Re: GRE specification going to last call

[Tim Gleeson <tgleeson@cisco.com>]

At 15:20 00/1/6 -0500, David Oran wrote:
> Since UDLR uses GRE for its tunnels, it is important the some folks inthe
> UDLR group review this version of the GRE spec to make sure it meets the
> needs of UDLR.
> 
> Please respond to the IETF last call (to be started imminently) if there are
> issues and/or problems, and copy both me and Thomas Narten, as well as the
> WG, on any issues.
> 
> This will remove a barrier to the progression of UDLR, so expeditious review
> will be very helpful.
> 
> Thanks, Dave Oran (your friendly AD)

Dave,

I took a look at gre-update-02 and sent a few minor editorial comments 
direct to Dave Meyer.

There don't seem to be implications, as far as I can see, for the UDLR 
spec; apart from protocol types.

This was quite loosely defined in 1701:
  "In general... additional values may be defined in other documents."
but gre-update-02 is rather stricter:
  "These protocol types are defined in [RFC1700]... and in [ETYPES]."

This may limit some scope for UDLR implementations which want to use GRE, 
but use a MAC type which doesn't have a number, e.g. ETSI MPE, which I 
believe Patrick brought up as an issue some time ago. This doesn't really 
affect UDLR itself though.

Just as a note to those concerned, the new draft also says "IANA SHOULD NOT 
encourage the assignment of additional ETHER TYPES (GRE Protocol Types) for 
use with GRE."

Tim