Bug 323522 – e-table-CRITICAL when switching component in Evolution

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 323522 - e-table-CRITICAL when switching component in Evolution


Summary:	e-table-CRITICAL when switching component in Evolution


Status:	RESOLVED FIXED

Product:	evolution
Classification:	Applications
Component:	Do Not Use
Version:	2.6.x
Hardware:	Other Linux

Importance:	High critical
Target Milestone:	---
Assigned To:	Harish Krishnaswamy
QA Contact:	Evolution QA team

URL:
Whiteboard:

Depends on:
Blocks:	327508 327510

Reported:	2005-12-08 05:02 UTC by Kjartan Maraas
Modified:	2013-09-13 12:24 UTC

See Also:
GNOME target:	---
GNOME version:	2.13/2.14

Description Kjartan Maraas 2005-12-08 05:02:13 UTC

e-table-CRITICAL **: eti_unfreeze: assertion `eti->frozen_count > 0' failed
aborting...

Program received signal SIGTRAP, Trace/breakpoint trap.
IA__g_logv (log_domain=Variable "log_domain" is not available.
) at gmessages.c:503
503               g_private_set (g_log_depth, GUINT_TO_POINTER (depth));
(gdb) bt

+ Trace 64481

#0 IA__g_logv
at gmessages.c line 503
#1 IA__g_log
at gmessages.c line 517
#2 IA__g_return_if_fail_warning
at gmessages.c line 532
#3 eti_unfreeze
at e-table-item.c line 1062
#4 eti_table_model_changed
at e-table-item.c line 1112
#5 IA__g_cclosure_marshal_VOID__VOID
at gmarshal.c line 77
#6 IA__g_closure_invoke
at gclosure.c line 490
#7 signal_emit_unlocked_R
at gsignal.c line 2449
#8 IA__g_signal_emit_valist
at gsignal.c line 2208
#9 IA__g_signal_emit
at gsignal.c line 2252
#10 e_table_model_changed
at e-table-model.c line 460
#11 ets_proxy_model_rows_deleted
at e-table-sorted.c line 289
#12 etss_proxy_model_rows_deleted
at e-table-subset.c line 389
#13 e_util_marshal_VOID__INT_INT
at e-util-marshal.c line 813
#14 IA__g_closure_invoke
at gclosure.c line 490
#15 signal_emit_unlocked_R
at gsignal.c line 2449
#16 IA__g_signal_emit_valist
at gsignal.c line 2208
#17 IA__g_signal_emit
at gsignal.c line 2252
#18 e_table_model_rows_deleted
at e-table-model.c line 586
#19 e_table_model_row_deleted
at e-table-model.c line 603
#20 e_cal_view_objects_modified_cb
at e-cal-model.c line 1419
#21 IA__g_cclosure_marshal_VOID__POINTER
at gmarshal.c line 601
#22 IA__g_closure_invoke
at gclosure.c line 490
#23 signal_emit_unlocked_R
at gsignal.c line 2449
#24 IA__g_signal_emit_valist
at gsignal.c line 2208
#25 IA__g_signal_emit
at gsignal.c line 2252
#26 objects_modified_cb
at e-cal-view.c line 90
#27 IA__g_cclosure_marshal_VOID__POINTER
at gmarshal.c line 601
#28 IA__g_closure_invoke
at gclosure.c line 490
#29 signal_emit_unlocked_R
at gsignal.c line 2449
#30 IA__g_signal_emit_valist
at gsignal.c line 2208
#31 IA__g_signal_emit
at gsignal.c line 2252
#32 impl_notifyObjectsModified
at e-cal-view-listener.c line 157
#33 _ORBIT_skel_small_GNOME_Evolution_Calendar_CalViewListener_notifyObjectsModified
at Evolution-DataServer-Calendar-common.c line 24
#34 ORBit_POAObject_invoke
at poa.c line 1145
#35 ORBit_OAObject_invoke
at orbit-adaptor.c line 336
#36 ORBit_small_invoke_adaptor
at orbit-small.c line 835
#37 ORBit_POAObject_handle_request
at poa.c line 1354
#38 ORBit_POAObject_invoke_incoming_request
at poa.c line 1422
#39 giop_thread_queue_process
at giop.c line 774
#40 giop_recv_buffer_get
at giop-recv-buffer.c line 705
#41 ORBit_small_invoke_stub
at orbit-small.c line 657
