Environment

Environment variables are used to pass information into the container at run time.

This allows you to build an image once but run it with different configurations (API Keys, passwords, etc) or for different environments (development vs production).

Typically, environment variables flow like this:

.env --> docker-compose.yml --> container --> application

.env

Environment variables are typically placed in a .env file at the base of the repo. Docker automatically looks for .env when running the compose configuration. Here are the contents of this project's .env:

#mkdocs
SITE_NAME="Plebnet Compose (Oct 21, 2023)"

#local db
DB_HOST_LOCAL=db_local
DB_HOST_LOCAL_PROD=db_local_prod
DB_USER_LOCAL=postgres_user
DB_PASS_LOCAL=mypass
DB_PORT_LOCAL=5432
DB_NAME_LOCAL=postgres

Warning

.env contains sensitive information, you should NEVER add a .env file to git control. .env files are often included in .gitignore to prevent accidental upload.

docker-compose.yml

You can pass environment variables to your services using curly braces ${MY_ENV_VAR}, like this:

# docker-compose.yml
    ...
    environment:
      SITE_NAME: ${SITE_NAME}

container

Once provisioned by the compose file, the container's environment variables have been set. You can log in and verify these values for the docs service:

docker compose run docs bash

(base) root@4abd4921d273:/code# echo $SITE_NAME
Plebnet Compose (Oct 21 '03)

application

Your application can (finally) access these values. How this is done depends on the language your app is written in. For example, in python it would be

import os

print(os.environ['SITE_NAME'])

For this project, SITE_NAME is picked up by this secion of mkdocs.yml (in the base of this repo):

# Project information
site_name: !ENV [SITE_NAME, "Plebnet Compose"] # if SITE_NAME is undefined, use Plebnet Compose

Note

Your app needs to handle cases where an environment variable is missing