yes
You now may have no kernel to boot with - the process did not finish.
Uninstalled an old kernel and ran mapare again.
Now there was just the warnings about possible missing firmware
pacdiff -o
only lists grub and pacman. Should not affect how the pc runs. (grub has resume from hibernate, pacman has reduced number of parallel downloads)
New test user - same problem
Since you ran out of space a minute ago - is there enough of it?
root has got 5% more than regular users on default ext4 file systems …
Is a problem in multiple ways, and will continue to be until you free up some space.
I dont know what cruft is all on your system … videos, browser cache, SNAPs?
This will get rid of all uninstalled package cache, and all but 2 of installed cache;
paccache -rvuk0
paccahce -rvk2
EDIT.
Oh, looking, its specifically your boot partition thats rather full?
In that case the above package cache wont have any impact.
You will want to look at what is using the space … lots of kernels?
Or ultimately you may even need to provide more for the partition, if there is not much to clean.
/boot has about 100mb free after removal of not used kernel. Shouldn’t not be a problem.
root has several GB free.
What kind of settings is used when xfwm4.xml does not exist? I’d like to compare this to the file in /etc/skel. I’d expect the config that is used as running config when the file doesn’t exist is the settings that is written to xfwm4.xml at login when the file doesn’t exist at login. But then - why does WM fail to start when this autogenerated file is present?
Could it be permissions issue? Trying
chmod -R go+rw /home/test8/.config/xfce
still the same issue
This won’t even work - the directory name is: /home/test8/.config/xfce4
You should have received an error message in response to your command.
also - verify whether what you did had the desired effect:
ls -al /home/test8/.config/xfce4
Is it the correct owner?
I think you are looking at the wrong thing when you persist focusing on that one file.
You can even launch a working session totally without /home/user/.config/xfce4
Nothing at all.
You’ll still get a working session, the default session, which you can configure yourself.
It looks ugly (a matter of taste) - but it works.
I will recommend installing and using mc one more time.
It will make everything more transparent than the sole use of command line is - especially when you are not in a graphical session but in a TTY or terminal.
Guess I was missing last character when copying the command from terminal to forum.
My focus is that WM fail to start if it has that config file present. That leads me thinking it is a config issue.
Any way to get more logging from WM startup?
I don’t know about it.
How about that?
… you are sure? - you didn’t share …
We are blind here - if you need help … share
If your shell startup files are ok - they come from /etc/skel when you create a new user -
I have no clue.
How do you create a new user - to test with whether it’ll work there?
One idea:
if you (temporarily) disable the display manager
systemctl disable lightdm.service
systemctl stop lightdm.service
you then land on a TTY - log in, and run
startxfce4
then you should get some text output - but I don’t know how to capture it
I guess you’ll have to be a quick reader 
… just kidding: simply log out of the session and you’ll be back at the TTY you started from and see the output
systemctl enable lightdm.service
to re-enable the service …
Owner is test8:test8
Permissions are rw?rw?rw? (? to show that some has x, some has not - directories)
I create new users using Manjaro Settings Manager.
WIll try to disable lightdm…
I see a lot of
warning: could not resolve keysym xxxx
xxx includes: XF86CameraAccessToggle, XF86NextElement, XF86PreviousElement, AutopilotEngageToggle, XF86MarkWaypoint, XF86Sos, XF86NavChart, XF86FishingChart… and many more.
And I see:
** (xiccd:28673): CRITICAL **: 12.23.23.718: failed to create colord device: failed to obtain org.freedesktop.color-manager.create-device auth
** (light-locker:20708): ERROR **: 12.23.24:360: Environment variable XDG_SESSION_PATH not set. Is LightDM running?
Turns out there is no longer a scroll function in tty. Will try again, try to catch output in a file...
Another possible solution;
→ only if your /home is on a separate partition:
- Boot from the latest available Manjaro installer, and reinstall using the manual partitioning method.
- Adjust the sizes of all other partitions as needed; do not format
/home. - Finish reinstalling, reboot, enjoy Manjaro.
Cheers.
There never was one - you can scroll command output, but not the TTY itself.
It’s not important anyway. There are only xorg messages.
The point of the exercise was to see:
does it work differently without lightdm, does it make a difference?
Your chown -R command is useless - the user has got the permissions anyway.
Adding group and others doesn’t help anything.
