Re: draft-rosen-l3vpn-mvpn-spmsi-joins-00.txt

[Eric Rosen <[e-mail]>]

> Note also that draft-ietf-2547bis-mcast and
> draft-ietf-l3vpn-mvpn-considerations together do provide a specification
> of a standard

Please note that "considerations" is not a standards track document, and
hence is not part of the standard.  Additionally, since it is not part of
the WG charter, it cannot be used to prevent the WG from considering
proposals on their merits.  

> with a control plane that can be used independently of the forwarding
> plane that is used.

I believe that RFC 4384 requires that any standard control plane be usable
with any standard forwarding plane technology, so I think it does require
that the UDP-based S-PMSI signaling be usable with MPLS forwarding.

I do agree that the BGP S-PMSI A-D route mechanism has advantages, and we
are likely to see the industry migrate in this direction.  But as a
practical matter, failure to officially publish the code points for the
UDP-based S-PMSI Join extensions is not going to speed up that migration.

On the IPv6 issue, if the WG decides it does not want to honor the promise
made to the IESG to do the IPv6 extensions, then the WG should really
request that draft-ietf-2547bis-mcast be removed from the RFC Editors' queue
and returned to IETF last call status.  Then we can add another year of
delay while we argue with the IESG about IPv6 support.  But I find it
difficult to see how such a step would help the industry.