search

microsoft / WSL

Issues found on WSL
https://docs.microsoft.com/windows/wsl
MIT License
17.44k stars 822 forks source link

Chrome fails to launch #648

Closed Manouchehri closed 8 years ago

Manouchehri commented 8 years ago

Chrome fails to launch.

10.0.14385

/usr/bin/google-chrome

https://gist.githubusercontent.com/Manouchehri/da19ce4b9874c0a893210fa509fc1a33/raw/5fea50f4e81233e1cbe65ccdd92a295352ae9197/strace-chrome-wsl-fail.txt

wget -q -O - https://dl-ssl.google.com/linux/linux_signing_key.pub | sudo apt-key add -
sudo apt-get update
sudo apt-get install google-chrome-unstable
lypanov commented 7 years ago

This is getting especially interesting given the release of chrome-headless. Thanks everyone for your work on this!

therealkenc commented 7 years ago

Dev channel 60.0.3080.5 Headless Chrome runs out of the box with the libudev-stub workaround. Below is a screencap of headfull chrome talking to headless chrome, both running simultaneously on WSL. The headless side is totally a "CLI development use case scenario". The other isn't. Or something.

Notable side remark from the Google developer in that link above, quote: "Since I'm on Mac, I created convenient aliases..." Something to think about.

wsl-headless-chrome

jgoux commented 7 years ago

@therealkenc Thanks for your awesome work! I followed your instructions and compiled pulseaudio by myself with the fix, but I still get the same error as @iz0eyj : Assertion 'pthread_mutex_unlock(&m->mutex) == 0' failed at pulsecore/mutex-posix.c:108, function pa_mutex_unlock(). when trying to launch chrome or vscode.

therealkenc commented 7 years ago

@jgoux - If you are getting that assertion at line 108 of mutex-posix.c, it probably means you don't have the #undef HAVE_PTHREAD_PRIO_INHERIT at around line 23 as described here (or it wouldn't trip).

Since PulseAudio is clearly blocking folks, I'll see if I can do a PPA like @aseering did for Trusty sometime this week. I have been loath to do one because, knowing my luck, #1006 will be squashed by Stephen about 24 hours after I put it up.

therealkenc commented 7 years ago

Okay I've done a new libpulse build for WSL 16.04 Xenial. I jumped through the Launchpad PPA hoops, but I gather it takes time for the build bots to do their thing. It might or might not show up here eventually. I also put up a .deb package in the libudev-stub github release page here.

So the low tech steps, if you are feeling lucky, might be:

sudo apt-get install ubuntu-desktop       # and in general get firefox running with VcXsrv
sudo dpkg -i google-chrome-unstable_current_amd64.deb
sudo apt -f install                       # probably
sudo dpkg -i libpulse0_8.0-0ubuntu3.2ppa1_amd64.deb
sudo dpkg -i libudev-stub-0.9.0-WSL.deb   # doing this **last** is seriously important
google-chrome-unstable --no-sandbox

Again a heads up that the libudev-stub symbolic link will get blown away by apt-get <almost anything>, so if you get an error unrelated to pthread_mutex_unlock(&m->mutex) right out of the gate it probably means the link has been clobbered. Just reinstall the libudev-stub .deb to fix it up.

Bonne chance.

therealkenc commented 7 years ago

The PulseAudio PPA appeared. So alternately for the PA bits:

sudo add-apt-repository ppa:therealkenc/wsl-pulseaudio
sudo apt-get update && sudo apt-get upgrade
sudo apt-get install -f   # might need to do this if you installed the .deb manually
sudo dpkg -i libudev-stub-0.9.0-WSL.deb  # because the symlink will be clobbered

Personally I would just use the .deb.

jgoux commented 7 years ago

And...it works! Thanks a ton for the libpulse deb!!!

I'm able to run Chrome now. 🏃

I tried both code and code-insiders but still no luck.

This is the logs I have : image

Did you succeed in launching vscode from WSL?

