[OpenSIPS-Users] 180 Ringing crashes OpenSIPs

[Jeff Pyle]

I did a fresh update of opensips-1.4 SVN, and a recompile at about 15:00
UTC today.  It still blows up with early media.


-----Original Message-----
From: users-bounces at lists.opensips.org
[mailto:users-bounces at lists.opensips.org] On Behalf Of Bogdan-Andrei
Iancu
Sent: Wednesday, November 05, 2008 3:30 AM
To: Richard Revels
Cc: users at lists.opensips.org
Subject: Re: [OpenSIPS-Users] 180 Ringing crashes OpenSIPs

Hi Richard,

Thanks for the update - the backtrace you get shows an older code (it
was fixed two days ago)

Program received signal SIGSEGV, Segmentation fault.
0x003e3cbf in tmcb_func (t=0xb610ef00, type=2, ps=0x184b94) at
acc_logic.c:259
259  if ( !(early_media && code<200 &&


That line looks now like:
     if ( code<200 && !(early_media &&

Please be sure you have the latest 1.4.2 (SVN 1.4 branch).

Regarards,
Bogdan

Richard Revels wrote:
> I'm getting the same backtrace from core dumps I'm having.  Updating 
> from svn didn't help but turning off the early media accounting seems 
> to be keeping it from happening.
>
> Richard Revels
>
>
> On Oct 30, 2008, at 11:17 AM, Bogdan-Andrei Iancu wrote:
>
>> Hi Jeff,
>>
>> It might be related to a fix I did in the ACC module for early_media 
>> - could you disable early_media accounting to see if it still crashes
?
>>
>> Thanks and regards,
>> Bogdan
>>
>> Jeff Pyle wrote:
>>> Hello,
>>>
>>> We've got a handful of Asterisk boxes that register to today's build

>>> of opensips_1_4.  All works well.  But, when we call from any of 
>>> these Asterisk boxes to one particular one, OpenSIPs crashes.  
>>> Sometimes it relays the 180 Ringing just before crash, sometimes it
crashes first.
>>>
>>> Here's the backtrace:
>>>
>>> Program received signal SIGSEGV, Segmentation fault.
>>> 0x003e3cbf in tmcb_func (t=0xb610ef00, type=2, ps=0x184b94) at
>>> acc_logic.c:259
>>> 259  if ( !(early_media && code<200 &&
>>> (gdb) bt
>>> #0  0x003e3cbf in tmcb_func (t=0xb610ef00, type=2, ps=0x184b94) at
>>> acc_logic.c:259
>>> #1  0x0015c057 in run_trans_callbacks (type=2, trans=0xb610ef00, 
>>> req=0xb610fea8, rpl=0x81cff58, code=180) at t_hooks.c:205
>>> #2  0x0016653c in t_reply_matching (p_msg=0x81cff58,
>>> p_branch=0xbfc737f4) at t_lookup.c:840
>>> #3  0x001669dc in t_check (p_msg=0x81cff58, param_branch=0xbfc737f4)

>>> at t_lookup.c:911
>>> #4  0x00177136 in reply_received (p_msg=0x81cff58) at t_reply.c:1288
>>> #5  0x080651ca in forward_reply (msg=0x81cff58) at forward.c:507
>>> #6  0x08095536 in receive_msg (
>>>    buf=0x817a0a0 "SIP/2.0 180 Ringing\r\nVia: SIP/2.0/UDP
>>>
60.70.82.45;branch=z9hG4bK9027.cfa92ba.0;received=60.70.82.45\r\nVia:
>>> SIP/2.0/UDP
>>> 208.157.201.66:5060;received=208.157.201.66;branch=z9hG4bK3206a4aa;r
>>> port=5060\r\nRecor"...,
>>>
>>> len=697, rcv_info=0xbfc73924) at receive.c:203
>>> #7  0x080d7ef7 in udp_rcv_loop () at udp_server.c:449
>>> #8  0x0806d94e in main (argc=1, argv=0xbfc73b14) at main.c:780
>>>
>>> Here's a packet that made it crash.  Not the time that I got this 
>>> particular backtrace, but it crashed nonetheless:
>>>
>>> U +0.008071 208.157.208.67:5060 -> 60.70.82.45:5060 SIP/2.0 180 
>>> Ringing.
>>> Via: SIP/2.0/UDP
>>> 60.70.82.45;branch=z9hG4bK28b3.c9a41341.0;received=60.70.82.45.
>>> Via: SIP/2.0/UDP
>>>
208.157.201.66:5060;received=208.157.201.66;branch=z9hG4bK5de91597;rport
=5060. 
>>>
>>> Record-Route:
<sip:60.70.82.45;lr=on;ftag=as1a627d69;did=092.a565c3d2>.
>>> From: "Jeff Pyle" <sip:02511 at 208.157.201.66>;tag=as1a627d69.
>>> To: <sip:02061 at sip.fakenet.net>;tag=as70e3a685.
>>> Call-ID: 3974f19662afbc8a7f20983c6a21218a at 208.157.201.66
>>> <mailto:3974f19662afbc8a7f20983c6a21218a at 208.157.201.66>.
>>> CSeq: 103 INVITE.
>>> User-Agent: Asterisk PBX MFLD.
>>> Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, SUBSCRIBE, NOTIFY.
>>> Supported: replaces.
>>> Contact: <sip:02061 at 208.157.208.67>.
>>> Remote-Party-ID: "Office"
>>> <sip:02061 at 208.157.201.66>;party=called;privacy=off;screen=no.
>>>
>>> This same configuration of Asterisk boxes works fine on OpenSER 
>>> 1.3.2.  Still in the process of migration...
>>>
>>> Any thoughts?
>>>
>>>
>>> Thanks,
>>> Jeff
>>>
>>> --------------------------------------------------------------------
>>> ----
>>>
>>>
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.opensips.org
>>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>>
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.opensips.org
>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>


_______________________________________________
Users mailing list
Users at lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users