You are not logged in.
Apologies if this is the incorrect category for this post. If it is in the wrong category, please let me know where I should post it and I will attempt to move it there. This is my first full post the forums.
I first noticed this issue a couple days ago and have gone through various troubleshooting steps to try to resolve.
The issue first presented itself as my computer did not want to "wake" the next day after going to bed. (2025-02-12)
Notes:
I do not fully "shutdown / suspend" my PC. This is a precaution due to a "flaky" water pump that did not want to restart from power-off one time.
Only the screens will go black / shut off after some time. (I think this is called dpms? It is accomplished via Hyprland's hypridle program.)
"wake" means I tried shaking my mouse / pressing keys on my keyboard to turn on my screens and let me login via screen lock. (hyprlock)
I ended up hitting the reset button on my PC to power cycle / restart the machine. Boot process process and things seemed normal.
I had some package updates. I did updates via `pacman -Syu` and rebooted. No issues booting, continued using PC through the day.
2025-02-13 - Same issue as the day before; PC did not want to "wake" normally. Hit reset button, booted fine, no issues.
Later that day, I had several package updates and I again installed via `pacman -Syu`. Rebooted normally.
However, shortly (~15-30min?) my computer just "locked up". Only applications open were Firefox and Alacritty terminal at that time.
Clarifications:
"locked up" in this instance means I still had video display on both monitors. But none of my inputs were resulting in any actions.
Nothing on my screens were updating / refreshing. I have a Waybar script that shows fan speed for example, and it was frozen / not updating.
I hit the reset button again. Rebooted to desktop without issue as before. At this point, I started to troubleshoot in earnest. (See Troubleshooting Steps below). However, while doing this troubleshooting, my PC kept doing this "locking up" roughly every 15-20 minutes...
I started with a `journalctl -b -1` and looked at the logs leading up to the end of that boot. I saw nothing higher than an "INFO" level, and all "INFO" messages were normal/expected. I can supply these messages if requested. Just let me know how I should format them. (E.g., What the journalctl args should be.) After every "lock up" I would check the logs and still see nothing unexpected.
At this point, my PC was unusable. So I decided to take a second look at the packages updated and see if I could isolate the issue.
I used
grep -i upgrade /var/log/pacman.logto figure out what was updated 2025-02-13. These were the packages updated that day:
[2025-02-13T18:45:04-0500] [PACMAN] starting full system upgrade
[2025-02-13T18:45:08-0500] [ALPM] upgraded aardvark-dns (1.13.1-1 -> 1.14.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded netavark (1.13.1-1 -> 1.14.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded containers-common (1:0.61.1-1 -> 1:0.62.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded curl (8.12.0-3 -> 8.12.1-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded lib32-curl (8.12.0-1 -> 8.12.1-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libblockdev (3.2.1-3 -> 3.3.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libblockdev-crypto (3.2.1-3 -> 3.3.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libblockdev-fs (3.2.1-3 -> 3.3.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libblockdev-loop (3.2.1-3 -> 3.3.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libblockdev-mdraid (3.2.1-3 -> 3.3.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libblockdev-nvme (3.2.1-3 -> 3.3.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libblockdev-part (3.2.1-3 -> 3.3.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libblockdev-swap (3.2.1-3 -> 3.3.0-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded libcurl-gnutls (8.12.0-3 -> 8.12.1-1)
[2025-02-13T18:45:08-0500] [ALPM] upgraded tailscale (1.80.0-1 -> 1.80.2-1)
[2025-02-13T21:51:34-0500] [PACMAN] starting full system upgrade
[2025-02-13T21:51:36-0500] [ALPM] upgraded gtk3 (1:3.24.48-1 -> 1:3.24.48-2)
[2025-02-13T21:51:36-0500] [ALPM] upgraded gvfs (1.56.1-2 -> 1.56.1-3)
[2025-02-13T21:51:36-0500] [ALPM] upgraded gvfs-smb (1.56.1-2 -> 1.56.1-3)
[2025-02-13T21:51:36-0500] [ALPM] upgraded liblc3 (1.1.2-1 -> 1.1.3-1)
[2025-02-13T21:51:36-0500] [ALPM] upgraded nodejs (23.7.0-1 -> 23.8.0-1)I then edited `/etc/pacman.d/mirrorlist` by commenting out all the servers and adding:
Server=https://archive.archlinux.org/repos/2025/02/12/$repo/os/$archI then did a "downgrade" via `pacman -Syyuu` and rebooted.
System rebooted and I was able to log in and reach desktop as normal. However, the system locked up again within the similar time frame of ~20min. Reset PC, booted, and decided to downgrade another day back.
