[asterisk-bugs] [Asterisk 0016531]: [regression] chan_local audio crash

[Asterisk Bug Tracker]

A NOTE has been added to this issue. 
====================================================================== 
https://issues.asterisk.org/view.php?id=16531 
====================================================================== 
Reported By:                john8675309
Assigned To:                
====================================================================== 
Project:                    Asterisk
Issue ID:                   16531
Category:                   Channels/chan_local
Reproducibility:            always
Severity:                   crash
Priority:                   normal
Status:                     acknowledged
Asterisk Version:           SVN 
JIRA:                       SWP-615 
Regression:                 Yes 
Reviewboard Link:            
SVN Branch (only for SVN checkouts, not tarball releases): N/A 
SVN Revision (number only!): 237133 
Request Review:              
====================================================================== 
Date Submitted:             2010-01-01 09:32 CST
Last Modified:              2010-01-05 11:45 CST
====================================================================== 
Summary:                    [regression] chan_local audio crash
Description: 
Starting with asterisk 1.4.26 we started experiencing crashes whenever we
did an originate from a local channel that played audio.

The trunk 237133
no longer crashes, but it only starts 1/2 of the call (line 998)  I will
attach my dial plan and core dumps from earlier versions.
====================================================================== 

---------------------------------------------------------------------- 
 (0116072) palbrecht (reporter) - 2010-01-05 11:45
 https://issues.asterisk.org/view.php?id=16531#c116072 
---------------------------------------------------------------------- 
I think I found the problem or at least part of the problem. The
application that plays mp3 streams wasn't updated when the frame_list field
was added to the frame structure. As a result, it seems the field contains
random data and segfaults on the first frame write. I changed the code to
clear the frame structure before use and the problem went away. What do you
think? Make sense? Now I'm getting the log error for exceptionally long
frames. Does the frame need to be adjusted or broken up like the other
applications that were changed with 200991? 

Issue History 
Date Modified    Username       Field                    Change               
====================================================================== 
2010-01-05 11:45 palbrecht      Note Added: 0116072                          
======================================================================