Hello Daniel,<div><br></div><div>I think your guess about failure_route is correct. Here is the related piece of config. Actually I can not find any append_branch() call in config (which is not originally developed by me).</div>
<div><font class="Apple-style-span" face="'courier new', monospace"><br></font></div><div><div><font class="Apple-style-span" face="'courier new', monospace">failure_route[1] {</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> if (t_was_cancelled()) {</font></div>
<div><font class="Apple-style-span" face="'courier new', monospace"> exit;</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> }</font></div><div><font class="Apple-style-span" face="'courier new', monospace"><br>
</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> if (t_check_status("408|403|404|488|480|415|50[0-4]")) {</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> if (next_gw()) {</font></div>
<div><font class="Apple-style-span" face="'courier new', monospace"> t_on_failure("1");</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> route(6);</font></div>
<div><font class="Apple-style-span" face="'courier new', monospace"> }</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> else</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> {</font></div>
<div><font class="Apple-style-span" face="'courier new', monospace"> route(6);</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> }</font></div>
<div><font class="Apple-style-span" face="'courier new', monospace"> }</font></div><div><font class="Apple-style-span" face="'courier new', monospace">}</font></div><div><font class="Apple-style-span" face="'courier new', monospace"><br>
</font></div><div><font class="Apple-style-span" face="'courier new', monospace">route[6] {</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> if (!t_relay()) {</font></div>
<div><font class="Apple-style-span" face="'courier new', monospace"> sl_reply_error();</font></div><div><font class="Apple-style-span" face="'courier new', monospace"> }</font></div><div>
<font class="Apple-style-span" face="'courier new', monospace"> exit;</font></div><div><font class="Apple-style-span" face="'courier new', monospace">}</font></div><div><br></div></div><div>Cheers,</div>
<div>Max.</div><div><br><div class="gmail_quote">On Tue, Jul 19, 2011 at 3:37 PM, Daniel-Constantin Mierla <span dir="ltr"><<a href="mailto:miconda@gmail.com">miconda@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Hello,<div><div></div><div class="h5"><br>
<br>
On 7/13/11 5:51 PM, Max Doronin wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hello guys,<br>
<br>
I have strange behavior on my Kamailio-1.5.1-notls instance.<br>
<br>
I have LCR and TM modules enabled.For a particular destination I have<br>
4 routes to try. When I call to non-existing number, it tries<br>
- route 1 (404 Not Found)<br>
- route 2 (404 Not Found)<br>
- route 3 (503 Service unavailable)<br>
<br>
And the problem is with the last 503.<br>
<br>
Kamailio ACKs it, immediately drops these 2 lines into syslog<br>
Jul 13 04:14:47 /usr/local/sbin/kamailio[<u></u>23572]:<br>
ERROR:tm:t_forward_nonack: no branch for forwarding<br>
Jul 13 04:14:47 /usr/local/sbin/kamailio[<u></u>23572]: ERROR:tm:w_t_relay:<br>
t_forward_nonack failed<br>
And sends<br>
SIP/2.0 500 Server error occurred (19/SL).<br>
To the call originator.<br>
<br>
30 microseconds later it sends<br>
SIP/2.0 500 Service Unavailable.<br>
to the originator again<br>
<br>
Latter 500 looks like the relayed original "503 Service unavailable".<br>
I think so because the Reason header is the same:<br>
Reason: SIP;cause=503;text="Service<br>
Unavailable";icodetext="<u></u>NoCircuitAvailable";iintcode=<u></u>10034;isubsystem=3.<br>
<br>
Obviously call is terminated.<br>
<br>
I noticed that that last peer rewrites my Via headers like this:<br>
Original:<br>
==<br>
Via: SIP/2.0/UDP 187.45.214.132;branch=<u></u>z9hG4bK3dac.fae9b793.2.<br>
Via: SIP/2.0/UDP<br>
192.168.2.33:5622;received=49.<u></u>49.59.23;branch=z9hG4bK-<u></u>d8754z-a6ba40014f97fb7c-1---<u></u>d8754z-;rport=23832.<br>
==<br>
<br>
Rewritten<br>
==<br>
Via: SIP/2.0/UDP<br>
187.45.214.132;received=187.<u></u>45.214.132;branch=z9hG4bK3dac.<u></u>fae9b793.2,SIP/2.0/UDP<br>
192.168.2.33:5622;received=49.<u></u>49.59.23;branch=z9hG4b<br>
==<br>
<br>
I tried to modify via1_matching parameter (1 -> 0) but no changes.<br>
<br>
My questions are:<br>
- What can be the reason of that 500 with 19/SL<br>
- Can kamailio properly handle this 1 line Via header?<br>
- What can be the reason of that "t_forward_nonack: no branch for forwarding"?<br>
</blockquote>
<br></div></div>
it seems you try to forward one more time, but there is no new branch where to send. I guess there is some issue in the failure_route, like calling t_relay() even if the new branch is not added. If you can paste it here, I can tell more clear if something is wrong there.<br>
<br>
Cheers,<br>
Daniel<br><font color="#888888">
<br>
-- <br>
Daniel-Constantin Mierla -- <a href="http://www.asipto.com" target="_blank">http://www.asipto.com</a><br>
Kamailio Advanced Training, Oct 10-13, Berlin: <a href="http://asipto.com/u/kat" target="_blank">http://asipto.com/u/kat</a><br>
<a href="http://linkedin.com/in/miconda" target="_blank">http://linkedin.com/in/miconda</a> -- <a href="http://twitter.com/miconda" target="_blank">http://twitter.com/miconda</a><br>
<br>
</font></blockquote></div><br></div>