Chairs: Dorothy Gellert, Margaret Wasserman, M Mani

(nobody volunteering for minutes)

Minutes: M Mani

Attendance: ~40

 

DG: Dorothy Gellert, MM: Mani, MW: Margaret, PC: Pat Calhoun, DS: Dorothy Stanley, DR: Dan Romascanu, SK: Scott Kelly, CC: Charles Clancy, DP: David Perkins

DH: Dave Hetherington, MN: Madjid; CL: Clint Chaplin

 

·        Agenda bashing & Chairs’ report (DG)

o       MIB draft: Richard Young volunteered; call for addl volunteers

o       Issue 138 will be the highest priority item and will be discussed for sense of consensus (no objections to agenda)

o       Brief recap of Jan 07 interim and issues discussed thereat.

o       EPCglobal (forum) coming up with RFID binding for CAPWAP

§         MW: the ditor of RFID bindings in good hands of Bob O’Hara.

o       farewell to David Kessens (outgoing Ops AD), welcome to Ron Bonica (incoming Ops AD)

o       Is MIB still important to WG? (no response)

§         Show of hands (3) that we should do MIB

§         How many want to review: 1

§         How many do not want MIB (0)

§         DR: explain how devices will be configured without MIB(someone said CLI). CLI is one option, but not guaranteed.

§         MW: AC config? DR: yes – that’s where MIBs are applied

§         DP: config for AC yes, some WTP-config through AC.

o       DP: started long ago; got mibs from vendors; Richard has some sent; if anyone has MIB input to add to review send to list.

o       WGLC on current doc or close off current issues within next 2 weeks or so that we can issue a refined draft to send to IEEE cross-SDO expert review

o       Can work on issue138 after or before WGLC

o       PC: If we can’t agree on 138 what WGLC will solve is not clear.

o       DG: try to do consensus call over open issues in next 2 weeks.

·        Issue138 discussion (MW)

o       Issues & args were fairly complicated and had discussions w/o consensus in interim

o       Encryption supported both wtp & ac? If both - which is mandatory where?

o       Had discussed theoretically possible combo of discussions. Need agreement on an interoperable mode mandatory (or more than 1 mandatory mode)

§         (MW lists the combinations referring to the slide of table(ppt presentation available online)

o       Need to say what is mandatory and what is optional

§         Have only one mode or both; at both ends or one end. Some don’t make sense (such as optional on both ends)

§         In favor of wtp supporting both modes – ac-central encryption can be favored in high-assurance environments (though not clear exactly what they are) – centralized encryption optional for modes. Both modes are widely aaialble in wtp.

§         For AC to do mandatory central-enc. Some AC’s out there don’t h/w encrypt-decrypt and hence problematic. No definitive answers yet from all chipset/wtp vendors whether they can make enc/dec a pass-thro.

§         Not many have understood the options available or their impact. (referring to the table).option 1 and option 2.

·        option 1 has been put in the spec (05/02). – wtp must support enc/dec; ac can support either.

·        Option 2: WTP-enc. Termination is mandatory. AC-enc is optional.

§         Consensus-chk: “1, 2 or either is OK with me”

§         Hou Zhao: AC must support enc at WTP – table must be modified to say so.

·        In split-mac mode support either; in local-mac mode support wtp-enc.

·        MW: local-mac case is implicit (wtp-mandatory)

§         MW: [not certain about what Scott proposes]

§         PC: not valid for issue.

·        Exact text going under split-mac section is what matters. Local-mac is already handled.

§         DS: shd deal with what overall func needs supported; not where it goes in doc. options covering both as a whole: - agree with scott (lower right of option1 is MUST).

§         DS: whether wtp must support capability for ac-enc. – that is th only question.

§         MW: who thinks WTP and AC require to support both wtp-enc? - MUST

§         MW: how many think it is optional for AC to support WTP-enc? - MUST

§         MW: (how) should we mandate centralAC-enc?

·        Editors report on 05/02 drafts

o       Issue 173:

o       Issue 227:

o       Issue 236:

o       Issue 226:

o       Goals for 06/03: 138, 217, 246-250 to be resolved and relevant proposed-text addressed.

·        Security analysis (Charles Clancy – CC)

o       Should this be an end-end trust model or transitive trust model?

o       We have had very little review on the doc. do review and provide feedback on the list.

·        Call for volunteers to interoperability-bakeoff

o       No response.