we used to have this in the caddy file:
:3000 {

# Health check para Railway

handle /health {

respond 200 "Healthy"

}

# Manejo API: si la ruta comienza con /api, va al backend

@api {

path /api/*

}

handle @api {

reverse_proxy https://nimva-backend-production.up.railway.app {

transport http {

tls_insecure_skip_verify

}

header_up Host {http.reverse_proxy.upstream.hostport}

}

}

# Resto: frontend

handle {

reverse_proxy https://nimva-front-production.up.railway.app {

header_up Host {http.reverse_proxy.upstream.hostport}

}

}

}

but after getting the error up there "Same machine same service in the most recent hop on"... we changed it to:
{

auto_https off

}

:3000 {

# Health check

handle /health {

respond 200 "Healthy"

}

# API -> Backend (Port 8000)

handle /api/* {

reverse_proxy http://nimva-backend.railway.internal:8000

}

# Rest -> Frontend (Port 8080)

handle {

reverse_proxy http://nimva-front.railway.internal:8080

}

}

and now we get the logs from the attached image for both, front and backend even though those are the right private domains and ports

Terribly sorry about this, but it should be fixed now. Can you confirm?

can't really tell, we stopped using caddy, this is the second issue we have gotten with it though the first one seemed to be due to the domain provider, but now we use the public domain for the backend provided by railway instead