How to Deploy MySQL on Kubernetes

deploy mysql on kubernetes

This tutorial shows detailed steps of deploying MySQL on Kubernetes. I’ll be using minikube here to demonstrate Kubernetes MySQL examples.

We all know the big importance of data persistence and almost, all our applications rely hugely on some sort of Database Management Systems (DBMS). Setting up a DBMS on Kubernetes helps the DevOps team and Database administrators leverage and scale the Database easily.

Prepare the environment

Follow this tutorial you need to have Minikube installed on your Ubuntu Linux.

You can verify whether the Minikube has been successfully up and running by the following command:

$ minikube status

Output:

minikube
type: Control Plane
host: Running
kubelet: Running
apiserver: Running
kubeconfig: Configured

Create Secret for MySQL

Kubernetes uses Secret to store and manage sensitive information such as passwords, ssh keys and OAuth tokens. In this tutorial, we use base64 encoded to store 'MYSQL_ROOT_PASSWORD'. For example:

$ echo -n 'admin' | base64

Output:

YWRtaW4=

Create a mysql-secret.yaml file for MySQL that will be mapped as an environment variable as follows:

apiVersion: v1
kind: Secret
metadata:
  name: mysql-pass
type: Opaque
data:
  password: YWRtaW4=

Apply the manifest:

$ kubectl create -f mysql-secret.yaml

secret/mysql-pass created

Verify that the Secret has just been created successfully:

$ kubectl get secrets

NAME                  TYPE                                  DATA   AGE
default-token-l7t7b   kubernetes.io/service-account-token   3      4h24m
mysql-pass            Opaque                                1      1m

Deploy MySQL

Create the mysql-pod.yaml file to deploy MySQL pod on Kubernetes cluster:

apiVersion: v1
kind: Pod
metadata:
  name: k8s-mysql
  labels:
    name: lbl-k8s-mysql
spec:
  containers:
  - name: mysql
    image: mysql:latest
    env:
    - name: MYSQL_ROOT_PASSWORD
      valueFrom:
        secretKeyRef:
          name: mysql-pass
          key: password
    ports:
    - name: mysql
      containerPort: 3306
      protocol: TCP
    volumeMounts:
    - name: k8s-mysql-storage
      mountPath: /var/lib/mysql
  volumes:
  - name: k8s-mysql-storage
    emptyDir: {}

Apply the manifest file:

$ kubectl create -f mysql-pod.yaml

pod/k8s-mysql created

Verify that the pod is running:

$ kubectl get pod

NAME        READY   STATUS    RESTARTS   AGE
k8s-mysql   1/1     Running   0          30s

Now, we can connect to the k8s-mysql pod:

$ kubectl exec k8s-mysql -it -- bash

root@k8s-mysql:/# echo $MYSQL_ROOT_PASSWORD
admin
root@k8s-mysql:/# mysql --user=root --password=$MYSQL_ROOT_PASSWORD
mysql: [Warning] Using a password on the command line interface can be insecure.
Welcome to the MySQL monitor.  Commands end with ; or \g.
Your MySQL connection id is 11
Server version: 8.0.22 MySQL Community Server - GPL

Copyright (c) 2000, 2020, Oracle and/or its affiliates. All rights reserved.

Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

mysql> show databases;
+--------------------+
| Database           |
+--------------------+
| information_schema |
| mysql              |
| performance_schema |
| sys                |
+--------------------+
4 rows in set (0.00 sec)

mysql> 

Kubernetes use Service to expose pods to other pods or external systems. We will use the following manifest file mysql-service.yaml to make the k8s-mysql pod be reachable:

apiVersion: v1
kind: Service
metadata:
  name: mysql-service
  labels:
    name: lbl-k8s-mysql
spec:
  ports:
  - port: 3306
  selector:
    name: lbl-k8s-mysql
  type: ClusterIP

Apply the manifest to create the service:

$ kubectl create -f mysql-service.yaml

service/mysql-service created

Verify that the service has been successfully created:

$ kubectl get svc

NAME            TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)    AGE
kubernetes      ClusterIP   10.96.0.1               443/TCP    5h4m
mysql-service   ClusterIP   10.110.22.182           3306/TCP   30s

Creating a NodeJS Api to hit mysql

In order to be able to connect to mysql from an other pod we need to have the IP address of our pod which can be done using:

$ kubectl get pod k8s-mysql -o template --template={{.status.podIP}}

172.17.0.5

Alright, now I’m going to create a sample NodeJS app, to store a set of messages in database MESSAGES table, the app will have two endpoints:

  • '/ping': to check the server health
  • '/msg-api/all': to get all the stored messages

To keep things simple... the table will have only one column called TEXT.

First thing first, node app:

// api.js -> endpoints goes here

var express = require('express')
var mysql = require('mysql')

var Router = express.Router();
var ConnPool = mysql.createPool({
host: '172.17.0.5',
user: 'root',
password: 'admin',
database: 'k8smysqldb'
})

// create database and MESSAGE table if not exist
ConnPool.query('CREATE DATABASE IF NOT EXISTS k8smysqldb', function (err) {
if (err) throw Error('\n\t **** error creating database **** ' + err)

console.log('\n\t ==== database k8smysqldb created !! ====')

ConnPool.query('USE k8smysqldb', function (err) {
if (err) throw Error('\n\t **** error using database **** ' + err);

console.log('\n\t ==== database k8smysqldb switched !! ====')

ConnPool.query('CREATE TABLE IF NOT EXISTS messages('
+ 'id INT NOT NULL AUTO_INCREMENT,'
+ 'PRIMARY KEY(id),'
+ 'text VARCHAR(100)'
+ ')', function (err) {
if (err) throw Error('\n\t **** error creating table **** ' + err);
})
})
})

