Generic Jar Infra CDK Java project

Project using CDK development with Java.
Automation AWS Infrastructure As a Code (IaaC), for the particular scenario: we want to deploy a Spring Boot jar, supplying a previously filled database and authorization to access S3 buckets.

Stacks

• VPS: integration of the AWS objects;
• EC2: hosting a Java 17 executable jar using Spring Boot on the port 5000;
  • Amazon Linux 2023;
  • Java 17 (Amazon Corretto);
  • AWS-CLI;
• RDS: in which Spring Boot jar's got to be connected;
  • Postgresql 15;
  • Until the moment this document was writen no further version available in Amazon Linux package repository :-(
• IAM Role: allowing the project in EC2 the following AWS communications in the cloud:
  • RDS;
  • S3;
  • Secrets Manager;
• Security Group: allowing the folowing connections:
  • Inbound:
    • SSH;
    • releasing port 5000 for access to the site;
  • Outbound:
    • Internet updates, and integration with external webservices;

Requirements

All this objects will be created from scratch.
The only exisiting AWS objects required are:

• key pair;
• S3 bucket containing:
  • .jar file to be deployed;
  • .sql script file with data.

Custom commands

Environment initialization:

cdk bootstrap

Deploy this stack to your default AWS account/region:

cdk deploy --parameters RdsStack:project={project-name} RdsStack:dbPassword={database-password} --parameters EC2Stack:jars3arn={jars3arn} --parameters EC2Stack:sqls3arn={sqls3arn} --parameters EC2Stack:keypair={keypair} --all

• {project-name}: put the project name and database name are going to be based on this keyword;
• {database-password}: put database password for the user postgres;
• {jars3arn}: Bucket S3 ARN containing the executable jar;
• {sqls3arn}: Bucket S3 ARN containing the SQL script to generate and populate data in database;
• {keypair}: Existing key pair name without .pem extension;
  

Uninstall all together:

cdk destroy --all

Example:

cdk deploy --parameters RdsStack:project=marques --parameters RdsStack:dbPassword=postgres --parameters EC2Stack:jars3arn=s3://marquesjars/marques_create_db.jar --parameters EC2Stack:sqls3arn=s3://marquesjars/import.sql --parameters EC2Stack:keypair=sshlinux --all

will generate:

• RDS named marques-db, with inner database named marquesdb, for user postgres and password postgres;
• EC2 named EC2Linux, hosting an executable jar named project.jar;

JAR Project Prerequisites

The EC2 stack will generate the environment variables SPRING_DATASOURCE_URL and SPRING_DATASOURCE_PASSWORD, so the Spring Boot project can connect to the database.

#application.properties

spring.datasource.url=${SPRING_DATASOURCE_URL}
spring.datasource.username=postgres
spring.datasource.password=${SPRING_DATASOURCE_PASSWORD}

Using passwords in environment variables is not recomended; it'd rather using some other tool like AWS Secrets Manager SDK inside the project. This case was used for development and testing environments. But AWS Secrets Manager SDK can be used to overwrite the password with a secure password in production deployments.

Original SDK reference

It is a Maven based project, so you can open this project with any Maven compatible Java IDE to build and run tests.

• cdk.json file tells the CDK Toolkit how to execute your app.
• mvn compile -q compile and first check
• mvn package compile and run tests
• cdk ls list all stacks in the app
• cdk synth emits the synthesized CloudFormation template
• cdk deploy deploy this stack to your default AWS account/region
• cdk diff compare deployed stack with current state
• cdk docs open CDK documentation