Re: [http-auth] draft-ietf-httpauth-digest-update and SHA-3

Aargh. What I tell you three times is true?

Sorry about that.

Yoav

On Jul 7, 2013, at 6:56 PM, Yoav Nir <ynir@checkpoint.com> wrote:

> [with implementer hat on]
> 
> My implementation has it, although this makes little difference as it's not a public library.
> 
> OIDs are already defined for this algorithm (2.16.840.1.101.3.4.2.6 and 1.2.840.113549.1.1.16) but I'm not aware of anyone issuing certificates with it.
> 
> We could define a simple truncation of SHA2-512 ourselves, but I guess there was some method behind defining it a little differently. Anyone know?
> 
> Yoav
> 
> On Jul 7, 2013, at 5:51 PM, Yaron Sheffer <yaronf.ietf@gmail.com> wrote:
> 
>> Hi,
>> 
>> The draft mistakenly cites FIPS 186, rather than FIPS 180, for SHA2.
>> 
>> The correct notation is "SHA-512/256".
>> 
>> And most worryingly, reading Sec. 5.3.6 of the FIPS pub, I wonder if common implementations of SHA2 support this algorithm, which is *not* the naive truncation that I expected it to be.
>> 
>> Thanks,
>> 	Yaron
>> 
>> On 2013-07-07 17:44, Paul Hoffman wrote:
>>> On Jul 7, 2013, at 6:24 AM, Rifaat Shekh-Yusef <rifaat.ietf@gmail.com> wrote:
>>> 
>>>> I have just submitted a new version of the draft that removes SHA3 as a back algorithm and uses SHA2-512/256 instead. The draft also mentions that a future version might add SHA3 as a backup algorithm.
>>>> https://datatracker.ietf.org/doc/draft-ietf-httpauth-digest-update/
>>>> 
>>>> Please, review the new draft and let me know if you have any further comments.
>>> 
>>> This looks very good. We need a reference for SHA-512-256, but I think I remember that NIST already has one.
>>> 
>>> --Paul Hoffman
> 
> _______________________________________________
> http-auth mailing list
> http-auth@ietf.org
> https://www.ietf.org/mailman/listinfo/http-auth