Project

General

Profile

Bug #7216

Asterisk crashing due to too many TCP connections

Added by James Cardenas about 1 year ago. Updated about 1 year ago.

Status:
New
Priority:
Normal
Assignee:
-
Category:
Asterisk
Target version:
-
Security issue:
No
In versions:
18.03
Read documentation?:
Yes

Description

This is in reference to AST-2018-005 bug report as referenced:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=891228

This issue reportedly affected up to Asterisk 15.2.1. In Wazo 18.03, Asterisk package 15.2.2 is installed and so the bug possibility exists in this version as well. The latest version of Asterisk is 15.5 and we are informed this issue has been resolved in the newest build.

We'd like to confirm if the issue has been replicated in other installations of 15.2.2 and if the next Wazo package will include the bug fix.

vsw13-threads.txt View (66.9 KB) James Cardenas, 2018-07-18 22:57

History

#1 Updated by James Cardenas about 1 year ago

James Cardenas wrote:

This is in reference to AST-2018-005 bug report as referenced:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=891228

This issue reportedly affected up to Asterisk 15.2.1. In Wazo 18.03, Asterisk package 15.2.2 is installed and so the bug possibility exists in this version as well. The latest version of Asterisk is 15.5 and we are informed this issue has been resolved in the newest build.

We'd like to confirm if the issue has been replicated in other installations of 15.2.2 and if the next Wazo package will include the bug fix.

I'm showing the issue was fixed in 15.2.2. What happened today was we had switched several end points to TCP as part of our TCP conversion process. We noticed that after an extension after being registered/unregistered a few times due to changes on the end point itself, Asterisk crashed and had to be restarted. It has not crashed currently and we still continue to plan our migration over to TCP.

#2 Updated by James Cardenas about 1 year ago

Here is what shows up in

/var/log/syslog

Jul 18 15:08:07 vsw13 kernel: [573062.001301] asterisk2454: segfault at 18 ip 0000560230877570 sp 00007f8fbc196378 error 4 in asterisk[560230720000+306000]

This is the exact time it crashed.

#3 Updated by James Cardenas about 1 year ago

I'll post a back trace this evening after close of business.

#4 Updated by James Cardenas about 1 year ago

James Cardenas wrote:

I'll post a back trace this evening after close of business.

As promised, here's the backtrace as an attachment.

An excerpt is below:

[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Core was generated by `/usr/sbin/asterisk -g -U asterisk'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  ast_iostream_get_fd (stream=0x0) at iostream.c:75
[Current thread is 1 (Thread 0x7ff902e0b700 (LWP 24066))]
#0  ast_iostream_get_fd (stream=0x0) at iostream.c:75
No locals.
#1  0x00007ff90a79a3c2 in sip_prepare_socket (p=0x7ff98c014918) at chan_sip.c:29305
        th = 0x0
        ca = <optimized out>
        launched = 0
        s = 0x7ff98c014b78
        sa_tmp = {ss = {ss_family = 14128, __ss_padding = "\022@\371\177\000\000\276\372O[\000\000\000\000l\242\003\000\000\000\000\000\220fE\276\323U\000\000 7\022@\371\177\000\000\226L\265\274\3$
#2  __sip_xmit (p=0x7ff98c014918, data=0x7ff9400c22f0) at chan_sip.c:3758
        res = 0
        __PRETTY_FUNCTION__ = "__sip_xmit" 
#3  0x00007ff90a79b0a4 in __sip_reliable_xmit (p=p@entry=0x7ff98c014918, seqno=seqno@entry=119, resp=resp@entry=0, data=<optimized out>, fatal=fatal@entry=1, sipmethod=<optimized out>) at chan_sip$
        pkt = 0x7ff9401c6190
        siptimer_a = 32000
        xmitres = 0
        respid = 0
        __PRETTY_FUNCTION__ = "__sip_reliable_xmit" 

Also available in: Atom PDF