https://bugs.kde.org/show_bug.cgi?id=416739
Alain Knaff <k...@kde.lka.org.lu> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |k...@kde.lka.org.lu
--- Comment #40 from Alain Knaff <k...@kde.lka.org.lu> ---
I am experiencing the same problem.
Debian Version: 10
Kernel: 4.19.0-10-amd64
Plasma Version: 5.14.5-1
Desktop: Terra
Monitors: 2 Terra 2463W monitors, connected via Display Port
These are monitors that have 2 inputs each, Display Port and HDMI. On Mondays
and Fridays, I work from home, and connect my work laptop to the HDMI inputs.
The home computer is still connected to the display port inputs, but the
monitors are switched to HDMI input.
I've got the impression that the problem mostly occurs when the work computer
is connected, and has the screens.
When that happens, the ~/.config/plasmashellrc gets an extra :0.0 entry for the
screen connector:
Before:
[ScreenConnectors]
0=DP-1
1=DP-2
After:
[ScreenConnectors]
0=DP-1
1=:0.0
2=DP-2
=> i.e. the bogus sneaked right into the middle of the real entries, causing
DP-2 to get renumbered from 1 to 2.
However, the lastScreen settings in
.config/plasma-org.kde.plasma.desktop-appletsrc would not get updated, i.e. all
panels on DP-2 would still keep lastScreen=1, causing them to disappear (as
plasma tried to display them on the non-existant :0.0 screen)
N.B. monitors just going to sleep don't cause the problem by itself (unless
~/.config/plasmashellrc is already corrupted), it actually needs monitor's
input to be switched to another computer to make .plasmashellrc to become
corrupted. But once plasmashellrc *is* corrupted, then shenanigans happen
whenever monitors go to sleep.
I've now write-protected the plasmashellrc file, let's see whether this stops
the problem from happening next Friday :-)
--
You are receiving this mail because:
You are watching all bug changes.
