Re: I-D ACTION:draft-ietf-smime-examples-04.txt

Dr S N Henson <[email protected]> Wed, 25 Oct 2000 01:53:53 +0100
Newsgroups gmane.ietf.smime-examples
Organization S N Henson
Message-ID <[email protected]>
Paul Hoffman / IMC wrote:
> 
> 
> Because I had also lamely filed the comments people had sent me a
> year ago about -03 into different folders, I am not sure what need to
> be fixed. I know that much does need to be fixed, but I am not sure
> what. Please send (or, probably, re-send) all comments to me and the
> [email protected] list. And I really, really promise to get
> -05 out before the cutoff for the San Diego IETF.
> 

OK let me start with a comment I made ages ago about the DH parameters
used in the certificates of the examples.

I can't reproduce the DH parameters: that is using the algorithm of
RFC2631 2.2.1 I cannot reproduce the parameters from the given seed and
counter. It may be a problem with my implementation but I don't think
so.

To test my code I have asked everywhere I can think of for some
independent test vectors that aren't equivalent to FIPS186 (DSS)
parameter generation so far without success. My implementation works
fine on FIPS186 compatible parameters.

Various people have also stated that the parameter generation algorithm
of RFC2631 is not exactly optimal and various other techniques are more
efficient.

I'm willing to do tests with anyone who has the relevant implementation
or indeed give a step by step explanation as to why I think the
parameters are incorrect.

I consider this important because if we are including parameters in
certificates with an invalid seed and counter then a compliant
implementation would be perfectly justified in rejecting the
certificates on those grounds.

However if almost no one is implementing the RFC2631 parameter
generation algorithm and confirmation cannot be obtained then IMHO it
would be best if they were simply omitted from the certificates.

A secondary issue is that if no one is implementing RFC2631 2.2.1 why
are we saying that agents SHOULD use it?

Steve.
-- 
Dr Stephen N. Henson.   http://www.drh-consultancy.demon.co.uk/
Personal Email: [email protected] 
Senior crypto engineer, Celo Communications: http://www.celocom.com/
Core developer of the   OpenSSL project: http://www.openssl.org/
Business Email: [email protected] PGP key: via homepage.