evolution uses up to 95% of cpu usage

[Bertrand Rousseau]

when I start evolution, it hangs for several minutes with almost 100% cpu usage.
When it stops, any actions refreshing the content of the messages list pane
(e.g. clicking on a folder) makes this happen again.

The same problem arise when I go to the prefs panel.

[Bertrand Rousseau (bertrand-rousseau)]

I forgot to say that it concerns evolution 2.1.2, sorry

[Sebastien Bacher (seb128)]

(please don't change the milestone, that's a tool used by the developers to know
which bugs have to be worked on for a release)

BTw I've no idea at all on how to get details on this problem. Do you have a lot
of mail ? Is this bug specific of a new evolution version ?
could you run it in gdb and get a backtrace ?
$ gdb evolution
(gdb) r
ctrl-C
(gdb) bt

[Bertrand Rousseau (bertrand-rousseau)]

Yes I have quite a lot of mails (almost 2000). The problem has happened since I
upgraded from warty to hoary, and switched from evolution 2.0 to 2.1.2, so I
suppose it is specific to this version.

When the CPU starts to rise (just after the GUI appears), and until the cpu
usage falls again (+/- 5min later), the output displayed in gdb is :

[New Thread -1321845840 (LWP 6862)]
requesting object classid: attachment.0x80f6f98.9610.mixed.1
object_found: 1
requesting object classid:
bonobo-unknown:///em-format-html-display/.0x80f6f98.9610.mixed.1
---Type <return> to continue, or q <return> to quit---
(no debugging symbols found)
inside eab_vcard_control_new
object_found: 1
loading error file /usr/share/evolution/2.2/errors/addressbook-errors.xml
loading error file /usr/share/evolution/2.2/errors/e-system-errors.xml
loading error file /usr/share/evolution/2.2/errors/shell-errors.xml
loading error file /usr/share/evolution/2.2/errors/filter-errors.xml
loading error file /usr/share/evolution/2.2/errors/mail-composer-errors.xml
loading error file /usr/share/evolution/2.2/errors/calendar-errors.xml
loading error file /usr/share/evolution/2.2/errors/mail-errors.xml
loading error file
/usr/share/evolution/2.2/errors/org-gnome-mailing-list-actions-errors.xml
[New Thread -1330332752 (LWP 6893)]
[Thread -1330332752 (LWP 6893) exited

Here's the backtrace:
---------------------

(gdb) bt
#0 0xffffe410 in __kernel_vsyscall ()
#1 0xb6e52155 in poll () from /lib/tls/i686/cmov/libc.so.6
#2 0xb6ed995a in g_main_loop_get_context () from /usr/lib/libglib-2.0.so.0
#3 0xb6ed8fd0 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0
#4 0xb6ed959a in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#5 0xb7667d2f in bonobo_main () from /usr/lib/libbonobo-2.so.0
#6 0x0806724b in main ()

[Bertrand Rousseau (bertrand-rousseau)]

The bug has gone away since last update of my hoary (2 days ago). I suspect the
new version of libgtkhtml of being responsible for this. Anyway, everything back
to normal now. Thanks for the work!

[Sebastien Bacher (seb128)]

nice, bug closed so, thanks

[Magankie (magankie)]

This problem returns on Hardy. CPU is at 95-100% used by gnome-keyring. I'm using a exchange plugin, I think that it's the problem now...

[Sebastien Bacher (seb128)]

the hardy issue is rather bug #236171