Tumbleweed Won't Shut Down

I’ve been running Tumbleweed on this computer for several months. There has been a persistent but intermittent problem on shutdown, where Tumbleweed will hang on the last screen before shutdown with the Gigabyte Motherboard screen displaying with the Tumbleweed logo at the bottom, and will never shutdown. I’ve tried to remedy this by researching on the internet and haven’t yet hit on the complete solution. I added the last two parameters to the grub configuration, as such:

splash=silent quiet security=apparmor nosimplefb=1 apm=power_off acpi=force

They made the problem better, but the computer still hangs with the Gigabyte Motherboard screen displaying with Tumbleweed at the bottom and won’t shut down, just less frequently.

I’m running:
Tumbleweed 20230502
KDE Plasma version 5.27.4
KDE Frameworks Version 5.105.0
Qt Version 5.15.9
Kernel Version 6.2.12-1-default (64bit)
Graphics Platform X11

I look forward to the forum’s help to fix this.
Mark

We need your help. What hardware is this? inxi -Faz input/output pasted here using PRE tags may shed some light on possible issues.

Hi,
I tried to educate myself about

 tags last evening, but haven’t been able figure out how to get them to work with a bash console readout.  I’ve created a paste of the readout which I hope will provide the information you requested, in a similar format.

https://paste.opensuse.org/pastes/90e6cc5283bc

Mark

To use PRE tags, click the </> icon above the input window.

There’s nothing in inxi output that I recognize would cause the problem.

You might try appending plymouth=0 or noplymouth or plymouth.enable=0 to the linu line at the Grub menu after striking the E key, to keep boot and shutdown in framebuffer text mode, to see if it helps. If it works, uninstall plymouth, or append it to the GRUB_CMDLINE_LINUX_DEFAULT= line in /etc/default/grub so that next grub.cfg regeneration (via grub2-mkconfig, or YaST, or kernel removal or installation), will include it.

Here’s the plymouth-shutdown-debug.log

https://paste.opensuse.org/pastes/2255f349915b

Here’s the plymouth-shutdown.log

https://paste.opensuse.org/pastes/a70528f3e01c

I don’t see anything in either of them that could show a problem, but please look them over.

I never have Plymouth installed, and never knew such logs existed, so don’t recognize anything in either that might indicate trouble. :stuck_out_tongue:

So you are using nvidia card.
Do you have this in the services manager?

Here’s what I have corresponding to your question and screenshot:

Can you please switch all nvidia services to manual and reboot.
Mine is a bit different from what you have because i installed the nvidia-persistenced.

You have strong preconceptions where to look. However there is nothing to see.

Startup and shutdown are governed by systemd. Thus you may consider to post its actions by running the following commands:

  • As root: journalctl -b -1 -u init.scope -g 'Start|Stop' -o short-monotonic --no-pager.
  • As user: journalctl --user -b -1 -u init.scope -g 'Start|Stopped' -o short-monotonic --no-pager

You post lots of output, but omit the command producing it. Guessing your command lines turns me off.

The output of your commands:

journalctl -b -1 -u init.scope -g ‘Start|Stop’ -o short-monotonic --no-pager

https://paste.opensuse.org/pastes/4827d6d7dd49

~journalctl --user -b -1 -u init.scope -g ‘Start|Stopped’ -o short-monotonic --no-pager

https://paste.opensuse.org/pastes/4cf07b354134

I’m not sure what you were referring to regarding not posting commands. In the paste
https://paste.opensuse.org/pastes/90e6cc5283bc
please note that the command inxi -Faz is included as the first line on the paste.

In the other two pastes showing the plymouth logs, I opened the two logs up in Dolphin. I individually did a “select all” from the drop-down list, then I copied them to paste them in susepaste. There were no commands to produce the output, it was information already present in the logs.

I hope these explanations will ease your stress.

I switched all nvidia services to manual and rebooted. I haven’t noticed a difference so far, but I’ll keep checking and let you know as time goes on.

Great. Pastes of the previous boot show a prefect shutdown. Repost when you encounter a failing shutdown.

Results from after the computer failed to shutdown. I had to shutoff power to the computer because it hung on the Motherboard manufacturer display screen with the Tumbleweed logo at the bottom.

As User
~> journalctl --user -b -1 -u init.scope -g ‘Start|Stopped’ -o short-monotoni
c --no-pager

https://paste.opensuse.org/pastes/98d68cba4bbb

As root:

journalctl -b -1 -u init.scope -g ‘Start|Stop’ -o short-monotonic --no-pager

https://paste.opensuse.org/pastes/729f5423ee48

plymouth-debug.log (remember, this is a copied log and wasn’t originated from bash commands)

https://paste.opensuse.org/pastes/a71e5dc14323

plymouth-shutdown-debug.log (remember, this is a copied log and wasn’t originated from bash commands)

https://paste.opensuse.org/pastes/22afd5592d90

User journal looks fine. System journal looks fine too. However a mess starts after “Stopped target Local File Systems”.

