Re: [OpenSER-Users] To-tag of 200 for CANCEL - sr-users

26 Mar 2008

Corrected list address.

Andreas Granig wrote:
...
  Hi Bogdan,

 Yes, and I also read some older discussions about this topic like 
 http://www.ietf.org/mail-archive/web/sipping/current/msg03071.html but 
 still I'm quite uneasy regarding the phrasing in RFC3261 saying "that 
 the To tag of the response to the CANCEL and the To tag in the response 
 to the original request SHOULD be the same". It doesn't clearly state to 
 which response it refers - the final 487 for the INVITE or maybe to 
 possible 180/183 (although I understand the consequences and 
 implications of parallel forking etc.).

 I currently don't have a trace available, but is the To-tag in the 200 
 to the CANCEL really the same as the To-tag in 487 to the INVITE, as the 
 RFC suggests, or is the 487 just passed through?

 Andreas

 Bogdan-Andrei Iancu wrote:
  Hi Andreas,

 Have you went through this:

http://sourceforge.net/tracker/index.php?func=detail&aid=1818469&gr…

 There is also a reference to a discussion (on the topic) on sip 
 implementers list.

 Regards,
 Bogdan

 Andreas Granig wrote:
  Hi,

 Actually, someone pointed out a previous discussion about that issue. 
 Interesting read. 
 http://www.openser.org/pipermail/users/2007-October/014049.html

 Andreas

 Andreas Granig wrote:
  Hi,

 One more question regarding the CANCEL processing:

 RFC says in §9.2 regarding the 200 for a CANCEL:

 #+
     This response is
     constructed following the procedures described in Section 8.2.6
     noting that the To tag of the response to the CANCEL and the To tag
     in the response to the original request SHOULD be the same.
 #-

 OpenSER (1.3.1) on the other hand does generate a different To-tag 
 and doesn't use the one from the response of the original request. 
 Why is that? Anything we can do about this?

 Best regards,
 Andreas

 _______________________________________________
 Users mailing list
 Users(a)lists.openser.org
 http://lists.openser.org/cgi-bin/mailman/listinfo/users 