not worth the bytes. My other thought is
> that we can infer the string lengths based on the various offsets
> scattered throughout the bundle, in other words, if all the offsets
> (i.e. src, dest, custodian, etc) were sorted, we could then infer the
> various lengths based on the differences. But this is probably harder
> to read (though more efficient) than just scanning for the null.
> Susan: It occurs to me that when a bundle is fragmented, we won't always be
> able to infer the various dictionary string lengths based on the various
> offsets scattered through the bundle, because not all headers (and therefore
> not all offsets) are guaranteed to be in every fragment.  The CH and PSH are
> two examples of headers that may contain an offset but that most likely will
> not be in every fragment. In the new FLAGS field we defined last week, one
> bit is for indicating whether the header must be duplicated on all fragments
> or just the first.  So, it seems like the PSH and CH may not be the only
> examples. 
> So, if terminating the strings with nulls is not sufficient, then it seems
> we need something like a length field to go along with the offset.
> -susan
> _______________________________________________
> dtn-security mailing list
> dtn-security@mailman.dtnrg.org
> http://mailman.dtnrg.org/mailman/listinfo/dtn-security