Again I used
grep -i upgrade /var/log/pacman.logto figure out what was updated 2025-02-12.
[2025-02-12T17:41:02-0500] [PACMAN] starting full system upgrade
[2025-02-12T17:41:06-0500] [ALPM] upgraded aom (3.11.0-1 -> 3.12.0-1)
[2025-02-12T17:41:07-0500] [ALPM] upgraded go (2:1.23.6-1 -> 2:1.24.0-1)
[2025-02-12T17:41:08-0500] [ALPM] upgraded openssl (3.4.0-1 -> 3.4.1-1)
[2025-02-12T17:41:08-0500] [ALPM] upgraded gpgme (1.24.1-3 -> 1.24.2-1)
[2025-02-12T17:41:08-0500] [ALPM] upgraded lib32-openssl (1:3.4.0-1 -> 1:3.4.1-1)
[2025-02-12T17:41:08-0500] [ALPM] upgraded node-gyp (11.1.0-2 -> 11.1.0-3)
[2025-02-12T17:41:08-0500] [ALPM] upgraded nvidia-open (570.86.16-5 -> 570.86.16-6)
[2025-02-12T17:41:08-0500] [ALPM] upgraded podman (5.3.2-1 -> 5.4.0-1)
[2025-02-12T17:41:08-0500] [ALPM] upgraded python-click (8.1.7-4 -> 8.1.8-1)
[2025-02-12T17:41:08-0500] [ALPM] upgraded python-pip (25.0-1 -> 25.0.1-1)
[2025-02-12T17:41:08-0500] [ALPM] upgraded xkeyboard-config (2.43-1 -> 2.44-1)I then edited `/etc/pacman.d/mirrorlist` by commenting out all the servers and adding:
Server=https://archive.archlinux.org/repos/2025/02/11/$repo/os/$archI then did a "downgrade" via `pacman -Syyuu` and rebooted.
System stopped "locking up" after the short duration. I took some notes, browsed the web, and then went to bed for the evening without the system having issue.
This morning when I tried to login, I encountered the same issue that started this series. PC wouldn't "wake" from monitor off.
At this point, I'm not sure where to look or what to do next... I could try downgrading further, but I don't know how to replicate the "wake" issue intentionally. This is just a band-aid solution though as it's preventing me from upgrading things. Additionally, not seeing anything in the journal is leaving me without something to track down.
Any help with this would be appreciated. If you need further information, please let me know, but please be specific / provide steps/commands to get this information. (I am still a bit of a novice when it comes to Linux)
Thank you.
OS: Arch Linux x86_64
Kernel: Linux 6.13.2-arch1-1
Uptime: 2 hours, 8 mins
Packages: 857 (pacman), 18 (flatpak)
Shell: bash 5.2.37
Display (AORUS FO48U): 3840x2160 @ 120 Hz in 32" [External]
Display (DELA0FA): 3440x1440 @ 100 Hz in 34" [External]
WM: Hyprland 0.47.2 (Wayland)
Theme: catppuccin-mocha-blue [Qt], Adwaita-dark [GTK2/3]
Icons: Papirus-Dark [GTK2/3]
Font: Noto Sans (12pt) [Qt], Cantarell (11pt) [GTK2/3]
Cursor: catppuccin-mocha-dark (35px)
Terminal: tmux 3.5a
CPU: AMD Ryzen 9 5950X (32) @ 5.08 GHz
GPU: NVIDIA GeForce RTX 3090 [Discrete]
Memory: 4.65 GiB / 31.25 GiB (15%)
Swap: 0 B / 38.00 GiB (0%)
Disk (/): 33.72 GiB / 97.87 GiB (34%) - ext4
Disk (/home): 510.34 GiB / 884.81 GiB (58%) - ext4
Local IP (enp5s0): 192.168.1.10/24
Locale: en_US.UTF-8Last edited by Ac5000 (2025-03-13 22:58:07)
Offline
So it took me ~3 weeks, but I figured out a solution.
What I discovered after making the initial post was that I was experiencing a "true" crash of the system. I learned about the Kernel (SysRq) keybinding and was never able to get it to work when my system was locking up. From searching around online I determined that this meant my system was crashing fully.
I eventually found this wiki section on Ryzen system halts.
I remembered when I was running Windows and had just put the system together that I had memory stability issues. I ended up setting PBO curve to +4 initially, and I finally saw positive results. System was able to idle throughout a night without locking up.
It didn't make it through the next night though. So I bumped PBO curve to +6 and it's remained solid ever since. (Set on 2025-03-06)
My only assumption is that something that was patched/updated was causing my system to use even less resources when idle and the low voltage was causing system halts.
Last edited by Ac5000 (2025-03-13 22:57:09)
Offline