X Window System error on run

[SekoiaTree]

When running a built deb image:

(WebKitWebProcess:103134): Gdk-ERROR **: 20:32:04.545: The program 'WebKitWebProcess' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadValue (integer parameter out of range for operation)'.
  (Details: serial 220 error_code 2 request_code 152 (GLX) minor_code 34)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the GDK_SYNCHRONIZE environment
   variable to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error() function.)

The window just stays gray. I'm fully updated and I've restarted my computer. I am on Pop!_OS 22.04 LTS.

[FabianLars]

A few questions from me i asked in thread with a very similar error: #5085 (comment)

Errors like this are basically impossible to debug (for us/me) tbh, but a few questions anyway:

WSL or an actual Linux installation?
What GPU are you using?
If AMD: Are you using the open-source drivers (that are part of the kernel) or amdgpu-pro?
Are all your system dependencies up-to-date?

Edit: Another thing i saw on the internet, can you try setting this env var: WEBKIT_FORCE_SANDBOX=0 ? Probably doesn't work but worth a try :/

And another thing you can try is WEBKIT_DISABLE_COMPOSITING_MODE=1 which i saw in #3407

Would you mind trying these things and answering my questions? Then we can try to figure out the problem, even though i'm not really optimistic tbh.

[FabianLars]

I just ran WEBKIT_DISABLE_COMPOSITING_MODE=1 in the commandline and ran the program again (from the same console), same with FORCE_SANDBOX).

If it's 2 separate commands you need to add an export like:

export WEBKIT_DISABLE_COMPOSITING_MODE=1

alternatively you can combine them into one command like

WEBKIT_DISABLE_COMPOSITING_MODE=1 ./your-app

[SekoiaTree]

Okay, so WEBKIT_DISABLE_COMPOSITING_MODE worked, but SANDBOX didn't

[rivy]

@FabianLars , thanks for taking the time to address this concern.

I've had the same problem on Win10 using ssh with a local X11 server to an Ubuntu host in multiple WebKit-based apps.
This fixes the failure but, as noted, causes other graphics degradation.
Is this fundamentally an issue with WebKit? And, if so, where should we post the bug report/issue?

[FabianLars]

Is this fundamentally an issue with WebKit?

I think so, yes. Of course not 100% sure but i can't think of anything that Tauri does that could cause these issues.

And, if so, where should we post the bug report/issue?

Their bug tracker is over here https://bugs.webkit.org/enter_bug.cgi

[mberry]

Was facing this issue trying to debug with vscode and setting WEBKIT_DISABLE_COMPOSITING_MODE=1 was the fix, haven't noticed any graphics issues since. This is in a container and using X11 Forwarding.

Running tauri normally inside the container via command line worked fine without disabling webkit compositing so I'm not sure what is happening there.

[Rohit19060]

Can anyone tell me how to do this in windows? WEBKIT_DISABLE_COMPOSITING_MODE=1

[FabianLars]

This is a Linux-only setting.

[Rohit19060]

What is the solution for windows for this issue, can we avoid the startup white screen in any way?

[FabianLars]

You can spawn the window hidden (visible: false) and then show it at a later point in javascript. Note that if you're using v2 beta you may have to wait until the next release for this to work again.

[0rvar]

Hit this on Ubuntu 22.04 (Wayland + XWayland). Adding
XInitThreads() at the start of main fixed it for us:

fn main() {
    #[cfg(target_os = "linux")]
    unsafe { x11::xlib::XInitThreads(); }

    // ...
}

Never hit this on Ubuntu 24, so not sure what's going on here. But calling XInitThreads should be safe as it's supposed to be idempotent anyway