Self-hosting is just a generalized term for running your own stacks instead of just using the standard corpo stuff. It’s not a regimen or lifestyle or anything like that. Don’t worry about the terminology.

Okay, so you do, or DO NOT have access to btrfs-assistant right now?

If you’re starting from scratch, then there’s no risk. Run it, and if they are issues, switch to Core. Lemon-Squeezy.

Sure you can, they just have to be containerized add-ons that don’t need host access. Some of those docs were created during the period of time when HASSIOS was a bit new, so a lot of add-ons hadn’t been compatible yet. I thinks that’s a bygone issue for the most part though.

Check what you use now, then go check their compatibility. As a test you could also backup your current HA, boot liveUSB, import your backup, and see if there are any issues. If there are insurmountable issues, just boot back to your normal setup.

👍

So your config is importing other drop-in configs as mentioned here:

# This config file imports drop-in files from /etc/default/grub.d/.

Go look in that directory. You probably just have a duplicate file with your menuentries which you can just rename to be safez then run your mkconfig command again and reboot.

HASSIOS is a purely container-basednsituation that has limitations by design. It doesn’t give you SSH access to the host OS out of the box, but you can change that. If you’re not comfortable with containers, just stick with what you’re running.

What are you expecting to do “more” of by switching?

Yes, you need to be careful about syntax errors and dead links but otherwise, any major issues should be caught by mkconfig. You can always just boot from a liveUSB if you mess something up and fix it.

Yep. Edit /etc/default/grub, then run grub-mkconfig -o /boot/grub/grub.cfg to generate the new menu. Reboot.

Use Ghost to generate a static site and you can host it lots of places for free. Cloudflare Pages, Digitalocean, Surge…etc.

Otherwise just kick up stack and serve for yourself.

Check the cost.

Graylog is probably one of the more simple log aggregators out there, so if that was a bit much, you’re in for a surprise.

SigNoz is probably the best all in one tooling out there at the moment. It’s also a bit heavy, but it’s an open source alternator DataDog, so there’s a lot happening.

Harbor Innovations, the Paper 7 maker? Damn, I was just starting to get interested in trying one out.

It’s all errors. Address them.

Don’t look at the browser, look at the service logs. Turn on debug if needed. You’re only seeing one side of the conversation.

Yeah, but you have a wildcard in your caddy file. Does your SSL cert respond to wildcard requests? One of these services logs has the real issue logged.

These seem like startup logs. Start the server, make a request, and then see if this output changes.

Edit: oh wait, I see you have custom domain stuff happening in there. I’m guaranteeing this is because the server instance is having issues with your personal domain. Check that wherever the server is running can access whatever hostname and port you have there for your auth.

I’m also positive these errors are being logged somewhere. What you posted is not the meat of the issue.

What’s the server log saying though? I’m guessing since it’s throwing a 400 that there is a very clear error there. Probably DB connection.

Very confusing phrasing, but are you just talking about the stock Arch package repo? It’s as safe as I stalling the OS in the first place. The browser is as much a risk as any other package you’ve already installed from the main repo.

Can you give more info on the PC you’re running this on? Model name/number, and the current kernel version you’re on?