How I Deployed My First Production App on AWS EC2 — Every Mistake I Made

I am a third-year computer science student at IIIT Sonepat. Recently, I deployed my chat application, FastChat, on a live AWS EC2 server with HTTPS support, a domain name, and a proper Nginx reverse proxy. This blog is going to be a description of exactly what I did, how it all fits together, and what I did wrong so you don’t have to. What I Built FastChat is a REST + WebSocket chat API built with: App: Node.js, Express.js, Socket.io, MongoDB, PostgreSQL, Redis, AWS S3 (avatar storage), Jest + Supertest (testing) Infrastructure: Docker, Docker Compose, Nginx, AWS EC2, Let's Encrypt (SSL), DuckDNS (free domain) The live API is running at https://fastchat.duckdns.org Note: This URL may not always be live as I shut down the EC2 instance when not in use to avoid AWS charges. If you want to see the code instead, check out the GitHub repo at codephoenix86. Architecture Overview Before jumping into steps, here's how everything connects: The key security decision: only ports 22 (SSH), 80 (HTTP