Link Management Protocol (LMP) for Dense Wavelength Division Multiplexing (DWDM) Optical Line Systems
draft-ietf-ccamp-lmp-wdm-03

Revision seems to have addressed all issues except one comment from Ted Hardie. That comment has been addressed with an RFC-Editor note.

AD is now checking of IANA is also happy

[Ballot discuss]
I should have seen the below in my AD review. Appology that I missed it.

From Pekka (OPS directorate):

> o draft-ietf-ccamp-lmp-wdm-02.txt
>
>I only glaced this through for nits, but I noticed a bigger issue.
>IANA considerations section does not define how the name spaces being
>created will be managed (IETF Consensus, Protocol Action, IESG
>Approval, etc.).

The IANA considerations is not very clear indeed.
But the good news is that it does NOT specify a new namespace.
In fact it requests allocation in an existing namespace as defined
in draft-ietf-ccamp-lmp-10.txt, which was approved on Oct 17th.
The latter doc does have the appropriate IANA instructions.

So this one should probably be fixed with:
OLD:
    6. IANA Considerations
 
      LMP defines the following name spaces that require management:

      - LMP Message Type.
      - LMP Object Class.
      - LMP Object Class type (C-Type) unique within the Object Class.
      - LMP Sub-object Class type (Type) unique within the Object Class.

      This memo introduces the following name spaces which need
      assignment:

            o CONFIG
                - LMP-WDM_CONFIG (suggested C-Type = 2)

            o CHANNEL_STATUS
                - LINK_GROUP (suggested C-Type = 4)

      LMP sub-object Class type (C-Type) should be assigned from the range
      0-127.

            o DATA_LINK
                - Link_GroupId (suggested sub-object Type = 3)
                - SRLG (suggested sub-object Type = 4)
                - BER_Estimate (suggested sub-object Type = 5)
                - Optical_Protection (suggested sub-object Type = 6)
                - Total_Span_Length (suggested sub-object Type = 7)
                - Administrative_Group (suggested sub-object Type = 8)

NEW:
      LMP [LMP] defines the following name spaces and how IANA can
      make assignments in those namespaces:

      - LMP Message Type.
      - LMP Object Class.
      - LMP Object Class type (C-Type) unique within the Object Class.
      - LMP Sub-object Class type (Type) unique within the Object Class.

      This memo introduces the following new assignments:

      LMP Object Class Types:

            o under CONFIG class name (as defined in [LMP]):
                - LMP-WDM_CONFIG (suggested C-Type = 2)

            o uncer CHANNEL_STATUS class name (as defined in [LMP]):
                - LINK_GROUP (suggested C-Type = 4)

      LMP Sub-object Class names:

            o under DATA-LINK Class name (as defined in [LMP]), Sub-object
                Class names should be assigned from the range 0-127:
                - Link_GroupId (suggested sub-object Type = 3)
                - SRLG (suggested sub-object Type = 4)
                - BER_Estimate (suggested sub-object Type = 5)
                - Optical_Protection (suggested sub-object Type = 6)
                - Total_Span_Length (suggested sub-object Type = 7)
                - Administrative_Group (suggested sub-object Type = 8)


>Nit:
>
>The contributors section lists 23 persons. Does this document confuse
>"contributors" with an "acknowledgements" section? Some have probably
>been more substantial contributors, and some others less, and if
>possible splitting them might be useful, not not a critical thing to
>me at least..

I'd like to not request changes here (too much history to break it open)

[Ballot discuss]
Very mild discuss, as it may well be a misunderstanding on my part.  In reading section 2.4, on
Fault Management, I understood the first section on Fault Detection to be specific to
faults detected between peers.  It was not clear whether the same section would apply
to faults detected on links between a peer and an OLS.  The introduction says:

  Although there are many similarities between an LMP session between
  two peer nodes and an LMP session between a peer node and an OLS,
  there are some differences as well. The former type of LMP session
  is used to provide the basis for GMPLS signaling and routing. The
  latter type of LMP session is used to augment knowledge about the
  links between peer nodes.

Though that might imply the same considerations apply here, and explicit statement on whether
fault management on peer-ols links has the same characteristics as peer-peer links seems
like it is needed.  My apologies if I have misunderstood here, and this covered by other
text.

AD review comments posted to WG list

-----Original Message-----
From: Wijnen, Bert (Bert) [mailto:bwijnen@lucent.com]
Sent: donderdag 13 maart 2003 23:12
To: Ccamp-wg (E-mail)
Subject: AD review of: draft-ietf-ccamp-lmp-wdm-01.txt


I am basically OK with this document except for:

- I see no IPR statement which is required for stds track
  documents
- I see no IANA considreations section, while the document
  does need IANA to make assignments.
- You may want to add the copyright verbage at the end
  of the document.
- I suspect that security ADs may find the sec.considerations
  section weak. You may at least want to state that you recommend
  to use the same security considerations as in the [LMP] document

Maybe the authors can do a quick rev, since I do not intend
to ask for IETF Last Call before the meeting in SF is over.

Thanks,
Bert