1. Proposed title of this feature request

kube-state-metrics scraping for observability operator

2. What is the nature and description of the request?

Customer has an external endpoint that they send metrics to. The external receiver is resource limited and they need to be restrictive about what projects send metrics. This is currently being achieved with the following two-pronged solution:

• Using remoteWrite on the Cluster Monitoring stack and using namespace filters to allow or deny certain namespaces
• Using Cluster Observability Operator to allow users to "opt-in" sending their custom application metrics using remoteWrite

However, COO does not scrape default metrics like CPU and Memory which are exposed by kube-state-metrics. While it is possible to scrape KSM using this method this requires cluster-monitoring-view clusterrole which should not be given to users in a multitenant environment. Thus there is no real way for end users to "opt-in" to scraping CPU/Memory with COO and requires manual intervention by administration team for each new namespace.

Please let me know if I missed anything or another way to do this out-of-the-box!

3. Why does the customer need this? (List the business requirements here)

Flexibility for users to capture their own basic metrics.

4. List any affected packages or components.

Cluster Observability Operator. NOTE: I didnt see this as a component option, so I chose Monitoring, let me know if this is the wrong component!