Summary: | ASTERISK-17955: Segmentation fault with LibSS7 | ||||
Reporter: | Ian M Sherman (shermster) | Labels: | |||
Date Opened: | 2011-06-02 16:55:30 | Date Closed: | 2011-10-26 18:05:22 | ||
Priority: | Critical | Regression? | No | ||
Status: | Closed/Complete | Components: | Channels/chan_dahdi/SS7 | ||
Versions: | 1.8.4 | Frequency of Occurrence | |||
Related Issues: |
| ||||
Environment: | Attachments: | ( 0) gdb-11918.txt ( 1) gdb-8636.txt | |||
Description: | Multiple segmentation faults occur randomly throughout the day. Using asterisk-1.8.4.1, libss7-1.0.2, dahdi-linux-complete-2.4.1.2+2.4.1 ****** ADDITIONAL INFORMATION ****** Compiled with DONT_OPTIMIZE Running on 2.6.18-128.el5 #1 SMP Wed Jan 21 10:44:23 EST 2009 i686 i686 i386 GNU/Linux Could not find BETTER_BACKTRACES in compile flags - am I missing a library to enable this? | ||||
Comments: | By: David Woolley (davidw) 2011-06-03 06:21:08 The first backtrace is not a segmentation fault (it's an abort) and indicates a memory management problem, so you need to enable thread debugging and run under valgrind. The second one probably is a segmentation fault, but that is not certain, but, in any case, may be a consequence of the memory management problem. By: Ian M Sherman (shermster) 2011-06-03 06:34:25 Thanks David. I'll do this. Quick question - when running under Valgrind, will Asterisk automatically restart if there is a crash? By: Kenneth Van Velthoven (kvveltho) 2011-07-21 13:07:53.137-0500 I have the same problem. Also on 1.8.5 RC 1. Probably the problem will also exist on 1.8.5 Any progress on this issue? By: Richard Mudgett (rmudgett) 2011-09-16 13:45:07.481-0500 I have attached some patches to ASTERISK-17966 that may help. By: Richard Mudgett (rmudgett) 2011-10-26 16:25:00.721-0500 Any status update for this issue in light of the activity on ASTERISK-17966? By: Ian M Sherman (shermster) 2011-10-26 17:18:52.523-0500 @Richard unfortunately I don't have access to this system anymore so am unable to test or provide any feedback on those patches. By: Richard Mudgett (rmudgett) 2011-10-26 18:05:22.256-0500 Closing issue as reporter can no longer supply feedback. |