Re: [OPSEC] minutes part 2

On  17 Dec 2008, at 20:06, Glen Kent wrote:
>> I object to the assertions in those minutes that:
>>       - the SHA approaches are better/stronger than MD5 approaches.
>>         [We have agreed that the evidence does NOT support such a
>>         claim -- which claim appears to be in the meeting minutes.]
>>
>>       - the WG should recommend/encourage/promote SHA approaches
>>         over MD5 approaches. [which claim again appears to be
>>         in the meeting minutes]
>
> So you're asserting that Md5 is as strong as SHA. Can you please  
> confirm this?

No.  I'm asserting that:

   At present, there is no data indicating that MD5
   (as used for IGP authentication) is either stronger
   or weaker than SHA (as used for IGP authentication).

There was a time (now past) where there were published
concerns about the compression function of MD5
[Dobbertin1996] but there were not equivalent concerns
about SHA.  I cited specific published papers that now
indicate strong concerns with the compression function
of SHA.  I also cited a NIST formal statement indicating
that there were "serious attacks" on the compression
function of SHA.  (In both cases, I provided specific
URLs yesterday in email to the OPsec list.)

Yours,

Ran Atkinson
rja@extremenetworks.com

_______________________________________________
OPSEC mailing list
OPSEC@ietf.org
https://www.ietf.org/mailman/listinfo/opsec