Hey all!
I have a bunch of services running on my home server and was looking to expose some of them publicly via Cloudflare tunnel. This is done and working great using the origin server certificate and strict TLS.
Up until now, I’ve been using self-signed certs internally but now I don’t want to deal with the “proceed anyway” crap on browsers. I have Traefik set up to get certs from Cloudflare using DNS challenge and that seems to be working.
So, now my problem is: how do I switch between these certificates for the same URL when I’m internal vs public? I’d rather keep that traffic local if I’m at home, which is also working, I just can’t figure out how to get Traefik to use the appropriate certificate depending on if the request is coming from my LAN or Cloudflare.
Any suggestions? Is there a better way to accomplish what I want to do?
EDIT: Looks like I’m just going full Cloudflare on this one, thanks for your help everyone!
You must log in or # to comment.
Just serve the CloudFlare certs. If the URL is the same, it won’t matter. Doesn’t matter if you’re talking to a local private address like 192.166.1.100 or a public IP. If you’re accessing it via a DNS name, that is what is validated, not the underlying IP.
PS. If you tried this and are having issues. We need more details about how things are set up, and how you are accessing them.
If I use the Cloudflare origin server certs, the browser shows insecure and the message is “certificate not trusted” which is the same message as self-signed, if I’m not mistaken. I’m not sure what other details are relevant as I’m still new-ish to the networking portion of this home server thing. I’m happy to answer any questions if you suspect something.
You said Traefik is getting certs from Cloudflare, but do you mean it’s getting Let’s Encrypt certs using a CF DNS challenge? And if that is the case, then your browser should trust the Traefik endpoint since LE certs are publicly trusted.
Are you sure you’re hitting Traefik when you get a cert warning? You need to update your internal DNS if not.
You’re right, I’m using the cloudflare DNS challenge to get let’s encrypt certs. I’m definitely hitting traefik. I’m testing by turning the Wi-Fi on my phone off/on and opening the page after. I get the same cert every time but it’s not trusted when on Wi-Fi. This makes sense since it’s the origin server cert which is meant to encrypt traffic between my server and cloudflare. To add more certainty, when Wi-Fi is on, a traceroute shows only one hop to my server and shows a bunch of hops when it’s off.
If you, Traefik, and your origin server are on the same network, then it’s going to be one hop regardless of whether you’re hitting the Traefik proxy or the origin server. If Traefik is serving up the origin server’s cert and not the LE cert, then Traefik is misconfigured to pass through instead of proxy, but I’m still not sure that’s the case as it’s almost harder to configure it that way than the correct way as a proxy.
What IP:port is your origin server listening on, what IP:port is Traefik listening on, and how is Traefik configured to reach the origin server?
When I turn off Wi-Fi, I’m not on the same network as my server, it’s my carrier network so all the internet hops are expected.
The way it’s working now is I have a domain (example.com) that is set up on cloudflare DNS. I added a tunnel in cloudflare zero trust, which generates certificates you add to your server to encrypt traffic from your server to cloudflare. I have added these to traefik to be served with my service url (service.example.com). Then, I added a route in cloudflare for service.example.com.
This works fine. But, what I’ve also done is add a local DNS entry for service.example.com so when I’m on my LAN, I access it without going out to the internet and back (seems like a waste). However, this is serving the origin server certs from cloudflare, which causes trust issues
I’m using docker for everything: traefik, cloudflared tunnel, and my services on the same hardware. The tunnel just runs, and it’s configured on cloudflare zero trust to talk directly to the container:port over the docker network.
In that case, if CF is taking to Traefik and not the actual origin server, you just need to forget about the origin certs altogether and use LE certs in Traefik.
I somewhat wonder if CloudFlare is issuing two different certs. An “internal” cert your servers use to serve to CloudFlare, which uses a private CA only valid for CloudFlare’s internal services. CloudFlare’s tunnel service validates against that internal CA, and then serves traffic using an actual public CA signed cert to public internet traffic.
Honestly though, I kinda think you should just go with serving everything entirely externally. Either you trust CloudFlare’s tunnels, or you don’t. If you don’t trust CloudFlare to protect your services, you shouldn’t be using it at all.
That’s what I’m settling on. However, it’s not just about trust, some of the services I’m exposing deal with moving files and I’m mostly interested in higher speeds associated with local transfers as well as not using up my internet data cap.
Here’s a drawing of what I think might be happening to your private traffic: traffic diagram
One major benefit to this approach is CloudFlare does not need to revoke an entire public certificate authority (CA) if a singular private tunnel’s Certificate Authority is compromised.
That sounds like Cloudflare is giving you certificates intended only to be used for talking to Cloudflare.
You might be able to do it if Cloudflare sends a different SNI. It’s probably better if you get real certificates from Let’s Encrypt and just use those.
Why would you need to continue using the self-signed certificates for the Cloudflare connection? Just use the valid certificate for all connections.
I’m not using self-signed anymore, I’m getting them from Cloudflare via DNS challenge
So what’s the problem? They’re valid for the same DNS name.
There’s two ways around the symptoms you’re trying to treat:
- Don’t bother with internal vs external. Always route through external which gets encrypted by the origin cert to CloudFlare and then CloudFlare to your browser. This is simplest in that you don’t need to manage two sets of DNS records and you don’t end up with different certificates for the same domain (in the odd event where you end up needing to do something like certificate pinning). Or;
- Just add the origin cert to your systems’ trust store. You know the certificate, it will encrypt the traffic anyway, also you’re accessing the service via intranet so there’s really no attack vector here.
Probably worth calling out that although 1 feels like there’s more hops (and there absolutely are), with any decent internet, you’re probably not going to feel it. This is because the edge server is probably situated very close to your ISP (that’s how they make sure everything responds quickly) so your over all round trip should only be affected by a negligible amount of time that you most likely won’t notice.
Barring any Traefik tricks that allows me to accomplish what I’m after, I was thinking of going with your “third” option of just letting it use Cloudflare for everything but, I had to check with the experts first before just doing it.
I have some apps that complain or, in one case, flat out doesn’t work if the cert is invalid. I’ve been working around it (sort of) but it would be nice to have it set up “correctly” for once. If routing all traffic through Cloudflare is the answer, so be it ¯_(ツ)_/¯
Acronyms, initialisms, abbreviations, contractions, and other phrases which expand to something larger, that I’ve seen in this thread:
Fewer Letters More Letters CA (SSL) Certificate Authority CF CloudFlare DNS Domain Name Service/System IP Internet Protocol SSL Secure Sockets Layer, for transparent encryption
4 acronyms in this thread; the most compressed thread commented on today has 5 acronyms.
[Thread #743 for this sub, first seen 10th May 2024, 01:55] [FAQ] [Full list] [Contact] [Source code]
