Chrome quits when any key is pressed (Bionicpup64) solved

[mikewalsh]

Same issue here on Bionic 64 after updating Chrome to google-chrome-stable_current_amd.deb
Chrome halts when any key is pressed.
Any help will be much appreciated.

@CelsoIII :-

You ARE aware, I take it, that installing Chrome directly from the .deb package buggers-up read-write permissions on several of the aufs 'layers' in Puppy's file-system?

Google-chrome-stable, as packed, is built for the standard 'multi-user' set-up as found in most mainstream distros. Puppy, being a single-user hobbyist distro, and not using the sudo mechanisms - or anything related - doesn't function the same.

Mike.

[snoring_cat]

Hi Amaponian,

I downloaded bionicpup64-8.0-uefi.iso, ran it and installed the newest 64 bit deb from https://www.google.com/chrome/. There were no problems running it as user root or spot. Also, the DEB was converted to SFS and did not encounter any problems running Chrome that way either. Can you please run the following tests.

User spot
Test 1a: Check to see if some libs are causing a conflict by seeing if Chrome works via typing the following commands in a Terminal window:

Code: Select all

export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
run-as-spot /opt/google/chrome/chrome

Test 1b See if there is a problem with a Chrome profile for use spot

Code: Select all

export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
cd /home/spot/.config
mv google-chrome google-chrome_backup
run-as-spot /opt/google/chrome/chrome

User root
Test 2a: Check to see if some libs are causing a conflict by seeing if Chrome works via typing the following commands in a Terminal window:

Code: Select all

export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
/opt/google/chrome/chrome --no-sandbox

Test 2b: See if there is a problem with a Chrome profile for use root

Code: Select all

export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
cd /root/.config
mv google-chrome google-chrome_backup
/opt/google/chrome/chrome --no-sandbox

[Amaponian]

Thanks for your interest.

I downloaded bionicpup64-8.0-uefi.iso, ran it and installed the newest 64 bit deb from https://www.google.com/chrome/. There were no problems running it as user root or spot. Also, the DEB was converted to SFS and did not encounter any problems running Chrome that way either.

I did the same, as some others above, and we did have problems. This most likely means a hardware "condition" that affects only a few and prevents them from upgrading to a recent version of chrome. I downloaded my last working chrome (Versión 91.0.4472.164) in July 17, 2021.

Can you please run the following tests.

Yes, of course.
Below is the result of every test.
In all cases chrome halts when any key is pressed.

User spot
Test 1a: Check to see if some libs are causing a conflict by seeing if Chrome works via typing the following commands in a Terminal window:

Code: Select all

export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH.

run-as-spot /opt/google/chrome/chrome

Test 1a, terminal output

Code: Select all

root# export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
root# run-as-spot /opt/google/chrome/chrome
[31613:31613:1222/202325.175225:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.
[31537:31858:1222/202330.782820:ERROR:object_proxy.cc(642)] 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
[31537:31858:1222/202330.783531:ERROR:object_proxy.cc(642)] 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
[31537:31858:1222/202330.783776:ERROR:object_proxy.cc(642)] 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
[31537:31537:1222/202330.815729:ERROR:gpu_process_host.cc(979)] GPU process exited unexpectedly: exit_code=11
[31884:31884:1222/202330.977296:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.
[31780:1:1222/202331.216774:ERROR:command_buffer_proxy_impl.cc(125)] ContextResult::kTransientFailure: Failed to send GpuControl.CreateCommandBuffer.
[31537:31537:1222/202331.276913:ERROR:gpu_process_host.cc(979)] GPU process exited unexpectedly: exit_code=11
[31922:31922:1222/202331.390238:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.
[31537:31537:1222/202331.670932:ERROR:gpu_process_host.cc(979)] GPU process exited unexpectedly: exit_code=11
[31932:31932:1222/202331.839957:ERROR:gpu_init.cc(453)] Passthrough is not supported, GL is swiftshader, ANGLE is 
[31780:7:1222/202331.934450:ERROR:command_buffer_proxy_impl.cc(125)] ContextResult::kTransientFailure: Failed to send GpuControl.CreateCommandBuffer.
[31865:1:1222/202332.325865:ERROR:command_buffer_proxy_impl.cc(125)] ContextResult::kTransientFailure: Failed to send GpuControl.CreateCommandBuffer.
[31537:31746:1222/202336.238922:ERROR:chrome_browser_main_extra_parts_metrics.cc(230)] crbug.com/1216328: Checking Bluetooth availability started. Please report if there is no report that this ends.
[31537:31746:1222/202336.238963:ERROR:chrome_browser_main_extra_parts_metrics.cc(233)] crbug.com/1216328: Checking Bluetooth availability ended.
[31537:31746:1222/202336.238977:ERROR:chrome_browser_main_extra_parts_metrics.cc(236)] crbug.com/1216328: Checking default browser status started. Please report if there is no report that this ends.
[31537:31746:1222/202336.263779:ERROR:chrome_browser_main_extra_parts_metrics.cc(240)] crbug.com/1216328: Checking default browser status ended.
Segmentation fault
root# 

