OpenShift Container Platform 4 - Node rebooting due to new MachineConfig being roll-out, even though no change was applied

Issue

• We noticed compute nodes are switching in between two rendered MachineConfigs, that were generated on the same day but actually some days ago. The only change that was applied was to doubled the instance size of the master nodes. But that should not trigger such an effect. When checking the MachineConfig it was found that nodeStatusUpdateFrequency is being updated in /etc/kubernetes/kubelet.conf even though that option is not specified in any of the MachineConfig provided.
• a couple days ago, we started seeing workers nodes in one cluster rebooting. This is kind of weird as the last openshift upgrade was rolled out about a month ago and there was no new MachineConfig generated since then.