The MPLS-OPS Archive[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index] Re: : Newbie Questions
Hi, thanks everyone so far for the replies. I think my first question on the start and endpoint of the nested LSP caused a little bit of confusion, as I haven't quoted every detail of the example from the FAQ on this website. <http://www.mplsrc.com/faq2.shtml#MPLS Components> Again the picture: |------| 1 1 |-----| | R1 |--\ /---| R5 | |------| \ 2 2 2 / |-----| \|------| |-----| |-----| / | R2 |---| R3 |---| R4 |/ |------| |-----| |-----| Most of the replies assumed that the level 2 label is pushed at R2, and then I agree that the nested LSP starts at R2. However, the example states that *R1* pushes the new label for R2 (and also swaps the label for R5), and that the nested LSP is R2-R3-R4. If this is correct, I don't understand how this matches the RFC3031, which says that a LSP begins with the LSR that pushes the new label onto the label stack, which was basically the core of my question when I wrote "...the R2-R3-R4 path should start at R1 and finish at R5...". I think M. Elk's reply said what I meant (here LSP2 start at R1): "we have one LSP ,say LSP1 as R0,R1 and R5 . we have another LSP , say LSP2 as R1,R2,R3,R4,R5 . when R1 receive label packet over LSP1 , it swap the label and push the labeled packe inside LSP2 to reach R5 . so R1 will transmit the packet with 2 label , the top is for LSP2 and the bottom is for LSP1." Do I still miss something or is the FAQ-example just not built properly (when saying that R1 pushes the new label, but the LSP starts at R2)? Sorry for the confusion and thanks again for the replies to the other points! Cheers, Andreas ------- The MPLS-OPS Mailing List Subscribe/Unsubscribe: http://www.mplsrc.com/mplsops.shtml Archive: http://www.mplsrc.com/mpls-ops_archive.shtml MPLScon 2005 - May 16-19, NYC, NY http://www.mplscon.com/
|
|