WP configuration

[Able2Code]

By installing Wordpress and using ob_php-fpm container, i'm getting the mixed-content error over the entire site. How to correct this error? Thank you in advance !

[Able2Code]

Solved by editing wp-config.php define('FORCE_SSL_ADMIN', true); if ($_SERVER['HTTP_X_FORWARDED_PROTO'] == 'https') $_SERVER['HTTPS']='on'; but not i'm trowing "HTTP 502 / No backend servers found at the moment" after some plugins registration. How to solve this?

[Able2Code]

Hi @tspicer , how to prevent nginx to cache wordpress pages when i'm logged as administrator? i'm using the php configuration with redis, in many cases when i visit any site page, the cache save also the wp admin toolbar header , by visiting in incognito i can see the cached wp admin toolbar. I see also you wordpress dedicated git, this config will fix my problem? can i use you installer in any vps ?

[wualex]

@Able2Code how did you solve the 502 issue? have the same problem

[tspicer]

Since this is an NGINX (not Wordpress) repo, I will do my best to assist.

It is hard to troubleshoot these errors given the variations of configs. I'm happy to help, but please post configs for NGINX and wordpress. For example, in NGINX you can set it to not cache anything /wp-admin*. Also, if you have a WP installer you are using, please share that as well.

[wualex]

This would be my docker compose file

`version: '3'

services: web: depends_on:

• wp
• redis image: openbridge/nginx environment: NGINX_DOCROOT: /var/www/html NGINX_CONFIG: php NGINX_SERVER_NAME: somedomain.tld PHP_FPM_UPSTREAM: wp:9000 NGINX_PROXY_UPSTREAM: wp:9000 REDIS_UPSTREAM: redis:6379 ports:
• '80:80'
• '443:443' volumes:
• ./wp-web:/var/www/html
• ./dhparam.pem:/etc/pki/tls/dhparam.pem
• /etc/letsencrypt/live/somedomain.tld/fullchain.pem:/etc/letsencrypt/live/somedomain.tld/fullchain.pem
• /etc/letsencrypt/live/somedomain.tld/privkey.pem:/etc/letsencrypt/live/somedomain.tld/privkey.pem
• /etc/letsencrypt/live/somedomain.tld/chain.pem:/etc/letsencrypt/live/somedomain.tld/chain.pem restart: always networks:

• net

  db: image: mysql:5.7 volumes:

• ./wp-db:/var/lib/mysql restart: always environment: MYSQL_ROOT_PASSWORD: passrandom MYSQL_DATABASE: wordpress MYSQL_USER: wordpress MYSQL_PASSWORD: randompass networks:

• net

  wp: depends_on:

• db image: wordpress:php7.3-fpm-alpine restart: always volumes:
• ./wp-web:/var/www/html environment: WORDPRESS_DB_HOST: db:3306 WORDPRESS_DB_USER: wordpress WORDPRESS_DB_PASSWORD: randompass networks:

• net

  redis: image: redis:alpine container_name: redis networks:

• net restart: unless-stopped networks: net volumes: wp-db:`

docker ps

CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 44468c130c9b openbridge/nginx "/docker-entrypoint.…" 3 days ago Up 3 days *:80->80/tcp, *:443->443/tcp rec_web_1 c0cb54a5774a wordpress:php7.3-fpm-alpine "docker-entrypoint.s…" 3 days ago Up 3 days 9000/tcp rec_wp_1 ab84677495fb mysql:5.7 "docker-entrypoint.s…" 3 days ago Up 3 days 3306/tcp, 33060/tcp rec_db_1 4257ef159d62 redis:alpine "docker-entrypoint.s…" 3 days ago Up 3 days 6379/tcp redis

But when i try to access somedomain.tld i get "HTTP 502 / No backend servers found at the moment"

[tspicer]

Hi, I noticed you set your PROXY address to your PHP service. Is there a reason you are directing NGINX to connect to PHP as a PROXY? You have NGINX_PROXY_UPSTREAM: wp:9000. I would normally expect a reverse proxy to be internal, so something like NGINX_PROXY_UPSTREAM=localhost:8080

The docs dig into this: https://github.com/openbridge/nginx#digging-into-the-env-file

Take a look at where this setting is used: https://github.com/openbridge/nginx/tree/master/conf/php/nginx/upstream.d

In a nutshell, it appears you are treating the proxy and php-fpm as the same thing. Maybe this is a typo?

I can envision certain configurations where the PROXY is a different connection, but this is an advanced setup. Also, you may want to try using this container for PHP-FPM: https://github.com/openbridge/ob_php-fpm

It affords more flexibility and is optimized for this WP.

[tspicer]

In nginx.conf notice the upstream includes are distinct for each service. If PHP and PROXY are the same, I'm not sure how it would work properly.

upstream php-fpm      {include /etc/nginx/upstream.d/php-fpm.conf;}
upstream proxy      {include /etc/nginx/upstream.d/proxy.conf;}
upstream redis     {include /etc/nginx/upstream.d/redis.conf;}

[wualex]

@tspicer changed NGINX_PROXY_UPSTREAM to localhost:8080 and that made it work i can't wrap my head around it at the moment but i'll read some more docs to understand how it really works

thank you!

[tspicer]

awesome, glad to hear.