ansible/roles/prometheus/templates/prometheus/alert.rules.yml.j2

# {{ ansible_managed }}
{# As this is also Jinja2 it will conflict without a raw block #}
{# Depending of Prometheus Node exporter version, rules can change depending of version #}
{% raw %}
groups:
- name: alert.rules
  rules:

  # Alert for any instance that is unreachable for >3 minutes.
  - alert: InstanceDown
    expr: up == 0
    for: 3m
    labels:
      severity: critical
    annotations:
      summary: "{{ $labels.instance }} est invisible depuis plus de 3 minutes !"

  # Alert for out of memory
  - alert: OutOfMemory
    expr: (node_memory_MemFree_bytes + node_memory_Cached_bytes + node_memory_Buffers_bytes) / node_memory_MemTotal_bytes * 100 < 10
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "Mémoire libre de {{ $labels.instance }} à {{ $value | printf "%.2f" }}%."

  # Alert for out of disk space
  - alert: OutOfDiskSpace
    expr: node_filesystem_free_bytes{fstype="ext4"} / node_filesystem_size_bytes{fstype="ext4"} * 100 < 10
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "Espace libre de {{ $labels.mountpoint }} sur {{ $labels.instance }} à {{ $value | printf "%.2f" }}%."

  # Alert for out of inode space on disk
  - alert: OutOfInodes
    expr: node_filesystem_files_free{fstype="ext4"} / node_filesystem_files{fstype="ext4"} * 100 < 10
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "Presque plus d'inodes disponibles ({{ $value }}% restant) dans {{ $labels.mountpoint }} sur {{ $labels.instance }}."

  # Alert for high CPU usage
  - alert: CpuUsage
    expr: (100 - avg by (instance) (irate(node_cpu_seconds_total{mode="idle"}[5m])) * 100) > 75
    for: 10m
    labels:
      severity: warning
    annotations:
      summary: "CPU sur {{ $labels.instance }} à {{ $value | printf "%.2f" }}%."

  # Check systemd unit (> buster)
  - alert: SystemdServiceFailed
    expr: node_systemd_unit_state{state="failed"} == 1
    for: 10m
    labels:
      severity: warning
    annotations:
      summary: "{{ $labels.name }} a échoué sur {{ $labels.instance }}"
  
  # Check UPS
  - alert: UpsOutputSourceChanged
    expr: upsOutputSource != 3
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "La source d'alimentation de {{ $labels.instance }} a changé !"

  - alert: UpsBatteryStatusWarning
    expr: upsBatteryStatus == 3
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "L'état de la batterie de {{ $labels.instance }} est faible !"

  - alert: UpsBatteryStatusCritical
    expr: upsBatteryStatus == 4
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "L'état de la batterie de {{ $labels.instance }} est affaibli !"

  - alert: UpsHighLoad
    expr: upsOutputPercentLoad > 70
    for: 5m
    labels:
      severity: critical
    annotations:
      summary: "La charge de {{ $labels.instance }} est de {{ $value }}% !"    

  - alert: UpsWrongInputVoltage
    expr: (upsInputVoltage < 210) or (upsInputVoltage > 250)
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "La tension d'entrée de {{ $labels.instance }} est de {{ $value }}V."

  - alert: UpsWrongOutputVoltage
    expr: (upsOutputVoltage < 220) or (upsOutputVoltage > 240)
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "La tension de sortie de {{ $labels.instance }} est de {{ $value }}V."  

  - alert: UpsTimeRemainingWarning
    expr: upsEstimatedMinutesRemaining < 15
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "L'autonomie restante sur {{ $labels.instance }} est de {{ $value }} min."

  - alert: UpsTimeRemainingCritical
    expr: upsEstimatedMinutesRemaining < 5
    for: 5m
    labels:
      severity: critical
    annotations:
      summary: "L'autonomie restante sur {{ $labels.instance }} est de {{ $value }} min."


{% endraw %}