This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed 18.380s CPU time. May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with result 'timeout'. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main process exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State 'stop-sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/0. May 01 10:43:03 Bree systemd[1]: run-user-0.mount: Deactivated successfully. May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/990. May 01 10:43:03 Bree systemd[1]: run-user-990.mount: Deactivated successfully. May 01 10:43:03 Bree systemd[1]: Stopped Network Manager Wait Online. ...
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
poc
On Sunday, 1 May 2022 12:14:00 BST Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed 18.380s CPU time.
May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with result 'timeout'.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main
process exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State 'stop-sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
There was an issue related to kde-connect that caused a similar issue. This has since been fixed though.
Do you use kde-connect by any chance?
Colin
On Sun, 2022-05-01 at 12:48 +0100, Colin J Thomson wrote:
On Sunday, 1 May 2022 12:14:00 BST Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed 18.380s CPU time.
May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with result 'timeout'.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main
process exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State 'stop- sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
There was an issue related to kde-connect that caused a similar issue. This has since been fixed though.
Do you use kde-connect by any chance?
No. I don't think I've every used it and certainly don't now.
poc
On 5/1/22 09:08, Patrick O'Callaghan wrote:
On Sun, 2022-05-01 at 12:48 +0100, Colin J Thomson wrote:
On Sunday, 1 May 2022 12:14:00 BST Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed 18.380s CPU time.
May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with result 'timeout'.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main
process exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State 'stop- sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
There was an issue related to kde-connect that caused a similar issue. This has since been fixed though.
Do you use kde-connect by any chance?
No. I don't think I've every used it and certainly don't now.
Try pressing the up arrow when the gui shutdown screen with the spinner is up. That should show you what process is taking so long. I have seen dnf/policykit stuff still running before on a long shutdown.
poc _______________________________________________ kde mailing list -- kde@lists.fedoraproject.org To unsubscribe send an email to kde-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/kde@lists.fedoraproject.org Do not reply to spam on the list, report it: https://pagure.io/fedora-infrastructure
On Sun, 2022-05-01 at 09:18 -0300, Patrick Boutilier wrote:
On 5/1/22 09:08, Patrick O'Callaghan wrote:
On Sun, 2022-05-01 at 12:48 +0100, Colin J Thomson wrote:
On Sunday, 1 May 2022 12:14:00 BST Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed 18.380s CPU time.
May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with result 'timeout'.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main
process exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State 'stop- sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
There was an issue related to kde-connect that caused a similar issue. This has since been fixed though.
Do you use kde-connect by any chance?
No. I don't think I've every used it and certainly don't now.
Try pressing the up arrow when the gui shutdown screen with the spinner is up. That should show you what process is taking so long. I have seen dnf/policykit stuff still running before on a long shutdown.
"A stop job is running for User Manager for UID 1000 (1m 20s...)"
UID 1000 is my account, and the delay counter matches the interval in the journalctl record.
poc
On Sun, May 1, 2022 at 6:35 AM Patrick O'Callaghan pocallaghan@gmail.com wrote:
On Sun, 2022-05-01 at 09:18 -0300, Patrick Boutilier wrote:
On 5/1/22 09:08, Patrick O'Callaghan wrote:
On Sun, 2022-05-01 at 12:48 +0100, Colin J Thomson wrote:
On Sunday, 1 May 2022 12:14:00 BST Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed 18.380s CPU time.
May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with result 'timeout'.
May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main
process exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State 'stop- sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
There was an issue related to kde-connect that caused a similar issue. This has since been fixed though.
Do you use kde-connect by any chance?
What was the solution/fix for kde-connect? This might be related, and might need a similar fix.
No. I don't think I've every used it and certainly don't now.
Try pressing the up arrow when the gui shutdown screen with the spinner is up. That should show you what process is taking so long. I have seen dnf/policykit stuff still running before on a long shutdown.
"A stop job is running for User Manager for UID 1000 (1m 20s...)"
