https://bugs.kde.org/show_bug.cgi?id=521871
Bug ID: 521871
Summary: ext_idle_notifier_v1 / org_kde_kwin_idle never fires;
screen never dims, turns off, or auto-locks from
genuine inactivity
Classification: Plasma
Product: kwin
Version First 6.7.0
Reported In:
Platform: Arch Linux
OS: Linux
Status: REPORTED
Severity: major
Priority: NOR
Component: wayland-generic
Assignee: [email protected]
Reporter: [email protected]
Target Milestone: ---
Created attachment 193442
--> https://bugs.kde.org/attachment.cgi?id=193442&action=edit
powerdevilrc
***
If you're not sure this is actually a bug or valid feature request, instead
post about it at https://discuss.kde.org
If you're reporting a crash, attach a backtrace with debug symbols; see
https://community.kde.org/Guidelines_and_HOWTOs/Debugging/How_to_create_useful_crash_reports
Please remove this comment after reading and before submitting - thanks!
***
DESCRIPTION
Since updating to Plasma 6.7, the system never detects idle state at all.
Screen dimming, "turn off screen" (manual DPMS works fine when triggered
directly), and "lock screen after inactivity" all fail to trigger, even after
leaving the system completely untouched for several minutes. This is not
specific to the lock screen, it affects ALL idle-triggered behavior, with or
without locking.
STEPS TO REPRODUCE
1. System Settings > Power Management > Energy Saving: set "Dim screen" and
"Turn off screen" to short timeouts (e.g. 20s) on AC power.
2. System Settings > Screen Locking: set "Lock screen automatically after
inactivity" to 1 minute.
3. Close all applications. Do not touch mouse/keyboard/any input device.
4. Wait several minutes without any input.
OBSERVED RESULT
Screen never dims.
Screen never turns off (despite `kscreen-doctor --dpms off` working correctly
when run manually).
Screen never auto-locks from inactivity.
`loginctl show-session $XDG_SESSION_ID -p IdleHint` stays at IdleHint=no
indefinitely (though this may reflect a separate logind mechanism rather than
KWin's idle-notify protocol directly).
A third-party client (`swayidle`) registered an idle timeout directly against
KWin's idle-notify protocol and never received a notification, even though KWin
advertises both org_kde_kwin_idle (v1) and ext_idle_notifier_v1 (v2) per
`wayland-info`.
EXPECTED RESULT
Screen dims/turns off per configured timeouts, and the session auto-locks per
the configured inactivity timeout.
SOFTWARE/OS VERSIONS
Operating System: Arch Linux
KDE Plasma Version: 6.7.0
KDE Frameworks Version: 6.27.0
Qt Version: 6.11.1
Kernel Version: 7.0.12-arch1-1 (64-bit)
Graphics Platform: Wayland
Processors: 16 × AMD Ryzen 7 9800X3D 8-Core Processor
Memory: 32 GiB of RAM (30.3 GiB usable)
Graphics Processor 1: AMD Radeon RX 9070 XT
Graphics Processor 2: AMD Ryzen 7 9800X3D 8-Core Processor
ADDITIONAL INFORMATION
Manual DPMS control works correctly (`kscreen-doctor --dpms off` successfully
blanks the display).
powerdevilrc contains correct, valid config
(TurnOffDisplayIdleTimeoutWhenLockedSec=20, etc. under [AC][Display]).
No idle/sleep inhibitors present (checked via `systemd-inhibit --list` and the
Plasma power-management tray applet, which shows no "applications preventing
power management" message).
Wallpaper is a static image, not animated/video — ruled out continuous
compositor repaints from that source.
KWin's own CPU usage while idle is low (~2%), no runaway repaint loop observed
via `top`.
`sudo libinput debug-events` shows zero input events during the observed idle
window — ruling out phantom/spurious input from hardware (wireless
mouse/keyboard dongles, etc.).
Tested with swayidle directly against KWin's idle-notify protocol:
$ swayidle -d -w timeout 8 'echo IDLE-FIRED' resume 'echo RESUMED'
2026-06-20 09:40:25 - Register idle timeout at 8000 ms
2026-06-20 09:40:25 - Setup idle
2026-06-20 09:40:25 - Enable idle timeouts
2026-06-20 09:40:25 - Register with timeout: 8000
Timeout registered without error against the protocol; after waiting well past
the 8-second timeout with zero input, no "IDLE-FIRED" was ever printed.
This affects dim, turn-off, and auto-lock identically and simultaneously, which
suggests the underlying idle-notify signal from KWin to clients is not being
delivered at all, rather than an issue isolated to PowerDevil or
kscreenlocker's handling of it.
Note: this also has a security implication beyond inconvenience as it silently
defeats the "lock screen after inactivity" protection with no error or
indication to the user that idle-triggered locking isn't working.
--
You are receiving this mail because:
You are watching all bug changes.
