We want UWM users to be able to leverage the Alertmanager running in the openshift-monitoring namespace. They configure their alert routing and notification settings by deploying AlertmanagerConfig resources in their namespace(s).

The prometheus operator running in openshift-monitoring reconciles the user-defined AlertmanagerConfig resources and updates the Alertmanager configuration accordingly.

From https://github.com/openshift/enhancements/pull/937, the CMO configmap should be extended:

apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |-
    enableUserWorkload: true
    alertmanagerMain:
      enableUserAlertmanagerConfig: true

DoD

• AlertmanagerConfig resources from user namespaces (without the “openshift.io/cluster-monitoring: true” and “openshift.io/user-monitoring: false” labels) are reconciled in the Alertmanager configuration.