[MIB-DOCTORS] MIBDoc review of draft-ietf-bess-l2l3-vpn-mcast-mib-07

[Glenn Mansfield Keeni <[e-mail]>]

Dear Tsuno/Zhang
      Thanks for waiting. The review of
  draft-ietf-bess-l2l3-vpn-mcast-mib-07 follows.

Glenn

C1. Abstract:
     The draft now defines 2 MIB modules.  Please revise the abstract
     and probably the title of the document too.

C2. The MIBs (L2L3-VPN-MCAST-TC-MIB, L2L3-VPN-MCAST-MIB) compile OK.
     (Three {type-unref} warnings are there, may be ignored.)

C3. Page 4:
       s/3.  Summary of MIB Module/
         3.  Summary of MIB Modules/

C4. Page 6: L2L3VpnMcastProviderTunnelPointer DESCRIPTION
       s/"Denotes a pointer to the row pertaining to a table entry/
        /"This textual convention represents a pointer to a row in
         the table represented by the following object of type
         L2L3VpnMcastProviderTunnelPointerType./

C5. Page 7: L2L3VpnMcastProviderTunnelPointer DESCRIPTION
         The explanation in the last paragraph seems out of place.
         It may be removed.

C6  Page 7: L2L3VpnMcastProviderTunnelPointerType DESCRIPTION
         it is unclear when the value 'null(0)' will be used.
         Is this allowed only when the corresponding object of type
         L2L3VpnMcastProviderTunnelPointer has a value that is a
         zero-length string ? If yes, please make that clear.

C7. Page 9: l2L3VpnMcastPmsiTunnelAttributeTable DESCRIPTION
         s/created by a PE router/maintained by a PE router/

C8. Page 12: l2L3VpnMcastPmsiTunnelAttributeId
          Do you really want to keep this object in L2L3-VPN-MCAST-MIB.
          It will change every time a new "tunnel type" is added to
          L2L3VpnMcastProviderTunnelType. That will defeat the purpose
          of separating L2L3-VPN-MCAST-TC-MIB from L2L3-VPN-MCAST-MIB
          It may be a good idea to define a textual convention like
              L2L3VpnMcastPmsiTunnelAttributeId
          in the L2L3-VPN-MCAST-TC-MIB and use that textual convention
          in the L2L3-VPN-MCAST-MIB

C9. Page 13: l2L3VpnMcastPmsiTunnelAttributeId DESCRIPTION
A.       s/Thus, the size of this object is 16 octets in IPv4/
            Thus, the size of this object is 8 octets in IPv4/
B.       The last 2 paragraphs do not tie up well with the previous
          paragraphs in the DESCRIPTION.
C.       In the last paragraph
          "this object is a pair of source and group IP addresses"
          is unlcear. Please clarify.

C10. Page 15: Security Considerations
          I would think that any field that reveals information about
          a Multicast VPN and/or its members is sensitive.
          Does the l2L3VpnMcastPmsiTunnelAttributeId field reveal
          information about the participating members? If yes, it must
          be marked as sensitive.

C11. Editorial:

A. This does not pertain to the MIBs as such,
    but I am uncomfortable with the  several variations
    of the phrase
    "Layer 2 and Layer 3 Virtual Private Networks (VPN)
     that support multicast"
    that appears in the text. I do not have a solution
    on hand but it would be perhaps be more readable to
    define and use a simple name/notation the beast for
    which the MIB is being designed (e.g. "L2L3VPNMCast").

B. Same with the phrase
     "Layer 2 (L2) and Layer 3 (L3) VPN (Virtual Private
      Network)
     it would be probably be easier on the reader if an
     abbreviation like L2L3VPNs was used where ever
     applicable.

C12. P2:Para4: s/within MIB module specifications//;

C13. General:
A.      The DESCRIPTION clauses could do would some more
         packing. (Too much space at the end of lines)
B.      Please check the articles a/an/the once again. Some
         usages do not seem right to me.