Re: Polling for two new working group I-Ds

["tom.petch" <cfinss@dial.pipex.com>]

----- Original Message -----
From: "Adrian Farrel" <adrian@olddog.co.uk>
To: <ccamp@ops.ietf.org>
Sent: Monday, August 13, 2007 5:33 PM
Subject: Polling for two new working group I-Ds

> 2. draft-nadeau-ccamp-gmpls-oam-requirements-01.txt
> This work is a charter milestone, and Deborah and I are both co-authors, so
> we are likely to be influenced in favour of adoption. So please voice your
> opinions loud and clear.
>

Mmmmm ... I agree that this will, most likely, be the basis of the charter
milestone but it does seem to lack clarity and consistency, especially from
about 4.3 onwards.  From many possible examples,

- if the reader is assumed to be familiar with [RFC3945, RFC4377] then I expect
those to be normative references.

- are requirements subject to RFC2119? I've seen this argued both ways recently;
here the approach seems to be 'yes' up to 4.2.3, 'no' for 4.3 onwards

-  "... between control planes of different layers.  For the operational
convenience,

4.10 Support for OAM Interworking for Fault Notification ..."

Am I missing something?

- "4.3 Interaction between the GMPLS Control Plane and Data Plane. ...The
operational state of one plane should be automatically reflected to that of
another plane. "

Reflecting control plane operational state to data plane does not sound like OAM
as I understand it.  Like, does that mean when the control plane goes down, the
data plane is taken down:-(?

 -" 4.2.2 The Management plane SHALL provide the capability to verify routing
consistency in the control plane with the data plane."

I am struggling to imagine what this is (especially given the emphasis elsewhere
in this document on the separation of data and control planes).

-  "4.2.1 The change of recovery type of LSPs is required, for example, from
unprotected to 1+1 protected LSP, or from full LSP rerouting to pre-planned LSP
rerouting. Other capabilities include changing a recovery type of LSPs [RFC4426,
Segment]"

What is the difference between 'the change of recovery type of LSPs' and
'changing a recovery type of LSPs' (apart from one being a noun and the other a
verbal noun)?

Like I said, mmmmmmm

Tom Petch

> Thanks,
> Adrian
>
>
>