As said in the title, I’m in the process of moving some of my servers onto different networks. Upon moving my tunnel server, which now has 2 interfaces on 2 different networks instead of 1 for everything, any attempt to visit tunneled sites results in a not found error.
I’ve tried to re-install the tunnel server multiple times, including reinstalling the operating system. I’ve tried to install it both natively on the server and in a docker container. Every time I’ve installed it sudo cloudflared status shows the following:
It looks like your configuration file (config.yml or config.yaml) is missing. I would go through the steps to set up your tunnel again if you don’t have a backup of that file:
Perhaps I’m missing something, but I had set up these through the Zero Trust Dashboard, not through the CLI. As far as I can see the instructions do not mention a config.yml for that route.
My assumption is you’ve lost some configuration in the move, but I’m not sure exactly what you mean by “moving some of my servers”.
I would recommend visiting the tunnels dashboard and access the Overview page of your tunnel. From there under the Install and run a connector section copy &n run the sudo cloudflared service install <TOKEN> command you see there. This should ensure your chosen tunnel configuration is used by your server.
I’m currently using a tunnel server to allow external access to 2 different servers. Previously all three were connected to the same network. At the moment the servers being tunneled have both been re-connected to 2 different networks, with the tunnel server being connected to both networks. All three servers had not been altered in any way outside of the network connections, with only the tunnel server being re-installed from scratch after the problem had started. Currently there are no firewall rules that would prevent communication between the networks.
I tried to run the command you mentioned, and it gave the following error:
As mentioned above, I had attempted a complete re-install of everything on the tunnel server after encountering the issue, including deleting the previous tunnel from the dashboard and re-installing the operating system of the tunnel server. This did not appear to have any effect as far as I can tell. I have not yet tried to set up a tunnel through the CLI.
I think I have it working now. I made a configuration file and built a tunnel through the CLI. Both sites appear to be working now.
From what I can tell, it was entirely an error on my part. One of the servers I was tunneling was a Home-Assistant instance. When I changed networks, I was supposed to change the Home-Assistant config file to accept proxy connections from the new address of the tunnel server. I had forgotten that this needed to be done, and since I was only testing that particular server instead of setting up both, it always appeared to not be working. But since I had switched over to using the other server to test the CLI tunnel, it worked immediately.
