Your app works on your machine, and nobody else can open it. Production is the leap to a real server the world can reach. This chapter gets you there. It also hands you the one mindset that makes every later step in this part sane: the whole setup is rebuildable from files, never clicked together by hand.
13.1.1Production is rebuildable, not clicked
Production is not a machine you configure by hand and hope never dies. Anything you set by clicking around in a dashboard is gone the day that server does, and you are rebuilding it from memory.
Real production is described in files. You run those files and the whole setup comes back, identical, on a fresh machine. That single idea is what the rest of this part builds, so treat every by-hand click as a note you still have to turn into a file.
13.1.2Pick a host you can afford to keep
The honest default is cheap, own-the-metal hosting, not the big-cloud maze. A small server on a provider like Hetzner costs a fraction of the same box on a hyperscaler. You also stop paying a tax for a hundred managed services you will never touch.
Rule of thumb: skip the AWS managed maze unless you have a specific reason to be in it. If you must be on AWS, a plain server (an EC2 box) beats wiring ten services together to run one app.
13.1.3Deploy by hand once, then never again
The first time, do it by hand. Stand up a server, point a domain (the web address people type) at it, turn on HTTPS, then get the app running and open it from another device. Doing it once shows you every moving part, and automating a thing you have never done by hand only hides where it breaks.
Then you never do it by hand again. Every step you just took becomes a file in the chapters that follow, and from then on a deploy is one command, not a lost afternoon.
This prompt turns your first deploy into a written plan:
Do this now: paste the prompt, get your manual deploy plan, and stand the app up on a real host once, by hand.
