Re: [Curdle] AD Review of draft-ietf-curdle-gss-keyex-sha2-05

[denis bider <[e-mail]>]

I think this misunderstands the scope of the GSS draft.

The GSS draft discusses GSS key exchange methods for SSH. These are not the
main key exchange methods for SSH, they are a *version* of the main key
exchange methods for use with GSS.

Because these are a *version* of the main key exchange methods, the main
value of this draft is that it updates the GSS key exchange methods to be
in line with how the main key exchange methods for SSH have evolved.

As Mark has pointed out, we have already added these key exchange methods
to SSH in RFC 8268:

diffie-hellman-group14-sha256
diffie-hellman-group15-sha512
diffie-hellman-group16-sha512
diffie-hellman-group17-sha512
diffie-hellman-group18-sha512

The whole argument about which groups and which hashes we should have has
already been staged, multiple times, and this is what we arrived at.

The GSS draft is merely mirroring what we have already done for use with
the GSS key exchange methods. It does not make sense for the GSS key
exchange methods to do something entirely different than what the main key
exchange methods do. It makes sense for the GSS key exchange methods to be
consistent with what we have already done for the main ones.


On Sat, Apr 7, 2018 at 6:38 PM, Eric Rescorla <[e-mail]> wrote:

>
>
> On Sat, Apr 7, 2018 at 7:23 AM, Salz, Rich <[e-mail]> wrote:
>
>> I see no point in adding another hash.  What’s the reason?
>>
>
> So that you can have an implementation that spans a variety of groups with
> just SHA-256.
>
>>
>>
>
>