We have created a new forum here to talk about Cluster API: https://discuss.kubernetes.io/c/contributors/cluster-api. We encourage everyone to come participate there for discussions specific to Cluster API. This does not replace other tools such as Slack or GitHub issues; instead, it is meant to be a place where we can organize and have asynchronous discussions on high level design topics or anything else that doesn’t necessarily fit into Slack or GitHub.
I also wanted to mention that at today’s Cluster API meeting, we discussed several new workstreams to help us drive toward v1alpha2. These are documented in the meeting minutes document, and I’m summarizing them below:
How does Cluster API interact with providers? webhooks, gRPC, CRDs?
Refine the data model for the project, including concepts such as machines, clusters, control planes.
Make the cluster control plane a first class resource. Support scaling up/down of control plane. Support control plane upgrades.
Define the extension points for bootstrapping control plane and non control plane nodes. Provide a reference implementation using kubeadm. Figure out implementations with cluster autoscaler.
If you feel there are other workstreams that are not currently listed, please create new topics at https://discuss.kubernetes.io/c/contributors/cluster-api and/or bring them up at the next weekly meeting.
We are looking for co-facilitators for each workstream. The role of the facilitators is to set up the initial kickoff meeting for each workstream, moderate meetings, and report workstream progress summaries to the overall Cluster API community. They are not meant to “own” the workstream.
I’m really looking forward to seeing how quickly we can make progress on all of these items!