I just recommended in antoher post to increase the timer and do an ngrep
trace to be certain what happens. If the CANCEL is not processed by SER,
the timer will hit, and if the timer is low, it may hit before the CANCEL
reaches SER.
There was some code work on handling of 40x a while back in head (serdev, I
think). You may want to look back to that thread if it's related.
g-)
----- Original Message -----
From: "Antonio Rabena" <antonio(a)lgatelecom.net>
To: "serusers" <serusers(a)lists.iptel.org>
Sent: Tuesday, February 14, 2006 10:16 AM
Subject: Re: Re: Re: [Serusers] a question about "cancel" in
forwardingscenario
Hi,
Im also experiencing the same problem. The only diff. is ua2 and ua3 are
using same sip account so when ua1 sends invite to that sip account, both
ua2 and ua3 ring, and when ua1 cancels the call, it receives 408 Request
Timeout".
At 04:41 PM 2/14/2006, you wrote:
Dear Greger V. Teigre,
I've tried t_on_failure("2").
failure_route[2] {
#when caller cancel, terminate the call
if (t_check_status("487")) {
log(1,"cancel received\n");
t_reply("487", "Request Terminated");
break;
}
}
There are similar lines in failure_route[1]. When ua2 is ringing and ua1
cancel the call, the value of "t_check_status("487")" in
failure_route[1]
will be true. However, When ua3 is ringing and ua1 cancel the call, the
value of "t_check_status("487")" in failure_route[2] will be false.
This
time, t_check_status() returns "408". I don't know why SER generates
"408"
under this circumstance. My ser's version is 0.9.3 and I just run
update_from_cvs. And I found similar description here:
http://bugs.sip-router.org/browse/SER-63
Is it an unsolved problem?
>Do some logging, what happens in the failure route? To me, it looks like
>your config appends a branch. BTW, run update_from_cvs in your SER
>source
>dir top make sure you get the latest bug fixes!
>g-)
>
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers