0

[Note: I know there are many questions on nvidia and blank screen, but none so far have brought me closer to a solution, I've tried a lot, see below]

Problem: Running Xubuntu 18.10 on a Dell XPS 9570 (which has a GeForce GTX 1050 Ti card, along with built-in Intel graphics), the nvidia 410 and 415 drivers stopped working. All I get when I start Linux after the scrolling boot messages, is a black screen with (non-blinking) white cursor on the upper left corner. nouveau drivers work fine, but they only allow me to use the built-in laptop screen but not ones attached to the laptop via HDMI.

It used to work (if you can call it that, when it boots to a working desktop environment 2 out of 3 times, and stays black otherwise) until today morning (when I did an apt upgrade, which installed linux-image-4.18.0-13.

What I have tried so far to resolve this:

  • tried booting the linux-image-4.18.0-12, same issue
  • as recommended here, I added nomodeset kernel option to be set through grub (I already set this a while ago, this was required to get it working in the first place)
  • as recommended most of the time on blank screen related nvidia questions, I tried purging everything nvidia-related (everything listed by dpkg --list | grep nvidia), and reinstalling 410 drivers
  • I also added graphics-drivers ppa and installed 415 drivers from there (https://launchpad.net/~graphics-drivers/+archive/ubuntu/ppa) (a few times with purging in between)
  • Tried Bumblebee: https://wiki.ubuntu.com/Bumblebee (the guide there doesn't seem to activate the nvidia driver, right? At least for me it didn't, nouveau still was used)
  • As for the issue mentioned above, when the GUI would only load every 2 out of 3 times, I tried xdm, gdm, lddm - but with them, I got a black screen all the time; only lightdm (where I remember an issue a while back causing exactly the symptoms I saw) seems to work togehter with nvidia driver...

As for errors in the journal, I see something regard to the nvidia-persistenced and some bumblebee reference (the latter only after installing bumblebee, of course). There don't seem to be any real nvidia-related errors now (and there weren't previously on the occasions when the screen stayed blank):

