Re: [MOBILE-IP] New GRE Draft Extensions

[Thomas Narten <narten@raleigh.ibm.com>]

Folks,

I'm redirecting followups to the gre list. To join:

    majordomo@ops.ietf.org
    subscribe gre

Note: there are folks interested in gre outside of mobile-ip. The UDLR
WG has also expressed an interest in this area.

> I do not know if the IESG figures into this since GRE is
> informational. However I would like to know the reason behind the
> updates because draft-meyer + draft-dommety looks pretty close to
> RFC 1701.  The changes appear to be the semantics of the key field,
> which was somewhat semantic free before since its use as an
> authenticator was always suspect, and the split sequence number
> field.  What's the use of deprecating some fields in one draft and
> extending it with the same deprecated fields in another draft?

RFC 1701 is informational. It cannot be referenced by standards track
documents in a normative way.

draft-meyer-gre-update-03.txt has been approved as a Proposed Standard
by the IESG an will be out as an RFC any day now. This document is
backwards compatable with 1701 in the sense that existing
implementations of 1701 are already compliant with the new
standard. However, draft-meyer-gre-update does not include some of the
more exotic extensions that 1701 has. Thus, the new GRE document isn't
for everyone, as it doesn't include the key and sequence
fields. meyer-gre-update is just the basics, no fluff.

The purpose of draft-dommety is to define some new options to GRE that
may in fact look suprisingly like some of the fields in 1701 that
don't appear in draft-meyer-gre-update. One problem with 1701's
definition of those fields is that they are underspecified (you can't
implement them from 1701) and its not even clear whether everyone
would agree on what semantics those fields should have. draft-dommety
is intended to find consensus on what those other fields should be (at
least from the perspectives of mobile IP).

Questions like should GRE extensions use a new version number, etc.,
need to hashed out on the gre list. 

Thomas