<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Hello,<br>
<br>
On 2/8/10 11:09 PM, Dmitri wrote:
<blockquote cite="mid:4B708B97.8020200@festart.ee" type="cite">
  <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
Hello,<br>
The PC has private address on eth0 interface and public on ppp0<br>
default route looks like:<br>
default dev ppp0  scope link<br>
  <br>
Tomorrow I'll try to play with ngrep.<br>
</blockquote>
<br>
it is looping on the IP you can see in the Record-Route headers:<br>
<pre wrap="">SUBSCRIBE <a class="moz-txt-link-freetext"
 href="sip:dima@voice.djuk.ee;transport=UDP">sip:dima@voice.djuk.ee;transport=UDP</a> SIP/2.0\r\nRecord-Route: <a
 class="moz-txt-link-rfc2396E" href="sip:90.190.13.159;lr=on">&lt;sip:90.190.13.159;lr=on&gt;</a>\r\nRecord-Route: <a
 class="moz-txt-link-rfc2396E" href="sip:90.190.13.159;lr=on">&lt;sip:90.190.13.159;lr=on&gt;</a>\r\nRecord-Route: <a
 class="moz-txt-link-rfc2396E" href="sip:90.190.13.159;lr=on">&lt;sip:90.190.13.159;lr=on&gt;</a>\r\nRecord-Route: &lt;<a
 class="moz-txt-link-freetext" href="sip:90">sip:90</a>"...

Is 90.190... a local IP?

Cheers,
Daniel
</pre>
<br>
<blockquote cite="mid:4B708B97.8020200@festart.ee" type="cite">BR,<br>
Dmitri<br>
  <br>
08.02.2010 23:59, Daniel-Constantin Mierla пишет:
  <blockquote cite="mid:4B708956.7090806@gmail.com" type="cite">
    <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
Hello,<br>
    <br>
On 2/8/10 10:46 PM, Dmitri wrote:
    <blockquote cite="mid:4B708628.7090601@festart.ee" type="cite">
      <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
Hello,<br>
No, it is not local, it is reacheable over internet, real dns domain.<br>
    </blockquote>
why is it looping then? I see many Record Route headers with same IP,
should be one since it is no transport layer gatewaying.<br>
    <br>
You can use ngrep to watch the network and see what happens with
SUPBSCRIBE (watch as well looback interface).<br>
    <br>
    <blockquote cite="mid:4B708628.7090601@festart.ee" type="cite">Waiting
for
your
instructions.<br>
    </blockquote>
    <br>
Going to do a commit to make it easily via makefile.<br>
    <br>
Cheers,<br>
Daniel<br>
    <br>
    <blockquote cite="mid:4B708628.7090601@festart.ee" type="cite"><br>
BR,<br>
Dmitri.<br>
      <br>
08.02.2010 23:27, Daniel-Constantin Mierla пишет:
      <blockquote cite="mid:4B7081A4.4040909@gmail.com" type="cite">
        <meta content="text/html; charset=UTF-8"
 http-equiv="Content-Type">
Hello,<br>
        <br>
might be a buffer overflow somewhere else. I will send you some
instructions to compile with memory debug one.<br>
        <br>
Meanwhile, this seems to be a particular case (has to be found and
fixed anyhow), the SUBSCRIBE is looping. Is voice.djuk.ee a local
domain (matching the 'myself' condition)? If is local domain and not in
dns then add:<br>
        <br>
alias=voice.djuk.ee<br>
        <br>
Cheers,<br>
Daniel<br>
        <br>
        <br>
On 2/8/10 10:14 PM, Dmitri wrote:
        <blockquote cite="mid:4B707E99.9030701@festart.ee" type="cite">
          <meta content="text/html; charset=UTF-8"
 http-equiv="Content-Type">
Hello,<br>
          <br>
Updated gdb output in attached file.<br>
          <br>
BR,<br>
Dmitri<br>
          <br>
08.02.2010 22:51, Daniel-Constantin Mierla пишет:
          <blockquote cite="mid:4B707952.5000207@gmail.com" type="cite">
            <meta content="text/html; charset=UTF-8"
 http-equiv="Content-Type">
Hello,<br>
            <br>
once you are in gdb, run 'bt' to get the backtrace. That is very useful
in the first step to see what functions were executed before the crash.<br>
            <br>
            <br>
Thanks,<br>
Daniel<br>
            <br>
On 2/8/10 9:23 PM, Dmitri wrote:
            <blockquote cite="mid:4B7072CF.8080502@festart.ee"
 type="cite">Hello
Klaus, <br>
              <br>
Thank you for suggestion. Now I got core, and output of gdb you may see
in attached file. <br>
Core was about 40M of size. <br>
I have to say that i never used kamailio before and this crash is my
first experience. <br>
              <br>
BR, <br>
Dmitri <br>
              <br>
08.02.2010 21:37, Klaus Darilion пишет: <br>
              <blockquote type="cite"><br>
                <br>
Am 08.02.2010 20:00, schrieb Dmitri: <br>
                <blockquote type="cite">Hello, <br>
1. Yes, there is at least 2 times more space than core dump needs. <br>
                </blockquote>
                <br>
                <br>
                <br>
# Try specifying things manually, as user root: <br>
echo "/tmp/core.%e.sig%s.%p" &gt; /proc/sys/kernel/core_pattern <br>