As suggested previously, perhaps a clue is here:
LANG=C ls -al /etc/skel
LANG=C ls -al /etc/skel/.config
it looks like this in my system:
(the templates for the shell startup files are here)
LANG=C ls -al /etc/skel
total 60
drwxr-xr-x 3 root root 4096 Feb 21 22:19 .
drwxr-xr-x 108 root root 12288 Apr 5 13:03 ..
-rwxr-xr-x 1 root root 100 Oct 30 2017 .Xclients
-rw-r--r-- 1 root root 21 Jan 23 22:22 .bash_logout
-rw-r--r-- 1 root root 57 Jan 23 22:22 .bash_profile
-rw-r--r-- 1 root root 3270 Jan 23 22:22 .bashrc
drwxr-xr-x 10 root root 4096 Feb 18 15:36 .config
-rw-r--r-- 1 root root 4855 Oct 30 2017 .dir_colors
-rw-r--r-- 1 root root 53 Jul 24 2023 .nanorc
-rw-r--r-- 1 root root 141 Feb 16 02:52 .profile
-rw-r--r-- 1 root root 1508 Feb 16 02:52 .xinitrc
-rw-r--r-- 1 root root 382 Jul 25 2023 .zshrc
and
LANG=C ls -al /etc/skel/.config
total 44
drwxr-xr-x 10 root root 4096 Feb 18 15:36 .
drwxr-xr-x 3 root root 4096 Feb 21 22:19 ..
drwxr-xr-x 2 root root 4096 Feb 18 15:36 Kvantum
drwxr-xr-x 2 root root 4096 Feb 18 15:36 Thunar
drwxr-xr-x 2 root root 4096 Mar 2 20:40 autostart
drwxr-xr-x 3 root root 4096 Aug 16 2022 falkon
drwxr-xr-x 2 root root 4096 Feb 18 15:36 gtk-2.0
-rw-r--r-- 1 root root 293 Feb 16 02:52 mimeapps.list
drwxr-xr-x 2 root root 4096 Feb 18 15:36 qt5ct
drwxr-xr-x 2 root root 4096 Feb 18 15:36 volumeicon
drwxr-xr-x 5 root root 4096 Feb 18 15:36 xfce4
I ran
$ startxfce4 1>xfce-log.log 2>xfce-err.log
I did it twice - with and without the xfwm4.xml.
The difference in xfce-err.log is that with the xfwm4.xml present, the following lines are included:
(xfwm4:29776): Gdk-WARNING **: 12:50:05.946: The program 'xfwm4' 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 2700 error_code 2 request_code 151 (GLX) minor_code 24)
(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.)
xfsettingsd: No window manager registered on screen 0.
xfce4-panel: No window manager registered on screen 0. To start the panel without this check, run with --disable-wm-check.
That is basically the same as found in .xsessions-errors
# cd /etc/skel/.config
# ls -al
returns same as yours, same permissions, different timestamps. In addition I have
drwxr-xr-x 2 root root 4096 apr. 3 23:39 vlc
# cd /etc/skel
# ls -al
returns same as yours, same permissions, different timestamps.
I suspect that’s what I will do… And possible it won’t help. I’m starting to think this is caused by some hardware no longer supported.
If you do, please use the latest available Manjaro installer; this alone could make all the difference by avoiding potential complication with older packages.
Cheers.
Will the installer environment use same drivers as what will be installed? (Is a working installer environment a confirmation that the installed environment will work?)
I’m currently at a location with slow internet via mobile phone hotspot (barely any LTE reception). Is there a way to install without downloading more than actually needed?
(I see an installer iso that is over 2.7GB - 3.5GB. On top of that I expect 3GB to be downloaded during the installation. I’d suspect that what is downloaded during installation is already in the iso running the installer environment. Is there a way to make the installer use packages that is already in the iso?)
I believe the installer already does that; at least, up until the first update is performed. The latest release ISO is usually recommended to avoid many issues. If you are using a 6-month-old ISO to install (or repair) Manjaro, for example, there is no wonder that you might have problems of some kind.
Required updates are frequent, and can also vary in size (anywhere between 1GB to 8GB is likely); and this is something that one accepts by virtue of choosing a rolling release distribution.
If you haven’t been updating regularly as required, due to restricted Internet, then this factor could arguably be a root cause of many problems.
Just a few points for consideration. Cheers.
When I installed a manjaro pc some months ago, I got the impression it did download everything during the installation. The installer was downloaded just a few hours before the installation.
This pc was updated on December 18, then the next was n February 12th I think, and on February 27th. The last one was when it screwed up. So, 7 weeks without updates (pc was not in use) went ok. The next 2 weeks did not go well.