Centralization is bad for everyone everywhere.
That bring said… I just moved my homeserver to another city… and I plugged in the power, then I plugged in the ethernet, and that was the whole shebang.
Tunnels made it very easy. No port forwarding no dns configuration no firewall fiddling no nothing.
Why do they have to make it so so easy…
Sure it’s easy to set up, but the same behaviour is what I get with my handrolled solution. I rent a cheap VPS with a fixed IP solely for forwarding all traffic through wireguard. My DNS entries all point to the VPS and my servers connect to the VPS to be reachable. It is absolutely network agnostic and does not require any port shenanigans on the local network nor does it require a fixed IP for the internet connection of my home server.
Data security wise the HTTPS terminates on my own hardware (homeserver with reverse proxy) and the wireguard connection is additionally encrypted. There are no secrets or certificates on the rented VPS beyond the bare minimum for the wireguard tunnel and my public key for SSH access.
Shuttling the packets on the VPS (inet to wireguard) is done by socat because I haven’t had the will or need to get in the weeds with nftables/iptables. I am just happy that it works reliably and am happy to loose some potential bandwidth to the kernelspace/userspace hoops.
Does this cause all traffic at the reverse proxy to appear to come from the source IP of your VPS or does it preserve the original source IP?
I’ve been working on setting up a similar setup myself and am trying to figure out specifically how to handle the forwarding on the VPS.
I also have a similar setup to maiskanzler. But I use iptables to forward the traffic over wireguard and I am able to preserve the original client IP by not snat the packets. I then have to use policy based routing to make sure that traffick goes back out through the wg tunnel.
I’m happy to share info on how to get this working.
Yes, I do loose the origin IP and I’m a little bugged by it. It also means that ALL traffic incoming on a specific port of that VPS can only go to exactly ONE private wireguard peer. You could avoid both of these issues by having the reverse proxy on the VPS (which is why cloudflare works the way it does), but I prefer my https endpoint to be on my own trusted hardware. That’s totally my personal preference though.
I trust my VPS provider to not be interested enough in my data to setup special surveillance tooling for each and every possible software combination their customers might have. Cloudflare on the other hand only has their own software stack to monitor and all customers must adhere to it. It’s by design much easier for them to do statistics or snooping.
You can do reverse proxy on the VPS and use SNI routing (because the requested domain is in clear text over HTTPS), then use Proxy Protocol to attach the real source IP to the TCP packets.
This way, you don’t have to terminate HTTPS on the VPS, and you can load balance between a couple wireguard peers so you have redundancy (or direct them to different reverse proxies or whatever).
On your home servers, you will need an additional frontend(s) that accepts Proxy Protocol from the VPS (as Proxy Protocol packets aren’t standard HTTP/S packets, so standard HTTPS reverse proxies will drop them as unknown/broken/etc).
This way, your home reverse proxy knows the original IP and can attach it to the decrypted http requests as x-forward-for. Or you can do ACLs based on original client IP. Or whatever.
I haven’t found a way to get a firewall that pays attention to Proxy Protocol TCP headers, but I haven’t found that to really be an issue. I don’t really have a use case
It also means that ALL traffic incoming on a specific port of that VPS can only go to exactly ONE private wireguard peer. You could avoid both of these issues by having the reverse proxy on the VPS (which is why cloudflare works the way it does), but I prefer my https endpoint to be on my own trusted hardware.
For TLS-based protocols like HTTPS you can run a reverse proxy on the VPS that only looks at the SNI (server name indication) which does not require the private key to be present on the VPS. That way you can run all your HTTPS endpoints on the same port without issue even if the backend server depends on the host name.
This StackOverflow thread shows how to set that up for a few different reverse proxies.
If you want something cheap check out RackNerd yearly deals. Last I checked they still have listings for black Friday and other old deals once you’ve made an account. I got a server for like $12 a year with 5tb monthly bandwidth. I have 3 servers total with them and haven’t had a problem for the 3 years I’ve been using them.
That would be awesome, currently it’s 500GB for their cheaper option which starts at 23/year. I didn’t find an option to increase the bandwidth before completing the order. Also it needs to be deployed in NY (which would be possibly slow for me in Europe). Finally their isos are somewhat old, the latest Ubuntu they have is 20.04 (which has an EoL next year).
All that being said, 23/year is very cheap for a VPS, and for people in the US that use less than 500GB/month that’s the best deal I’ve ever seen.
