GNOME Bugzilla – Bug 589568
Crash in model_row_changed_cb at e-tasks.c line 353
Last modified: 2010-09-14 10:21:20 UTC
What were you doing when the application crashed? checked TODO (mark as done) Distribution: Gentoo Base System release 2.0.1 Gnome Release: 2.26.3 2009-07-13 (Gentoo) BugBuddy Version: 2.26.0 System: Linux 2.6.29-tuxonice-r2 #1 SMP Mon May 18 22:30:28 CEST 2009 i686 X Vendor: The X.Org Foundation X Vendor Release: 10602000 Selinux: No Accessibility: Disabled GTK+ Theme: Clearlooks Icon Theme: gnome GTK+ Modules: canberra-gtk-module, gnomebreakpad Memory status: size: 328949760 vsize: 328949760 resident: 104017920 share: 30855168 rss: 104017920 rss_rlim: 18446744073709551615 CPU usage: start_time: 1248351843 rtime: 33682 utime: 30161 stime: 3521 cutime:119 cstime: 590 timeout: 0 it_real_value: 0 frequency: 100 Backtrace was generated from '/usr/bin/evolution' [Thread debugging using libthread_db enabled] [New Thread 0xb5071720 (LWP 3232)] [New Thread 0xa3efcb70 (LWP 17531)] [New Thread 0xa7aafb70 (LWP 17786)] [New Thread 0xa56ffb70 (LWP 17785)] [New Thread 0xb2b9fb70 (LWP 4638)] [New Thread 0xaa471b70 (LWP 4636)] [New Thread 0xa846db70 (LWP 3289)] [New Thread 0xa8c6eb70 (LWP 3288)] [New Thread 0xaac72b70 (LWP 3283)] [New Thread 0xab473b70 (LWP 3282)] [New Thread 0xabc74b70 (LWP 3281)] [New Thread 0xac475b70 (LWP 3280)] [New Thread 0xadee9b70 (LWP 3279)] [New Thread 0xad6e8b70 (LWP 3278)] [New Thread 0xae6eab70 (LWP 3276)] [New Thread 0xaeeebb70 (LWP 3275)] [New Thread 0xaf6ecb70 (LWP 3274)] [New Thread 0xafef9b70 (LWP 3273)] [New Thread 0xb1efdb70 (LWP 3269)] [New Thread 0xb26feb70 (LWP 3268)] [New Thread 0xb402ab70 (LWP 3264)] [New Thread 0xb482bb70 (LWP 3262)] 0xb7f12424 in __kernel_vsyscall ()
+ Trace 216597
Thread 1 (Thread 0xb5071720 (LWP 3232))
---- Critical and fatal warnings logged during execution ---- ** Gdk **: gdk_x11_atom_to_xatom_for_display: assertion `atom != GDK_NONE' failed ----------- .xsession-errors (60932 sec old) --------------------- ** (vino-server:22597): WARNING **: Remote Desktop server already running; exiting ... ** (vino-server:22598): WARNING **: Remote Desktop server already running; exiting ... ** (vino-server:22599): WARNING **: Remote Desktop server already running; exiting ... ** (vino-server:22600): WARNING **: Remote Desktop server already running; exiting ... ...Too much output, ignoring rest... --------------------------------------------------
Thanks for the bug report. There had been done some changes in this within 2.27, which will be available soon, in 2.28. Could you try there, please, and report back, as I'm unable to reproduce it with those changes? Thanks in advance.
