Rook is a great operator for Kubernetes. We've been using it in our enterprise Kubernetes distribution for a while now and haven't had any problems with it. Ceph is a complex problem, and Rook manages it flawlessly for us.

It's great that when someone wants to have shared storage or they want to have an object store (we only deploy it as block store), the app developer can include K8s YAML to provision block storage or shared storage.

It might be nice to have a "What does Rook do?" explanation. What does it provide over the helm charts for minio, nfs-server-provisioner, cockroachdb, etc.

Do I use it to provision persistent volumes? Databases? Key value stores? S3 compatible blob APIs? Those things all have "persistent storage" in common, but what does Rook do on top of them?

Note that rook is still in beta. I've had pretty bad luck with it not supporting features that I need. First it was not supporting adding additional OSDs on the same node, though I believe that's now fixed. In the end it was that PVCs can't be mounted by multiple pods at a time, even when set to RWX.

See also: https://rook.io/

The name is not nautical in theme, seems sketchy.