Cognito Workshop REST API

This is another Express + TypeScript + DDD (Domain Driven Design patterns) + IoC/DI (Inversion of control and Dependency injection) + API REST boilerplate.

Author: Gonzalo Plaza [email protected]

Installation and Configuration

yarn install

Generate .env file

cp .env.dist .env

You have to configure local environment variables with your own parameters inside .env file. These are the default values:

PORT=3000
APP_NAME="cognito-workshop"
APP_LOG_LEVEL="debug"
AWS_ACCESS_KEY_ID="your_aws_access_key_id"
AWS_SECRET_ACCESS_KEY="your_aws_secret_access_key"
COGNITO_USER_POOL="your_cognito_user_pool"
COGNITO_CLIENT_ID="your_cognito_client_id"
COGNITO_REGION="your-aws-region-x"

Development with nodemon (Runs on 3000 port)

yarn dev

Run tests and generate coverage report

yarn test #In progress

Build/Compile JS (to /dist folder)

yarn build

Authentication with AWS Cognito

AWS Cognito is implemented as default authentication service. Please, check documentation on how to configure it for more details: https://aws.amazon.com/cognito/getting-started/

---

Docker Compose (for development)

This project is ready to work with docker-compose 3.8 to initialize the needed stack during development process. To start working run the following commands:

docker-compose build
docker-compose up

To shut down local stack:

docker-compose down

Working docker compose environment (Port 8080)

curl http://localhost:8080/

Example JSON response (/):

{
  "success": true,
  "message": "Hello Cognito Workshop",
  "timestamp": 1648208765884
}

Build Docker image for production

You can build an optimized Docker production-ready image with the standard command:

docker build --no-cache -t cognito-workshop .

And then run the container passing environment variables within the initialization:

docker run --rm -it -p 3000:3000 \
  -e NODE_ENV=production \
  -e AWS_ACCESS_KEY_ID="your_aws_access_key_id" \
  -e AWS_SECRET_ACCESS_KEY="your_aws_secret_access_key" \
  -e COGNITO_USER_POOL="your_cognito_user_pool" \
  -e COGNITO_CLIENT_ID="your_cognito_client_id" \
  -e COGNITO_REGION="your-aws-region-x" \
  -e APP_LOG_LEVEL="info" \
  --name cognito-workshop cognito-workshop

Deploy Stack to AWS

In order to create/deploy the current stack to AWS using CloudFormation you'll need to generate your own .aws-envs file that you can copy from the provided template:

cp .aws-envs.example .aws-envs

Note: This process will need a proper role created into your AWS Account with enough permissions to create all needed resources, including IAM roles. You will need also to specify a working AWS VPC and 2 subnets. To execute it:

yarn deploy-stack # This is valid for stack create or update.

And to remove stack from cloudformation:

yarn remove-stack

Todo

• Decouple dependency on aws key and secret variables
• Link AWS LoadBalancer to Route53 domain and SSL certificate
• Add tests and coverage

Help and references

• https://nathanpeck.com/speeding-up-amazon-ecs-container-deployments
• https://docs.aws.amazon.com/es_es/AWSCloudFormation/latest/UserGuide/AWS_ECS.html
• https://docs.aws.amazon.com/elasticloadbalancing/latest/application/target-group-health-checks.html
• https://docs.aws.amazon.com/es_es/AWSCloudFormation/latest/UserGuide/AWS_ElasticLoadBalancingV2.html
• https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-elasticloadbalancingv2-targetgroup.html
• https://github.com/aaronwht/aws-ecs-typescript