keyboard crashed with SIGSEGV in g_settings_get_value()

[Hanna Maksimava]

Was opening "Keyboard" from system settings app.

ProblemType: Crash
DistroRelease: elementary OS 0.2
Package: switchboard-plug-keyboard 0.1-0~31~precise1 [origin: LP-PPA-elementary-os-daily]
ProcVersionSignature: Ubuntu 3.2.0-36.57-generic-pae 3.2.35
Uname: Linux 3.2.0-36-generic-pae i686
ApportVersion: 2.0.1-0ubuntu17.1+elementary3~precise1
Architecture: i386
CrashDB: switchboard_plug_keyboard
Date: Wed Jan 23 00:20:35 2013
EcryptfsInUse: Yes
ExecutablePath: /usr/lib/plugs/pantheon/keyboard/keyboard
InstallationMedia: elementary OS 0.2 "Luna" - Beta 1 i386 (20121114)
MarkForUpload: True
ProcCmdline: /usr/lib/plugs/pantheon/keyboard/keyboard
ProcEnviron:
 PATH=(custom, no user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
SegvAnalysis:
 Segfault happened at: 0xb71c92db <g_settings_get_value+43>: mov (%esi),%edx
 PC (0xb71c92db) ok
 source "(%esi)" (0x00000019) not located in a known VMA region (needed readable region)!
 destination "%edx" ok
SegvReason: reading NULL VMA
Signal: 11
SourcePackage: switchboard-plug-keyboard
StacktraceTop:
 g_settings_get_value () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
 g_settings_get_strv () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
 keyboard_shortcuts_settings_get_val ()
 keyboard_shortcuts_tree_construct ()
 keyboard_shortcuts_tree_new ()
Title: keyboard crashed with SIGSEGV in g_settings_get_value()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo

[Julien Spautz (julien-spautz)]

Thanks for reporting the bug!

Could you please run the program from terminal and paste the output here?

switchboard -o keyboard.plug

[Adam Wrzosek (amx-h)]

This is what I recived in terminal:

[_LOG_LEVEL_INFO 22:24:01.584091] Switchboard.vala:592: Welcome to System Settings
[_LOG_LEVEL_INFO 22:24:01.584945] Switchboard.vala:593: Version: 0.9
[_LOG_LEVEL_INFO 22:24:01.585169] Switchboard.vala:594: Report any issues/bugs you mind find to lp:switchboard
[_LOG_LEVEL_INFO 22:24:01.798256] Application.vala:75: Switchboard version: 0.9
[_LOG_LEVEL_INFO 22:24:01.799626] Application.vala:77: Kernel version: 3.2.0-36-generic-pae

(keyboard:3514): GLib-GIO-CRITICAL **: g_settings_get_value: assertion `G_IS_SETTINGS (settings)' failed

(keyboard:3514): GLib-CRITICAL **: g_variant_get_type: assertion `value != NULL' failed

(keyboard:3514): GLib-CRITICAL **: g_variant_type_is_subtype_of: assertion `g_variant_type_check (type)' failed

(keyboard:3514): GLib-CRITICAL **: g_variant_dup_strv: assertion `g_variant_is_of_type (value, G_VARIANT_TYPE_STRING_ARRAY)' failed

(keyboard:3514): GLib-CRITICAL **: g_variant_unref: assertion `value != NULL' failed

[RabbitBot (rabbitbot-a)]

StacktraceTop:
 g_settings_get_value () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
 g_settings_get_strv () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
 keyboard_shortcuts_settings_get_val ()
 keyboard_shortcuts_tree_construct ()
 keyboard_shortcuts_tree_new ()

[Emptywell (emptyhousewilliams)]

this is what apears, the program just keep in loading ....

[_LOG_LEVEL_INFO 23:47:34.149408] Switchboard.vala:592: Welcome to System
Settings
[_LOG_LEVEL_INFO 23:47:34.149531] Switchboard.vala:593: Version: 0.9
[_LOG_LEVEL_INFO 23:47:34.149608] Switchboard.vala:594: Report any
issues/bugs you mind find to lp:switchboard
[_LOG_LEVEL_INFO 23:47:34.339667] Application.vala:75: Switchboard version:
0.9
[_LOG_LEVEL_INFO 23:47:34.339768] Application.vala:77: Kernel version:
3.2.0-36-generic-pae

On 23 January 2013 14:21, Julien Spautz <email address hidden> wrote:

> Thanks for reporting the bug!
>
> Could you please run the program from terminal and paste the output
> here?
>
> switchboard -o keyboard.plug
>
> --
> You received this bug notification because you are subscribed to a
> duplicate bug report (1102802).
> https://bugs.launchpad.net/bugs/1103179
>
> Title:
> keyboard crashed with SIGSEGV in g_settings_get_value()
>
> Status in Switchboard Plug Keyboard:
> Triaged
>
> Bug description:
> Was opening "Keyboard" from system settings app.
>
> ProblemType: Crash
> DistroRelease: elementary OS 0.2
> Package: switchboard-plug-keyboard 0.1-0~31~precise1 [origin:
> LP-PPA-elementary-os-daily]
> ProcVersionSignature: Ubuntu 3.2.0-36.57-generic-pae 3.2.35
> Uname: Linux 3.2.0-36-generic-pae i686
> ApportVersion: 2.0.1-0ubuntu17.1+elementary3~precise1
> Architecture: i386
> CrashDB: switchboard_plug_keyboard
> Date: Wed Jan 23 00:20:35 2013
> EcryptfsInUse: Yes
> ExecutablePath: /usr/lib/plugs/pantheon/keyboard/keyboard
> InstallationMedia: elementary OS 0.2 "Luna" - Beta 1 i386 (20121114)
> MarkForUpload: True
> ProcCmdline: /usr/lib/plugs/pantheon/keyboard/keyboard
> ProcEnviron:
> PATH=(custom, no user)
> LANG=en_US.UTF-8
> SHELL=/bin/bash
> SegvAnalysis:
> Segfault happened at: 0xb71c92db <g_settings_get_value+43>: mov
> (%esi),%edx
> PC (0xb71c92db) ok
> source "(%esi)" (0x00000019) not located in a known VMA region (needed
> readable region)!
> destination "%edx" ok
> SegvReason: reading NULL VMA
> Signal: 11
> SourcePackage: switchboard-plug-keyboard
> StacktraceTop:
> g_settings_get_value () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
> g_settings_get_strv () from /usr/lib/i386-linux-gnu/libgio-2.0.so.0
> keyboard_shortcuts_settings_get_val ()
> keyboard_shortcuts_tree_construct ()
> keyboard_shortcuts_tree_new ()
> Title: keyboard crashed with SIGSEGV in g_settings_get_value()
> UpgradeStatus: No upgrade log present (probably fresh install)
> UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo
>
> To manage notifications about this bug go to:
>
> https://bugs.launchpad.net/switchboard-plug-keyboard/+bug/1103179/+subscriptions
>

[Julien Spautz (julien-spautz)]

I could reproduce the "infinite loading screen" behaviour, but it only happens sometimes. This happens when the plug tries to get a certain key in gsettings. Seems like other users are affected by the same bug but with other keys. I wrote a little workaround that simply discards those shortcuts, but in that case, they won't be displayed in the plug. It shouldn't crash anymore, as soon as the fix is released, but I'll have to further investigate this.

Thanks for reporting the bug!

[Julien Spautz (julien-spautz)]

Found something that might very well have caused the segmentation fault. It seems to have fixed the unresponsiveness/crash problem. I'll mark this as fixed, feel free to reopen the bug if it still hasn't been fixed. Don't forget to update to revision 34.