[SIPForum-discussion] urgent 491

雨 陈 chen.yu26 at yahoo.com.cn
Wed Mar 19 06:48:54 UTC 2008


Hi, Herve and all,
  Now I am recording the SIP call flowing, but there is a  problem about the response 491 .
  The call flowing as follows:
    UA1                        UA2     TIME
       

               Both way RTP Media
  ç==========================è  
  [254]INVITE[SDP]   
  ç--------------------------     09:18:31:052
  [255]INVITE[SDP]  
  ----------------------------è   09:18:31:065
      [256]100
  ----------------------------è   09:18:31:065
      [257]491
  ----------------------------è   09:18:31:065
      [259]100
  ç-------------------------------------       09:18:31:071
      [260]491
  ç-------------------------------------       09:18:31:071
                      [261] ACK
  ç----------------------------   09:18:31:071
                      [263]ACK
  -----------------------------è   09:18:31:085
            [2639] INVITE [SDP]
  ---------------------------------------è      09:19:01:092
  …………
  As we can see, the session has already set up and the entire message in this flow have the same dialog-ID. Now both UA1 and UA2 re-invite each other just simultaneously, then they get 491 as response from the other one. Well, after 30s, the UA1 send a INVITE again which is NO.2639 with SDP. My question is :
  1.       Since no relevant message during the 30s space time, does the NO.2639 INVITE still belong to the call flow ?
  2.       Is the response 491 like 487, which means the existed session will be terminated?   
   
  In fact, the RFC3621 have just mentioned the UAC should start a timer when it received a 491 response, but not described what will happen, if the timer fires and the UAC don’t attempt the re-INVITE once more.
   
  I look forward to the answer urgently,anything is helpful !
   
  RE
   
  Nora

       
---------------------------------
雅虎邮箱传递新年祝福,个性贺卡送亲朋! 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sipforum.org/pipermail/discussion/attachments/20080319/812c43aa/attachment-0002.html>


More information about the discussion mailing list