Cloudflare tunnel is always down after booting up

What is the name of the domain?

defender.zone

What is the error number?

1033, 502

What is the error message?

Bad gateway

What is the issue you’re encountering

The tunnel always went down shortly after a clean boot up of my machine.

What steps have you taken to resolve the issue?

Reinstall the tunnel or reinitiate token. Verify my device’s network works fine, was able to access my service on localhost. Verified my home router connection is working fine as I have another mini pc that is connected to the same and running another domain (zerohack.io) through Cloudflare tunnel (which works just fine). Unsure if this is just an issue on proxmox machine as I am running the tunnel from the proxmox container.

What are the steps to reproduce the issue?

Reinstall the tunnel or reinitiate token. Install on a fresh new container.

Screenshot of the error

Currently, it looks like you’re missing the CNAME record for your tunnel, or you’ve somehow removed it:

;QUESTION
defender.zone. IN CNAME
;ANSWER
Record not found!

;QUESTION
defender.zone. IN A
;ANSWER
Record not found!

Have you installed it and are you running it as a service or not?

How about running cloudflared tunnel list, do you see them listed or not?

Did you tried to re-authenticate yourself again with cloudflared tunnel login?

May I ask if you’ve created and manage your cloudflared tunnel via Dashboard (remote) or locally (CLI)? :thinking:

I’d suggest remote for easier setup:

Anything in the log files for the tunnel connection and particular hostname?
If you run the cloudflared tunnel command with -loglevel debug.

502 error you’re experiencing is usually an issue at the origin host:

Is your Website running fine over HTTPS (bound to the port 443) and with a valid SSL certificate at the origin host/server or rather not?

If so, then noTLSVerify option should be enabled from the Zero Trust dashboard for your tunnel public hostname.

Otherwise, might be your ISP might not support Quic protocol. Could you select other protocol e.g. http2, using the command and retry again as follows on the instructions from below article? :thinking:

Since 502, looks to me like an issue with missconfigured serving of the domain/hostname on the origin (with SSL) :thinking:

The related sub-domain pve.defender.zone I am presented with Cloudflare Zero Trust Access page - as expected I guess, to protect the direct login page to the Proxmox.

Since running Proxmox, see my posts from below for help:

1 Like

Thanks for the detailed explanation, I managed to figure out the issue was due to internal network and subnet!

1 Like

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.