Summary: | ASTERISK-15973: Asterisk core dumps using MOH | ||
Reporter: | dovid (dovid) | Labels: | |
Date Opened: | 2010-04-19 08:38:45 | Date Closed: | 2015-03-27 15:23:57 |
Priority: | Minor | Regression? | No |
Status: | Closed/Complete | Components: | Resources/res_musiconhold |
Versions: | Frequency of Occurrence | ||
Related Issues: | |||
Environment: | Attachments: | ( 0) ast_crash.log ( 1) core_dump2.log ( 2) full_dump.log | |
Description: | After Asterisk plays MOH several times in core dumps. I am using MOH+RT+MadPlay. This randomly happens after Asterisk initiates the MOH several times. I am not able to re-produce it on demand. As per SeanBright on irc, there may be an issue because there is a double free (not sure 100% what that means). ****** ADDITIONAL INFORMATION ****** Using CentOs 5.4. | ||
Comments: | By: Leif Madsen (lmadsen) 2010-04-19 13:28:30 Acknowledging issue as the backtrace seems to be complete. By: alric (alric) 2010-04-19 15:35:29 I just installed madplay on my 1.6.2.6 machine and I'm using it with Realtime. Are you using the cachrtclasses option in musiconhold.conf? By: dovid (dovid) 2010-04-19 18:34:43 Alric: No I am not. By: dovid (dovid) 2010-04-22 06:31:28 FTI: Upgraded to 1.6.2.6 and I still have the same issue. By: dovid (dovid) 2010-04-22 06:41:18 The last file that I uploaded (ast_crash.log) is from the console where I started asterisk when asterisk core dumped. It does seem to be a double free issue. By: dovid (dovid) 2010-04-25 10:20:08 I started using files instead of real time and for the most part it has been a lot better. Every so often when I do moh reload I still get a core dump. This can be seen in core_dump2.log By: dovid (dovid) 2010-05-11 02:15:06 Ping. By: Leif Madsen (lmadsen) 2010-05-17 10:34:33 This issue has been imported into the Digium internal issue tracker and as soon as development resources are available to move this issue forward it will be assigned to a developer. Until then, anyone in the community is welcome to move this issue forward. By: dovid (dovid) 2010-12-24 06:44:45.000-0600 Been a few months. Any update on this one ? By: antonio (antonio) 2011-01-27 05:26:16.000-0600 I'm seeing something similar on 1.6.2.16 as well. It's indeed a double free(), and has to do with referencing. I'm trying to track this down. By: dovid (dovid) 2011-11-15 09:16:08.001-0600 Ping. Any way I can help move this along? By: Joshua C. Colp (jcolp) 2015-03-27 15:23:57.374-0500 After looking at this I believe it was fixed as of revision 410044 in 11. If this is not the case please feel free to reopen. |