Test 1b See if there is a problem with a Chrome profile for use spot

Code: Select all

export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
cd /home/spot/.config
mv google-chrome google-chrome_backup
run-as-spot /opt/google/chrome/chrome

Test 1b, terminal output

Code: Select all

root# export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
root# cd /home/spot/.config
root# mv google-chrome google-chrome_backup
root# run-as-spot /opt/google/chrome/chrome
[3373:3373:1222/202637.059462:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.
[3373:3373:1222/202637.173832:ERROR:gl_surface_egl.cc(780)] EGL Driver message (Error) eglCreateWindowSurface: Failed to get the XVisualInfo for the child window.
[3373:3373:1222/202637.174559:ERROR:gl_surface_egl.cc(1510)] eglCreateWindowSurface failed with error EGL_BAD_NATIVE_WINDOW
[3331:3451:1222/202637.270043:ERROR:object_proxy.cc(642)] 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
[3331:3451:1222/202637.270486:ERROR:object_proxy.cc(642)] 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
[3331:3451:1222/202637.270965:ERROR:object_proxy.cc(642)] 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
LaunchProcess: failed to execvp:
xdg-settings
[1222/202653.765484:WARNING:minidump_to_upload_parameters.cc(32)] duplicate key guid, discarding value f23fc6ae718d4292828408763f734185
Segmentation fault
root# 

User root
Test 2a: Check to see if some libs are causing a conflict by seeing if Chrome works via typing the following commands in a Terminal window:

Code: Select all

export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
/opt/google/chrome/chrome --no-sandbox

Test 2a, terminal output

Code: Select all

root# export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
root# /opt/google/chrome/chrome --no-sandbox
[9742:9764:1222/203146.492364:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[9742:9764:1222/203146.493087:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[9777:9777:1222/203146.638210:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.
[9742:9764:1222/203146.654169:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[9742:9764:1222/203146.654224:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[9742:9864:1222/203146.921618:ERROR:object_proxy.cc(642)] 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
[9742:9864:1222/203146.922097:ERROR:object_proxy.cc(642)] 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
[9742:9864:1222/203146.922414:ERROR:object_proxy.cc(642)] 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
[9839:9839:1222/203147.066974:ERROR:command_buffer_proxy_impl.cc(125)] ContextResult::kTransientFailure: Failed to send GpuControl.CreateCommandBuffer.
[9742:9742:1222/203147.088458:ERROR:gpu_process_host.cc(979)] GPU process exited unexpectedly: exit_code=11
[9868:9868:1222/203147.169550:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.
[9742:9742:1222/203147.583161:ERROR:gpu_process_host.cc(979)] GPU process exited unexpectedly: exit_code=11
[9901:9901:1222/203147.681755:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.
[9742:9742:1222/203147.715944:ERROR:gpu_process_host.cc(979)] GPU process exited unexpectedly: exit_code=11
[9742:9742:1222/203147.719507:ERROR:command_buffer_proxy_impl.cc(328)] GPU state invalid after WaitForGetOffsetInRange.
[9915:9915:1222/203147.800360:ERROR:gpu_init.cc(453)] Passthrough is not supported, GL is swiftshader, ANGLE is 
[9839:9856:1222/203147.843366:ERROR:command_buffer_proxy_impl.cc(125)] ContextResult::kTransientFailure: Failed to send GpuControl.CreateCommandBuffer.
[9742:9764:1222/203148.359095:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[9742:9764:1222/203148.359702:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[9742:9836:1222/203151.121892:ERROR:chrome_browser_main_extra_parts_metrics.cc(230)] crbug.com/1216328: Checking Bluetooth availability started. Please report if there is no report that this ends.
[9742:9836:1222/203151.121935:ERROR:chrome_browser_main_extra_parts_metrics.cc(233)] crbug.com/1216328: Checking Bluetooth availability ended.
[9742:9836:1222/203151.121950:ERROR:chrome_browser_main_extra_parts_metrics.cc(236)] crbug.com/1216328: Checking default browser status started. Please report if there is no report that this ends.
[9742:9836:1222/203151.133364:ERROR:chrome_browser_main_extra_parts_metrics.cc(240)] crbug.com/1216328: Checking default browser status ended.
Segmentation fault
root# [1222/203155.590043:ERROR:nacl_helper_linux.cc(307)] NaCl helper process running without a sandbox!
Most likely you need to configure your SUID sandbox correctly
root#

