Skip to content

coder/envbuilder

envbuilder

discord release godoc license

Build development environments from a Dockerfile on Docker, Kubernetes, and OpenShift. Allow developers to modify their environment in a tight feedback loop.

  • Supports devcontainer.json and Dockerfile
  • Cache image layers with registries for speedy builds
  • Runs on Kubernetes, Docker, and OpenShift

Quickstart

The easiest way to get started is to run the envbuilder Docker container that clones a repository, builds the image from a Dockerfile, and runs the $INIT_SCRIPT in the freshly built container.

/tmp/envbuilder directory persists demo data between commands. You can choose a different directory.

docker run -it --rm \
    -v /tmp/envbuilder:/workspaces \
    -e GIT_URL=https://github.com/coder/envbuilder-starter-devcontainer \
    -e INIT_SCRIPT=bash \
    ghcr.io/coder/envbuilder

Edit .devcontainer/Dockerfile to add htop:

$ vim .devcontainer/Dockerfile
- RUN apt-get install vim sudo -y
+ RUN apt-get install vim sudo htop -y

Exit the container, and re-run the docker run command... after the build completes, htop should exist in the container! πŸ₯³

Note

Envbuilder performs destructive filesystem operations! To guard against accidental data loss, it will refuse to run if it detects that KANIKO_DIR is not set to a specific value. If you need to bypass this behavior for any reason, you can bypass this safety check by setting FORCE_SAFE=true.

Git Branch Selection

Choose a branch using GIT_URL with a ref/heads reference. For instance:

GIT_URL=https://github.com/coder/envbuilder-starter-devcontainer/#refs/heads/my-feature-branch

Container Registry Authentication

envbuilder uses Kaniko to build containers. You should follow their instructions to create an authentication configuration.

After you have a configuration that resembles the following:

{
  "auths": {
    "https://index.docker.io/v1/": {
      "auth": "base64-encoded-username-and-password"
    }
  }
}

base64 encode the JSON and provide it to envbuilder as the DOCKER_CONFIG_BASE64 environment variable.

Alternatively, if running envbuilder in Kubernetes, you can create an ImagePullSecret and pass it into the pod as a volume mount. This example will work for all registries.

# Artifactory example
kubectl create secret docker-registry regcred \
  --docker-server=my-artifactory.jfrog.io \
  --docker-username=read-only \
  --docker-password=secret-pass \
  [email protected] \
  -n coder
resource "kubernetes_deployment" "example" {
  metadata {
    namespace = coder
  }
  spec {
    spec {
      container {
        # Define the volumeMount with the pull credentials
        volume_mount {
          name       = "docker-config-volume"
          mount_path = "/envbuilder/config.json"
          sub_path   = ".dockerconfigjson"
        }
      }
      # Define the volume which maps to the pull credentials
      volume {
        name = "docker-config-volume"
        secret {
          secret_name = "regcred"
        }
      }
    }
  }
}

Docker Hub

Authenticate with docker login to generate ~/.docker/config.json. Encode this file using the base64 command:

$ base64 -w0 ~/.docker/config.json
ewoJImF1dGhzIjogewoJCSJodHRwczovL2luZGV4LmRvY2tlci5pby92MS8iOiB7CgkJCSJhdXRoIjogImJhc2U2NCBlbmNvZGVkIHRva2VuIgoJCX0KCX0KfQo=

Provide the encoded JSON config to envbuilder:

DOCKER_CONFIG_BASE64=ewoJImF1dGhzIjogewoJCSJodHRwczovL2luZGV4LmRvY2tlci5pby92MS8iOiB7CgkJCSJhdXRoIjogImJhc2U2NCBlbmNvZGVkIHRva2VuIgoJCX0KCX0KfQo=

Docker-in-Docker

See here for instructions on running Docker containers inside environments built by Envbuilder.

Git Authentication

Two methods of authentication are supported:

HTTP Authentication

If the GIT_URL supplied starts with http:// or https://, envbuilder will supply HTTP basic authentication using GIT_USERNAME and GIT_PASSWORD, if set.

For access token-based authentication, follow the following schema (if empty, there's no need to provide the field):

Provider GIT_USERNAME GIT_PASSWORD
GitHub [access-token]
GitLab oauth2 [access-token]
BitBucket x-token-auth [access-token]
Azure DevOps [access-token]

If using envbuilder inside of Coder, you can use the coder_external_auth Terraform resource to automatically provide this token on workspace creation:

data "coder_external_auth" "github" {
    id = "github"
}

resource "docker_container" "dev" {
    env = [
        GIT_USERNAME = data.coder_external_auth.github.access_token,
    ]
}

SSH Authentication

If the GIT_URL supplied does not start with http:// or https://, envbuilder will assume SSH authentication. You have the following options:

