Random system freezes

Maybe someone has seen this… it is my first time encountering a system freeze like this.

I installed CL to an Intel NUC via the server ISO. I then add a desktop-lxqt, lightdm, basic gui stuff. At random times, the NUC totally locks up, it will not even ping. As far as I can tell, the NUC becomes totally frozen. It will not respond in any way, other than to physically power it off… by pulling the power.

My first thought was a hardware issue. So I tried a second NUC, same issue.

Edit: I should add that I have not experienced this issue when installing via a desktop ISO.

1 Like

Unfortunately, these issues are very difficult to debug. You can try to inspect journalctl -a after restarting the system, attaching a serial console, etc…

The best advice is to check for BIOS upgrades and apply any that are available. As for hardware support, I honestly don’t even know where one would go… oops :frowning:

1 Like

Intel NUC8i7 with CL also freezes sometimes. Also no ping possible to the machine anymore. No other console with keys Alt-Ctl-F2/F3 possible. Only hard power-off.

With Fedora it freezes at least once a day. This damaged file system with loss of data.

Then changed to CL. With CL less freezes. Not daily anymore. Yesterday rearranged Gnome application menu. Suddenly it freezes. No ping anymore.

With CL display gets black regularly once a day for 5-10 seconds, but NUC not freezing. It’s not a complete loss of video signal. Signal indicator/LED on monitor remains blue. Yellow when NUC is off.

Brave Browser locks my computer up within minutes. It’s my preferred browser, but there are some big issues.

I have the same issues.

I suspect that issues appear when I use Brave Browser to play youtube videos(Gnome in X11)
Now I will try in Wayland to see when I have the same issue.

The same issue in Wayland,
I found this in journalctl --dmesg:
Feb 03 08:39:22 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:39:24 s3 kernel: i915 0000:00:02.0: GPU recovery timed out, cancelling all in-flight rendering.
Feb 03 08:39:24 s3 kernel: i915 0000:00:02.0: Resetting chip for hang on rcs0
Feb 03 08:57:30 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:57:30 s3 kernel: [drm:gen8_reset_engines] ERROR rcs0 reset request timed out: {request: 00000001, RESET_CTL: 00000001}
Feb 03 08:57:30 s3 kernel: i915 0000:00:02.0: Resetting chip for hang on rcs0
Feb 03 08:57:30 s3 kernel: [drm:gen8_reset_engines] ERROR rcs0 reset request timed out: {request: 00000001, RESET_CTL: 00000001}
Feb 03 08:57:30 s3 kernel: [drm:gen8_reset_engines] ERROR rcs0 reset request timed out: {request: 00000001, RESET_CTL: 00000001}
Feb 03 08:57:34 s3 kernel: GpuWatchdog[3240]: segfault at 0 ip 0000559bda3e413d sp 00007f76ee8bb400 error 6 cpu 3 in brave[559bd5f0a000+7683000]
Feb 03 08:57:34 s3 kernel: Code: 48 c1 c9 03 48 81 f9 af 00 00 00 0f 87 c9 00 00 00 48 8d 15 f9 90 3d fb f6 04 11 04 0f 84 b8 00 00 00 be 01 00 00 00 ff 50 30 04 25 00 00 00 00 37 13 0>
Feb 03 08:57:38 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:57:46 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:57:48 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:57:50 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:57:52 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:57:54 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:57:56 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:57:58 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:58:00 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:58:02 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:58:04 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
Feb 03 08:58:06 s3 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0

@Adrian_K your issue might be a different one that is upstream, not Clear Linux-spcific:

Hi puneetse,this is weird because this is first time when I see this log,that happened when I have brave open but not playing YouTube.I have this issue before with brave playing YouTuber and after freeze I can only reboot from power button.
With no playing video after minutes the system work again,no need reboot.So,this is a big issue and my fault was that I don’t record the log before because I erase CL.Now I have Manjaro to play(kernel 5.4.15) and here is no issue.I will reinstall CL and test again.

So,i have a fresh CL install,install Brave Browser and play 2 hours youtube,no issue so far…
But this is not a solution…everytime a new fresh install.Somehow,maybe with gnome upgrade the issue appear again…is here an intention to make Brave Browser official in CL?

