1679170 - Firefox crashes on tiling windows managers

Status: RESOLVED DUPLICATE of bug 1673767

(Core :: Widget: Gtk, defect)

Description

[contact]

Attachment: Crash log

User Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/13.0 Safari/605.1.15

Steps to reproduce:

1. Run a tiling wm (dwm, i3, bspwm)
2. Launch Firefox
3. Crashes within a few seconds

I doubt this is an issue that happens for other people as it'd likely have many reports, it should also be noted going back to v80 it still has the same behaviour. And again this doesn't occur on floating wm's, such as XFCE.

This also isn't because of my profile, clearing it, making a new one, and running with safe mode doesn't fix anything, crashes no matter what.

Actual results:

Firefox crashes

Expected results:

Firefox shouldn't crash

Comment 1

[contact]

Attachment: When launching with --jsconsole, this is in the console, right before it crashes

Comment 2

[e.sovetkin]

I confirm the same problem with xmonad.

Here is an output to the command line

(firefox:4103132): Gdk-ERROR **: 16:27:24.655: The program 'firefox' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadWindow (invalid Window parameter)'.
  (Details: serial 680 error_code 3 request_code 20 (core protocol) minor_code 0)
  (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.)
ExceptionHandler::GenerateDump cloned child 4103490
ExceptionHandler::WaitForContinueSignal waiting for continue signal...
ExceptionHandler::SendContinueSignalToChild sent continue signal to child

Comment 3

[contact]

Randomly Firefox started to function again, I have absolutely no clue on what fixed it as of right now. Possibly a dependency issue? No clue.

Comment 4

[contact]

I've absolutely no clue why but it stopped working again...

Comment 5

[Martin Stránský [:stransky] (ni? me)]

Can you run firefox with GDK_SYNCHRONIZE=1 env variable set like:

GDK_SYNCHRONIZE=1 firefox

and post any crashes from about:crashes?
Thanks.

Comment 6

[contact]

(In reply to Martin Stránský [:stransky] from comment #5)

Can you run firefox with GDK_SYNCHRONIZE=1 env variable set like:

GDK_SYNCHRONIZE=1 firefox

and post any crashes from about:crashes?
Thanks.

(firefoxdeveloperedition:2508621): Gdk-ERROR **: 12:29:26.256: The program 'firefoxdeveloperedition' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadWindow (invalid Window parameter)'.
  (Details: serial 1219 error_code 3 request_code 20 (core protocol) minor_code 0)
  (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.)
ExceptionHandler::GenerateDump cloned child 2512250
ExceptionHandler::SendContinueSignalToChild sent continue signal to child
ExceptionHandler::WaitForContinueSignal waiting for continue signal...
Exiting due to channel error.
Exiting due to channel error.
Exiting due to channel error.
Trace/breakpoint trap (core dumped)

Comment 7

[Martin Stránský [:stransky] (ni? me)]

Thanks but this does not carry any useful info. Do you have anything at about:crashes page?

Comment 8

[contact]

(In reply to Martin Stránský [:stransky] from comment #7)

Thanks but this does not carry any useful info. Do you have anything at about:crashes page?

On an unrelated reason I switched my distro to Artix from Arch, and it fixed it, but all my files in my home directory is the same. So it's some system files that probably caused this issue (Im guessing? Maybe drivers who knows...)

Either way here's the crash log:
https://crash-stats.mozilla.org/report/index/cfaa2480-3396-4a01-bca4-759dd0201120#tab-details

I installed Artix on a new drive for now, so I can still go back debugging it if necessary.

Comment 9

[Martin Stránský [:stransky] (ni? me)]

So this is a crash at XGetWindowProperty(). I wonder what can cause that. Karl, have you seen similar crashes?

Comment 10

[Karl Tomlinson (:karlt)]

I can't find a tag for 84.0b0 but it looks like this is during a call to WorkspaceManagementDisabled().

Bug 1682402 would provide better messaging, but "error_code 3" is BadWindow.

@contact, can you find for us the output of xprop -root _NET_SUPPORTING_WM_CHECK and then use that for xprop -id <window id> _NET_WM_NAME, please?

Comment 11

[Julien Cristau [:jcristau]]

(In reply to Karl Tomlinson (:karlt) from comment #10)

@contact, can you find for us the output of xprop -root _NET_SUPPORTING_WM_CHECK and then use that for xprop -id <window id> _NET_WM_NAME, please?

Comment 12

[contact]

(In reply to Karl Tomlinson (:karlt) from comment #10)

I can't find a tag for 84.0b0 but it looks like this is during a call to WorkspaceManagementDisabled().

Bug 1682402 would provide better messaging, but "error_code 3" is BadWindow.

@contact, can you find for us the output of xprop -root _NET_SUPPORTING_WM_CHECK and then use that for xprop -id <window id> _NET_WM_NAME, please?

Doesn't seem like anything is found for some reason...

__NET_WM_NAME: no such atom on any window.

Comment 13

[Karl Tomlinson (:karlt)]

I haven't been able to reproduce with bspwm and XDG_CURRENT_DESKTOP unset, where I saw this:

% xprop -root _NET_SUPPORTING_WM_CHECK
_NET_SUPPORTING_WM_CHECK(WINDOW): window id # 0x200000
% xprop -id 0x200000 _NET_WM_NAME
_NET_WM_NAME(UTF8_STRING) = "bspwm"
% xprop -id 0x200000 _NET_SUPPORTING_WM_CHECK                      
_NET_SUPPORTING_WM_CHECK(WINDOW): window id # 0x200000

That last command should be unaffected by missing _NET_WM_NAME.
What does xprop -id <window id> _NET_SUPPORTING_WM_CHECK print on your system?

Comment 14

[Karl Tomlinson (:karlt)]

This will be the same flaw as bug 1673767.