sankar2389 / sample-serverless-node

Table of contents

• Sample serverless node
  • Introduction
  • Tech stack
  • Code structures
  • Microservices workflows
  • Database ER diagram
  • Architecture diagram
  • TODOs
  • Steps to run application
  • API endpoints
  • Curl commands to test apis
  • Commands

Sample serverless node

Introduction

This node application is used to demonstrate

• Building microservices on AWS
• Serverless backend
• Communication between services using
  • Direct http calls
  • Message queue (SQS)
  • Near real-time data streaming using AWS Kinesis
• Infrastructure as code (defined in serverless.yaml)
• Integration tests using supertest

Tech stack

• AWS Lambda + with IAM role
• Amazon SQS
• Amazon Kinesis
• AWS Cloudformation
• supertest library for integration test
• AWS DynamoDB
• PostgreSQL
• Docker

Code structures

• At root directory: contains docker-compose.yaml for local development environment. This contains several images
  • postgres database
  • sqs
  • kinesis
  • dynamodb
• Each service is separated into its own folder and contains
  • serverless.yaml: define service resource using Cloudformation syntax
  • tests: unit test for application
  • src/*: contains code for the service

Example for products service

products
├── __tests__ <-- teet folder
│   └── productHandler.test.js
├── prisma
├── src
│   └── handler.js <-- main handler file
├── .env <-- dot env file
├── .env.dev <-- dot env file
├── package.json
└── serverless.yaml <-- define service resources

Microservices workflows

• CRUD product
• Direct HTTP: from products service, when search products, the search term is logged by calling http api of activity_logs service with payload is search term. This is for demonstration purpose only. We shouldn't do this in production.
• SQS workflow: when filter products by color or brand, the filters are sent as a SQS message to a ActivityLogsQueue, the activity_logs service process messages from the queue and process it (saving into database)
• Kinesis workflow: when view a product, the a record is put to kinese ActivityLogsStream, activity_logs service poll records and process them (saving into database)

Database ER diagram

Architecture diagram

TODOs

• Products CRUD
  • Create
  • Update
  • Delete
  • Get
  • List
  • List with custom pagination
• Seed data
  • Products
  • Brands
• Product search
  • By name
• Product filter
  • By color
  • Brand
• Streaming and store data for analytics.
  • Store search products logs - use direct http call
  • Store view product logs - use SQS
  • Stream view products - use Kinesis
• Other
  • Draw architecture diagram
  • Draw ER model
  • Explanation for code structure
  • Steps to run the application
  • List curl commands to test apis

Steps to run application

• global requirements
  • node version v14.14.0 or above
  • Docker
  • npm install -g serverless@latest
• from root folder
  • docker-compose up
• from products folder
  • npm install
  • npm start
  • npm test (optional)
  • if there is a problem starting the service, try running npm run db:reset
• from activity_logs folder
  • npm install
  • serverless dynamodb install
  • serverless dynamodb start
  • npm start
  • npm test (optional)

Demo video: https://www.loom.com/share/c7ae0540a17248cfa0ddcb9fb3d71a42

API endpoints

Entity	Endpoint	Method	Description
Product	http://localhost:3000/dev/products	POST	Create a new product
Product	http://localhost:3000/dev/products/{id}	GET	Get a product
Product	http://localhost:3000/dev/products/{id}	DELETE	Detete a product
Product	http://localhost:3000/dev/products/{id}	PUT	Update a product
Product	http://localhost:3000/dev/products	GET	List products
Product	http://localhost:3000/dev/products?search=productName	GET	Search products
Product	http://localhost:3000/dev/products?page=2&pageSize=10	GET	Custom pagination
Product	http://localhost:3000/dev/products?color=red,blue	GET	Filter by colors
Product	http://localhost:3000/dev/products?brand=dior	GET	Filter by brand
Activity Logs	http://localhost:3000/activityLogs	GET	List activity logs
Activity Logs	http://localhost:3000/activityLogs	POST	Create activity logs

Curl commands to test apis

Located in curl-command.md file

Commands

• reset: npx prisma migrate reset --preview-feature

• migrate: npx prisma migrate dev --preview-feature

• seed: npx prisma db seed --preview-feature

• generate: npx prisma generate

• send a sqs message: aws sqs --endpoint-url http://localhost:9324 send-message --queue-url http://localhost:9324/queue/ActivityLogsQueue --message-body "MyFirstMessage"

• put a kinesis record, data must be base64: aws kinesis --endpoint-url http://localhost:4567 put-record --stream-name ActivityLogsStream --partition-key "view product" --data "TXlGaXJzdE1lc3NhZ2UK"

• sls dynamodb install

• sls dynamodb start