On 7/11/11 6:42 AM, Brett Tate wrote:
>> Is it OK to send a subsequent NOTIFY in a SUBSCRIBE initiated
>> dialog, before receiving a successful response for the earlier
>> NOTIFY request?
>
> Yes; unless specifically restricted by the event package,
> draft-ietf-sipcore-event-rate-control, or similar RFC.
>
> However doing so can complicate some things such as retargeting.
Building on what Brett has said...
If you send a subsequent notify before having received the response to
the prior one, the subsequent one may be received *before* the earlier
one. Or the earlier one may never be received at all. If the new one is
received *before* the earlier one, then the earlier one will be rejected
with a 500 response, and so not processed.
You should consider the implications of that when deciding whether to
wait before sending or not. This is primarily a function of the design
of the event package. If the new event renders the earlier one moot,
then its fine to do this. If all events must be processed in order, the
its not fine.
Thanks,
Paul
>> Does the specification mention otherwise?
>
> The following indicates how to respond when such behavior leads to the
> requests being delivered out of order.
>
> RFC 3261 section 12.2.2: "If the remote sequence number was not empty, but
> the sequence number of the request is lower than the remote sequence number,
> the request is out of order and MUST be rejected with a 500 (Server Internal
> Error) response."
>
>
>
> _______________________________________________
> Sip-implementors mailing list
> Sip-implementors@lists.cs.columbia.edu
> https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
>
_______________________________________________
Sip-implementors mailing list
Sip-implementors@lists.cs.columbia.edu
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
