Caddy-docker-proxy metrics + prometheus help wanted

Help
1

1. The problem I’m having:

I am trying to use this page (Monitoring Caddy with Prometheus and Grafana - TheDeveloperCafe) to set up Prometheus and monitor my Caddy docker container, then eventually monitor my various web facing containers as well. However, I cannot test the metrics because it is locked to only accepting requests from localhost which is the container.

2. Error messages and/or full log output:

$ curl -X GET localhost:2019/metrics
curl: (52) Empty reply from server

3. Caddy version:

2.8.4

4. How I installed and ran Caddy:

docker pull lucaslorentz/caddy-docker-proxy
docker create network caddy

a. System environment:

Ubuntu 24.04.1 LTS
Docker version 27.3.1
lucaslorentz/caddy-docker-proxy 2.9.2

b. Command:

docker compose up -d

c. Service/unit/compose file:

services:
  caddy:
    image: lucaslorentz/caddy-docker-proxy:ci-alpine
    container_name: caddy
    ports:
      - 0.0.0.0:80:80
      - 0.0.0.0:443:443
      - 0.0.0.0:443:443/udp
      - 0.0.0.0:2019:2019
    environment:
      - CADDY_INGRESS_NETWORKS=caddy
      - TZ="Europe/London"
    networks:
      - caddy
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
      - ./data:/data
      - ./config:/config
    restart: unless-stopped
    labels:
      caddy.email: info@mydomain.co.uk
networks:
  caddy:
    external: true

d. My complete Caddy config:

{
	email info@mydomain.co.uk
}

5. Links to relevant resources:

2

There are multiple ways to do it. You can change the listening address of the admin endpoint to something different.

You can also define another site on a different port and use the metrics directive to serve metrics on that address/port.

The Caddyfile configuration has changed a bit since that blog post was published. I recommend you refer to our official documentation.

3

Thanks for your quick reply!

Your first suggestion to “change the listening address of the admin endpoint to something different” didn’t seem to work. Here are my settings and output:-

services:
  caddy:
    image: lucaslorentz/caddy-docker-proxy:ci-alpine  ## <<
    container_name: caddy
    ports:
      - 0.0.0.0:80:80
      - 0.0.0.0:443:443
      - 0.0.0.0:443:443/udp
      - 0.0.0.0:2020:2020  ## <<
    environment:
      - TZ="Europe/London"
      - CADDY_INGRESS_NETWORKS=caddy
      - CADDY_ADMIN=0.0.0.0:2020  ## <<
    networks:
      - caddy
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
      - ./data:/data
      - ./config:/config
    restart: unless-stopped
    labels:
      caddy.email: info@mydomain.co.uk
      caddy.admin: 0.0.0.0:2020  ## <<
      caddy.metrics: ""
networks:
  caddy:
    external: true

… which gave me this Caddyfile (made by the lucaslorentz docker image) …

{
	admin 0.0.0.0:2020
	email info@mydomain.co.uk
	metrics
}

… and this is the network connections list …

Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name    
tcp        0      0 0.0.0.0:80              0.0.0.0:*               LISTEN      10720/docker-proxy  
tcp        0      0 0.0.0.0:443             0.0.0.0:*               LISTEN      10725/docker-proxy  
tcp        0      0 0.0.0.0:2020            0.0.0.0:*               LISTEN      10741/docker-proxy

… but no joy with telnet or curl …

$ telnet 0.0.0.0 2020
Trying 0.0.0.0...
Connected to 0.0.0.0.
Escape character is '^]'.
Connection closed by foreign host.

$ curl -X GET localhost:2020/metrics
curl: (56) Recv failure: Connection reset by peer

and this is the network connections list in the caddy container (showing the restricted 127.0.0.0 localhost for the admin port 2019) …

/ $ netstat -ntap
Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name    
tcp        0      0 127.0.0.1:2019          0.0.0.0:*               LISTEN      1/caddy
tcp        0      0 :::443                  :::*                    LISTEN      1/caddy
tcp        0      0 :::80                   :::*                    LISTEN      1/caddy

… so that did not work :frowning:

NEXT, to “define another site on a different port and use the metrics directive to serve metrics on that address/port” …

back tomorrow.