RFC: Add basic serviceMonitor support into library chart about helm-charts HOT 3 CLOSED

Not clear to me, where you would place the release: kube-prometheus-stack type label needed for auto-discovery?

Would this work???

    # Full Options for Service Monitor
    serviceMonitor:
      enabled: true
      endpoints:
        - port: http
          path: /metrics
          interval: 1m
          scrapeTimeout: 10s
      # -- Additional labels for the Kubernetes `ServiceMonitor` object
      labels:
        release: kube-prometheus-stack

Side note - can you also consider creating a podMonitor equivalent? If the pod doesn't require a service in front of it, then no reason to create one just for the sake of using a serviceMonitor.

from helm-charts.

I am guessing you don't need the ServiceMonitor labels for auto-discovery if you have the kps chart configured with the following? I have never had to include those labels on a ServiceMonitor before.

https://github.com/onedr0p/home-ops/blob/66096c2802daf96e48c7843ea1ef23ba76242084/cluster/apps/monitoring/kube-prometheus-stack/helm-release.yaml#L208-L211

PodMonitor could totally be added as an option too, however currently the only way to define ports is by creating a service. Implementing a way to define ports without a service might be challenging but that is an entirely separate issue.

from helm-charts.

Thanks for the reference to the ...SelectorNilUsesHelmValues didn't know about them, will research.

from helm-charts.