Beginning with Spring Cloud Data Flow v2.0.0, we have consolidated the Local, Cloud Foundry, and Kubernetes servers into a single server. Instead of building, maintaining, and shipping individual server-implementations, we have consolidated them into a single server, and all the supported platform-specific deployer implementation libraries are bundled in it.

That would mean, as a user, you don’t need to pick different JARs or Docker-images depending on what platform you want to run.

The other notable benefit to this approach is the ability to configure the single-server with multiple platform backends. The same server can be configured against Local, "n" number of Cloud Foundry org/space combinations, and as well against "n" number of Kubernetes clusters and the associated namespace. At the time of stream-deployment or task-launches, you would be able to pick-and-choose the platform where you want to orchestrate the deployment. It would further simplify and promote CI/CD practices for data pipelines.

The getting-started steps for each platform remains mostly the same. One significant change is how Tasks are configured, as we are opening up the capability for Tasks to be launched across different platforms as well.

In summary, please DO NOT use the master branch of this repository; switch over to Spring Cloud Data Flow v2.0 instead. We intend to continue to support the Spring Cloud Data Flow’s v1.x branch of Kubernetes-server for a period of at least a year from the general availability of v2.0 GA release.

A Helm Chart is available for deploying the Spring Cloud Data Flow server and its required services to a Kubernetes Cluster. Checkout the getting-started guide for more details.

We love contributions. Follow this link for more information on how to contribute.

Clone the repo and type

To build the docker image for the Data Flow Server

For more information on building, see this link.