Hello friends,
My EOS/XFCE (fully updated) can never shut down properly if the docker service is running. When I shut it down via the GUI shutdown button or “shutdown -h now” or “systemctl poweroff” etc., the shutdown process always gets stuck with a black screen (blinking cursor) regardless of how long I wait.
However, as long as I stop the docker service prior to initiating the shutdown, it can always shut down properly. Any idea please?
Thanks.
Welcome! I’m not on Xfce myself, so my advice will be limited to this, but if you haven’t already, I would test out to see if this issue is present in both the mainline kernel and the LTS kernel. Sometimes bugs and weird regressions show up in the mainline kernel (currently at 5.13), so sometimes it’s advisable to use the LTS kernel (currently at 5.10) if some issues happen. Give that a try if you haven’t already, hopefully it may help you.
If you won’t find any “smooth” solution, you could perhaps look into creating a systemd service that kill off the process before shutdown.
You might want to remove quiet from the kernel boot parameters in /etc/default/grub to make visible bootup and shutdown messages. It might give some more clue. If you do so, don’t forget to update the grub after editing the file.
Is there a container running in the background?
Is there anything in the journal?
Hello,
I followed your advice and made changes to the grub config to display the bootup and shutdown messages. The last shutdown message was “Reached target system power off” but it never actually powered off the machine. I then switched to the LTS kernel per Scotty_Trees’ advice, and everything was back to normal.
Thank you very much for all your help.
Thanks,
Eric
Mark @Scotty_Trees’s post as the solution.
“Thanks!” is not a solution.
IF they don’t have any more issues, we may never see them again on here 
Might actually have to have a mod possible edit the solution part before the issue is closed if that’s a possibility.
Personally, I think self-solving should be turned off. It is abused too much.
“Give all the power for the user to decide” is the EndeavourOS way after all
It happens sometimes that a user posts on the forum about an issue and while waiting for a response they will find the solution on their own and post about it. Then it’s not more than fair to mark their own solution as solution.
Except that happens once in 100 times. And out of those 1%, most are just people trying to bump up their solutions score.
The genuine cases of self-solutions are exceptionally rare. Negligible.
I wouldn’t know that for sure. Or what the actual statistics are for self-solved threads.
However there are sometimes legitimate cases for marking your own solution as solution if you have really done it yourself.
In case of abuse or marking the wrong post, perhaps a moderating invention should be enough.
At any rate, it is not something I am bothered with (apart from these two posts in this thread). I just try to
Please don’t make a big deal about this mistake. I’ve hardly used any forum so I’m not very familiar with how things should be dealt with in forums.
Since EOS booted and shut down extremely quickly on my machine, I did not even pay attention to the fact that EOS disables bootup/shutdown messages by default. All other distros I used in the past enabled them by default. It was only when pebcak mentioned about enabling them in the grub config that I finally noticed that these messages were not there in the first place (journalctl did not give any information about the failure after a forceful physical power off/on). After enabling the kernel messages and seeing the last shutdown message "Reached target system power off”, I immediately knew it was a problem with the latest mainline 5.13.x kernel. Scotty_Trees’ suggestion was a confirmation. As a matter of fact, I would prefer to be able to mark both suggestions from pebcak and Scotty_Trees as the solution. RCA by log messages is always the way to go.
Thanks again and please take things lightly in life. Doing so will make us feel happier.
No worries mate, I think what others meant wasn’t directly geared at you personally, but just from a culmination of recurrences that they have noticed over time build up. Once a post is solved & therefore locked, if anyone views that post weeks or months later, they may see the incorrect solution selected and assume that that solution will work for them. It’s always good to select the post with the most appropriate solution first. You can always add or edit a post to clarify that one or more posts worked for your solution. It’s my understanding that you didn’t do anything wrong, it was simply just a misunderstanding of sorts, so I wouldn’t worry about it, it is just an ongoing issue that predates your time here.
But thank you for selecting the solution that I mentioned and I’m happy it worked out for you. By selecting that solution it will give any user in the future that sees this the best solution for them to try as well so they may get back to a working system as quickly as possible 
This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.