Right now CMO deploys telemeter-client that federates metrics from the in-cluster Prometheus and forwards them to the Telemetry service using the remote write protocol. We tried early 2020 to have the in-cluster Prometheus talking directly to the Telemetry service but the change had to be rolled back [1]. A follow-up change [2] introduced a flag that would allow a gradual roll out of native remote write but we never took the time to wrap this up and the CMO code base has changed a lot since then.
It is time to reassess whether the code is still functional.

[1] https://github.com/openshift/cluster-monitoring-operator/pull/615
[2] https://github.com/openshift/cluster-monitoring-operator/pull/617
 

DoD

• Verify that CMO can still talk to the telemetry service using remote write (e.g. when started with "--enabled-remote-write").
• Verify that additional remote write configurations aren't disrupted.