Lab 3: Create MySQL deployment

In the Docker 101 workshop this was the command to start MySQL.

Do not run this command! It is only here for reference.

docker run -d \
    --network todo-app --network-alias mysql \
    -v todo-mysql-data:/var/lib/mysql \
    -e MYSQL_ROOT_PASSWORD=secret \
    -e MYSQL_DATABASE=todos \
    mysql:8

This is the information from the docker command:

• network and network-alias are not needed in Kubernetes
• Kubernetes has volumes, too, but for simplicity we will start without volumes (but add them later)
• The container image name is mysql:8 (and is implicitly located on Docker Hub)
• There are two environment variables, MYSQL_ROOT_PASSWORD and MYSQL_DATABASE

Using this information, the Kubernetes deployment and service configuration for MySQL looks like this (deploy/mysql-v1.yaml):

apiVersion: apps/v1
kind: Deployment
metadata:
  name: mysql
  labels:
    app: mysql
spec:
  replicas: 1
  selector:
    matchLabels:
      app: mysql
  template:
    metadata:
      labels:
        app: mysql
    spec:
      containers:
      - name: mysql
        image: mysql:8
        ports:
        - containerPort: 3306
        env:
        - name: MYSQL_ROOT_PASSWORD
          value: secret
        - name: MYSQL_DATABASE
          value: todos
---
apiVersion: v1
kind: Service
metadata:
  name: mysql
spec:
  selector:
    app: mysql
  type: NodePort
  ports:
    - protocol: TCP
      port: 3306

This isn’t too different from the todo deployment, the names and labels are changed, and of course the container image reference is different. Also Kubernetes Deployment and Service are handled in a single file which could be simpler to handle.

NOTE: As you can see, we are creating a Kubernetes Deployment for MySQL. Deployments are typically used for stateless applications which a database like MySQL IS NOT. Stateful applications like a database should use Kubernetes Stateful Sets, instead. What we build here is a demo scenario and for a demo it is sufficient (and easier) to use a Deployment. For a real production MySQL environment with primary and secondary instances and replication you would use a MySQL Kubernetes operator which in turn would most likely result in MySQL StatefulSets. For examples see Percona Operators.

The section for environment variables is new:

        env:
        - name: MYSQL_ROOT_PASSWORD
          value: secret
        - name: MYSQL_DATABASE
          value: todos

It contains 2 key-value pairs in a YAML array called env.

1. Deploy the app and check the logs

   In another, second shell start:

    stern mysql
   

   and in your first shell issue:

    kubectl apply -f deploy/mysql-v1.yaml
   

   It will take some time (the MySQL image needs to be downloaded) but then there will be a lot of output in the stern shell. It should state in the end:

    mysql-7bf656bfc9-hpn7v mysql 2023-01-11T13:50:06.461633Z 0 [System] [MY-010931] [Server] /usr/sbin/mysqld: ready for connections.
    Version: '8.0.31'  socket: '/var/run/mysqld/mysqld.sock'  port: 3306  MySQL Community Server - GPL.
   

So MySQL is running in its own pod and waiting for a connection on port 3306 which is MySQL’s default.

Next Step: Connect ToDo with MySQL using Environment Variables