Test 2b: See if there is a problem with a Chrome profile for use root

Code: Select all

export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
cd /root/.config
mv google-chrome google-chrome_backup
/opt/google/chrome/chrome --no-sandbox

Test 2b, terminal output

Code: Select all

root# export LD_LIBRARY_PATH=/opt/google/chrome:/opt/google/chrome/swiftshader:$LD_LIBRARY_PATH
root# cd /root/.config
root# mv google-chrome google-chrome_backup
root# /opt/google/chrome/chrome --no-sandbox
[12243:12264:1222/203333.362693:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[12243:12264:1222/203333.362831:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[12243:12264:1222/203333.475927:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[12243:12264:1222/203333.475999:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[12275:12275:1222/203333.540029:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.
[12275:12275:1222/203333.629971:ERROR:gl_surface_egl.cc(780)] EGL Driver message (Error) eglCreateWindowSurface: Failed to get the XVisualInfo for the child window.
[12275:12275:1222/203333.630403:ERROR:gl_surface_egl.cc(1510)] eglCreateWindowSurface failed with error EGL_BAD_NATIVE_WINDOW
[12243:12336:1222/203333.650862:ERROR:object_proxy.cc(642)] 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
[12243:12336:1222/203333.651156:ERROR:object_proxy.cc(642)] 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
[12243:12336:1222/203333.651371:ERROR:object_proxy.cc(642)] 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
LaunchProcess: failed to execvp:
xdg-settings
[12243:12264:1222/203339.469355:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[12243:12264:1222/203339.470054:ERROR:bus.cc(392)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[12243:12392:1222/203343.383555:ERROR:chrome_browser_main_extra_parts_metrics.cc(230)] crbug.com/1216328: Checking Bluetooth availability started. Please report if there is no report that this ends.
[12243:12392:1222/203343.384617:ERROR:chrome_browser_main_extra_parts_metrics.cc(233)] crbug.com/1216328: Checking Bluetooth availability ended.
[12243:12392:1222/203343.384924:ERROR:chrome_browser_main_extra_parts_metrics.cc(236)] crbug.com/1216328: Checking default browser status started. Please report if there is no report that this ends.
[12243:12392:1222/203343.445083:ERROR:chrome_browser_main_extra_parts_metrics.cc(240)] crbug.com/1216328: Checking default browser status ended.
[1222/203345.812132:WARNING:minidump_to_upload_parameters.cc(32)] duplicate key guid, discarding value b7b74a968b964bed91233b8acb2fed83
Segmentation fault
root# 

[snoring_cat]

Ahh,

I think I have seen this before. Amaponian, the problem I think has to do with missing packages, specifically libgtk. Can you please update your Puppy Package Manager,and then install (with dependencies) anything that starts off with libgtk-3.

if you still have problems, other possible packages would be

• libatk-bridge2

• libatspi2

• libavahi

• libcups

• libgbm

• libwayland

[snoring_cat]

Same issue here on Bionic 64 after updating Chorme to google-chrome-stable_current_amd.deb

Celsolll, BologneChe, and Amaponian, I tested on a fresh Bionic 64 8.0 booted from ISO. The newest Google Chrome 64bit deb was downloaded and installed from Google. Chrome works for me fine. This deb file was obtained by using the menu option "Setup" > "Puppy Quickpet" which directs you to download Google Chrome from Google.