I'm using VcXsrv, which runs Chrome well.

EDIT : I tried resolving my errors by installing nvidia drivers in Ubuntu, now I'm down to these errors : image

The only differences from Chrome's log are the last two warnings about x11_utils.cc BadMatch.

therealkenc commented 7 years ago

Unfortunately vscode is a no-go at the moment. It is impractical to debug on WSL right now due to gdb limitations; and even on a good day without gdb limitations it would be a bear to debug down at this level because of all the libchromiumcontent threads and message passing between nodejs and the electron renderer process. The strace output doesn't show anything glaringly broken, so I can't even pinpoint places in the source to add debug output lines.

I have been waiting for #1006 to flip and then perhaps try to find someone motivated enough on the vscode team to take a look and maybe make some educated guesses. It fails early in vscode-loader, which itself is a complicated package. Either vscode-loader is failing itself, or it tries to load some node native library that in turn hits a WSL gap not disturbed by Chrome proper. I also have an eye on #1910.

Atom gets a little further (at least it lights up), but can't open files. I don't use Atom so I didn't track the fail very far.

jgoux commented 7 years ago

Well, I suppose I'll have to wait! :p

You already got us really far so thanks again for all the tools!

therealkenc commented 7 years ago

Per video drivers (saw your edit), what you want is mesa llvmpipe, which is also typical in a VM scenario.

sudo add-apt-repository ppa:paulo-miguel-dias/mesa
sudo apt-get update
sudo apt-get dist-upgrade                # yes dist-upgrade or it won't bump
sudo apt-get install mesa-utils
sudo dpkg -i libudev-stub-0.9.0-WSL.deb  # because the symlink will be clobbered
glxinfo -B | grep OpenGL

If all goes well you should see:

OpenGL vendor string: VMware, Inc.
OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 5.0, 256 bits)
OpenGL core profile version string: 3.3 (Core Profile) Mesa 17.2.0-devel - padoka PPA
...
bmayen commented 7 years ago

What would be the reason to run VSCode through WSL anyway? Chrome I understand for running tests against, but what's the benefit of running VSCode through it vs native?

therealkenc commented 7 years ago

NodeJS Linux Development. DotNet Linux development. C/C++ Linux development.

You can in principle use Windows VSCode though WSL interop, but then you'll run headlong into DrvFS limitations (777 perms, symlinks, NTFS unfriendly filenames, files pinned by open handles). Chrome on WSL isn't all that useful (possibly notwithstanding headless), because all interaction goes through HTTP, which means it can just as well be running on Windows. VSCode talks to the filesystem, and there is the rub.

aseering commented 7 years ago

@therealkenc , "gdb": For what it's worth, I've observed significant improvements in gdb's functionality in post-Creators Insider builds. I'd be curious if vscode is now more debuggable. (I've been traveling lately and haven't had a chance to update relevant tickets.)

therealkenc commented 7 years ago

@aseering - Unfortunately as of 16188 I still can't break with ctrl-c to see where vscode gets stuck.

ken@DESKTOP-9BDQA3V:~/Devel/vscode/vscode$ gdb ../electron/out/D/electron
GNU gdb (Ubuntu 7.11.1-0ubuntu1~16.04) 7.11.1
Copyright (C) 2016 Free Software Foundation, Inc.
[...]
Reading symbols from ../electron/out/D/electron...done.
(gdb) set args .
(gdb) run
Starting program: /home/ken/Devel/vscode/electron/out/D/electron .
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7fffd9820700 (LWP 320)]
^C^C^C^C^C^C^C^C^C^C^C^C^C^C
evil-shrike commented 7 years ago

