The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2003-Mar> msg00325



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

[PWE3] MPLS PID

  • From: "David Allan" <dallan@nortelnetworks.com>
  • Date: Wed, 26 Mar 2003 14:47:54 -0500

Title: RE: [PWE3] MPLS PID

Hi Scott:

I'm sure you can guarantee coordination. Not sure as to how folks feel about a 4 bit version number being subsumed as a protocol id as it is a rather scarce resource, therefore the "few we need" actually consititues potentially a significant amount of that resource (and one will note that there is already pressure on it just in this discussion). Yes, sure we can chain bits (if MSbit set then it's really an 8 bit value) bit that's not pretty if we're planning on it up front, its simply how you get around a long standing situation.

If we go on the assertion that MPLS should have had a PID (as opposed to current practice of binding protocols to labels) then we need to scope what we are really trying to achieve....

1) Make the payload self describing to end points.
2) Multiplex payloads over a single LSP.
3) Make the payload self-describing to intermediate systems.

At the moment, 3 appears to be the exclusive motivation for this but has implications in 1 & 2 as we already have mechanisms for achieving those goals today (indirectly via signalling (1) & label stacking (2)).

IMHO if we were to add a payload identification to MPLS in particular for intermediate systems to inspect (bottom label is now really 36 or more bits that happen to overlap with payload), I would want a lot more from it. For example, it was a problem folks wrestled with when trying to figure out how to identify OAM flows (where we went with reserved label as a PID and therefore only had e2e significance), and I'm sure other applications will come up.

cheers
Dave

> -----Original Message-----
> From: Scott W Brim [mailto:sbrim@cisco.com]
> Sent: Wednesday, March 26, 2003 1:56 PM
> To: pwe3@ietf.org; mpls@uu.net
> Subject: Re: [PWE3] MPLS PID
>
>
> On Wed, Mar 26, 2003 01:38:20PM -0500, David Allan allegedly wrote:
> > We don't have 14 values, we are talking inferring payload
> on the basis
> > of defacto IETF protocol payloads (IPv4 and IPv6). PWE3 and MPLS do
> > not own IPv4/v6 or any subsequent network layer protocols
> the IETF may
> > choose to develop.
>
> Are you trying to say that we can't guarantee coordination
> between a group developing new versions of IP and the MPLS
> community?  This is a small logistical problem.  In the
> absolute worst case we get assignments from IANA to cover the
> few we need.
>
> (See http://www.iana.org/assignments/version-numbers).
>
> _______________________________________________
> pwe3 mailing list
> pwe3@ietf.org
> https://www1.ietf.org/mailman/listinfo/pwe3
>