The MPLS WG Archive

["Busschbach, Peter B (Peter)" <busschbach@lucent.com>]

Hi Rishi,

I have a comment on your statements regarding PWE. Please see below.

Peter

> -----Original Message-----
> From: Kulkarni, Hrishikesh (Hrishikesh) [mailto:hkulkarni@lucent.com]
> Sent: Thursday, October 14, 2004 2:00 AM
> To: mpls@ietf.org
> Subject: [mpls] Label Space Query
> 
> 
> Hi all,
> 
> Does LDP or RSVP provide a method to signal a label as a 
> per-interface or
> per-platform label?
> 
> PWE3 signaling assumes that a PSN tunnel is available before 
> setting up a
> PWE3.

I don't think that is true. The PWE3 drafts are not very clear on this subject, but you could read section 2 of draft-ietf-pwe3-control-protocol-11.txt as a reference.

It mentions that a PSN tunnel is not required if a PW is setup between adjacent nodes. Furthermore, it does not say that a PSN tunnel must be available when a PW is set up. It merely mentions that a PSN Tunnel must be available when packets are forwarded from one PW end-point to another. In other words, two PEs could establish a PW before they establish a PSN Tunnel, as long as they don't transmit packets.

In general, in MPLS networks, the notion of binding inner labels to outer labels is meaningless, because Pen-ultimate Hop Popping will lead to removal of the outer label before the packet reaches the tunnel endpoint.

> But is there way 
> to signal an inner label as per-interface or per-platform 
> between two peers,
> already connected by PSN tunnel.
> 
> 
> thanks
> rishi.
> 
> _______________________________________________
> mpls mailing list
> mpls@lists.ietf.org
> https://www1.ietf.org/mailman/listinfo/mpls
> 

_______________________________________________
mpls mailing list
mpls@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/mpls