Pi Cluster
estimated read time: 2 minutes
Creating a Home Kubernetes Service
History
In 2021/2022, I made the transition from running Pi-hole on a single Raspberry Pi 4 Model B—a setup that served me well for years—to running it as a service on K3S in a highly available cluster with multiple nodes.
This blog series covers my setup, exploring the different layers of the Kubernetes cluster episode by episode. Each article is linked below as a single index for future reference.
Why Bother?
A fair question. If Pi-hole was working perfectly well on a single Pi, why add complexity?
The honest answer is partly “because I can”, but there’s more to it than that. Running workloads on Kubernetes introduces high availability, allows you to share compute across multiple projects, and creates a more economical solution. You also get a proper delivery pipeline around your containers—a more complete and repeatable process.
I’ve recently been chatting with others online who are trying to decide what to do with spare compute lying around. Personally, this has been a brilliant use of my three Raspberry Pi 4 Model B 8GB units. It also offered the right level of complexity and learning curve to avoid being just another software install—something I wouldn’t have found challenging enough.