Traffic Engineering Link Management Information Base
draft-ietf-mpls-telink-mib-07

[Ballot discuss]
I'd like to understand why in table

  1.3.6.1.2.1.10.200.1.4      teLinkBandwidthTable
  1.3.6.1.2.1.10.200.1.4.1    teLinkBandwidthEntry
  1.3.6.1.2.1.10.200.1.4.1.1  teLinkBandwidthPriority
  1.3.6.1.2.1.10.200.1.4.1.2  teLinkBandwidthUnreserved
  1.3.6.1.2.1.10.200.1.4.1.4  teLinkBandwidthRowStatus
  1.3.6.1.2.1.10.200.1.4.1.5  teLinkBandwidthStorageType

there is a gap between column 2 and 4. I.e. why was ccolumn 3 skipped?

Same for:

  1.3.6.1.2.1.10.200.1.7  componentLinkBandwidthTable
  1.3.6.1.2.1.10.200.1.7.1  componentLinkBandwidthEntry
  1.3.6.1.2.1.10.200.1.7.1.1  componentLinkBandwidthPriority
  1.3.6.1.2.1.10.200.1.7.1.2  componentLinkBandwidthUnreserved
  1.3.6.1.2.1.10.200.1.7.1.4  componentLinkBandwidthRowStatus
  1.3.6.1.2.1.10.200.1.7.1.5  componentLinkBandwidthStorageType

Also...
If you do not intend to represent fractional values, then it would
be good to add some text to this TC:

  TeLinkBandwidth ::= TEXTUAL-CONVENTION
    STATUS      current
    DESCRIPTION
      "This type is used to represent link bandwidth in bps. This
        value is represented using a 4 octet IEEE floating point
        format."
    REFERENCE
      "IEEE Standard for Binary Floating-Point Arithmetic,
        Standard 754-1985"
    SYNTAX      OCTET STRING (SIZE(4))

[Ballot comment]
Section 4 should add to the list of requirements met:  Create and bundle TE links.
(RFC Editor Note material).

The Security Considerations warning about the read-write/read-create objects
could be stronger, given how powerful this MIB is.

Does the IANA search through MIBs still, or should there be an IANA Considerations?

Gone through MIB-doc review, Bert shepharded it.
There are some outstanding comments from the mib doctor (Dave Thaler), but would like to IETF LC it and address
all potential issues together.

MIB Doctor review

-----Original Message-----
From: Wijnen, Bert (Bert) [mailto:bwijnen@lucent.com]
Sent: maandag 1 september 2003 20:43
To: 'Martin Dubuc'; Mpls (E-mail)
Subject: MIB Doctor review:draft-ietf-mpls-telink-mib-03.txt


- Interesting that title page claims that doc expires feb 2003?
  You porobably mean feb 2004

- I get this WMICng warning:
  W: f(telink.mi2), (1564,19) MIN-ACCESS value identical to access
    specified for "teLinkBandwidthUnreserved"
  Seems to me you can just remove that MIN-ACCESS from the MODULE
  COMPLIANCE.

- I see
    TeLinkSonetSdhIndication ::= TEXTUAL-CONVENTION
      STATUS      current
      DESCRIPTION
          "SONET/SDH indication type."
      SYNTAX      INTEGER {
                    standard(0),
                    arbitrary(1)
                }
  Since we normallyh do not start with zero (but with 1), I assume
  there is a reason you start with zero. Could that reason be described
  and is there a doc that explains this, so that you refernece it?

NITS:

- I see
    teLinkGroups
      OBJECT IDENTIFIER ::= { teLinkConformance 1 }

    teLinkCompliances
      OBJECT IDENTIFIER ::= { teLinkConformance 2 }
  Normally we do it the other way around, first Compliances, then Groups,
  See page 35, appendix D of draft-ietf-ops-mib-review-guidelines-02.txt

- I see in OBJECT-GROUP statement things like:
    DESCRIPTION
          "Collection of objects needed for the monitoring of
          resources associated with TE links."
  I would think the objects *at least a subset) are also usefull for
  configuration. How about:
    DESCRIPTION
          "Collection of objects for management of
          resources associated with TE links."
  Most OBJECT-GROUP descritpion clauses have similar "problem".

- I see:
    teLinkModuleFullCompliance MODULE-COMPLIANCE
      STATUS current
      DESCRIPTION
      "Compliance statement for agents that support the
        configuration and monitoring of TE Link MIB module."
  Mmmm. I would word it a bit different:
      "Compliance statement for agents that support read-create
        so that both configuration and monitoring of TE Links can
        be accomplished via this MIB module."
  Matter of taste I guess.

- I see:
    teLinkModuleReadOnlyCompliance MODULE-COMPLIANCE
      STATUS current
      DESCRIPTION
      "Compliance statement for agents that support the
        monitoring of TE link MIB module."
      MODULE -- this module

      -- The mandatory groups have to be implemented
      -- by all devices supporting TE links. However, they may all
      -- be supported as read-only objects in the case where manual
      -- configuration is unsupported.

      MANDATORY-GROUPS    { teLinkGroup,
                            teLinkBandwidthGroup,
                            componentLinkBandwidthGroup }

  It seems to me that that all of those 4 comment lines are redundant.
  The idea of the MODULE-COMPLIANCE statements is that they are both
  human and machine readable.

- I see hyphenation. That is something the RFC-Editor does not want/like.

- Section 6 starts with:
    6.  Brief Description of MIB Objects

      Sections 6.1-6.4 describe objects pertaining to TE links.  The MIB
      objects were derived from the link bundling document [BUNDLING].
  How abaout the section 6.5-6.7 ??

Thanks,
Bert