In ZTP, sometimes CGU became completed before the spoke cluster is ready, causing no policies to be applied to spoke.  

Hub OCP: 4.18 rc3
ACM: 2.12.2
MCE: 2.7.3
Spoke OCP: 4.18 rc3
TALM: 4.18/4.17

Intermittent - so far only seen a couple times on MNO spoke cluster

    1. MNO cluster was previously installed via ZTP with all nodes up
    2. Clean up resources on hub via cascade deletion of argocd apps (spoke NS, managedcluster, policies, etc)
    3. Trigger new ZTP deployment by pushing contents to git and deploy/configure argocd apps.
    4. Wait for ZTP to complete 
    

    OCP payload installed, but policies are not applied to spoke. CGU shows completed before the cluster install started. 

    CGU should not start before spoke is ready

CGU started/completed before the cluster installation started. 

    history:
    - completionTime: "2025-01-06T20:02:15Z"
      image: registry.hlxcl11.lab.eng.tlv2.redhat.com:5000/openshift-release-dev/ocp-release@sha256:668c92b06279cb5c7a2a692860b297eeb9013af10d49d2095f2c3fe9ad02baaa
      startedTime: "2025-01-06T19:32:31Z"


[kni@registry ~]$ oc get cgu -n ztp-install kni-qe-26 -o yaml
apiVersion: ran.openshift.io/v1alpha1
kind: ClusterGroupUpgrade
...
spec:
  actions:
    afterCompletion:
      addClusterLabels:
        ztp-done: ""
      deleteObjects: true
      removeClusterLabels:
      - ztp-running
    beforeEnable:
      addClusterLabels:
        ztp-running: ""
  backup: false
  clusters:
  - kni-qe-26
  enable: true
  managedPolicies:
  - common-config-policy
  - common-subscriptions-policy
  - group-du-standard-config-policy
  - group-du-standard-site-config-policy
  - group-du-standard-validator-du-policy
  preCaching: false
  preCachingConfigRef: {}
  remediationStrategy:
    maxConcurrency: 1
    timeout: 240
status:
...
  status:
    completedAt: "2025-01-06T19:02:04Z"
    startedAt: "2025-01-06T19:01:23Z"