CDK LAB for RDS

This is a simple Lab environment for quickly building a Database and its peripheral infrastructures.

• Components to be deployed
• Prerequisite: Set environment variables
• Quickstart
• Usage
• References

Components to be deployed

• IAM Stack (cdk-<prefix>-iam-stack)
  • A secret contains the necessary information to login database, including database identifier, username and password
  • A role called EC2SSMInstanceProfile with some policies
    • AmazonSSMManagedInstanceCore managed policy to connect to the EC2 instance used as development environment
    • ReadDBSecretPolicy custom policy to read the secret
    • DescribeDBInstancesPolicy custom policy to list the database's name
• VPC Stack (cdk-<prefix>-vpc-stack)
  • A brand new VPC with 2 Availability Zones to create subnets
  • 2 public subnets in each AZ
  • 2 private subnets in each AZ
  • A security group for EC2 instances
  • A security group for RDS database instances, allow inbound connection to connect database from the instance with EC2's security group
• EC2 Stack (cdk-<prefix>-ec2-stack)
  • A t3.micro free instance with
    • A EC2SSMInstanceProfile role attached
    • Pre-installed database command line client
    • 2 scripts to help users manipulate the database:
      • /srv/install_example_database.sh: Load the example database into RDS database instance
      • /srv/login_database.sh: Login to the database
• RDS Stack (cdk-<prefix>-rds-stack)
  • A RDS for MySQL/MariaDB/PostgreSQL database instance in private subnets

Prerequisite: Set environment variables

Either provide a .env file or set the environment variables before deploy the stacks. The easy way is copying the .env.example file to .env and replace the variables, including:

• REGION: The default region to create the resources (e.g. us-east-2)
• STACKNAME_PREFIX: The prefix string of each stack
• DB_ENGINE: Choose one from MySQL, MariaDB or PostgreSQL
• DB_IDENTIFIER: Database identifier
• DB_USERNAME: Database's default username
• DB_PASSWORD: Database's default password. Please make sure the password meets the complexity requirement of the database engine

Here is a sample .env file:

REGION=us-east-2
STACKNAME_PREFIX=mylab
DB_ENGINE=MySQL
DB_IDENTIFIER=database
DB_USERNAME=username
DB_PASSWORD=password

Quickstart

1. Set environment variables
2. Deploy all stacks at once by executing cdk deploy --all command
3. Login to AWS console, enter EC2 service, check the instance we just created and click "Connect" button, navigate to "Session Manager" tab and click "Connect" button
4. Execute /srv/install_example_database.sh to load the example database into RDS database instance
5. Execute /srv/login_database.sh to login to the database with the command line client

Usage

Deploy stacks:

cdk deploy <stack name>  # Deploy specitic stack
cdk deploy cdk-iam  # e.g: Deploy the IAM stack if no STACKNAME_PREFIX specified
cdk deploy cdk-mylab-iam  # e.g: Deploy the IAM stack if the STACKNAME_PREFIX is assigned as "mylab"
cdk deploy --all # Deploy all stacks
cdk deploy --all --require-approval=never  # Deploy all stacks without asking yes or no

Destroy stacks:

cdk destroy <stack name>  # Destroy specitic stack
cdk destroy cdk-iam  # e.g: Destroy the IAM stack if no STACKNAME_PREFIX specified
cdk destroy cdk-mylab-iam  # e.g: Destroy the IAM stack if the STACKNAME_PREFIX is assigned as "mylab"
cdk destroy --all # Destroy all stacks

All available stacks:

• cdk-<STACKNAME_PREFIX>-iam
• cdk-<STACKNAME_PREFIX>-ec2
• cdk-<STACKNAME_PREFIX>-vpc
• cdk-<STACKNAME_PREFIX>-rds

References

• Create EC2 in an existing VPC with AWS CDK Python
• Create VPC, EC2 ASG, and RDS instance with AWS CDK Python
• AWS CDK (v2) Python Reference
• AWS CDK (v2) Reference Documentation (TypeScript)