cookbooks/prometheus/templates/default/alert_rules.yml.erb

   1 # DO NOT EDIT - This file is being maintained by Chef
   2
   3 groups:
   4   - name: alertmanager
   5     rules:
   6       - alert: prometheus target missing
   7         expr: up == 0
   8         for: 5m
   9         labels:
  10           alertgroup: "prometheus"
  11   - name: hwmon
  12     rules:
  13       - alert: hwmon fan alarm
  14         expr: node_hwmon_fan_alarm == 1
  15         for: 5m
  16         labels:
  17           alertgroup: "{{ $labels.instance }}"
  18         annotations:
  19           fan_rpm: "{{ with printf \"node_hwmon_fan_rpm{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}rpm{{end}}"
  20           fan_min_rpm: "{{ with printf \"node_hwmon_fan_min_rpm{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}rpm{{end}}"
  21       - alert: hwmon temperature alarm
  22         expr: node_hwmon_temp_alarm == 1
  23         for: 5m
  24         labels:
  25           alertgroup: "{{ $labels.instance }}"
  26         annotations:
  27           temp_celsius: "{{ with printf \"node_hwmon_temp_celsius{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}C{{end}}"
  28           temp_max_celsius: "{{ with printf \"node_hwmon_temp_max_celsius{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}C{{end}}"
  29           temp_crit_celsius: "{{ with printf \"node_hwmon_temp_crit_celsius{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}C{{end}}"
  30       - alert: hwmon voltage alarm
  31         expr: node_hwmon_in_alarm == 1
  32         for: 5m
  33         labels:
  34           alertgroup: "{{ $labels.instance }}"
  35         annotations:
  36           in_volts: "{{ with printf \"node_hwmon_in_volts{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}V{{end}}"
  37           in_min_volts: "{{ with printf \"node_hwmon_in_min_volts{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}V{{end}}"
  38           in_max_volts: "{{ with printf \"node_hwmon_in_max_volts{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}V{{end}}"
  39   - name: ipmi
  40     rules:
  41       - alert: ipmi fan alarm
  42         expr: ipmi_fan_speed_state > 0
  43         for: 5m
  44         labels:
  45           alertgroup: "{{ $labels.instance }}"
  46         annotations:
  47           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
  48       - alert: ipmi temperature alarm
  49         expr: ipmi_temperature_state > 0
  50         for: 5m
  51         labels:
  52           alertgroup: "{{ $labels.instance }}"
  53         annotations:
  54           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
  55       - alert: ipmi voltage alarm
  56         expr: ipmi_voltage_state > 0
  57         for: 5m
  58         labels:
  59           alertgroup: "{{ $labels.instance }}"
  60         annotations:
  61           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
  62   - name: mdadm
  63     rules:
  64       - alert: mdadm array inactive
  65         expr: node_md_state{state="inactive"} > 0
  66         for: 0m
  67         labels:
  68           alertgroup: "{{ $labels.instance }}"
  69         annotations:
  70           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  71           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  72           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  73           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  74       - alert: mdadm disk failed
  75         expr: node_md_disks{state="failed"} > 0
  76         for: 0m
  77         labels:
  78           alertgroup: "{{ $labels.instance }}"
  79         annotations:
  80           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  81           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  82           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  83           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  84   - name: memory
  85     rules:
  86       - alert: low memory
  87         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes * 100 < 10
  88         for: 5m
  89         labels:
  90           alertgroup: "{{ $labels.instance }}"
  91         annotations:
  92           memory_free: "{{ $value }}%"
  93       - alert: memory pressure
  94         expr: rate(node_vmstat_pgmajfault[1m]) > 1000
  95         for: 5m
  96         labels:
  97           alertgroup: "{{ $labels.instance }}"
  98         annotations:
  99           major_page_faults: "{{ $value }} faults/s"
 100       - alert: oom kill detected
 101         expr: increase(node_vmstat_oom_kill[1m]) > 0
 102         for: 0m
 103         labels:
 104           alertgroup: "{{ $labels.instance }}"
 105         annotations:
 106           new_oom_kills: "{{ $value }}"