Hello,
I have an issue with Xorg becoming completely unresponsive once I switched to a VT, and then back. The cursor moves, but nothing else responds. At this point I can switch back to a TTY and reboot. Restarting xdm service sometimes works, sometimes doesn’t with Xorg failing to launch.
I am using Intel display drivers, running on 5-th gen Haswell.
xf86-video-intel 2.99.917
xorg-x11-server 7.6_1.18.1
kernel 4.4.1-1-default
Only entries in journal during the VT switch:
> Feb 27 11:31:31 alienware13 gnome-shell.desktop[2240]: Gjs-Message: JS LOG: _onFocus [object instance proxy GType:MetaWindowX11 jsobj@0x7fa55721a340 native@0x358ee40]
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) AIGLX: Suspending AIGLX clients for VT switch
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 13:65
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 13:70
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 226:0
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 13:69
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 13:72
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 13:68
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 13:67
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 13:64
> Feb 27 11:31:32 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got pause for 13:79
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 13:65
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 13:70
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 226:0
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) AIGLX: Resuming AIGLX clients after VT switch
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) intel(0): switch to mode 1920x1080@60.0 on eDP1 using pipe 0, position (0, 0), rotation normal, reflection none
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (–) synaptics: SynPS/2 Synaptics TouchPad: touchpad found
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 13:69
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 13:72
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 13:68
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 13:67
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 13:64
> Feb 27 11:31:35 alienware13 /usr/lib/gdm/gdm-x-session[2092]: (II) systemd-logind: got resume for 13:79
Most likely a GDM problem…
To avoid a few chances for trouble, Leap “downgraded” GDM to 3.14 while using Gnome 3.16: the result is rock-solid here on Haswell Mobile…
If Leap fits your needs, stick to it until known issues are solved in the coming versions of Gnome.
Actually seeing as Gnome has been absolutely stable I have added the Gnome repository and didn’t downgrade any Gnome packages.
Just in case someone might have the same issue.
Just to avoid misunderstandings to others reading this, gdm 3.14.2-6.1 is the current standard update in the Leap “OSS - update” repo, no user triggered “downgrade” is needed: this was one of the Distribution choices at release time.
On the other side, it is not clear to me whether you are now running Tumbleweed with (which?) Gnome repo enabled or you are still running Leap with Gnome 3.16 (and gdm 3.14.2, BTW), or you are now running a base Leap (kernel 3.15.8… say) with some sort of upgraded Gnome (Gnome:Next?) on top.
As you helpfully state…
“Just in case someone might have the same issue.”
Currently (when I looked into related about 4 mths ago),
There are Haswell features that enhance virtualization and process execution ever since gen 1(4 yrs ago) that even recently are still considered unstable. There are relatively few virtualization technologies that have tried to take advantage of these features and these features may or may not be enabled by default.
VMware has been one of the few virtualization technologies that have tried to take advantage from the beginning, and there are documents that describe how to turn off the feature with micro-code instructions if necessary.
You didn’t describe which virtualization technology you’re using, so best I can advise is that this is something you might want to look into. Note that even if enabled by default in the kernel, if you’re not running virtualization or you’re running virtualization and the technology isn’t trying to use these bleeding edge features you wouldn’t be affected by this.
Just to make it clear. I am running Leap 42 with the Gnome repository on top. Likewise I have the Kernel stable repository, The Leap 4.1 kernel is missing some drivers for my Alienware.
