Terraform Deployment of MariaDB Galera

About

This is an example of a Terraform deployment for MariaDB Galera with MaxScale. It is intended as a way to get a self-hosted Galera cluster running quickly for troubleshooting, demos, or learning. It should not be used in production.

How To Use

1. Setup

Configure AWS Credentials by creating a file named aws_creds in the terraform directory with the information from AWS
Configure SSH Keys by creating a key pair in AWS and downloading the private key. Then edit terraform/vars.tf to set the ssh_key_path and aws_key_name
Find your IP address by going to https://ifconfig.me and edit terraform/vars.tf to set the ssh-location variable

Finally, review all settings terraform/vars.tf to ensure they are correct.

2. Terraform

Run terraform plan -out a.plan to validate Terraform and create the plan. Then run terraform apply a.plan to apply it.

Note: If you run into errors, read them carefully twice and then do what they say.

3. Configure Machines

First, run deploy/setup.sh. This will install the SSH config on the jumpbox, copy the deploy directory, and connect via SSH with a SOCKS proxy. You will be dropped into the shell of the jumpbox.

From here, you can connect to the machines in the cluster via SSH to install Galera Manually:

ssh galera[0-N]
ssh maxscale[0-N]

Or, you can run this from the jumpbox to setup Galera and Maxscale:

cd ~/deploy
./maxscale_setup.sh