GNOME Bugzilla – Bug 517601
segfault when exiting f-spot
Last modified: 2008-02-25 08:57:55 UTC
Please describe the problem: When I exit f-spot, it crashes with a segmenation fault. Steps to reproduce: 1. Fetch photos.db from http://public-files.askwar.s3.amazonaws.com/photos.db.bz2 2. Start f-spot 0.4.2 with this database 3. Exit f-spot Actual results: F-Spot crashes. Expected results: No segfault. Does this happen every time? Yes. Other information: Using f-spot 0.4.2 on Gentoo Linux.
(In reply to comment #0) > Please describe the problem: > When I exit f-spot, it crashes with a segmenation fault. please attach some console output, obviously including the segfault. I couldn't reproduce this with your db. is this sistematic or happening only sometimes?
It happens all the time, yes. But what should I show you? All that's printed is "Speicherzugriffsfehler" (segmentation fault). Can I run f-spot in some sort of debug mode, so that it prints out more information? What's interesting is, that you don't get that error message. Hmm.... I also get it, when I move the photos.db out of the way. So it's NOT related to the photos.db. --($:~/.gnome2/f-spot)-- f-spot Initializing Mono.Addins Starting new FSpot server Query: SELECT photos.id, photos.time, photos.uri, photos.description, photos.roll_id, photos.default_version_id, photos.rating FROM photos WHERE photos.id NOT IN (SELECT photo_id FROM photo_tags WHERE tag_id = 2) ORDER BY photos.time Query: SELECT photos.id, photos.time, photos.uri, photos.description, photos.roll_id, photos.default_version_id, photos.rating FROM photos WHERE photos.id NOT IN (SELECT photo_id FROM photo_tags WHERE tag_id = 2) ORDER BY photos.time Query: SELECT photos.id, photos.time, photos.uri, photos.description, photos.roll_id, photos.default_version_id, photos.rating FROM photos WHERE photos.id NOT IN (SELECT photo_id FROM photo_tags WHERE tag_id = 2) ORDER BY photos.time Query: SELECT photos.id, photos.time, photos.uri, photos.description, photos.roll_id, photos.default_version_id, photos.rating FROM photos WHERE photos.id NOT IN (SELECT photo_id FROM photo_tags WHERE tag_id = 2) ORDER BY photos.time Reloading Query: SELECT photos.id, photos.time, photos.uri, photos.description, photos.roll_id, photos.default_version_id, photos.rating FROM photos WHERE photos.id NOT IN (SELECT photo_id FROM photo_tags WHERE tag_id = 2) ORDER BY photos.time item changed System - gnome-dev-harddisk - Mountpoint file:///Windows True True Harddrive Harddrive Daten - gnome-dev-harddisk-usb - Mountpoint file:///media/Daten_ True True Harddrive Harddrive PConKey4AS - gnome-dev-harddisk-usb - Mountpoint file:///media/PConKey4AS True True Harddrive Harddrive (f-spot:29175): GdkPixbuf-WARNING **: GdkPixbufLoader finalized without calling gdk_pixbuf_loader_close() - this is not allowed. You must explicitly end the data stream to the loader before dropping the last reference. Speicherzugriffsfehler It's also not related to some external plugin - I removed ~/.gnome2/f-spot and still get that "Speicherzugriffsfehler". Do you want me to do a "debug build" of f-spot? If so, please tell me how.
When I exit f-spot, I also get the following in my kern.log/syslog: Feb 20 11:38:47 winnb000488 f-spot[21387]: segfault at b51e8968 eip b51e8968 esp bfeb0aec error 4 I'm using the TuxOnIce kernel sources with Kernel 2.6.24 (actually tuxonice-sources-2.6.24-r2 from Gentoo).
Hello, I have the same problem : when closing f-spot I got "Segmentation fault" in my console, but nothing more. I think it is more a mono related problem : I have this message with other mono programs such Tomboy. I am running f-spot 0.4.1 with mono 1.2.6 on Archlinux. After investigation I think that using Gnome libraries (using Gnome; in C#) causes the problem. Here is a very simple program to reproduce the problem : using System; using Gtk; using Gnome; public class MainClass { static void Main (string[] args) { new MyProgram (args); } } class MyProgram { public MyProgram(string[] args) { Program program = new Program ("Demo", "0.1", Modules.UI, args); App app = new App ("gnome-demo", "Main demo Window"); app.DeleteEvent += Window_Delete; app.ShowAll (); program.Run (); } private void Window_Delete (object o, DeleteEventArgs args) { Application.Quit (); } } Save the code as demo.cs and compile with : gmcs demo.cs -pkg:gtk-sharp-2.0,gnome-sharp-2.0 Launch the program in a console : mono demo.exe Close the opened window and you will see a "Segmentation fault" message. If you use only Gtk-sharp (no Gnome-sharp related stuffs) you do not have the segfault :-( Anybody confirms my investigation ? I am really new to mono and I do not want to say silly things on mono's bug tracker : I prefere say them here ;-) I hope it helped ! Cheers, Chicha
(In reply to comment #4) > Anybody confirms my investigation ? > I am really new to mono and I do not want to say silly things on mono's bug > tracker : I prefere say them here ;-) ACK that. --($:~/Source)-- mono demo.exe Speicherzugriffsfehler ==> ./kern.log <== Feb 22 07:27:14 winnb000488 mono[8795]: segfault at b791da28 eip b791da28 esp bfdba37c error 4 > I hope it helped ! At least I know that I'm not all alone :) Do you also get a message in the syslog when you exit your tiny program? What kernel are you using? What distribution? Best regards, Alexander
Yes, I have the kernel error too : Feb 22 08:11:49 Grogros mono[10237]: segfault at b78f7848 eip b78f7848 esp bf92884c error 4 I am using Archlinux with kernel 2.6.24.1 (with patches from upstream to fix the recent security issue).
Hi all ! I have opened a bug upstream at Novell's : https://bugzilla.novell.com/show_bug.cgi?id=363948 Lets see what they think about this bug. I will keep you inform of their investigations. ++ Chicha
Confirmed in gtk-sharp2-2.10.2. This problem doesn't happen in gtk-sharp2-2.10.3. Please See : https://bugzilla.redhat.com/show_bug.cgi?id=434609
(In reply to comment #8) > Confirmed in gtk-sharp2-2.10.2. > This problem doesn't happen in gtk-sharp2-2.10.3. True for me as well. This also fixed the segfault-upon-exit of f-spot.
Since it is an upstream bug and there is a solution (upgrade to latest gtk-sharp) I think this bug can be closed :-) Thanks to all of you for your help :-) ++ Chicha
thanks for your work