[Home]

Summary:ASTERISK-24758: Lengthy CDR chain crashes during object destruction
Reporter:JoshE (n8ideas)Labels:
Date Opened:2015-02-04 11:36:04.000-0600Date Closed:2015-04-07 16:14:57
Priority:MajorRegression?
Status:Closed/CompleteComponents:CDR/General
Versions:13.1.0 Frequency of
Occurrence
Related
Issues:
Environment:Attachments:( 0) backtrace.txt
Description:Observed this crash in __ast_string_field_init today for the first time.

Not sure on a repro for it yet, but the box was quite lightly loaded.
Comments:By: JoshE (n8ideas) 2015-02-04 11:37:52.777-0600

Very lengthy text crash BT attached.
By: Matt Jordan (mjordan) 2015-02-04 14:13:46.040-0600

Somehow, you've managed to get several thousand CDRs generated for a single channel. Things break down when the channel hangs up, and the chain of CDRs tries to unwind during its destruction:

{noformat}
#8897 0x00000000004956cd in cdr_object_dtor (obj=0x7f0f28172bb8) at cdr.c:852
#8898 0x000000000045af83 in internal_ao2_ref (user_data=0x7f0f28172bb8, delta=-1, file=0x699584 "astobj2.c", line=513, func=0x6997b1 <__FUNCTION__.8336> "__ao2_ref") at astobj2.c:442
#8899 0x000000000045b24d in __ao2_ref (user_data=0x7f0f28172bb8, delta=-1) at astobj2.c:513
#8900 0x000000000045b2bd in __ao2_cleanup (obj=0x7f0f28172bb8) at astobj2.c:526
#8901 0x000000000049a15e in _dtor_cdr (v=0x7f0fe8a6cb78) at cdr.c:2060
#8902 0x000000000049a522 in handle_channel_cache_message (data=0x0, sub=
{noformat}

To fix this, we're going to need to know how you're produced the situation. Specific steps that result in the crash and the errant CDR chain will be needed.
By: JoshE (n8ideas) 2015-02-13 14:54:37.507-0600

I think I figured out how we got into this state.  I'll be working on a repro scenario for this shortly.
By: Rusty Newton (rnewton) 2015-03-06 15:54:57.381-0600

Thanks. I'll put this into Waiting for Feedback until you get reproduction steps. Remember to use "Enter Feedback" to send it back to us.
By: Rusty Newton (rnewton) 2015-03-23 07:35:55.561-0500

[~n8ideas] were you able to reproduce?
By: Rusty Newton (rnewton) 2015-04-07 16:14:49.304-0500

Suspended due to lack of activity. Please request a bug marshal in #asterisk-bugs on the IRC network irc.freenode.net to reopen the issue should you have the additional information requested. Further information on issue tracker usage can be found in the Asterisk Issue Guidlines [1].

[1] https://wiki.asterisk.org/wiki/display/AST/Asterisk+Issue+Guidelines