Application of PerformanceProfile with invalid cpuset in one of the reserved/isolated/shared/offlined cpu fields causing webhook validation to panic instead of returning an informant error.

4.18.0-0.nightly-2024-11-22-231049

Apply a PerformanceProfile with invalid cpu values

Apply the following PerformanceProfile with invalid cpu values:

apiVersion: performance.openshift.io/v2
kind: PerformanceProfile
metadata:
  name: pp
spec:
  cpu:
    isolated: 'garbage'
    reserved: 0-3
  machineConfigPoolSelector:
    pools.operator.machineconfiguration.openshift.io/worker-cnf: ""
  nodeSelector:
    node-role.kubernetes.io/worker-cnf: ""     

On OCP >= 4.18 the error is:

Error from server: error when creating "pp.yaml": admission webhook "vwb.performance.openshift.io" denied the request: panic: runtime error: invalid memory address or nil pointer dereference [recovered]
  

On OCP <= 4.17 the error is:

Validation webhook passes without any errors. Invalid configuration propogates to the cluster and breaks it.

We expect to pushback an informant error when invalid cpuset has been entered, without panicking or accepting it!