🐳📦 Run any build step in a Docker container
Clone or download
Latest commit b0e5c6e Jan 23, 2019

README.md

Docker Buildkite Plugin Build status

A Buildkite plugin for running pipeline steps in Docker containers.

Also see the Docker Compose Buildkite Plugin which supports docker-compose.yml, multiple containers, and overriding many of Docker’s defaults.

Example

The following pipeline will build a binary in the dist directory using golang Docker image and then uploaded as an artifact.

steps:
  - command: "go build -o dist/my-app ."
    artifact_paths: "./dist/my-app"
    plugins:
      - docker#v2.2.0:
          image: "golang:1.11"

Windows images are also supported:

steps:
  - command: "dotnet publish -c Release -o published"
    plugins:
      - docker#v2.2.0:
          image: "microsoft/dotnet:latest"
          always-pull: true

If you want to control how your command is passed to the docker container, you can use the command parameter on the plugin directly, which also disables the default volume mounts:

steps:
  - plugins:
      - docker#v2.2.0:
          image: "mesosphere/aws-cli"
          always-pull: true
          command: ["s3", "sync", "s3://my-bucket/dist/", "/app/dist"]
          volumes: [ "./:/app" ]
    artifact_paths: "dist/**"

You can pass in additional environment variables and customize what is mounted into the container:

steps:
  - command:
      - "yarn install"
      - "yarn run test"
    plugins:
      - docker#v2.2.0:
          image: "node:7"
          always-pull: true
          workdir: "/app"
          volumes:
            - "./code:/app"
          environment:
            - "MY_SECRET_KEY"
            - "MY_SPECIAL_BUT_PUBLIC_VALUE=kittens"

Environment variables available in the step can also automatically be propagated to the container:

steps:
  - command:
      - "yarn install"
      - "yarn run test"
    env:
      MY_SPECIAL_BUT_PUBLIC_VALUE: kittens
    plugins:
      - docker#v2.2.0:
          image: "node:7"
          always-pull: true
          workdir: "/app"
          volumes:
            - "./code:/app"
          propagate-environment: true

You can pass in additional volumes to be mounted. This disables the default mount behaviour of mounting $PWD to /workdir. This is useful for running Docker :

steps:
  - commands:
      - "docker build . -t image:tag"
      - "docker push image:tag"
    plugins:
      - docker#v2.2.0:
          image: "docker:latest"
          always-pull: true
          volumes:
            - ".:/work"
            - "/var/run/docker.sock:/var/run/docker.sock"

You can disable all mounts, including the default by setting volumes to false:

steps:
  - command: "npm start"
    plugins:
      - docker#v2.2.0:
          image: "node:7"
          always-pull: true
          volumes: false

Configuration

Required

image (required, string)

The name of the Docker image to use.

Example: node:7

Optional

additional-groups (optional, array)

Additional groups to be added to the user in the container, in an array of group names (or ids). See https://docs.docker.com/engine/reference/run/#additional-groups for more details.

Example: docker

always-pull (optional, boolean)

Whether to always pull the latest image before running the command. Useful if the image has a latest tag.

Default: false

command (optional, array)

Sets the command for the Docker image, and defaults the shell option to false. Useful if the Docker image has an entrypoint, or doesn't contain a shell.

This option can't be used if your step already has a top-level, non-plugin command option present.

Examples: [ "/bin/mycommand", "-c", "test" ], ["arg1", "arg2"]

debug (optional, boolean)

Enables debug mode, which outputs the full Docker commands that will be run on the agent machine.

Default: false

entrypoint (optional, string)

Override the image’s default entrypoint, and defaults the shell option to false. See the docker run --entrypoint documentation for more details.

Example: /my/custom/entrypoint.sh

environment (optional, array)

An array of additional environment variables to pass into to the docker container. Items can be specified as either KEY or KEY=value. Each entry corresponds to a Docker CLI --env parameter. Values specified as variable names will be passed through from the outer environment.

Example: [ "BUILDKITE_MESSAGE", "MY_SECRET_KEY", "MY_SPECIAL_BUT_PUBLIC_VALUE=kittens" ]

propagate-environment (optional, boolean)

Whether or not to automatically propagate all pipeline environment variables into the docker container. Avoiding the need to be specified with environment.

Note that only pipeline variables will automatically be propagated (what you see in the Buildkite UI). Variables set in proceeding hook scripts will not be propagated to the container.

privileged (optional, boolean)

Whether or not to run the container in privileged mode

init (optional, boolean)

Whether or not to run an init process inside the container. This ensures that responsibilities like reaping zombie processes are performed inside the container.

See Docker's documentation for background and implementation details.

Default: true for Linux and macOS, false for Windows.

mount-buildkite-agent (optional, boolean)

Whether to automatically mount the buildkite-agent binary from the host agent machine into the container. Set to false if you want to disable, or if you already have your own binary in the image.

Default: true for Linux, and false for macOS and Windows.

mount-ssh-agent (optional, boolean)

Whether to automatically mount the ssh-agent socket from the host agent machine into the container (at /ssh-agentand /root/.ssh/known_hosts respectively), allowing git operations to work correctly.

Default: false

network (optional, string)

Join the container to the docker network specified. The network will be created if it does not already exist. See https://docs.docker.com/engine/reference/run/#network-settings for more details.

Example: test-network

runtime (optional, string)

Specify an explicit docker runtime. See the docker run options documentation for more details.

Example: nvidia

shell (optional, array or boolean)

Set the shell to use for the command. Set it to false to pass the command directly to the docker run command. The default is ["/bin/sh", "-e", "-c"] unless you have provided an entrypoint or command.

Example: [ "powershell", "-Command" ]

shm-size (optional, string)

Set the size of the /dev/shm shared memory filesystem mount inside the docker contianer. If unset, uses the default for the platform (typically 64mb). See docker run’s runtime constraints documentation for information on allowed formats.

Example: 2gb

tty (optional, boolean)

If set to false, doesn't allocate a TTY. This is useful in some situations where TTY's aren't supported, for instance windows.

Default: true for Linux and macOS, and false for Windows.

user (optional, string)

Allows a user to be set, and override the USER entry in the Dockerfile. See https://docs.docker.com/engine/reference/run/#user for more details.

Example: root

volumes (optional, array or boolean)

Extra volume mounts to pass to the docker container, in an array. Items are specified as SOURCE:TARGET. Each entry corresponds to a Docker CLI --volume parameter. Relative local paths are converted to their full-path (e.g .:/app).

To disable the default mount mounts, set volumes to false.

Default: true Example: [ ".:/app", "/var/run/docker.sock:/var/run/docker.sock" ]

tmpfs (optional, array)

Tmpfs mounts to pass to the docker container, in an array. Each entry corresponds to a Docker CLI --tmpfs parameter. See Docker's tmpfs mounts documentation for more information on this feature.

Example: [ "/tmp", "/root/.cache" ]

workdir(optional, string)

The working directory to run the command in, inside the container. The default is /workdir.

Example: /app

Developing

You can use the bk cli to run the test pipeline locally, or just the tests using Docker Compose directly:

docker-compose run --rm tests

License

MIT (see LICENSE)