totem crashed with SIGSEGV in gst_pulsering_context_state_cb()

[Stephen Monteith]

Binary package hint: totem

I opened a .MOV file from Nautilus that is about 1 minute long. When it completed, I opened another MOV file from Nautilus which caused the crash of the existing instance of Totem

ProblemType: Crash
Architecture: i386
Date: Wed Oct 28 21:34:00 2009
DistroRelease: Ubuntu 9.10
ExecutablePath: /usr/bin/totem
Package: totem 2.28.1-0ubuntu4
ProcCmdline: totem /home/username/Pictures/2007-12-06--21.32.28/100_2200.MOV
ProcEnviron:
 LANG=en_CA.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-14.48-generic
SegvAnalysis:
 Segfault happened at: 0xd74917 <gst_pulsering_context_state_cb+151>: mov 0x1c4(%esi),%eax
 PC (0x00d74917) ok
 source "0x1c4(%esi)" (0x000001c4) not located in a known VMA region (needed readable region)!
 destination "%eax" ok
SegvReason: reading NULL VMA
Signal: 11
SourcePackage: totem
StacktraceTop:
 gst_pulsering_context_state_cb (c=0xb3413008,
 ?? () from /usr/lib/libpulse.so.0
 pa_context_disconnect () from /usr/lib/libpulse.so.0
 gst_pulsering_destroy_context (pbuf=0xb3413008)
 gst_pulseringbuffer_finalize (object=0x9f48ee8)
Title: totem crashed with SIGSEGV in gst_pulsering_context_state_cb()
Uname: Linux 2.6.31-14-generic i686
UserGroups: adm admin audio cdrom dialout dip floppy fuse lpadmin netdev plugdev powerdev scanner video
XsessionErrors:
 (gnome-settings-daemon:2154): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (gnome-settings-daemon:2154): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed
 (nautilus:2288): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed
 (polkit-gnome-authentication-agent-1:2318): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
 (nautilus:3810): Eel-CRITICAL **: eel_preferences_get_boolean: assertion `preferences_is_initialized ()' failed

[Apport retracing service (apport)]

StacktraceTop:gst_pulsering_context_state_cb (c=0xb3413008,
pa_context_set_state (c=0xb3413008,
pa_context_disconnect (c=0xb3413008) at pulse/context.c:1037
gst_pulsering_destroy_context (pbuf=0x9f48ee8)
gst_pulseringbuffer_finalize (object=0x9f48ee8)

[xteejx (xteejx)]

I can confirm this from my duplicate.

[xteejx (xteejx)]

Marking Triaged, apport retraced this fine - should be enough debugging information here for a developer to debug this.

[madbiologist (me-again)]

I've reported this bug upstream as per the instructions at https://wiki.ubuntu.com/Bugs/Upstream/GNOME

[madbiologist (me-again)]

@Stephen Monteith - One of the developers has asked a question in the upstream bug:

what version of gst-plugins-good do you have ?

This might have been fixed by the following commit:

commit 55b5e2e3f38803cd50930db6fd6c7fdb09f0e815
Author: Wim Taymans <email address hidden> Wed Jan 27 13:26:46 2010
Committer: Wim Taymans <email address hidden> Fri Feb 12 13:53:57
2010

pulsesink: avoid segfault when shutting down

when we are shutting down, we might still receive state updates from pulseaudio
but since we are unparented we should not do anything with the NULL parent
anymore.

[Sebastien Bacher (seb128)]

the bug reports there are on karmic and the upstream commit timestamp suggests the bug is fixed in lucid

[Stephen Monteith (stephen-monteith)]

I cannot reproduce this bug anymore, I have since upgraded to Ubuntu 10.04 Lucid. The version of gstreamer0.10-plugins-good I have installed now is 0.10.21-1ubuntu3

[madbiologist (me-again)]

Sebastien, I'm not completely sure of the context of the "there" to which you refer. With regards to the upstream commit timestamp, are you referring to the upstream bug's question and supporting information which I pasted into comment #7 of this bug? While that particular commit was indeed included in gstreamer0.10-plugins-good 0.10.19 (see http://gstreamer.freedesktop.org/releases/gst-plugins-good/0.10.19.html ), I interpret that as the developer saying that the commit might have possibly fixed the reported bug, but they cannot be certain. I would be hesitant to close this bug based on that premise, or indeed to use such a premise as the basis for closing any bug. On the other hand Stephen Monteith's comment #9 of this bug seems a much more reliable basis for closing this bug, a basis which is further reinforced by him being the original reporter.

I have forwarded Stephen's answer upstream and closed the upstream bug.

[Dart (danhart)]

I have upgraded to Ubuntu 10.04 Lucid Netbook and I can not recreate the
issue.

gstreamer0.10-ffmpeg

version 0.10.10-1

On Sat, 2010-06-19 at 16:19 +0000, madbiologist wrote:
> @Stephen Monteith - One of the developers has asked a question in the
> upstream bug:
>
> what version of gst-plugins-good do you have ?
>
> This might have been fixed by the following commit:
>
> commit 55b5e2e3f38803cd50930db6fd6c7fdb09f0e815
> Author: Wim Taymans <email address hidden> Wed Jan 27 13:26:46 2010
> Committer: Wim Taymans <email address hidden> Fri Feb 12 13:53:57
> 2010
>
> pulsesink: avoid segfault when shutting down
>
> when we are shutting down, we might still receive state updates from pulseaudio
> but since we are unparented we should not do anything with the NULL parent
> anymore.
>