[BUG] Lockscreen died....

[Neo-29]

Describe the bug

As decribed

Steps to reproduce

wait for lockscreen to trigger

Expected behavior

working lockscreen

Screenshots

Distribution

Arch Linux

If Other, specify

No response

Additional context

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[mylinuxforwork]

Does it also happens when running hyprlock manually?

[AngeloGian]

Hello,
I've been having the same issue as @Neo-29 for at least a year now, but for me this happens only when I suspend Hyprland. Locking the screen or triggering hyprlock manually doesn't seem to trigger the issue. I am on RTX 4070, and I am currently using nvidia-open-dkms (v595.58.03-1) drivers. This could be something related to NVIDIA and hyprland in general...

[mylinuxforwork]

Yes. This is probably the reason.
Maybe you need to adjust hypridle.conf

[Neo-29]

Yes. This is probably the reason. Maybe you need to adjust hypridle.conf

It's not as I'm not using Nvidia here but same issue

Only happens on suspend

Manual triggering hyprlock is ok

[mylinuxforwork]

I have learned that suspend is an extremely individual setup and depends a lot on the hardware. On my Lenovo everything is working fine.

I would suggest that you experiment with the hypridle.conf

For the version after 2.12.2 I could provide a setting to hide the suspend button if not needed.

[AngeloGian]

Thanks for the suggestion, @mylinuxforwork . However, I tried a lot of different hypridle configurations, I’ve also tested a 'naked' configuration by stripping hypridle.conf to a completely empty general {} block with no listeners at all. The 'Oopsie Daisy' crash still happens exactly the same way, so it doesn't seem to be caused by config complexity or specific listeners. Logs indicate the compositor backend loses the DRM master lease before the power cycle completes. Even with PreserveVideoMemoryAllocations enabled and attempting manual SIGSTOP/SIGCONT process syncing, the session fails to recover. This seems to be a deeper architectural issue that hypridle adjustments can't reach (probably not dotfile-related as well...). Just wanted to flag this for further investigation.