PromRules

apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  labels:
    release: monitoring
  name: daniel-custom-rules-prod
  namespace: monitoring
spec:
  groups:
    - name: "pod-error-rule"
      rules:
        - alert: PodErrorAlert
          for: 3m
          expr: count(kube_pod_container_status_terminated_reason{reason='Error'} > 0) by (namespace, pod)
          labels:
            severity: critical
          annotations:
            summary: Pod has been in error state for more than 10 minutes.
            message: "pod {{ $labels.pod }} in the {{ $labels.namespace }} namespace has been in error state for more than 10m"
    - name: "pod-pending-rule"
      rules:
        - alert: PodPendingFor30mAlert
          for: 2m
          expr: count(kube_pod_status_phase{phase="Pending"} > 0) by (namespace, pod)
          labels:
            severity: critical
          annotations:
            summary: Pod has been pending for more than 30 minutes.
            message: "pod {{ $labels.pod }} in the {{ $labels.namespace }} namespace has been in pending state for more than 30m"