The beauty of Kubernetes is that it essentially frees an application developer from having to worry about the underlying infrastructure upon which the app runs. If only we could give operators the confidence that Kubernetes will run on their platform of choice—from private infrastructure to public providers—and their host operating system of choice. Oh, and while we’re at it, let’s bundle in the additional operational tooling to monitor the health of Kubernetes itself.
Today, we’re pleased to announce that AppFormix has contributed to the Kubernetes community the testing results on a variety of operating environments. These have been merged as part of Kubernetes 1.3. The continuous integration job now executes the e2e tests nightly. In addition, we are now including operational tooling to monitor the health of Kubernetes cluster in any deployment scenario.
Travis Newhouse has written in detail about the contributions and the direct benefits of this work in a series of blog posts on Kubernetes.io. You can find Part 1 of “Bringing End-to-End Kubernetes Testing to Azure” here and Part 2 here. And, none of this work would have been brought to fruition without the contributions of Weixu Zhuang and Parantap Roy.
Essentially, what we’re offering to the community is confidence: it’s the confidence to know that Kubernetes is stable and runs on your environment of choice. Whether you’re running container workloads on a bare metal cloud or on a public cloud (AWS, Azure, GCE), Kubernetes can run there. What’s more, you also can choose your host operating system and then bundle in the operational tooling that monitors the health of Kubernetes (control plane).
While we were at it, we also added a nightly end-to-end test job for Kubernetes on Azure. This has helped to define the process to contribute a federated test to the Kubernetes project. During the course of the work, we also saw the immediate benefit of expanding test coverage to more platforms when our Azure test job identified compatibility issues.
Kubernetes has many moving parts. At the base of it all is the hard work of how to deploy a distributed system on a number of heterogeneous environments, with different operating systems, network implementations, IaaS, and so on. Even small differences in the environment can create big problems.
More Testing, More Operating Systems, More Cloud Providers
Our work continues as we contribute more tests to the Kubernetes community to cover more operating systems and cloud providers.
Moreover, to give operators confidence that they will be able to manage and maintain a healthy Kubernetes environment, AppFormix has contributed Cloud-Health as an open source project. We thank Fabio Yeon at Google for his insight on how to efficiently and effectively monitor Kubernetes in production.
We also want to thank Aaron Crickenberger, Erick Fejta, Joe Finney, and Ryan Hutchinson for their help incorporating the results of our Azure end-to-end tests into the Kubernetes test history.
We are passionate members of the Kubernetes community and are proud to join forces with others members of the community to move the project forward. Together we possess a great deal of technical talent and enthusiasm, while maintaining an openness and accessibility to new contributors.
If you’d like to get involved with testing to create stable, high-quality releases of Kubernetes, join us in the Kubernetes Testing SIG (sig-testing).