Can anyone still having problems please post to this topic with the following

• Hardware model (e.g. Dell Latitude 1234)

• Video card type

• What version of Puppy Linux being run (e.g. Bionic 64 8.0

[Amaponian]

Ahh,

I think I have seen this before. Amaponian, the problem I think has to do with missing packages, specifically libgtk. Can you please update your Puppy Package Manager,and then install (with dependencies) anything that starts off with libgtk-3.

if you still have problems, other possible packages would be

• libatk-bridge2

• libatspi2

• libavahi

• libcups

• libgbm

• libwayland

Updated Package Manager and installed all recommended packages (with dependencies) that were missing. Chrome still refuses to move forward.

[snoring_cat]

Let's see if you are having problems with non-Google Chrome browsers as well. I am uploading some SFS files for the newest 64bit Vivaldi, Firefox, Chromium and Chrome. If possible, please try these out as well

Here is the link

[mikewalsh]

@snoring_cat :-

Ahh,

I think I have seen this before. Amaponian, the problem I think has to do with missing packages, specifically libgtk. Can you please update your Puppy Package Manager, and then install (with dependencies) anything that starts off with libgtk-3.

if you still have problems, other possible packages would be

• libatk-bridge2

• libatspi2

• libavahi

• libcups

• libgbm

• libwayland

Actually, insofar as my own portable packages go, that makes some kinda sense.

With assistance from baraka, I eventually un-earthed the cause of the mysterious "ailment" that had plagued my own builds for several releases. When I first started building Chrome packages for the community some years ago, everything was plain sailing. Then, in the middle of the 60-series, Google rebuilt the thing with a new build environment and different runtime requirements.....essentially building expressly for the multi-user approach employed by mainstream distros. The 'run-as-spot' approach then had to be employed, and a couple of releases later, the thing wanted newer versions of many things that prevented it from running in older Puppies. So battleshooter - one of the very few lasses apart from Jemimah that ever took more than a passing interest in Puppy - helped me figure out the included 'lib' directory that has formed an integral part of the Chrome builds ever since.....including a much newer version of libgtk-3.0.

This was fine until recently, when Big Brother decided to move the goalposts once again, this time including requirements for a version of libgtk-3.0 that seemed to have been compiled in a different way to what had previously been needed. Whatever the reason, it was eventually determined that older Puppies were perfectly content to run Chrome/Chromium-based browsers if permitted to do so with their own native libs, rather than using "LD_LIBRARY_PATH" to point them at the included 'lib' directory!

With the sole exception of Tahrpup64, which still wants SOME of the included deps, most Pups now run Chrome more or less without assistance. Of course, those dependencies need to be there to start with.....which I still find hard to comprehend, since Bionicpup64 comes with libgtk-3/libgdk-3.so.0.2200.30 OOTB, which runs Chrome quite happily in my own install without any issues.

I can't help thinking we're still "barking up the wrong tree" here.....

More research is required. Thanks for "holding the fort", BTW; this really ought to be my "shout", I know, but all kinds of crap has been happening here at home this last couple of weeks, and I just haven't been able to keep an eye on things as closely as I would have liked.

Mike.

[Amaponian]

Please, let's not miss the point here.

The problem:

Some users of Puppy are having an issue with the last version of Chrome, it halts when any key is pressed, nobody seems to know why. The affected, a minority of a unknown number, are not getting much help.

I downloaded a new iso, Bionicpup64, installed a fresh puppy and a fresh copy of chrome. The result is that chrome halts when any key is pressed, no matter how I change the settings.

I understand that many puppy users install chrome daily without any problem, so I guess it's a hardware problem. I have two different machines, chrome fails in them both. Either I have bad luck or this problem has something to do with my region (Venezuela).

I still have to do some tests
and provide here information of my second machine.

[mikewalsh]

@Amaponian :-

Please don't think we're not trying to help you find a solution. But with you only able to get on once a week, it does kinda mean that the 'momentum' falters a bit with long-ish breaks between replies. I'm not saying that's your fault; it's not. Many here are limited to the times they can access the forum, and it's something we just have to work round as best we can.

As for your issue; TBH, you ARE the first person I've come across who's got this problem, and TBH, I really don't know what else we can try. We've suggested everything we can think of, and whatever we suggest, you still have the issue.

The only thing I can think of is that someone in your geographic region is running some kind of 'man-in-the-middle' scam with the local Google servers, or at the the very least, the routing to & from them. I cannot think of any other explanation.

I don't know of anybody that is able to think up solutions to things that haven't yet occurred to them.

