Closed mfisher31 closed 1 month ago
I think just got this issue, with my freshly build version on current Debian Sid. In my case, the whole Gnome session crashes though (no nvidia involved, so it might be different).
It seems to me it's an issue that only occurs when using Wayland. In that case, the session (Gnome in my case) crashes completely with this error:
sept. 22 18:02:07 gnome-shell[64089]: Window manager warning: Buggy client sent a _NET_ACTIVE_WINDOW message with a timestamp of 0 for 0xa00013
sept. 22 18:02:08 gnome-shell[64089]: Received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadWindow (invalid Window parameter)'.
(Details: serial 9958 error_code 3 request_code 2 (core protocol) minor_code 0)
It does work when using a native X session instead of Wayland + XWayland, so if your desktop of choice provide this option (e.g for Gnome, you can choose a Gnome on X11 session), that might be a workaround. Of course you'll loose the Wayland advantages, and more and more desktop environment have switched to Wayland by default.
Strange. I get the exact opposite. In Linux MINT, this only happens if I'm on X11. When I change to Wayland (experimental), the problem goes away. It's clearly a JUCE problem (at least in my mind)
Well, actually it doesn't seem to be that consistent, since now I have the crash on both… So my workaround isn't working after all. I think it strange it did work at some point now :sweat_smile:
I managed to narrow it out a bit. The crash only happens under Gnome Shell if:
<VALUE name="systrayKey" val="1"/>
(For the casual onlooker, setting the above property with val="0"
in .config/Kushview/Element/Element.conf
allows to change this without having to disable the Systray extension first).
It doesn't matter, in my case, if Xorg or Wayland is used in the end. I was under the false impression it worked under Xorg because at that point, Gnome started in failsafe mode with all extension disabled, and the above extensions aren't really enabled until you restart the session, so the bug didn't trigger at first.
Looking at the log, Window manager warning: Buggy client sent a _NET_ACTIVE_WINDOW message with a timestamp of 0 for 0xa00013
shows up even when there's no crash, so I guess this one could be ignored (it does look like a bug in Juce as well, though).
I'm guessing this is the same as https://github.com/juce-framework/JUCE/issues/1421
Thanks for testing that. The saga continues..... I updated MINT and now it works perfectly fine in X11 or Wayland:
System:
Kernel: 5.15.0-121-generic x86_64 bits: 64 compiler: gcc v: 11.4.0 Desktop: Cinnamon 6.0.4
tk: GTK 3.24.33 wm: muffin vt: 7 dm: LightDM 1.30.0 Distro: Linux Mint 21.3 Virginia
base: Ubuntu 22.04 jammy
....
Interesting. Does latest Mint implement a systray in their mutter fork?
If so, it might be a bug in mutter (though still caused by a misbehaving Juice).
Guess the upgrade had nothing to do with it. Started happening again for me. Now it seems if I turn off the "System Tray" applet (right click an empty spot in the desktop panel -> "Applets" -> "System Tray")
Hopefully this issue is fixed. Recent changes in develop seem to have fixed it for me...
Running element in Linux Mint causes the whole desktop to black out. This is probably a JUCE issue...
Same thing has been reported in the juce forum. See: https://forum.juce.com/t/display-issues-on-linux-with-all-juce-applications/60676