Bug 751584 - evolution crashes on start
Summary: evolution crashes on start
Keywords:
Status: CLOSED UPSTREAM
Alias: None
Product: Fedora
Classification: Fedora
Component: evolution
Version: 16
Hardware: x86_64
OS: Linux
unspecified
unspecified
Target Milestone: ---
Assignee: Matthew Barnes
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2011-11-05 23:28 UTC by Zoltán Vigh
Modified: 2011-12-03 16:46 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2011-12-01 10:23:07 UTC
Type: ---


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
GNOME Bugzilla 644695 0 None None None Never

Description Zoltán Vigh 2011-11-05 23:28:42 UTC
When I start evolution it randomly crashes. I get this error message "camel-CRITICAL **: camel_folder_summary_lock: assertion `CAMEL_IS_FOLDER_SUMMARY (summary)' failed
Segmentation fault". I don't know how to reproduce this bug.

Kernel: 3.1.0-7.fc16.x86_64
OS: Fedora 16
Evolution version: 3.2.1-2.fc16

Comment 1 Milan Crha 2011-11-07 07:42:37 UTC
Thanks for a bug report. Unfortunately, without a reproducer it's hard to tell what failed. It would be good to get a backtrace and possibly a valgrind log, whose may help to identify the issue. Could you follow below steps, please?
a) install debug info packages for all evolution-related packages, which
   involves gtkhtml3, evolution-data-server, evolution, and any other
   evolution-* packages you may have installed and be using
b) run evolution from console under gdb and get a backtrace in case it'll
   crash. You can use this command to achieve it:
   $ G_SLICE=always-malloc gdb --batch evolution --ex r --ex "t a a bt" \
     --ex q &>bt.txt
c) get a valgrind log by running evolution under valgrind, with command:
   $ G_SLICE=always-malloc valgrind --num-callers=50 evolution &>log.txt

I hope that one of these will be helpful in identifying your issue.

Comment 2 dominik 2011-11-25 13:51:07 UTC
Hey,

same Problem.

Kernel: Linux 3.0.0-1-486
OS: Debian Wheezy
Evolution: 3.0.3-2

(evolution:7245): camel-CRITICAL **: camel_folder_summary_lock: assertion `CAMEL_IS_FOLDER_SUMMARY (summary)' failed
Speicherzugriffsfehler

b)
via
http://www2.abteilung2.de/Bugreport/Evolution/bt_20111126.txt

c)
break after 2 minutes
via
http://www2.abteilung2.de/Bugreport/Evolution/log_20111126.txt

(In reply to comment #1)

> b) run evolution from console under gdb and get a backtrace in case it'll
>    crash. You can use this command to achieve it:
>    $ G_SLICE=always-malloc gdb --batch evolution --ex r --ex "t a a bt" \
>      --ex q &>bt.txt
> c) get a valgrind log by running evolution under valgrind, with command:
>    $ G_SLICE=always-malloc valgrind --num-callers=50 evolution &>log.txt

Comment 3 Milan Crha 2011-11-28 10:08:51 UTC
Thanks for the update. I see the backtrace is missing debug information for evolution-data-server and evolution. If you could install them then it'll give more detailed information. The valgrind log doesn't show any crash, and is clean of anything from camel. Do I read it right that the evolution didn't crash when run under valgrind? It's possible this is related to "proper" timing of actions, and when the evolution is run under valgrind then because of all the memory checking and slower running the timing is different and the crash is "avoided".

Comment 4 dominik 2011-11-28 12:12:59 UTC
> more detailed information. The valgrind log doesn't show any crash, and is
> clean of anything from camel. Do I read it right that the evolution didn't
> crash when run under valgrind? 

Right, under valgrind runs Evo under
- Debian
- Mint and
- Fedora

I break these Tests after 2 minutes and 10 syncs. 

> Thanks for the update. I see the backtrace is missing debug information for
> evolution-data-server and evolution. If you could install them then it'll give
> more detailed information. The valgrind log doesn't show any crash, and is

This Week i test other Distris and debug installed/enabled.

Comment 5 Andrei Trifan 2011-11-28 16:31:23 UTC
same problem here...

