Running a simple Django website in Docker

Docker is a near perfect solution for having project portability across a wide variety of host platforms. I use docker to run my websites on both development computers and servers. For Django my go to Dockerfile looks a little something like this.

# django
#
# I use this to run most of my django projects in a single container in
# production. If you wish to seriously reduce the size of this image and you
# don't need it you can remove the chromium line. I use it for screenshots and
# pdf creation.
#
# Make sure to change the below ENV variables to fit your needs and the gunicorn
# path to your applications asgi file.
#
# You can run this with:
#   docker build --tag overshard/django:latest .
#   docker run -d -p 80:8000 -e DJANGO_SETTINGS_MODULE=project.settings.production \
#     -v /srv/data:/data django:latest

FROM alpine:3.16

RUN apk add --update --no-cache \
      sqlite \
      python3 py3-pip \
      nodejs yarn \
      chromium libstdc++ nss harfbuzz freetype font-noto font-noto-extra font-noto-emoji && \
    pip install pipenv

COPY Pipfile Pipfile.lock package.json yarn.lock /app/

RUN yarn install && pipenv install --system

COPY . .

RUN yarn webpack:production && \
    rm -rf node_modules && \
    python3 manage.py collectstatic --noinput

RUN addgroup -S -g 1000 app && \
    adduser -S -h /app -s /sbin/nologin -u 1000 -G app app && \
    chown -R app:app /app

USER app:app

WORKDIR /app

VOLUME /data

EXPOSE 8000

ENV DJANGO_SETTINGS_MODULE=project.settings.production

CMD ["gunicorn", "project.asgi:application", "-k", "uvicorn.workers.UvicornWorker", "-w", "4", "-b", ":8000", "--access-logfile", "-", "--error-logfile", "-"]

A few notes on my choices:

This assumes you are just using an sqlite database but this can easily scale into using PostgreSQL in coordination with docker-compose
I use webpack to build all of my static files on all of my sites hence why nodejs and yarn are included
Chromium is used in most of my projects for generating PDFs and screenshots, I've found it the most reliable and consistent way of handling that functionality
You'll need both gunicorn and uvicorn installed

You could remove Chromium and save ~400MB of space on a roughly ~450MB image if you have no use for it. It is by far the largest dependency here. I also often use docker-compose in conjunction with this Dockerfile.

This can be used directly in production pretty well however I do put most of my websites behind Caddy using a reverse proxy. If you'd like to see my most up-to-date alpine-docker files you can check them out on my overshard/dockerfiles GitHub project.