RFE: For the ability to set up persistent storage for Prometheus

[franMredhat]

Introduction
The customer would like the ability to set up persistent storage for Prometheus

Issue
Configuring Prometheus is not supported in ARO as shown in the Azure Red Hat OpenShift support policy.
Azure Red Hat OpenShift support policy [ https://learn.microsoft.com/en-us/azure/openshift/support-policies-v4#cluster-configuration-requirements ]

Business impact on the customer

• Can not predict reliable Capacity needs like PVC/Disk, CPU or memory.
• They update or replace the Nodes continuously. So they lose continuously metric history and are unable to use trend analysis.
• Unable after a simple pod restart of Prometheus to see anomalies inside our user workload and this disturbs incident handling.

Request / Customers Expectations
The customer would like to have persistent ARO cluster core metrics in the built-in Prometheus, so that I can calculate trends for at least over 60 days.
Acceptance crirteria:

• persistent also when pods are restarting
• out of the box with a standard retention time
• adjustable retention time

[GijsvanDulmen]

Is this on the roadmap already? We are also in need of this!