OKD 4.7: Nodes not boot: clocksource: Switched to clocksource tsc

[ToibayevNariman]

Describe the bug

Cluster: 3-master, 3 worker nodes
Version: 4.7.0-0.okd-2021-03-07-090821
Method: UPI
Platform: vSphere
VM created from vSphere template: fedora-coreos-33.20210217.ova

At some point, two cluster nodes (1-master,1-worker) became unavailable. The SSH connection attempt was unsuccessful. In vSphere I see the OS is not boot. OS does not receive an IP address from a DHCP server

[vrutkovs]

Please attach log bundle