The MPLS WG Archive

[George Swallow <swallow@cisco.com>]

>   The solution SHOULD allow for make-before-break re-optimization of
>   any subdivision of the P2MP LSP (S2L sub-tree, S2X sub-LSP, S2L
>   sub-LSP, X2L sub-tree, B2L sub-tree, X2L tree, or B2L tree) with no
>   impact on the rest of the P2MP LSP (no label reallocation, no change
>   in identifiers, etc.).

Clearly in order to do make-before-break, some ID has to change.  And in
most cases the message that initiates the change will have to pass
through some nodes that aren't affected by the change.  So the
requirement seems to be a bit over-specified.  

I'm glad it's just a *SHOULD*.  It would be nice if you could say which
identifiers, but as were trying to be protocol neutral, that's a bit
difficult. 

Perhaps a more abstract requirement should be made?  Such as:

>   The solution SHOULD allow for make-before-break re-optimization of
>   any subdivision of the P2MP LSP (S2L sub-tree, S2X sub-LSP, S2L
>   sub-LSP, X2L sub-tree, B2L sub-tree, X2L tree, or B2L tree).
>   Further is SHOULD do so minimizing the signalling impact on the rest
>   of the P2MP LSP and without affecting the ability of the management
>   plane to the LSP.

...George


========================================================================
George Swallow             Cisco Systems                  (978) 936-1398
                           1414 Massachusetts Avenue
                           Boxborough, MA 01719

_______________________________________________
mpls mailing list
mpls@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/mpls