Containers and deploying

Engineering Toolkit, lesson 11

Package an API with Docker, keep secrets out, and read its logs.

"HTTP APIs with FastAPI" for the API being packaged, and "Pipes, variables and scripts" for environment variables.

Part 1. Why a container

The runs API works on your laptop. To run it on a server, the server needs the right Python, the right packages, the files, the start command, and none of the leftovers on your machine that you've forgotten it depends on. A container image packs all of that into one thing that runs the same everywhere Docker does.

Example 1.1

From a folder to a running container

Dockerfile+your files
A Dockerfile is a recipe: start from a base image, copy files in, install packages, name the start command.
docker build→image
Building runs the recipe and saves the result: a read-only filesystem plus the command to start.
docker run→container
Running an image starts a container: a process that sees only the image's files, with its own network address.
the same image→laptopservera teammate's machine
Push the image to a registry, and any machine with Docker runs exactly the same thing.

Your turn

The image starts FROM python:3.12-slim. Your laptop has Python 3.14. Which Python runs inside the container?

A container sees only its image's files, and the image brought its own Python 3.12. That's the point: the server and your laptop run the same version whatever each machine has installed. The container does share the machine's operating system kernel, which is why it starts much faster than a virtual machine.

Show the answer

A. A container sees only its image's files, and the image brought its own Python 3.12. That's the point: the server and your laptop run the same version whatever each machine has installed. The container does share the machine's operating system kernel, which is why it starts much faster than a virtual machine.

Image

A packaged, read-only filesystem (an operating system's basic files, Python, your packages and code) plus the command to start. Built from a Dockerfile, stored in a registry.

Container

A running instance of an image: a process that sees only the image's files and gets its own network. Delete it and anything it wrote is gone; the image is unchanged.

Part 2. A Dockerfile, line by line

Keep going with Pro

You've done the first part. Pro opens the rest of this lesson, and every other one.

Still to come

A Dockerfile, line by line
Settings and secrets come from outside
Deploying, and reading the logs

See Pro plans I have Pro: sign in

Next lessonCapstone: a script becomes a deployed APIA word counter, tested, packaged, containerised and deployed. Start lesson 12

Engineering Toolkit overviewSpotted a mistake? Tell us.

11 · Containers and deploying