[sr-dev] Segmentation fault if using multi_leg_info

[Daniel-Constantin Mierla]

Please keep the mailing list cc-ed. Private emails from the lists are 
the last checked and usually dropped after first notice.

Are you running the 3.1.0 from tarball? There was a fix related to 
cancel processing with new addition of reason header support. Although 
it does not look to be right there, might be related, since it is 
processing of a cancelled transaction.

Could you try with latest version of GIT branch 3.1? If you use debs, 
then you can take the nightly builds. Anyhow, for any eventuality, here 
is a guide:
http://www.kamailio.org/dokuwiki/doku.php/install:kamailio-3.1.x-from-git

Cheers,
Daniel


On 10/26/10 6:08 PM, Mino Haluz wrote:
> Hm maybe it is something with radiusclient ...
>
> (gdb) bt
> #0  0xb70dbf73 in rc_avpair_assign () from /usr/lib/libradiusclient-ng.so.2
> #1  0xb70dc059 in rc_avpair_new () from /usr/lib/libradiusclient-ng.so.2
> #2  0xb70dc1ae in rc_avpair_add () from /usr/lib/libradiusclient-ng.so.2
> #3  0xb70ec745 in acc_rad_request (req=0xb7292fc0) at acc.c:585
> #4  0xb70ef9d0 in on_missed (t=<value optimized out>, req=0xb7292fc0,
>      reply=0x8353da8, code=487) at acc_logic.c:328
> #5  0xb7239067 in run_trans_callbacks_internal (cb_lst=0xb5161b70, type=64,
>      trans=0xb5161b30, params=0xbf896cbc) at t_hooks.c:290
> #6  0xb72392fa in run_trans_callbacks (type=64, trans=0xb5161b30,
>      req=0xb7292fc0, rpl=0x8353da8, code=487) at t_hooks.c:317
> #7  0xb725df6e in run_failure_handlers (t=0xb5161b30, rpl=0x8353da8, code=487,
>      extra_flags=64) at t_reply.c:948
> #8  0xb725f28c in t_should_relay_response (Trans=0xb5161b30,
>      new_code=<value optimized out>, branch=0, should_store=0xbf896ec8,
>      should_relay=0xbf896ecc, cancel_data=0xbf8970a0, reply=0x8353da8)
>      at t_reply.c:1212
> #9  0xb7262bf1 in relay_reply (t=0xb5161b30, p_msg=0x8353da8, branch=0,
>      msg_status=487, cancel_data=0xbf8970a0, do_put_on_wait=1) at t_reply.c:1615
> #10 0xb7265c15 in reply_received (p_msg=0x8353da8) at t_reply.c:2284
> #11 0x080929cc in forward_reply (msg=0x8353da8) at forward.c:768
> #12 0x080d57d5 in receive_msg (
>      buf=0x827e740 "SIP/2.0 487 Request Cancelled\r\nVia: SIP/2.0/UDP
> 1.1.1.1;branch=z9hG4bK01a2.35a963e6.0,SIP/2.0/UDP
> 192.168.10.4:5070;received=2.2.2.2;rport=1081;branch=z9hG4bK1902564\r\nFrom:
> \"zxc\"<sip:004"...,
>      len=492, rcv_info=<value optimized out>) at receive.c:266
> #13 0x0815b532 in udp_rcv_loop () at udp_server.c:532
> #14 0x080a2917 in main_loop () at main.c:1554
> #15 0x080a5142 in main (argc=1, argv=0xbf8975b4) at main.c:2398
>
> On Tue, Oct 26, 2010 at 5:43 PM, Daniel-Constantin Mierla
> <miconda at gmail.com>  wrote:
>> Hello,
>>
>> On 10/26/10 3:18 PM, Mino Haluz wrote:
>>
>> Hi,
>>
>> there seems to be a bug when using multi_leg_info (version: kamailio 3.1.0
>> (i386/linux) bf8b8d):
>>
>> modparam("acc", "multi_leg_info", "Src-Leg=$avp(i:901);Dst-Leg=$avp(i:902)")
>>
>> When the call ends, kamailio suddenly segfaults:
>>
>> WARNING: no fork mode
>> Segmentation fault (core dumped)
>>
>> you got a core file, locate it in / or in the working directory.
>>
>> The do:
>>
>> gdb /path/to/kamailio /path/to/core
>>
>> bt
>>
>> Send the output here.
>>
>> Cheers,
>> Daniel
>>
>> If you do not know to reproduce the error, please provide a howto that can
>> lead me to retrieve another useful information (gdb..). Thank you very much.
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> sr-dev mailing list
>> sr-dev at lists.sip-router.org
>> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
>>
>> --
>> Daniel-Constantin Mierla
>> http://www.asipto.com

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