Internet and Management Support for Storage (imss) WG
Meeting - Vancouver, Canada
Tuesday, November 8, 2005: 1740-1840
------------------------------------------------------
Administrivia, agenda bashing, draft status review, etc.: 15 min
David L. Black, EMC (new imss WG chair)
Blue sheets
Note Well
Milestones (see WG charter page on IETF web site)
WG Draft status
- IP over FC draft is in IETF Last Call
- FAM and NSM MIBs should go to IETF Last Call by the
end of next week (Keith McCloghrie and Bert Wijnen
will coordinate to make this happen).
- Remaining 3 WG MIBs will go to WG Last Call by end
of November.
--> Record status
T11.5 Status, Fibre Channel MIBs under development: 20 min
Roger Cummings (Symantec, T11.5 chair)
See slides.
T11.5 has 3 more MIBs that will be sent to the imss WG - Zone Server
MIB, Registered State Change Notification MIB, and Fabric Configuration
Server MIB. The imss WG's current Apr 06 milestone to determine what
to do next will be replaced by a milestone (later in 2006) to work on
these 3 MIBs. These 3 MIBs may not be ready for handoff to imss prior
to the next IETF meeting, but the mechanism used in Paris for the VF
MIB can be reused - the imss WG can decide to accept the drafts
as official WG drafts subject to a pending T11 formal votes to pass
change control to the IETF.
T11.5 is considering assigning MIB development responsibility to protocol
development work groups instead of the current approach of having a separate
group that works on MIBs. This will be discussed at the T11 meetings in
early December.
FLIP (FAIS Line Interface Protocol) Conceptual Discussion: 25 min
Roger Cummings (Symantec)
draft-cummings-imss-flip-00.txt
See presentation. Netconf appears promising for the configuration
aspects of FLIP, but currently contains no support for events or
notifications from the configured device. Use of netconf would
result in functionality at the level of the configuration model
instead of a 1-1 match between FAIS API calls and RPCs. The
required event/notification mechanism needs to be capable of
carrying a fair amount of structured information - it is somewhat
analogous to the use of COPS to ask a policy decision point
"I just received this RSVP reservation request, what should I do
about it?". There are also concerns about effective transmission
of bulk binary data (encoding in XML is unworkable, but encapsulation
or wrapping in XML is probably ok). The netconf group is not currently
standardizing schemas, but will want/need to, so this could be a
timely interaction between imss and netconf. Roger will send a
note to the netconf list soliciting interest in helping with
schema definition.
The FLIP concept and suggested use of netconf will be discussed
further at T11 before the next imss WG meeting (March 2006, Dallas).
|