@jaideepr97 A few questions:

• Do we have any possible installation failures and underlying causes or symptoms for the same? If yes, can we have troubleshooting tips for them?
• If not installed properly, what status(es) are available to see for the user? Please provide doc inputs with the description of the statues (es). Also, provide information on what action the users must take for each of them.

Thanks!

@jaideepr97 I see that the user has to just select/enable the OpenShift monitoring option while installing the GitOps Operator in the openshift-gitops-operator namespace.

• Do we need any documentation or information like a concept/overview, or anything that we need in the Downstream documentation from the user's perspective, about monitoring metrics?

@jaideepr97 Can you please respond to my queries here? Thanks!

@Srivaralakshmi I don't know what the best place to put this info is so I'm just putting it here :

Installation instructions to be able to use this feature:

• fresh installations of the operator will see the option to choose the namespace to install the operator in, with the recommended option already selected as openshift-gitops-operator namespace
• users must select the check box saying Enable Operator recommended cluster monitoring on this Namespace to enable cluster monitoring

Usage instructions:

• Once operator is installed, users can leverage the Observe > metrics page in the admin view to query the metrics exposed by the operator
• At present the new operator performance metrics exposed are:
• active_argocd_instances_total [Guage] - This metric produces the graph that tracks the total number of active argo-cd instances being managed by the operator at a given time
• active_argocd_instances_by_phase{phase=<phase>} [Guage] - This metric produces the graph that tracks the count of active Argo CD instances by their phase [Available/Pending/Failed/unknown]
• active_argocd_instance_reconciliation_count{namespace="<argocd-instance-ns>"} [Counter] - This metric produces the graph that tracks total number of reconciliations that have occurred for the instance in the given namespace at any given point in time
• controller_runtime_reconcile_time_seconds_per_instance_bucket{namespace="<argocd-instance-ns>",le="0.5"} [Histogram]- This metric tracks the number of reconciliations that took under 0.5s to complete for a given instance

Any of these metrics can be queried against prometheus as described above, or can be used for more complex queries as is true with any other promql queries