ETCD error of dataDir has been destroyed and must be removed in OpenShift Container Platform 4 cluster.

Issue

• While replacing an unhealthy master node, etcd is degraded with following error:

ETCD POD Logs:
2022-06-06T04:55:19.695291547Z #### attempt 0
2022-06-06T04:55:19.696317283Z       member={name="master02.ocp-dev.example.com", peerURLs=[https://192.168.0.2:2380}, clientURLs=[https://192.168.0.2:2379]
2022-06-06T04:55:19.696317283Z       member={name="master03.ocp-dev.example.com", peerURLs=[https://192.168.0.3:2380}, clientURLs=[https://192.168.0.3:2379]
2022-06-06T04:55:19.696317283Z       member={name="master01.ocp-dev.example.com", peerURLs=[https://192.168.0.1:2380}, clientURLs=[https://192.168.0.1:2379]
2022-06-06T04:55:19.696335095Z       target={name="master02.ocp-dev.example.com", peerURLs=[https://192.168.0.2:2380}, clientURLs=[https://192.168.0.2:2379]
2022-06-06T04:55:19.696368271Z member "https://192.168.0.2:2380" dataDir has been destroyed and must be removed from the cluster

• ETCD Pod failed to start and getting restarted multiple times:

etcd-master01.ocp-dev.example.com                 4/4    Running    0         5d
etcd-master02.ocp-dev.example.com                 3/4    Running    553       1d    <---- node which has been replaced recently.
etcd-master03.ocp-dev.example.com                 4/4    Running    0         5d
etcd-quorum-guard-c97444699-kqw72                           0/1    Running    0         1d
etcd-quorum-guard-c97444699-lkjmm                           1/1    Running    0         1d
etcd-quorum-guard-c97444699-zsq4c                           1/1    Running    0         1d