apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  name: k8up
  labels:
    prometheus: k8s
    role: alert-rules
spec:
  groups:
  - name: k8up.rules
    rules:
    - alert: baas_last_errors
      expr: baas_backup_restic_last_errors > 0
      for: 1m
      labels:
        severity: critical
      annotations:
        summary: Amount of errors of last restic backup
        description: This alert is fired when error number is > 0
    - alert: K8upBackupFailed
      expr: rate(k8up_jobs_failed_counter[1d]) > 0
      for: 1m
      labels:
        severity: critical
      annotations:
        summary: "Job in {{ $labels.namespace }} of type {{ $labels.jobType }} failed"
    - alert: K8upBackupNotRunning
      expr: sum(rate(k8up_jobs_total[25h])) == 0 and on(namespace) k8up_schedules_gauge > 0
      for: 1m
      labels:
        severity: critical
      annotations:
        summary: "No K8up jobs were run in {{ $labels.namespace }} within the last 24 hours. Check the operator, there might be a deadlock"
    - alert: K8upJobStuck
      expr: k8up_jobs_queued_gauge{jobType="backup"} > 0 and on(namespace) k8up_schedules_gauge > 0
      for: 24h
      labels:
        severity: critical
      annotations:
        summary: "K8up jobs are stuck in {{ $labels.namespace }} for the last 24 hours."