Implement WordPress load balancing on multiple Availability Zones with one MySQL RDS instance on AWS

This is a diagram that I have used to deploy this lab.

There are a couple of main steps that I have used in the lab.

  • Create both private subnets on both AZs
  • Create a WordPress instance on the first AZ.
  • Create a new MySQL instance.
  • Create an AMI for the WordPress instance.
  • Launches a new WordPress instance 2 on the second AZ by using your customized AMI.
  • Create an Application Load Balancer for WordPress on multiple AZs
  • Set up a checkhealth.html file to test the Application Load Balancing

Create a new VPC with CIDR

Create both private subnet, on us-east-1a and on us-east-1b.

Create a new Internet Gateway and attach this to your VPC.

Add a static route on your Internet gateway.

Launches a new Linux instance to run WordPress on AZ1.

Copy it into the User data setting.

yum update -y
# Install Apache web service
yum install httpd -y
# Download WordPress 
tar -zxf latest.tar.gz
# Install php7.4
amazon-linux-extras install php7.4 -y

On Security Group, allow SSH, HTTP, HTTPS, and MySQL/Audora from

SSH to Linux instance.

Check httpd, php is installed on the machine.

rpm -qa | grep httpd
rpm -qa | grep php
sudo yum install php -y
sudo systemctl start httpd
sudo systemctl enable httpd
netstat -antp

Copy all files on WordPress directory to /var/www/html

cd /wordpress
sudo cp -r * /var/www/html

Go to Amazon RDS, create subnet groups on Amazon RDS.

Create a new Database instance on AWS.

Choose the Free tier.

Enter wordpress on “DB instance identifier”, “master user name and password”

DB instance class is d2.t2.micro.

Public access is No.

Choose the Availability zone as the following screenshot.

Enter “wordpress” on the initial database name.

Backup retiontion period: 0, then click “Create database”.

Wait a couple of minutes to completely create the database instance.

Access WordPress site via the public IP address of WP.

Database, username, password is wordpress.

Database Host is the endpoint address of the RDS database on AWS on the previous screenshot.

Click Submit.

Copy entire content, open SSH shell on Linux instance. Create a new wp-config.php under /var/www/html.

sudo nano wp-config.php 

Back to WP web interface set up, click “Run the installation”.

Click “Install WordPress”.

Log in WP.

Now, create a new AMI image for this WP. Right-click the WP instance, on Actions – Image and templates – Create image.

Right-click AMI. Click Actions – Launch an instance from AMI.

Go to the load balancer, and create a new application load balancer.

Create a new WordPress ALB SG. Allow HTTP from on this Security Group.

Create a target group.

Select “Instances”.

Enter “AP-ALB” on the target group name and checkhealth.html for the health check WP instance.

Change the settings as the screenshot below. Click Next.

Select both instance IDs and click “Include as pending below”.

Create a target group.

Back to the Application Load Balancer setup, choose the “WP-ALB” on the target group.

Create a load balancer.

Wait a few minutes to see “Health status” is Healthy.

SSH to Linux instance on WordPress server 2.

Change the checkhealth.html to make the difference between WP1 and WP2.

On WP1.

<h1> This is health check from the WordPress Server 1 </h1>

On WP server 2.

Start the httpd daemon.

sudo systemctl start httpd 

Do the same on WP1 to make sure the httpd daemon is running after making the AMI template.

Access WP health check on WP server 1.

Make sure both WP servers have Healthy status on WP-ALB.

Copy the Amazon ALB link into your web browser.

Refresh, it can be seen that the web traffic is loaded balancing on the WP server 2.

Check the connection from WP instances to the Amazon RDS database.

I will set up Amazon Route 53, Amazon CloudFront with a real domain name such as, and load balancing via HTTPS (WordPress SSL certificate will be issued by Amazon), not HTTP. Also. I will configure a strict Security Group policy to strengthen security from WordPress to the Amazon RSD database in the next labs.