So this isn’t really an answer on how to migrate but this solution has worked well for me.

I define the toolboxes config with it, and if it gets destroyed, recreating it is as simple as creating a new one with the same name. You could upgrade or downgrade by just creating a new toolbox with the version you want.

Reminds me of this Steve Huges joke

Youtube

it still mounts and at first glance seems to be working

What makes you say that?

Show us the output of things like lsblk, mount and cat /etc/fstab to give us a fuller picture

If you’re making backups of things you care about and not running sudo rm -rf the command isn’t really dangerous.

But +1 for having it in /tmp I have a bash function I call tempd that is basically cd $(mktemp -d) I use it so much for stuff I dont really care to keep.

That’s really weird. I set up a test system and I couldnt reproduce. The only thing I noticed errors flooding dmesg about elogind already running when I enabled it following the docs. I guess sddm is already starting it?

I dont see how that would cause your issue though. I would probably just reinstall lol

If you CTRL-ALT-F3 and login to a non graphical session does everything work as intended?

Did anybody think that they did?

I always assumed they were just easier to set up

It also depends on the viewer. I remember using prctl() in C to chamge a process name and top showed my change but htop didn’t. I’m sure a competent malware writer would be able to trick it though

Have you tried taking the metwork config out of the compose file and just letting podman handle it?

From wikipedia

Adult cats rarely meow to each other. Thus, an adult cat meowing to human beings is generally considered a post-domestication extension of meowing by kittens: a call for attention.

Gender really is a complex topic huh?

I am root I am admin I am user I am all.

Holy shit I almost died

Anybody else notice the first graph goes from 2020 to 1996?

Hyprland is an official package as of fedora 39

Was thinking “Oh shit now I have to become vegan”, but the article is paywalled so I didn’t have to go on the guilt trip.

Yeah you’re right, it’s time for bed lol

The default start timeout is disabled by default for oneshot.

You could try setting TimeoutStopSec=“infinity” for the service. There may be a default timeout for services and its killing rclone before it can finish because the oneshot type is considered “starting” until the program exits.

Can you share your service file?

I haven’t yet! Today I did a kernel update with it, I was kind of hoping something would go wrong so I would have a bug to report. But nope. Everything worked flawlessly. I’m not really sure how to break it but I’m going to try (in a vm lol)

Yeah pretty much, but it’s wayyyy faster. There’s times where it feels like dnf is hanging trying to download metadata that’s 25KB. I have 1Gb down and it takes like 2 minutes, its ridiculous. I know in the grand scheme of things I’m being petty. But it’s frustrating when the metadata step takes longer than downloading 500MB of packages lol