(*shrug*)

Mike.

[Feek]

Maybe it's nonsense, but:
what if there is a problem with the keyboard?

Be in your place, I would try to play with the keyboard settings, or try to connect another external keyboard via the usb port (if you have one).

And also try how the built-in virtual keyboard will behave (xvkbd, if I'm not mistaken).

There's nothing to lose, you will see.

[mikewalsh]

Maybe it's nonsense, but:
what if there is a problem with the keyboard?

Be in your place, I would try to play with the keyboard settings, or try to connect another external keyboard via the usb port (if you have one).

And also try how the built-in virtual keyboard will behave (xvkbd, if I'm not mistaken).

There's nothing to lose, you will see.

Mm. Yeah; we're getting "tunnel-vision" where the software's concerned. None of us has so far stopped to consider it could actually BE the keyboard itself!

Good catch, Feek. We can't rule anything out, yet....

Mike.

[thinkpadfreak]

Some users of Puppy are having an issue with the last version of Chrome, it halts when any key is pressed, nobody seems to know why. The affected, a minority of a unknown number, are not getting much help.

I seem to be one of the affected.
The Chrome I use is an sfs package I made from the official .deb package. Nothing is added to the original.
The issue began at version 94 of Chrome. As a test, I booted Bionicpup64 with RAM only mode. Chrome worked without a problem.
So I suspect that some change of the settings I made or some program / library I added causes the issue. Unfortunately, I don't know the cause. If I check every program, library, and change of the settings I added after installation of Bionicpup64, I might find the cause. But actually I am reluctant to do so. I would rather use Firefox or another Puppy.

[Amaponian]

Maybe it's nonsense, but:
what if there is a problem with the keyboard?

Be in your place, I would try to play with the keyboard settings, or try to connect another external keyboard via the usb port (if you have one).

And also try how the built-in virtual keyboard will behave (xvkbd, if I'm not mistaken).

There's nothing to lose, you will see.

Mm. Yeah; we're getting "tunnel-vision" where the software's concerned. None of us has so far stopped to consider it could actually BE the keyboard itself!

Good catch, Feek. We can't rule anything out, yet....

Mike.

That was a very basic guess that I've already tested. Two physical keyboards and some different configurations. The virtual keyboard behave the same. The last test I did was installing puppy without keyboard configuration (just the updates). The problem with chrome remained the same. I keep doing tests, feeling that I'm alone in this (but I know I'm not).

@mikewalsh:

Is there a way to move this topic to a place different from Bugs Report? (may be to Users). I feel somehow bad the discussion of this topic has got this long in this chanel.

[Amaponian]

Some users of Puppy are having an issue with the last version of Chrome, it halts when any key is pressed, nobody seems to know why. The affected, a minority of a unknown number, are not getting much help.

I seem to be one of the affected.
The Chrome I use is an sfs package I made from the official .deb package. Nothing is added to the original.
The issue began at version 94 of Chrome. As a test, I booted Bionicpup64 with RAM only mode. Chrome worked without a problem.
So I suspect that some change of the settings I made or some program / library I added causes the issue. Unfortunately, I don't know the cause. If I check every program, library, and change of the settings I added after installation of Bionicpup64, I might find the cause. But actually I am reluctant to do so. I would rather use Firefox or another Puppy.

That's very very interesting, my friend.
Thanks for the information.
Now I'm sure we're gonna solve this problem,
so keep in touch.

[mikewalsh]

@Amaponian :-

Is there a way to move this topic to a place different from Bugs Report? (may be to Users). I feel somehow bad the discussion of this topic has got this long in this chanel.

Nah, it's fine right here; that's what this section is for. And if this isn't a bug, I don't know what else you'd call it....

Don't give it a second thought.

Mike.

[Amaponian]

Some users of Puppy are having an issue with the last version of Chrome, it halts when any key is pressed, nobody seems to know why. The affected, a minority of a unknown number, are not getting much help.

I seem to be one of the affected.
The Chrome I use is an sfs package I made from the official .deb package. Nothing is added to the original.
The issue began at version 94 of Chrome. As a test, I booted Bionicpup64 with RAM only mode. Chrome worked without a problem.
So I suspect that some change of the settings I made or some program / library I added causes the issue. Unfortunately, I don't know the cause. If I check every program, library, and change of the settings I added after installation of Bionicpup64, I might find the cause. But actually I am reluctant to do so. I would rather use Firefox or another Puppy.

