Cloudflare and Docker problems Issues with Cloudflare setup Docker configuration challenges
Cloudflare and Docker problems Issues with Cloudflare setup Docker configuration challenges
I’ll keep an eye on this discussion to the fullest and any advice or support is welcome. Recently I experienced some challenging conditions, and after a while I’m reaching out to fellow users. Earlier we faced intense weather and I normally turned off my server, but when I restarted it all my sites went offline. I set A records in Cloudflare and Nginx for every app I use across different devices such as Overseer, Portainer, and Nginx itself. All domains functioned smoothly before the outage; once restored, they stopped working entirely. Upon inspection, nothing seemed altered when the system came back online. Since then, only Cloudflare error 522 appears. I’m struggling to pinpoint the problem or its path. At first I assumed an SSL authentication failure between Cloudflare and Nginx, but that didn’t resolve anything. Please note, none of these sites are operational at the moment. I managed six of them, so it doesn’t seem like a single configuration issue within a Docker container would be the cause.
Cloudflare: An A record for Nginx.example.com using my local IP is redirected. HTTPS is enabled and SSL/TLS encryption is set to Full. I attempted Full (strict) mode without success. API token was provided for all Nginx SSL zones, but it didn’t help.
Nginx: The domain was created in Cloudflare. Current destination points to 127.0.0.1:81 or localhost:81. A note: I’m using a wildcard certificate (*.example.com) that’s linked to Cloudflare, and supplied the API token. I ran curl to verify its validity.
Portainer: Version 2.18.2. I’m using the nginx image: jc21/nginx-proxy-manager:latest. Ports 80, 81, and 443 are published in the container and forwarded through my ISP router. I’ve tried multiple approaches—starting fresh, adding domains to the network, disabling SSL, even attempting an unsecured connection—but the 522 error persists.
Potential concerns: During research, I saw Overseer running on 10.0.0.208 with a port number, while Nginx is on 127.0.0.1. I haven’t specified IPs for these containers, and in Portainer they aren’t part of a defined network. When checking port forwarding on my ISP router, it reserved an IP (10.0.0.208) but that doesn’t appear elsewhere, and the connected Ethernet interface shows 192.168.x.x instead.
Hardware details: Ubuntu 22.04.2 LTS, i3-9100F CPU, AMD Radeon RX 560, 16GB RAM, RealTek Gig/Ethernet NIC, 40TB storage.
You noted a power issue and asked about your public IP. Could you verify it by using a hotspot to fetch your public IP via curl, and ensure Nginx is functioning correctly from there? If successful, setting up a Mitmproxy container between Cloudflare and Nginx might help you analyze traffic and troubleshoot further.
After some research and troubleshooting, I realized what happened about two weeks ago. When the power went out and the system came back, everything seemed normal except for the config file. It got corrupted, and even after redeploying a new image, it stayed messed up because my friend copied the wrong settings without understanding how things work. He used a config file that needed port 81 to forward Nginx traffic instead of the original setup. Since port 81 is the admin port for Nginx, I didn’t use it as a forward port and thought I’d fixed the issue. All other parts were fine.