3 rules for backups:

1. Keep 3 copies of your data.
2. Keep 2 of those copies on different types of media
3. Keep 1 of those copies off-site.

I keep a backup on an array of RAIDZ disks. I keep a second copy on a very large hard disk; one that is powered down 99% of the time.

And I keep another at the in-laws house that I can upload to remotely.

You mean the one that was just recently…taken down?

Yeah, didn’t last as long as I’d hoped it would.

Yeah, I couldn’t find that on ollama; but I did find it in text-generation-webui - which is a little more complicated, but for me, I think it might help springboard me into understanding a few more things.

The biggest thing that I want to learn is how to either A: add “tools” for the AI to run, or B: “fine-tune” the model by feeding it data that’s relevant to me.

I bought a CN62 Chromebox, and put MrChromebox’s Bios on it – I did the rounds comparing it with a Pi 4 and it was 2.5x faster, and could easily saturate my gigabit connection. It came with 16gb of storage, and 2gb of ram; but using ACTUAL DRAM slots. I could upgrade it to 16gb if I needed to down the line.

The whole thing, cost me like $45 shipped; power supply, storage, everything needed…and it’s an X86 instruction set - so I can use whatever version of Linux I want, without any crazy Raspberry Pi specific patches/builds.

Proton Pass is a password manager? I’m failing to see why it has anything to do with a web service that you’re running. You should be asking in support for Proton Pass.

Yeah, that’s certainly the truth.

I don’t watch TV, just shows and movies, so I didn’t ever need the DVR functionality. So I get that. NVENC encoding was as simple as choosing it and hitting save; so I’m not sure why you were having troubles there unless you were trying to set up docker or some shit, but that’s on you for using containerization, not on jellyfin.

And the UI is short, sweet, and to the point - exactly what I want to select a show and have it get out of my way. It looks almost exactly like AndroidTV did when it was introduced. Just a nice, clean way to select and start what you want.

Hardware acceleration was as simple as choosing NVENC and saving for me. What are you guys doing wrong?

Just learn a simple reverse proxy and swap out for jellyfin. Other than Plex not handling the user subscription/account side (privacy!) it’s basically the same thing with some small edge cases like people with WebOS TVs and shit.

I’d still think it’s a power issue. I’ve got a bunch of 500gig laptop drives, and ended up getting a 10A 5v supply with a powered hub. Also if you have the chance, power the rpi by the 5v GPIO pins rather than USB, as often the PMIC on the Rpi is anemic and loves to STILL drop under recommended voltage. I run 5.2v 5A PSU on the 5v rail, and haven’t had issues.

If these are 2.5" HDDs, (laptop sized) then maybe not. If they’re the full sized 3.5" HDDs, they need their own external PSU.

I tried it out, and it looks like they include a “data” directory that it reads from and tries to answer questions based on the ‘data’ that’s included. It’s a bunch of text files, so it looks almost as if it’s been made to easily import your own text stuff to guide the LLM.

I’ve replaced reconnaissance commands (a handful of them found here: https://www.cybrary.it/blog/linux-commands-used-attackers) – whoami, uname, id, uptime, last, etc

With shell scripts which run the command but also send me a notification via pushover. I’m running several internet-facing services, and the moment those get run because someone is doing some sleuthing inside the machine, I get notified.

It doesn’t stop people getting in, I’ve set up other things for that – but on the off chance that there is some zero-day that I don’t know about yet, or they’ve traversed the network laterally somehow, the moment they run one of those commands, I know to kill-switch the entire thing.

The thing is, security is an on-going process. Leave any computer attached to the internet long enough and it’ll be gotten into. I don’t trust being able to know every method that can be used, so I use this as a backup.

Honestly, I installed Ombi, so friends can request movies - and gave them all jellyfin logins as well. I’m not running any kind of pay-for service, I’m just giving them access to my library. Additionally, my kids will sometimes spend the night at friends, etc - and their friend won’t have an anime, or a crunchyroll subscription, so they’ll pull it up on jellyfin. It’s easy to remember for them because it’s just jellyfin.mydomain.com

They don’t know anything about how the backend gets the movies/tv shows, just that they go to ombi, and it shows up on jellyfin if they want something ;)

Very nice setup imho. Quite a bit more complicated than mine - mine is basically just the left box without being behind a VPS or anything. I don’t expose anything through Caddy except Jellyfin. I’m also running fail2ban in front of my services, so that if it gets hit with too many 404s because someone is poking around, they get IP banned for 30d

Just what I want to do, move files from A to B in an easy 8 step process!

I kid, I kid…

Docker is wonderful when it is used for the right purposes.

That’s the best thing about Namecheap for me. I just don’t ever have to deal with them. They just kind of exist in the background, not giving me any shit.

You missed CVE – Common Vulnerabilities and Exposures

I’m kinda weirded out by all the people suggesting a VPN here.

Like – if you’re hosting Nextcloud, Jellyfin, etc and you want friends/family to use it, having them VPN into shit is a hurdle that none of them are going to overcome.

You need to make sure you’re not behind CGNAT first, if not, don’t use Nextcloud on port 80, put it on another port, and then open that port to the outside world.

Just be aware, you REALLY want these things to be isolated from your home environment if you’re going to host them, and you NEED to be on some sort of CVE notification list for the software you currently use. Not all CVEs are “YOU MUST UPGRADE NOW”, but some of them can be pretty severe.

I’ve set up fail2ban on my isolated network, and it does a pretty good job of banning any IPs that are probing for things. So much so that I’ve accidentally locked myself out of my own network a few times, lol

IF you ARE behind a CGNAT - what you’ll want to do is likely rent the cheapest VPS you can find, and then set up a VPN not on the VPS, but on your home network, and have the VPS be your public entry point to the network, as it will have a public facing IP and can mask your home IP address. – https://github.com/fractalnetworksco/selfhosted-gateway

Edit: THEN - once you’ve accomplished all that, you’ll probably want to buy a domain name, and reverse-proxy subdomains to forward to the services on specific ports.

Speaking in an offensive manner is not trolling. Trolling is low-effort posts designed to elicit a response. Reddit and Lemmy both have a huge problem with everyone thinking that because someone disagrees with them, and they’re not being “nice about it”, that it means they’re a troll.

That’s not what a troll is. Being nice isn’t a requirement for discussion.