The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2002-Jan> msg00139



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

[Fwd: LDP protocol problem?]

  • From: Praveen Kumar <praveen@it.iitb.ac.in>
  • Date: Mon, 21 Jan 2002 17:19:38 +0530 (IST)
  • cc: MPLS Mailing List <mpls@UU.NET>
  • X-SMTP-Sending-IP: 144.16.116.2


On Fri, 18 Jan 2002, Jack Brennen wrote:

> If "update" Label Mapping messages were syntactically different than normal
> unsolicited Label Mapping messages, the whole problem could be avoided.  That
> is how I would fix the specification if it were my decision.

I agree with this.

Label Mapping message allows an optional mesg id field which can indicate
the mesg id of corresponding Label Request Message. Here either 
"update" or "unsolicited" mapping messages are not result of any label
request. So, how about carrying special mandatory (reserved) ids in these
fields, for these two cases. Thus this reserved id can logically/ 
syntactically differentiate these two cases.

For the cases mentioned by jack, LSR A can determine whether the received
release message is result of recently sent update mapping or previous
mapping by using the approach suggested by "gabhijit". 

The differentiation of mapping messages that are sent as a result of label
request, result of update or result of advertisement mode gives more
flexibility to an receiving LSR to take any further (may be 
future) decisions.

Correct me if am wrong.

-- 
S.Praveen Kumar
Graduate Student
KReSIT, IIT-Bombay
web: http://www.it.iitb.ac.in/~praveen