Kernel: 3.1.0-1.2-default 
OS: Opensuse 12.1
Evolution: 3.2.1

(evolution:7245): camel-CRITICAL **: camel_folder_summary_lock: assertion
`CAMEL_IS_FOLDER_SUMMARY (summary)' failed

bt.txt http://pastebin.com/VxEf63za

log.txt http://pastebin.com/0HVVjPwn

(In reply to comment #1)
> Thanks for a bug report. Unfortunately, without a reproducer it's hard to tell
> what failed. It would be good to get a backtrace and possibly a valgrind log,
> whose may help to identify the issue. Could you follow below steps, please?
> a) install debug info packages for all evolution-related packages, which
>    involves gtkhtml3, evolution-data-server, evolution, and any other
>    evolution-* packages you may have installed and be using
> b) run evolution from console under gdb and get a backtrace in case it'll
>    crash. You can use this command to achieve it:
>    $ G_SLICE=always-malloc gdb --batch evolution --ex r --ex "t a a bt" \
>      --ex q &>bt.txt
> c) get a valgrind log by running evolution under valgrind, with command:
>    $ G_SLICE=always-malloc valgrind --num-callers=50 evolution &>log.txt
> 
> I hope that one of these will be helpful in identifying your issue.

Comment 6 Milan Crha 2011-11-29 11:04:08 UTC
Thanks for the update. The both logs contain the information, unfortunately without debug information, thus it shows it happened in a library, but doesn't tell in what part it happened in an enough detailed level (source file names and line numbers are missing). Could you repeat with installed debug info packages for evolution and evolution-data-server of the same version as the binary packages, please? You can check with:
  $ rpm -qa | grep evolution | sort
and package, with its corresponding -debuginfo package should have exactly the same version, like here:
   evolution-3.2.1-2.fc16.x86_64
   evolution-data-server-3.2.1-1.fc16.x86_64
   evolution-data-server-debuginfo-3.2.1-1.fc16.x86_64
   evolution-debuginfo-3.2.1-2.fc16.x86_64

Comment 7 dominik 2011-11-29 12:42:32 UTC
1. Test

Kernel: Linux 3.0.0-1-486
OS: Debian Wheezy
Evolution: 3.0.3-2

Installation:
i   evolution   
i A evolution-common
i A evolution-data-server
i A evolution-data-server-common
i   evolution-data-server-dbg
i   evolution-dbg
i A evolution-exchange
i   evolution-exchange-dbg
i A evolution-plugins  
i   evolution-plugins-experimental
i A evolution-webcal
i A libevolution
i A libsyncevolution0
i   mail-notification-evolution

bt.txt -> http://pastebin.com/fA5XABCJ

with option
> G_SLICE=always-malloc valgrind --num-callers=50 evolution &>log.txt
evo dosent start. permanently curls mails but no error was report.

log.txt -> http://pastebin.com/CjiL73b5

Comment 8 Andrei Trifan 2011-11-29 14:16:24 UTC
$ rpm -qa | grep evolution | sort
evolution-3.2.1-2.2.2.i586
evolution-data-server-3.2.1-2.1.2.i586
evolution-data-server-debuginfo-3.2.1-2.1.2.i586
evolution-data-server-debugsource-3.2.1-2.1.2.i586
evolution-data-server-devel-3.2.1-2.1.2.i586
evolution-data-server-doc-3.2.1-2.1.2.i586
evolution-data-server-lang-3.2.1-2.1.2.noarch
evolution-debuginfo-3.2.1-2.2.2.i586
evolution-debugsource-3.2.1-2.2.2.i586
evolution-devel-3.2.1-2.2.2.i586
evolution-ews-0.31.0-2.1.2.i586
evolution-ews-debuginfo-0.31.0-2.1.2.i586
evolution-ews-debugsource-0.31.0-2.1.2.i586
evolution-ews-devel-0.31.0-2.1.2.i586
evolution-ews-lang-0.31.0-2.1.2.noarch
evolution-exchange-3.2.1-2.1.2.i586
evolution-exchange-debuginfo-3.2.1-2.1.2.i586
evolution-exchange-debugsource-3.2.1-2.1.2.i586
evolution-exchange-lang-3.2.1-2.1.2.noarch
evolution-groupwise-3.2.1-2.1.2.i586
evolution-groupwise-debuginfo-3.2.1-2.1.2.i586
evolution-groupwise-debugsource-3.2.1-2.1.2.i586
evolution-groupwise-devel-3.2.1-2.1.2.i586
evolution-lang-3.2.1-2.2.2.noarch
evolution-mono-plugins-3.2.1-2.2.2.i586
evolution-mono-plugins-debuginfo-3.2.1-2.2.2.i586
evolution-plugin-rss-0.2.90-6.1.2.i586
evolution-plugin-rss-debuginfo-0.2.90-6.1.2.i586
evolution-plugin-rss-lang-0.2.90-6.1.2.noarch
evolution-python-plugins-3.2.1-2.2.2.i586
evolution-python-plugins-debuginfo-3.2.1-2.2.2.i586
evolution-rss-debugsource-0.2.90-6.1.2.i586
evolution-sharp-0.21.1-27.1.2.i586
evolution-sharp-debuginfo-0.21.1-27.1.2.i586
evolution-sharp-debugsource-0.21.1-27.1.2.i586
evolution-tray-0.0.8-3.1.2.i586
evolution-tray-debuginfo-0.0.8-3.1.2.i586
evolution-tray-debugsource-0.0.8-3.1.2.i586
evolution-webcal-2.32.0-8.1.2.i586
evolution-webcal-debuginfo-2.32.0-8.1.2.i586
evolution-webcal-debugsource-2.32.0-8.1.2.i586
evolution-webcal-lang-2.32.0-8.1.2.noarch
glade-catalog-evolution-3.2.1-2.2.2.i586
libopensync-plugin-evolution2-0.22-216.1.2.i586
libopensync-plugin-evolution2-debuginfo-0.22-216.1.2.i586
libopensync-plugin-evolution2-debugsource-0.22-216.1.2.i586
libopensync-plugin-evolution2-devel-0.22-216.1.2.i586
nautilus-sendto-plugin-evolution-3.0.1-2.1.2.i586
nautilus-sendto-plugin-evolution-debuginfo-3.0.1-2.1.2.i586
pidgin-evolution-2.10.0-8.1.3.i586
pidgin-evolution-debuginfo-2.10.0-8.1.3.i586
planner-evolution-0.14.5-3.1.2.i586
planner-evolution-debuginfo-0.14.5-3.1.2.i586
python-evolution-2.32.0-11.1.2.i586
python-evolution-debuginfo-2.32.0-11.1.2.i586
tracker-miner-evolution-0.12.5-3.4.1.i586
tracker-miner-evolution-debuginfo-0.12.5-3.4.1.i586