                <br>
# verify configuration changes <br>
cat /proc/sys/kernel/core_pattern <br>
                <br>
# set ulimit <br>
ulimit -c unlimited <br>
                <br>
# start kamailio manually <br>
kamailio <br>
                <br>
now kamailio should be running as user root and should be able to core
dump into /tmp <br>
                <br>
                <blockquote type="cite">2. I could find nothing,
unfortunately... <br>
I can only see that problem is related to tm.so library: <br>
kamailio kernel: [ 1051.432646] kamailio[2678]: segfault <br>
at 12 ip 00007fa8b592ae43 sp 00007fff48ff9f40 error 4 in <br>
tm.so[7fa8b5917000+72000] <br>
                </blockquote>
                <br>
Daniel, I can't remember. Is it necessary to turn off some compiler
optimizations in Makefile.defs to get a useful backtrace? (or was that
in Asterisk?) <br>
                <br>
regards <br>
klaus <br>
                <br>
                <br>
                <blockquote type="cite"><br>
BR, <br>
Dmitri <br>
                  <br>
                  <blockquote type="cite">
                    <blockquote type="cite">I don't know it this
applies
to
you
but
also check the following: <br>
                      <br>
1. You have enough space in the partition where $COREDIR is created <br>
2. Check that the user that runs kamailio (usually openser) has read <br>
write access on that directory <br>
                      <br>
Also if you grep for core in the /var/log/ directory I believe that <br>
there might be some lines either in dmesg or in messages that tell <br>
you if a core was created or not. <br>
                      <br>
Hope what I said is not redundant and it helps. <br>
                      <br>
Marius <br>
                    </blockquote>
                    <br>
                  </blockquote>
                  <br>
                  <br>
_______________________________________________ <br>
Kamailio (OpenSER) - Users mailing list <br>
                  <a moz-do-not-send="true"
 class="moz-txt-link-abbreviated" href="mailto:Users@lists.kamailio.org">Users@lists.kamailio.org</a>
                  <br>
                  <a moz-do-not-send="true"
 class="moz-txt-link-freetext"
 href="http://lists.kamailio.org/cgi-bin/mailman/listinfo/users">http://lists.kamailio.org/cgi-bin/mailman/listinfo/users</a>
                  <br>
                  <a moz-do-not-send="true"
 class="moz-txt-link-freetext"
 href="http://lists.openser-project.org/cgi-bin/mailman/listinfo/users">http://lists.openser-project.org/cgi-bin/mailman/listinfo/users</a>
                  <br>
                </blockquote>
              </blockquote>
              <br>
              <pre wrap=""><fieldset class="mimeAttachmentHeader"></fieldset>
_______________________________________________
Kamailio (OpenSER) - Users mailing list
<a moz-do-not-send="true" class="moz-txt-link-abbreviated"
 href="mailto:Users@lists.kamailio.org">Users@lists.kamailio.org</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
 href="http://lists.kamailio.org/cgi-bin/mailman/listinfo/users">http://lists.kamailio.org/cgi-bin/mailman/listinfo/users</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
 href="http://lists.openser-project.org/cgi-bin/mailman/listinfo/users">http://lists.openser-project.org/cgi-bin/mailman/listinfo/users</a></pre>
            </blockquote>
            <br>
            <pre class="moz-signature" cols="72">-- 
Daniel-Constantin Mierla
eLearning class for Kamailio 3.0.0
Starting Feb 8, 2010
* <a moz-do-not-send="true" class="moz-txt-link-freetext"
 href="http://www.asipto.com/">http://www.asipto.com/</a>
  </pre>
          </blockquote>
          <br>
        </blockquote>
        <br>
        <pre class="moz-signature" cols="72">-- 
Daniel-Constantin Mierla
eLearning class for Kamailio 3.0.0
Starting Feb 8, 2010
* <a moz-do-not-send="true" class="moz-txt-link-freetext"
 href="http://www.asipto.com/">http://www.asipto.com/</a>
  </pre>
      </blockquote>
      <br>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
Daniel-Constantin Mierla
eLearning class for Kamailio 3.0.0
Starting Feb 8, 2010
* <a moz-do-not-send="true" class="moz-txt-link-freetext"
 href="http://www.asipto.com/">http://www.asipto.com/</a>
  </pre>
  </blockquote>
  <br>
</blockquote>
<br>
<pre class="moz-signature" cols="72">-- 
Daniel-Constantin Mierla
eLearning class for Kamailio 3.0.0
Starting Feb 8, 2010
* <a class="moz-txt-link-freetext" href="http://www.asipto.com/">http://www.asipto.com/</a>
</pre>
</body>
</html>