Re: Comments on draft-ietf-kitten-krb5-gssapi-prf-03.txt

On Tue, May 24, 2005 at 12:43:37AM +0200, Martin Rex wrote:
> As with a similar discussion on krb-ietf list how a KDC cert should
> be verified in PKINIT, we should not ignore the fact that the affected
> spec (here) is going to be normative for both, the gssapi mechanism
> implementor and the application gssapi caller, and those two
> will have an entirely different background and needs.
> 
> So I would suggest to actually quantify
>  (a) what the mechanism implementor must provide as a minimum
>      to be at all useful

That's already there, in section 2 of
draft-ietf-kitten-gssapi-prf-03.txt:

   Mechanisms MUST be able to output at least up to 2^14 octets.

   If the implementation cannot produce the desired output due to lack
   of resources then it MUST output what it can and still return
   GSS_S_COMPLETE.

Perhaps we should sprinkle "cryptographically safe" there.

>  (b) what the application caller can rely on to be safely available.

I think the application should be prepared to fail if there's not enough
PRF output, or if the PRF is not available.

> We should add a reference to the document draft-eastlake-randomness2-10.txt
> that is sitting on the Editor's queue:

With respect to what?  The input to the PRF?

Nico
-- 

_______________________________________________
Kitten mailing list
Kitten@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/kitten