Skip to content
Distr
Star
Book DemoStart free trialLogin

Docker Environment Variables

Distr allows users to define environment variable templates for Docker applications. Users can configure and customize environment variables before deploying an application.

This guide builds on top of the create a Docker application guide and explains environment variable templates in more detail.

What is an Environment Variable Template?

Section titled “What is an Environment Variable Template?”

Environment variables are used to pass deployment-specific values to your application. With Distr, you can define a template for these variables, allowing users to customize them before deployment and ensuring all necessary values are set.

This setup consists of three parts:

  1. A parameterized Docker Compose file that references dynamic variables.
  2. A template file that provides default values for those variables, allowing end users to modify them in Distr before deployment.
  3. The actual environment configured during the deployment of the application.

Here’s an example of how it works:

The Docker Compose file will be the same for all deployments that use that specific version of the application. The environment file can be customized for each deployment.

services:
  web:
    image: '${WEB_IMAGE:?error}'
    container_name: '${WEB_NAME:?error}'
    ports:
      - '${WEB_PORT:?error}:80'
    environment:
      - APP_ENV=${APP_ENV:?error}

You can execute this example with the following command:

Terminal window
docker compose up

You can also create different environments using the --env-file flag:

Terminal window
docker compose --env-file .prod.env up

Referencing environment variables in multiple services

Section titled “Referencing environment variables in multiple services”

You need to pass the environment variables to the services in the docker-compose.yaml file.

If multiple services need the same environment variables, you can define them in x- sections of the docker-compose.yaml file.

x-shared-env: &shared-env
  SHARED_ENV_PRODUCT: ${SHARED_ENV_PRODUCT:?error}
  SHARED_ENV_DOCS_URL: ${SHARED_ENV_DOCS_URL:?error}


# Our partly shared env needs to extend the shared env
x-partly-shared-env: &partly-shared-env
  <<: *shared-env
  PARTLY_SHARED_ENV: ${PARTLY_SHARED_ENV:?error}


services:
  one:
    image: 'stefanprodan/podinfo'
    ports:
      - '3001:9898'
    environment:
      <<: *partly-shared-env
      SERVER_NAME: ${SERVER_NAME_ONE:?error}


  two:
    image: 'stefanprodan/podinfo'
    ports:
      - '3002:9898'
    environment:
      <<: *partly-shared-env
      SERVER_NAME: ${SERVER_NAME_TWO:?error}


  three:
    image: 'stefanprodan/podinfo'
    ports:
      - '3003:9898'
    environment:
      <<: *shared-env
      SERVER_NAME: ${SERVER_NAME_THREE:?error}

You can execute this example with the following command:

Terminal window
docker compose up

When running the Docker Compose file you will be able to see the following output for the webservers /env endpoint:

http://localhost:3001/env
[
  "HOME=/home/app",
  "HOSTNAME=7278f34f08d1",
  "PARTLY_SHARED_ENV=ONE_AND_TWO",
  "PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin",
  "SERVER_NAME=ONE",
  "SHARED_ENV_DOCS_URL=https://distr.sh/docs",
  "SHARED_ENV_PRODUCT=Distr"
]

Injecting the whole environment file into a service

Section titled “Injecting the whole environment file into a service”

When applying a deployment, the Distr agent writes the configured environment as a .env file next to the docker-compose.yaml. This means you can use the standard Docker Compose env_file directive to inject the entire environment file into a service, instead of listing every variable individually under environment:

services:
  webserver:
    image: nginx
    env_file:
      - .env