Unify binary, repo and image naming

[tnozicka]

We should unify the packages, binary and image name for consistency.

At this point the difference is confusing (caused by historical reasons when requesting the repo name): repo name: github.com/scylladb/k8s-local-volume-provisioner image name: docker.io/scylladb/k8s-local-volume-provisioner binary name: local-csi-driver

[scylla-operator-bot[bot]]

The Scylla Operator project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 30d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out

/lifecycle stale

[tnozicka]

/remove-lifecycle stale /triage accepted

[zimnx]

I collected pros and cons behind few options we discussed, here's a summary:

---

Change k8s-local-volume-provisioner to local-csi-driver

• (pro) name convention matches other drivers available
• (pro) simple user update, it requires changing an image used in manifests. Namespace and object names stay the same.

What needs to be done afterwards:

• rename the Github repo
• create new quay/docker repo, gain access for CI jobs through IT/Tomas
• update all jobs in CI to use new name

Change local-csi-driver to k8s-local-volume-provisioner

• (pro) name convention matches kubernetes-sigs local-static-volume-provisioner
• (con) conterargument about above, this project doesn't provide any provisioner implementation, it's being used as a dependency
• (con) k8s prefix might be an issue, as drivers are designed to be run not only on kubernetes - although there are no plans for having support for other orchestrators
• (con) update requires manually deleting namespace first and only then applying new manifests

What needs to be done afterwards:

• rename binary and names in manifests in CI, Operator and repo itself

Change k8s-local-volume-provisioner to scylladb-local-csi(-driver)

• (pro) scylla prefix prevents collision with other solutions
• (con) nothing is related to scylla, we are just the maintainers
• (con) scylla prefix may be confusing, users can use GCP, AWS CSI or other drivers and run scylla without known issues

What needs to be done afterwards:

• rename Github repo
• create new quay/docker repo, gain access for CI jobs through IT/Tomas
• update all jobs in CI to use new name
• rename binary and names in manifests in CI, Operator and repo itself

---

I had a meeting with @ylebi where we discussed them, and we are align on opting to the first option. @tnozicka what do you think?

[tnozicka]

Thanks for writing this down.

I'd generally be in favour of using scylla- prefix for our namespace

(con) nothing is related to scylla, we are just the maintainers

Kind of, but the name is generic enough for other people writing a CSI driver to pick the same name and conflict with us. Having such prefix is pretty common practice to avoid collisions on naming. Both our GH repo and the container registry already have the prefix in the organization part, so this is meant only for the namespace.

That said, going for the first option Change k8s-local-volume-provisioner to local-csi-driver is fine with me, the non-prefixed namespace is already there, this won't make it worse and it can be done separately later, if we decide to go there.

But please start by renaming the repo, so the opinionated part (which is how this discrepancy came to be) will be sorted out.