[Ietf-caldav] iTIP codes for CalDAV Schedule

[Cyrus Daboo]

Hi Arnaud,

--On September 19, 2008 4:49:43 PM +0200 Arnaud Quillaud 
<Arnaud.Quillaud at Sun.COM> wrote:

> When using CalDAV Schedule or iSchedule, for each recipient an iTIP
> status is returned in the response.
>
> Which iTIP status to use is not totally obvious from just reading the
> table at
> http://tools.ietf.org/html/draft-ietf-calsify-2446bis-07#section-3.6
>
> For example, what would be the code for:
> * a recipient in a domain not handled by this server ? 3.8 no authority ?
> * a recipient in a domain handled by this server but the recipient is
> unknown ? 3.7 invalid calendar user, 3.8 no authority ?....
> * a recipient in a domain handled by this server but the originator lacks
> privileges to send iTIP messages to this recipient ?
> * a recipient in a domain handled by this server but the recipient has
> reached his quota ?
>
> Then, assuming that the CalDAV Server is iSchedule enabled what would be
> the code for a recipient in a remote iSchedule domain but the remote
> iSchedule server is currently (temporarily) down ( CalDAV client ---->
> CalDAV Server --x--> remote iSchedule Server ) ?
>
> About that last case, and assuming a scheduling REQUEST, the CalDAV
> Server may elect to return SUCCESS to the client even if the remote
> server is down and try to redeliver asynchronously.
> Now how does the server notify the client that it eventually gave up
> trying ? iTIP seems to sugest that it should put a REPLY in the
> originator's inbox with a REQUEST-STATUS property set to 5.1 service
> unavailable. Should this type of behavior be mentioned in the spec ?
> What if the message to be sent is already a REPLY ?

Yes it would be better to be explicit about what iTIP codes appear in 
various cases. We have already done that to some extent in the new caldav 
scheduling draft which we should be posting later today or over the 
weekend. That includes defining some new 1.xx codes to cover delivery 
status information (including "server is still trying") as well as attendee 
reply status. Those cover some of the bullet points you had, so please 
review and comment when you see the new draft.

-- 
Cyrus Daboo