Created attachment 637631 [details]
File: core_backtrace

Created attachment 637632 [details]
File: environ

Created attachment 637633 [details]
File: limits

Created attachment 637634 [details]
File: backtrace

Created attachment 637635 [details]
File: cgroup

Created attachment 637636 [details]
File: executable

Created attachment 637637 [details]
File: maps

Created attachment 637638 [details]
File: dso_list

Created attachment 637639 [details]
File: proc_pid_status

Created attachment 637640 [details]
File: var_log_messages

Created attachment 637641 [details]
File: open_fds

Rhythmbox crashes anytime I start playback of _any_ media, regardless of codec, local or radio stream ...

The only useful debug output is:
"TypeError: can't convert return value to desired type
Segmentation fault (core dumped)"


backtrace_rating: 4
Package: rhythmbox-2.98-4.fc18
OS Release: Fedora release 18 (Spherical Cow)

Tried to start playback.

backtrace_rating: 4
Package: rhythmbox-2.98-4.fc18
Architecture: i686
OS Release: Fedora release 18 (Spherical Cow)

ladspa (pulseaudio-equalizer) cannot be removed i think this is crashing pulseaudio and sound is distorted

backtrace_rating: 4
Package: rhythmbox-2.98-4.fc18
OS Release: Fedora release 18 (Spherical Cow)

Crashing on playback, no matter what format file I've tried.

Error oappearing when i try to play any track

backtrace_rating: 4
Package: rhythmbox-2.98-4.fc18
OS Release: Fedora release 18 (Spherical Cow)

Marking as a blocker as per Final criteria "All applications listed under the Applications menu or category must withstand a basic functionality test and not crash after a few minutes of normal use. They must also have working Help and Help -> About menu items"

Rhythmbox (2.98-4) works for me, at least ogg playback.

From the backtrace it also looks like this might be related having the crossfade option enabled, which is disabled by default.
If that's the case, I think it's questionable whether this should actually be a blocker.

This is seemingly a bug in pygobject3 introspection triggered by the ReplayGain plugin.

It does this:

→       def create_stream_filter_cb(self, player, uri):•
→       →       print "creating rgvolume instance for stream %s" % uri•
→       →       rgvolume = Gst.ElementFactory.make("rgvolume", None)•
→       →       rgvolume.connect("notify::target-gain", self.xfade_target_gain_cb)•
→       →       self.set_rgvolume(rgvolume)•
→       →       return [rgvolume]•

as a handler for this signal:

→       signals[GET_STREAM_FILTERS] =•
→       →       g_signal_new ("get-stream-filters",•
→       →       →             G_OBJECT_CLASS_TYPE (object_class),•
→       →       →             G_SIGNAL_RUN_LAST,•
→       →       →             0,•
→       →       →             rb_signal_accumulator_value_array, NULL,•
→       →       →             rb_marshal_BOXED__STRING,•
→       →       →             G_TYPE_ARRAY,•
→       →       →             1,•
→       →       →             G_TYPE_STRING);•

So the assumption is that the "return [rgvolume]" will implicity create a GArray.  pygobject3 has a function pyg_value_from_pyobject which handles GValueArrays but not GArrays:

             else if (PySequence_Check(obj) &&                          
                        G_VALUE_HOLDS(value, G_TYPE_VALUE_ARRAY))       
                 return pyg_value_array_from_pyobject(value, obj, NULL);

This isn't really a blocker bug by the criteria, since we don't enable the ReplayGain plugin by default.

Discussed at 2012-12-19 blocker review meeting: http://meetbot.fedoraproject.org/fedora-bugzappers/2012-12-19/f18final-blocker-review-6.2012-12-19-17.02.log.txt . Rejected as a blocker on the grounds that you have to enable a non-default plugin to make it crash, but it'd be good to get a fix for this ASAP, be nice to have it as a 0-day. Some suggestion that we might need to file this upstream, as apparently the maintainer ignores RH bugs?

filed upstream here:

http://bugzilla.gnome.org/690514

(moving to pygobject since the bug is there really)

pygobject3-3.4.2-5.fc18 has been submitted as an update for Fedora 18.
https://admin.fedoraproject.org/updates/pygobject3-3.4.2-5.fc18

Package pygobject3-3.4.2-5.fc18:
* should fix your issue,
* was pushed to the Fedora 18 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing pygobject3-3.4.2-5.fc18'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2012-20727/pygobject3-3.4.2-5.fc18
then log in and leave karma (feedback).

Why we didn't consider this for NTH I've no idea. Obviously it'd be good to get it in as NTH so it's fixed on lives.

pygobject3-3.4.2-6.fc18 has been submitted as an update for Fedora 18.
https://admin.fedoraproject.org/updates/pygobject3-3.4.2-6.fc18

(In reply to comment #26)
> Why we didn't consider this for NTH I've no idea. Obviously it'd be good to
> get it in as NTH so it's fixed on lives.

IIRC, it was because the plugin in question isn't on the lives and therefore doesn't affect livecds and could be fixed with an update. I could be mis-remembering something here, though.

tflink, it is, in fact on the lives and is installed by default.  It is merely not enabled (unless you install to a system while preserving /home, and thus settings where I had previously enabled it).

Discussed at 2012-12-21 blocker review meeting: http://meetbot.fedoraproject.org/fedora-bugzappers/2012-12-21/f18final-blocker-review-7.2012-12-21-18.33.log.txt . Rejected as NTH - it does affect lives as the plugin is present there, but the feeling was that poking pygobject for such a relatively trivial bug was not a good plan.

Upstream bug is closed FIXED now.

pygobject3-3.4.2-6.fc18 has been pushed to the Fedora 18 stable repository.  If problems still persist, please make note of it in this bug report.