Cannot connect to Wordpress docker

[madsmh]

I have the following compose.yml :

networks:
  ip6net:
    enable_ipv6: true
    ipam:
      config:
        - subnet: fd00::/8
services:
  wordpress:
    image: wordpress:php8.5-fpm
    container_name: wordpress
    restart: always
    ports:
      - 127.0.0.1:9000:9000
    environment:
      WORDPRESS_DB_HOST: mariadb
      WORDPRESS_DB_USER: wordpress
      WORDPRESS_DB_PASSWORD: "XXXX"
      WORDPRESS_DB_NAME: blog
      WORDPRESS_DEBUG: 1
    volumes:
      - wordpress:/var/www/html  
  mariadb:
    container_name: mariadb
    image: mariadb:13-ubi
    restart: always
    environment:
      MARIADB_DATABASE: blog
      MARIADB_USER: wordpress
      MARIADB_PASSWORD: "XXXX"
      MARIADB_RANDOM_ROOT_PASSWORD: '1'
    volumes:
      - mariadb-data:/var/lib/mysql
    ports:
      - 127.0.0.1:3306:3306
volumes:
  mariadb-data:
  wordpress:

but when I tunnel locally via SSH to the server, I get a connection reset error in Firefox. The logs for both wordpress and mariadb show no errors.

What am I doing wrong?

[sueun-dev]

Port 9000 of the -fpm image is not an HTTP server. It is PHP-FPM, which speaks FastCGI, so a browser cannot talk to it directly, even through an SSH tunnel. The image docs say this variant needs a reverse proxy such as NGINX or Apache that speaks FastCGI. That would also explain why neither log shows an error.

I checked with wordpress:php8.5-fpm (sha256:cab80f4e) and no network: FPM listened on 9000, and a plain HTTP GET to it returned zero bytes with "Connection reset by peer", which matches your Firefox error.

Two ways forward: use the default wordpress:php8.5 tag, which is the Apache image and serves HTTP on port 80 (a maintainer suggested the same in #209), or put an NGINX container on the same compose network with fastcgi_pass wordpress:9000; and the same wordpress volume mounted so it can serve static files (see tianon's reply in #1019). I did not test an NGINX config.

[madsmh]

Thank you for your reply! I gave up on FPM.
I am now trying with the compose file:

networks:
  ip6net:
    enable_ipv6: true
services:
  wordpress:
    depends_on: 
      - mariadb
    image: wordpress:php8.5
    container_name: wordpress
    restart: always
    ports:
      - 127.0.0.1:8087:80
    environment:
      WORDPRESS_DB_HOST: mariadb
      WORDPRESS_DB_USER: wordpress
      WORDPRESS_DB_PASSWORD: "XXX"
      WORDPRESS_DB_NAME: blog
    volumes:
      - wordpress:/var/www/html
    networks:
      default:
        aliases:
          - blog.example.dk  
  mariadb:
    container_name: mariadb
    image: mariadb:13-ubi
    restart: always
    environment:
      MARIADB_DATABASE: blog
      MARIADB_USER: wordpress
      MARIADB_PASSWORD: "XXX"
      MARIADB_RANDOM_ROOT_PASSWORD: '1'
    volumes:
      - mariadb-data:/var/lib/mysql
    ports:
      - 127.0.0.1:3306:3306
volumes:
  mariadb-data:
  wordpress:

and with Nginx running on the host:

server {
    server_name blog.example.dk;
    error_log /var/log/nginx/blog.example.dk.error.log;
    location / { 
        proxy_pass http://127.0.0.1:8087;
	proxy_redirect  off;
	proxy_set_header  X-Real-IP $remote_addr;
	proxy_set_header  X-Forwarded-For $proxy_add_x_forwarded_for;
	proxy_set_header  X-Forwarded-Proto $scheme;
	proxy_set_header  Host $host;
	
	rewrite ^((?:/[^/\.?]+)*)$ $1/ break;
  
    } 
    
  
    
    listen 443 ssl; # managed by Certbot
    listen [::]:443 ssl; # managed by Certbot
    # http2  on;
    ssl_certificate /etc/letsencrypt/live/blog.example.dk/fullchain.pem; # managed by Certbot
    ssl_certificate_key /etc/letsencrypt/live/blog.example.dk/privkey.pem; # managed by Certbot
    include /etc/letsencrypt/options-ssl-nginx.conf; # managed by Certbot
    ssl_dhparam /etc/letsencrypt/ssl-dhparams.pem; # managed by Certbot

}

server {
    if ($host = blog.example.dk) {
        return 301 https://$host$request_uri;
    } # managed by Certbot


    server_name blog.example.dk;
    listen 80;
    listen [::]:80;
    return 404; # managed by Certbot

But I get these errors: