It’s pretty meek on specs. It depends on how much you’d be paying for it. If it was free, I’d say go for it. If it’s $100, you’re halfway to a much more capable machine. The onboard graphics on this are going to be abysmal, even for video.
Friend, I’ve literally linked the DBGate repo. You can see yourself there is no server component running, and it’s all in browser. It’s literally called “web-based”. Have a look here: https://docs.dbgate.io/web-app-config/
So in your world, you imagine that if you run this project, there is a server running…somewhere, and then it’s forwarding all requests from the browser to this server, and the server is making the connections to the DB endpoint? Lolzzzz 🤣🤣🤣
https://github.com/dbgate/dbgate/tree/master/plugins/dbgate-plugin-postgres
I don’t even know where to begin with this 😂
You had better alert the Internet at large and the developers of the apps being discussed here to let them that the very product they build is impossible then.
Oh…wait: https://reintech.io/blog/using-node-js-to-access-remote-database
A very basic example on how to do the very thing you said is not possible you say? While you’re at it, you better go alert Zoom, Google, Microsoft, and anyone else with a WebRTC app that they aren’t allowed to make connections to other things from the browser. It’s totally against the rules and impossible.
🤣
You joking?
Again, Tailscale.
PLEASE look back at the crypto mining rush of a decade ago. I implore you.
You’re buying into something that doesn’t exist.
Look at all your replies, chum. Everyone is already telling you.
Wow, so you want to use inefficient models super cheap. I guarantee nobody has ever thought of this before. Good move coming to Lemmy for tips on how to do so. I bet you’re the next Sam Altman 🤣
It wouldn’t even matter. OP doesn’t understand how any of this works, and is instead just running rampant calling everything bullshit 😂
I assume you’re talking about a CUDA implementation here. There’s ways to do this with that system, and even sub-projects that expand on that. I’m mostly pointing how pointless it is for you to do this. What a waste of time and money.
Edit: others are also pointing this out, but I’m still being downvoted. Mkay.
Cool. You got lucky. This is covered in the docs and is normal behavior.
The problems arise when this exchange doesn’t happen without issue though.
That is a LOCAL running interface. It’s not something being run as a server-side interface in the docker container.
I’m not sure what point you’re trying to make, but at this point, you’re original concern and question has been answered.
Why?
You’re trying to run a DC setup in your home for AI bullshit?
I…don’t think I need to. It’s all open source. Here is the DBGate repo right here.
If you’re unfamiliar with all of this, that’s your job to get educated. This is how browser-based JS software works. The “proof” is right there in all it’s glory for you to peruse.
No, it can run along anything, as long as you don’t conflict the IP space assigned to a VPN. It creates it’s own IP network space when running, so just don’t overlap with your other VPN software. Using it while at home is a bit wasteful on effort and power, but just use the Jellyfin LetsEncrypt setup and it’s the same thing.
You are missing a lot here. I think you’re confused on the difference between your LAN security, and how that fits into network connections. You don’t need an SSL cert to say that something is secure, that’s just one method of PUBLICLY securing something. Every connection on Tailscale is secure end-to-end, so if you run it on your Pi, any client that can connect to it is secured. No open ports, no lapses in security. The encryption happens between each client and the server. You’re secure.
Okay, so you might be unfamiliar with networking, so maybe some extra confusion there. Let me try to explain that a bit.
The Jellyfin server runs on LAN like normal. No need to use Tailscale if you’re just using your Wi-Fi or Ethernet.
Tailscale/Headscale creates it’s own VPN network which will need its own IP space. Same as any other VPN. It’s just a setting in the config, and the routing is pretty simplistic and mostly automatic.
Tailscale/Headscale can run anywhere. Doesn’t need to be on that Pi, but that Pi will need a Tailscale client to be on the “Tailnet” and communicate with other devices also connected to it.
ProtonVPN clients have their own IP space and network that go elsewhere. That’s its own separate thing.
Sorry, it may be confusing, but Headscale is ONLY the free server component. The client is still Tailscale’s open client. That’s why I’m saying just sign up and try it first with Tailscale, and then if you need more connections without paying, create a Headscale server and re-register your clients to that to skip charges.
Nope. Wireguard runs outside the same protocols.
Just give Tailscale a try first because it’s essentially free for a few nodes. If you need more and don’t want to pay, then investigate Headscale.
Okay, so let me explain a bit:
Tailscale is a commercial client that is semi-FOSS. It’s built on Wireguard, which is FOSS, but the cloud hosted architecture does cost money after I think 5 clients.
Headscale is a FOSS implementation of Tailscale, and totally free to host, skipping the above.
Tailscale itself is super easy to use, and you just install it on a node, register it, and then it has access to any other device on that secured network. So if you install it on your Jellyfin machine at home behind your normal firewall, then install it on your phone, you’ll be able to connect to it without forwarding ports for messing around with much.
It should be that simple.
Yeah, and also now, the value of $100 vs $200 in a minipc is vast. That’s a fairly new thing.