bt.txt http://pastebin.com/PN9imKwe

log.txt http://pastebin.com/B7fSvzMi

(In reply to comment #6)
> Thanks for the update. The both logs contain the information, unfortunately
> without debug information, thus it shows it happened in a library, but doesn't
> tell in what part it happened in an enough detailed level (source file names
> and line numbers are missing). Could you repeat with installed debug info
> packages for evolution and evolution-data-server of the same version as the
> binary packages, please? You can check with:
>   $ rpm -qa | grep evolution | sort
> and package, with its corresponding -debuginfo package should have exactly the
> same version, like here:
>    evolution-3.2.1-2.fc16.x86_64
>    evolution-data-server-3.2.1-1.fc16.x86_64
>    evolution-data-server-debuginfo-3.2.1-1.fc16.x86_64
>    evolution-debuginfo-3.2.1-2.fc16.x86_64

Comment 9 Milan Crha 2011-11-30 09:26:23 UTC
(In reply to comment #8)
> tracker-miner-evolution-0.12.5-3.4.1.i586
> tracker-miner-evolution-debuginfo-0.12.5-3.4.1.i586

Seeing tracker-miner-evolution in your list, and knowing it introduces some issues, will anything change if you remove this package, please?

Comment 10 Andrei Trifan 2011-11-30 10:31:27 UTC
(In reply to comment #9)
> (In reply to comment #8)
> > tracker-miner-evolution-0.12.5-3.4.1.i586
> > tracker-miner-evolution-debuginfo-0.12.5-3.4.1.i586
> 
> Seeing tracker-miner-evolution in your list, and knowing it introduces some
> issues, will anything change if you remove this package, please?

It does not solve my problem and evolution randomly crashes.