*** Bug 604172 has been marked as a duplicate of this bug. ***
last dupe in 2.28.x
Just had another crash in model_row_changed_cb with latest gnome-2-28 code. I only remember having edited a calendar event in the description field right at the top. Remember having inserted a date string like "20.01.2010". Then, when I pressed the Save button, evolution segfaulted: here is the stack trace: (gdb) btaf
+ Trace 220213
Thread 1 (Thread 0xb7fc5af0 (LWP 3577))
Oh, yes, I forgot: There was the following warning: (evolution:3577): GLib-GObject-WARNING **: invalid cast from `GdkGCX11' to `ETasks' Program received signal SIGSEGV, Segmentation fault. 0x016a84b3 in model_row_changed_cb (etm=0x9504360, row=20, data=0x8d968d0) at ../../../../evolution/calendar/gui/e-tasks.c:353 (gdb) p priv $1 = (ETasksPrivate *) 0x8 (gdb) [Thread 0xb047ab70 (LWP 5914) exited] $2 = (ETasksPrivate *) 0x8 (gdb) p tasks $3 = (ETasks *) 0x8d968d0 (gdb) p *4 Cannot access memory at address 0x4 (gdb) p *$ $4 = {table = {container = {widget = {object = {parent_instance = {g_type_instance = {g_class = 0x815e640}, ref_count = 1, qdata = 0x0}, flags = 0}, private_flags = 0, state = 0 '\000', saved_state = 0 '\000', name = 0x0, style = 0x0, requisition = {width = 134836248, height = 141769144}, allocation = {x = 134840584, y = 6291456, width = 0, height = 0}, window = 0x0, parent = 0x0}, focus_child = 0x0, border_width = 0, need_resize = 0, resize_mode = 0, reallocate_redraws = 0, has_focus_chain = 0}, children = 0x0, rows = 0x0, cols = 0x0, nrows = 0, ncols = 0, column_spacing = 0, row_spacing = 0, homogeneous = 1}, priv = 0x8} (gdb) p data $5 = (gpointer) 0x8d968d0 (gdb) up
(In reply to comment #4) > Just had another crash in model_row_changed_cb with latest gnome-2-28 code. > I only remember having edited a calendar event in the description field right > at the top. Remember having inserted a date string like "20.01.2010". Then, > when > I pressed the Save button, evolution segfaulted: > > here is the stack trace: > > (gdb) btaf > > Correction: I was in the calendar view, but I edited a task!!
(In reply to comment #5) > Oh, yes, I forgot: > There was the following warning: > (evolution:3577): GLib-GObject-WARNING **: invalid cast from `GdkGCX11' to > `ETasks' This looks like an obvious programming error. Can you try to capture a backtrace where this warning is emitted? After you're up and running under gdb, halt the program with Ctrl+C, set a breakpoint on the function "g_logv", continue, and then try to reproduce the crash again.
(In reply to comment #7) > (In reply to comment #5) > > Oh, yes, I forgot: > > There was the following warning: > > (evolution:3577): GLib-GObject-WARNING **: invalid cast from `GdkGCX11' to > > `ETasks' > > This looks like an obvious programming error. Can you try to capture a > backtrace where this warning is emitted? After you're up and running under > gdb, halt the program with Ctrl+C, set a breakpoint on the function "g_logv", > continue, and then try to reproduce the crash again. Just tried, to edit the same task again in calendar view, but could not reproduce this. But thank you for the hint, I'll keep that in mind next time!
*** Bug 612521 has been marked as a duplicate of this bug. ***
*** Bug 613862 has been marked as a duplicate of this bug. ***
*** Bug 615201 has been marked as a duplicate of this bug. ***
so Thomas, any luck?
(In reply to comment #12) > so Thomas, any luck? I haven't seen the problem anymore. I don't know if it is connected to this bug, but I remember having submitted a patch to bug 544187, https://bugzilla.gnome.org/attachment.cgi?id=157239, which checks if indexes are in bounds within some sorting routines for calendar/tasks.
Just had another one of those crashes in e_table_model_row_changed Just edited a few tasks, tried to save one, and evo segfaults. gdb tells me (evolution:3513): GLib-GObject-WARNING **: invalid cast from `GdkGCX11' to `ETasks' In the following stack trace, in frame 2, marshal_data is clearly NULL, but somehow when g_cclosure_marshal_VOID__INT is called the final argument is not NULL anymore and the callback is invalid and crashes. I am running evo 2.28.4 with a few patches of mine on ubuntu karmic. (gdb) bt
+ Trace 222682
(In reply to comment #14) > Just edited a few tasks, tried to save one, and evo segfaults. gdb tells me > > (evolution:3513): GLib-GObject-WARNING **: invalid cast from `GdkGCX11' to > `ETasks' That warning is coming from code that doesn't exist anymore in Evolution 2.30. Until you can reproduce this issue in a supported version this bug probably won't get much attention.
Please feel free to reopen the bug if the problem still occurs with a newer version of Evolution 2.30.3 or later, thanks.