Create, manage, snapshot, and scale Kubernetes infrastructure in the public cloud.
kubicorn is an unofficial project that solves the Kubernetes infrastructure problem and
gives users a rich golang library to work with infrastructure.
kubicorn is a project that helps a user manage cloud infrastructure for Kubernetes. With
kubicorn a user can create new clusters, modify and scale them, and take a snapshot of their cluster at any time.
NOTE: This is a work-in-progress, we do not consider it production ready. We're working hard towards the
v1.0.0 release and we're expecting it in the upcoming months.
Use at your own risk and if you're as excited about it as we are, maybe you want to join us on the
#kubicorn channel in the Kubernetes Slack community.
kubicorn is currently going through a breaking API changes as we adopt the upstream cluster API. We have moved
github.com/kubicorn/kubicorn. We are targeting a stable release shortly!
Previously, we mainly used a channel in the Gophers Slack community, which is still active, but we're moving to the Kubernetes Slack. You can also get involved and send your questions to our public mailing list.
We hold developer calls biweekly on Tuesdays, 1pm Pacific Time. By joining the mailing list, you'll get a calendar invite.
This is a community driven project. We love you, and respect you. We are here to help you learn, help you grow, and help you succeed. If you have an idea, please share it.
We are all software engineers, and we all work in many different code bases. We want the code to be stable, and approachable. We strive for clean and simple software, and we encourage refactoring and fixing technical debt.
We want our tool to work, and work well. If an operator is running
kubicorn it should feel comfortable and make sense
to them. We want operators to feel empowered.
Infrastructure as software.
We believe that the oh-so important layer of infrastructure should be represented as software (not as code!). We hope that our project demonstrates this idea, so the community can begin thinking in the way of the new paradigm.
Rainbows and Unicorns
We believe that sharing is important, and encouraging our peers is even more important. Part of contributing to
means respecting, encouraging, and welcoming others to the project.
kubicorn lets a user create a Kubernetes cluster in a cloud of their choice.
Define what you want, then apply it. That simple.
kubicorn can adopt any Kubernetes cluster at any time.
kubicorn is powered by a state enforcement pattern. A user defines the intended state of Kubernetes infrastructure,
kubicorn can enforce the intended state.
kubicorn allows a user to take a snapshot of a Kubernetes cluster, and run the image in any cloud at any
time. A snapshot is compressed file that will represent intended infrastructure and intended application
definitions. Take a snap, save a snap, deploy a snap.
kubicorn is built as a library and a framework. Thus allowing it to be easily vendored into operator and
controller patterns to enforce intended state of infrastructure.
- We use
kubeadmto bootstrap our clusters.
- We strive for developer empathy, and clean and simple code.
- We strive for operational empathy, and clean and simple user experience.
- We start with
structliterals for profiles, and then marshal into an object.
- We offer the tooling as a library, more than a command line tool.
- We are atomic, and will un-do any work if there is an error.
- We run on many operating systems.
- We allow users to define their own arbitrary bootstrap logic.
- We have no guarantee that anything works, ever, use at your own risk.
- We have no dependency on DNS.
- We believe in snapshots, and that a user should be able to capture a cluster, and move it.