[AFS3-std] rxgk (protocol) error codes

[Benjamin Kaduk]

On Tue, 15 Jan 2013, Michael Meffie wrote:

> On Fri, 11 Jan 2013 12:38:47 -0500
> Benjamin Kaduk <kaduk@MIT.EDU> wrote:
>
>> On Tue, 8 Jan 2013, Benjamin Kaduk wrote:
>>
>>>
>>> I still don't have a better idea than a new section.
>>
>> I pushed ac793f588b84f97038b433d61e3b43e9279fc547 to github with a new
>> section describing the use of the error codes, and removing the
>> COMPOUND_IDENTITY and PRINTED codes.
>> There have been a few other syntax/formatting commits in there, too,
>> but those should be uninteresting.
>
> Hello Ben,
>
> This looks fine to me. Hearing no objections, do you have any other updates
> planned before making a draft 03 of the rxgk i-d?  Otherwise, do you have a
> timeline in mind for the draft 03? I would like to make call for consesus.

[We discussed this on OpenAFS Jabber today]
I think Simon plans to submit a new I-D tomorrow, but there remain a few 
tweaks that will be needed before we finalize the document.
In particular, placing limits on the length of variable-length arrays on 
the wire and distinguishing which errors should be sent as aborts versus 
in-band codes come to mind.

That said, I don't know of any major issues outstanding with this 
document, and it may be time to move our focus to the rxgk-afs document. 
There is some chance that discussion on that latter document could cause 
us to want to change portions of the rxgk document.

-Ben