Dagger: the devops superglu, by the creators of docker

Dagger is a scripting environment for automating the delivery of your application to the cloud, from code to production.

If you are familiar with PaaS services like Heroku, Firebase or Openshift: Dagger makes it easy to create a custom PaaS that is perfectly tailored to each application, instead of forcing it into a rigid, homogeneous mold. The Dagger philosophy is that the platform should adapt to the application, not the other way around.

This means that, unlike traditional PaaS, you can use Dagger with all your applications, without having to change your tooling, software stack or infrastructure.

TODO: [... principles: productivity + compat] [.. do out-of-the-box things that would require months of work otherwise] [.. alternative to the dilemna: paas or build your own?]

Warning: alpha software

Please note that Dagger is alpha-quality software. This means it has many bugs, the user interface is minimal, and it may change in incompatible ways at any time.

If you are still willing to try it, thank you! We appreciate your help. Please do not hesitate to ask questions, report bugs, or share feedback, either in a github or in the discord chatroom.

Platform as code

The most important feature of Dagger is that it is programmable.

Thanks to its powerful scripting environment and growing catalog of reusable components, anyone with basic programming knowledge can assemble a custom platform for their application, and quickly start automating repetitive and complex tasks, so the entire team can spend less time getting deployment to work, and more time developing.

Whether you're a seasoned SRE building a custom PAAS for your organization, a hobbyist on a fun over-engineered week-end project, or a developer trying to setup CICD because, well, someone has to do it.. There's a whole community of fellow automation enthusiasts ready to help you write your first script.

Usage examples

A few examples of how Dagger is used in the wild:

• Use AWS Elastic Container Service to deploy the new API, while continuing to deploy the main app on Heroku.
• Deploy lightweight staging environments on-demand for QA, integration testing or product demos.
• Run integration tests on a live production-like deployment, automatically, for each pull request.
• Deploy the same app on Netlify for testing, and on Kubernetes for production
• Replace a 500-line deploy.sh with a 10-line configuration file
• Control sprawl of serverless functions on AWS, Google Cloud, Cloudflare, Netlify etc. by gradually moving them to a generic interface, and switching backend at will.
• When the ML team uploads a new model to their S3 bucket, automatically incorporate it into staging deployments, but not into production until manual confirmation!
• Rotate database credentials, and automatically re-deploy all staging environments with the new credentials.
• Allocate cool auto-generated URLs to development instances, and automatically configure your DNS, load-balancer and SSL certificate manager to route traffic to them.
• Orchestrate application deployment across 2 infrastructure siloes, one managed with CloudFormation, the other with Terraform.
• Migrate from Helm to Kustomize, without disrupting next week's big release.

Getting started

1. Build the dagger command-line tool. You will need Go version 1.16 or later.

$ make

2. Copy the dagger tool to a location listed in your $PATH. For example, to copy it to /usr/local/bin:

$ cp ./cmd/dagger/dagger /usr/local/bin

3. Run buildkitd on your local machine. The simplest way to do this is using Docker: docker run -d --name buildkitd --privileged moby/buildkit:latest

On a machine with Docker installed, run:

$ docker run -d --name buildkitd --privileged moby/buildkit:latest

4. Compute a test configuration

Currently dagger can only do one thing: compute a configuration with optional inputs, and print the result.

If you are confused by how to use this for your application, that is normal: dagger compute is a low-level command which exposes the naked plumbing of the Dagger engine. In future versions, more user-friendly commands will be available which hide the complexity of dagger compute (but it will always be available to power users, of course!).

Here is an example command, using an example configuration:

$ dagger compute ./examples/simple --input-string www.host=mysuperapp.com --input-dir www.source=.

Custom buildkit setup

Dagger can be configured to use an existing buildkit daemon, running either locally or remotely. This can be done using two environment variables: BUILDKIT_HOST and DOCKER_HOST.

To use a buildkit daemon listening on TCP port 1234 on localhost:

$ export BUILDKIT_HOST=tcp://localhost:1234

To use a buildkit daemon running in a container named "super-buildkit" on the local docker host:

$ export BUILDKIT_HOST=docker-container://super-buildkit

To use a buildkit daemon running on a remote docker host (be careful to properly secure remotely accessible docker hosts!)

$ export BUILDKIT_HOST=docker-container://super-buildkit
$ export DOCKER_HOST=tcp://my-remote-docker-host:2376

Comparison to other automation platforms

CICD

Github, Gitlab, Jenkins, Spinnaker, Tekton

Build systems

Bazel, Nix, Skopeo

Infrastructure automation

Terraform, Pulumi, Ansible

Traditional scripting

Bash, Make, Python

PaaS

Heroku, Elastic Beanstalk, Cloud Foundry, Openshift

Kubernetes management

Kustomize, Helm, jsonnet

Gitops

Flux, ...