UID 1000 is my account, and the delay counter matches the interval in the journalctl record.
poc
It's related to this.[1][2] It's related to the kernel version, but from what I see, everyone seems to just want to fix the symptom, and not fix the real problem. If you are running CentOS Stream 8, the problem is much worse.[3]
[1] - https://bugs.kde.org/show_bug.cgi?id=445449 [2] - https://bugs.kde.org/show_bug.cgi?id=445385 [3] - https://bugzilla.redhat.com/show_bug.cgi?id=2043771
Troy
On Mon, 2022-05-02 at 06:21 -0700, Troy Dawson wrote:
Try pressing the up arrow when the gui shutdown screen with the spinner is up. That should show you what process is taking so long. I have seen dnf/policykit stuff still running before on a long shutdown.
"A stop job is running for User Manager for UID 1000 (1m 20s...)"
UID 1000 is my account, and the delay counter matches the interval in the journalctl record.
poc
It's related to this.[1][2] It's related to the kernel version, but from what I see, everyone seems to just want to fix the symptom, and not fix the real problem. If you are running CentOS Stream 8, the problem is much worse.[3]
[1] - https://bugs.kde.org/show_bug.cgi?id=445449 [2] - https://bugs.kde.org/show_bug.cgi?id=445385
I should have mentioned that I'm not using Wayland. I'm on X11 and will stay there until Wayland gets session restore working correctly.
[3] - https://bugzilla.redhat.com/show_bug.cgi?id=2043771
That bug seems to be unrelated to what I'm asking about.
poc
On Sun, 2022-05-01 at 12:14 +0100, Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed > 18.380s CPU time. May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with > result 'timeout'. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main process > exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing > process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing > process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State > 'stop-sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/0. May 01 10:43:03 Bree systemd[1]: run-user-0.mount: Deactivated > successfully. May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/990. May 01 10:43:03 Bree systemd[1]: run-user-990.mount: Deactivated > successfully. May 01 10:43:03 Bree systemd[1]: Stopped Network Manager Wait Online. ...
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
I posted the above a couple of months back, but haven´t seen any resolution so far. Google finds a few references to similar issues but they seem related to Wayland, while I'm using exclusively X11.
The problem appears to be related to plasma-kded.service and happens even when I explicitly log out before shutting down.
I've reported it at: https://bugs.kde.org/show_bug.cgi?id=456713
Any
On Thu, 2022-07-14 at 17:36 +0100, Patrick O'Callaghan wrote:
On Sun, 2022-05-01 at 12:14 +0100, Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed
18.380s CPU time. May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with > result 'timeout'. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main process > exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing
process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing
process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State > 'stop-sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/0. May 01 10:43:03 Bree systemd[1]: run-user-0.mount: Deactivated > successfully. May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/990. May 01 10:43:03 Bree systemd[1]: run-user-990.mount: Deactivated
successfully. May 01 10:43:03 Bree systemd[1]: Stopped Network Manager Wait Online. ...
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
I posted the above a couple of months back, but haven´t seen any resolution so far. Google finds a few references to similar issues but they seem related to Wayland, while I'm using exclusively X11.
The problem appears to be related to plasma-kded.service and happens even when I explicitly log out before shutting down.
I've reported it at: https://bugs.kde.org/show_bug.cgi?id=456713
1) I tried logging out and manually killing the kded5 process from another console (using -9 as it ignored the standard kill). I then rebooted and saw no delay.
2) New session: logged out and killed kded5 as before, then tried logging in again. The login screen (SDDM) hung until I manually killed the kded5 process from another console (I didn't wait for it to time out), at which point the SDDM login continued.
3) Clearly the kded5 process should die when I log out. I have "KillUserProcesses=yes" in /etc/systemd/logind.conf, so why isn't this happening automatically? Is kded5 not a user process? If not, how can I ensure that this happens? There doesn't seem to be any valid reason for kded5 to be still active when the user isn't logged in.
poc
I vaguely recall that at some point KillUserProcesses=yes was not working. Maybe Google for that
On Sat, Jul 16, 2022, 7:04 AM Patrick O'Callaghan pocallaghan@gmail.com wrote:
On Thu, 2022-07-14 at 17:36 +0100, Patrick O'Callaghan wrote:
On Sun, 2022-05-01 at 12:14 +0100, Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed
18.380s CPU time. May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with > result 'timeout'. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main process > exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing
process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing
process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State > 'stop-sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/0. May 01 10:43:03 Bree systemd[1]: run-user-0.mount: Deactivated > successfully. May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/990. May 01 10:43:03 Bree systemd[1]: run-user-990.mount: Deactivated
successfully. May 01 10:43:03 Bree systemd[1]: Stopped Network Manager Wait Online. ...
