System Lockup on Ubuntu 24.04 with GNOME 47

Support and Help
,
1

System does a full lockup, mouse and keyboard doesn’t respond and hard drive activity light stops flashing, however the system is still responding to ICMP Pings and I can still SSH into the system if i run sudo shutdown now the system doesn’t do anything it just sits ther. The other day the system did a lockup while the radio stream was playing but it was a instant stop no suttering.

Ubuntu 24.04 is stock no modifications, would anyone have any ideas? this activity seemed to start when i upgraded from Ubuntu 22.04 to 24.04, did memtest left it running for 24 hours no errors, prime95 left for 48 hours no lockups.

There has been a few more ports about this.

Here’s what I have tried,
I set amdgpu.dc=0 in grub, updated grub via CLI as I have a RX570 8GB Model. The system has not done a lockup yet but time will tell. It’s mainly worst when I am on a Video Call on Firefox however using Brave Browser the issue seems to disspear, Thunderbird mail also causes the same problem lockup, leave it closed and the systems to be somewhat okay.

Looking at logs I ran the following,
sudo journalctl -b -1 | grep -i amdgpu

this log isn’t showing much same with dmesg or syslog although i do see the following.

Apr 11 23:19:47 VULTA.uk.routehosts.lan kernel: amdgpu 0000:01:00.0: amdgpu: IH ring buffer overflow (0x00087EE0, 0x0000C140, 0x00007EF0)
Apr 11 23:19:47 VULTA.uk.routehosts.lan kernel: amdgpu 0000:01:00.0: amdgpu: GPU fault detected: 147 0x0f8a8402
Apr 11 23:19:47 VULTA.uk.routehosts.lan kernel: amdgpu 0000:01:00.0: amdgpu:  for process rustdesk pid 637759 thread rustdesk:cs0 pid 637807
Apr 11 23:19:47 VULTA.uk.routehosts.lan kernel: amdgpu 0000:01:00.0: amdgpu:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0010C800
Apr 11 23:19:47 VULTA.uk.routehosts.lan kernel: amdgpu 0000:01:00.0: amdgpu:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x090C4002
Apr 11 23:19:47 VULTA.uk.routehosts.lan kernel: amdgpu 0000:01:00.0: amdgpu: VM fault (0x02, vmid 4, pasid 32772) at page 1099776, write from 'TC3' (0x54433300) (196)
Apr 12 13:20:06 VULTA.uk.routehosts.lan brave_brave.desktop[1274382]: /usr/share/libdrm/amdgpu.ids: No such file or directory
Apr 12 13:20:06 VULTA.uk.routehosts.lan brave_brave.desktop[1274382]: /usr/share/libdrm/amdgpu.ids: No such file or directory
Apr 12 13:20:06 VULTA.uk.routehosts.lan brave_brave.desktop[1274382]: /usr/share/libdrm/amdgpu.ids: No such file or directory

But the log relating to RustDesk I also see this on my Dell Laptop which uses onboard iGPU HD630 Graphics.

If anyone has any ideas please let me know, All I do know is I don’t believe this is a full system lockup and I think the original person should also test, the system responds to pings and you can still SSH into it. I think this is more down to either a bug with GNOME latest version or possibly a Graphics issue.

Regards

2

I’m experiencing a same freezing issue on my Ubuntu system that started suddenly after 3–4 months of stable use.

System Information:

  • Ubuntu Version: 24.04.2 LTS
  • Kernel Version: 6.11.0.24-generic
    (Also tried 6.11.0.21 via GRUB — same issue)
    (Tried installing kernels by mainline, but failed)
  • Graphics: Integrated (No dedicated GPU)
  • Laptop Model: VAIO VJS146
  • CPU: 13th Gen Intel® Core™ i7-1360P × 16 (Intel® Iris® Xe Graphics)
  • Memory: 32.0 GiB
  • SSD: 2TB, dual-boot setup with Windows
  • GNOME Version: 26

Problem Description:

Ubuntu randomly freezes when the system is left idle — sometimes after a few minutes, sometimes longer. There’s no consistent trigger. Sometimes it happens when Chrome is open, other times it’s just on the blank desktop. When I try to move the mouse or press a key, nothing responds.

In some cases, the system freezes at black screen(battery saving mode); other times, it freezes on the last visible screen. The system becomes completely unresponsive — even REISUB doesn’t work, although I confirmed it works when the system is healthy.

This has happened over 10 times now, and I’m really concerned about potential SSD or system wear due to forced shutdowns.

What I Tried:

I’d appreciate any help or suggestions.

3

This is the result of “journalctl -b -1 -p err”

journalctl
journalctl1920×860 240 KB

4

Have you checked if there is a BIOS update for your system? There are a lot of ACPI errors at the start of your screenshot…

5

Hello,

Where can I check for a BIOS update?

I regularly perform automatic system updates.

My laptop is quite rare, so I’m concerned that a BIOS update might not be available for it.

I’ve already tried the following, but the issue remains the same:

sudo nano /etc/default/grub
# I added the following line:
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash i915.enable_psr=0"
sudo update-grub

I checked the BIOS version using this command:

sudo dmidecode -s bios-version

The result was:

R0242ZR

I found this update program, but it is for windows. :

https://support.us.vaio.com/knowledge-base/vjs126-vjs146-bios-r0242zr-update-program/

Also, I’ve reached the 3-reply limit for new users, so I can’t respond further at the moment.

6

This is typically only something the hardware vendor can provide (so I guess Sony given this is a Vaio) and as you already found if they are not actually focusing on Linux you often need to use a Windows tool …

Though in times of UEFI BIOS there is often an update mechanism built into the BIOS and you only need to provide the binary file to it on a USB stick

7

I’ve resolved this problem now, after investigating some more, I found that it was a RAM issue which was causing the problem. Although the RAM was running at 3200Mhz instead of 3600Mhz but the RAM’s voltage was incorrect, was running at 1.2V instead of 1.35v, fixed the RAM issue and lock up hasn’t happened since neither errors relating to the GPU has been resolved too.

Regards

1 Like
8

I saw this issue a few months ago with a personal Framework 13 that I reinstalled w/ Ubuntu. I had installed with disk encryption. I was not able to ssh in to diagnose further as I hadn’t set that up yet. I reinstalled & disabled FDE and never saw the problem again. I just got a Framework 16 for work setup with FDE and it just happened. I definitely have the latest BIOS so there is no upgrade to potentially fix it. It hung while I was watching a video in my browser. I have a hard time believing this was shipped with a BIOS that doesn’t have the correct RAM voltage set, and it would be sad if I had to reinstall this OS without encryption just to use it.

9

So how exactly did you find that you had the very same ACPI errors causing your hang…

Just having similar (very generic) symptoms doesn’t mean you have in any way the same issue…

1 Like
10

This topic was automatically closed after 29 days. New replies are no longer allowed.