Customers are unable to scale-up the OCP nodes when the initial setup is done with OCP 4.8/4.9 and then upgraded to 4.15.22/4.15.23

At first customer observed that the node scale-up failed and the /etc/resolv.conf was empty on the nodes.
As a workaround, customer copy/paste the resolv.conf content from a correct resolv.conf and then it continued with setting up the new node.

However then they observed the rendered MachineConfig assembled with the 00-worker, and suspected that something can be wrong with the on-prem-resolv-prepender.service service definition.
As a workaround, customer manually changed this service definition which helped them to scale up new nodes.

4.15 , 4.16

100%

1. Install OCP vSphere IPI cluster version 4.8 or 4.9
2. Check "on-prem-resolv-prepender.service" service definition
3. Upgrade it to 4.15.22 or 4.15.23
4. Check if the node scaling is working 
5. Check "on-prem-resolv-prepender.service" service definition     

Unable to scaleup node with default service definition. After manually making changes in the service definition , scaling is working.

Node sclaing should work without making any manual changes in the service definition.

on-prem-resolv-prepender.service content on the clusters build with 4.8 / 4.9 version and then upgraded to 4.15.22 / 4.25.23 :
~~~
[Unit]
Description=Populates resolv.conf according to on-prem IPI needs
# Per https://issues.redhat.com/browse/OCPBUGS-27162 there is a problem if this is started before crio-wipe
After=crio-wipe.service
[Service]
Type=oneshot
Restart=on-failure
RestartSec=10
StartLimitIntervalSec=0
ExecStart=/usr/local/bin/resolv-prepender.sh
EnvironmentFile=/run/resolv-prepender/env
~~~

After manually correcting the service definition as below, scaling works on 4.15.22 / 4.15.23 :
~~~
[Unit]
Description=Populates resolv.conf according to on-prem IPI needs
# Per https://issues.redhat.com/browse/OCPBUGS-27162 there is a problem if this is started before crio-wipe
After=crio-wipe.service
StartLimitIntervalSec=0                -----------> this
[Service]
Type=oneshot
#Restart=on-failure                    -----------> this
RestartSec=10
ExecStart=/usr/local/bin/resolv-prepender.sh
EnvironmentFile=/run/resolv-prepender/env
~~~

Below is the on-prem-resolv-prepender.service on a freshly intsalled 4.15.23 where sclaing is working fine :
~~~
[Unit]
Description=Populates resolv.conf according to on-prem IPI needs
# Per https://issues.redhat.com/browse/OCPBUGS-27162 there is a problem if this is started before crio-wipe
After=crio-wipe.service
StartLimitIntervalSec=0
[Service]
Type=oneshot
Restart=on-failure
RestartSec=10
ExecStart=/usr/local/bin/resolv-prepender.sh
EnvironmentFile=/run/resolv-prepender/env
~~~

Observed this in the rendered MachineConfig which is assembled with the 00-worker