#10183 NORM Not Tri: Record-82 crashes on os204 while saving a just-recorded audio clip

[Zarro Boogs per Child]

#10183: Record-82 crashes on os204 while saving a just-recorded audio clip
---------------------------------------+------------------------------------
           Reporter:  mikus            |       Owner:  dsd                              
               Type:  defect           |      Status:  new                              
           Priority:  normal           |   Milestone:  Not Triaged                      
          Component:  record-activity  |     Version:  Development build as of this date
         Resolution:                   |    Keywords:                                   
        Next_action:  diagnose         |    Verified:  0                                
Deployment_affected:                   |   Blockedby:                                   
           Blocking:                   |  
---------------------------------------+------------------------------------
Changes (by Quozl):

  * next_action:  reproduce => diagnose


Comment:

 Replying to [comment:3 Quozl]:
 >  * reproduce the symptom myself on Record-82, e.g. by trying your manual
 install process,

 It was easy to reproduce, it happens here on two C3 units on build os204
 with ''Software update'' to Record-82 followed by a restart of Sugar.
 (The version of Record was not consistent with the ''Software update''
 action when viewed in the activity list, hence the Sugar restart).

 The symptom does not happen first time every time, but generally happens
 within ten attempts, even within the same instance of the activity.  It is
 more likely to happen if there is significant movement seen by the camera.

 >  * attempt to reproduce the symptom on Record-81,

 The symptom does not reproduce on Record-81.

 >  * examine the source code changes from Record-81 to Record-82,

 The change in git seems unremarkable.

 >  * correlate the symptom against the source changes,

 Nothing figured out.

 >  * ensure the source that is unpacked on Mikus' system matches the
 source in Record-82,

 Skipping this one since I can reproduce it fine.  The symptom should have
 nothing to do with your method of installation.

 >  * compare the installed RPM list between Mikus' system and mine,

 Skipping this one since I can reproduce it fine.

 >  * run the activity manually from Terminal so as to capture any unusual
 output,

 The last output is

 {{{
 Segmentation fault
 }}}

 >  * run the activity with strace to capture any signals or exit status.

 The strace shows a SIGSEGV hits all threads at about the time the
 photoShutter.wav file is being read, although the thread that got the
 SIGSEGV first had just finished a video related ioctl(2) and was in
 futex(2).  I've no clue what that means.

-- 
Ticket URL: <http://dev.laptop.org/ticket/10183#comment:5>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system