The workaround for dealing with unsynced secret, documented here:
https://github.com/stolostron/rhacm-docs/blob/2.9_stage/clusters/release_notes/known_issues.adoc#on-bare-metal-platforms-agent-resources-might-fail-to-pull-ignition

Is not working properly.
 

    [kni@ocp-edge77 ~]$ oc version
Client Version: 4.14.0
Kustomize Version: v5.0.1
Server Version: 4.14.0-0.nightly-2024-05-10-212837
Kubernetes Version: v1.27.13+fd36fb9
[kni@ocp-edge77 ~]$ ~/hypershift_working/hypershift/bin/hcp -v
hcp version openshift/hypershift: 2ec8cfbc768a1088c62bd538a46029f33b982c1f. Latest supported OCP: 4.16.0

TBD    

    1.deploy a hosted cluster, I used this job:https://auto-jenkins-csb-kniqe.apps.ocp-c1.prod.psi.redhat.com/job/CI/job/job-runner/2803/

    2.enable autoscalin min:1,max:4
    3. folowing the workaround mentioned above :
     a. [kni@ocp-edge77 ~]$ oc delete secret/agent-user-data-hosted-0-c5cf52dc -n clusters-hosted-0 --kubeconfig ~/clusterconfigs/auth/hub-kubeconfig
secret "agent-user-data-hosted-0-c5cf52dc" deleted
     b. add a lable to one of the agent (I did 2)

    in the edited agents err :
status:
  conditions:
  - lastTransitionTime: "2024-05-13T07:20:49Z"
    message: 'The Spec could not be synced due to backend error: Failed to get user-data
      secret: failed to get secret clusters-hosted-0/agent-user-data-hosted-0-c5cf52dc
      from API: secrets "agent-user-data-hosted-0-c5cf52dc" not found'
    reason: BackendError
    status: "False"
    type: SpecSynced

In others,no error

Error is reflected in the GUI as well 

    a new secret will be created, and the nodes ignition will start dropping unused nodes.