Docker for beginners: what it's for

The problem it solves

The classic situation: the project runs for you and not for your colleague. The cause is different versions of the language, libraries or operating system. Diagnosing it takes a day.

A container solves this by packaging the environment along with the code — the right language version, the libraries, the settings. It then runs identically for anyone and on any server.

"It works on my machine" stops being an argument: if it works in the container, it works everywhere.

What a container is, plainly

Not a virtual machine. A virtual machine is a whole computer inside a computer: heavy and slow.

A container is an isolated box that shares your system's kernel but has its own file system and processes. It starts in seconds and weighs little.

Two terms to keep apart:

• An image — the recipe and the built artefact. An immutable snapshot with your application and its environment.
• A container — a running instance of an image. One image can produce many.

First commands

Pull a ready image and run it:

docker run --rm hello-world

Run a database without installing it on your system:

docker run --name pg -e POSTGRES_PASSWORD=secret -p 5432:5432 -d postgres

Reading that: a container name, an environment variable with the password, a port exposed outward, and detached execution.

Incidentally that's the easiest way to get a database for study: nothing to install and nothing to clean up afterwards.

See what's running:

docker ps

Stop and remove:

docker stop pg
docker rm pg

A minimal build file

A file of instructions for building your project's image. For a Python application, roughly:

FROM python:3.12-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY . .
CMD ["python", "main.py"]

Line by line: start from a ready image with the language, set a working directory, copy the dependency list, install them, copy the code, declare the start command.

The order matters: dependencies are copied and installed before the code, so that changing the code doesn't reinstall them. That speeds up rebuilds noticeably.

Build and run:

docker build -t myapp .
docker run --rm -p 8000:8000 myapp

What not to put in an image

• Passwords and keys. Pass those as environment variables.
• Your machine's virtual environment.
• The version history folder and other junk — there's an exclusion file for that.

Why a beginner should care

First, it appears in postings increasingly often, and being able to package your own application sets you apart.

Second, it's convenient for yourself: databases and helper services for study come up with one command and go away just as easily, without cluttering your system.

Third, a containerised project is easy to deploy anywhere — you aren't tied to one platform.