Blank screen using nvidia drivers

[prothis]

I am experiencing the ‘blank screen, missing codec’ problem that some others referred to earlier this year. But those threads either do not contain a solution or the proposed solution did not work for me.

The problem: connecting to my Linux remote using a Windows client works fine, as long as the former has the nouveau display driver installed. As soon as I switch to the nvidia proprietary display driver I experience the ‘blank screen’ when connecting. The screen is entirely white and I cannot interact with it. I can access the menu by navigating to the corner of the screen. When clicking on ‘Display’ there most options are greyed out and at the bottom it says ‘Unknown codec’.

Based on other, similar threads I have tried the following:
– Disabling hardware encoding and H264
– Uncommenting DesktopCommandLibraryPath in node.cfg and directing it to where I believe it should be directed: DesktopCommandLibraryPath “/usr/lib/x86_64-linux-gnu/”

Neither of these have helped. The only solution has been to switch back to the nouveau driver, which solves the problem.

Thank you in advance for any tips about how to solve this!

Server OS: Linux Mint 19.2 Tina, Cinnamon
Server display: physical
Client OS: Windows 10
NoMachine version: 6.8.1, free (both machines)

[fra81]

Hi,

is the Linux server headless?

Please gather server side logs (possibly after disabling hardware encoding) as explained in https://www.nomachine.com/AR10K00697. You can send them to forum[at]nomachine[dot]com.

[prothis]

Thanks for your response!

The server is not headless.

Will send the logs ASAP.

[prothis]

Just to confirm: the logs have now been sent your way. Of course let me know if there is anything else I can send. Thanks!

[Britgirl]

I’ve checked our forum mail address and have not received anything.

[prothis]

That’s odd – I’ve resent it now. Email subject is “Logs for post/issue #23878”

[Britgirl]

Received 🙂

[prothis]

Great, thanks for confirming!

[fra81]

Hi,

it seems the nxnode process on the server keeps crashing, and this problem is unknown. Please check for core files (usually located in the user’s home directory) and follow instructions in https://www.nomachine.com/AR09L00810 to generate a backtrace.

[prothis]

Okay will do. Having some trouble figuring out how to enable core dumps in my distro (disabled by default). But will try this weekend and get back.

[prothis]

Apologies for the delay in getting back! Unfortunately I do not have a core dump to share, but:

This problem is now resolved. I believe the problem may have been related to the fact that I installed NoMachine while running the nouveau driver, and then subsequently switched to the nvidia driver. That may have caused a messed up config somewhere. I installed a new GPU today (something I was planning anyway), and as part of that process I uninstalled NoMachine, switched to the nvidia driver, then reinstalled NoMachine (with reboots in between each step). Now it works fine.

Thanks for your support with this issue, much appreciated!

[Author]

Posts