Summary: | ASTERISK-24820: Asterisk Cert 13-1 crashing | ||
Reporter: | Fabian Borot (fborot) | Labels: | |
Date Opened: | 2015-02-23 08:20:01.000-0600 | Date Closed: | 2015-03-26 15:53:53 |
Priority: | Major | Regression? | |
Status: | Closed/Complete | Components: | Channels/chan_sip/General |
Versions: | 13.1.1 | Frequency of Occurrence | Frequent |
Related Issues: | |||
Environment: | Asterisk 13.1-cert1 built by root @ AW-MIA-CORE-07 on a x86_64 running Linux on 2015-02-22 04:21:54 UTC Linux AW-MIA-CORE-07 2.6.18-308.4.1.el5 #1 SMP Tue Apr 17 17:08:00 EDT 2012 x86_64 x86_64 x86_64 GNU/Linux | Attachments: | ( 0) backtrace_Feb_20_submitted_to_digium.txt |
Description: | Sometimes this function is called with a null pointer and the system crashes:
ast_format_get_codec_id (format=0x0) at format.c:332 I am submitting a backtrace as well showing the stack at the moment. We applied a fix in our system to check for null before using the pointer | ||
Comments: | By: Matt Jordan (mjordan) 2015-02-23 08:23:51.050-0600 You are never supposed to have a NULL format. Having one indicates an error elsewhere in the code. Merely adding a NULL check obfuscates the root cause of the problem. Can you provide: # The {{sip.conf}} for your configuration # A pcap showing the message traffic leading up to the crash # A debug log with 'sip set debug on' that shows the SIP message traffic leading up to the crash Thanks! By: Fabian Borot (fborot) 2015-02-23 09:13:21.197-0600 thank you, I will see if we can provide the information although I am not sure if we can because of the fix that we applied so it wont happen again most likely. This is a production system so we can not risk to leave it running in a way that can crash. We added a verbose line to the code so maybe we can provide it when a call comes in that could create the crash By: Matt Jordan (mjordan) 2015-02-23 11:10:50.940-0600 At the very least, we need the SIP configuration and the SIP debug showing the request that came in. The crash is occurring due the calculated joint format capabilities having nothing within it, which shouldn't occur due to a format capability structure always having at least one format in it (the "<none>" format). We'll need the message traffic to solve this correctly. By: Rusty Newton (rnewton) 2015-03-11 17:34:56.394-0500 [~fborot] we haven't received a response back from you. We can't move forward without the debug Matt requested, so if you are unable to respond within the next few weeks then we'll go ahead and close out the issue. Thanks! By: Rusty Newton (rnewton) 2015-03-26 15:54:05.352-0500 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 |