I followed comment by @therealkenc, but getting some error in /opt/google/chrome-unstable/libosmesa.so (the same was when I tried stable Chrome):

root@host:/mnt/c/temp/lx# google-chrome-unstable --no-sandbox --headless --remote-debugging-port=9222 https://chromium.org
UDEV STUB: Using libudev-stub 0.9.0
UDEV STUB: udev_new()
UDEV STUB: stubbing udev...
UDEV STUB: udev_monitor_new_from_netlink() name: udev
UDEV STUB: udev_monitor_filter_add_match_subsystem_devtype() subsystem: sound devtype: (null)
UDEV STUB: udev_monitor_filter_add_match_subsystem_devtype() subsystem: video4linux devtype: (null)
UDEV STUB: udev_monitor_enable_receiving()
UDEV STUB: udev_monitor_get_fd()
Segmentation fault (core dumped)
root@host:/mnt/c/temp/lx# [0525/171206.593849:ERROR:gl_implementation.cc(246)] Failed to load /opt/google/chrome-unstable/libosmesa.so: /opt/google/chrome-unstable/libosmesa.so: cannot open shared object file: No such file or directory

How can it be fixed?

therealkenc commented 7 years ago

@evil-shrike -- Sound a lot like this. Try adding:

$ google-chrome-unstable --disable-gpu --no-sandbox --headless --remote-debugging-port=9222 https://chromium.org

You can also try installing ppa:paulo-miguel-dias/mesa mentioned above and see if that is a big hammer fix.

Are you running stock Ubuntu Xenial or some other distro? I ask because it should run out of the box like for jgoux so long as you have a slate that would also succeed on Real Linux. Anyway looks like you are close because you are getting the libudev-stub traces.

evil-shrike commented 7 years ago

@therealkenc thanks. adding --disable-gpu helped, the last error (libosmesa.so) has gone, other things stayed and Chrome just exists. Will try with mesa later. yes, I'm running stock Ubuntu for Creators Update (16)

UPDATE: installing mesa-utils helped remove other errors (I just apt-get i mesa-utils). Now Chrome output only one line: "Trace/breakpoint trap (core dumped)" and exits.

root@host:/mnt/c/temp/lx# google-chrome-unstable --disable-gpu --no-sandbox --headless --remote-debugging-port=9222 http://chromium.org
Trace/breakpoint trap (core dumped)
root@host:/mnt/c/temp/lx#
therealkenc commented 7 years ago

Chrome now works out of the box with sandbox on 16273 thanks to @benhillis work in #2242. There are a couple of caveats (there are always caveats). First, the pthread_mutex_unlock(&m->mutex) fail (which is #486) will still trigger if you have pulse audio installed. Which you do, if you installed the whole ubuntu-desktop meta package. This had me confused earlier as well (see the strikeout). There are two ways around. One is to push though it with the PulseAudio 8 PPA. The other is to purge pulse from your system. I'd recommend the latter.

The other caveat, which is of the soft variety, is that chrome would really rather be living in an environment with a functioning dbus and opengl (ie mesa llvmpipe). Putting it all together, something like the following should get you there:

sudo update && sudo apt upgrade
sudo apt install ubuntu-desktop wget mesa-utils
sudo apt purge pulseaudio libpulse0 && sudo apt autoremove
wget https://dl.google.com/linux/direct/google-chrome-unstable_current_amd64.deb
sudo dpkg -i google-chrome-unstable_current_amd64.deb
sudo apt -f install           # probably
export DISPLAY=localhost:0
export NO_AT_BRIDGE=1
exec dbus-run-session -- bash
sudo service dbus start
google-chrome-unstable        # look mom, no flags

And there you have it.

iz0eyj commented 7 years ago

GREAT!!! it works fine!!! A problem: xfce4 needs pulseaudio :( So, if I install Chrome... bye bye desktop integration with xfce :(

SOLVED! I don't know how, but now I have xfce4 without Pulse :) desktop

iz0eyj commented 7 years ago

Hey @therealkenc this is my preferred browser, Vivaldi, that runs out of the box without any hack. Many thanks to you and to @benhillis

desktop

benhillis commented 7 years ago

@therealkenc - Unfortunately I didn't have enough time to get the priority futex work done for this release. I'm hoping to knock this one out very early for the next release.