$ rpm -qa | grep evolution | sort
evolution-3.2.1-2.2.2.i586
evolution-data-server-3.2.1-2.1.2.i586
evolution-data-server-debuginfo-3.2.1-2.1.2.i586
evolution-data-server-debugsource-3.2.1-2.1.2.i586
evolution-data-server-devel-3.2.1-2.1.2.i586
evolution-data-server-doc-3.2.1-2.1.2.i586
evolution-data-server-lang-3.2.1-2.1.2.noarch
evolution-debuginfo-3.2.1-2.2.2.i586
evolution-debugsource-3.2.1-2.2.2.i586
evolution-devel-3.2.1-2.2.2.i586
evolution-lang-3.2.1-2.2.2.noarch
evolution-mono-plugins-3.2.1-2.2.2.i586
evolution-mono-plugins-debuginfo-3.2.1-2.2.2.i586
evolution-plugin-rss-0.2.90-6.1.2.i586
evolution-plugin-rss-debuginfo-0.2.90-6.1.2.i586
evolution-plugin-rss-lang-0.2.90-6.1.2.noarch
evolution-python-plugins-3.2.1-2.2.2.i586
evolution-python-plugins-debuginfo-3.2.1-2.2.2.i586
evolution-rss-debugsource-0.2.90-6.1.2.i586
evolution-sharp-0.21.1-27.1.2.i586
evolution-sharp-debuginfo-0.21.1-27.1.2.i586
evolution-sharp-debugsource-0.21.1-27.1.2.i586
evolution-tray-0.0.8-3.1.2.i586
evolution-tray-debuginfo-0.0.8-3.1.2.i586
evolution-tray-debugsource-0.0.8-3.1.2.i586
evolution-webcal-2.32.0-8.1.2.i586
evolution-webcal-debuginfo-2.32.0-8.1.2.i586
evolution-webcal-debugsource-2.32.0-8.1.2.i586
evolution-webcal-lang-2.32.0-8.1.2.noarch
glade-catalog-evolution-3.2.1-2.2.2.i586
nautilus-sendto-plugin-evolution-3.0.1-2.1.2.i586
nautilus-sendto-plugin-evolution-debuginfo-3.0.1-2.1.2.i586
pidgin-evolution-2.10.0-8.1.3.i586
pidgin-evolution-debuginfo-2.10.0-8.1.3.i586
planner-evolution-0.14.5-3.1.2.i586
planner-evolution-debuginfo-0.14.5-3.1.2.i586
python-evolution-2.32.0-11.1.2.i586
python-evolution-debuginfo-2.32.0-11.1.2.i586

Comment 11 Milan Crha 2011-12-01 09:04:57 UTC
(In reply to comment #10)
> (In reply to comment #9)
> > Seeing tracker-miner-evolution in your list, and knowing it introduces some
> > issues, will anything change if you remove this package, please?
> 
> It does not solve my problem and evolution randomly crashes.

Is it still this same crash, please? Also, I still do not understand why your last backtrace shows things like this:
#0  0xb6b4b401 in camel_offline_journal_write () from /usr/lib/libcamel-provider-1.2.so.29
which means there is something wrong with your debuginfo packages, event the versions you've shown above seems correct, except of those 'debugsource' packages. For example, where does the evolution-data-server-debugsource package come from, please? I do not have it available and the standard fedora package with debug information is evolution-data-server-debuginfo.

Comment 12 Milan Crha 2011-12-01 10:23:07 UTC
I just got a good backtrace with debug symbols at an upstream bug [1], thus let's move there.

[1] https://bugzilla.gnome.org/show_bug.cgi?id=644695

Comment 13 Milan Crha 2011-12-02 14:00:38 UTC
(In reply to comment #11)
> (In reply to comment #10)
> > (In reply to comment #9)
> > > Seeing tracker-miner-evolution in your list, and knowing it introduces some
> > > issues, will anything change if you remove this package, please?
> > 
> > It does not solve my problem and evolution randomly crashes.
> 
> Is it still this same crash, please?

Checking within the upstream bug I'm pretty sure the tracker is on the fault, thus my question from comment #11 stands, as I would like to know whether it's on your side too and if evolution crashes, then what the backtrace is. Also, when you run evolution from console, there should not be any messages like below two shown there, because they are coming from the tracker plugins, thus it can be ensured that the plugin is really removed and not used here.

    Tracker-DEBUG: Tracker Evolution plugin enabled
    Tracker-DEBUG: Tracker plugin creating new object...

Comment 14 Andrei Trifan 2011-12-03 16:46:17 UTC
(In reply to comment #11)
> (In reply to comment #10)
> > (In reply to comment #9)
> > > Seeing tracker-miner-evolution in your list, and knowing it introduces some
> > > issues, will anything change if you remove this package, please?
> > 
> > It does not solve my problem and evolution randomly crashes.
> 
> Is it still this same crash, please? Also, I still do not understand why your
> last backtrace shows things like this:
> #0  0xb6b4b401 in camel_offline_journal_write () from
> /usr/lib/libcamel-provider-1.2.so.29
> which means there is something wrong with your debuginfo packages, event the
> versions you've shown above seems correct, except of those 'debugsource'
> packages. For example, where does the evolution-data-server-debugsource package
> come from, please? I do not have it available and the standard fedora package
> with debug information is evolution-data-server-debuginfo.

It's the same crash. I have all debuginfo packages from opensuse 12.1 repo. I'm using two imap accounts from my freebsd server. 
Now I found a solution to use my email. I remove evolution directory from my home folder and make all configurations again. If I leave special folders untouched in imap account editor and use them locally on my computer everything is working normal without errors and evolution is not crashing anymore.
http://imageshack.us/photo/my-images/6/screenshotat20111203181.png/
If I use drafts, sent, junk, trash folders from my imap account evolution is crashing randomly.


Note You need to log in before you can comment on or make changes to this bug.