Sorry again, there’s no relation between the HTTP route port and the port that’s already in use, I was just thinking loud.
One issue is that there’s a port already in use, another issue is that eturnal isn’t a webapp so there’s no need for an FQDN and traefik HTTP routes.
What you could try is to change the port range.
Edit the service:
runagent -m eturnal1 systemctl --user edit --full eturnal
Change the port range so 50373 isn’t included anymore:
--publish 50000-50100:50000-50100/udp \