Ubuntu 24.10 random hang/crash on SER9

oddhack

I’ve been running the latest Ubuntu 24.10 ISO (which has linux-firmware 20240913) on my SER9 (DHX375 SN, which I think is the V2 BIOS?) for a couple of days. No issues with video while installing, and as far as I can tell at casual inspection it’s accelerated - gets around 7500 fps for vkgears -present-mode in 4K fullscreen. So that’s the good part.

The major bad part is that after a couple of hours of uptime, the system just halts or crashes. I cannot tell what happened, because I can’t sit there for hours watching, and there is literally nothing in any of the system logs at the approximate time it goes down (which I established by running a ping every few minutes from another machine). It is repeatable although I’m not sure if the uptime is repeatable (took 3.5 hours the first time I measured, 10.5 hours the second time). This might be something trivial but unless someone recognizes this behavior, I have no clue how to debug it. If I can’t sort it out I’ll have to return the device, which would be sad.

The minor annoyance part is that on boot, I get dropped into a Busybox command prompt under grub, instead of the usual selection-of-boot-options screen. I can just ‘exit’ to get to the boot choices, but no clue why this happened - have never seen anything quite like it in 25+ years of running Linux. Maybe I should just reinstall grub and see if that helps?

oddhack

@gukin @RandomName (tagging since you talked about Ubuntu 24.10 in the other thread) - have you had any stability issues running 24.10? I installed it a few days ago and my SER9 randomly halts after some variable number of hours of uptime.

support6

oddhack

Don’t worry, the PC has one-year warranty. Please send us a picture of SN on the bottom of the PC.
Here are some potential solutions: First, Please turn off the PC , unplug the power supply, and press the CMOS button (if no CMOS button, please put the pin in the CMOS pinhole and press it ) for about 10 seconds , restart the PC after ten minutes. Here is the tutorial https://mega.nz/file/XypwTISY#DVT0BsjOgiPkZ-drQaj88JH9Z7f7qpKGP2E3nMHW8ms or https://url.bee-link.cn/5FdW If the problem persists, please try other solutions:

①Please take out the RAM ,please clean the contacts of the RAM with a rubber. Here is the tutorial https://mega.nz/file/XqZX0TIR#_DtO2oa4aRP4bWtvxgKLj4QGhBpQ959cg5U3Yw4u1Cc or https://url.bee-link.cn/AzzB**

②Please take out one RAM and leave the other one in it to test if the RAM is defective or not.** You can also use Memtest software to test the RAM.

③Please repeatedly press delete key as soon as you turn on the PC, go to BIOS. Please send us a picture of the Main page, we will check if you need to update the BIOS.**

④Please change another power supply.**

⑤Overheating. If your PC is overheating, please check if the fan spins. You can also turn off the turbo boost. Go to control panel —power option—change to High performance— change plan setting—advanced setting —processor power management— change both 100% to 99% , click save.

⑥Update graphics driver: Please uninstall the graphic driver in the device manager and install the new one. After you send us the SN code, we’ll send you the corresponding driver.

Best regards!
Yue
Customer Support

oddhack

@support6 thank you. At the moment, I think this is more likely to be an interaction with Ubuntu 24.10, than a hardware or power problem. I switched to Manjaro Linux, and have been running vkgears for nearly 24 hours without issues. So my speculation is that there is some instability in the video or DRI drivers packaged with Ubuntu 24.10 which is interacting with the hardware configuration of the SER9 and occasionally crashing. Quite difficult to diagnose, obviously. Assuming Manjaro does not exhibit this crashing behavior in the coming days, I’ll just stay on that distribution for the time being.

support10

oddhack

Hello there,

Thank you for sharing.

If there are any updates, please keep us informed.

Have a nice day!

old486whizz

@oddhack - I run Fedora on a V2 SER9 (Serial DHX…) . I’ve only had mine running for a max of ₁₅ hours with no issues so far.
I use KDE, not sure what you may be running in ubuntu (is it Mate or Gnome maybe?).
Also, the Window System (if it’s using X.org still, some newer wayland based one or their own that they used to use… PIR or something?).

It could be that it’s interacting with the graphic drivers and it could be the graphical system crashing - could you SSH in from another machine (maybe a laptop) and simply “sudo journalctl -f” ? That way, as soon as it crashes, you’ll get the error messages from the crash that aren’t able to be written to disk…
You can also open up another SSH terminal and use “watch sensors” which might give you a clue as to whether the temperature was high when it crashed (although we would need to do some looping etc to get it to record a history and get it as a CSV for analysis).

Have you altered any BIOS settings to possibly introduce any instability (other than changing the VRAM size)?

I am unaware of what services might be set up on ubuntu and what they might do (maybe a disk scanner etc).. So not sure if you’ve tried stress testing cpu/ram/disk ?

-
old486whizz

support4

old486whizz

Hi there
Check for errors: Use SSH to run sudo journalctl -f to capture error messages when the system crashes.
This can give you insights into what went wrong.

Monitor temperature:
Use watch sensors to see if temperature spikes correlate with the crash.
You could set up a history log for more detailed analysis.

System stress test:
Stress test CPU, RAM, and disk to check if they might be contributing to the crashes.