What is kine?

Kine is a tool that lets you run Kubernetes on top of databases you might already be familiar with, like MySQL, Postgres, or SQLite, instead of the default option, which is a specialized database called etcd. If you've ever wanted to simplify your Kubernetes setup or use infrastructure you already have, this project bridges that gap. Under normal circumstances, Kubernetes requires etcd to store its internal state: which applications are running, their configurations, and so on. Kine acts as a translator sitting between Kubernetes and your preferred database. When Kubernetes asks to read or write data in the way it normally talks to etcd, Kine intercepts those requests and reroutes them to MySQL, Postgres, SQLite, or NATS instead, handling the create, update, and delete operations behind the scenes. This is useful for teams or solo builders who want a lightweight Kubernetes setup without introducing an entirely new database technology into their stack. For example, if your company already operates Postgres heavily and your team knows how to back it up and scale it, using Kine means you can lean on that existing expertise rather than also learning to operate etcd. It can also make small-scale or experimental Kubernetes clusters simpler to manage, since SQLite requires almost no setup. The project notes that it implements only a subset of the etcd API, enough for Kubernetes to function, but not enough to serve as a general-purpose replacement for etcd in other contexts. The README doesn't go into deeper detail on tradeoffs, but the core appeal is straightforward: it gives you flexibility in where Kubernetes stores its data, and it can run on its own so it works with standard Kubernetes, not just K3s.