davidmaxwaterman commented 7 years ago

FYI, I've been using WSL since late last year. I would normally wipe off Windows and install Ubuntu, but WSL has been working for me since then. However, I am now trying to write some tests for my Google Polymer based web apps, and the tools use Web Component Tester, and so I hit this problem where it can't launch the browser to run my tests. It does seem like there is a solution trying to find it's way out, but the whole 'no GUI' stance makes me think it'll not be supported for some time, even if it is made to work soon (which seems likely). As such, I'm think I'm going to switch back to Ubuntu, which is a bit of a shame, tbh....WSL is kind of awesome really.

jehon commented 6 years ago

I have the impression that google chrome works if you:

But since that's the conclusion of a long work, with lots of trials, I would rather hope that someone else could confirm that...

bitcrazed commented 6 years ago

@davidmaxwaterman - I know that @TheLarkInn was looking into how one could launch a browser from within WSL.

therealkenc commented 6 years ago

how one could launch a browser from within WSL.

this. There is nothing special about launching a Windows browser (Chrome/Edge/Firefox) versus any other Windows application (Notepad). Native Linux Chrome has worked since 16273 and this issue is fixedfallcreatorsupdate. There are no known issues filed in github related to Chrome, with the nonblocking exceptions of #1006 and #1353.

davidmaxwaterman commented 6 years ago

Thanks but I wrote that 5 months ago and have long since given up with WSL and gone back to Ubuntu. I can quite believe things have changed since then.

rsshilli commented 6 years ago

This works great for me. I installed VcXsrv , started "XLaunch" in windows, and then started up my Windows Subsystem Ubuntu bash console and followed @therealkenc 's comment from Aug 24, 2017.

I'm using it to run automated tests in Chrome. It all works wonderfully.

Thank you!

yxy1996 commented 6 years ago

I came across with this situation When I launched roboware. I tried to use the latter method, but it will purge so many packages relative to ros-kinetic. I tried it eventually and my roboware worked magically, however, roscd couldn't work any more. @therealkenc

illyaMs commented 5 years ago

Combining the tips from @therealkenc (https://github.com/Microsoft/WSL/issues/648#issuecomment-324562271) and @jehon (https://github.com/Microsoft/WSL/issues/648#issuecomment-372794217) worked for me as well. Had issue running automated tests based on Chromium from win10/WSL.

Thanks!

caseyjkey commented 5 years ago

Chrome now works out of the box with sandbox on 16273 thanks to @benhillis work in #2242. There are a couple of caveats (there are always caveats). First, the pthread_mutex_unlock(&m->mutex) fail (which is #486) will still trigger if you have pulse audio installed. Which you do, if you installed the whole ubuntu-desktop meta package. This had me confused earlier as well (see the strikeout). There are two ways around. One is to push though it with the PulseAudio 8 PPA. The other is to purge pulse from your system. I'd recommend the latter.

The other caveat, which is of the soft variety, is that chrome would really rather be living in an environment with a functioning dbus and opengl (ie mesa llvmpipe). Putting it all together, something like the following should get you there:

sudo update && sudo apt upgrade
sudo apt install ubuntu-desktop wget mesa-utils
sudo apt purge pulseaudio libpulse0 && sudo apt autoremove
wget https://dl.google.com/linux/direct/google-chrome-unstable_current_amd64.deb
sudo dpkg -i google-chrome-unstable_current_amd64.deb
sudo apt -f install           # probably
export DISPLAY=localhost:0
export NO_AT_BRIDGE=1
exec dbus-run-session -- bash
sudo service dbus start
google-chrome-unstable        # look mom, no flags

And there you have it.

I want to use Capybara to run some automated tests with Chrome, how do I choose to run the google-chrome-dev instead of the default Chrome? Only Chrome 78 (unstable) will run, however, Selenium (for Capybara) requires webdriver. Webdriver only support version 77.

