The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2006-Nov> msg00146



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

[mpls] Unique Identififcation of LSP

  • From: <neil.2.harrison@bt.com>
  • Date: Thu, 30 Nov 2006 08:39:11 -0000
  • Cc: mpls@lists.ietf.org
  • Thread-Index: AccT6BI4Pv8jd+ZESe2kwQWCU9gQTgAb4JRA
  • Thread-Topic: [mpls] Unique Identififcation of LSP
  • X-MIME-Autoconverted: from quoted-printable to 8bit by cell.onecall.net id kAU8b6s27985
  • X-OriginalArrivalTime: 30 Nov 2006 08:39:11.0672 (UTC)FILETIME=[024FEB80:01C7145B]

I believe we should add these observations:

-	this only applies to this particular signalling protocol;

-	it does not apply in the traffic data-plane.  That is, one does
not have such information here, just a single (20 bit) forwarding label
that can be changed on each link connection and which is proxying for
some distant LSP termination point.  This has implications wrt
identifying/distinguishing misconnectivity between LSPs..... noting that
this can occur in both a horizontal sense (ie between LSPs created at
the same level in a nested stack) and a vertical sense (ie between LSPs
created at different levels in the stack).....be these generated by the
same or different signalling protocols.

regards, Neil

> -----Original Message-----
> From: George Swallow [mailto:swallow@cisco.com] 
> Sent: 29 November 2006 18:53
> To: Anergy Virt
> Cc: mpls@lists.ietf.org
> Subject: Re: [mpls] Unique Identififcation of LSP 
> 
> 
> > I have read that the 5-tuple {tunnel-endpoint address,tunnel-ID, 
> > extended tunnel-ID, sender address, and LSP ID} can 
> uniquely identify 
> > an LSP. My question is it locally unique between two nodes 
> or globally 
> > unique identifcation
> > 
> > A----B-----C------E
> >        \     /
> >          D
> > 
> > For example two LSPs are set up from A to E via the same 
> tunnel. i.e. 
> > the have same end points and same path ABDE. Will the 5 
> tuple uniquely 
> > identify any LSP across any two nodes e.g. B and D? Will the values 
> > remain same for this LSP across two other nodes in the path 
> e.g. C and 
> > E?
> 
> Yes, provided that the extended tunnel ID is a globally 
> unique id. (Normally this is set to the head-end's router ID).
> 
> ...George
> 
> ==============================================================
> ==========
> George Swallow             Cisco Systems                  
> (978) 936-1398
>                            1414 Massachusetts Avenue
>                            Boxborough, MA 01719
> 
> _______________________________________________
> 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