EKSA Plugin Upgrade to Cluster leaves one of the ETCD Machine in Provisioned state and Control Plane in Scaling State

[rammarepally]

What happened:
Applying anywhere plugin update (v0.16.5) to EKSA Cluster(vSphere) leaves one of the etcd machine in provisioned state and also has etcdadm cluster point to old etcd IPs.Similarly the CP points to the old etcd IPs and on few occassions has KubeadmControlplane in Scaling state .

What you expected to happen:
New ETCD Machines to be in running state and CP pointing to the new ETCD IPs.

How to reproduce it (as minimally and precisely as possible):
NA

Anything else we need to know?:
Deleted etcd machines in provisioned state multiple times but still see the same issue.
Raised a support case but no luck with resolving the issue yet.

Environment:

• EKS Anywhere Release: v0.16.5
• EKS Distro Release:v1.24.12-eks-76dc719