After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 138997 - Epiphany crashes at startup
Epiphany crashes at startup
Status: RESOLVED FIXED
Product: gtk+
Classification: Platform
Component: Class: UIManager / Actions
2.4.x
Other Linux
: High critical
: ---
Assigned To: gtk-bugs
gtk-bugs
: 139865 142826 145514 146230 146598 146742 146890 146919 147082 147112 147170 147323 149551 149852 151993 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2004-04-04 01:10 UTC by Cody Russell
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
Backtrace (1.42 KB, text/plain)
2004-04-04 01:11 UTC, Cody Russell
  Details
possible fix (869 bytes, patch)
2004-04-05 15:24 UTC, Christian Persch
none Details | Review
updated patch, also fixes disconnection (1.94 KB, patch)
2004-04-09 10:03 UTC, Christian Persch
accepted-commit_now Details | Review

Description Cody Russell 2004-04-04 01:10:16 UTC
Start Ephiphany...

*boom*
Comment 1 Cody Russell 2004-04-04 01:11:36 UTC
Created attachment 26301 [details]
Backtrace
Comment 2 Cody Russell 2004-04-04 01:15:13 UTC
I was told it's better to put the backtrace in the bug report itself rather than
as an attachment, so here it is:

(gdb) run
Starting program: /usr/bin/epiphany-bin
(no debugging symbols found)...
Program received signal SIGSEGV, Segmentation fault.
0x0000002a99ea1e21 in strncmp () from /lib/libc.so.6
(gdb) bt
  • #0 strncmp
    from /lib/libc.so.6
  • #1 g_object_connect
    from /usr/lib/libgobject-2.0.so.0
  • #2 gtk_ui_manager_insert_action_group
    from /usr/lib/libgtk-x11-2.0.so.0
  • #3 ephy_window_get_type
  • #4 ephy_window_get_type
  • #5 g_type_create_instance
    from /usr/lib/libgobject-2.0.so.0
  • #6 g_object_new_valist
    from /usr/lib/libgobject-2.0.so.0
  • #7 g_object_newv
    from /usr/lib/libgobject-2.0.so.0
  • #8 g_object_new_valist
    from /usr/lib/libgobject-2.0.so.0
  • #9 g_object_new
    from /usr/lib/libgobject-2.0.so.0
  • #10 ephy_window_new
  • #11 ephy_shell_new_tab
  • #12 ephy_nautilus_view_new_component
  • #13 _ORBIT_skel_small_GNOME_EphyAutomation_loadurl
  • #14 ORBit_c_stub_invoke
    from /usr/lib/libORBit-2.so.0
  • #15 GNOME_EphyAutomation_loadurl
  • #16 ephy_shell_get_type
  • #17 ephy_shell_startup
  • #18 main

Comment 3 Cody Russell 2004-04-04 01:18:44 UTC
By the way, this is while running on AMD64 in Gentoo.  I mentioned this when
filing the bug report initially, but I don't see it on here now and I want to
make sure you know since it could be the issue.
Comment 4 Christian Persch 2004-04-04 10:23:11 UTC
You said you're starting epiphany, but the trace has
  • #12 ephy_nautilus_view_new_component

Can you please recompile epiphany (and glib and gtk+) with debugging symbols to
get a better trace?
Comment 5 Marco Pesenti Gritti 2004-04-04 10:38:41 UTC
I cant look at the code right now. Maybe ephy_nautilus_view_new_component is 
just used for the registration of the bonobo component ? In that case it's 
done every time at startup, not only when creating a view object for real.
Comment 6 Cody Russell 2004-04-04 22:26:01 UTC
What do you mean "recompile with debugging symbols"?  Those are debugging
symbols up there.

To get this backtrace, I'm running "gdb epiphany-bin".  I'm not executing it
from within Nautilus or anything.
Comment 7 Christian Persch 2004-04-05 10:12:17 UTC
> What do you mean "recompile with debugging symbols"?  Those are debugging
> symbols up there.

I mean with full debuggin infos, so that the trace shows the function call
arguments and the source code line numbers.
Comment 8 Cody Russell 2004-04-05 14:07:28 UTC
I'm not sure off hand how to build that in.  What additional CFLAGS should I
use?  I currently have my CFLAGS set to "-g3"
Comment 9 Cody Russell 2004-04-05 14:56:08 UTC
My apologies.. I just learned how to get Gentoo to not strip stuff.  So here is
a better backtrace for you.

