docker-wordpress | Demostrating dev workflow | Infrastructure Automation library

Hello Gonçalo Figueiredo,

Reproducing this issue, it worked just fine when I used kubectl create secret docker-registry and deployed the chart in my GKE cluster in my linux vm.

When I tried using Minikube on my local machine, I did come across a ImagePullBackOff error. The problem was that I had recently changed my dockerHub password and the credentials in my host machine were obsolete. Doing a new docker login solved the issue.

Not sure if it could be happening something similar on your end. If not, could you please check that your private repo is up and try using another method to create the Secret containing the Docker credentials?

When you are in another namespace be sure to put the following in the helm values file:

Try this

Docker-compose need to be defined everything explicitly, consider it as a docker run parameter, it only applies those configurations or parameter on each container that is defined in the docker-compose file.

So one should avoid overriding ENV if they are already declared in Dockerfile so that is the why you need to explicitly define these configurations for your self.

when we build docker we set some default configuration in the docker-image, but docker run command and docker-compose provide the functionality to override these to change that take effect when the image starts up as a container.

A service definition contains configuration that is applied to each container started for that service, much like passing command-line parameters to docker container create. Likewise, network and volume definitions are analogous to docker network create and docker volume create.

compose-file

env_file

Add environment variables from a file. Can be a single value or a list.

If you have specified a Compose file with docker-compose -f FILE, paths in env_file are relative to the directory that file is in.

Environment variables declared in the environment section override these values – this holds true even if those values are empty or undefined.

Your wordpress image used in this stack is default image of Docker Hub and they have very detail documentation of how to install new PHP extension.

First we have to edit some changes on your docker-compose.yml file to make new wordpress custom build:

What worked finally was to adding a proxy_set_header directive to my nginx.conf file:

proxy_set_header Host $http_host;

dockerwp in http://dockerwp is a domain name that does not exist.

You example is only valid if you deploy your wp site to their platform which will give you a subdomain called "dockerwp". The label om.theimpossiblecode.expose.host is read by their reverse proxy to create this effect.

In conclusion you can't set hostname with this configuration. You will be better off just access your site with localhost:port.

First of all, working_dir and volumes need to go into the web section of your compose file, not in the mysql section.

Secondly, the volume mapping is :.

As you specified /var/www/html as your working dir, the wordpress image uses /var/www/html/wp-content as the base directory. So you need to mount the directory on your host with static files into /var/www/html/wp-content. I assuming this to be ~/wordpress.