ServiceMonitor's serverName is wrong when not in the default namespace

Hi,

when you've deployed the operator through the operatorhub on openshift 4.10 in a different namespace than 'group-sync-operator'
prometheus gets a tls error when trying to scrape for metrics.

that is because the servicemonitor get the wrong serverName :

❯ oc -n my-namespace get servicemonitor group-sync-operator-controller-manager-metrics-monitor -o yaml | kubectl-neat
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
labels:
control-plane: controller-manager
name: group-sync-operator-controller-manager-metrics-monitor
namespace: my-namespace
spec:
endpoints:

• bearerTokenFile: /var/run/secrets/kubernetes.io/serviceaccount/token
  interval: 30s
  port: https
  scheme: https
  tlsConfig:
  caFile: /etc/prometheus/configmaps/serving-certs-ca-bundle/service-ca.crt
  serverName: group-sync-operator-controller-manager-metrics-service.group-sync-operator.svc
  selector:
  matchLabels:
  control-plane: controller-manager

seems that the namespace part of the serverName is hardcoded don't you think ?

regards

Ludovic

Start by locating the ServiceMonitor generation that produces group-sync-operator-controller-manager-metrics-monitor and compare it with the YAML shown here. Trace where the serverName namespace is selected; done means it uses the deployed namespace instead of group-sync-operator and Prometheus can scrape the metrics endpoint without a TLS error.