Ok.
I made the same test,
I booted in Puppy RAM mode
and got the same result: Chrome works just fine.
This put some light to this problem.

This may still require a lot of testing,
still may be a hardware condition of a few,
but now it's clear that it has a sotfware workaround.

[HerrBert]

FYI:
Brave Version 1.34.81 Chromium: 97.0.4692.99 (Offizieller Build) (64-Bit)
and
Slimjet Version 33.0.0.0 (basiert auf Chromium 94.0.4606.81) (Offizieller Build) (64-Bit)
also segfault when a key is pressed.

Wanted to test some other stuff in Bionicpup64 and stumbled upon these crashes and this topic.
May be of interest: Min v1.22.2 Chromium v94.0.4606.81 works (posting from it...)

This Bionicpup64 8.0 runs in PUPMODE=5 with customized adrv. Quickpet fixes are installed.

[HerrBert]

At least there are now three topics about this issue in different places.

@bigpup links from viewtopic.php?p=48334#p48334 to this topic suggesting there may be a fix for the problem.

I don't see a fix here. Am i missing something?

I also did the test suggested by @thinkpadfreak and @Amaponian .
Not surprising that Slimjet and Brave Browser do work here in RAM mode too.

At shutdown i created a savefolder with just the settings made in quicksetup.
Rebooting with these minimal changes, again Slimjet segfaults.

Code: Select all

Jan 30 10:34:05 puppypc6775 kern.info kernel: slimjet[6408]: segfault at 8 ip 00007efe95f6ed60 sp 00007ffe4937aee8 error 4 in libX11.so.6.3.0[7efe95f42000+133000]

(did not try Brave Browser after reboot - just guessing it will also fail...)

If running Bionicpup64 8.0 in pristine live mode is supposed to be the fix, this Puppy will become unusable for many users.

[mikewalsh]

There may perhaps be a fix, IF you can get to the Settings panel.

Google caused quite an upset last year when they announced the introduction of the FLoC trials. (FLoC = "Federated Learning of Cohorts"). This is all part of Google's big push toward greater internet privacy, and the elimination of 3rd-party cookies.

Basically, instead of ads being individually targeted at users, FLoC notices when a group of people all have the same interests, and sends suitable adverts to that group as a whole. Google received a lot of flak over this, and have since replaced it with "Privacy Sandbox" trials instead. The annoying thing about this is that it's enabled by default, and you have to deliberately opt out (as opposed to opting-in).

I disabled this in Chrome months ago. When Slimjet 33 appeared the other week, it also seemed very unstable. I've just now disabled Privacy Sandbox trials here, too; touch wood, it seems more settled. It's early days yet, though.

---------------------------------------------

To disable it, go into Settings->Privacy & Security. 'Privacy Trials' is down the bottom. If it's on - which it probably IS - follow the link, and disable it. Then shut-down, and re-start the browser.

As I say, this MAY be barking up the wrong tree, but.....it's worth a try. Apparently, early instances of this caused widespread crashing of Chrome in the middle of last year when it was first introduced, so it's possible there's still bugs in the system; Slimjet, in particular, is always built around versions of Chromium that are a few releases behind.

(@all :- Guys, is this ONLY occurring for you in Bionicpup64, or is anybody having issues in other Puppies, too?)

Mike.

[xenial]

Just to add to the discussion with a small point,In regard to chromium browsers i also enter the flags and disable some of the calling out in there too and if i recall there is a flag pertaining to privacy sandbox plus a webID flag which i disabled.

[HerrBert]

Although I actually use Slacko64 7.0 as my OS of choice, i've been banging my head this afternoon with this topic

Not a fix but a workaround:
Go to Menu -> Desktop -> compton, click Stop and Remove Autostart

I haven't got a clue how compton is related to graphic card drivers, but it seems to be an issue for some of us.

EDIT:
Remove Autostart in compton settings window doesn't stop compton from beeing started on reboot... but that may be considered another bug

EDIT 2:
Found that compton is started in /root/Autostart and /etc/xdg/autostart/lxqt-compton.desktop
Menu -> Desktop -> compton, click Remove Autostart only disables /root/Startup/compton
To prevent compton from starting at boot you also have to disable /etc/xdg/autostart/lxqt-compton.desktop by renaming it to something like /etc/xdg/autostart/lxqt-compton.desktop~ or delete it.
It may also be possible to exclude Chrom{e,ium} based browsers in /root/.config/compton.conf, but that would go beyond the scope.

