Not at all. Anything an attacker can connect to is a potential vector.

Keep in mind, this goes for outbound connections as well. If something is compromised server-side, and you or your software reach out to it, that’s also a potential vector. Less likely, since the attacker has to compromise that server and pass their attack off as legit traffic, but it’s possible.

Regardless of what you choose, consider what happens when you lose Internet service or power. A copper line needs neither. If a bad storm comes through and someone gets hurt, you need to be able to call 911.

Just use Immich. It’s easy and apes Google Photos.

Ente is the other main option, but I never evaluated it because Immich worked so well.

Connect via VPN instead of reverse proxy. Don’t expose services directly if you can avoid it.

Maybe there’s some kind of filter software that could limit the visual whiplash. Like you can adjust brightness and contrast on the monitor, but maybe there’s some other software that could add motion blur to fast changes, kind of like volume leveling in audio. Even if a game doesn’t support it, you might be able to use something like ReShade or some system software. Just spitballing, you know your body better than I do.

Plenty of online games that don’t do that. Hell, you can play chess online. Any turn-based game shouldn’t have fast-moving graphics, for a start.

Why can’t you?

2 tb HDD I bought on aliexpress

Well there’s the first red flag. Who’s it made by? Have you verified the capacity with one of those verifier tools? Fake ones will report a huge capacity while having very little, and writes past it just disappear. The error you’re seeing certainly sounds like that kind of behavior.

I don’t. But I do sync the backup data to cloud storage.

So no advantage over something known like Netbird and Wireguard, then

Never heard of it. Why should I use it instead of established systems?

In any system, the human is usually the weakest link.

You don’t need ipv6 for that. You can assign as many ipv4 addresses to an interface as you want.

Edit: this is also certainly not the simplest way, especially since I have no idea how it would interact with docker. It’s a significantly larger management headache than a single rule-based reverse proxy.

Yes, unfortunately 8tb on one drive is not necessarily the exact same as 8tb on any other model drive, even from the same manufacturer. Where possible, you should replace with as close a model as you can get.

If you’re expecting to mix and match, it’s a good idea to leave a little slack space at the end to account for this. Else you may have to fiddle with the partitions as you found, or at worst start a whole new array with the smaller drive and migrate the others.

Drives don’t die every few years. Some will, just out of pure statistics, but generally once they’re broken in, they’ll run for years and years. I buy mine used with up to 40k hours on them (~4.5 years) and they keep running for years more. Usually I end up upgrading them before they actually have issues.

Noise is not really a specification for this kind of drive. I would recommend just putting the server somewhere it won’t matter, if possible.

Most drives are mostly reliable. Backblaze publishes data on reliability if you’re curious, but I would just say to always be prepared to replace a drive. That means RAID and/or backups (but RAID itself is not a backup).

The only really important thing is to avoid SMR drives. Performance is terrible because any write involves a bunch of additional writes.

Test it, find out.

I use a volume bind mount for the container’s config directory, then back that up. And the docker-compose file.

Tenable owns Nessus. But they have a free tier.

The FOSS equivalent is Greenbone OpenVAS: https://greenbone.github.io/docs/latest/

Plus it enables you to access everything. If you have radarr or sonarr or whatever, you can get to those and add media while out and about.

Personally I use Mealie and pull up ingredient lists while I’m im at the grocery store.