The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2002-Sep> msg00049



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

Clarification on Handling a specific event in LDP

  • From: "Ramia, Kannan Babu" <kannan.babu.ramia@intel.com>
  • Date: Tue, 17 Sep 2002 22:07:45 -0700

Hi all

 

    I am encountering a peculiar situation where I need some clarification.

 

    Here is my problem

 

    My LSR A acts in DOD ordered mode (also assumed all the other LSRs in the domain acts in the same mode). For a particular FEC X it has found itself as an Egress LSR and it knows that the penultimate LSR B supports IMPLICIT NULL LABEL and it distributed a NULL Label for that FEC X. Now in the data plane at my LSR it will receive only unlabeled packets since the penultimate LSR pops the label in the packet (assuming no stacking of labels).

 

    Now in later point of time if some other LSR C gets added to this MPLS Domain which happens to be my Next Hop for the FEC X. Now my LSR will become an intermediate LSR for that particular FEC-X and it will request a label from LSR C and install it in the forwarding plane. Now the problem arises because it had distributed an NULL Label in the upstream, so before installing this new label it needs to withdraw the NULL label from its penultimate LSR B and redistribute a new label so that LSR B wont acts like a proxy egress for this FEC-X. If we send a label withdraw message then that message will progressively travel in the upstream direction and the whole Lsp context will be lost. Since LSR A now send a new label map for the same FEC-X the LSR B which receives it, will not propagate because it has been configured as a DOD ordered mode.

 

    Can anyone suggest any solution to this problem…

 

 

I have a solution for this problem.

Basically if any LSR receives an Label withdraw message from its valid Next hop neighbor then before propagating to its upstream peers it need to check the value of the label being withdrawn, if the label happens to be an NULL label then it should not propagate the label withdraw to its upstream peers. Now this LSR should now assume itself as an egress for that LSP.

 

Second one could be there should be a provision to change the label for the particular FEC. I.e any point of time the valid next hop LSR can advertise a change of label for that FEC. Here problems are we need to maintain the context of that advertisement (LM).

 

Kannan Babu

Senior Network Software Engineer

Network Software Division

Intel Corporation        

Bangalore

841 2333 extn 2506