Re: reinvite response's dialog is not recognized

Becky,

Cool! No problem about the trouble. Indeed there were some issues with 
dialog identity that I had to work out. It was not perfect when you 
reported the problem :-)  Thanks for helping me to work out the problems 
as they arise. I do appreciate your switching over to Jain-SIP 1.2. 
Incidentally, you may want to cvs update -- especially  if you are doing 
anything with subscribe and Notify. Been a lot of hacking there lately 
to take care of race conditions (subscribe response racing with notify).

Anybody (Jeroen?)  know why subscribe does not use a three way handshake 
like Invite ? Could solve a lot of problems with race conditions.

We are only a test or two away from TCK completion.

Regards,

Ranga

Becky McElroy wrote:

> Hi Ranga,
>
> Your CVS update is working just fine - the problem was with my code. 
> The re-invite OK response was actually never going out because I 
> didn't include a Contact header in the response, and I wasn't checking 
> the return code from my send response method so the exception thrown 
> by the stack was going in a black hole. Haste really does make waste.
>
> Thanks alot for your help, and sorry to trouble you.
> Becky
>
> M. Ranganathan wrote:
>
>> Becky,
>>
>> Can you cvs update again and try it. First, the presence proxy is way 
>> way out of date. I will fix it right after I complete the 
>> JAIN-SIP-1.2  TCK and am happy with the RI ( still a few kinks left). 
>> I shall roll around to the presence proxy in two weeks and upgrade 
>> that to JAIN-SIP 1.2 ( should eat my own cooking  I suppose).
>>
>> The reason for multiple route headers is that some element (SipUnit 
>> perhaps)  is using Strict route ( i.e. not setting lr in the route 
>> header of the request URI). Can you check if that is the case.
>>
>> Regards,
>>
>> Ranga
>> Becky McElroy wrote:
>>
>>> Hello:
>>>
>>> I just tested my new REINVITE methods and saw the same Exception 
>>> encountered below also. I updated from CVS (last night) and no 
>>> longer got the exception, but there are a couple of issues still, or 
>>> I'm doing something wrong.
>>>
>>> 1. The setup is: user a (amit) establishes a call with user b 
>>> (becky) - INVITE->, <-OK, ACK->
>>> It is through the sip-nist presence proxy.
>>> Then amit sends reinvite, becky receives it and responds OK. Amit 
>>> receives just TRYING (two of them - you can see them at the end of 
>>> the sipunit.log) but never receives the OK. The second TRYING 
>>> received in the sipunit.log doesn't have the To-tag (and also comes 
>>> much later, by 4+ seconds). Interestingly, neither of the TRYING 
>>> messages has a To-tag if you look in the endpoint-stacktrace.log....!
>>>
>>> 2. After calling dialog.createRequest(Request.INVITE) and adding a 
>>> few headers, SipCall.sendReinvite() prints out the request before 
>>> sending and it has some extra 'Route' headers (the target of the 
>>> re-invite has been added twice to the Route list). You can see it in 
>>> the sipunit.log - search for: "We have created this RE-INVITE....".  
>>> Is it something I've done to cause that to happen?
>>>
>>> Attached is a zip containing:
>>>     sipunit.log (output of the test method execution)
>>>     endpoint-stacktrace.log (stack trace for the stack used by 
>>> endpoints a and b)
>>>     test-method.txt (excerpt containing SipUnit test setUp(), 
>>> tearDown(), and testReinvite() methods)
>>>     SipCall.java (has the methods sendReinvite(), 
>>> waitReinviteResponse(), sendReinviteOkAck() for the REINVITE sending 
>>> side and for the receive side: waitForReinvite(), 
>>> respondToReinvite() - these are called by the test method)
>>>     SipPhone.java (the parent of SipCall)
>>>     SipSession.java (the base class of SipPhone, the only method of 
>>> interest used here is sendRequestWithTransaction()).
>>>
>>>
>>> Any help or tips appreciated!
>>> Thanks alot.
>>> Becky
>>>
>>> M. Rangnathan wrote:
>>>
>>>> Somebody set a to tag on your 100 respose so you see this rather 
>>>> descriptive error message :-) . In any case cvs update and you will 
>>>> be OK
>>>>
>>>> Ranga
>>>>
>>>> Hagai Sela (TA) wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> I have the following 3pcc flow: my nist-sip based server sends an 
>>>>> invite to endpoint 1, which is OK’d and ACK’d, then sends an 
>>>>> invite to endpoint 2, which is also OK’d and ACK’d, and then it 
>>>>> sends a re-invite to endpoint 1. When I receive the 200 OK 
>>>>> response for the reinvite I have a different dialog than the one I 
>>>>> had on the first request, and I also get this exception:
>>>>>
>>>>> Exception in thread "UDPMessageChannelThread" 
>>>>> java.lang.RuntimeException: bad things happened 100 in 
>>>>> setLastResponse
>>>>>
>>>>> at 
>>>>> gov.nist.javax.sip.stack.SIPDialog.setLastResponse(SIPDialog.java:1986) 
>>>>>
>>>>>
>>>>> at gov.nist.javax.sip.stack.SIPDialog.<init>(SIPDialog.java:806)
>>>>>
>>>>> at 
>>>>> gov.nist.javax.sip.stack.SIPClientTransaction.processResponse(SIPClientTransaction.java:1184) 
>>>>>
>>>>>
>>>>> at 
>>>>> gov.nist.javax.sip.stack.UDPMessageChannel.run(UDPMessageChannel.java:439) 
>>>>>
>>>>>
>>>>> at java.lang.Thread.run(Thread.java:595)
>>>>>
>>>>> logs are attached.
>>>>>
>>>>> Hagai.
>>>>>
>>>>> ------------------------------------------------------------------------ 
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> nist-sip mailing list
>>>>> [email protected]
>>>>> http://www-x.antd.nist.gov/mailman/listinfo/nist-sip
>>>>>  
>>>>>
>>>>
>>>>
>>>
>>
>>
>
>


-- 
M. Ranganathan 

Advanced Networking Technologies Division,
National Institute of Standards and Technology (NIST),
100 Bureau Drive, Stop 8920, Gaithersburg, MD 20899. 
tel:301 975 3664 , fax:301 590 0932 http://w3.antd.nist.gov/
Advanced Networking Technologies For the People!