2xsaiko

Most computers with (at least) two network interfaces will do. If it’s something too crappy your throughput will be limited by CPU speed but I can’t tell you exact recommendations here. Here’s OPNsense’s hardware recommendations for example, they’re not high at all. Off-the-shelf devices that allow you to do this should probably be fine too.

I’d put Linux on it and use nftables but BSD PF seems to be very popular for firewalls (OPNsense/pfSense are built on this) which I have never used so consider that too.

Not a professional networking guy either but here’s my opinion.

What I would do is use the ISP router as is, open all ports on it (except to itself, hopefully it doesn’t do that…), and put a firewall in between the router and everything else that controls the actual access to everything behind it (in bridge mode between the two network interfaces of the firewall, so you only have the one network).

Could a potential second router also assign addresses to devices in that globally routable space directly?

Devices in IPv6 assign addresses themselves via SLAAC, you just need one device advertising the prefix which the ISP router should already do. The firewall should be able to just purely be there for packet filtering. If you need fixed addresses for public facing servers I would just assign them manually to the respective boxes as you likely also need to add them to public DNS manually anyway.

Huh, I thought I looked through them all when I tried it last time. I’ll check again.

Do you self-host Jitsi? The public instance has absolutely unusable FPS for streaming gameplay which is pretty much the only thing I still use discord for because it’s the only thing that seems to do it well. I read somewhere you can turn up the FPS on a self-hosted Jitsi though.

Yup!

Never seen this before, but you can enable NFS debugging with ‘rpcdebug -m nfs -s all’ (or nfsd on the server, or rpc for the underlying protocol). It prints to dmesg.

IIRC Keepass2Android does have that feature.

What they suggest sounds like setting up a bridge interface between your LAN and the VPN interface to connect the VPS with your LAN. That’s actually a good idea since it would not need you to have a separate /64 for your local network. In this case I’m pretty sure that your VPN needs to be a layer 2 VPN, i.e. transports whole ethernet frames instead of TCP/UDP only, for this to work correctly. Wireguard doesn’t do this, OpenVPN can for example.

To make the VPS a gateway, you need to configure it to forward packets between networks and then set it as your default route on the clients (with IPv6, default route is usually published using router advertisements, set up radvd service on your VPS for that). That’s pretty much it IIRC except for the firewall rules. Here’s an article that’s some cloud stuff but is also applicable to your situation: https://www.linode.com/docs/guides/linux-router-and-ip-forwarding/#enable-ip-forwarding

I had the network before moving here (created it when I did have a public IPv4). Can’t test creating one new since it will only allow me to make one per IP.

Hm, it doesn’t? I’m not behind CGNAT but I’m in a network I don’t control (university dorm) so my gateway is just another device in the local network and I don’t have a public IP which I control, which I feel like should effectively be the same thing as CGNAT, and it works for me. Maybe it isn’t the same.

The easy way is to just use tunnelbroker.net, that is what I currently have (this would use one of their assigned net blocks, not the one from the VPS). Set it up on the Pi, set up IP forwarding with appropriate firewall rules, make the Pi serve RA so clients can assign themselves an IP, done (IIRC).

If you want to set up the v6/v4 gateway yourself, I would do this with a /64 you can fully route to your home network like you would get with tunnelbroker.net because then you don’t have to deal with the network split and essentially two gateways for the same network (your Pi and the VPS), because otherwise your clients would assume the VPS is directly reachable since it’s in the same network when in reality it would have to go through the gateway (you would have to set up an extra route in that case on every client, I think). You’d need a second network from Oracle for this.

But it’s pretty much the same thing I would assume plus the setup on the VPS side, make the VPN route your /64 block (or use 6in4 which is what tunnelbroker.net uses), configure IP forwarding on the Pi and the VPS between the VPN interface and local/WAN respectively.

You’re looking for an OAuth-compatible identity provider (personally I use Kanidm, if Keycloak does that too that works, I’ve never used it). And then set it up as the auth mechanism for Immich, and whatever else you want: https://immich.app/docs/administration/oauth

How about GNU M4 + Make (output)?

(to be clear this is a joke suggestion. but yes it is what I legitimately use)

Don’t use passwords for public SSH in the first place. Disable password authentication and use pubkeys.

Registrars (or DNS providers if you don’t use the one that comes with your registrar) worth using have an API to manage DNS entries. That’s basically all there is to DynDNS.

It offers no practical benefit to small networks at the moment.

The internet is not a “small network”, and I assume your small network is connected to it. You need local IPv6 routing to have access to IPv6-only hosts which are becoming more and more because it’s reasonable in terms of price to get an IPv6 block unlike IPv4 blocks which are being auctioned for tens of thousands of dollars at this point (!!!).

Also restoring global addressing is a huge benefit. P2P communications in IPv4 has become an insane mess of workarounds due to lack of addresses and this becomes worse the more layers of NAT you stick behind each other to try to save your ass from the rising tide.

I’m really sick of hearing these idiotic excuses over and over, “it’s hard” this, “it’s unsafe” that, “it’s expensive”, “understanding the eldritch secrets of IPv6 has driven 5 of my colleagues into madness” skill issue. THERE ARE NO MORE IPV4 ADDRESSES. So unless your network is so fucked that you haven’t managed to fix it in 26 years, since IPv6 has been standardized, or it really is just an internal network with no outward facing services where it doesn’t matter when someone who just has IPv6 can’t access it because they wouldn’t be able to access it anyway, and you’re not some kind of ISP, you have no reason not to have support for it at this point and you absolutely never have a reason to tell people it’s not “useful” because that is straight up wrong in the general case even if it might be true for your situation.

Borg is great and I use it myself but afaik there is no Windows version and there is only remote support over SSH, not HTTPS.

fcgiwrap is what you want for CGI in nginx.

I use it for Mullvad and a couple internal things but yeah it works for me.