Comment 3 for bug 1636564

Revision history for this message
Jan Lachnitt (pepalogik) wrote :

Another report and some thoughts below

HW: Asus Vivobook (touchscreen, Intel graphics), external monitor attached.
OS: Linux Mint 18 Cinnamon x64.
State before crash: Running X session using both displays ("big desktop"), several applications, Vivaldi browser active.
My aim: Lock the computer.
Course of action, the USUAL part: Pressed Ctrl+Alt+L, not locked - Vivaldi displayed its menu (to be displayed upon pressing Alt), tapped the touchscreen to get rid of the menu and thus actually lock the computer.
Course of action, the UNUSUAL part: Black screen for a few seconds, then login screen, old X session gone.

syslog:
Mar 20 14:43:20 VivoBook org.a11y.atspi.Registry[1729]: XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0"
Mar 20 14:43:20 VivoBook org.a11y.atspi.Registry[1729]: after 2925 requests (2925 known processed) with 0 events remaining.
Mar 20 14:43:20 VivoBook cinnamon-session[1640]: Gdk-WARNING: t+4492,79569s: cinnamon-session: Fatal IO error 11 (Prostředek je dočasně nepřístupný) on X server :0.
Mar 20 14:43:20 VivoBook org.gtk.vfs.Daemon[1711]: A connection to the bus can't be made
Mar 20 14:43:20 VivoBook org.gtk.vfs.Daemon[1711]: A connection to the bus can't be made
Mar 20 14:43:20 VivoBook mdm[1212]: WARNING: mdm_slave_xioerror_handler: Kritická chyba X - Restartování :0
Mar 20 14:43:20 VivoBook kernel: [ 4538.677519] vivaldi-bin[4100]: segfault at 968 ip 00007f61ac384643 sp 00007ffc09f25b50 error 4 in libX11.so.6.3.0[7f61ac35c000+135000]
Mar 20 14:43:20 VivoBook ag[4118]: mdm-superpost Starting...
Mar 20 14:43:20 VivoBook ag[4122]: mdm-superpost Finished
Mar 20 14:43:20 VivoBook console-kit-daemon[1566]: (process:4128): GLib-CRITICAL **: g_slice_set_config: assertion 'sys_page_size == 0' failed
Mar 20 14:43:21 VivoBook acpid: client 1219[0:0] has disconnected
Mar 20 14:43:21 VivoBook acpid: client connected from 4131[0:0]
Mar 20 14:43:21 VivoBook acpid: 1 client rule loaded
Mar 20 14:43:22 VivoBook ag[4139]: mdm-superinit Starting...
Mar 20 14:43:22 VivoBook ag[4140]: mdm-superinit Finished

Xorg.0.log covers the new session, Xorg.0.log.old has no record of the event.

The segfault really seems to affect multiple users, especially with Chromium-based browsers, e.g.:
https://bbs.archlinux.org/viewtopic.php?id=131328 - from 2011 already
https://bbs.archlinux.org/viewtopic.php?id=221888 - solution?

I don't know if syslog preserves the order of events (within a second), but if yes, then the segfault is just a consequence of another problem, e.g.:
https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers/+bug/999191