Bug 651099 – [regression] hang in name-completion in git-master

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 651099 - [regression] hang in name-completion in git-master


Summary:	[regression] hang in name-completion in git-master


Status:	RESOLVED DUPLICATE of bug 651147

Product:	evolution
Classification:	Applications
Component:	Mailer
Version:	3.2.x (obsolete)
Hardware:	Other Linux

Importance:	Normal critical
Target Milestone:	---
Assigned To:	evolution-mail-maintainers
QA Contact:	Evolution QA team

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2011-05-25 22:05 UTC by David Ronis
Modified:	2013-09-13 01:06 UTC

See Also:
GNOME target:	---
GNOME version:	---

Attachments
interim eds fix (4.38 KB, patch) 2011-05-26 07:23 UTC, Milan Crha	rejected	Details \| Review

Description David Ronis 2011-05-25 22:05:45 UTC

I've just updated the git master and tried sending a short message (one created when I hit send-link in my browser (firefox)).   Evolution hangs (both the compose message UI and main evolution UI).

I attached gdb and got the following backtrace:

(gdb) thread apply all bt full

+ Trace 227258

Comment 1 Milan Crha 2011-05-26 07:19:04 UTC

Thanks for a bug report. This is caused due to main loop interleaving. The new DBus templates are running their own main loops while waiting for a "done" signal from the server, where is delivered a timeout from the entry completion and another invocation of the same is initiated, so there are actually three main loops in the stack now (the third is hidden in the Thread 1 in the top).

Comment 2 Milan Crha 2011-05-26 07:23:21 UTC

Created attachment 188635 [details] [review]
interim eds fix

for evolution-data-server;

I tried to repeat this, but I was unable to, I do not know why exactly, maybe just a coincidence. This patch makes sure that the timeout callbacks initiated by ENameSelectorEntry aren't delivered in the stacked main loops. The real fix will be when it stops using EBook and starts using EBookClient instead, but till I get to that change I would like to have this working a bit better (and the real fix is supposed to come to both eds and evolution mailer's part for contact photo search and "is in address book" checker).

Comment 3 Milan Crha 2011-05-26 07:25:09 UTC

Just to mention, I see lacks of response while working with autocompletion, but it usually recovers after few seconds (mostly within 2-3 seconds).

Comment 4 David Ronis 2011-05-27 01:44:10 UTC

I applied the patch to the git/master.   It doesn't work.   Now I get hung on a 

formatting message...  which won't go away even when I explicitly cancel it (via the red icon in the notification area).  I'm not 100% sure that this is the same  bug.
I attached gdb and got the following backtrace:

(gdb) thread apply all bt full

+ Trace 227279

Comment 5 David Ronis 2011-05-30 18:10:23 UTC

I just updated evo and friends to the lastest git-master.    Just to try something different, I started evolution as:

evolution -c calendar


The UI never maps!   This is now a blocker for me, since I can't read my e-mail any longer.
I've attached gdb and got the following backtrace:

(gdb) thread apply all bt full

+ Trace 227323

Thread 1 (Thread 0xb5a00860 (LWP 3699))

