Monitor Reconnects regularly when using Kanshi (wlr-output-management) (May Be due to VRR)

[5ila5]

I use Kanshi to configure my monitors: My main monitor (DP-3 in config) regularly reconnects (Gets Black for 1-3 seconds) I did not find a reliable way to reproduce it, it happens anywhere between a few seconds after each other up to an hour.

It also happens after switching to a non vrr profile (novrr in config [kanshictl switch novrr]) so I'm not sure if this is a vrr related issue. But it only happens on my VRR capable monitor, so it might be.

It does not happen when not using Kanshi but the Hyprland monitor settings to configure my monitors (No matter if I enable/disable vrr)

The Hyprland logs show that hyprland sets the vrr and all the other variables as expected by my Kanshi config.

I created a DRM debugging log (as described here) where I put messages before starting Kanshi, but I don't know if this helps, but I'll attach it:

(The monitor Reconnect did not happen during creation of this dmesg log, but I started kanshi, so there might or might not be something hidden in there)

Messages inserted by me

• Starting kanshi...
• killing kanshi...
• stopping Hyperland

dmesg.log

Kanshi config:

profile default {
	output DP-2 enable position 0,540
	output DP-3 enable position 1920,300 mode 3840x2160@144Hz adaptive_sync on scale 1.25
	output DP-1 enable position 4992,204 transform 90
  exec xrandr --output DP-3 --primary
}

profile novrr {
	output DP-2 enable position 0,540
	output DP-3 enable position 1920,300 mode 3840x2160@144Hz adaptive_sync off scale 1.25
	output DP-1 enable position 4992,204 transform 90
  exec xrandr --output DP-3 --primary
}

The Hyprland monitor configuration where the reconnection issue seems not to happen:

monitor = DP-2,1920x1080,0x540,1
monitor = DP-3,3840x2160,1920x300,1.25,vrr,1
monitor = DP-1,1920x1080,4992x204,1,transform,1
exec-once= xrandr --output DP-3 --primary

Let me know if there is any other debugging step I could try.