Create ServiceMonitor for operator metrics programmatically #3371
Conversation
...nifests/opentelemetry-operator-prometheus-rules_monitoring.coreos.com_v1_prometheusrule.yaml
Show resolved
Hide resolved
|
@iblancasa I tested the PR with operator deployed in non-standard namespaces and enabling the in-cluster monitoring by adding label openshift.io/cluster-monitoring="true". The metrics are scraped by the in-cluster monitoring as well as the targets are shown up for the servicemonitors that are created. We already have added test case to check the metrics exported by prometheus exporter and using the user workload monitoring stack. I'll add the operator monitoring as well to that test case and verify the metrics using the in-cluster monitoring stack. |
|
Thanks @IshwarKanse ! |
...openshift/manifests/opentelemetry-operator-prometheus_rbac.authorization.k8s.io_v1_role.yaml
Outdated
Show resolved
Hide resolved
| tlsConfig = &monitoringv1.TLSConfig{ | ||
| SafeTLSConfig: monitoringv1.SafeTLSConfig{ | ||
| // kube-rbac-proxy uses a self-signed cert by default | ||
| InsecureSkipVerify: &t, |
There was a problem hiding this comment.
I think we talked about this internally as a bug that we should not be using the InsecureSkipVerify it is also not present here https://github.com/os-observability/konflux-opentelemetry/blob/main/bundle-patch/manifests/opentelemetry-operator-metrics-monitor_monitoring.coreos.com_v1_servicemonitor.yaml#L22
There was a problem hiding this comment.
If you review the code, you will see this will be only executed if the ConfigMap containing the CA doesn´t exist. This covers non-OpenShift scenarios.
Signed-off-by: Israel Blancas <[email protected]>
Signed-off-by: Israel Blancas <[email protected]>
Signed-off-by: Israel Blancas <[email protected]>
|
@pavolloffay anything else? |
Link to tracking Issue(s):