During the Supervisor Cluster upgrade some namespace workloads are not upgraded.

Problem

The Supervisor Cluster upgrade suceeds but some namespace workloads are skipped during the upgrade. The Kubernetes resources indicate insufficient resources and the newly created tier-1 gateway is in ERROR state.

Cause

The load balancer capacity is insufficient to support the workloads.

Solution

Perform one of the following steps to resolve the problem:
  • Delete unused workloads, restart NCP, and run the upgrade again.
  • Add additional Edge nodes to the cluster and trigger a re-allocation to the tier-1 gateway. Restart NCP and run the upgrade again.