1. Proposed title of this feature request
Support for scaling KEDA operator with CMA to 2 to achieve HA

2. What is the nature and description of the request?
There is a potential downtime with just one replica (higher probability). If we want to use the CMA for critical production services the scaling is mandatory all the time. If it is not and we rely on the CMA client requests can potentially not be fulfilled.

3. Why does the customer need this? (List the business requirements here)
There is impact for company reputation and also SLAs.

4. List any affected packages or components.
OpenShift Container Platform 4 - Custom Metrics Autoscaler