Re: Doubts on draft-ietf-mpls-rsvp-te-p2mp-01

["Adrian Farrel" <adrian@olddog.co.uk>]

Hi Mohan,

The correct place for your question is really the MPLS mailing list (see
http://www.ietf.org/html.charters/mpls-charter.html)

Brief answers to your questions below.

> Draft: draft-ietf-mpls-rsvp-te-p2mp-01.txt

> 1) In this draft what is the role of sub-group ID. What will be the
> value of this? I mean IP address or some other number? If it is IP
> address which one?

The sub-group ID allows you to distinguish different sub-trees of the same
P2MP tree. this allows the signaling state to be separated and managed
distinctly (for example, using separate Path messages). The ability to
provide such a distinction is important when more than one Path message
for the same P2MP LSP might be received by the same LSR.

For more details, see the section 3.3 of the draft.

Sub-Group Originator ID should be set to the TE Router ID of the transit
LSR.
Sub-Group ID should be chosen as unique within the context of the
Sub-Group Originator ID.

> 2) Any vendor is already supporting this feature?
> Setup of RSVP-TE P2MP LSPs? Can anyone list the vendors?

I am aware of three vendors with "early" implementations of this draft. I
must let them come forward and announce their own names.

> 3) Which one of the following is best suitable for CSPF in P2MP?
> Core Based Trees, Reverse Path Forwarding and Spanning trees?

There are many choices.
Since this is traffic engineering, all algorithms are acceptable, and
vendors might expect to make significant marketing through their
implementation of path computation.

Regards,
Adrian