#42 ORBit_small_invoke_stub_n
at orbit-small.c line 575
#43 ORBit_c_stub_invoke
at poa.c line 2644
#44 ConfigDatabase_set
at GConfX-stubs.c line 114
#45 gconf_engine_set
at gconf.c line 1259
#46 error_checked_set
at gconf.c line 3318
#47 gconf_client_set_string
at gconf-client.c line 1722
#48 sidebar_button_selected_callback
at e-shell-window.c line 311
#49 IA__g_cclosure_marshal_VOID__INT
at gmarshal.c line 216
#50 IA__g_closure_invoke
at gclosure.c line 490
#51 signal_emit_unlocked_R
at gsignal.c line 2449
#52 IA__g_signal_emit_valist
at gsignal.c line 2208
#53 IA__g_signal_emit
at gsignal.c line 2252
#54 e_sidebar_select_button
at e-sidebar.c line 537
#55 e_shell_window_switch_to_component
at e-shell-window.c line 939
#56 menu_component_selected
at e-shell-window.c line 588
#57 marshal_VOID__USER_DATA_STRING
at bonobo-ui-component.c line 221
#58 IA__g_closure_invoke
at gclosure.c line 490
#59 bonobo_closure_invoke_va_list
at bonobo-types.c line 404
#60 bonobo_closure_invoke
at bonobo-types.c line 467
#61 impl_Bonobo_UIComponent_execVerb
at bonobo-ui-component.c line 158
#62 _ORBIT_skel_small_Bonobo_UIComponent_execVerb
at Bonobo-common.c line 332
#63 ORBit_c_stub_invoke
at poa.c line 2631
#64 Bonobo_UIComponent_execVerb
at Bonobo-stubs.c line 544
#65 impl_emit_verb_on
at bonobo-ui-engine.c line 1533
#66 IA__g_cclosure_marshal_VOID__POINTER
at gmarshal.c line 601
#67 g_type_class_meta_marshal
at gclosure.c line 567
#68 IA__g_closure_invoke
at gclosure.c line 490
#69 signal_emit_unlocked_R
at gsignal.c line 2487
#70 IA__g_signal_emit_valist
at gsignal.c line 2208
#71 IA__g_signal_emit
at gsignal.c line 2252
#72 bonobo_ui_engine_emit_verb_on_w
at bonobo-ui-engine.c line 3040
#73 exec_verb_cb
at bonobo-ui-sync-menu.c line 445
#74 IA__g_cclosure_marshal_VOID__VOID
at gmarshal.c line 77
#75 IA__g_closure_invoke
at gclosure.c line 490
#76 signal_emit_unlocked_R
at gsignal.c line 2449
#77 IA__g_signal_emit_valist
at gsignal.c line 2208
#78 IA__g_signal_emit
at gsignal.c line 2252
#79 closure_accel_activate
at gtkwidget.c line 3151
#80 IA__g_closure_invoke
at gclosure.c line 490
#81 signal_emit_unlocked_R
at gsignal.c line 2449
#82 IA__g_signal_emit_valist
at gsignal.c line 2218
#83 IA__g_signal_emit
at gsignal.c line 2252
#84 IA__gtk_accel_group_activate
at gtkaccelgroup.c line 766
#85 IA__gtk_accel_groups_activate
at gtkaccelgroup.c line 804
#86 IA__gtk_window_activate_key
at gtkwindow.c line 7506
#87 gtk_window_key_press_event
at gtkwindow.c line 4543
#88 bonobo_window_key_press_event
at bonobo-window.c line 267
#89 _gtk_marshal_BOOLEAN__BOXED
at gtkmarshalers.c line 83
#90 g_type_class_meta_marshal
at gclosure.c line 567
#91 IA__g_closure_invoke
at gclosure.c line 490
#92 signal_emit_unlocked_R
at gsignal.c line 2487
#93 IA__g_signal_emit_valist
at gsignal.c line 2218
#94 IA__g_signal_emit
at gsignal.c line 2252
#95 gtk_widget_event_internal
at gtkwidget.c line 3735
#96 IA__gtk_propagate_event
at gtkmain.c line 2143
#97 IA__gtk_main_do_event
at gtkmain.c line 1406
#98 gdk_event_dispatch
at gdkevents-x11.c line 2291
#99 IA__g_main_context_dispatch
at gmain.c line 1913
#100 g_main_context_iterate
at gmain.c line 2544
#101 IA__g_main_loop_run
at gmain.c line 2748
#102 bonobo_main
at bonobo-main.c line 312
#103 main
at main.c line 602
#104 __libc_start_main
at ../sysdeps/generic/libc-start.c line 231
#105 _start

