this post was submitted on 10 Jan 2025

83 points (95.6% liked)

Selfhosted

41132 readers

635 users here now

A place to share alternatives to popular online services that can be self-hosted without giving up privacy or locking you into a service you don't control.

Rules:

Be civil: we're here to support and learn from one another. Insults won't be tolerated. Flame wars are frowned upon.
No spam posting.
Posts have to be centered around self-hosting. There are other communities for discussing hardware or home computing. If it's not obvious why your post topic revolves around selfhosting, please include details to make it clear.
Don't duplicate the full text of your blog or github here. Just post the link for folks to click.
Submission headline should match the article title (don’t cherry-pick information from the title to fit your agenda).
No trolling.

Resources:

selfh.st Newsletter and index of selfhosted software and apps
awesome-selfhosted software
awesome-sysadmin resources
Self-Hosted Podcast from Jupiter Broadcasting

Any issues on the community? Report it using the report flag.

Questions? DM the mods!

founded 2 years ago

MODERATORS

[email protected]

How do you handle SSL certs and internet access in your setup? (lemmy.world)

submitted 1 week ago* (last edited 1 week ago) by [email protected] to c/[email protected]

99 comments fedilink hide all child comments

tldr: I'd like to set up a reverse proxy with a domain and an SSL cert so my partner and I can access a few selfhosted services on the internet but I'm not sure what the best/safest way to do it is. Asking my partner to use tailsclae or wireguard is asking too much unfortunately. I was curious to know what you all recommend.

I have some services running on my LAN that I currently access via tailscale. Some of these services would see some benefit from being accessible on the internet (ex. Immich sharing via a link, switching over from Plex to Jellyfin without requiring my family to learn how to use a VPN, homeassistant voice stuff, etc.) but I'm kind of unsure what the best approach is. Hosting services on the internet has risk and I'd like to reduce that risk as much as possible.

I know a reverse proxy would be beneficial here so I can put all the services on one box and access them via subdomains but where should I host that proxy? On my LAN using a dynamic DNS service? In the cloud? If in the cloud, should I avoid a plan where you share cpu resources with other users and get a dedicated box?
Should I purchase a memorable domain or a domain with a random string of characters so no one could reasonably guess it? Does it matter?
What's the best way to geo-restrict access? Fail2ban? Realistically, the only people that I might give access to live within a couple hundred miles of me.
Any other tips or info you care to share would be greatly appreciated.
Feel free to talk me out of it as well.

(page 2) 49 comments

sorted by: hot top controversial new old

[–] [email protected] 7 points 1 week ago* (last edited 1 week ago) (1 children)

5). Hey OP, don't worry, this can seem kind of scary at first, but it is not that difficult. I've skimmed some of the other comments and there are plenty of good tips here.

2). Yes, you will want your own domain and there is no fear of other people "knowing it" if you have everything set up correctly.

1b). Any cheap VPS will do and you don't need to worry about it being virtualized rather than dedicated. What you really care about is bandwidth speed and limits because a reverse proxy is typically very light on resources. You would be surprised how little CPU/memory it needs.

1a). I use a cheap VPS from RackNerd. Once you have access to your VPS, just install your proxy directly into the OS or in Docker. Whichever is easier. The most important thing for choosing a reverse proxy is automatic TLS/Let's Encrypt. I saw a comment from you about certbot... don't bother with all that nonsense. Either Traefik, Caddy, or Nginx Proxy Manager (not vanilla Nginx) will do all this for you--I personally use Traefik unless for some reason I can't. Way less headaches. The second most important thing to decide is how your VPS in the cloud will connect back to your home securely... I personally use Tailscale for that and it works perfectly fine.