#0 nanosleep
from /lib/libpthread.so.0
#1 g_usleep
at gtimer.c line 253
#2 gdbus_proxy_call_sync
at e-gdbus-templates.c line 1260
#3 e_gdbus_cal_call_get_backend_property_sync
at e-gdbus-cal.c line 302
#4 set_local_attachment_store
at e-cal.c line 1085
#5 e_cal_new
at e-cal.c line 1189
#6 e_auth_new_cal_from_source
at authentication.c line 145
#7 e_memo_shell_sidebar_add_source
at e-memo-shell-sidebar.c line 870
#8 memo_shell_sidebar_row_changed_cb
at e-memo-shell-sidebar.c line 350
#9 _gtk_marshal_VOID__BOXED_BOXED
at gtkmarshalers.c line 1310
#10 g_closure_invoke
at gclosure.c line 771
#11 signal_emit_unlocked_R
at gsignal.c line 3256
#12 g_signal_emit_valist
at gsignal.c line 2987
#13 g_signal_emit
at gsignal.c line 3044
#14 gtk_tree_model_row_changed
at gtktreemodel.c line 1757
#15 e_source_selector_select_source
at e-source-selector.c line 1568
#16 memo_shell_sidebar_restore_state_cb
#17 g_cclosure_marshal_VOID__OBJECT
at gmarshal.c line 638
#18 g_closure_invoke
#19 signal_emit_unlocked_R
at gsignal.c line 3256
#20 g_signal_emit_valist
at gsignal.c line 2987
#21 g_signal_emit
at gsignal.c line 3044
#22 e_shell_window_get_shell_view
at e-shell-window.c line 1049
#23 cal_shell_content_constructed
at e-cal-shell-content.c line 356
#24 g_object_newv
at gobject.c line 1521
#25 g_object_new_valist
at gobject.c line 1610
#26 g_object_new
at gobject.c line 1325
#27 e_cal_shell_content_new
at e-cal-shell-content.c line 666
#28 shell_view_constructed
at e-shell-view.c line 589
#29 cal_shell_view_constructed
at e-cal-shell-view.c line 53
#30 g_object_newv
at gobject.c line 1521
#31 g_object_new_valist
at gobject.c line 1610
#32 g_object_new
at gobject.c line 1325
#33 shell_window_create_shell_view
at e-shell-window.c line 642
#34 e_shell_window_get_shell_view
at e-shell-window.c line 1047
#35 e_shell_window_set_active_view
at e-shell-window.c line 1289
#36 shell_window_set_property
at e-shell-window.c line 198
#37 object_set_property
at gobject.c line 1199
#38 g_object_set_property
#39 prop_binding_sync_pref_to_prop
at gconf-bridge.c line 242
#40 gconf_bridge_bind_property_full
at gconf-bridge.c line 540
#41 e_shell_window_private_constructed
at e-shell-window-private.c line 430
#42 shell_window_constructed
#43 g_object_newv
at gobject.c line 1521
#44 g_object_new_valist
at gobject.c line 1610
#45 g_object_new
at gobject.c line 1325
#46 e_shell_window_new
at e-shell-window.c line 989
#47 e_shell_create_shell_window
at e-shell.c line 1515
#48 idle_cb
at main.c line 263
#49 g_idle_dispatch
#50 g_main_dispatch
at gmain.c line 2477
#51 g_main_context_dispatch
at gmain.c line 3050
#52 g_main_context_iterate
at gmain.c line 3128
#53 g_main_loop_run
at gmain.c line 3336
#54 gtk_main
at gtkmain.c line 1358
#55 main
at main.c line 691

Comment 6 David Ronis 2011-05-30 19:47:08 UTC

One clarification: the UI still maps if I start in mail mode. Nonetheless, Ican't read any e-mail. No matter wich folder I click on, I get "Formatting message..." in the notification area, which never completes.

There's a fair bit of console output as well:

(evolution:22020): evolution-mail-WARNING **: Couldn't get service: David.Ronis@McGill.CA: No provider available for protocol ''

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate EMailShellContent's child EAlertBar 0x8449da0. Allocation is 786x0, but minimum required size is 0x45.

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate EAlertBar's child GtkBox 0x8453538. Allocation is 786x1, but minimum required size is 0x45.

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate GtkBox's child GtkVBox 0x84535b8. Allocation is 786x1, but minimum required size is 0x45.

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate EMailShellContent's child EAlertBar 0x8449da0. Allocation is 786x0, but minimum required size is 0x45.

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate EMailShellContent's child EAlertBar 0x8449da0. Allocation is 970x0, but minimum required size is 0x45.

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate EAlertBar's child GtkBox 0x8453538. Allocation is 970x1, but minimum required size is 0x45.

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate GtkBox's child GtkVBox 0x84535b8. Allocation is 970x1, but minimum required size is 0x45.

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate EMailShellContent's child EAlertBar 0x8449da0. Allocation is 970x0, but minimum required size is 0x45.

(evolution:22020): GLib-WARNING **: GError set over the top of a previous GError or uninitialized memory.
This indicates a bug in someone's code. You must ensure an error is NULL before it's set.
The overwriting error message was: no such column: msg_security

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate EMailShellContent's child EAlertBar 0x8449da0. Allocation is 970x0, but minimum required size is 0x45.

(evolution:22020): camel-WARNING **: Could not open converter for 'unknown' to 'UTF-8' charset

(evolution:22020): camel-WARNING **: Could not find key entry for word 'shhhhhhhhhhh': No such file or directory

(evolution:22020): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to underallocate EMailShellContent's child EAlertBar 0x8449da0. Allocation is 970x0, but minimum required size is 0x45.

(evolution:22020): camel-WARNING **: Could not find key entry for word '0x00000000': No such file or directory

(evolution:22020): camel-WARNING **: Could not find key entry for word 'd0000000': No such file or directory

(evolution:22020): camel-WARNING **: Could not find key entry for word '0000000100000000': No such file or directory

(evolution:22020): camel-WARNING **: Could not find key entry for word '00000000': No such file or directory