$ journalctl --since today | grep -i nvidia | grep -i "\(err\|fail\)"
Dez 21 07:04:45 nertha kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  415.23  Thu Dec  6 21:34:12 CST 2018 (using threaded interrupts)
Dez 21 13:44:00 nertha kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  415.23  Thu Dec  6 21:34:12 CST 2018 (using threaded interrupts)
Dez 21 13:44:00 nertha nvidia-persistenced[743]: nvidia-persistenced failed to initialize. Check syslog for more details.
Dez 21 13:44:00 nertha nvidia-persistenced[749]: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 121 has read and write permissions for those files.
Dez 21 13:44:00 nertha systemd[1]: nvidia-persistenced.service: Failed with result 'exit-code'.
Dez 21 13:44:00 nertha systemd[1]: Failed to start NVIDIA Persistence Daemon.
Dez 21 13:47:49 nertha kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  415.23  Thu Dec  6 21:34:12 CST 2018 (using threaded interrupts)
Dez 21 13:47:49 nertha nvidia-persistenced[772]: nvidia-persistenced failed to initialize. Check syslog for more details.
Dez 21 13:47:49 nertha nvidia-persistenced[780]: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 121 has read and write permissions for those files.
Dez 21 13:47:49 nertha systemd[1]: nvidia-persistenced.service: Failed with result 'exit-code'.
Dez 21 13:47:49 nertha systemd[1]: Failed to start NVIDIA Persistence Daemon.
Dez 21 13:47:50 nertha nvidia-persistenced[875]: Failed to unlink PID file: No such file or directory
Dez 21 13:47:50 nertha nvidia-persistenced[912]: Failed to unlink PID file: No such file or directory
Dez 21 13:47:50 nertha nvidia-persistenced[951]: Failed to unlink PID file: No such file or directory
Dez 21 13:47:50 nertha nvidia-persistenced[971]: Failed to unlink PID file: No such file or directory
Dez 21 13:47:50 nertha systemd[1]: nvidia-persistenced.service: Failed with result 'start-limit-hit'.
Dez 21 13:47:50 nertha systemd[1]: Failed to start NVIDIA Persistence Daemon.
Dez 21 13:47:50 nertha systemd[1]: nvidia-persistenced.service: Failed with result 'start-limit-hit'.
Dez 21 13:47:50 nertha systemd[1]: Failed to start NVIDIA Persistence Daemon.
Dez 21 13:58:06 nertha kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  415.23  Thu Dec  6 21:34:12 CST 2018 (using threaded interrupts)
Dez 21 14:09:52 nertha kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  415.25  Wed Dec 12 10:22:08 CST 2018 (using threaded interrupts)
Dez 21 14:09:52 nertha nvidia-persistenced[754]: nvidia-persistenced failed to initialize. Check syslog for more details.
Dez 21 14:09:52 nertha nvidia-persistenced[759]: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 121 has read and write permissions for those files.
Dez 21 14:09:53 nertha systemd[1]: nvidia-persistenced.service: Failed with result 'exit-code'.
Dez 21 14:09:53 nertha systemd[1]: Failed to start NVIDIA Persistence Daemon.
Dez 21 14:10:53 nertha kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  415.25  Wed Dec 12 10:22:08 CST 2018 (using threaded interrupts)
Dez 21 14:10:53 nertha nvidia-persistenced[809]: nvidia-persistenced failed to initialize. Check syslog for more details.
Dez 21 14:10:53 nertha nvidia-persistenced[811]: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 121 has read and write permissions for those files.
Dez 21 14:10:53 nertha systemd[1]: nvidia-persistenced.service: Failed with result 'exit-code'.
Dez 21 14:10:53 nertha systemd[1]: Failed to start NVIDIA Persistence Daemon.
Dez 21 14:10:54 nertha nvidia-persistenced[912]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[920]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[926]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[956]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[975]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[988]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1016]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1058]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1068]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1089]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1105]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1113]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1121]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1131]: Failed to unlink PID file: No such file or directory
Dez 21 14:10:54 nertha nvidia-persistenced[1142]: Failed to unlink PID file: No such file or directory
Dez 21 14:31:05 nertha kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  415.25  Wed Dec 12 10:22:08 CST 2018 (using threaded interrupts)
Dez 21 14:31:05 nertha nvidia-persistenced[756]: nvidia-persistenced failed to initialize. Check syslog for more details.
Dez 21 14:31:05 nertha nvidia-persistenced[759]: Failed to query NVIDIA devices. Please ensure that the NVIDIA device files (/dev/nvidia*) exist, and that user 121 has read and write permissions for those files.
Dez 21 14:31:05 nertha systemd[1]: nvidia-persistenced.service: Failed with result 'exit-code'.
Dez 21 14:31:05 nertha systemd[1]: Failed to start NVIDIA Persistence Daemon.
Dez 21 14:31:06 nertha systemd-udevd[403]: Process '/bin/mknod -m 666 /dev/nvidiactl c 195 255' failed with exit code 1.
Dez 21 14:31:06 nertha systemd-udevd[403]: Process '/bin/mknod -m 666 /dev/nvidia0   c 195 0' failed with exit code 1.
Dez 21 14:31:06 nertha bumblebeed[783]: [    5.548701] [ERROR]Failed to unload module 'nvidia_drm' (ref count: 2).
Dez 21 14:31:06 nertha bumblebeed[783]: [    5.548716] [ERROR]Failed to unload module 'nvidia_modeset' (ref count: 2).
Dez 21 14:31:06 nertha bumblebeed[783]: [    5.548724] [ERROR]Failed to unload module 'nvidia' (ref count: 77).
Dez 21 14:48:44 nertha kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  410.78  Sat Nov 10 22:09:04 CST 2018 (using threaded interrupts)
Dez 21 14:48:45 nertha systemd-udevd[427]: Process '/bin/mknod -m 666 /dev/nvidiactl c 195 255' failed with exit code 1.
Dez 21 14:48:45 nertha systemd-udevd[427]: Process '/bin/mknod -m 666 /dev/nvidia0   c 195 0' failed with exit code 1.
Dez 21 14:48:45 nertha bumblebeed[804]: [    5.564388] [ERROR]Failed to unload module 'nvidia_drm' (ref count: 2).
Dez 21 14:48:45 nertha bumblebeed[804]: [    5.564420] [ERROR]Failed to unload module 'nvidia_modeset' (ref count: 2).
Dez 21 14:48:45 nertha bumblebeed[804]: [    5.564428] [ERROR]Failed to unload module 'nvidia' (ref count: 77).

I really would like to use nvidia drivers again, as with nouveau, external screens don't seem to be recognized. What else can I try? If you need more info, let me know!

Part of this seems to be a lightdm problem. As noted in the answer below, for some reason, lightdm doesn't seem to start properly the first time. The only problem related to lightdm turning up in the journal (journalctl -b) is:

PAM unable to dlopen(pam_kwallet.so): /lib/security/pam_kwallet.so: cannot open shared object file: No such file or directory
PAM adding faulty module: pam_kwallet.so

I suppose this message is not related to this issue, it's probably a non-issue as far as I see from issues mentioning this message...

0

I got the nvidia-driver-390 to work with the previous kernel, 4.18.0-12 (but still didn't get multi-monitor support running).

Then I decided to try again with 4.18.0-13, but without nomodeset option - and that fixed it!

So the solution in my case: - nvidia-driver 390 drivers - Remove nomodeset from the line starting with GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub.

The issue with the occasional blank screen when starting still appears. It seems to have grown more frequent actually. A workaround in this situation is to log in on tty1 and execute: sudo systemctl restart lightdm.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.