A modular, scalable backend system for searching flights, locking seats, creating bookings, processing payments, applying discounts, generating tickets/invoices, and sending notifications.
Built using TypeScript, Node.js, PostgreSQL, and Redis, with clean class-based architecture, secure JWT auth, Redis caching, and containerized using Docker Compose.
This is a complete backend for a flight booking platform (like MakeMyTrip). It includes 9 decoupled microservices handling:
- Authentication
- Flight search (supports 2-hop logic)
- Seat inventory & locking
- Booking system
- Payment gateway (mock)
- Discount validation
- Ticket & Invoice generation
- Notification dispatch (mock)
- Centralized API Gateway
| Layer | Stack |
|---|---|
| Language | TypeScript |
| Runtime | Node.js |
| API Framework | Express.js |
| Database | PostgreSQL |
| Cache | Redis |
| Auth | JWT (JSON Web Token) |
| Containerization | Docker + Docker Compose |
[ Client / Postman ]
│
┌────────────┐
│ API Gateway│ ← http://localhost:8080
└─────┬──────┘
┌────────────┬────────────┬────────────┬────────────┬────────────┐
│ Auth │ Flight │ Seat │ Booking │ Ticket │
│ Service │ Search │ Inventory │ Service │ Service │
└────────────┴─────┬──────┴────┬───────┴──────┬─────┴────────────┘
│ │ │
Payment Discount Notification
Service Service Service
| Service | Description | Port |
|---|---|---|
auth-service |
Signup, login, JWT authentication | 3000 |
flight-search |
Direct & 2-hop flight search | 4000 |
seat-service |
Lock and confirm seats via Redis | 5000 |
booking-service |
Booking flow management | 6000 |
payment-service |
Simulate payments, update status | 7000 |
discount-service |
Validate discount tokens | 8000 |
ticket-service |
Generate ticket + invoice | 9000 |
notification |
Mock SMS/email notifications | 9100 |
api-gateway |
Central entrypoint & routing | 8080 |
- Node.js ≥ 18
- Docker and Docker Compose
.envfiles (see below)
NOTE: I have added .env files in the code so that its easy for anyone testing and refrencing the code. DO NOT DO THIS IN PRODUCTION ENVIRONMENT
git clone <repo-url>
cd flight-booking-system
docker compose up --build
npm run test:allThis starts all 9 services, Postgres DBs, and Redis via Docker Compose.
Run this for running each service individually
npm run test
Run this for running all services at once
npm run test:all
USE POSTMAN COLLECTION FOR TESTING
Each service contains a .env file. Most include:
NOTE: I have added .env files in the code so that its easy for anyone testing and refrencing the code. DO NOT DO THIS IN PRODUCTION ENVIRONMENT
PORT=XXXX
DATABASE_URL=postgres://postgres:postgres@<db-service>:5432/<dbname>
JWT_SECRET=your-secret (for auth-service only)
REDIS_URL=redis://redis:6379 (for flight-search and seat-service)The api-gateway also includes:
AUTH_SERVICE_URL=http://auth-service:3000
FLIGHT_SEARCH_URL=http://flight-search:4000
SEAT_SERVICE_URL=http://seat-service:5000
BOOKING_SERVICE_URL=http://booking-service:6000
PAYMENT_SERVICE_URL=http://payment-service:7000
DISCOUNT_SERVICE_URL=http://discount-service:8000
TICKET_SERVICE_URL=http://ticket-service:9000
NOTIFICATION_SERVICE_URL=http://notification-service:9100Below is a complete end-to-end simulation of a flight booking using Postman or cURL. All APIs are routed via the API Gateway (http://localhost:8080):
-
Signup
POST /api/auth/signup Content-Type: application/json { "name": "John", "email": "[email protected]", "password": "test123" } -
Login
POST /api/auth/login Content-Type: application/json { "email": "[email protected]", "password": "test123" }🔐 Returns a JWT token and
userId. Use the token in all further requests as:Authorization: Bearer <token> -
Search Flights
GET /api/flights/search?source=DEL&destination=BOM&date=2025-08-05 Authorization: Bearer <token> -
Check Available Seats
GET /api/seats/1 Authorization: Bearer <token> -
Lock a Seat
POST /api/seats/lock Authorization: Bearer <token> Content-Type: application/json { "flightId": 1, "userId": 1, "seatNumber": "A1" } -
Create Booking
POST /api/bookings/create Authorization: Bearer <token> Content-Type: application/json { "userId": 1, "flightId": 1, "price": 5000, "seats": 2 } -
Get Booking Details
GET /api/bookings/1/1 Authorization: Bearer <token> -
Initiate Payment
POST /api/payments/pay Authorization: Bearer <token> Content-Type: application/json { "bookingId": 1, "amount": 5000, "userId": 1 } -
Apply Discount Code (Optional)
POST /api/discounts/apply Authorization: Bearer <token> Content-Type: application/json { "bookingId": 1, "code": "FLY100" } -
Generate Ticket
POST /api/tickets/generate
Authorization: Bearer <token>
Content-Type: application/json
{
"booking_id": 1
}
- Get Ticket
GET /api/tickets/1
Authorization: Bearer <token>
- Send Notification
POST /api/notifications/send
Authorization: Bearer <token>
Content-Type: application/json
{
"userId": 1,
"message": "ticket booked"
}
flight-booking-system/
├── auth-service/
├── flight-search-service/
├── seat-service/
├── booking-service/
├── payment-service/
├── discount-service/
├── ticket-service/
├── notification-service/
├── api-gateway/
├── docker-compose.yml
└── README.md
-
Flight search results are cached in Redis
-
Cache is invalidated if:
- Flight has already departed
- Flight is in the cancelled list
- Flight is fully booked
-
Every service (except signup/login) checks for valid JWT
-
Booking flow is strictly linear:
- Search → Lock → Booking → Pay → Ticket
- Discount is optional, all else is mandatory
-
Redis ensures race-free seat locking
- Reverse proxy: NGINX or Traefik in front of API Gateway
- Use HTTPS (TLS termination at gateway)
- Run Redis/Postgres as managed services (e.g., GCP Cloud SQL, Memorystore)
- Add retry logic and exponential backoff for network calls
- Add monitoring (Prometheus + Grafana)
- Structured logging (e.g., Winston + FluentBit)