Hi all,

To supplement the concern that forwarding of non-matched *requests* by the
stateful proxy is wrong, here is the excerpt from rfc3261 regarding the
processing of *responses *after the changes introduced in rfc6026 are
applied:


16.7 Response Processing

   When a response is received by an element, it first tries to locate a
   client transaction (Section 17.1.3) matching the response.   If a
   transaction is found, the response is handed to the client transaction.
*   If none is found, the element MUST NOT forward the response.*


The attention is to the last sentence. But there was no clarification
introduced to clarify the behavior with CANCEL.


Regards,

Brez

On Tue, Apr 19, 2011 at 3:48 PM, Worley, Dale R (Dale) <[email protected]>wrote:

> ________________________________________
> From: [email protected] [
> [email protected]] On Behalf Of Iñaki Baz
> Castillo [[email protected]]
>
> It doesn't check the server transaction state. I expect that a
> stateful proxy must only reply 200 for a CANCEL in case its associated
> INVITE server transaction is in Proceeding state (no final response
> has been sent yet).
>
> So, if a proxy receives a CANCEL for an INVITE transaction whose
> server transaction is in state Completed, Confirmed, Accepted or
> Terminated, should the proxy reply 200? or 481? or nothing?
> _______________________________________________
>
> I don't remember the chapter-and-verse justification, but the practice
> is that a CANCEL for an INVITE whose dialog is confirmed receives a
> 200 response, even though it does not modify the dialog.
> The UAC determines whether the CANCEL terminated
> the INVITE attempt based on the response to the INVITE (200 vs. 487).
>
> Dale
>
> _______________________________________________
> Sip-implementors mailing list
> [email protected]
> https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
>
_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

Reply via email to