GNOME Bugzilla – Bug 330467
crash in ogg demux using Rhythmbox
Last modified: 2006-03-15 14:43:55 UTC
Distribution: Fedora Core release 4 (Rawhide) Package: rhythmbox Severity: critical Version: GNOME2.13.90 0.9.x Gnome-Distributor: Red Hat, Inc Synopsis: crashed after import, bug-buddy dialog presented, works fine if I ignore bug-buddy Bugzilla-Product: rhythmbox Bugzilla-Component: Importing Bugzilla-Version: 0.9.x BugBuddy-GnomeVersion: 2.0 (2.13.3) Description: Description of the crash: After importing music files, bug buddy pops up saying rhythmbox crashed but if I ignore the dialog, rhythmbox plays music fine. Steps to reproduce the crash: 1. import lots of files 2. 3. Expected Results: How often does this happen? seems to happen every time although this is the first time I've ignored the dialog Additional Information: Debugging Information: Backtrace was generated from '/usr/bin/rhythmbox' (no debugging symbols found) Using host libthread_db library "/lib/libthread_db.so.1". (no debugging symbols found) `shared object read from target memory' has disappeared; keeping its symbols. (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1208956736 (LWP 19284)] [New Thread -1374704736 (LWP 20985)] [New Thread -1364214880 (LWP 20983)] [New Thread -1332745312 (LWP 20563)] [New Thread -1290806368 (LWP 19874)] [New Thread -1269826656 (LWP 19521)] [New Thread -1301296224 (LWP 19520)] [New Thread -1343235168 (LWP 19519)] [New Thread -1322255456 (LWP 19518)] [New Thread -1353725024 (LWP 19491)] [New Thread -1244251232 (LWP 19286)] [New Thread -1233761376 (LWP 19285)] (no debugging symbols found) 0x00a5b410 in __kernel_vsyscall ()
+ Trace 66023
Thread 10 (Thread -1353725024 (LWP 19491))
------- Bug created by bug-buddy at 2006-02-08 21:04 -------
This is crashing inside gstreamer, so moving there.
Stack trace seems unique. Could you launch "rhythmbox -d" to see if the crash happens always on a specific file ? It could be related to http://bugzilla.gnome.org/show_bug.cgi?id=322769 What version of RB and GStreamer are you using ?
I don't see an actual crash anywhere in the stack trace, but some of the traces look rather messed up (but that might just be the lack of debugging symbols, dunno).
The crash is in thread 10, however the track is definitely messed up, as it lists "gnome_init_with_popt_table" as the signal handler.
Unfortunately there's not much we can do about this one really. The traces are just too messed up. There have been some fixes in the latest releases though for things that could have caused this, so it would be great if you could try again with a current version of GStreamer core (=0.10.4) and gst-plugins-base (=0.10.5) to check if this still happens. Just out of curiosity: is "but if I ignore the dialog, rhythmbox plays music fine" even possible in case of a segfault?
Yeah, if it wasn't the streaming thread that actually segfaulted. I've seen that other threads than the segfaulting one can sometimes continue running until the program actually terminates.
This is working in the current fedora rawhide version.