subgraph-os-issues
subgraph-os-issues copied to clipboard
regression: can't boot to desktop
The following problems did not occur with the 2016-06-16 alpha but do with the recent alpha 3.
Problem #1: With monitor connected to DVI output of Nvidia GTX 970 graphics card, after about 3 lines of text briefly flash at the top of the screen, the video signal dies, the monitor goes into black screen powersave and stays there. To avoid this, I add nomodeset to the GRUB line, like a lot of people in the Debian ecosystem have to do. This however leads to
Problem #2: With nomodeset, the boot process gets stuck with a systemd log on the screen and desktop never comes up. The text on the screen flashes for a while, and then the flashing stops but it's still stuck. Apparently it is looping on a failure to launch the X server. kern.log has a lot of:
subgraph gnome-session-binary[13661]: Unrecoverable failure in required component org.gnome.Shell.desktop
user.log has a lot of x.org fails:
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) Failed to load module "nv" (module does not exist, 0)
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) [drm] Failed to open DRM device for pci:0000:01:00.0: -19
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) open /dev/dri/card0: No such file or directory
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) open /dev/dri/card0: No such file or directory
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) open /dev/fb0: No such file or directory
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) open /dev/fb0: No such file or directory
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) Screen 0 deleted because of no matching config section.
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) Screen 0 deleted because of no matching config section.
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) VESA(0): Cannot read int vect
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) Screen(s) found, but none have a usable configuration.
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE)
Jan 3 20:40:51 subgraph /usr/lib/gdm3/gdm-x-session[13639]: (EE) no screens found(EE)
And daemon.log has a lot of this:
Jan 3 20:40:50 subgraph systemd[1]: Created slice User Slice of Debian-gdm.
Jan 3 20:40:50 subgraph systemd[1]: Starting User Manager for UID 111...
Jan 3 20:40:50 subgraph systemd[1]: Started Session c499 of user Debian-gdm.
Jan 3 20:40:50 subgraph systemd[13502]: Starting D-Bus User Message Bus Socket.
Jan 3 20:40:50 subgraph systemd[13502]: Reached target Paths.
Jan 3 20:40:50 subgraph systemd[13502]: Reached target Timers.
Jan 3 20:40:50 subgraph systemd[13502]: Listening on D-Bus User Message Bus Socket.
Jan 3 20:40:50 subgraph systemd[13502]: Reached target Sockets.
Jan 3 20:40:50 subgraph systemd[13502]: Reached target Basic System.
Jan 3 20:40:50 subgraph systemd[13502]: Reached target Default.
Jan 3 20:40:50 subgraph systemd[13502]: Startup finished in 10ms.
Jan 3 20:40:50 subgraph systemd[1]: Started User Manager for UID 111.
Jan 3 20:40:50 subgraph systemd[13502]: Started D-Bus User Message Bus.
Jan 3 20:40:50 subgraph gnome-session-binary[13511]: WARNING: Could not parse desktop file orca-autostart.desktop or it references a not found TryExec binary
Jan 3 20:40:50 subgraph gnome-session-binary[13511]: WARNING: App 'org.gnome.Shell.desktop' exited with code 1
Jan 3 20:40:50 subgraph systemd[1]: Stopping User Manager for UID 111...
Jan 3 20:40:50 subgraph systemd[13502]: Stopped target Default.
Jan 3 20:40:50 subgraph systemd[13502]: Stopping D-Bus User Message Bus...
Jan 3 20:40:50 subgraph systemd[13502]: Stopped D-Bus User Message Bus.
Jan 3 20:40:50 subgraph systemd[13502]: Stopped target Basic System.
Jan 3 20:40:50 subgraph systemd[13502]: Stopped target Sockets.
Jan 3 20:40:50 subgraph systemd[13502]: Stopped target Timers.
Jan 3 20:40:50 subgraph systemd[13502]: Stopped target Paths.
Jan 3 20:40:50 subgraph systemd[13502]: Closed D-Bus User Message Bus Socket.
Jan 3 20:40:50 subgraph systemd[13502]: Reached target Shutdown.
Jan 3 20:40:50 subgraph systemd[13502]: Starting Exit the Session...
Jan 3 20:40:50 subgraph systemd[13502]: Received SIGRTMIN+24 from PID 13526 (kill).
Jan 3 20:40:50 subgraph systemd[1]: Stopped User Manager for UID 111.
Jan 3 20:40:50 subgraph systemd[1]: Removed slice User Slice of Debian-gdm.
Can you try this in another tty?
sudo dpkg-reconfigure gdm3
and if that fails, more aggressively
sudo apt install --reinstall gdm3
Let me know if that helps? This is very similar to an issue I've seen with nvidia and "plain" Debian in the past.
Booting with nomodeset, I can switch tty with Alt-F2, but the screen keeps flashing back and forth between the login prompt and systemd log, and keypresses are randomly captured by the systemd tty. I have to wait for the flashing to stop before I can log in successfully.
The first command had no effect. The second command failed with "Reinstallation of gdm3 is not possible, it cannot be downloaded."
This flashing behavior is identical to what I've seen before on "pure" Debian.
You will need to use nmcli
(nmcli connection up <connid>
) to bring up a network interface, or will need to pre-download the gdm3 package from the Debian repositories to reinstall it.
Manually downloaded stretch gdm3 for amd64 657246 Jan 5 19:56 gdm3_3.22.1-1_amd64.deb
Installed with sudo dpkg -i *.deb
Made no difference.
Seeing the same problem with the December build in a newly created VirtualBox VM.
The same problem here with build 2016/12/30 (clean install) i can't boot.
I am having the same problem as well on a Mac host with Virtualbox. Doesn't boot after install goes fine, screen flashes
December's alpha in virtualbox on linux mint serena boots as far as trying to start gdm but then falls to the flashing behaviour described above. Trying to start sgos in recovery mode fails because a console cannot be opened. The root account is said to be locked.
It is also a problem in Debian Stretch/Testing with their recent builds. We'll look at it further and try to address it.
Debian Stretch with Virtualbox. Stopped booting @ "Started GNOME Display Manager"
I also have the same problem, after a clean install on VBox during the first boot my screen started flashing after [ OK ] Started GNOME Display Manager. After 2 minutes the screen stops flashing. When I then press Ctrl+Alt+F2 I can login with "user" and then type "sudo startx" to start the desktop. But it's pretty annoying to wait 2-3 minutes every time I boot just to wait until the flashing stops...
See also: https://www.jamieweb.net/blog/subgraph-os-overview/#virtualmachines
Same here. Very annoying.
same too
With alpha release 4, after a clean install wait some minutes and after a screen stops blinking, press ctrl alt f2 or 3 or 4... and enter as credentials, so insert the command sudo startx.
If anyone here is still having this issue (even with the latest alpha), this worked for me:
When the screen stops flickering, press VirtualBox Host Key
(Right Ctrl in my case) + Ctrl
+ Alt
+ F2
(or cycle through F3, F4). You should see the bash prompt.
If you do sudo startx
, nothing works. So instead do: sudo chmod u+s /usr/bin/xinit
(solution from https://unix.stackexchange.com/questions/159124/xf86enableioports-failed-to-set-iopl-for-i-o-operation-not-permitted) and then do startx
.
Anyone find a way to fix this issue yet?