Summary:	ASTERISK-06508: Asterisk 1.2.5 crashed when using call queues.
Reporter:	Vladimir S. Blazhkun (vovan)	Labels:
Date Opened:	2006-03-09 05:44:05.000-0600	Date Closed:	2006-05-04 16:23:41
Priority:	Critical	Regression?	No
Status:	Closed/Complete	Components:	Applications/app_queue
Versions:		Frequency of Occurrence
Related Issues:
Environment:		Attachments:	( 0) crash-10.log ( 1) crash-11.log ( 2) crash-12.log ( 3) crash-9.log
Description:	Hardware information: HP DL380 G3, 1GB RAM (fully tested with memtest86). Software information: Redhat Linux AS 4.0, kernel 2.6.9-5.ELsmp. Some logs before crash: Mar  7 16:14:50 WARNING[17546] chan_sip.c: Asked to transmit frame type 64, while native formats is 8 (read/write = 8/8) Mar  7 16:14:50 WARNING[17546] chan_sip.c: Asked to transmit frame type 64, while native formats is 8 (read/write = 8/8) Mar  7 16:14:50 WARNING[17546] chan_sip.c: Asked to transmit frame type 64, while native formats is 8 (read/write = 8/8) Mar  7 16:14:50 WARNING[17546] chan_sip.c: Asked to transmit frame type 64, while native formats is 8 (read/write = 8/8) Mar  7 16:14:50 WARNING[17546] chan_sip.c: Asked to transmit frame type 64, while native formats is 8 (read/write = 8/8) Mar  7 16:14:50 WARNING[17546] chan_sip.c: Asked to transmit frame type 64, while native formats is 8 (read/write = 8/8) Mar  7 16:14:50 WARNING[17546] chan_sip.c: Asked to transmit frame type 64, while native formats is 8 (read/write = 8/8) Mar  7 16:14:50 WARNING[17546] chan_sip.c: Asked to transmit frame type 64, while native formats is 8 (read/write = 8/8) Debug attached as a file.
Comments:	By: Vladimir S. Blazhkun (vovan) 2006-03-10 04:56:24.000-0600 Another crash, seems 1.2.5 isn't very stable. Rolling back to 1.2.4. By: Clod Patry (junky) 2006-04-11 22:52:34 vovan: still an issue with 1.2.6? Can you paste me your dialplan for that extension? Do you have that problem only with SIP? By: Vladimir S. Blazhkun (vovan) 2006-04-24 07:03:55 junky: yes, still got crashes with 1.2.7.1. We are using SIP and Agent channels only. Which extension do you interested in? By: Jared Smith (jsmith) 2006-04-24 11:37:31 Can you provide a full backtrace of the problem?  Can you turn up the verbosity in the logs (see logger.conf), and see if there are any more relevant lines in the logs right before the crash? By: Vladimir S. Blazhkun (vovan) 2006-04-25 09:19:46 Latest asterisk log messages: Apr 25 17:12:29 DEBUG[16379] pbx.c: Launching 'Set' Apr 25 17:12:29 DEBUG[9814] chan_sip.c: Checking device state for peer 1611 Apr 25 17:12:29 VERBOSE[16379] logger.c:     -- Executing Set("SIP/10.0.0.61-b7d6c558", "LANGUAGE()=ru") in new stack Apr 25 17:12:29 DEBUG[20210] app_queue.c: Device 'SIP/1611' changed to state '1' (Not in use) Apr 25 17:12:29 DEBUG[16379] pbx.c: Launching 'Playback' Apr 25 17:12:29 DEBUG[9814] devicestate.c: Changing state for Agent/1611 - state 1 (Not in use) Apr 25 17:12:29 VERBOSE[16379] logger.c:     -- Executing Playback("SIP/10.0.0.61-b7d6c558", "please-hang-up-and-try-again") in new stack Apr 25 17:12:29 DEBUG[20211] app_queue.c: Device 'Agent/1611' changed to state '1' (Not in use) Apr 25 17:12:29 DEBUG[16379] channel.c: Set channel SIP/10.0.0.61-b7d6c558 to write format slin Apr 25 17:12:29 DEBUG[9814] devicestate.c: Changing state for Agent/1611 - state 1 (Not in use) coredump for this crash is attached as crash-11.log. By: Vladimir S. Blazhkun (vovan) 2006-04-27 13:45:10 Are there any solutions for this problem? By: Vladimir S. Blazhkun (vovan) 2006-05-02 03:22:50 Two months passed from bug submission, why are no answers here? Did i do any mistakes in bugreporting? Please help us to solve this instability. By: BJ Weschke (bweschke) 2006-05-02 15:13:53 vovan: your crash from crash-11.log was fixed with the patch applied to the 1.2 branch today from bug ASTERISK-6882. pls update and let us know if you're still having trouble as it's not immediately clear to me if crash-9, crash-10, and crash-12 isn't something else not related to what happened in crash-11. By: BJ Weschke (bweschke) 2006-05-04 16:23:27 vovan: I'm going to close this one out. If you crash again with regard to the ast_frfree problem, please open a new bug and we'll look at the backtrace outside of this.