3). Honestly, I think Fail2Ban and geo restrictions are overdoing it. Fail2ban has never gotten me any lift because any sort of modern brute force attack will come from a botnet that has 1000s of unique IPs... never triggering Fail2ban because no repeat offenders. Just ensure your VPS has a firewall enabled and you know what ports you are exposing from Docker and you should be good. If your services don't natively support authentication, look into something like Authelia or Authentik. Rather than Fail2Ban and/or geo restrictions, I would be more inclined to suggest a WAF like Caddy WAF before I reached for geo restrictions. Again, assuming your concern is security, a WAF would do way more for you than IP restrictions which are easily circumvented.

4). Have fun!

EDIT: formatting

[–] [email protected] 1 points 1 week ago

I appreciate the info, thanks

[–] [email protected] 7 points 1 week ago (2 children)

If security is one of your concerns, search for "HTTP client side certificates". TL;DR: you can create certificates to authenticate the client and configure the server to allow connections only from trusted devices. It adds extra security because attackers cannot leverage known vulnerabilities on the services you host since they are blocked at http level.

It is a little difficult to find good and updated documentation but I managed to make it work with nginx. The downside is that Firefox mobile doesn't support them, but Firefox PC and Chrome have no issues.

Of course you want also a server side certificate, the easiest way is to get it from Let's Encrypt

load more comments (2 replies)

[–] [email protected] 6 points 1 week ago (2 children)

I use a central nginx container to redirect to all my other services using a wildcard let's encrypt cert for my internal domain from acme.sh and I access it all externally using a tailscale exit node. The only publicly accessible service that I run is my Lemmy instance. That uses a cloudflare tunnel and is isolated in it's own vlan.

TBH I'm still not really happy having any externally accessible service at all. I know enough about security to know that I don't know enough to secure against much anything. I've been thinking about moving the Lemmy instance to a vps so it can be someone else's problem if something bad leaks out.

[–] [email protected] 3 points 1 week ago

Don't fret, not even Microsoft does.

You're not as valuable as a target as Microsoft.

It's just about risk tokerance. The only way to avoid risk is to not play the game.

[–] [email protected] 1 points 1 week ago* (last edited 1 week ago) (1 children)

wildcard let’s encrypt cert

I know what "wildcard" and "let's encrypt cert" are separately but not together. What's going on with that?

How do you have your tailscale stuff working with ssl? And why did you set up ssl if you were accessing via tailscale anyway? I'm not grilling you here, just interested.

I know enough about security to know that I don’t know enough to secure against much anything

I feel that. I keep meaning to set up something like nagios for monitoring and just haven't gotten around to it yet.

[–] [email protected] 4 points 1 week ago (1 children)

So when I ask Let's Encrypt for a cert, I ask for *.int.teuto.icu instead of specifically jellyfin.int.teuto.icu, that way I can use the same cert for any internally running service. Mostly I use SSL on everything to make browsers complain less. There isn't much security benefit on a local network. I suppose it makes harder to spoof on an external network, but I don't think that's a serious threat for a home net. I used to use home.lan for all of my services, but that has the drawback of redirecting to a search by default on most browsers. I have my tailscale exit node running on my router and it just works with SSL like anything else.

[–] [email protected] 1 points 1 week ago* (last edited 6 days ago) (1 children)

Ok so I currently have a cert set up to work with:

domain.com

www.domain.com (some browsers seemingly didn't like it if I didn't have www)

subdomain.domain.com

Are you saying I could just configure it like this:

domain.com

*.domain.com

The idea of not having to keep updating the cert with new subdomains (and potentially break something in the process) is really appealing

[–] [email protected] 3 points 1 week ago* (last edited 1 week ago) (1 children)

Yes. If you're using lets encrypt then note that they do not support wildcard certs with the HTTP-01 challenge type. You will need to use the DNS-01 challenge type. To utilize it you would need a domain registrar that supports api dns updates like cloudflare and then you can use the acme.sh package. Here is an example guide i found.