  1. Public/Private key authentication: set GIT_SSH_KEY_PATH to the path of an SSH private key mounted inside the container. Envbuilder will use this SSH key to authenticate. Example:

     docker run -it --rm \
       -v /tmp/envbuilder:/workspaces \
       -e [email protected]:path/to/private/repo.git \
       -e GIT_SSH_KEY_PATH=/.ssh/id_rsa \
       -v /home/user/id_rsa:/.ssh/id_rsa \
       -e INIT_SCRIPT=bash \
       ghcr.io/coder/envbuilder
  2. Agent-based authentication: set SSH_AUTH_SOCK and mount in your agent socket, for example:

  docker run -it --rm \
    -v /tmp/envbuilder:/workspaces \
    -e [email protected]:path/to/private/repo.git \
    -e INIT_SCRIPT=bash \
    -e SSH_AUTH_SOCK=/tmp/ssh-auth-sock \
    -v $SSH_AUTH_SOCK:/tmp/ssh-auth-sock \
    ghcr.io/coder/envbuilder

Note: by default, envbuilder will accept and log all host keys. If you need strict host key checking, set SSH_KNOWN_HOSTS and mount in a known_hosts file.

Layer Caching

Cache layers in a container registry to speed up builds. To enable caching, authenticate with your registry and set the CACHE_REPO environment variable.

CACHE_REPO=ghcr.io/coder/repo-cache

To experiment without setting up a registry, use LAYER_CACHE_DIR:

docker run -it --rm \
  -v /tmp/envbuilder-cache:/cache \
  -e LAYER_CACHE_DIR=/cache
  ...

Each layer is stored in the registry as a separate image. The image tag is the hash of the layer's contents. The image digest is the hash of the image tag. The image digest is used to pull the layer from the registry.

The performance improvement of builds depends on the complexity of your Dockerfile. For coder/coder, uncached builds take 36m while cached builds take 40s (~98% improvement).

Image Caching

When the base container is large, it can take a long time to pull the image from the registry. You can pre-pull the image into a read-only volume and mount it into the container to speed up builds.

# Pull your base image from the registry to a local directory.
docker run --rm \
  -v /tmp/kaniko-cache:/cache \
  gcr.io/kaniko-project/warmer:latest \
    --cache-dir=/cache \
    --image=<your-image>

# Run envbuilder with the local image cache.
docker run -it --rm \
  -v /tmp/kaniko-cache:/image-cache:ro \
  -e BASE_IMAGE_CACHE_DIR=/image-cache

In Kubernetes, you can pre-populate a persistent volume with the same warmer image, then mount it into many workspaces with the ReadOnlyMany access mode.

A sample script to pre-fetch a number of images can be viewed here. This can be run, for example, as a cron job to periodically fetch the latest versions of a number of base images.

Setup Script

The SETUP_SCRIPT environment variable dynamically configures the user and init command (PID 1) after the container build process.

Note

TARGET_USER is passed to the setup script to specify who will execute INIT_COMMAND (e.g., code).

Write the following to $ENVBUILDER_ENV to shape the container's init process:

  • TARGET_USER: Identifies the INIT_COMMAND executor (e.g.root).
  • INIT_COMMAND: Defines the command executed by TARGET_USER (e.g. /bin/bash).
  • INIT_ARGS: Arguments provided to INIT_COMMAND (e.g. -c 'sleep infinity').
# init.sh - change the init if systemd exists
if command -v systemd >/dev/null; then
  echo "Hey πŸ‘‹ $TARGET_USER"
  echo INIT_COMMAND=systemd >> $ENVBUILDER_ENV
else
  echo INIT_COMMAND=bash >> $ENVBUILDER_ENV
fi

# run envbuilder with the setup script
docker run -it --rm \
  -v ./:/some-dir \
  -e SETUP_SCRIPT=/some-dir/init.sh \
  ...

Custom Certificates

  • SSL_CERT_FILE: Specifies the path to an SSL certificate.
  • SSL_CERT_DIR: Identifies which directory to check for SSL certificate files.
  • SSL_CERT_BASE64: Specifies a base64-encoded SSL certificate that will be added to the global certificate pool on start.

Local Development

Building envbuilder currently requires a Linux system.

On MacOS or Windows systems, we recommend either using a VM or the provided .devcontainer for development.

Additional Requirements:

  • go 1.21
  • make
  • Docker daemon (for running tests)

Makefile targets:

  • build: builds and tags envbuilder:latest for your current architecture.
  • develop: runs envbuilder:latest against a sample Git repository.
  • test: run tests.
  • test-registry: stands up a local registry for caching images used in tests.

