Crash on Debian Linux Bullseye

[Whattteva]

I searched both on Google and the forums but the closest I can find is a crashed session, which isn’t quite the same.

I’ve been running NoMachine 8.4.2 on Debian Linux Bullseye VM installed on Proxmox.

Once every 1-2 days, I noticed that my VM would crash and completely power off the VM. This may also be a coincidence, but everytime it happens, it seems like I’m in the middle of scrolling. I know for a fact that this is due to NoMachine because if I switch to the SPICE client, the once every 1-2 day crash would cease completely. Unfortunately, I can’t keep using the SPICE because copy/paste/audio don’t work reliably there and it works flawlessly on NoMachine.

Has anyone come across this issue or have an idea of where to look in the logs so I can figure out what is causing this instability?

My relevant specs are as follows:

Server:

• Headless Proxmox 7.3-3
• SPICE QXL display
• NoMachine server 8.4.2 (free)
• Debian 11 Bullseye VM
• i3wm 4.19.1-1
• 16 GiB RAM

Client:

• NoMachine client 8.4.2 (free)
• Linux Mint 21.1
• Cinnamon 5.6.7

Thank you for all the feedbacks!

[Britgirl]

Please submit the full logs from both server and player sides as per the instructions here. Thanks!
https://kb.nomachine.com/DT07S00244

Submit them directly via email to forum[at]nomachine[dot]com making sure to reference the title of the this topic (in the subject).

[Whattteva]

Thanks for the reply. I will post the logs once I’ve reproduced the problem.

[Britgirl]

In the meantime we have not been able to reproduce the same behaviour specifically when scrolling. It would be useful to see system logs (journalctl) so submit that as well when you are able to reproduce it.

There was another user that was experiencing something similar which you can see here: https://forums.nomachine.com/topic/session-terminated-unexpectedly, and we reproduced it with the same set-up and steps.

That crash is due to an Xorg bug that is exposed by NoMachine registering for receiving ButtonPress events from the system. You can find the workaround in that topic.

[Whattteva]

Hi, I have just sent the email with the requested logs. I have referenced this forum post in the email as well.

[Britgirl]

Ok received thank you.

[Whattteva]

Thank you for all your help.

If you would be so kind to update this thread if there are any updates, I will be following this closely.

Once again, thank you so much for your help.

[Britgirl]

Hi, the logs didn’t show anything strange occurring. Can you also send us the system logs (journalctl)? These would be useful. How about the workaround I suggested in my earlier reply. Have you tried that since submitting the logs?

[Whattteva]

That’s unfortunate. I haven’t tried that because I wanted as close as possible reproduction of the problem. I’d just modified the node.cfg file and restarted the server. Will observe the next few days what happens.

As for the journalctl, is there any specific thing you want me to grep filter it with or any special cli flags because there’s a lot of other non-relevant things in it.

[ikfredje]

I have the same problem on Debian Bullseye X11 (not Wayland) with NoMachine 8.4.2 but I only sporadically use it so I don’t know when/which update is responsible.

I can connect to the server without problems from other linuxes (Rocky linux 8 / X11).

The modification of /usr/NX/etc/node.cfg did not solve the problem.

journalctl shows:

Mar 13 11:32:09 maxwell kernel: nxplayer.bin[5366]: segfault at 7fb1d8ceddc0 ip 00007fb22fd1fdf4 sp 00007fb1a0ff8378 error 4 in libc-2.31.so[7fb22fc9d000+15a000]
Mar 13 11:32:09 maxwell kernel: Code: 29 4f 10 0f 29 57 20 0f 29 5f 30 48 83 c7 40 48 83 fa 40 77 d0 0f 11 29 0f 11 71 f0 0f 11 79 e0 44 0f 11 41 d0 41 0f 11 23 c3 <0f> 10 26 0f 10 6e 10 0f 10 76 20 0f 1>

 

[Britgirl]

ikfredje,

 

your issue us on the player side, so it’s best to deal with the problem in another topic.

I see that you have opened a separate topic https://forum.nomachine.com/topic/client-on-debian-bullseye-aborts-on-connect, so I’ve linked it here.

 

[Britgirl]

Whatteva, no need to filter anything. Just send the whole file to us as it is 🙂

[Whattteva]

Ok I’ve sent it in. It’s a massive file (102 MiB uncompressed), though it compresses down to 3.3 MiB for sending.

[Britgirl]

They didn’t reveal anything useful either.

So to sum up: the NoMachine logs don’t show anything strange (just a couple of restarts) and the system logs didn’t show anything that could suggest reasons for a crash.

So, either the problem is not there or the logs are not catching it.

All we can suggest is that you keep debug enabled and then should the issue occur, you should collect the logs and then send them in (nomachine player and server, as well as system logs).

[Whattteva]

Thank you so much for the help.

I have indeed implemented the work-around and since I put that in, the frequency has definitely gone down significantly. I have only had 1 crash so far within at least a week now. I hope that this cuts it down severely that if it just happens like once a month, I’ll be fine with it.

[Author]

Posts