Summary

In this chapter, we looked into the core components of HA. We explored the ideas of availability, uptime, and fragility. We took those concepts and explored how we could achieve five nines of uptime.

Additionally, we explored the key components of a highly available cluster, the etcd and control plane nodes, and left room to imagine the other ways that we'd build HA into our clusters using hosted PaaS offerings from the major cloud providers.

Later, we looked at the cluster life cycle and dug into advanced capabilities with a number of key features of the Kubernetes system: admission controllers, the workload API, and CRS.

Lastly, we created a CRD on a GKE cluster within GCP in order to understand how to begin building these custom pieces of software.