Control and Provisioning of
Wireless Access Points (capwap) Working Group, March 26, 2009 3:10pm-4:10pm
Notes: Dorothy Stanley
Chair (Mahalingam Mani,
Dorothy Gellert) called the meeting to order at 3:15 pm.
- Chairs obtained volunteers for a note taker (Dorothy
Stanley) and a jabber scribe (Dan Romascanu).
- Review of agenda, no changes identified
- Intro/Administrivia (3
min, Chairs): Note Well,
Minute-taker, Jabber-scribe.
- CAPWAP
Protocol: CAPWAP Standards & Informational RFC (5 min, Chairs):
5415-5418
- Report on MIB drafts (30 min, Yang
Shi/Yong) http://www.ietf.org/internet-drafts/draft-ietf-capwap-base-mib-04.txt
and http://www.ietf.org/internet-drafts/draft-ietf-capwap-802dot11-mib-03.txt
- Implementation Information – updates, if
any (5min)
i.
EPCGlobal
ii.
WLAN
- Next Steps (5-10 min, Chairs)
i.
MIB drafts –
ready for IESG?
- Other work proposals in WG
- Chair: Review of recently published RFCs
- Chairs reviewed the list of document published,
and congratulated the Group on the milestone.
- Dan: The three individual submissions can now be
released for publication. Expect them to be published shortly.
- Richard: MIB Status
- Status of CAPWAP Base MIBCurrent
version: 04
i.
Total number Issues found by WG: 43
ii.
Issues closed: 39
iii.
Open issues: 4
- CAPWAP Dot11 MIB
i.
Current version:
03
ii.
Total number
Issues found by WG: 18
iii.
Issues closed: 12
iv.
Open issues: 6
- Comment: Have all of the issues been posted to
the list? Some still need to be posted, will do this.
- Open issue: Need a mechanism that preserves the
values of ifIndex. Dan explains the issue as he raised it. Richard
brought a solution; keep the IFIndex value persistent. This is not a
common restriction. Implementations will need to take standard SNMP
packages and change to have persistent values. Text has been added to the
document to explain this. Another solution was proposed, which would move
the problem to the operator.
Believe we can live with the IF Index restriction.
- Open Issue: Dot11 MIB should add a new Terminology “WLAN Profile”. Add to the overview
section: “WLAN Profile”, a configuration profile which includes
several parameters such as MAC
Mode. The value scope of profile’s
Id is 1 to 512. Also, it should explain the relationship between WLAN
Profile Id and WLAN ID. For MIB
table, Current WLAN ID object in the capwapDot11WlanTable should be replaced with WLAN Profile Id. and capwapDot11WlanBindTable
should add a WLAN Profile Id Object.
- Open Issue: Add a WTP Profile Function to the Base MIB. The WTP Profile
includes several Configuration parameters such as static IP for a
specific WTP. Would be created by an operator before a specific WTP
connects to AC. Discussed the rationale for the changes, and an
explanation of how the WTP profile will be used.
- Chair: Is there any
objection to these changes. They are being made late in the process, and
will require another WGLC.
i.
No objection to the changes.
- Continue with
presentation on how the WTP profile is used.
- Describes impact on
the current MIB objects: impact to capwapbaseWtpTable – the operator will
only query the WPT properties when the WTP is in the RUN status. Configuration
related functions should be moved to the capwapBaseWtpProfileEntry.
- Describe Impact on the
current CAPWAPBaseWtpStateTable – add a new object, the WTP profile ID
- Describes impact to
capwapBaseWirelessBindingEntry. Change the index valueto the
capwapBaseWtpProfileId, since the creation of the Virtual radio interface
is triggered by the profile table.
- Next steps are to
close all open issues in the next revision within two weeks, including
updating the references for the published RFCs (5415 and 5416). Then go
to WGLC again.
- How many days should
we wait for feedback on the mailing list? First post the solution text on
the list. Update the documents with the final proposed resolution. Then
wait for 2 weeks for any objections. Then issue a WGLC on the 2
documents.
- Chair: Encourage participants
to review the documents.
- Comment, CoKo Musaza: Began looking at CAPWAP in 2006. Is CAPWAP a protocol or a
software? Is there any implementation working on Unix or Linux? Testing
mechanisms? Were they tested using IPv4, IPv6?
i.
Is a protocol, not a set of software.
ii.
Have heard that there are 4-5 implementations in progress for WLAN,
unknown number for EPCGlobal. Best to ask these questions on the list, ask for
folks to respond to the list, or to you directly.
iii.
Not aware of any IPv6 implementations, again ask on the list.
iv.
Dan: The protocol is designed to work with IPv6. Up to vendors and
customers to require IPv6.
- Implementation Information – updates
- Chair: Is anyone willing to give information on
implementations? Ask this question on the list.
- No information contributed.
- Chair asked for any more questions
- Comment, Hassana Moustafa: Is there any second
phase? There was discussion on this at the last meeting.
- Chair: Asked at the last meeting, so far have
been no contributions or expressions of interest.
- Dan: AD perspective is that we need to finish
the MIB, then at the next IETF meeting see if there is interest, or contributions.
Also discuss on the list.
- Meeting adjourned at 4:10pm.