HW11: Docker Web App due Sat 11 Nov 23:59

The Big Picture

1. Copy the singlewp directory created while watching this week's videos and name the new directory doublewp.
2. Create a .env file from which we will provide passwords to our systems.
3. Modify the MySQL container so that we set the root password and write a script that will create a database and user for the WordPress app as needed.
4. Add a second WordPress container (mapping it to port 8081). We'll keep a single MySQL container and simply add a new database/user to support the application.

Using an ENV File

In that directory create a new file named .env. In this file we are going put environment variables that we want to use in our docker-compose.yml file. In the .env file you will create environment variables to hold the following values: root MySQL password, name of wordpress database, name of wordpress database user, the password for the database user. It is conventional to use all CAPS for environment variables. NOTE: We will (later) be adding a second WordPress container. That second container will make use of the same MySQL container as the first, but will need access to a different database, database user, and password. So, you may want to consider naming your variables something like: WP_DB_1, WP_USER_1, WP_PASSWORD_1, WP_DB_2, etc.

Once the variables have been set you can modify your docker-compose.yml file to use them. So, the line that reads:

MYSQL_DATABASE: exampledb

MYSQL_DATABASE: ${WP_DB_1}

Also, you'll no longer want MYSQL to use a random root password. Instead it should use the value you specified in the env file.

Test your new YAML file.

Modifying the MySQL Container Setup

Things to remember about how this script will have to run:

• The script will need to be run once for each WordPress instance.
• The values for the database/user/password are found in the .env file on the host server (not in the MySQL container).
• To access the .env file we need to run the script on the host server.
• The commands to actually create a new database/user/password have to be executed inside the MySQL container by the MySQL root user, however.
• So, we'll use two scripts.
  1. The host server script will look up the database/user/password values and use docker exec to connect to the container passing the values as environment variables. The script will require a parameter (1, 2, 3, etc.) to identify which set of environment variables should be passed.

     Once connected to the container it will execute the MySQL container script.

  2. The MySQL container script will read the passed environment variables, connect to MySQL and create the appropriate entries.

  Here are links to my versions of these scripts:

  https://josephus.hsutx.edu/classes/sa/source/installdb

  https://josephus.hsutx.edu/classes/sa/source/add-new-wp-db

  You'll need to change them to match your variable names and configuration. I used a Dockerfile to install the MySQL container script into the container. If you do that you'll create a subdirectory for the Dockerfile and the script and then use the build directive (rather than image) in your YAML file.

  To test this functionality you'll need to delete the existing volume that contains the mysql data and remove the commands in the YAML file that cause a new MySQL user to be created. Then make changes a needed and test.

Add a Second WordPress Container

1. Edit the env file to have a database/user/password with a suffix of 2.
2. Edit the YAML file to have another wordpress definition. It will basically be a copy of the original one. You'll need to change the name, and map it to a different port than the first (perhaps 8081). It will also need a separate volume.
3. Execute the installdb script (with a parameter of 2) to create the database/user/password in the MySQL container.
4. Execute: docker-compose up -d

Test this configuration and verify that you can connect to both sites. You may want to use the lynx text-based browser from your server to remove the need for a second layer of port-forwarding.

Turning In Your Work