Infamous host erlangen unmounts all file systems and readily stops:

erlangen:~ # journalctl -b -1 |tail -18
May 07 14:32:23 erlangen systemd[1]: Stopped target Local File Systems.
May 07 14:32:23 erlangen systemd[1]: Unmounting /.snapshots...
May 07 14:32:23 erlangen systemd[1]: Unmounting /boot/efi...
May 07 14:32:23 erlangen systemd[1]: Unmounting /boot/grub2/i386-pc...
May 07 14:32:23 erlangen systemd[1]: Unmounting /boot/grub2/x86_64-efi...
May 07 14:32:23 erlangen systemd[1]: Unmounting /home...
May 07 14:32:23 erlangen systemd[1]: Unmounting /home-SSD...
May 07 14:32:23 erlangen systemd[1]: Unmounting /opt...
May 07 14:32:23 erlangen systemd[1]: Unmounting /root...
May 07 14:32:23 erlangen systemd[1]: Unmounting /run/credentials/systemd-tmpfiles-setup.service...
May 07 14:32:23 erlangen systemd[1]: Unmounting /run/credentials/systemd-tmpfiles-setup-dev.service...
May 07 14:32:23 erlangen umount[7749]: umount: /run/credentials/systemd-tmpfiles-setup.service: no mount point specified.
May 07 14:32:23 erlangen systemd[1]: Unmounting /run/credentials/systemd-vconsole-setup.service...
May 07 14:32:23 erlangen systemd[1]: Unmounting /srv...
May 07 14:32:23 erlangen systemd[1]: Unmounting Temporary Directory /tmp...
May 07 14:32:23 erlangen systemd[1]: Unmounting /usr/local...
May 07 14:32:23 erlangen systemd[1]: Stopping Flush Journal to Persistent Storage...
May 07 14:32:23 erlangen systemd[1]: \x2esnapshots.mount: Deactivated successfully.
erlangen:~ # 

You may want to disable lvm2:

erlangen:~ # systemctl list-unit-files lvm2*
UNIT FILE             STATE    PRESET 
lvm2-lvmpolld.service static   -      
lvm2-monitor.service  disabled enabled
lvm2-lvmpolld.socket  disabled enabled

3 unit files listed.
erlangen:~ # 

HI,
here is a screenshot of the services with the lvm2 choices as I have presently configured them at your suggestion:

I tried rebooting the computer several times and this suggestion hasn’t remedied the shutdown problem.

Here are the current logs:

journalctl -b -1 -u init.scope -g ‘Start|Stop’ -o short-monotonic --no-pager

https://paste.opensuse.org/pastes/73a571b59743

~journalctl --user -b -1 -u init.scope -g ‘Start|Stopped’ -o short-monotonic --no-pager

https://paste.opensuse.org/pastes/6238c4d65d9d

A quick question. How much time have you given it to shut down? In Leap and now only recently in Tumbleweed I sporadically get a delayed shutdown that I have never been able to trace, even with considerable troubleshooting instruction. The system runs a 90 second timer while trying to shut down something that is not shutting down normally (apparently). Once the 90 second timer ends, shutdown proceeds normally. I’m not asking for assistance. Just pointing out a possibility if you have not given it considerable time to shut down.

Yea, I’m seeing that behaviour too on one of our [four] TW machines. It’s sporadic, and there’s nothing obvious, so I just turn my attention elsewhere - it shuts down without issue, about 60 second later.
.
.
To @Munguasafiwe … have you tapped the ESC key at GUI shutdown screen to see the terminal output ?

In response to the last two posts:

  1. How long have I waited for it to shut down?
    Sometimes I haven’t caught the failure to completely shutdown because I was tired and went to bed immediately upon shutting down the computer. When I looked at the computer 6-10 hours later it still hadn’t completely shutdown, remaining hung up on the Motherboard Manufacturer’s Logo Screen with the Tumbleweed name and logo at the bottom of the screen. I’ve tried to wait to see if the computer would shutdown when this problem first started happening, but found no matter how long I waited it wouldn’t shut down.

  2. Have I tapped the ESC key at GUI shutdown screen to see the terminal output?
    I think I have some months ago, but I don’t remember for sure. I’ll try it again the next time I get a boot failure and post the results here.

If you wait until the boot failure, the system may not respond to the ESC key. It’s best to tap ESC as soon as you see the GUI-based Shutdown screen (usually displayed within a couple seconds of confirming the shutdown).

That way you can watch the detailed scrolling shutdown output at a console screen … and hopefully the last few lines at lock up might provide a clue.