Environment Variables

Flag Environment variable Default Description
--setup-script ENVBUILDER_SETUP_SCRIPT The script to run before the init script. It runs as the root user regardless of the user specified in the devcontainer.json file. SetupScript is ran as the root user prior to the init script. It is used to configure envbuilder dynamically during the runtime. e.g. specifying whether to start systemd or tiny init for PID 1.
--init-script ENVBUILDER_INIT_SCRIPT The script to run to initialize the workspace. Default: sleep infinity.
--init-command ENVBUILDER_INIT_COMMAND The command to run to initialize the workspace. Default: /bin/sh.
--init-args ENVBUILDER_INIT_ARGS The arguments to pass to the init command. They are split according to /bin/sh rules with https://github.com/kballard/go-shellquote.
--cache-repo ENVBUILDER_CACHE_REPO The name of the container registry to push the cache image to. If this is empty, the cache will not be pushed.
--base-image-cache-dir ENVBUILDER_BASE_IMAGE_CACHE_DIR The path to a directory where the base image can be found. This should be a read-only directory solely mounted for the purpose of caching the base image.
--layer-cache-dir ENVBUILDER_LAYER_CACHE_DIR The path to a directory where built layers will be stored. This spawns an in-memory registry to serve the layers from.
--devcontainer-dir ENVBUILDER_DEVCONTAINER_DIR The path to the folder containing the devcontainer.json file that will be used to build the workspace and can either be an absolute path or a path relative to the workspace folder. If not provided, defaults to .devcontainer.
--devcontainer-json-path ENVBUILDER_DEVCONTAINER_JSON_PATH The path to a devcontainer.json file that is either an absolute path or a path relative to DevcontainerDir. This can be used in cases where one wants to substitute an edited devcontainer.json file for the one that exists in the repo.
--dockerfile-path ENVBUILDER_DOCKERFILE_PATH The relative path to the Dockerfile that will be used to build the workspace. This is an alternative to using a devcontainer that some might find simpler.
--build-context-path ENVBUILDER_BUILD_CONTEXT_PATH Can be specified when a DockerfilePath is specified outside the base WorkspaceFolder. This path MUST be relative to the WorkspaceFolder path into which the repo is cloned.
--cache-ttl-days ENVBUILDER_CACHE_TTL_DAYS The number of days to use cached layers before expiring them. Defaults to 7 days.
--docker-config-base64 ENVBUILDER_DOCKER_CONFIG_BASE64 The base64 encoded Docker config file that will be used to pull images from private container registries.
--fallback-image ENVBUILDER_FALLBACK_IMAGE Specifies an alternative image to use when neither an image is declared in the devcontainer.json file nor a Dockerfile is present. If there's a build failure (from a faulty Dockerfile) or a misconfiguration, this image will be the substitute. Set ExitOnBuildFailure to true to halt the container if the build faces an issue.
--exit-on-build-failure ENVBUILDER_EXIT_ON_BUILD_FAILURE Terminates the container upon a build failure. This is handy when preferring the FALLBACK_IMAGE in cases where no devcontainer.json or image is provided. However, it ensures that the container stops if the build process encounters an error.
--force-safe ENVBUILDER_FORCE_SAFE Ignores any filesystem safety checks. This could cause serious harm to your system! This is used in cases where bypass is needed to unblock customers.
--insecure ENVBUILDER_INSECURE Bypass TLS verification when cloning and pulling from container registries.
--ignore-paths ENVBUILDER_IGNORE_PATHS The comma separated list of paths to ignore when building the workspace.
--skip-rebuild ENVBUILDER_SKIP_REBUILD Skip building if the MagicFile exists. This is used to skip building when a container is restarting. e.g. docker stop -> docker start This value can always be set to true - even if the container is being started for the first time.
--git-url ENVBUILDER_GIT_URL The URL of the Git repository to clone. This is optional.
--git-clone-depth ENVBUILDER_GIT_CLONE_DEPTH The depth to use when cloning the Git repository.
--git-clone-single-branch ENVBUILDER_GIT_CLONE_SINGLE_BRANCH Clone only a single branch of the Git repository.
--git-username ENVBUILDER_GIT_USERNAME The username to use for Git authentication. This is optional.
--git-password ENVBUILDER_GIT_PASSWORD The password to use for Git authentication. This is optional.
--git-ssh-private-key-path ENVBUILDER_GIT_SSH_PRIVATE_KEY_PATH Path to an SSH private key to be used for Git authentication.
--git-http-proxy-url ENVBUILDER_GIT_HTTP_PROXY_URL The URL for the HTTP proxy. This is optional.
--workspace-folder ENVBUILDER_WORKSPACE_FOLDER The path to the workspace folder that will be built. This is optional.
--ssl-cert-base64 ENVBUILDER_SSL_CERT_BASE64 The content of an SSL cert file. This is useful for self-signed certificates.
--export-env-file ENVBUILDER_EXPORT_ENV_FILE Optional file path to a .env file where envbuilder will dump environment variables from devcontainer.json and the built container image.
--post-start-script-path ENVBUILDER_POST_START_SCRIPT_PATH The path to a script that will be created by envbuilder based on the postStartCommand in devcontainer.json, if any is specified (otherwise the script is not created). If this is set, the specified InitCommand should check for the presence of this script and execute it after successful startup.
--coder-agent-url CODER_AGENT_URL URL of the Coder deployment. If CODER_AGENT_TOKEN is also set, logs from envbuilder will be forwarded here and will be visible in the workspace build logs.
--coder-agent-token CODER_AGENT_TOKEN Authentication token for a Coder agent. If this is set, then CODER_AGENT_URL must also be set.
--coder-agent-subsystem CODER_AGENT_SUBSYSTEM Coder agent subsystems to report when forwarding logs. The envbuilder subsystem is always included.

About

Build development environments from a Dockerfile on Docker, Kubernetes, and OpenShift. Enable developers to modify their development environment quickly.

Resources

License

Code of conduct

Security policy

Stars

Watchers

Forks

Languages