Goal

Test NHC/SNR on Compact Cluster before our customers do.
We should start the testing as soon as the operator is available to us, even before released.

Identify pitfalls that arise in a compact cluster due to differences between control plane and worker nodes.
I.e.

• NHC/SNR can not fence ctl plane nodes today
• Any special considerations for networking between workers and ctl plane nodes?
• Anything to consider for affinity due to the different node pools
• Implications of the different node pools on update flow?
• …

User Stories

• As a RHV Cluster owner I want to run OCP with CNV on a similar BM footprint so that I do not need to get more or more expensive hardware.
• As a RHV cluster owner I want to have HA for VMs on my compact cluster so that I get comparable functionality than RHV
• As a RHV cluster owner I would like to minimize the downtime for any of my VMs in case a node failed
• As a RHV cluster owner I would like to understand the different timeouts I can set, what are "safe" values and what are the risks if selecting timeouts that are lower than the "safe" ones
• As a RHV cluster owner I would like to understand how to calculate the minimal values that are HW dependent 

On a compact cluster with Node Remediation (poison pill) operator installed and shared storage:

• As a VM owner I would like my VM restart on another node within the same amount of time it takes or non-compact cluster in case the node it's running on fails.

Non-Requirements

• List of things not included in this epic, to alleviate any doubt raised during the grooming process.

Notes

• Any additional details or decisions made/needed

Done Checklist