CAPWAP Meeting

Chair's Report:
- call for interest in interoperability testing:
- There are so many open issues that it is too early for interoperability testing. 
Perhaps we'd be ready at the next draft.
- We have to consider how many companies are going to implement this protocol. We need
to find out how many companies are willing to participate in interoperability testing.
- THere are about 10 entities that are willing to implement CAPWAP: 3 in hardware and 7 
in software
- The working group can mandate that there be working code before you submit to the IESG.
- The AD's want to close on the 3rd CAPWAP chair by the end of this week.

IEEE Liason Report:
- There were approximately 12000 comments on the initial IEEE 802.11
- IEEE 802.11T and IEEE 802.11y do not impact CAPWAP.
- We are not dependent on any of the existing IEEE 802.11 amendments 
- We should be dependent on IEEE 802.11n

Editor's report:
< presentation >

Discussion on Encryption at the AC versus WTP
- There are cases where data encryption is required end-to-end at the AC.
- There are trade-offs to the different encryption mechanisms

Postpone the MUX issues, security discussion and evaluation to the mailing list.

Discussion on the CAPWAP MUX
- If you encrypt all the way to the AC, you can't tell whether the packet was marked correctly.
- We are going to resolve the problem with QoS and interference,
- We have not clearly defined the boundary between QoS domains: wire, air; air+wire.
- QoS doesn't matter on the data channel between the WTP and the AC.
- Ports are a way to distinguish an endpoint. If the traffic goes to more than one endpoint, 
then different ports should be used.
- There is a valid case to route control traffic to a different endpoint.
- We could make multi-port an optional case.