stacks-instances/otc/edp.buildth.ing/stacks/observability/victoria-k8s-stack/manifests/alerts.yaml

apiVersion: operator.victoriametrics.com/v1beta1
kind: VMRule
metadata:
  name: forgejo-alerts
  namespace: observability
spec:
  groups:
    - name: forgejo
      rules:
        - alert: forgejo down
          expr: sum by(cluster_environment) (up{pod=~"forgejo-server-.*"}) < 1
          for: 30s
          labels:
            severity: critical
            job:  "{{ $labels.job }}"
          annotations:
            value: "{{ $value }}"
            description: 'forgejo is down in cluster environment {{ $labels.cluster_environment }}'
    - name: forgejo-backup
      rules:
        - alert: forgejo s3 backup job failed
          expr: max by(cluster_environment) (kube_job_status_failed{job_name=~"forgejo-s3-backup-.*"}) != 0
          for: 30s
          labels:
            severity: critical
            job:  "{{ $labels.job }}"
          annotations:
            value: "{{ $value }}"
            description: 'forgejo s3 backup job failed in cluster environment {{ $labels.cluster_environment }}'
    - name: disk-consumption-high
      rules:
        - alert: disk consumption high
          expr: 1-(kubelet_volume_stats_available_bytes / kubelet_volume_stats_capacity_bytes) > 0.6
          for: 30s
          labels:
            severity: major
            job:  "{{ $labels.job }}"
          annotations:
            value: "{{ $value }}"
            description: 'disk consumption of pvc {{ $labels.namespace }}/{{ $labels.persistentvolumeclaim }} is high in cluster environment {{ $labels.cluster_environment }}'