Re: Concerns about individual submissions process (c.f. RFC5647 AES-GCM for Secure Shell)

[Nicolas Williams <Nicolas.Williams%Sun.COM@localhost>]

On Wed, Sep 02, 2009 at 02:54:33PM -0500, Nicolas Williams wrote:
> All that said, I'm reasonably happy with RFC5647, but there are several
> issues that I think should have been addressed prior to publication:
> 

>     - 'fixed' appears to be fixed per-_key exchange_, not for the life
>       of the connection.  This one, in particular, is a complete and
>       utter guess on my part.

I should point out that it's obvious enough that 'fixed' is fixed
per-_key exchange_ simply because otherwise security would be
compromised after more than 2^64 invocations.

Other comments:

 - The algorithm names registered by this RFC with the IANA do NOT match
   the style of naming of previously existing algorithm names.

   The previous alg names were all lower-case LDH (letters digits and
   hyphens).  The new ones are all upper-case letters, digits, and
   underscore.

   Earlier versions of the I-D had alg names with the "proper" style.
   What happened?

 - The Security Considerations section is silent on re-keying
   requirements (i.e., when and how often to re-key).  It should not
   have been.

Nico
--