Note that you could still request multiple explicit subdomains in the same issue/renew commands so it's not a huge deal either way but the wildcard will be more seamless in the future if you don't know what other services you might want to selfhost.

load more comments (1 replies)

[–] [email protected] 5 points 1 week ago

Nginx Proxy Manager + LetsEncrypt.

[–] [email protected] 5 points 1 week ago (1 children)

I use nginx proxy manager and let's encrypt with a porkbun domain, was very easy to set up for me. Never tried caddy/traefik/etc though. Geo blocking happens on my OPNsense with the built in tools.

[–] [email protected] 0 points 1 week ago (1 children)

Do you have instructions on how you set that up?

[–] [email protected] 2 points 1 week ago* (last edited 1 week ago) (1 children)

At a high level you forward ports 80 and 443 to NPM from your router. In NPM you set up your proxy by IP address and port and you can also set up automatic SSL certs when you create the proxy via letsencrypt. I also run a DDNS auto update that tells porkbun if my IP changes. I'd be happy to get into some more specifics if there's a particular spot you're stuck. This is all assuming you have a public IPv4 and aren't behind cgnat. If you have cgnat you're not totally fucked but it makes it more complicated. If it's OPNsense related struggles that shit is mysterious to me, I've only been running it a few weeks and it's not fully configured. Still learning.

[–] [email protected] 1 points 1 week ago* (last edited 1 week ago) (2 children)

Why am I forwarding all http and https traffic from WAN to a single system on my LAN? Wouldn't that break my DNS?

[–] [email protected] 1 points 1 week ago* (last edited 1 week ago)

The reverse proxy is th single system because it tells the incoming traffic where to go. It also doesn't really do anything unless the incoming traffic is requesting one of the domains you set up. it doesn't affect your internal DNS. You are able to redirect from the public address to your internal server through DNS though.

load more comments (1 replies)

[–] [email protected] 5 points 1 week ago* (last edited 1 week ago) (1 children)

A fairly common setup is something like this:

Internet -> nginx -> backend services.

nginx is the https endpoint and has all the certs. You can manage the certs with letsencrypt on that system. This box now handles all HTTPS traffic to and within your network.

The more paranoid will have parts of this setup all over the world, connected through VPNs so that "your IP is safe". But it's not necessary and costs more. Limit your exposure, ensure your services are up-to-date, and monitor logs.

fail2ban can give some peace-of-mind for SSH scanning and the like. If you're using certs to authenticate rather than passwords though you'll be okay either way.

Update your servers daily. Automate it so you don't need to remember. Even a simple "doupdates" script that just does "apt-get update && apt-get upgrade && reboot" will be fine (though you can make it more smart about when it needs to reboot). Have its output mailed to you so that you see if there are failures.

You can register a cheap domain pretty easily, and then you can sub-domain the different services. nginx can point "x.example.com" to backend service X and "y.example.com" to backend service Y based on the hostname requested.

[–] [email protected] 5 points 1 week ago (1 children)

I would recommend automating only daily security updates, not all updates.

Ubuntu and Debian have “unattended-upgrades” for this. RPM-based distros have an equivalent.

[–] [email protected] 0 points 1 week ago

Agree - good point.

[–] [email protected] 5 points 1 week ago

It doesn’t improve security much to host your reverse proxy outside your network, but it does hide your home IP if you care.

If your app can exploited over the web and through a proxy it doesn’t matter if that proxy is on the same machine or over the network.

[–] [email protected] 4 points 1 week ago* (last edited 1 week ago) (1 children)

On my home network I have nginxproxymanager running let's encrypt with my domain for https, currently only for vaultwarden (I'm testing it for a bit for rolling it out or migrating wholly over to https). My domain is a ######.xyz that's cheap.

For remote access I use Tailscale. For friends and family I give them a relay [raspberry pi with nginx which proxys them over tailscale] that sits on their home network, that way they need "something they have"[the relay] and "something they know" [login credentials] to get at my stuff. I won't implement biometrics for "something they are". This is post hoc justification though, and nonesense to boot. I don't want to expose a port and a VPS has low WAF and I'm not installing tailscale on all of their devices so s relay is an unhappy compromise.

