README.md

Multicluster Special Interest Group

A Special Interest Group focused on solving common challenges related to the management of multiple Kubernetes clusters, and applications that exist therein. The SIG will be responsible for designing, discussing, implementing and maintaining API’s, tools and documentation related to multi-cluster administration and application management. This includes not only active automated approaches such as Cluster Federation, but also those that employ batch workflow-style continuous deployment systems like Spinnaker and others. Standalone building blocks for these and other similar systems (for example a cluster registry), and proposed changes to kubernetes core where appropriate will also be in scope.

The charter defines the scope and governance of the Multicluster Special Interest Group.

Meetings

Joining the mailing list for the group will typically add invites for the following meetings to your calendar.

• Regular SIG Meeting: Tuesdays at 9:30 PT (Pacific Time) (bi-weekly). Convert to your timezone.
  • Meeting notes and Agenda.
  • Meeting recordings.

Leadership

Chairs

The Chairs of the SIG run operations and processes governing the SIG.

• Jeremy Olmsted-Thompson (@jeremyot), Google
• Paul Morie (@pmorie), Apple

Emeritus Leads

• Quinton Hoole (@quinton-hoole)

Contact

• Slack: #sig-multicluster
• Mailing list
• Open Community Issues/PRs
• GitHub Teams:
  • @kubernetes/sig-multicluster-api-reviews - API Changes and Reviews
  • @kubernetes/sig-multicluster-bugs - Bug Triage and Troubleshooting
  • @kubernetes/sig-multicluster-feature-requests - Feature Requests
  • @kubernetes/sig-multicluster-misc - General Discussion
  • @kubernetes/sig-multicluster-pr-reviews - PR Reviews
  • @kubernetes/sig-multicluster-test-failures - Test Failures and Triage
  • @kubernetes/sig-mutlicluster-proposals - Design Proposals
• Steering Committee Liaison: Bob Killen (@mrbobbytables)

Subprojects

The following subprojects are owned by sig-multicluster:

Kubefed

• Owners:
  • kubernetes-sigs/kubefed

about-api

• Owners:
  • kubernetes-sigs/about-api

kubemci

• Owners:
  • GoogleCloudPlatform/k8s-multicluster-ingress

mcs-api

• Owners:
  • kubernetes-sigs/mcs-api

work-api

• Owners:
  • kubernetes-sigs/work-api

Subprojects

Kubefed

Control Plane for newer Multicluster-specific APIs. Discussions are ongoing to focus future development on multi-cluster specific Kubernetes APIs rather than a reimplementation of single-cluster Kubernetes APIs.

Recommended for Production	Not yet.
General Availability	Beta by 2018-Q4. GA TBD.
Current Level of Activity	Once to twice weekly discussions as part of Kubefed working group (most active SIG-Multicluster members).
Owner(s)	Working Group
Where to find it?	https://github.com/kubernetes-sigs/kubefed
Meeting Agenda	https://docs.google.com/document/d/1v-Kb1pUs3ww_x0MiKtgcyTXCAuZlbVlz4_A9wS3_HXY

Cluster Registry

Common abstraction for a Registry of Clusters that can store per-Cluster metadata and supports Kubernetes label selection. The Cluster Registry can be deployed as a standalone or an aggregated API server and currently provides a Registry of Clusters without any actively reconciling Kubernetes controller. The API design is documented here and is intended to serve as a basis to develop multicluster controllers.

Recommended for Production	Not yet.
General Availability	Beta by 2018-Q1, GA by 2018-Q2 (see Project plan for details).
Current Level of Activity	Alpha, active development for Beta and GA milestones.
Owner(s)	@perotinus, @font
Where to find it?	https://github.com/kubernetes/cluster-registry

Kubemci (Kube Multi-cluster Ingress)

Ability to program Global multicluster load balancers with two deliverables:

• Standalone CLI tool without a centralized control plane. Allows ingress configurations that span multiple clusters. It needs to be invoked whenever there is a change in the number of clusters (healthchecking is still maintained in the background). Much of this work will be folded back into the next deliverable.
• Kubernetes style multicluster API and self-healing controller with cluster registry

Recommended for Production	Not yet.
General Availability	CLI: Beta by 2018-Q1, Controller: Alpha by 2018-Q1
Current Level of Activity?	CLI is alpha and team is looking for feedback. Active development is on to bring it to beta.
Owner(s)	@nikhiljindal, @G-Harmon
Where to find it?	https://github.com/GoogleCloudPlatform/k8s-multicluster-ingress

Federation v1 (a.k.a. Cluster Federation)

Control Plane that coordinates configuration across multiple clusters by presenting a subset of the existing single-cluster Kubernetes APIs. Initially released as part of Kubernetes 1.3, Cluster Federation has implemented parts of the existing “single-cluster” Kubernetes API so that they may be used transparently in a multi-cluster fashion. As of Kubernetes 1.9, many of the non-Federated Kubernetes Workloads APIs have moved to GA (Deployments, ReplicaSets, Daemonset, StatefulSet). However, under Federation, these workloads have not reached the maturity of “GA” since they are re-implemented by different federated controllers. The SIG is currently re-thinking how to support Kubernetes APIs in a Federation fashion and this discussion is reflected under Kubefed.

Recommended for Production	No. Between Alpha and Beta levels of functionality depending on Kubernetes APIs used.
General Availability	No target set date.
Current Level of Activity	On Hold. See Kubefed.
Owner(s)	@shashidharatd
Where to find it?	Out of main Kubernetes repository as of 1.9 and into its own kubernetes/federation repository (kubefed is no longer distributed as part of Kubernetes client tools). Users have to build their own Federation Control Plane and kubefed client tool or wait on issue #192 to be resolved.