Mageia forum

Hello,

I've run into an issue after the last set of updates with one of my laptops running Mageia 9. I cannot log into my profile with either Plasma (X11) or with icewin.


In both cases, the system hangs and refuses to proceed. I tried to do a bit of troubleshooting before posting here. I used the hot-key combination

Code: Select all

Crtl Alt F2

to bring up a command line and could log into my profile from there. I ran a couple of commands to see if the system needed a further update and to see if I could manually start Plasma:

Code: Select all

 sudo urpmi --auto-update

and

Code: Select all

kstart

The command, "kstart", threw some interesting error messages:

There appears to be a needed Qt platform plug-in "xcb" couldn't load. The error message suggests that "reinstalling the application may fix this problem"

How do I fix this issue?

Thanks,
Brad

Could you list what was updated for it to break?

Have you tested to simply try the previous kernel in the boot menu?

Is there something interesting in the system journal?

Hm, I wonder if kstart is a correct method to start Plasma.

Hello,

I'm not quite sure how to answer your questions or find the information you're asking about. If someone can walk me through the steps, I can post what I find here on this thread.

Have you tested to simply try the previous kernel in the boot menu?

I can try that and post my findings on this thread.

Thanks,
Brad

Hello again,

Have you tested to simply try the previous kernel in the boot menu?

I tried booting to the previous kernel and ran into the same issue. We can cross that off the list.

Thanks,
Brad

Best create a new user, and try logging in as that user.
Also, for the last updates:

Code: Select all

rpm -qa --last | head -250

kstart is not the way to start a plasma session. You probably want to run startplasma-x11

Hello @doktor5000!

Best create a new user, and try logging in as that user.

How do I do this from the command line?

Also, for the last updates:
rpm -qa --last | head -250

Okay, I ran a variation of this command so I got fewer records on the screen:

Code: Select all

rpm -qa --last | head -45

and I got the following:

kstart is not the way to start a plasma session. You probably want to run startplasma-x11

I ran this command and got the following error message:

Code: Select all

[brad@localhost ~]$ startplasma-x11
$DISPLAY is not set or cannot connect to the X server
brad@localhost ~]$


Please let me know what I should do next.

Thanks,
Brad

linuxbrad wrote:

Best create a new user, and try logging in as that user.

How do I do this from the command line?

The following should be sufficient if you want testbrad to be the new username, both run as root

Code: Select all

useradd testbrad
passwd testbrad

I see in screenshot that proprietary nvidia driver is updated.

Maybe try to switch to Xorg nouveau, or modesetting.
see https://wiki.mageia.org/en/How_to_chang ... hic_driver

Hello, everyone,

doktor5000 wrote:
Best create a new user, and try logging in as that user.

How do I do this from the command line?


The following should be sufficient if you want testbrad to be the new username, both run as root

Code: Select all

useradd testbrad
passwd testbrad

I was able to create a new user "testbrad" and log in so it appears that the issue is with my "brad" user profile. How do I recover my "brad" profile so I can log in with plasma?

Thanks,
Brad

Well it would be helpful to know why login with your regular user hangs. Can you try again like in your first post after a fresh boot, and when the login hangs, wait a few minutes.
After that, Ctrl+Alt+F2 and login as your user or as root, and then run

Code: Select all

journalctl -ab > /tmp/journal.log

as root. Then later attach /tmp/journal.log here.

Hi doktor5000! How do I extract /tmp/journal.log from the laptop that is having an issue and get it on the laptop that I'm running this email thread on with only a command line?

Thanks,
Brad

I'd do scp, secure copy via ssh. Although you need to install openssh server beforehand and start the sshd service on the box where you want to copy from.
On the broken box as root:

Code: Select all

urpmi openssh-server
systemctl start sshd

On the other box where you have email (also needs openssh-clients installed, although that is in the default installation) as normal user:

Code: Select all

scp user@brokenbox:/tmp/journal.log /tmp/journal.log

That will copy the file to your email box as /tmp/journal.log

Otherwise on the broken box you can also use something like fpaste or pastebinit to send the file contents to a pastebin service.
Although be aware that the contents of that logfile will be publicly available to everybody you give the pastebin link.

How do I figure out the address of my broken box?

