Control plane machine set operator (CPMSO) fails to roll-out control plane nodes, hanging, because the new etcd pod fails with Init:CrashLoopBackOff.

The error comes from the init container "etcd-ensure-env-vars" because environment parameters are missing from the deployed static yaml. For example:

~~~
  etcd-ensure-env-vars:
...
    Command:
      /bin/sh
      -c
      #!/bin/sh
      set -euo pipefail
      
      : "${NODE_rugouvei_cluster1_bkw7w_master_kxp8p_1_ETCD_URL_HOST?not set}"
      : "${NODE_rugouvei_cluster1_bkw7w_master_kxp8p_1_ETCD_NAME?not set}"
      : "${NODE_rugouvei_cluster1_bkw7w_master_kxp8p_1_IP?not set}"
...  
    State:      Terminated
      Reason:   Error
      Message:  /bin/sh: line 4: NODE_rugouvei_cluster1_bkw7w_master_kxp8p_1_ETCD_URL_HOST: not set
~~~

The error is correct. The only "NODE_*" parameters listed are:

~~~
      NODE_rugouvei_cluster1_bkw7w_master_1_ETCD_NAME:      rugouvei-cluster1-bkw7w-master-1
      NODE_rugouvei_cluster1_bkw7w_master_1_ETCD_URL_HOST:  10.44.135.201
      NODE_rugouvei_cluster1_bkw7w_master_1_IP:             10.44.135.201
      NODE_rugouvei_cluster1_bkw7w_master_2_ETCD_NAME:      rugouvei-cluster1-bkw7w-master-2
      NODE_rugouvei_cluster1_bkw7w_master_2_ETCD_URL_HOST:  10.44.135.240
      NODE_rugouvei_cluster1_bkw7w_master_2_IP:             10.44.135.240
      NODE_IP:                                               (v1:status.podIP)
~~~

"rugouvei_cluster1_bkw7w_master_0" was the first machine/node to be deleted, and the new node "rugouvei-cluster1-bkw7w-master-kxp8p-1" is not in the list causing the container to fail.
  

Reproduced with freshly installed VMware IPI cluster 4.13.23 in Lab.

    1. Install IPI VMware cluster 4.13.23.
    2. Configure CPMSO following documentation: https://docs.openshift.com/container-platform/4.13/machine_management/control_plane_machine_management/cpmso-about.html
    3. Manually delete "master-0".

$ ./oc get machines -n openshift-machine-api -l machine.openshift.io/cluster-api-machine-role=master
NAME                                     PHASE      TYPE   REGION   ZONE   AGE
rugouvei-cluster1-bkw7w-master-1         Deleting                          5h9m
rugouvei-cluster1-bkw7w-master-2         Running                           5h9m
rugouvei-cluster1-bkw7w-master-gx9pd-0   Running                           34m
rugouvei-cluster1-bkw7w-master-kxp8p-1   Running                           34m

$ ./oc get nodes -l node-role.kubernetes.io/master=
NAME                                     STATUS   ROLES                  AGE    VERSION
rugouvei-cluster1-bkw7w-master-1         Ready    control-plane,master   5h7m   v1.26.9+636f2be
rugouvei-cluster1-bkw7w-master-2         Ready    control-plane,master   5h7m   v1.26.9+636f2be
rugouvei-cluster1-bkw7w-master-gx9pd-0   Ready    control-plane,master   30m    v1.26.9+636f2be
rugouvei-cluster1-bkw7w-master-kxp8p-1   Ready    control-plane,master   30m    v1.26.9+636f2be

$ ./oc -n openshift-etcd get pod -l app=etcd --show-labels
NAME                                          READY   STATUS                  RESTARTS         AGE   LABELS
etcd-rugouvei-cluster1-bkw7w-master-1         4/4     Running                 0                61m   app=etcd,etcd=true,k8s-app=etcd,revision=9
etcd-rugouvei-cluster1-bkw7w-master-2         4/4     Running                 0                63m   app=etcd,etcd=true,k8s-app=etcd,revision=9
etcd-rugouvei-cluster1-bkw7w-master-kxp8p-1   0/4     Init:CrashLoopBackOff   10 (2m42s ago)   29m   app=etcd,etcd=true,k8s-app=etcd,revision=9    

New masters to roll-out.