What were you trying to do that didn't work?

During the upgrade of OpenShift from version 4.14.31 to 4.15.22, we have experienced intermittent freezes and crashes across both Windows and Linux Fedora virtual machines.

One particular Fedora VM freezes and displays the message "marking tsc unstable due to clocksource watchdog" in the console. This issue has also been observed on other Fedora VMs that did not freeze. Additionally, some Windows VMs have experienced Event ID 41 reboots.

I have come across a related bug report that appears similar to our situation, though it involves AMD cores my servers use Intel : 

https://bugzilla.redhat.com/show_bug.cgi?id=2125671

Please provide the package NVR for which bug is seen:

sh-5.1# lscpu | grep -i 'vendor id'
Vendor ID:                          GenuineIntel
BIOS Vendor ID:                     Intel
sh-5.1# uname -r
5.14.0-284.75.1.el9_2.x86_64

sh-5.1# cat /etc/redhat-release 
Red Hat Enterprise Linux CoreOS release 4.15

How reproducible:

sometimes 

Steps to reproduce

1. Create an Open Shift system version 4.14.31 with 56 windows and 10 fedoras
2. Upgrade to 4.15.22

Expected results

VMS do not crash or freeze

Actual results

VMS crash and freeze