Closed
Description
Cluster autoscaler is running out of memory at 200+ nodes. Raise the limits of cluster-autoscale to allow the cluster-autoscaler to use more memory than requested if available. It might get evicted from the node but the operator nodegroup will scale up and will be provisioned on the new instance.
Metadata
Metadata
Assignees
Labels
No labels