[asterisk-bugs] [JIRA] (ASTERISK-25168) Random Core Dumps on Asterisk 13.4 PJSIP, in ast_channel_name at channel_internal_api.c

Tue Jun 30 14:09:32 CDT 2015

     [ https://issues.asterisk.org/jira/browse/ASTERISK-25168?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Richard Mudgett updated ASTERISK-25168:
---------------------------------------

    Attachment: jira_asterisk_25168_v13_test.patch

[^jira_asterisk_25168_v13_test.patch] - This should fix both crashes.  I have placed some {{BUGBUG}} messages in a couple locations dealing with FAX.  Let me know which ones you see if you see any in the logs.

# The NULL channel crash.  Let me know if you see this message in the log:
{{BUGBUG T.38 auto reject timer expired without a session channel.}}
If it doesn't crash then that is proof positive that the NULL channel crash is fixed.  This crash happened when the caller called a FAX number and then hung up immediately after the call was answered.
# The MWI subscription crash.  I needed to give a datastore on the subscription a reference to the MWI subscription object to prevent it going away.  What makes this tricky is that it creates a circular object reference chain.  The circular reference can never go away without explicit code to break the loop such as when setting up the subscription fails or the subscription shuts down.

> Random Core Dumps on Asterisk 13.4 PJSIP, in ast_channel_name at channel_internal_api.c
> ---------------------------------------------------------------------------------------
>
>                 Key: ASTERISK-25168
>                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-25168
>             Project: Asterisk
>          Issue Type: Bug
>      Security Level: None
>          Components: Resources/res_pjsip
>    Affects Versions: 13.4.0
>         Environment: Our software:
> Asterisk Realtime 13.4 PJSIP Driver
> mysql Ver 14.14
> pjproject 2.3
> spandsp 0.0.6
> jansson 2.7
> CentOS 6.6 64 bits on Vmware
> Number of endpoints : > 700
> Numbers of calls : 3000/day
> Our Hardware:
> Phones : Cisco SPA514G FW: 7.5.7
> ATA : Audiocodes MP124
> T1 : Mediatrix 3532 ISDN to SIP gateway
> CPU : Quadcore Intel(R) Xeon(R) CPU E5-2650
> RAM : 3 GB
>            Reporter: Carl Fortin
>            Assignee: Richard Mudgett
>         Attachments: backtrace_1108.txt, backtrace-1_1741.txt, backtrace-2_1741_24.txt, backtrace-3_1741_44.txt, backtrace-4_1742.txt, backtrace-5_1743_19.txt, backtrace5.txt, backtrace-6_1744_56.txt, backtrace-7_1745_08.txt, jira_asterisk_25168_v13_test.patch, Last_backtrace.txt, mmlog, myDebugLog2, myDebugLog_asterisk.zip, Putty console before crash.txt, valgrind.txt
>
>
> We are running Asterisk 13.4 PJSIP in a production environment and get random crashes. Everything seemed to be working fine for a couple of days and we now get crashes more and more.
> I have also noticed that crashes happens even though no calls are made, so I don't think it has anything to do with the dialplan. I've really tested everything in a test machine before moving my dialplan to PJSIP. Most of the phones use direct media, so the load on asterisk is low.
> I get no errors when doing asterisk -cvvvvvvvvvvvvvv.
> [Edit by Rusty - For reference here I'm extracting from your debug log the last lines before the crash.]
> {noformat}
> [Jun 17 11:08:19] DEBUG[23136][C-0000020e] res_rtp_asterisk.c: Got RTCP report of 76 bytes
> [Jun 17 11:08:19] DEBUG[23136][C-0000020e] acl.c: Not an IPv4 nor IPv6 address, cannot get port.
> [Jun 17 11:08:19] DEBUG[23136][C-0000020e] netsock2.c: Splitting 'dti-asterisk.cegep-fxg.qc.ca' into...
> [Jun 17 11:08:19] DEBUG[23136][C-0000020e] netsock2.c: ...host 'dti-asterisk.cegep-fxg.qc.ca' and port ''.
> [Jun 17 11:08:19] DEBUG[23136][C-0000020e] acl.c: Not an IPv4 nor IPv6 address, cannot get port.
> [Jun 17 11:08:19] DEBUG[23136][C-0000020e] acl.c: Attached to given IP address
> [Jun 17 11:08:19] DEBUG[11851] netsock2.c: Splitting '206.80.250.101:40631' into...
> [Jun 17 11:08:19] DEBUG[11851] netsock2.c: ...host '206.80.250.101' and port '40631'.
> [Jun 17 11:08:19] DEBUG[11851] netsock2.c: Splitting '206.167.100.36:0' into...
> [Jun 17 11:08:19] DEBUG[11851] netsock2.c: ...host '206.167.100.36' and port '0'.
> [Jun 17 11:08:28] Asterisk 13.4.0 built by root @ dti-asterisk.cegep-fxg.qc.ca on a x86_64 running Linux on 2015-06-16 23:48:22 UTC
> {noformat}
> Excerpt from the trace
> {noformat}Core was generated by `/usr/sbin/asterisk -f -vvvg -c'.
> Program terminated with signal 11, Segmentation fault.
> #0  0x00000000004c9c81 in ast_channel_name (chan=0x0) at channel_internal_api.c:476
> 476	DEFINE_STRINGFIELD_GETTER_FOR(name);
> #0  0x00000000004c9c81 in ast_channel_name (chan=0x0) at channel_internal_api.c:476
> No locals.
> #1  0x00007f4b13724ffa in t38_automatic_reject (obj=0x7f4b441c50e8) at res_pjsip_t38.c:201
>         session = 0x7f4b441c50e8
>         datastore = 0x7f4b7c0fcc98
>         session_media = 0x7f4b440d48b8
>         __PRETTY_FUNCTION__ = "t38_automatic_reject"
> #2  0x00000000005dc343 in ast_taskprocessor_execute (tps=0x7f4b440a7fe8) at taskprocessor.c:769
>         local = {local_data = 0x2fc00649ca0, data = 0x649ca0}
>         t = 0x7f4b20003000
>         size = 0
>         __PRETTY_FUNCTION__ = "ast_taskprocessor_execute"
> #3  0x00000000005e481c in execute_tasks (data=0x7f4b440a7fe8) at threadpool.c:1157
>         tps = 0x7f4b440a7fe8
> #4  0x00000000005dc343 in ast_taskprocessor_execute (tps=0x2fad768) at taskprocessor.c:769
>         local = {local_data = 0x7f4b01d07c90, data = 0x0}
>         t = 0x7f4b20002fd0
>         size = 0
>         __PRETTY_FUNCTION__ = "ast_taskprocessor_execute"
> {noformat}

--
This message was sent by Atlassian JIRA
(v6.2#6252)