[mpls] MPLS-RT comments on draft-ali-mpls-inter-domain-p2mp-rsvp-te-lsp-08

[Gregory Mirsky <[e-mail]>]

Dear All,
Please find my MPLS-RT comments on draft-ali-mpls-inter-domain-p2mp-rsvp-te-lsp-08 below:

*       Is the document coherent?
I found that terminology and identification for Re-merge Recording Request flag need improvement. At least once the flag seem to be referred as Re-merge Recording Flag. And several versions of capitalization and use of quotation marks appear throughout the document.
Document can benefit from tightening up and stricter use of RFC 2119 terms in sections 4.3 Signaling Procedure and 5.0 Reoptimization Signaling Procedure. E.g. the last sentence of second paragraph in Section 4.3 "In this case, before the Resv message is sent to the upstream node, the node adds the RRO Attributes sub-object to the RRO and sets the "P2MP-TE Re-merge Recording Request" Flag" not clear whether operations are MUST, SHOULD or merely CAN. I can suggest more similar places on- or off-line.
Then there's reference to path error 25 which, as I understand, Notify Error defined in RFC 3209.

*       Is it useful?
Document proposes solution to realistic, protocol-wise, scenarios. Though without confirmation from network operator community it is rather difficult to conclude how real scenarios are and how helpful the proposed solution.

*       Is it technically sound?
As mentioned in the first bullet, normative language in sections 4.3 and 5 needs to be more definite.
Another technical aspect of proposed solution - generality. It might be beneficial to look for solution of described scenarios as part of GMPLS, not only as MPLS-TE.

*       Is the document ready to be considered for WG adoption?
I think that input from network operators is needed.

Regards,
        Greg