Re: [mpls] Re: draft-ietf-mpls-rsvp-te-p2mp-06.txt [P2MP ID]

["Adrian Farrel" <adrian@olddog.co.uk>]

Just FWIW

  P2MP-ID (P2ID):
     A unique identifier of a P2MP TE LSP, which is constant for the
     whole LSP regardless of the number of branches and/or leaves.

Extracted from RFC4461.
We do not have to use that description in the protocol fields, but if we 
choose not to, I suggest that we change the name of the field to avoid 
confusion.

Historically, the reason for the text "The P2MP ID provides an identifier 
for the set of destinations of the P2MP TE Tunnel." is that some folk hoped 
to use something akin to a group address in this space. Thus the tunnel 
would be identified by tunnel ID, extended tunnel ID and group destination 
address. Multiple tunnels to the same destination group address would have 
different tunnel IDs (if from the same sender) and could have different 
extended tunnel IDs (if the senders made that choice).

Adrian

----- Original Message ----- 
From: "Lou Berger" <lberger@labn.net>
To: "Yakov Rekhter" <yakov@juniper.net>
Cc: <p2mp@labn.net>; <mpls@ietf.org>; <ccamp@ops.ietf.org>
Sent: Wednesday, June 28, 2006 3:01 AM
Subject: [mpls] Re: draft-ietf-mpls-rsvp-te-p2mp-06.txt [P2MP ID]


> Yakov,
>         See below.
>
> At 05:59 PM 6/27/2006, Yakov Rekhter wrote:
>
> > Lou,
> >
> > [clipped...]
> >
> > > > > > > >    The P2MP ID provides an identifier for the set of
> > > > > > > >    destinations of the P2MP TE Tunnel.
> > > > > > > >
> > > > > > > >The last sentence above has to be deleted.
> > > > > > >
> > > > > > > why, what's incorrect about it?
> > > > > >
> > > > > >To begin with, the last sentence states that the P2MP ID, *by 
> > > > > >itself*
> > > > > >"provides an identifier for the set of destinations of a given 
> > > > > >P2MP
> > > > > >TE Tunnel". However, since a given P2MP ID, *by itself*, may not
> > > > > >be unique, how could it unambiguously identify the set of 
> > > > > >destinations
> > > > > >of such tunnel ?
> > > > >
> > > > > You are 100% correct, how about fixing it by saying:
> > > > > "The P2MP ID together with Tunnel ID, Extended Tunnel ID provide an
> > > > > identifier for the set of destinations of the P2MP TE Tunnel."
> > > >
> > > >Given that P2MP ID is unique within the scope of the root, why a
> > > >tuple <Extended Tunnel ID, P2MP ID> is not sufficient to identify
> > > >the set of destinations of a P2MP tunnel ? After all, you agreed
> > > >further down that such a tuple identifies a P2MP tunnel.
> > >
> > > Because this would preclude the use of tunnel ID to further partition
> > > the session space (as is done today for P2P-TE.)  The tuple I
> > > mentioned included all three.
> >
> > You, yourself agreed that "A combination of this address and P2MP
> > ID provides a globally unique identifier for the P2MP tunnel."
>
> Well actually my point was: "... the uniqueness of the tuple <P2MP ID, 
> Tunnel ID, Extended Tunnel ID>. "
>
> > Since the set of destinations of a P2MP tunnel is just a part of
> > the tunnel, identifying the tunnel should be sufficient in order
> > to identify the set of destinations.
>
> right, and all three fields are needed.
>
> > Since a tuple <Extended Tunnel ID, P2MP ID> is sufficient to identify
> > the tunnel, the same tuple is sufficient to identify the set of
> > destinations of the tunnel. Therefore, the Tunnel ID is not
> > needed to identify the set of destinations of the tunnel.
>
> we disagree.
>
> > Yakov.
> >
> > P.S. I would also caution against trying to draw too much analogy
> > with the P2P tunnels. As I explained to you in one of my previous
> > e-mails, the semantics of some of the fields carried in the Session
> > object in P2MP tunnels is quite different from the semantics of
> > their counterparts in P2P tunnels.
>
> Sure, but you've not stated any reason or need to change the semantics wrt 
> tunnel ID...
>
> Lou
>
> _______________________________________________
> mpls mailing list
> mpls@lists.ietf.org
> https://www1.ietf.org/mailman/listinfo/mpls