Thanks,
Brad

Basically before you run journalctl - Ctrl+Alt+F2 and login as your user or as root, then

Code: Select all

ip -o a

As an example:

[doktor5000@Mageia9]─[16:08:01]─[~] ip -o a
1: lo inet 127.0.0.1/8 scope host lo\ valid_lft forever preferred_lft forever
2: eno2 inet 192.168.2.121/24 brd 192.168.2.255 scope global dynamic noprefixroute eno2\ valid_lft 1812364sec preferred_lft 1812364sec

Hello doktor5000,

I apologize for the delayed response but I got buried under work and other stuff. Please find the requested journal file attached.

Thanks,
linuxbrad

It's hard to tell why it hangs. plasma is starting for like 10 seconds and seems to immediately exit.
You could try to remove ~/.config/plasma-org.kde.plasma.desktop-appletsrc to check if that helps

Apart from that, I'd say you should disable your NordVPN connection for now to test if that helps. As root

Code: Select all

systemctl disable nordvpnd


Also one autostart entry seems to be broken, you should move away or delete /home/brad/.config/autostart/lbry.desktop for now.

May 15 19:42:13 localhost systemd-xdg-autostart-generator[2301]: Exec binary '/tmp/.mount_LBRY_0XxMKfB/lbry' does not exist: No such file or directory
May 15 19:42:13 localhost systemd-journald[564]: /var/log/journal/76186261c6c2454783d1a9786354174c/user-1001.journal: Monotonic clock jumped backwards relative to last journal entry, rotating.
May 15 19:42:13 localhost systemd-xdg-autostart-generator[2301]: /home/brad/.config/autostart/lbry.desktop: not generating unit, executable specified in Exec= does not exist.

Hi doktor5000,

I did all of the suggestions in your last email and the same issue still occurs. I'm still stuck at the log-in screen.

Should I generate a new "journal" file?

Thanks,
linuxbrad

Hello again,

I'm uploading a new journal.log file (see attached). I did a simple search through the file for "plasma" and got one hit:

Code: Select all

...
May 22 17:36:18 localhost sddm-helper[2254]: Starting: "/usr/share/sddm/scripts/Xsession \"/usr/bin/startplasma-x11\""
May 22 17:36:27 localhost chronyd[1179]: Selected source 172.97.213.179 (pool.ntp.org)
May 22 17:36:28 localhost systemd[1]: Stopping user@986.service...
May 22 17:36:28 localhost systemd[2203]: Activating special unit exit.target...
May 22 17:36:28 localhost systemd[2203]: Stopped target default.target.
May 22 17:36:28 localhost systemd[2203]: Stopping dbus.service...
May 22 17:36:28 localhost systemd[2203]: Stopping pipewire.service...
May 22 17:36:28 localhost systemd[2203]: Stopping xdg-document-portal.service...
May 22 17:36:28 localhost systemd[2203]: Stopping xdg-permission-store.service...
May 22 17:36:28 localhost systemd[2203]: Stopped dbus.service.
May 22 17:36:28 localhost systemd[2203]: Stopped pipewire.service.
May 22 17:36:28 localhost systemd[1]: run-user-986-doc.mount: Deactivated successfully.
May 22 17:36:28 localhost systemd[2203]: Stopped xdg-desktop-portal.service.
May 22 17:36:28 localhost systemd[2203]: xdg-document-portal.service: Main process exited, code=exited, status=20/n/a
May 22 17:36:28 localhost systemd[2203]: xdg-document-portal.service: Failed with result 'exit-code'.
May 22 17:36:28 localhost systemd[2203]: Stopped xdg-document-portal.service.
May 22 17:36:28 localhost systemd[2203]: xdg-permission-store.service: Main process exited, code=exited, status=1/FAILURE
...


I found no references to "kde" in the file. I don't know if that might help.

Thanks,
linuxbrad

Well I really don't know. You could try to manually start an X session to see if that behaves differently. Ctrl+Alt+F2, login as your user and then try startx as I believe startplasma-x11 itself can only be run inside the X server as session.

You can also try to move away more plasma configuration ( stolen from https://forum.manjaro.org/t/cannot-log- ... /153622/20 )

Code: Select all

cd ~/.config ; for j in plasma*; do mv -- "$j" "${j%}.bak"; done; mv Trolltech.conf Trolltech.conf.bak; kbuildsycoca5 --noincremental


Then reboot and try normal login again.

But otherwise I'm afraid you'd need to manually redo your configuration under your new user and then copy over your documents and such.