For bonus points I run pihole to pretty up the domain names to service.swirl and run a homarr instance so no-one needs to remember anything except home.swirl, but if they do remember immich.swirl that works too.

If there are many ways to skin a cat I believe I chose to use a spoon, don't be like me. Updating each dockge instance is a couple minutes and updating diet pi is a few minutes more which, individually, is not a lot on my weekly/monthly maintence respectfully. But on aggregate... I have checklists. One day I'll write a script that will ssh into a machine > update/upgrade the os > docker compose pull/rebuild/purge> move on to the next relay... That'll be my impetus to learn how to write a script.

[–] [email protected] 2 points 1 week ago* (last edited 1 week ago) (1 children)

That’ll be my impetus to learn how to write a script.

This part caught my eye. You were able to do all that other stuff without ever attempting to write a script? That's surprising and awesome. Assuming you are running everything on a linux server, I feel like a bash script that is run via a cronjob would be your best bet, no need to ssh into the server, just let it do it on it's own. I haven't tested any of this but I do have scripts I wrote that do automatic ZFS backups and scrubs; the order should go something like:

open the terminal on the server and type

mkdir scripts

cd scripts

nano docker-updates.sh

type something along the lines of this (I'm still learning docker so adjust the commands to your needs)

#!/bin/bash

cd /path/to/scripts/docker-compose.yml
docker compose pull && docker compose up -d
docker image prune -f

save the file and then type sudo chmod +x ./docker-updates.sh to make it executable

and finally set up a cronjob to run the script at specific intervals. type

crontab -e

sudo crontab -e (this is if you want to run the script as root but ideally, you just add your user to the docker group so this shouldn't be needed)

and at the bottom of the file type this and save, that's it:

# runs script at 1am on the first of every month
0 1 1 * * /path/to/scripts/docker-updates.sh

this website will help you choose a different interval

For OS updates you basically do the same thing except the script would look something like: (I forget if you need to type "sudo" or not; it's running as root so I don't think you need it but maybe try it with sudo in front of both "apt"s if it's not working. Also use whatever package manager you have if you aren't using apt)

while in the scripts folder you created earlier

nano os-updates.sh

#!/bin/bash

apt update -y && apt upgrade -y
reboot now

save and don't forget to make it exectuable

then use

sudo crontab -e (because you'll need root privileges to update. this will run the script as root without requiring you to input your password)

# runs script at 12am on the first of every month
0 0 1 * * /path/to/scripts/os-updates.sh

[–] [email protected] 2 points 1 week ago* (last edited 1 week ago) (2 children)

I did think about cron but, long ago, I heard it wasn't best practice to update through cron because the lack of logs makes things difficult to see where things went wrong, when they do.

I've got automatic-upgrades running on stuff so it's mostly fine. Dockge is running purely to give me a way to upgrade docker images without having to ssh. It's just the monthly routine of "apt update && apt upgrade -y" *5 that sucks.

Thank you for the advice though. I'll probably set cron to update the images with the script as you suggest. I have a "maintenance" homarr page as a budget uptime kuma so I can quickly look there to make sure everything is pinging at least. I made the page so I can quickly get to everyone's dockge, pihole and nginx but the pings were a happy accident.

load more comments (2 replies)

[–] [email protected] 4 points 1 week ago (2 children)

The biggest reason to use VPN is that some ISPs may take issue with you running a web server over a residential service when they see incoming HTTP requests to your IP. If you don't want to require VPN, then Cloudflare tunnels are perfect for this and they also solve the need for dynamic DNS if you want to use static domain because your domain points to the Cloudflare edge servers and they route it to you wherever your tunnel endpoint is running.