I will now boot to my Slacko64 7.0 and wish you all good luck.
Maybe someone jumps in here and delivers a real fix for this...

[thinkpadfreak]

I haven't got a clue how compton is related to graphic card drivers, but it seems to be an issue for some of us.

After turning off compton, the current version of Chrome came to work fine. I had given up updating Chrome 93.
Thanks a lot.

Because an older version of Chrome worked fine with compton on, I suspect that some change in Chrome has also something to do with this issue.

[mikewalsh]

@HerrBert / @thinkpadfreak :-

TBH, this makes sense, now I think about it.

Google (via the Chromium Project) have been tweaking the GPU settings in Chrome a LOT recently. It's quite possible one of these is conflicting with picom/compton.

This was one of the very first things I disabled in Bionicpup64 when I first set it up, many months ago; I've never liked 'compositors', since I always have issues with the damn things.....and 'fancy' window effects have never really interested me. Probably explains why I've never noticed any problems with recent Chrome releases.....

Ah, well. Ya live & learn.

------------------------
@Amaponian :-

Can you try the above 'fix', and let us know if it works for you, too? Thanks.

Mike.

[mikeslr]

Hi All,

Thanks, bigpup, for directing us to this thread. And thanks, HerrBert, for the work-around. Just to add to the data we have on this situation:
As the paraphrased title above indicates, Iron Web-browser was also affected. I'd add recently. I didn't happen when I restructured Bionicpup64 a couple week ago: primarily using Save2SFS to convert my old Save into an adrv.sfs. And I don't recall adding anything since which might have triggered it.

To answer Mike Walsh's question: I use his Iron-portable under Bionicpup64, Xenialpup64, Fossapup64 and Vpup64. Each Puppy has two entries on the boot-menu: one with pfix=ram, the other using either a SaveFile or SaveFolder. Only under Bionicpup64 with a SaveFile did Iron crash when I tried to use it.

[xenial]

Just to add my own experiences in regard to picom and compton.I have recently started using fossapup64 again as my last usage had high cpu usage issues.

As a spur of the moment decision i disabled compton and the picom and fossapup is running as sweet as a nut with no high cpu usage at all and floating between 1-2% even when browsing.

I don't need the bling and so far there has been no adverse effects.

[mikewalsh]

@all:-

I know my own install of Fossapup was as laggy as hell on the desktop when I first began to use it.....especially using drag'n'drop, or when dragging windows around the screen. I run an Nvidia card, and Phil advised me to install the Nvidia driver .pet he'd built for it; he confessed he'd had a hell of a time of it trying to get the compositor to "play nice" with the in-kernel 'nouveau' driver.....and more or less gave up on it after discovering the 'official' driver fixed things.

IIRC, he also said Intel cards/onboard GPUs didn't display the issue at all. Go figure...

Out of the 6 Pups in the regular kennels - not including 'test' Pups - 4 of them have the official driver installed. Of the remaining 2, Sailor's 'take' on the original Slacko 5.7.0, 5.7.1, runs happily on the nouveau driver.....as does jrb's 'lite' spin on Barry's original Quirky64 'April' 7.0.1 (this one is currently running with Fossapup's k5.4.53 kernel).

Mike.

[ndujoe2]

Bionicpup64

Solved:

I followed the instructions to click remove Autosave in Compton. Then instead of renaming the file in the etc/xdg/autostart directory, I deleted it.

Now I am typing from the Version 98.0.4758.80 (Official Build) (64-bit) of Google Chrome once again.

Now to understand the mystery why compton and the new version of Google Chrome do not get along with each other when they did previously.

Joe

[Vverve]

Bionicpup64

Solved:

Work here too,latest chrome...Should be mark as solved

[mikewalsh]

@Vverve :-

Bionicpup64

Solved:

Work here too,latest chrome...Should be mark as solved

.....except that the OP is one of those who may not put in an appearance for months at a time. I could mark it as solved, but I feel it's a bit much when the Puppian who posed the question hasn't yet reported back whether or not it actually works for them. (I suspect it will, but 'twould be nice to have confirmation.)

D'you see my point?

Mike.