[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Soliciting comments on moving drafts to WG status



hi richard, see in-line

1. Loose Path Re-optimization draft-vasseur-ccamp-loose-path-reopt-02.txt This draft is stable and has an implementation. The work is predominantly pertinent to inter-domain signaling, but could also be used within a domain. The meeting in San Diego reported relatively few as having read the draft, but no objection to it becoming a
WG draft.

Yes. Though I would like the authors to mention GMPLS and drop the focus on MPLS since they say in the abstract that this applies to "packet and non-packet TE LSPs".

agree, i don't know if this has already being pointed out but the statement could even become somehow confusing to the audience, so it requires to explicitly state [RFC 3209] and [RFC 3473] for packet LSPs (as GMPLS support by definition PSC LSPs)


now concerning non-PSC LSPs, there is also a point to be addressed
that in order to achieve non-disruptive re-optimization using MBB
one would require double counting as a parallel non-PSC LSP would
be required (traffic is then send over both LSP before tearing the
old one)

Agree on these points. Would the authors address these issues in the next revision?

imho, these changes should be committed in the next revision and this independently of its promotion as WG I-d


And while you're at it, can you fix the double quotes throughout?

this has been recorded in the meeting minutes and hope it will be addressed in the next revision


2. A Transport Network View of LMP

Not sure. Adrian mentioned that this would possibly identify items of work for ITU and IETF. What is the thinking of the authors about the draft after the protocol modifications are finished?


If the expected outcome is an alignment of the IETF and ITU views
on LMP, then the draft would have served its purpose and would
not need publication as Informational.

your question is sensible, the reason is that in order to exchange views we need first to agree 1) that we want to work on it then 2) that we are in agreement about these views (you will also find part of the response to your in section 6.4) and finally 3) that we are in agreement on how to progress the work

I support making it a WG draft, at least to be able to initiate liaisons with SG15. I'm wondering though what the authors would like
the final outcome to be.

as part of the outcome, if there are commonalities in terms of discovery functionality, then an interoperable/compatible LMP mechanism may be considered within the scope of the process


If all issues are resolved, then in my opinion this draft should be
made to expire eventually. Otherwise, documenting the areas of
divergence as Informational would be more than appropriate.

nice way to conclude the discussion point because if there are no commonalities in terms of functionality and scope, we would have (with this i-d) closely documented why we didn't do so (but yes, i agree with you in both cases this document is meant promoted in the long term for the record)