Automate and Configure Your RDS Database With Terraform

Execute post-deployment scripts based on your needs

Photo by Ruchindra Gunasekara on Unsplash

I’ve been using a lot CloudFormation and enjoying the journey so far. But as you all know, Terraform is trendy so I thought giving a shot at the tool and the different available options to automate a deployment may be worth it.

In this article, I will explain how you can deploy an RDS database, and run post deployment script such as create database, tables… I will assume that you already have your aws credentials configured, and terraform installed as well, and you’re also familiar with configuring an S3 bucket as Terraform backend — so I will mainly focus on the modules sections, and go thru the deployment part briefly at the end.

The project will deploy few items

1. One bastion host group which will be located in a public subnet, that will allow you to ssh tunnel to your database + the associated security group
2. One RDS PostGres database and associated security groups to allow only your VPC subnets to access the db
3. One Lambda deployed within your VPC, because you want the lambda to be able to connect to the database in a secure way
4. Some SSM parameters such as username / password for your database by the Lambda first and your application in the future
5. A Lambda invocation trigger after all the previous items are deployed

Project tree overview — the whole project is available on GitHub

Resources creation

This code block snippet (main.tf — folder module/rds-database-lambda) will create for you some of the items I mentioned previously:

• Your bastion security group. Note that you will need to allow your IP line 17 instead of the dummy one — l5
• The ec2 instance which requires as input a keypair (not…