/usr/bin/mate-display-properties was killed by signal 11 about mate-control-center

[szesch]

Looks very similar to #52

[dmashal]

This is related to glib.

(process:5246): GLib-CRITICAL **: g_slice_set_config: assertion `sys_page_size == 0' failed
abrt-applet: repeated problem in mate-control-center-1.6.0-1.fc19, not showing the notification

(process:5313): GLib-CRITICAL **: g_slice_set_config: assertion `sys_page_size == 0' failed

[szesch]

Perhaps this will help our bug squashers:
https://bugzilla.mozilla.org/show_bug.cgi?id=833117

[dmashal]

Any update? This is happening since 1.5.0 and continues with mate-control-center-1.6.0

https://bugzilla.redhat.com/show_bug.cgi?id=879663

[raveit65]

a current backtrace.
https://dl.dropboxusercontent.com/u/49862637/Mate-desktop/Bugs/backtrace-mate-display
This happend on both funktions.
a) change resolution
b) try to rotate the display

[rosset]

This bug is still present on mate 1.7.90

mate installed versions: http://paste.fedoraproject.org/83124/94134735

Related bugs on Fedora 21 (rawhide):
https://retrace.fedoraproject.org/faf/reports/313745/
https://bugzilla.redhat.com/show_bug.cgi?id=1058429

[rosset]

Also reproducible in MATE 1.8.0

http://paste.fedoraproject.org/83244/16328913

[stefano-k]

Finally I am able to reproduce this with my new laptop. I will try to fix this

[rosset]

Great, thank you Stefano. Please let me know if you need help to validate the patches when you fix it.

[samtygier]

Still seeing this in mate-control-center-1.8.1-2.fc21

(process:2633): GLib-CRITICAL **: g_slice_set_config: assertion 'sys_page_size == 0' failed

[stefano-k]

@samtygier are you able to get a backtrace running gdb and mate-display-properties with -g-fatal-warnings argument?

[samtygier]

http://www.fpaste.org/100880/ (gdb is asking for more dbg packages, but the backtrace looks complete to me. let me know if you need more info)

[flexiondotorg]

@stefano-k and @infirit Here are some bugs raised against Ubuntu MATE with logs of debug.

• https://bugs.launchpad.net/ubuntu-mate/+bug/1351890
• https://bugs.launchpad.net/ubuntu-mate/+bug/1338854

Regards, Martin.

[raveit65]

in a Mate GTK3 enviroment i don't see the issue anymore

[monsta]

in a Mate GTK3 enviroment i don't see the issue anymore

This might indicate we've hit some old bug in GTK+2...

[monsta]

Now I can confirm it indeed doesn't happen in GTK+3 build.
(Using GTK+ 3.14.5 in Debian Testing.)

[raveit65]

new backtraces:
fedora 20 https://dl.dropboxusercontent.com/u/49862637/Mate-desktop/Bugs/mate-display-properties-f20

with debian jessie

rave@jeesie:~$ gdb mate-display-properties 
GNU gdb (Debian 7.7.1+dfsg-5) 7.7.1
Copyright (C) 2014 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from mate-display-properties...(no debugging symbols found)...done.
(gdb) run
Starting program: /usr/bin/mate-display-properties 
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7fffed586700 (LWP 1228)]
[New Thread 0x7fffeccd3700 (LWP 1229)]

Program received signal SIGSEGV, Segmentation fault.
0x00007ffff577207d in ?? () from /usr/lib/x86_64-linux-gnu/libgdk-x11-2.0.so.0
(gdb) 

Full backtrace informations are in fedora ones.

[raveit65]

@stefano-k same as this report.
#106
Can we focus our work a bit on those reports please?
This is a regression since 1.6.x
Every user in fedora who change the screen resolution after installation will get a alarm from fedora's bug reporting tool.
The list of users who are affected in downstream report is huge.

[monsta]

I suspect this is caused by https://bugzilla.gnome.org/show_bug.cgi?id=635380.

[raveit65]

this commit is in gtk3
https://git.gnome.org/browse/gtk+/commit/?id=323df2b2800383832ed3c2e43626f2c6821c33ec
still wondering why this have an effect in gtk2.
or maybe this is missing in gtk2.

[monsta]

Yes, because they didn't push that commit to GTK+2 branch.

[monsta]

Ok, I've made a patch for GTK+2 and tested it in Debian (Jessie) and Ubuntu (Precise/Trusty/Utopic).
@NiceandGently also tested it in Fedora a few days ago.

If you use some other distro, you can get the patch from here and try it for yourself.

[raveit65]

..end of a long story :)
Patch is pushed to the gtk-2-24 branch upstream.
https://git.gnome.org/browse/gtk+/commit/?h=gtk-2-24&id=ee95f3d7259c0859ce41189b781b4339b4cd64aa
Thanks to monsta and m.clasen.