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
I posted the above a couple of months back, but haven´t seen any resolution so far. Google finds a few references to similar issues but they seem related to Wayland, while I'm using exclusively X11.
The problem appears to be related to plasma-kded.service and happens even when I explicitly log out before shutting down.
I've reported it at: https://bugs.kde.org/show_bug.cgi?id=456713
- I tried logging out and manually killing the kded5 process from
another console (using -9 as it ignored the standard kill). I then rebooted and saw no delay.
- New session: logged out and killed kded5 as before, then tried
logging in again. The login screen (SDDM) hung until I manually killed the kded5 process from another console (I didn't wait for it to time out), at which point the SDDM login continued.
- Clearly the kded5 process should die when I log out. I have
"KillUserProcesses=yes" in /etc/systemd/logind.conf, so why isn't this happening automatically? Is kded5 not a user process? If not, how can I ensure that this happens? There doesn't seem to be any valid reason for kded5 to be still active when the user isn't logged in.
poc _______________________________________________ kde mailing list -- kde@lists.fedoraproject.org To unsubscribe send an email to kde-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/kde@lists.fedoraproject.org Do not reply to spam on the list, report it: https://pagure.io/fedora-infrastructure
On Sat, 2022-07-16 at 13:15 -0400, Neal Becker wrote:
I vaguely recall that at some point KillUserProcesses=yes was not working. Maybe Google for that
I found: https://bugzilla.redhat.com/show_bug.cgi?id=1341837
which dates from nearly 6 years ago and relates to gdm. It doesn't appear to have been resolved.
It looks like this isn't a KDE problem as such, so I'll take it to the Fedora User list.
poc
On Thu, 2022-07-14 at 17:36 +0100, Patrick O'Callaghan wrote:
On Sun, 2022-05-01 at 12:14 +0100, Patrick O'Callaghan wrote:
This has been happening for IIRC) several months now at least:
$ journalctl --boot=-1 --reverse ... May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Consumed
18.380s CPU time. May 01 10:44:26 Bree systemd[1210]: Stopped KDE Daemon. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Failed with > result 'timeout'. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Main process > exited, code=killed, status=9/KILL May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing
process 1791 (ProfilesWatcher) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: Killing
process 1722 (kded5) with signal SIGKILL. May 01 10:44:26 Bree systemd[1210]: plasma-kded.service: State > 'stop-sigterm' timed out. Killing. <-----------------------* May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/0. May 01 10:43:03 Bree systemd[1]: run-user-0.mount: Deactivated > successfully. May 01 10:43:03 Bree systemd[1]: Unmounted /run/user/990. May 01 10:43:03 Bree systemd[1]: run-user-990.mount: Deactivated
successfully. May 01 10:43:03 Bree systemd[1]: Stopped Network Manager Wait Online. ...
Any idea why? I see some apparently related BZ entries dating back a couple of years, but nothing recently.
I posted the above a couple of months back, but haven´t seen any resolution so far. Google finds a few references to similar issues but they seem related to Wayland, while I'm using exclusively X11.
The problem appears to be related to plasma-kded.service and happens even when I explicitly log out before shutting down.
I've reported it at: https://bugs.kde.org/show_bug.cgi?id=456713
See the above URL for details, but basically the Users list suggested a workaround to shorten the timeout.
That still doesn't explain why kded is hanging. It's clearly waiting for *something* but there's no indication of exactly what.
poc
-
Colin J Thomson -
Neal Becker -
Patrick Boutilier -
Patrick O'Callaghan -
Troy Dawson