[Kamailio-Devel] 1.4 crash (maybe xlmrpc related)

[Daniel-Constantin Mierla]

On 12/12/08 20:36, Juha Heinanen wrote:
> i'm sending PUBLISH request via xmlrpc mi interface to kamailio and i get a
> crash:
>
> python program that sends the xmlrpc request reports:
>
> Dec 12 20:19:29 lohi sems[9345]: Error: system error error: type <class 'xmlrpclib.ProtocolError'> 
>
> and kamailio reports:
>
> Dec 12 20:19:29 lohi /usr/sbin/kamailio[10931]: CRITICAL:core:receive_fd: EOF on 16 
> Dec 12 20:19:31 lohi /usr/sbin/kamailio[10850]: INFO:core:handle_sigs: child process 10857 exited by a signal 11 
>
> gdb does not tell much:
>
> (gdb) where
> #0  0xb7fb0424 in __kernel_vsyscall ()
> #1  0xb7cb6640 in raise () from /lib/i686/cmov/libc.so.6
> #2  0xb7cb8018 in abort () from /lib/i686/cmov/libc.so.6
> #3  0x08067496 in sig_alarm_abort (signo=14) at main.c:422
> #4  <signal handler called>
> #5  fm_status (qm=0xb3629000) at mem/f_malloc.c:512
> #6  0x08067126 in cleanup (show_status=1) at main.c:356
> #7  0x08067b92 in handle_sigs () at main.c:523
> #8  0x0806ab5e in main (argc=15, argv=0xbffcccc4) at main.c:865
>
> looks like a child has died when accessing an unassigned memory
> location.
>
> any ideas on how i could debug this further and how to find where the
> problem is?
>   
the core seems from main process, perhaps overwrites the one from the 
process that crasshed. I think on some OS, you can specify to dump the 
core in core.pid files.

Do you know if that the crashed process was the MI listener?

Cheers,
Daniel

-- 
Daniel-Constantin Mierla
http://www.asipto.com