The MPLS WG Archive[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index] Comments on draft-mpls-rsvp-lsp-tunnel-06.txt
Philip Matthews wrote: > > Thanks for the information. > I guess the answer is that an implementation SHOULD make an effort to > handle such cases, and not just reject the change with a PathErr or > ResvErr. IMO, if the hardware (or its operating system) is capable of changing the egress label for an established tunnel, it should attempt to do so. If the hardware can't do this, or if the attempt fails, then you must report this fact to the ingress node so that it can take appropriate action (probably involving tearing down the tunnel and reestablishing it). This report would be in the form of a PathErr message. A ResvErr message to the egress node might also be useful, but I don't think it will be necessary. -- David
|
|