Multicluster-scheduler is multicluster-controller's prime application example. It uses patterns that could be generalized and hosted here as library code for re-use by other projects. Here are some of the patterns used by multicluster-scheduler:
[ ] push/pull observations (snapshots of objects as custom resources)
[ ] push/pull decisions (specs of objects to be applied in another cluster)
[x] cross-cluster garbage collection with finalizers (foreground deletion and orphan)
Multicluster-scheduler is multicluster-controller's prime application example. It uses patterns that could be generalized and hosted here as library code for re-use by other projects. Here are some of the patterns used by multicluster-scheduler:
Other interesting patterns: