Default environment variables

How to create and manage default environment variables for your Release environments

Why to use environment variables

Environment variables hold the environment-specific information that an application needs at runtime.

For instance, a web application deployed in a production environment has to know how to access the production database. The same application running in a staging environment would have to know about a staging database instead.

By storing configuration as environment variables instead of in code, you'll make your applications portable, secure, and easy to deploy on Release environments.

How to create default environment variables for your environments

When you create a new application, Release generates a default environment variables file, based on your code.

For example, if your repository has a docker-compose.yml file, Release will populate the default environment variables for your new application based on the environment variables in that file.

If your repository does not have a docker-compose.yml file, you can create your Release application's default environment variables manually, by editing your application's settings.

How to edit an application's default environment variables

To edit the default environment variables for an application, visit the application's App Settings page, then click the Edit button in the “Default Environment Variables” section.

Edit the default environment variables

This opens a YAML editor, which you can use to edit your default environment variables file.

Default environment variables file structure

A Release application's default environment variables file is formatted in YAML, and consists of three sections:

Section	Type	Required	Description
mapping	Hash	false	Maps Release variables to your app's environment variables
defaults	Array	true	Default environment variables for your app
services	Hash	false	Service-specific environment variables

Here's an example of a default environment variables file:

mapping:
  COMMIT_HASH: RELEASE_COMMIT_SHA
  AWS_REGION: RELEASE_CLUSTER_REGION
defaults:
  - key: DEFAULT_LANG
    value: en
  - key: DB_HOSTNAME
    value: db
services:
  admin:
  - key: DB_USERNAME
    value: admin_rw
  - key: DB_PASSWORD
    secret: true
  frontend:
  - key: DB_USERNAME
    value: user_ro
  - key: DB_PASSWORD
    secret: true

Environment variables consist of a key, a value, and an optional secret flag. Mark an environment variable as secret using secret=true. Secrets are encrypted at rest in a vault and hidden in the UI. You may also import secret environment variables from external secrets managers. See the instructions here for more on secrets.

Individual environment variable schema

Field	Type	Required	Description
key	String	True	The environment variable's name. It is convention to use uppercase names for environment variables. Example: DB_USERNAME
value	String or Null	True	The value for this environment variable. Required if secret is false. If secret is true, this field is optional, and Release defaults to the previously saved value if this field is omitted.
secret	Boolean	False	Value is secret and should be encrypted. Value will be hidden from Release UI. Defaults to false.

Here's an example of three environment variables:

defaults:
  - key: DEFAULT_LANG
    value: en
  - key: DB_HOSTNAME
    value: db
  - key: DB_PASSWORD
    value:
    secret: true

How Release uses an application's default environment variables file

When you create a new environment, Release uses the application's default environment variables file to generate an environment-specific environment variables file for your new environment.

Think of an application's default environment variables file as the blueprint for each new environment's environment variables.

Release's environment variables

Apart from the environment variables lifted from your docker-compose.yml, and the variables you created yourself, Release provides additional environment variables.

You can map Release's environment variables to your app's environment variable names so that you do not need to change your application's code when deploying on Release or elsewhere.

Environment variables provided by Release have names that start with RELEASE_.

Here's an overview of the variables provided by Release:

RELEASE_ACCOUNT_ID

Release account ID: Each Release account has a unique numerical identifier.

RELEASE_APP_NAME

Release application name: Each Release application has a unique string identifier.

RELEASE_BRANCH_NAME

Git branch: The name of the git branch from your repository deployed to this environment.

RELEASE_BUILD_ID

Release build ID: Each Release build has a unique numerical identifier.

RELEASE_CLOUD_PROVIDER

Cloud provider: The cloud provider that hosts this environment. For example, aws or gcp.

RELEASE_CLUSTER_REGION

Cloud provider region: The cloud provider region for this environment. For example, us-west-2 or us-east-1.

RELEASE_COMMIT_SHA

Git hash: The SHA hash of the git commit deployed in this environment. This refers to a commit in your repository.

RELEASE_COMMIT_SHORT

Short git hash: The short hash of the git commit deployed in this environment. This refers to a commit in your repository.

