Plasma 6 takes a long time to start both panel and background image

Booting into a recently updated Plasma 6 desktop, it now takes a long time to initialize the plasma shell. I thought this was being caused by remmina because, when the desktop did initialize it showed two systray icons. However, upon inspecting the system logs I found that the plasma shell job was failing multiple times before finally becoming available.

This isn’t really blocking as OpenSUSE allows me to alt + f2/alt + tab and return to my rofi days to start my programs as needed and just wait until everything else starts. But it may be nicer if my desktop showed up.

I extracted this dump from KSystemLog (it’s trimmed):

28/9/24 10:24	user@1000.service	Failed connect to PipeWire: Couldn't connect to PipeWire
28/9/24 10:24	user@1000.service	Choosing kwallet.portal for org.freedesktop.impl.portal.Secret via the deprecated UseIn key
28/9/24 10:24	user@1000.service	The preferred method to match portal implementations to desktop environments is to use the portals.conf(5) configuration file
28/9/24 10:24	krunner	kf.kunitconversion: currency conversion table network error "Equipo www.ecb.europa.eu no encontrado"
28/9/24 10:25	systemd	plasma-plasmashell.service: start-pre operation timed out. Terminating.
28/9/24 10:25	systemd	plasma-plasmashell.service: Control process exited, code=killed, status=15/TERM
28/9/24 10:25	systemd	plasma-plasmashell.service: Failed with result 'timeout'.
28/9/24 10:25	systemd	Failed to start KDE Plasma Workspace.
28/9/24 10:25	systemd	plasma-plasmashell.service: Scheduled restart job, restart counter is at 2.
28/9/24 10:25	systemd	Starting KDE Plasma Workspace...
28/9/24 10:25	accounts-daemon.service	g_dbus_interface_skeleton_get_object_path: assertion 'G_IS_DBUS_INTERFACE_SKELETON (interface_)' failed
28/9/24 10:26	systemd	plasma-plasmashell.service: start-pre operation timed out. Terminating.
28/9/24 10:26	systemd	plasma-plasmashell.service: Control process exited, code=killed, status=15/TERM
28/9/24 10:26	systemd	plasma-plasmashell.service: Failed with result 'timeout'.
28/9/24 10:26	systemd	Failed to start KDE Plasma Workspace.
28/9/24 10:26	systemd	plasma-plasmashell.service: Scheduled restart job, restart counter is at 3.
28/9/24 10:26	systemd	Starting KDE Plasma Workspace...
28/9/24 10:26	kded6	org.kde.plasma.dataengine.geolocation: gpsd not found
28/9/24 10:28	user@1000.service	Failed to load module "canberra-gtk-module"

Similar issues outside didn’t shed any light as:

  • Almost all libcanberra packages are installed in my system, except for deveopment ones.
  • xdg-desktop-portal-kde is installed and doesn’t have any issues
  • I’m using the default plasma login animation provided by OpenSUSE
  • I’m not using a third-party theme. I’m using breeze-dark both theme and icons.
  • Pipewire seems to be installed. I do not understand why is it failing.
  • The systemd plasma-plasmashell.service restart counter can go up to 6 until the desktop is shown.

Maybe a widget
https://bbs.archlinux.org/viewtopic.php?id=294710

1 Like

I deleted the file. But the plasma shell is still taking up a long time to initialize. However the number of retries was cut up to 3 retries. Where the third booted the Plasma Desktop

So it seems like a setting/leftover. Did you try with a fresh user profile? If that works, you may find out which app/setting/leftover breaks your system.

1 Like

I’ll check if further logs show something. I’ll clean up my ~/.config directory and update this thread if I find anything new.

I found another bug in the update. As the screensaver no longer works as well and restoring the session results in a screen text reading:

The screen locker is damaged and cannot be unlocked anymore.

To unlock it, open a new virtual terminal (for example Ctrl + Alt + F1)
login from there and run the command:

loginctl unlock-session 7

Then, logout of the terminal pressing Control + D and [redacted]

I’ll try to see if I’m missing some dependencies or if the update did mess things up.

28/9/24 17:20	kded6	org.kde.pulseaudio: context kaput
28/9/24 17:20	kded6	kf.bluezqt: PendingCall Error: "Could not activate remote peer 'org.bluez': unit failed"
28/9/24 17:20	obexd	OBEX daemon 5.77
28/9/24 17:20	systemd	Started Bluetooth OBEX service.
28/9/24 17:20	systemd	dbus-:1.2-org.kde.kded.smart@0.service: Deactivated successfully.
28/9/24 17:20	systemd	plasma-plasmashell.service: Control process exited, code=exited, status=2/INVALIDARGUMENT
28/9/24 17:20	systemd	plasma-plasmashell.service: Failed with result 'exit-code'.
28/9/24 17:20	systemd	Failed to start KDE Plasma Workspace.
28/9/24 17:20	systemd	plasma-plasmashell.service: Scheduled restart job, restart counter is at 4.
28/9/24 17:20	systemd	Starting KDE Plasma Workspace...
28/9/24 17:21	systemd	plasma-plasmashell.service: start-pre operation timed out. Terminating.
28/9/24 17:21	systemd	plasma-plasmashell.service: Control process exited, code=killed, status=15/TERM
28/9/24 17:21	systemd	plasma-plasmashell.service: Failed with result 'timeout'.
28/9/24 17:21	systemd	Failed to start KDE Plasma Workspace.
28/9/24 17:21	systemd	plasma-plasmashell.service: Scheduled restart job, restart counter is at 5.
28/9/24 17:21	systemd	Starting KDE Plasma Workspace...
28/9/24 17:22	systemd	plasma-plasmashell.service: start-pre operation timed out. Terminating.
28/9/24 17:22	systemd	plasma-plasmashell.service: Control process exited, code=killed, status=15/TERM
28/9/24 17:22	systemd	plasma-plasmashell.service: Failed with result 'timeout'.
28/9/24 17:22	systemd	Failed to start KDE Plasma Workspace.
28/9/24 17:22	systemd	plasma-plasmashell.service: Scheduled restart job, restart counter is at 6.
28/9/24 17:22	systemd	Starting KDE Plasma Workspace...
28/9/24 17:22	kded6	org.kde.plasma.dataengine.geolocation: gpsd not found

Such is life with all rolling releases.

Fixed by deleting some kde/plasma files inside ~/.config

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.