groups:
- name: kube-state-metrics
  rules:
  - alert: KubeStateMetricsListErrors
    annotations:
      message: kube-state-metrics is experiencing errors at an elevated rate in
        list operations. This is likely causing it to not be able to expose metrics
        about Kubernetes objects correctly or at all.
      runbook_url: https://github.com/kubernetes-monitoring/kubernetes-mixin/tree/master/runbook.md#alert-name-kubestatemetricslisterrors
    expr: |
      (sum(rate(kube_state_metrics_list_total{job="kube-state-metrics",result="error"}[5m]))
        /
      sum(rate(kube_state_metrics_list_total{job="kube-state-metrics"}[5m])))
      > 0.01
    for: 15m
    labels:
      component: kubernetes
      oncall_team: dedicated_service
      severity: major
      response_class: ticket
  - alert: KubeStateMetricsWatchErrors
    annotations:
      message: kube-state-metrics is experiencing errors at an elevated rate in
        watch operations. This is likely causing it to not be able to expose metrics
        about Kubernetes objects correctly or at all.
      runbook_url: https://github.com/kubernetes-monitoring/kubernetes-mixin/tree/master/runbook.md#alert-name-kubestatemetricswatcherrors
    expr: |
      (sum(rate(kube_state_metrics_watch_total{job="kube-state-metrics",result="error"}[5m]))
        /
      sum(rate(kube_state_metrics_watch_total{job="kube-state-metrics"}[5m])))
      > 0.01
    for: 15m
    labels:
      component: kubernetes
      oncall_team: dedicated_service
      severity: major
      response_class: ticket
  - alert: KubeStateMetricsDown
    expr: up{job="kube-state-metrics"} == 0
    for: 15m
    labels:
      component: kubernetes
      oncall_team: dedicated_service
      service: kube-state-metrics
      severity: critical
      response_class: page
    annotations:
      description: kube-state-metrics is down