Closed Bug 71833 Opened 25 years ago Closed 25 years ago

GTK related crash

Categories

(Core :: XUL, defect)

x86
Linux
defect
Not set
critical

Tracking

()

RESOLVED DUPLICATE of bug 72069

People

(Reporter: spam, Assigned: pavlov)

Details

(Keywords: crash)

CVS build from march 13th, 03:30 I've had many GTK related crashes the past 24 hours or so, 4 different builds from this timespan all had them. Hard to say what exactly triggers it. Has happened on startup, when back-button was clicked, when link was clicked, when i hit ctrl+n. Program received signal SIGSEGV, Segmentation fault. 0x40b2a08e in handle_gdk_event () from /home/dark/DISK/mozilla/dist/bin/components/libwidget_gtk.so (gdb) bt #0 0x40b2a08e in handle_gdk_event () from /home/dark/DISK/mozilla/dist/bin/components/libwidget_gtk.so #1 0x4057053b in gdk_event_dispatch () from /usr/lib/libgdk-1.2.so.0 #2 0x4059d186 in g_main_dispatch () from /usr/lib/libglib-1.2.so.0 #3 0x4059d751 in g_main_iterate () from /usr/lib/libglib-1.2.so.0 #4 0x4059d8f1 in g_main_run () from /usr/lib/libglib-1.2.so.0 #5 0x404c57b9 in gtk_main () from /usr/lib/libgtk-1.2.so.0 #6 0x40b22630 in nsAppShell::Run () from /home/dark/DISK/mozilla/dist/bin/components/libwidget_gtk.so #7 0x410e647a in nsAppShellService::Run () from /home/dark/DISK/mozilla/dist/bin/components/libnsappshell.so #8 0x804dc60 in main1 () #9 0x804e465 in main () #10 0x4024b9cb in __libc_start_main (main=0x804e338 <main>, argc=1, argv=0xbffff5e4, init=0x804ae24 <_init>, fini=0x80546cc <_fini>, rtld_fini=0x4000aea0 <_dl_fini>, stack_end=0xbffff5dc) at ../sysdeps/generic/libc-start.c:92 ------ Now and then an error display in console sometime before crash - samples: Gtk-WARNING **: invalid cast from `(unknown)' to `GtkObject' Gtk-CRITICAL **: file gtkwidget.c: line 2725 (gtk_widget_event): assertion `GTK_IS_WIDGET (widget)' failed Not seen the problem before. I build moz fairly often. RH6.2/Gtk+ 1.2.8
It happens more "reliably" in connection with window-creation.
Adding crash keyword.
Keywords: crash
cc: blizzard, since it happens in event code (which may have nothing to do with his recent checkin).
->pav, gtk module owner
Assignee: trudelle → pavlov
when changing focus from/to a newly created loading window, i often see one of these: Gtk-CRITICAL **: file gtkmain.c: line 1279 (gtk_propagate_event): assertion `GTK_IS_WIDGET (widget)' failed. Gtk-CRITICAL **: file gtkwidget.c: line 2725 (gtk_widget_event): assertion `GTK_IS_WIDGET (widget)' failed.
FWIW: only using common build param's: ac_add_options --disable-tests ac_add_options --disable-debug ac_add_options --enable-optimize
Downloaded mozilla-i686-pc-linux-gnu-sea.tar.gz 2001-031321 as a sanity check: The errors and crash happens there too.
My build on March 14th doesn't get these errors. Also, my event patch wouldn't cause these errors. That path isn't in the stack trace provided.
simply hitting ctrl+n causes it. Browser set to load blank page as homepage. hit ctrl+n from a window i'd been browsing bugzilla with for some minutes. Got: Gtk-CRITICAL **: file gtkwidget.c: line 2725 (gtk_widget_event): assertion `GTK_IS_WIDGET (widget)' failed. Close that window. Hit ctrl+n again. Got: Gtk-CRITICAL **: file gtkmain.c: line 1279 (gtk_propagate_event): assertion `GTK_IS_WIDGET (widget)' failed. Close that too, hit ctrl+n again. Got: Gtk-WARNING **: invalid class type `(unknown)' in cast to `GtkObject' (All three times the window was created, partly with additional error message "Error loading URL about:blank: 804b0002". Closing the window and ctrl+n again spawns: Error loading URL about:blank: 804b0002 Document about:blank loaded successfully Gtk-WARNING **: invalid cast from `(unknown)' to `GtkObject' As i test this, error repeats 100% consistantly, but only the last error goes on displaying when i continue out over the third attempt.
I see the assertions under a clean KDE 2.1 for RH6.2 as well.
This crash happens so frequently here it smells like a messed up gtk or glib install. Upgraded to 1.2.9 now, rebuilding moz from scratch. Since nobody else see these crashes: Resolving as invalid.
Status: NEW → RESOLVED
Closed: 25 years ago
Resolution: --- → INVALID
Okay, thanks.
Status: RESOLVED → VERIFIED
Oh well.. just so others don't waste their time upgrading: it didn't help.
Should be marked dup
Status: VERIFIED → REOPENED
Resolution: INVALID → ---
*** This bug has been marked as a duplicate of 72069 ***
Status: REOPENED → RESOLVED
Closed: 25 years ago25 years ago
Resolution: --- → DUPLICATE
You need to log in before you can comment on or make changes to this bug.