Past that, Traefik is a great reverse proxy that can manage getting LetsEnrcypt SSL certificates for you even with wildcard domains and would still work fine with dynamic DNS.

[–] [email protected] 2 points 1 week ago (1 children)

Do you mind giving a high level overview of what a Cloudlfare tunnel is doing? Like, what's connected to what and how does the data flow? I've seen cloudflare mentioned a few other times in the comments here. I know Cloudflare offers DNS services via their 1.1.1.1 and 1.0.0.1 IPs and I also know they somehow offer DDoS protection (although I'm not sure how exactly. caching?). However, that's the limit of my knowledge of Cloudflare

load more comments (1 replies)

[–] [email protected] 4 points 1 week ago (1 children)

I got started with a guide from these guys back in 2020. I still use traefik as my reverse proxy and Authelia for authentication and it has worked great all this time. As someone else said, everything is in containers on the one host and it is super easy this way. It all runs on a single box using containers for separation. I should probably look into a secondary server as a live backup, but that’s a lot of work / expense. I have a Cloudflare dynamic DNS container running for that.
I would definitely advocate for owning your own domain, for the added use case of owning your own email addresses. I can now switch email providers and don’t have to worry about losing anything. This would also lean towards a more memorable domain, or at least a second domain that is memorable. Stay away from the country TLDs or “cute” generic TLDs and stay with a tried and true .com or .net (which may take some searching).
I don’t bother with this, I just run my server behind Cloudflare, and let them protect my server. Some might disagree, but it’s easy for me and I like that.
Containers, containers, containers! Probably Docker since it’s easy, but Podman if you really want to get fancy / extra secure. Also, make sure you have a git repo for your compose files, and a solid backup strategy from the start (so much easier than going back and doing it later). I use Backblaze for my backups and it’s $2/month for some peace of mind.
Do it!!!

load more comments (1 replies)

[–] [email protected] 4 points 1 week ago (1 children)

Either tailscale or cloudflare tunnels are the most adapted solution as other comments said.

For tailscale, as you already set it up, just make sure you have an exit node where your services are. I had to do a bit of tinkering to make sure that the ips were resolved : its just an argument to the tailscale command.

But if you dont want to use tailscale because its to complicated to your partner, then cloudlfare tunnels is the other way to go.

How it works is by creating a tunnel between your services and cloudlare, kind of how a vpn would work. You usually use the cloudlfared CLI or directly throught Cloudflare's website to configure the tunnel. You should have a DNS imported to cloudflare by the way, because you have to do a binding such as : service.mydns.com -> myservice.local Cloudlfare can resolve your local service and expose it to a public url.

Just so you know, cloudlfare tunnels are free for some of that usage, however cloudlfare has the keys for your ssl traffic, so they in theory could have a look at your requests.

best of luck for the setup !

load more comments (1 replies)

[–] [email protected] 3 points 1 week ago (2 children)

Tailscale is completely transparent on any devices I've used it on. Install, set up, and never look at it again because unless it gets turned off, it's always on.

[–] [email protected] 2 points 1 week ago (1 children)

I've run into a weird issue where on my phone, tailscale will disconnect and refuse to reconnect for a seemingly random amount of time but usually less than hour. It doesn't happen often but it is often enough that I've started to notice. I'm not sure if it's a network issue or app issue but during that time, I can't connect to my services. All that to say, my tolerance for that is higher than my partner's; the first time something didn't work, they would stop using it lol

[–] [email protected] 2 points 1 week ago

So I have it running on about 20 phones for customers of mine that use Blue Iris with it. But these are all Apple devices, I'm the only one with Android. I've never had a complaint except one person that couldn't get on at all, and we found that for some reason the Blue Iris app was blacklisted in the network settings from using the VPN. But that's the closest I've seen to your problem.

