A Platform Engineer is troubleshooting why a TKG cluster batch-processing is not scaling up, despite there being 50+ pods in the Pending state for over 30 minutes.
The engineer checks the autoscaler status annotations on the worker node pool: min-size: 3, max-size:
10. Current replicas: 5.
The engineer retrieves the logs from the CAPW (Cluster API Provider for vSphere) controller on the Supervisor and sees:
I1123 10:00:00.123 controller.go:100] Reconciling MachineSet "batch-processing-workers-md-0" E1123 10:00:00.125 controller.go:150] Failed to create Machine "batch-processing-workers-w9z4":
Quota "ns-resource-quota" exceeded.
Requested: cpu=4000m, memory=16Gi.
Used: cpu=20000m, memory=80Gi.
Limit: cpu=22000m, memory=88Gi.
Based on the logs, what is the root cause?