(gdb) run
Starting program: /usr/bin/epiphany-bin
 
Program received signal SIGSEGV, Segmentation fault.
0x0000002a99ea1e21 in strncmp () from /lib/libc.so.6
(gdb) bt
  • #0 strncmp
    from /lib/libc.so.6
  • #1 g_object_connect
    at gobject.c line 1349
  • #2 gtk_ui_manager_insert_action_group
    at gtkuimanager.c line 583
  • #3 setup_ui_manager
    at ephy-window.c line 940
  • #4 ephy_window_init
    at ephy-window.c line 1905
  • #5 g_type_create_instance
    at gtype.c line 1595
  • #6 g_object_constructor
    at gobject.c line 1044
  • #7 g_object_newv
    at gobject.c line 941
  • #8 g_object_new_valist
    at gobject.c line 984
  • #9 g_object_new
    at gobject.c line 822
  • #10 ephy_window_new
    at ephy-window.c line 2068
  • #11 ephy_shell_new_tab
    at ephy-shell.c line 577
  • #12 impl_ephy_automation_loadurl
    at ephy-automation.c line 100
  • #13 _ORBIT_skel_small_GNOME_EphyAutomation_loadurl
    at EphyAutomation-common.c line 32
  • #14 ORBit_c_stub_invoke
    from /usr/lib/libORBit-2.so.0
  • #15 GNOME_EphyAutomation_loadurl
    at EphyAutomation-stubs.c line 23
  • #16 open_urls
    at ephy-shell.c line 237
  • #17 ephy_shell_startup
    at ephy-shell.c line 414
  • #18 main
    at ephy-main.c line 173

Comment 10 Christian Persch 2004-04-05 15:24:14 UTC
Created attachment 26358 [details] [review]
possible fix

Reporter: can you please try again after applying the attached patch to gtk+
and rebuilding that? Thanks.
Comment 11 Christian Persch 2004-04-08 14:05:24 UTC
Ping?
Comment 12 Cody Russell 2004-04-09 04:44:25 UTC
Sorry, I was out of town for a few days.  I applied this patch and it appears to
have fixed the problem.  Epiphany started up correctly.
Comment 13 Cody Russell 2004-04-09 06:32:56 UTC
Okay, and also do the same thing to line 622 for the disconnection.  This fixes
a crasher that occurs when trying to add bookmarks.
Comment 14 Christian Persch 2004-04-09 10:02:26 UTC
Thanks for testing it, and for catching the disconnect problem too. Reopening...
Comment 15 Christian Persch 2004-04-09 10:03:37 UTC
Created attachment 26513 [details] [review]
updated patch, also fixes disconnection

... attaching updated patch ...
Comment 16 Christian Persch 2004-04-09 10:04:59 UTC
... and reassigning to gtk+.
Comment 17 Owen Taylor 2004-04-09 13:27:47 UTC
Comment on attachment 26513 [details] [review]
updated patch, also fixes disconnection

Looks good to commit
Comment 18 Christian Persch 2004-04-09 13:44:09 UTC
2004-04-09  Christian Persch  <chpe@cvs.gnome.org>

        * gtk/gtkuimanager.c: (gtk_ui_manager_insert_action_group),
        (gtk_ui_manager_remove_action_group): Terminate
        g_object_[dis]connect() calls with NULL instead of 0.
        Fixes #138997.
Comment 19 Christian Persch 2004-04-13 08:11:02 UTC
*** Bug 139865 has been marked as a duplicate of this bug. ***
Comment 20 Hypnos 2004-04-17 22:47:59 UTC
This patch also helps with Rhythmbox 8.0 when importing a folder (previously
crashing with null/0 confusion), on my Pentium 4.
Comment 21 Christian Persch 2004-05-20 10:51:21 UTC
*** Bug 142826 has been marked as a duplicate of this bug. ***
Comment 22 Tommi Komulainen 2004-07-06 17:27:34 UTC
*** Bug 145514 has been marked as a duplicate of this bug. ***
Comment 23 Christian Persch 2004-07-09 13:49:57 UTC
*** Bug 146230 has been marked as a duplicate of this bug. ***
Comment 24 Christian Persch 2004-07-09 14:18:30 UTC
*** Bug 146598 has been marked as a duplicate of this bug. ***
Comment 25 Christian Persch 2004-07-09 14:37:08 UTC
*** Bug 146919 has been marked as a duplicate of this bug. ***
Comment 26 Christian Persch 2004-07-09 15:59:21 UTC
*** Bug 147082 has been marked as a duplicate of this bug. ***
Comment 27 Christian Persch 2004-07-09 17:21:39 UTC
*** Bug 147112 has been marked as a duplicate of this bug. ***
Comment 28 Christian Persch 2004-07-09 17:26:42 UTC
*** Bug 147170 has been marked as a duplicate of this bug. ***
Comment 29 Christian Persch 2004-07-10 12:00:17 UTC
*** Bug 146742 has been marked as a duplicate of this bug. ***
Comment 30 Christian Persch 2004-07-11 12:54:31 UTC
*** Bug 147323 has been marked as a duplicate of this bug. ***
Comment 31 Crispin Flowerday (not receiving bugmail) 2004-07-11 19:28:23 UTC
*** Bug 146890 has been marked as a duplicate of this bug. ***
Comment 32 Christian Persch 2004-08-07 11:29:06 UTC
*** Bug 149551 has been marked as a duplicate of this bug. ***
Comment 33 Christian Persch 2004-08-10 23:16:41 UTC
*** Bug 149852 has been marked as a duplicate of this bug. ***
Comment 34 Christian Persch 2004-09-06 17:55:04 UTC
*** Bug 151993 has been marked as a duplicate of this bug. ***