Freeze again.:

Is maybe better to reinstall CL with LTS kernel?

Update:
I tryed with Google Chrome…no issue,no freeze.
Now started Brave Browser…this is what appear:

Now waiting to freeze…why freeze happens with Brave and not witht GoogleChrome and Firefox?

I use Vivaldi (Chromium-based) and also have this freezes for 1-2 month.
Gnome Version 3.34.3
Kernel 5.4.17-901.native
Mesa Intel® HD Graphics 530 (SKL GT2)

With the latest kernel-native 5.4.18 the system-coredump is gone when play Brave Browser.

After update to 5.4.18 graphic shell crash turned into friezes (2-3 sec).

фев 09 14:27:06 clr-1fc1c0bf41bb45ceb2acf44ba9aa1098 kernel: i915 0000:00:02.0: GPU HANG: ecode 9:1:0x00000000, hang on rcs0
фев 09 14:27:06 clr-1fc1c0bf41bb45ceb2acf44ba9aa1098 kernel: GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
фев 09 14:27:06 clr-1fc1c0bf41bb45ceb2acf44ba9aa1098 kernel: Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
фев 09 14:27:06 clr-1fc1c0bf41bb45ceb2acf44ba9aa1098 kernel: drm/i915 developers can then reassign to the right component if it's not a kernel issue.
фев 09 14:27:06 clr-1fc1c0bf41bb45ceb2acf44ba9aa1098 kernel: The GPU crash dump is required to analyze GPU hangs, so please always attach it.
фев 09 14:27:06 clr-1fc1c0bf41bb45ceb2acf44ba9aa1098 kernel: GPU crash dump saved to /sys/class/drm/card0/error
фев 09 14:27:06 clr-1fc1c0bf41bb45ceb2acf44ba9aa1098 kernel: i915 0000:00:02.0: Resetting rcs0 for hang on rcs0

unfortunate. the second time everything also hung tight

ethan@clr-1fc1c0bf41bb45ceb2acf44ba9aa1098~ $ uname -r
5.4.18-902.native

One of the big changes of late is that mesa has defaulted to using iris for gfx that it supports. When switching to iris prior to this change, it was able to hang my system completely when gaming (to the point of a forced reboot so no log) with similar journal messages. When I didn’t force iris, it didn’t lock the system.

2 things would be getting the log from /sys/class/drm/card0/error for a bug report so the issue can be fixed in mesa (if you can use the system after the hang/freezes). The other would be to remove iris so that it started using i965 again to see whether iris is the cause or not.

kernel was updated to 5.5

ethan@clr-1fc1c0bf41bb45ceb2acf44ba9aa1098~ $ uname -r
5.5.2-903.native
1 Like

It’s been nearly three years since I first encountered Clear Linux, but the freezing issue unique to Clear Linux remains a distinct and characteristic problem, unlike anything found in Debian or RHEL.

This problem occurs whether running Clear Linux on a hypervisor or on bare-metal hardware.

Even when trying different hardware platforms with entirely different applications, Clear Linux experiences freezing too frequently. In my life, Clear Linux is the only distribution that keeps me reaching for the physical reset button on my computer.

Despite many users reporting device messages and the like, it’s puzzling that a solution hasn’t been found. I suspect that the issue might be related to limited compatibility with certain hardware.

I have tested Clear Linux on three completely different devices over the course of three years, and the freezing issue persists consistently. The problem, which is not present in other Linux distributions, occurs in the same manner. The fact that there doesn’t seem to be a specific debugging method for addressing this issue is indeed a problem.

1 Like

Given that a substantial amount of time has elapsed since the initial mention of this issue, I had expected that this problem would have been resolved. With hopeful anticipation, I installed Clear Linux on my mini PC, but unfortunately, the freezing issue persists.

Similar to the experience with my old server where I had to press the physical button every week after installing Clear Linux, now my mini PC from an entirely different manufacturer is also encountering the same problem.

These days, even though I am still able to manually reset it once a week when I’m at home, if my business trips become more frequent, continuing this practice will no longer be feasible.

I intend to switch my Linux distribution to either RHEL or Debian in the near days.