WSL-Chrome-78

darrenm95 commented 4 years ago

I still get an issue when trying to run google chrome on windows subsystem for linux. Using ubuntu 16.04. Has anyone got any ideas why I am getting the followings errors?

[28834:28834:0206/151626.871150:ERROR:browser_switcher_service.cc(238)] XXX Init() [28918:32:0206/151627.109653:ERROR:platform_shared_memory_region_posix.cc(52)] Descriptor access mode (2) differs from expected (0) [28943:50:0206/151627.224939:ERROR:platform_shared_memory_region_posix.cc(52)] Descriptor access mode (2) differs from expected (0) [28929:36:0206/151627.235707:ERROR:platform_shared_memory_region_posix.cc(52)] Descriptor access mode (0) differs from expected (2) [28961:65:0206/151627.236391:ERROR:platform_shared_memory_region_posix.cc(52)] Descriptor access mode (2) differs from expected (0) [28868:28868:0206/151627.432404:ERROR:gl_surface_glx.cc(459)] GLX 1.3 or later is required. [28868:28868:0206/151627.432430:ERROR:gl_initializer_x11.cc(148)] GLSurfaceGLX::InitializeOneOff failed. [28868:28868:0206/151627.438697:ERROR:viz_main_impl.cc(159)] Exiting GPU process due to errors during initialization [28834:28834:0206/151628.837524:ERROR:CONSOLE(1)] "Uncaught SyntaxError: Unexpected token '<'", source: chrome://welcome/welcome_app.js (1) [28894:13:0206/151628.841397:ERROR:platform_shared_memory_region_posix.cc(52)] Descriptor access mode (0) differs from expected (2) [28834:28853:0206/151629.981332:FATAL:gpu_data_manager_impl_private.cc(439)] GPU process isn't usable. Goodbye. Trace/breakpoint trap (core dumped)

ambrt commented 4 years ago

Have same thing . I came to flags below. It flashes the website correctly (in VcXsrv) then crashes. Its at WSL 1. sudo DISPLAY=:0 google-chrome --disable-features=VizDisplayCompositor --no-sandbox --disable-setuid-sandbox --disable-dev-shm-usage --unlimited-storage --ignore-certificate-errors --enable-font-antialiasing --font-render-hinting=medium http://example.com

The error stack is this

[30371:30371:0319/144325.480839:ERROR:edid_parser.cc(102)] Too short EDID data: manufacturer id
[30371:30398:0319/144325.630878:ERROR:bus.cc(393)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[30371:30398:0319/144325.650220:ERROR:bus.cc(393)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[30371:30398:0319/144325.650262:ERROR:bus.cc(393)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[30371:30451:0319/144325.809533:ERROR:object_proxy.cc(632)] Failed to call method: org.freedesktop.DBus.Properties.Get: object_path= /org/freedesktop/UPower: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.UPower was not provided by any .service files
[30371:30451:0319/144325.810376:ERROR:object_proxy.cc(632)] Failed to call method: org.freedesktop.UPower.GetDisplayDevice: object_path= /org/freedesktop/UPower: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.UPower was not provided by any .service files
[30371:30451:0319/144325.811266:ERROR:object_proxy.cc(632)] Failed to call method: org.freedesktop.UPower.EnumerateDevices: object_path= /org/freedesktop/UPower: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.UPower was not provided by any .service files
libGL error: No matching fbConfigs or visuals found
libGL error: failed to load driver: swrast
[30396:30396:0319/144325.943947:ERROR:viz_main_impl.cc(161)] Exiting GPU process due to errors during initialization
[30371:30383:0319/144326.480351:ERROR:browser_gpu_channel_host_factory.cc(138)] Failed to launch GPU process.
Xlib:  extension "MIT-SHM" missing on display ":0".

