[mpls] MPLS-RT review of draft-rhd-mpls-tp-psc-priority

[Yaacov Weingarten <[e-mail]>]

Hi,

I was requested to conduct an initial review of
draft-rhd-mpls-tp-psc-priority-01 as part of the effort to determine if the
draft is suited for a WG adoption poll. I have read the draft and have the
following notes:

1. I find it rather disconcerting that the draft is referencing a LS that
was received from the ITU. While I am sure that the points rasied in the LS
as very well thought out and pertinent, I cannot see that the LS should be
considered a "standards document" that could be referenced by a future RFC.
At best, the LS could be considered a "contribution" at an ITU meeting, and
I do not believe that an ITU document would reference a contribution! I
therefore think that the authors should transfer into the draft whatever
information they feel is appropriate from said LS.

2. In general, the draft is addressing itself to two topics regarding
RFC6378,
a. Change of relative priority between SF-P and FS - the suggestion is
essentially to rollback the relevant changes that were made to the RFC
during the final IESG review to the version previous to that review.

b. Change of relative priority between ClearSF and SF/SD. This is based on
a particular use-case that is mentioned in the referenced LS, and that I
suggested be explicitly explained in the draft.

c. Introduce (in an appendix) the use of a Freeze command. Not sure what
the motivation for this in this context is.

In my opinion, all three of the points are valid for work by the WG and
should be fully discussed in the WG prior to publication of the draft - but
could be discussed as part of a WG draft.

3. The format of the draft is also rather interesting - it seems to be
presenting an errata note to RFC6378 rather than describing the desired
behavior. I leave it to the WG Chairs to decide what is the best format for
the draft moving forward.

4. One other note regarding this draft in the context of several other
drafts that are proposing changes to the PSC protocol for MPLS-TP Linear
Protection. I think that some of these should be combined into a more
robust proposal for the sake of the implementers. Otherwise, there may be a
need for a future "reader's guide" for an implementer to figure out what is
in PSC and what is no longer there!

Bottom line - I think that this draft presents valid points, it should be
corrected to not reference the ITU LS prior to acceptance as a WG draft.
And the WG should strongly consider consolidating this work with some of
the other drafts to make a more complete and consistent definition of PSC
available.

Hope this helps,

-- 
Thanx and BR,
yaacov

*Still looking for new opportunity*