Comment 1 Poornima 2005-12-08 08:11:27 UTC

Does this happen always ? R u doing any specific operation before switching
components in evolution.

This is not reproducible on suse 10.0 evolution 2.5.2.

Comment 2 Kjartan Maraas 2005-12-08 10:29:33 UTC

It happened when switching back to the mail component after going between
calendar/addressbook etc. I used ctrl+f<123> to switch between them

Comment 3 Kjartan Maraas 2005-12-18 11:32:10 UTC

Reopening this since the last comment.

Comment 4 Karsten Bräckelmann 2006-01-14 23:53:36 UTC

*sigh* This most likely is not even a crasher on SuSE 10, Evo 2.5.2.

Please not this is a "critical warning" bug, which is a crasher since 01/01/2006 in CVS. Adjusting Priority.

Comment 5 Karsten Bräckelmann 2006-02-26 02:28:52 UTC

I just found a way to reproduce this myself.

* Enabling a remote Task List (checking the checkbox next to its name in the tree) reproducibly crashes Evo for me. Stacktrace looks identical (top-most 40-50 lines at least, didn't check further). The task list seems to be completely displayed before the crash. (Could this be a red herring, err, a cached version from 2.4?)

* When starting Evo with that remote Task List still enabled it crashes again. It does not crash, if I quickly uncheck it.


During my early reproduction testing I got 2 stacktraces with interesting NULL pointers in different locations. etss and param_values, IIRC, somewhere hidden in the stacktrace within the first 40 lines. Can't reproduce them, though.


Evolution 2.5.91+, GNOME 2.13.91+

Comment 6 Karsten Bräckelmann 2006-02-28 20:53:13 UTC

Reassigning some important 2.5.x bugs to Varadhan.

Comment 7 Kjartan Maraas 2006-03-07 12:40:06 UTC

This is still biting me here.

static void
eti_unfreeze (ETableItem *eti)
{
        g_return_if_fail (eti->frozen_count > 0);

Can we change this to a test/return statement, or should the callers make sure that this never happens?

        eti->frozen_count --;
        d(g_print ("%s: %d\n", __FUNCTION__, eti->frozen_count));
        if (eti->frozen_count == 0 && eti->queue_show_cursor) {
                eti_show_cursor (eti, 0);
                eti_check_cursor_bounds (eti);
                eti->queue_show_cursor = FALSE;
        }
}

Comment 8 Kjartan Maraas 2006-05-15 12:08:44 UTC

Has anyone looked at this since last time?

Comment 9 André Klapper 2006-06-17 14:48:56 UTC

obviously not. code in /widgets/table/e-table-item.c hasn't changed. retargetting.

Comment 10 Kjartan Maraas 2006-08-29 20:56:36 UTC

I put this in one of my cleanup patches that went to evolution-patches@gnome.org today.

Comment 11 Srinivasa Ragavan 2007-07-27 08:55:53 UTC

I think it is fine to replace with if/return. Committed to head.