also frequent shared memory errors

[25681:25687:0319/172018.623518:ERROR:platform_shared_memory_region_posix.cc(52)] Descriptor access mode (0) differs from expected (2)

The network seems to work because if i will run with sudo DISPLAY=:1 xvfb-run (...) --dump-dom --headless http://example.com It dumps dom correctly then crashes with

[0319/144741.169122:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 1 time(s)
[0319/144741.313374:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 2 time(s)
[0319/144741.462192:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 3 time(s)
[0319/144741.611385:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 4 time(s)
[0319/144741.767465:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 5 time(s)
[0319/144741.926318:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 6 time(s)
[0319/144741.927170:ERROR:browser_gpu_channel_host_factory.cc(138)] Failed to launch GPU process.
G-Rath commented 4 years ago

@therealkenc I've followed your instructions in your original comment, but chrome seems to having some issues.

Here are some details:

Windows 10 Pro Version: 1909 OS Build: 18363.720

VcXsrv X Server: 1.20.6.0 (12 Jan 2020) Google Chrome 80.0.3987.149

I can run both google-chrome-stable & google-chrome-unstable if I use sudo & --no-sandbox, but --headless crashes:

g-rath@DESKTOP-MUGH739:/c/Users/G-Rath$ sudo google-chrome-stable --no-sandbox --headless https://google.com
[0320/095942.195820:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 1 time(s)
[0320/095942.337443:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 2 time(s)
[0320/095942.473931:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 3 time(s)
[0320/095942.610071:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 4 time(s)
[0320/095942.748664:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 5 time(s)
[0320/095942.882727:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 6 time(s)
[0320/095942.978725:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 7 time(s)
[0320/095943.069493:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 8 time(s)
[0320/095943.157845:WARNING:gpu_process_host.cc(1165)] The GPU process has crashed 9 time(s)
[0320/095943.157879:FATAL:gpu_data_manager_impl_private.cc(1034)] The display compositor is frequently crashing. Goodbye.
Failed to generate minidump.Illegal instruction (core dumped)

This is very annoying because headless is what I want, so I can run feature tests 😬

Heres the output for trying to run google-chrome-stable:

g-rath@DESKTOP-MUGH739:/c/Users/G-Rath$ /usr/bin/google-chrome-stable
Failed to move to new namespace: PID namespaces supported, Network namespace supported, but failed: errno = Permission denied
Trace/breakpoint trap (core dumped)
g-rath@DESKTOP-MUGH739:/c/Users/G-Rath$ stat /usr/bin/google-chrome-stable
  File: /usr/bin/google-chrome-stable -> /opt/google/chrome/google-chrome
  Size: 32              Blocks: 0          IO Block: 512    symbolic link
Device: 2h/2d   Inode: 844424932927321  Links: 1
Access: (0777/lrwxrwxrwx)  Uid: (    0/    root)   Gid: (    0/    root)
Access: 2020-03-19 17:46:48.000000000 +1300
Modify: 2020-03-17 10:50:01.000000000 +1300
Change: 2020-03-19 17:46:55.208091200 +1300
 Birth: -

If I then run it with --no-sandbox, Chrome opens and then crashes with a lot of permission errors.

For some reason it seems that Chrome is creating its cache (and I assume other files) in a way that means they end up being owned by root?

Let me know if you'd like me to create a new issue, or have any questions :)

Additionally, webpages don't actually load; I just get a white screen.

ambrt commented 4 years ago

The comment above about not working solution must be related to chrome version. The newest version 80.* is failing me also.

I used version 77 that i happened to have installed on wsl some time ago and it manages to open non https website, via VcXsrv. sudo DISPLAY=:0 /home/user/node_modules/puppeteer/.local-chromium/linux-674921/chrome-linux/chrome --disable-features=VizDisplayCompositor --no-sandbox --disable-setuid-sandbox --disable-dev-shm-usage --unlimited-storage --ignore-certificate-errors --disable-gpu --enable-font-antialiasing --font-render-hinting=medium --use-gl=wgl --disable-software-rasterizer http://example.com

It still opens https sites as a blank page.