I wonder if you set up a ping every 15 seconds from the device to the server if that would keep the tunnel active and prevent the disconnect. I don't think tailscale has a keepalive function like a wireguard connection. If that's too much of a pain, you might want to just implement Wireguard yourself since you can set a KeepAlive value and the tunnel won't go idle. Tailscale is probably wanting to reduce their overhead so they don't include a keepalive.

[–] [email protected] 1 points 1 week ago

relatable

[–] [email protected] 3 points 1 week ago

I use this https://github.com/ZoeyVid/NPMplus. I use unifi for goe-blocking.

[–] [email protected] 2 points 1 week ago

I use nginx manager in its own docker container on my unraid server. Was pretty simple to set up all things considered. I would call myself better with hardware than software but not a complete newb and I got it running with minimal headache.

[–] [email protected] 2 points 1 week ago (8 children)

nixos with nginx services does all proxying and ssl stuff, fail2ban is there as well

load more comments (8 replies)

[–] [email protected] 2 points 1 week ago (1 children)

I've tried 3 times so far in Python/gradio/Oobabooga and never managed to get certs to work or found a complete visual reference guide that demonstrates a complete working example like what I am looking for in a home network. (Only really commenting to subscribe to watch this post develop, and solicit advice:)

[–] [email protected] 3 points 1 week ago* (last edited 4 days ago) (1 children)

I've played around with reverse proxies and ssl certs and the easiest method I've found so far was docker. Just haven't put anything in production yet. If you don't know how to use docker, learn, it's so worth it.

Here is the tutorial I used and the note I left for myself. You'll need a domain to play around with. Once you figure out how to get NGINX and certbot set up, replacing the helloworld container with a different one is relatively straight forward.

DO NOT FORGET, you must give certbot read write permissions in the docker-compose.yml file which isn't shown in this tutorial
-----EXAMPLE, NOT PRODUCTION CODE----

    nginx:
        container_name: nginx
        restart: unless-stopped
        image: nginx
        depends_on:
            - helloworld
        ports:
            - 80:80
            - 443:443
        volumes:
            - ./nginx/nginx.conf:/etc/nginx/nginx.conf
            - ./certbot/conf:/etc/letsencrypt:ro
            - ./certbot/www:/var/www/certbot:ro

    certbot:
      image: certbot/certbot
      container_name: certbot
      volumes: 
        - ./certbot/conf:/etc/letsencrypt:rw
        - ./certbot/www:/var/www/certbot:rw
      command: certonly --webroot -w /var/www/certbot --keep-until-expiring --email *email* -d *domain1* -d *domain2* --agree-tos

[–] [email protected] 2 points 1 week ago (1 children)

I'd add that Traefik works even better with Docker because you tag your other containers that have web ports and Traefik picks that up from Docker and terminates the SSL connection for them. You don't even have to worry about setting up SSL on every individual service, Traefik will take care of that even for services that don't implement SSL.

[–] [email protected] 1 points 1 week ago (1 children)

You don’t even have to worry about setting up SSL on every individual service

I probably need to look into it more but since traefik is the reverse proxy, doesn't it just get one ssl cert for a domain that all the other services use? I think that's how my current nginx proxy is set up; one cert configured to work with the main domain and a couple subdomains. If I want to add a subdomain, if I remember correctly, I just add it to the config, restart the containers, and certbot gets a new cert for all the domains

load more comments (1 replies)

[–] [email protected] 2 points 1 week ago

I used to do a reverse proxy setup with caddy , but now I self host a Wireguard VPN. It has access to Nextcloud on the same machine, Home Assistant and Kodi on another. On our phones, Wireguard only has access to certain apps the rest of the network traffic is normal, so a nice simple setup.

[–] [email protected] 2 points 1 week ago (6 children)

Cloudflare

[–] [email protected] 3 points 1 week ago (1 children)

I presume you're referring to Cloudflare tunnel?

[–] [email protected] 2 points 1 week ago

Yep, cloudflare tunnel / Zero trust.

Dead easy to set up.

load more comments (5 replies)

load more comments