Preparing for Production

Production environments are not like your laptop.

A production-ready Flask app typically needs:

diagram what changes between the dev server and production mermaid

  flowchart TD
  subgraph Development
    A["flask run"] --> B["one process, reloader on"]
    B --> C["debug=True -- interactive debugger"]
  end
  subgraph Production
    D["nginx or the platform's proxy"] --> E["TLS, static files, timeouts"]
    E --> F["gunicorn / waitress, N workers"]
    F --> G["your app object"]
    G --> H["debug OFF, SECRET_KEY from the environment"]
  end
  C -.->|"never ship this"| I["the debugger console runs arbitrary code"]

flask run is a single-process development server with the reloader and debugger attached. In production the app is served by a WSGI server running several workers, usually behind a reverse proxy that terminates TLS and serves static files. The debugger must be off -- it executes arbitrary code by design.

Must-do checklist

• Debug mode disabled
• Secrets stored in environment variables
• A production WSGI server (Gunicorn) instead of flask run
• Proper logging
• HTTPS termination (usually via Nginx or platform)
• Error handling (friendly pages or JSON)
• Database migrations
• Static file handling

The biggest production mistake

Running:

• flask run

on a production server.

Use Gunicorn (or another WSGI server) instead.

Configuration mindset

Your app should be configured by:

• environment variables

Not by editing code per environment.

Health endpoints

Many deployments benefit from:

• /health endpoint returning 200

So monitoring systems can detect downtime.

Observability

Have some plan for:

• request logs
• error logs
• basic metrics

Even a simple setup is better than none.