Bug 732148 – evolution-3.12.3: evolution get stuck when I open send mail window

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 732148 - evolution-3.12.3: evolution get stuck when I open send mail window


Summary:	evolution-3.12.3: evolution get stuck when I open send mail window


Status:	RESOLVED NOTGNOME

Product:	evolution
Classification:	Applications
Component:	Mailer
Version:	3.12.x (obsolete)
Hardware:	Other Linux

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

URL:
Whiteboard:

Duplicates:	731680 (view as bug list)
Depends on:
Blocks:

Reported:	2014-06-24 10:00 UTC by Pacho Ramos
Modified:	2014-06-30 09:13 UTC

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

Attachments
gdb --batch --ex "t a a bt" -pid=`pidof evolution` &>bt-evolution.txt (16.91 KB, text/plain) 2014-06-27 09:19 UTC, Pacho Ramos	Details

Description Pacho Ramos 2014-06-24 10:00:26 UTC

evolution gets stuck when I try to forward and email *while* I am disconnected from network (I do this clicking to the button at bottom left in evolution window).

gdb -p "pid of evolution" shows me the following trace:

+ Trace 233721

#0 pthread_cond_wait
at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S line 185
#1 g_cond_wait
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/glib/gthread-posix.c line 756
#2 g_once_init_enter
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/glib/gthread.c line 649
#3 e_dbus_address_book_get_type
at e-dbus-address-book.c line 833
#4 book_client_connect_open_cb
at e-book-client.c line 1228
#5 g_simple_async_result_complete
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/gio/gsimpleasyncresult.c line 763
#6 reply_cb
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/gio/gdbusproxy.c line 2623
#7 g_simple_async_result_complete
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/gio/gsimpleasyncresult.c line 763
#8 g_dbus_connection_call_done
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/gio/gdbusconnection.c line 5498
#9 g_simple_async_result_complete
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/gio/gsimpleasyncresult.c line 763
#10 complete_in_idle_cb
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/gio/gsimpleasyncresult.c line 775
#11 g_main_dispatch
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/glib/gmain.c line 3064
#12 g_main_context_dispatch
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/glib/gmain.c line 3663
#13 g_main_context_iterate
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/glib/gmain.c line 3734
#14 g_main_loop_run
at /var/tmp/portage/dev-libs/glib-2.40.0-r1/work/glib-2.40.0/glib/gmain.c line 3928
#15 gtk_main
at gtkmain.c line 1192
#16 main
at main.c line 680

Comment 1 Pacho Ramos 2014-06-24 12:29:40 UTC

Looks like it's due address book, if I disable all the address books from "Contacts" -> "Autocompletion", the mail window doesn't cause evolution to hang :/

Comment 2 Pacho Ramos 2014-06-27 09:19:54 UTC

Created attachment 279368 [details]
gdb --batch --ex "t a a bt" -pid=`pidof evolution` &>bt-evolution.txt

Comment 3 Pacho Ramos 2014-06-27 09:23:13 UTC

$ gdb --batch --ex "t a a bt" -pid=`pidof evolution-addressbook-factory` &>bt-evolution-addressbook-factory.txt

Gets this:
$ cat bt-evolution-addressbook-factory.txt
/usr/libexec/evolution-addressbook-factory.#prelink#.KqMFJs (deleted): No existe el fichero o el directorio.

+ Trace 233734

Thread 1 (process 1131)

#0 ??



Maybe prelink is causing this :O

Comment 4 Pacho Ramos 2014-06-27 09:41:12 UTC

Indeed, undoing prelinking works :O, not sure why this has started to fail now, I was doing prelinking from ages :/

Comment 5 Milan Crha 2014-06-30 08:44:27 UTC

What precisely do you call prelink? Linux has a capability to overwrite running binaries and they will still not crash but then the backtrace of such processes result to something like at comment #3. Simply restarting newly built (or just installed) process makes that work again.

If "undoing prelink" fixes the issue, then I'd call this NotGnome.

Comment 6 Milan Crha 2014-06-30 08:49:07 UTC

*** Bug 731680 has been marked as a duplicate of this bug. ***

Comment 7 Pacho Ramos 2014-06-30 09:13:10 UTC

This one:
http://en.wikipedia.org/wiki/Prelink

I also tested with the same version used in Fedora that I guess you could be using in Redhat too :)
http://pkgs.fedoraproject.org/cgit/prelink.git/tree/prelink.spec