Abstract

One of the original expectations for the observability-operator project was that it would allow the deployment of monitoring stacks (so prometheus et.al.) that can rpvode newer features then the in-cluster stack of a particular OCP version supports.
In essence this request means that OBO and its operants should be independent of the monitoring related CRDs, that OCP installs

Problem

Since CRDs are cluster-scoped, this can only be achieved by not using the CMO supplied CRD scope, i.e. the various CRDs that are created in the monitoring.coreos.com group.

Goal

This epic gather all effort that goes to exploring solutions and tracking the chosen implementation.