Marked as can be closed.

[monsta]

Great! :)

[stefano-k]

Wonderful! Thank you @monsta and @NiceandGently !!!

[babak777]

How do I apply the patch to Debian. I have down loaded gtk2-debdiff.
The reason that I need to apply this patch is that I upgraded from Wheezy to Jessie. Apparently, they fixed problem on Jessie, but have not in Wheezy. Since the package has the same version number in Jessie and in Wheezy, it won't get updated on Jessie upgrade. So I need to apply the patch. But the question is how to apply the patch?

[monsta]

Since the package has the same version number in Jessie and in Wheezy, it won't get updated on Jessie upgrade.

I don't know which package you mean, but gtk+2.0 is updated during the upgrade just fine as it's newer than in Wheezy. The patch is already in it.

[babak777]

Hi Monsta,

I used to have mate on Wheezy and I used to be able to run all the programs that ran on gnome. I upgraded to Jessie. Now, I am not able run many programs on mate, while I am able to run the same programs on gnome. For example, when I want to run say synaptic the window is opened and then it crashes. I did research and it seem to be very similar to symptoms described here.

I described it more detail in another forum :
http://forums.mate-desktop.org/viewtopic.php?f=5&t=3701&p=11553#p11553

I was thinking to uninstall the mate and reinstall it, but if the problem is with gtk2 then it won't help.
Would you like to find out what went wrong in my system?

Is there a test that we can do to see if the patch is applied?

Thanks,

[monsta]

For example, when I want to run say synaptic the window is opened and then it crashes. I did research and it seem to be very similar to symptoms described here.

It doesn't look similar to me. Besides, Synaptic is GTK+3 app.

Your description hints at some more generic upgrade-related problem. Probably the best thing to do is to post a support request at Debian forums.

[babak777]

Is
http://forums.debian.net/
a good form to post the problem? If not which forum is the best for posting this problem.

Thanks.

[monsta]

Never used it, but since it's listed on the official support page, it should be ok.

[babak777]

Thanks Monsta,
I will post on that forum.

I also get the problem on Debian testing.

johan@johanba-nb ~ $ mate-display-properties
[1]    3176 segmentation fault  mate-display-properties

And the debug output:

    johan@johanba-nb ~ $ gdb mate-display-properties
    GNU gdb (Debian 7.7.1+dfsg-5) 7.7.1
    Copyright (C) 2014 Free Software Foundation, Inc.
    License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
    This is free software: you are free to change and redistribute it.
    There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
    and "show warranty" for details.
    This GDB was configured as "x86_64-linux-gnu".
    Type "show configuration" for configuration details.
    For bug reporting instructions, please see:
    <http://www.gnu.org/software/gdb/bugs/>.
    Find the GDB manual and other documentation resources online at:
    <http://www.gnu.org/software/gdb/documentation/>.
    For help, type "help".
    Type "apropos word" to search for commands related to "word"...
    Reading symbols from mate-display-properties...Reading symbols from /usr/lib/debug/.build-id/e5/433c1d030dcb2e24e07887bf88e9138dcee350.debug...done.
    done.
    (gdb) run
    Starting program: /usr/bin/mate-display-properties 
    [Thread debugging using libthread_db enabled]
    Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
    [New Thread 0x7fffec5a1700 (LWP 5417)]

    Program received signal SIGSEGV, Segmentation fault.
    list_connected_outputs (total_w=0x7fffffffbd68, total_h=0x7fffffffbd6c, app=<optimized out>) at xrandr-capplet.c:1075
    1075    xrandr-capplet.c: No such file or directory.
    (gdb) 

Distro info:

johan@johanba-nb ~ $ lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description:    Debian GNU/Linux testing (stretch)
Release:    testing
Codename:   stretch

Kernel:

johan@johanba-nb ~ $ uname -a
Linux johanba-nb 4.1.0-1-amd64 #1 SMP Debian 4.1.3-1 (2015-08-03) x86_64 GNU/Linux

Versions:

mate-control-center: 1.8.3+dfsg1-2

[raveit65]

Do you have a mix of 1.8 and 1.10 packages?
xrandr changes in mate-desktop package......

[monsta]

Unrelated but known issue in testing/unstable, will be fixed after m-c-c is updated to 1.10 there. Please consider checking Debian bug tracker for existing reports before creating a new one.

[babak777]

Hi johangwbasson,

I had a similar problem. It has to do with a bug in GTK+3. I found a fix that works for me. For more info, please follow my posts here:
http://forums.debian.net/viewtopic.php?f=6&t=124130

[Skamasle]

Same problem here on debian testing:

error 4 in mate-display-properties

So the issue is because mate-display-properties | mate-control-settings version is 1.8 on debian testing so I upgrade it from sid to 1.10 and the problem is solved, are a few ways, wait, upgrade from sid repos or downgrade mate-desktop to 1.8