Hello,
I tried running startx and then sudo startx and got the following errors:

Code: Select all

xauth:  file /home/brad/.serverauth.2484 does not exist
xauth:  timeout in locking authority file /home/brad/.Xauthority
xauth:  timeout in locking authority file /home/brad/.Xauthority
xauth:  timeout in locking authority file /home/brad/.Xauthority
xauth:  timeout in locking authority file /home/brad/.Xauthority

X: Only console users are allowed to run the X server
Authorization required, but no authorization protocol specified

xinit: giving up
xinit: unable to connect to X server: Resource temporarily unavailable
xinit: server error
xauth:  timeout in locking authority file /home/brad/.Xauthority


Code: Select all

auth:  file /root/.serverauth.2587 does not exist

_XSERVTransSocketUNIXCreateListener: ...SocketCreateListener() failed
_XSERVTransMakeAllCOTSServerListeners: server already running
(EE)
Fatal server error:
(EE) Cannot establish any listening sockets - Make sure an X server isn't already running(EE)
(EE)
Please consult the Mageia support
    at https://bugs.mageia.org
 for help.
(EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
(EE)
(EE) Server terminated with error (1). Closing log file.
Invalid MIT-MAGIC-COOKIE-1 key
xinit: giving up
xinit: unable to connect to X server: Resource temporarily unavailable
xinit: server error


Code: Select all

[   850.196] _XSERVTransSocketUNIXCreateListener: ...SocketCreateListener() failed
[   850.196] _XSERVTransMakeAllCOTSServerListeners: server already running
[   850.196] (EE)
Fatal server error:
[   850.196] (EE) Cannot establish any listening sockets - Make sure an X server isn't already running(EE)
[   850.196] (EE)
Please consult the Mageia support
    at https://bugs.mageia.org
 for help.
[   850.196] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[   850.196] (EE)
[   850.196] (EE) Server terminated with error (1). Closing log file.


Does that tell us anything?

If the only way to restore things is to copy over my files from the brad user to the testbrad user, is there an easy way to make a copy of let's say ~/Documents and have the permissions set to allow my testbrad user to access them? It's been ages since I seriously used the command line.

Thanks!

If the only way to restore things is to copy over my files from the brad user to the testbrad user, is there an easy way to make a copy of let's say ~/Documents and have the permissions set to allow my testbrad user to access them? It's been ages since I seriously used the command line.

Would this set of commands work?:

Code: Select all

sudo cp /home/USER1/FNAME /home/USER2/FNAME && sudo chown USER2:USER2 /home/USER2/FNAME

FROM: https://askubuntu.com/questions/551047/copying-files-from-one-user-to-another-in-a-single-machine

However, this doesn't look like it would copy all the files from /home/brad/Documents to /home/testbrad/Documents. I also found this version of the 'copy' command:

Code: Select all

cp -R t1/. t2/

The dot at the end tells it to copy the contents of the current directory, not the directory itself. This method also includes hidden files and folders.

FROM: https://stackoverflow.com/questions/3643848/copy-files-from-one-directory-into-an-existing-directory

This command doesn't address the 'chown' permissions change I need. Maybe I need a combination of the two?

Thanks, as always, for your patience and good advice!

linuxbrad wrote:and then sudo startx

Never, ever run X as root. Only as regular user.

linuxbrad wrote:and got the following errors:

Code: Select all

xauth:  timeout in locking authority file /home/brad/.Xauthority

Please delete /home/brad/.Xauthority and try logging in again normally.

You may need to stop sddm before that if an X server was already started beforehand, as root

Code: Select all

systemctl stop sddm
systemctl isolate multi-user.target

For copying over the files that's the easiest part basically, more on that later.

Never, ever run X as root. Only as regular user.

Ugh, I hope I didn't break anything!

Please delete /home/brad/.Xauthority and try logging in again normally.

And that solved it! Thank you so much, doktor5000! I am typing this from my "broken box".

I shut down the system and started it back up again without issue. Are there any other tests I should be doing? If not, I can mark this issue as "SOLVED".

linuxbrad wrote:Are there any other tests I should be doing?

Not really if your box is working again.

Regarding .Xauthority - this also happens when using the Test option in drakx11 unfortunately, so one should never use this. It's a longstanding bug which sadly has never been fully fixed.
https://bugs.mageia.org/show_bug.cgi?id=20835

Not really if your box is working again.

Alright, I'll mark this thread as "SOLVED".

Regarding .Xauthority - this also happens when using the Test option in drakx11 unfortunately, so one should never use this. It's a longstanding bug which sadly has never been fully fixed.
https://bugs.mageia.org/show_bug.cgi?id=20835

That is unfortunate. I'll make sure to stay away from that functionality.

Thanks again, doktor5000! I appreciate all your help with troubleshooting.