RELEASE_CONTEXT

Release context: Refers to the Kubernetes context for this deployment.

RELEASE_DOMAIN

Release domain: If you do not have a custom domain set up on your Release account, this will be rls.sh.

RELEASE_ENV_ID

Release environment ID: Each Release environment has a unique string identifier.

RELEASE_RANDOMNESS

Release random string: This short random string is used to create unique links.

Release's service-specific environment variables

Release also provides environment variables specific to each service in your environment.

When creating these variables, Release converts the service name to uppercase and prepends this to each service-specific environment variable's name.

For example, if you have two services, admin and frontend, Release provides environment variables as follows:

ADMIN_COMMIT_SHA

Short git hash: The SHA hash of the git commit deployed for this service. This refers to a commit in your repository.

ADMIN_COMMIT_SHORT

Short git hash: The short hash of the git commit deployed for this service. This refers to a commit in your repository.

ADMIN_INGRESS_HOST

Ingress hostname: The hostname of the instance/load balancer for this service, accessible to the internet.

ADMIN_INGRESS_URL

Ingress URL: The public URL for this service.

FRONTEND_INGRESS_URL, FRONTEND_INGRESS_HOST etc.

The equivalent environment variables are provided for the frontend service in this example.

Mapping environment variables in the mapping section

You can use the mapping section to map existing environment variables to new environment variables.

Mapping Release's variables

By mapping Release's environment variables to your app's environment variables, you do not need to reference Release-specific environment variable names in your code.

In the example below, your code expects an environment variable called AWS_REGION. Mapping allows you to pass RELEASE_CLUSTER_REGION to your application, adding its value to a new environment variable called AWS_REGION.

mapping:
  COMMIT_HASH: RELEASE_COMMIT_SHA
  AWS_REGION: RELEASE_CLUSTER_REGION
  REACT_APP_ADMIN_BASE_URL: ADMIN_INGRESS_URL
  REACT_APP_FRONTEND_BASE_URL: FRONTEND_INGRESS_URL

Mapping your app's environment variables

The mapping section also allows you to reference your app's environment variables. In the example below, REACT_APP_ADMIN_PATH gets its value from DEFAULT_ADMIN_PATH:

defaults:
  - key: DEFAULT_ADMIN_PATH
    value: /admin/
mapping:
  REACT_APP_ADMIN_BASE_URL: ADMIN_INGRESS_URL
  REACT_APP_ADMIN_PATH: DEFAULT_ADMIN_PATH

Combining existing variables and strings to create new variables

Release's mapping section allows you to create new environment variables through string interpolation.

In the example below, ADMIN_FULL_URL is created by concatenating ADMIN_INGRESS_URL and DEFAULT_ADMIN_PATH:

defaults:
  - key: DEFAULT_ADMIN_PATH
    value: /admin/
mapping:
  ADMIN_FULL_URL: "${ADMIN_INGRESS_URL}${DEFAULT_ADMIN_PATH}"

You can also provide default values if an environment variable is unset or empty.

In the example below, if MAYBE_EMPTY_BASE_URL does not exist, or is empty, the default https://example.com will take its place.

mapping:
  ADMIN_FULL_URL: "${MAYBE_EMPTY_BASE_URL:-'https://example.com'}/admin/"

The syntax to indicate a default is based on parameter expansion in bash: ${parameter:-defaultvalue}, but unlike in bash, defaultvalue is interpreted as a string. defaultvalue can't reference an environment variable.

String interpolation and parameter expansion is limited to the mapping section of your environment variables YAML.

The defaults section

The defaults section contains environment variables that are available to all services in an environment.

Here's an example of the defaults section:

defaults:
  - key: DEFAULT_LANG
    value: en
  - key: DB_HOSTNAME
    value: db

The services section

Use the services section to define environment variables for specific services only. Each service in this section has one or more environment variables.

Here's an example of service-specific environment variables:

services:
  admin:
  - key: DB_USERNAME
    value: admin_rw
  - key: DB_PASSWORD
    secret: true
  frontend:
  - key: DB_USERNAME
    value: user_ro
  - key: DB_PASSWORD
    secret: true