The best place to see what’s coming within a 1-2 month timeline is in the public milestones. All open items for the the next numbered milestone (e.g. 1.12) are visualized in the Milestone-Open project board view and planned at the very beginning of the 2-month release cycle. This planning and discussion begins at Cluster API Azure Office Hours after a major release. Active community PR contributions are prioritized throughout the release, but unplanned work will arise. Hence the items in the milestone are a rough estimate which may change. The “next” milestone is a very rough collection of issues for the milestone after the current numbered one to help prioritize upcoming work.
CAPZ is the official production-ready Cluster API implementation to administer the entire lifecycle of self-managed or managed Kubernetes clusters (AKS) on Azure. Cluster API extends the Kubernetes API to provide tooling consistent across on-premises and cloud providers to build and maintain Kubernetes clusters at scale while working with GitOps and the surrounding tooling ecosystem. See related blog post.
There are a number of large priority “Epics” which may span across milestones which we believe are important to providing CAPZ users an even better experience and improving the vision. The CAPZ project board roadmap view tracks the larger “epic issues” and their progress.
ManagedClusters - Provisioning new AKS (ManagedClusters) clusters at scale is a common use case we want to enable an excellent experience with stability and all of the same features available in AKS. Includes:
Latest features for self-managed - These larger or essential features enable a better experience for provisioning self-managed clusters on Azure. Includes:
- MachinePool autoscale support
- Azure CNI v2 support (waiting on functionality to be available from the Azure Networking team)
Supported adoption of existing AKS clusters - makes it easier to onboard to the CAPZ/ASO control plane. Includes: