For the complete documentation index, see llms.txt. This page is also available as Markdown.

Building Docker images

You can build Docker images for your Helm charts and define any number of Docker builds for your application. However, Dockerfiles must be located in the application's source control repository and you cannot specify Docker builds from remote source control repositories.

Here's an example of Docker builds for custom Nginx and backend containers:

builds:
- name: nginx
  dockerfile: Dockerfile
  context: nginx
  args:
  - MAX_UPLOAD_SIZE=60
- name: backend
  dockerfile: Dockerfile-backend
  context: app

Build environment variables

Each container built by Release will generate environment variables that can be used in your Helm charts. By default, Release adds environment variables for the Docker registry and path to the image built.

- key: RELEASE_REGISTRY_ENDPOINT
  value: 123456789000.dkr.ecr.us-east-1.amazonaws.com
  secret: false
- key: RELEASE_REGISTRY_PATH
  value: 123456789000.dkr.ecr.us-east-1.amazonaws.com/acme-org/acme-app
  secret: false

Two environment variables are emitted per build. The Ngnix build in our example above will generate the variables NGINX_REGISTRY_IMAGE_URL and NGINX_REGISTRY_IMAGE_SHA.

Here's an example of generated build environment variables:

Last updated

Was this helpful?