Frame Relay Service MIB WG (frnetmib) Minutes
Chair: Andy Malis <Andy.Malis@vivacenetworks.com>

The frnetmib working group met Tuesday, August 1, 2000 from 1415 to 1515.

AGENDA:

1. Administrivia, current documents status

    o The following two drafts have been approved by the IESG to be Proposed Standard RFCs:

      draft-ietf-frnetmib-atmiwf-06.txt,
      Definitions of Managed Objects for Monitoring and Controlling the Frame Relay/ATM PVC Service Interworking Function, approved by IETF for PS

      draft-ietf-frnetmib-frs-mib-12.txt
      Definitions of Managed Objects for Frame Relay Service, approved by IETF for PS

    o The following draft is in IESG review:

      draft-ietf-frnetmib-mfrmib-01.txt
      Definitions of Managed Objects for Monitoring and Controlling the UNI/NNI Multilink Frame Relay Function (review due by mid-August)

2. Rob Steinberger and Orly Nicklass, Frame Relay Circuit Interface MIB

    http://www.ietf.org/internet-drafts/draft-steinberger-frsi-00.txt

This MIB provides the ability to generate ifEntries for frame circuits.

The motivation is that neither RFC 2115 nor RFC 1604 create ifEntries for DLCIs.  RMON requires an ifIndex to monitor a data source, so the RMON WG requested that we do this work during the Adelaide meeting.

In addition, routing MIBs (such as the CIDR Routing Table and Net to Media table) require ifIndex to control routing.  Without it, you cannot control or display the routing to or on virtual circuits in a non-proprietary manner.  There is not a need to create ifIndex for every circuit, but it would be nice to have one for the interesting circuits.

Rob presented the MIB structure, which has two tables, a Circuit table and an Interface Map table.

Rob noted that the MIB requires a new ifType to be allocated by IANA.

The group consensus was to continue the work and to generalize it as much as possible, including folding in ATM.

The chair will ask on the list if there are any objections to making this a WG draft.

3. Rob Steinberger and Orly Nicklass,
    draft-ietf-frnetmib-frmrelay-service-01.txt, Definitions of Managed
    Objects for Frame Relay Service Level Definitions

    http://search.ietf.org/internet-drafts/draft-ietf-frnetmib-frmrelay-service-01.txt
    Also see http://www.frforum.com/5000/Approved/FRF.13/frf13.pdf

Since this MIB has been previously presented, no introductory slides were required.

Rob went over the list of major changes and issues that went into creating -01 as a result of working group last call on -00.  Several remaining open issues were discussed.  Those that resulted in action items were:

On issue 4, free running counters, the following four suggestions were made:
	1) Min/Max/Avg have no meaning over time. Remove them.
	2) Remove ALL free running counters.
	3) Use RFC 2493.
	4) Add period of relevance to the Min/Max/Avg variables.

The author had incorporated choice 4, but the consensus was to go with 1 instead.

Issue 5: table fragmentation: Don't keep availability in a separate table.

The original solution was to do nothing - needs further discussion.

Ken Rehbehn didn't see why this is a separate case, and this would be confusing in the future as well.

There was no compelling case for a separate table - having one table simplifies indexing.  We decided to join the tables without changing the indexing scheme.

Issue 6: Bidirectional flows:

The solution was to change the indexing scheme and modify the reference points to show remove and local reference points.  This was accepted.

However, Ken also asked for an example with a picture in the method of operations section for bidirectional flow.  This will be added.

No other changes to -01 were proposed in the meeting.  The MIB will be updated as discussed and forwarded to the IESG as the result of WG last call.

4. Open discussion

There being none, the group adjourned.