[RESPONDED][Framework Laptop 13 AMD on Arch] System fails to shut down

I’ve had this perplexing issue over the last few days where the system just doesn’t shut down. I’ve had this issue before intermittently and it’s been cropping up consistently as of late.

Nothing shows in journalctl:

May 23 08:18:37 phobos systemd[1]: Shutting down.
May 23 08:18:37 phobos systemd-shutdown[1]: Syncing filesystems and block devices.
May 23 08:18:37 phobos systemd-shutdown[1]: Sending SIGTERM to remaining processes...
May 23 08:18:37 phobos systemd-journald[684]: Journal stopped
-- Boot 7b16315f238c40308606eda64cb8da21 --

Kernel: 6.6.31-2-lts (linux-lts due to wifi issues)
systemd 255 (255.6-2-arch)
BIOS: 3.05

Has anyone else had this issue? It occurs both from hibernation and a simple shutdown command causes this issue. Thanks for any help.

Hi @WrenPorcaro ,

Can you confirm if it’s just happening on arch with a 6.6,31 kernel or also happening on say Fedora 39 Live which is supported distro?

cheers! :slight_smile:

I think this is a kernel bug introduced in 6.6.31. I was running 6.6.30 fine for a while (need 6.6 because 6.7+ has a bug related to the wifi module shipped by Framework). As soon as I upgraded to 6.6.31 I started seeing this issue.

Here is a link to a Void GH issue tracking the same problem: Cannot shutdown laptop with linux6.6-6.6.31_1 · Issue #50417 · void-linux/void-packages · GitHub
That link also mentions Arch, and I’m on Nix. It seems like a kernel issue, not a problem with a particular distro. I’m also on a Framework 16, so it may be related to one of the AMD updates in 6.6.31.

It seems like the workaround for now is to rollback to 6.6.30 or use 6.7+ if you are able to.

1 Like

It seems this is a known issue on this particular arch kernel. I had the same problem, then went back to 6.6.30-2-lts, and now all is fine.

Me too. I’m running on nixos. I have the same issue with 6.6.31. Rolling back to 6.630 is fine.

Looks like the breaking commit between 6.6.30 and 6.6.31 has been found. More details on the above linked Void linux issue report


Has anyone noticed occasional high power draw from Intel wifi adapter since this update?