The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-Nov> msg00257



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

LSP life time

  • From: Jeremy Lawrence <jlawrenc@cisco.com>
  • Date: Fri, 24 Nov 2000 09:51:35 +1100
  • Cc: "mpls uunet" <mpls@UU.NET>

At 00:03 11/24/2000 +0500, Faisal S. Naik wrote:
>What is the lifetime of an LSP?

A hop-by-hop routed LSP will exist without change for as long as
there is no change in the route for the FEC for which it is established.
In an ideal network which never experiences routing changes, the
hop-by-hop routed LSPs' lifetimes will be infinite.

A traffic engineering LSP will exist without change for as long as
there is no significant change in link loading, or other routing
state, which would affect the routing of the LSP.

Other uses of LSPs are possible, and some of those might involve
more frequent changes or set-ups of LSPs.

>What is the number of tunnels that can be created within an LSP.

It is effectively unlimited. If a single extra level of labelling
is used inside the LSP, 1M LSPs can be carried inside an LSP.
However, labels can be stacked to an arbitrary number of levels,
so an arbitrary number of LSPs can be carried inside an LSP.

(It is also possible to use other types of tunnels, e.g. IPSec
or GRE, inside an LSP.)

Jeremy Lawrence


  • Follow-Ups:
  • References: