After executing the replace unhealthy master node procedure[1], the new node is unable to finish the boot process. It get stuck after activating the main nic. 

[1] https://docs.openshift.com/container-platform/4.12/backup_and_restore/control_plane_backup_and_restore/replacing-unhealthy-etcd-member.html#restore-replace-stopped-baremetal-etcd-member_replacing-unhealthy-etcd-member

4.12.z

The problem has been detected when a cluster started in 4.5 version and has been upgraded to 4.12 version. Then, the replace an unhealthy master node is executed where the new node is created and started, but right after the nic is activated, it get stuck and never finish the boot process. Even if rebooted, the outcome is the same, it gets stuck in the very same stage. 

1. Deploy an Openshift 4.5 cluster deployed as IPI over a Openstack provider. 
2. Upgrade the cluster to 4.12.z
3. Execute the replace an unhealthy master node procedure[1]. 
4. After the step, creating a machine object from yaml file, the node is created, it obtains the ignition config  and starts booting, however after the nic is activated, the boot process gets stuck and never finishes. 

New master node gets stuck in the boot process, it never boots completely. 

New master node should start correctly