/**
* /all
*/
Router.get('/all', function (req, res) {
ConnPool.getConnection(function (errConn, conn) {
if (errConn) throw Error('error get connection : ' + errConn)

conn.query('SELECT * FROM messages', function (errSelect, rows) {
if (errSelect) throw Error('error selecting messages : ' + errSelect)
res.writeHead(200, {
'Content-Type': 'application/json'
});
var result = {
success: true,
rows: rows.length,
}
res.write(JSON.stringify(rows));
res.end();
})
})
})

module.exports = Router

// server.js -> fire expressjs server

var express = require('express')
var msgApi = require('./api')

var app = express()

app.use('/msg-api', msgApi)

app.get('/ping', function (req, res) {
res.write("hello there! I m up and running!");
res.end();
})

app.listen(8080, function () {
console.log('\n\t ==== Message API listening on 8080! ====')
})

// Dockerfile -> bundle docker image for our app

FROM node:latest

RUN mkdir -p /usr/src/app
WORKDIR /usr/src/app

COPY package.json /usr/src/app/package.json
RUN npm i

COPY . /usr/src/app/

EXPOSE 8080
CMD [ "node", "server.js" ]

Now we can build our docker images from the Dockerfile:

$ docker build -t linoxide/msg-api:v0.0.3 . --no-cache=true

Sending build context to Docker daemon   5.12kB
Step 1/8 : FROM node:latest
 ---> 2d840844f8e7
Step 2/8 : RUN mkdir -p /usr/src/app
 ---> Using cache
 ---> 1c29cda3dcd8
Step 3/8 : WORKDIR /usr/src/app
...

And push the built image to Docker Hub:

$ docker push linoxide/msg-api:v0.0.3

The push refers to a repository [docker.io/linoxide/msg-api]
c4477a160652: Pushed
32c1bac97782: Pushed
3d629e3d2e5a: Pushed
...

v1: digest: sha256:dba64e7ff64561f4af866fbbb657555cad7621688c7f312975943f5baf89efa2 size: 2628

Now we can create a pod of our NodeJS app, the below spec file msg-api-pod.yaml

apiVersion: v1
kind: Pod
metadata:
  name: k8s-msg-api
  labels:
    name: lbl-msg-api
spec:
  containers:
  - name: msg-api
    image: linoxide/msg-api:v0.0.1
    ports:
    - name: msg-api

Apply the manifest:

$ kubectl create -f msg-api-pod.yaml

pod/k8s-msg-api created

Make sure that the pod is running by checking the status:

$ kubectl get pod

NAME          READY     STATUS    RESTARTS   AGE
k8s-msg-api   1/1       Running   0          22s
k8s-mysql     1/1       Running   0          1h

At this level we need to expose the created pod so that can access it from outside. This time I will do it using only command line rather a spec file:

$ kubectl expose pod k8s-msg-api --port=8080 --name=k8s-srv-msg-api --type=NodePort

service/k8s-srv-msg-api exposed

Getting data from mysql database using nodejs api

At this level, I need to point out some important stuff, in order understand all the pieces, let’s first, summarize what we have done until now, so far, we have created a MySQL pod and we have exposed it through a service to make it accessible for other pods, second, we have created a sample nodejs app, we called it a messaging api, so that we can use it to hit the MySQL pod; similarly, to be able to access the messaging API we need to expose it through a service, I hope everything is clear until here!

Now the question is how can we call our messaging api from outside of our cluster mainly minikube? To do so, we need the IP address of our node, as I'm using minikube which create only one node so, the IP address is resolved, is the minikube ip address itself, just run:

$ minikube ip

192.168.99.100

And what about the port? Well good question! let’s describe our messaging api service to check that out:

$ kubectl describe service k8s-srv-msg-api

Name:           k8s-srv-msg-api
Namespace:      default
Labels:         name=lbl-msg-api
Selector:       name=lbl-msg-api
Type:           NodePort
IP:             10.0.0.170
Port:           <unset> 8080/TCP
NodePort:       <unset> 30887/TCP
Endpoints:      172.17.0.6:8080
Session Affinity:   None
No events.

So we have Port which is the port of our messaging API service. NodePort is the port on which the exposed service is available (accessible) i.e., the service is available on NodeIP:NodePort

Let’s try that out:

$ curl 192.168.99.100:30887/ping

hello there! I m up and running!%

$ curl 192.168.99.100:30887/msg-api/all

[]%

Very nice, so far we are able to hit our MySQL database, let’s insert some data into our database using terminal.

$ kubectl exec k8s-mysql -it -- bash

root@k8s-mysql:/# mysql --user=root --password=$MYSQL_ROOT_PASSWORD
mysql: [Warning] Using a password on the command line interface can be insecure.
...

mysql> use k8smysqldb;

Reading table information for completion of table and column names

You can turn off this feature to get a quicker startup with -A

Database changed
mysql> show tables;
+----------------------+
| Tables_in_k8smysqldb |
+----------------------+
| messages             |
+----------------------+
1 row in set (0.01 sec)

mysql> insert into messages(text) values ('this is the first msg!');

Query OK, 1 row affected (0.01 sec)

mysql> insert into messages(text) values ('this is the second msg!');

Query OK, 1 row affected (0.01 sec)

Let’s get this data through our nodejs API using curl:

$ curl 192.168.99.100:30887/msg-api/all

[{"id":1,"text":"this is the first msg!"},{"id":2,"text":"this is the second msg!"}]%

Conclusion

Containerizing the MySQL database and running DBMS on a Kubernetes cluster brings a lot of benefits to DevOps team, such as portability across environments, start/stop and update easier and having better security due to the services are isolated.

Thanks for reading and please leave your suggestion in the below comment section.

1 thought on “How to Deploy MySQL on Kubernetes”... add one

Leave a Comment