You are not logged in.
Hey. I've been having a problem for a while now with systemd hanging on shutdowns and reboots, running a stop job. I looked into journalctl -rb -1 and found that the hang is caused by systemd waiting its default timeout for kwin_x11 and QDBusConnection to shut down, before killing them:
Oct 04 22:17:31 lenovoarch systemd[1]: session-2.scope: Failed with result 'timeout'.
Oct 04 22:17:31 lenovoarch systemd[1]: session-2.scope: Killing process 6987 (QDBusConnection) with signal SIGKILL.
Oct 04 22:17:31 lenovoarch systemd[1]: session-2.scope: Killing process 6980 (kwin_x11) with signal SIGKILL.
Oct 04 22:17:31 lenovoarch systemd[1]: session-2.scope: Stopping timed out. Killing.
I can sidestep the issue a bit by lowering the timeout, but it's only a workaround. It doesn't happen every time (usually the less time since boot the less likely) but it's very frequent. Reinstalling Arch Linux has not solved the issue, and I haven't found a solution anywhere else. What could be causing this?
Edit: I'll provide a full log once it happens again cleanly. The last boot's log is a little polluted by me trying to switch to TTY2 while waiting for the shutdown, which seems to spam the log with messages of org.kde.kglobalaccel failing to start.
Last edited by limelier (2020-10-05 16:20:12)
Offline
https://gist.github.com/limelier/7694c3 … 97cfe84d75
Here's a full log. I'm not sure about the TTY2 thing anymore, the messages seem to still be there.
Offline
I'm having the same problem quite frequently
Offline
Same for me started, not a long ago after an update, alas I cannot say which update affected it as I haven't been updating for one month, and the updated and bug appeared. Any way to analyze the source of issue?
Offline
Your mention of kglobalaccel reminded me of a recent issue on kglobalaccel. It stuck logging in sometimes, so I think it might also stuck logging out: https://bugs.kde.org/show_bug.cgi?id=429426
Offline
I also have this issue - started seeing it when I switched from GNOME to KDE (though I only spent a few days using GNOME), so I reinstalled Arch completely with KDE after a week of putting up with it, and still see this issue. It's quite frequent, but like #1 states, its not every time and less likely the shorter the duration since boot. Journalctl shows kwin_x11 hanging but not QDBusConnection - kwin_x11 is the first entry after the timeout finishes. Does anyone know if kwin_x11 is actually the culprit? Thank you
Offline
I've also been getting this issue albeit with slightly different hanging processes.
Apr 12 15:58:07 arch01 systemd[1]: session-2.scope: Stopping timed out. Killing.
Apr 12 15:58:07 arch01 systemd[1]: session-2.scope: Killing process 1724 (kwin_x11) with signal SIGKILL.
Apr 12 15:58:07 arch01 systemd[1]: session-2.scope: Killing process 1759 (QQmlThread) with signal SIGKILL.
Apr 12 15:58:07 arch01 systemd[1]: session-2.scope: Killing process 1994 (CPMMListener) with signal SIGKILL.
Apr 12 15:58:07 arch01 systemd[1]: session-2.scope: Killing process 2004 (kwin_x11) with signal SIGKILL.
Apr 12 15:58:07 arch01 systemd[1]: session-2.scope: Failed with result 'timeout'.
Curious if it is kwin_x11 or something else.
Offline
Curious if it is kwin_x11 or something else.
Having the same issue, I disabled kwin from starting and haven't had a shutdown hang since.
$20 Free Credit Hetzner - https://hetzner.cloud/?ref=fuVilhv403fA
Offline
Same issue here. I don't want to disable desktop effects because I use them a lot for desktop grid and windows presentation.
EDIT : Good workaround here : https://forum.manjaro.org/t/kwin-x11-no … wn/73538/2
Last edited by zoui (2021-07-21 19:02:14)
Offline
Hi there - I realize that this is a few months old, but there is no solution yet, and I seem to have the exact same problem:
Nov 11 16:47:37 iris sddm[676]: Signal received: SIGTERM
Nov 11 16:47:38 iris sddm[676]: QProcess: Destroyed while process ("/usr/lib/sddm/sddm-helper") is still running.
Nov 11 16:47:38 iris systemd[1]: sddm.service: Consumed 18.465s CPU time.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Stopping timed out. Killing.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Killing process 953 (kwin_x11) with signal SIGKILL.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Killing process 1005 (n/a) with signal SIGKILL.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Killing process 1138 (CPMMListener) with signal SIGKILL.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Killing process 2238 (n/a) with signal SIGKILL.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Failed with result 'timeout'.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Consumed 1min 10.775s CPU time.
Any progress on that issue on your end?
(The workaround above just bypasses the 90 second timeout and kills kwin directly, I'd like to avoid that...)
Thanks!
- Alex
Last edited by Alex G (2021-11-12 15:31:54)
Offline
Any progress on that issue on your end? - Alex
Disabled my workaround service and can confirm i'm still having this issue.
$20 Free Credit Hetzner - https://hetzner.cloud/?ref=fuVilhv403fA
Offline
Did this ever get a resolution? I have been battling with the same issue on Manjaro.
Nov 11 16:47:37 iris sddm[676]: Signal received: SIGTERM
Nov 11 16:47:38 iris sddm[676]: QProcess: Destroyed while process ("/usr/lib/sddm/sddm-helper") is still running.
Nov 11 16:47:38 iris systemd[1]: sddm.service: Consumed 18.465s CPU time.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Stopping timed out. Killing.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Killing process 953 (kwin_x11) with signal SIGKILL.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Killing process 1005 (n/a) with signal SIGKILL.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Killing process 1138 (CPMMListener) with signal SIGKILL.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Killing process 2238 (n/a) with signal SIGKILL.
Nov 11 16:49:06 iris systemd[1]: session-1.scope: Failed with result 'timeout'.
Offline
Offline