Capwap Mailing List: Latest Status of CAPWAP Open Issues

[Pat Calhoun (pacalhou) (pcalhoun]

All,

Given the numerous messages dealing with all of issues raised by the
IESG, and the few additional ones discovered while fixing them, here is
an update on all of the issues, and what I'm waiting for in order to
move forward.

Issue 61: T20. The original request to change the enum lists to start at
zero was not acceptable to some due to the fact that it would be
incompatible with version -10 of the draft. Based on an agreement with
Dan, I have simply marked zero (0) as reserved. I am waiting for
confirmation that this is acceptable

Issue 103: Invalid Length Information Message Element. This was raised
on the list, and pointed out a valid issue that the length field for
this message element was incorrect. It needed to be 20, not 18. I am
waiting for confirmation that this is acceptable (although I can't see
why anyone would argue against it).

Issue 102: Join to Image Data state transition issue on the AC. This is
another issue raised on the list, and required that some new text be
added to this state transition description on the AC. While there are a
number of possible ways to fix this issue, I took the simplest and more
direct route to minimize the amount of text. I believe that anyone
implementing -10 would have assumed the change I made.

Issue Issue 53: T12. Dan raised an issue that he finds the way the
message element diagram was constructed was confusing. I have proposed
new text, and am waiting for confirmation that he is comfortable with
this change. If we opt to go down this route, I would also need to
change the WTP Board Data and WTP Descriptor message elements. This does
not introduce an compatibility issue with -10, it is simply an editorial
change that helps make the text clearer.

Issue 65: T24. The IESG raised an issue that they believed that the
Image Data and Image Identifier message elements should have a length
field. I made this change, but Margaret does not believe this change is
really necessary, and since it introduces an incompatibility issue with
-10, she would prefer that I back the change out. I am waiting for other
folks to provide their opinions. I agree the length is not necessary.

Issue 42: T1. This issue stated that any variable length field, or
length field, must have a stated maximum value. Many instances of this
was addressed via other technical comments from the IESG, but I went
through the draft and found a few more, which I combined into this
issue. I am waiting for the OK to close this issue.

Issue 99: CAPWAP Timers not referencing timers properly. This is a minor
change, but while fixing another issue, I found that the CAPWAP timers
message element was not referring to the timers in section 4.7. This fix
simply added the references.

Issue 73: T32. The IESG questioned whether the number of bits in a field
of the WTP Operational Statistics is sufficient. Reading the message
element, I couldn't really understand what this was trying to do, so I
asked the LWAPP developers, and this was an idea that never turned into
anything, and do not have any recommendations on how to make the
language clearer. Since I have no idea what someone would use this for,
I cannot really fix this message element - and would propose deleting
it. I am ok leaving it in, but I'd like to see the text cleaned up, and
would ask that whomever wants me to leave it in, provide text.

Issue 74: T33. This one was a question on whether the
statistics/counters in both the WTP Radio Statistics and the WTP Reboot
Statistics were intended to wrap around or reset. I modified the text to
state it is the former. I am waiting for confirmation that this change
is ok.

Issue 63: T22. The IESG asked why the length field was set to a specific
value. There was obviously some confusion about what information was
included, and I have provided a response on what this meant. I am
waiting to hear from Dan on how to proceed. I do not believe a fix is
required.

Issue 41: CAPWAP DTLS Timer Clarification. This is a small change that
provides some clarification on the use of timers. This does not
introduce an incompatibility issue with -10, as I suspect most people
ended up figuring out (the hard way) that the timers should have behaved
based on my change.

Issue 101: Error in message element figure in 4.6.32. This is a small
editorial change to a broken message element diagram. No incompatibility
with -10.

Issue 100: Editorial issue in 4.6.15. This is a small editorial change
to add a separator between an enumerated value and its description. No
technical change.

Issue 89: E8. I have rejected this issue since the creator of the
comment had simply not understood that the use of alpha/numeral/special
symbols was only for the AC, not the WTP. I am waiting for Dan's OK.

Issue 82: E1. I have changed the text to address the editorial issue
raised, and am waiting for Dan's OK.

Issue 86: E5. I have changed the text to address the editorial issue
raised, and am waiting for Dan's OK.

Issue 85: E4. I have changed the text to address the editorial issue
raised, and am waiting for Dan's OK.

Issue 97: E16. The IESG pointed out that some timers in section 4.7 were
not really timers, and needed to move to another place in the spec. Only
one of the sections they pointed to needed to move, and I have. I have
rejected the rest of the request. I am waiting for Dan's OK.

PatC