Daniel,

Thank you for the reply back. Unfortunately for the diagnosis with version 1.5.4-notls we had already upgraded to version 1.5.5-notls and the core files had been deleted before receiving your reply. I can state that version 1.5.5-notls has not exhibited the issue that we experienced prior to upgrading the software.

Thank You Stagg

On 12/3/10 5:57 AM, Daniel-Constantin Mierla wrote:

Hello,

probably is something triggered by a new kind of SIP message which was not encountered before, since there were no changes in that code. Can you send the output of 'bt full' in gdb?

Also, you should have the content of SIP request/reply available in core file, in one of gdb frames. Please send it as well.

Cheers, Daniel

On 12/2/10 2:27 AM, Stagg Shelton wrote:

...

Kamailio 1.5.4 segmentation faulted today. I was able to determine with gdb that the error occurred at mediaproxy.c line 1104. Line 1104 is a call to memcpy

Core was generated by `/sbin/kamailio -m 512'. Program terminated with signal 11, Segmentation fault. #0 0x00007f1469db6bb5 in replace_element (msg=0x77e470, old_element=<value optimized out>, new_element=0x7fffbd553210) at mediaproxy.c:1104 1104 memcpy(buf, new_element->s, new_element->len);

1.5.4 is new to our production environment, we were previously running 1.5.2 notls (also no mediaproxy) for more than a year without issue. 1.5.4 notls with mediaproxy has been running in our lab and preproduction QA environments since March of this year without any real segmentation faults.

Does anyone have any ideas of what may have caused this, or suggestions to get around this. I do see that 1.5.5 is available, although there does not appear to have been too many changes to mediaproxy.c

Thank you in advance

Stagg

---

SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list sr-users@lists.sip-router.org http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users

There isn't any useful information in the most recent core file backtraces. I'm rebuilding 1.5.5 with the mode=debug option so that more information will be available next time (if there is a next time).

Thanks Stagg

On 12/3/10 5:57 AM, Daniel-Constantin Mierla wrote:

Hello,

probably is something triggered by a new kind of SIP message which was not encountered before, since there were no changes in that code. Can you send the output of 'bt full' in gdb?

Also, you should have the content of SIP request/reply available in core file, in one of gdb frames. Please send it as well.

Cheers, Daniel

On 12/2/10 2:27 AM, Stagg Shelton wrote:

...

Kamailio 1.5.4 segmentation faulted today. I was able to determine with gdb that the error occurred at mediaproxy.c line 1104. Line 1104 is a call to memcpy

Core was generated by `/sbin/kamailio -m 512'. Program terminated with signal 11, Segmentation fault. #0 0x00007f1469db6bb5 in replace_element (msg=0x77e470, old_element=<value optimized out>, new_element=0x7fffbd553210) at mediaproxy.c:1104 1104 memcpy(buf, new_element->s, new_element->len);

1.5.4 is new to our production environment, we were previously running 1.5.2 notls (also no mediaproxy) for more than a year without issue. 1.5.4 notls with mediaproxy has been running in our lab and preproduction QA environments since March of this year without any real segmentation faults.

Does anyone have any ideas of what may have caused this, or suggestions to get around this. I do see that 1.5.5 is available, although there does not appear to have been too many changes to mediaproxy.c

Thank you in advance

Stagg

---

SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list sr-users@lists.sip-router.org http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users