MolloKhan commented 4 years ago

I downloaded the latest version today (81.0.4044) and I got this error as well. I don't know why this ticket is closed =/

savethebyte commented 4 years ago

I have used chrome 83 and 84, and I am still recieving the error in WSL.

[0709/221720.544454:ERROR:platform_shared_memory_region_posix.cc(46)] Descriptor access mode (0) differs from expected (2)
[0709/221720.548263:WARNING:crash_handler_host_linux.cc(366)] Could not translate tid - assuming crashing thread is thread group leader; syscall_supported=0
[0709/221720.675739:WARNING:crash_handler_host_linux.cc(366)] Could not translate tid - assuming crashing thread is thread group leader; syscall_supported=0
--2020-07-09 22:17:20--  https://clients2.google.com/cr/report
Resolving clients2.google.com (clients2.google.com)... 172.217.6.110, 2607:f8b0:4009:813::200e
Connecting to clients2.google.com (clients2.google.com)|172.217.6.110|:443... connected.
[1594351040.818][DEBUG]: DevTools WebSocket Event: Inspector.targetCrashed 5A70B9C03C3188FDCA221C0B94718B5F {

}
[1594351040.818][INFO]: Done waiting for pending navigations. Status: tab crashed
HTTP request sent, awaiting response... [1594351040.868][INFO]: [c6f38322f1211be529223d6f0a3aafb7] RESPONSE ClickElement ERROR unknown error: session deleted because of page crash
from tab crashed
  (Session info: headless chrome=83.0.4103.116)
[1594351040.868][DEBUG]: Log type 'driver' lost 0 entries on destruction
[1594351040.868][DEBUG]: Log type 'browser' lost 3 entries on destruction
200 OK
Length: unspecified [text/plain]
Saving to: ‘/dev/fd/4’

Crash dump id: aa6abc501aeb718e

     0K                                                         744K=0s
Ananthavijay commented 3 years ago

I get this error in wsl when I run google-chrome

[1023/204234.433299:ERROR:exception_handler_server.cc(361)] getsockopt: Invalid argument (22)
kohheepeace commented 2 years ago

I followed this docs https://docs.microsoft.com/en-us/windows/wsl/tutorials/gui-apps

And I got these errors

$ google-chrome
[2014:2014:1112/172745.466766:ERROR:browser_dm_token_storage_linux.cc(94)] Error: /etc/machine-id contains 0 characters (32 were expected).
[1112/172745.479719:ERROR:file_io_posix.cc(144)] open /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq: No such file or directory (2)
[1112/172745.479781:ERROR:file_io_posix.cc(144)] open /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq: No such file or directory (2)
Trace/breakpoint trap
ricklambrechts commented 2 years ago

For the people that are expierencing issues after using: https://docs.microsoft.com/en-us/windows/wsl/tutorials/gui-apps

Check the wsl version with:

wsl --list --verbose

If the wsl distro is version 1 then convert the distro to version 2 with: Check this documentation: https://docs.microsoft.com/en-us/windows/wsl/basic-commands#set-wsl-version-to-1-or-2

wsl --set-version <name> 2
mochammadrafi commented 2 years ago

Any solution?

simonLeary42 commented 1 year ago

A simple wsl --update fixed this issue for me. I'm not sure what version I had because wsl --version apparently wasn't a valid command back then. I'm now using 1.2.5.0

gelodefaultbrain commented 3 months ago

Hi, what did you guys do? I just did 

gelswsl@LAPTOP-MCLLQCCH:~$ google-chrome
Killed

what is happening?

gelodefaultbrain commented 3 months ago

Hi, what did you guys do? I just did

gelswsl@LAPTOP-MCLLQCCH:~$ google-chrome
Killed

what is happening?

Hi Had to update wsl as well as mentioned here. Well with that it caused my running application to restart as well 😢