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: apache
  12     rules:
  13       - alert: apache down
  14         expr: apache_up == 0
  15         for: 5m
  16         labels:
  17           alertgroup: "{{ $labels.instance }}"
  18       - alert: apache workers busy
  19         expr: sum(apache_workers{state="busy"}) by (instance) / sum(apache_scoreboard) by (instance) > 0.8
  20         for: 5m
  21         labels:
  22           alertgroup: "{{ $labels.instance }}"
  23         annotations:
  24           busy_workers: "{{ $value | humanizePercentage }}"
  25   - name: database
  26     rules:
  27       - alert: postgres replication delay
  28         expr: pg_replication_lag_seconds > 5
  29         for: 5m
  30         labels:
  31           alertgroup: database
  32         annotations:
  33           delay: "{{ $value | humanizeDuration }}"
  34   - name: hwmon
  35     rules:
  36       - alert: hwmon fan alarm
  37         expr: node_hwmon_fan_alarm == 1
  38         for: 5m
  39         labels:
  40           alertgroup: "{{ $labels.instance }}"
  41         annotations:
  42           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}}"
  43           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}}"
  44       - alert: hwmon temperature alarm
  45         expr: node_hwmon_temp_alarm == 1
  46         for: 5m
  47         labels:
  48           alertgroup: "{{ $labels.instance }}"
  49         annotations:
  50           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}}"
  51           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}}"
  52           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}}"
  53       - alert: hwmon voltage alarm
  54         expr: node_hwmon_in_alarm == 1
  55         for: 5m
  56         labels:
  57           alertgroup: "{{ $labels.instance }}"
  58         annotations:
  59           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}}"
  60           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}}"
  61           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}}"
  62   - name: ipmi
  63     rules:
  64       - alert: ipmi fan alarm
  65         expr: ipmi_fan_speed_state > 0
  66         for: 5m
  67         labels:
  68           alertgroup: "{{ $labels.instance }}"
  69         annotations:
  70           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
  71       - alert: ipmi temperature alarm
  72         expr: ipmi_temperature_state > 0
  73         for: 5m
  74         labels:
  75           alertgroup: "{{ $labels.instance }}"
  76         annotations:
  77           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
  78       - alert: ipmi voltage alarm
  79         expr: ipmi_voltage_state > 0
  80         for: 5m
  81         labels:
  82           alertgroup: "{{ $labels.instance }}"
  83         annotations:
  84           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
  85   - name: mdadm
  86     rules:
  87       - alert: mdadm array inactive
  88         expr: node_md_state{state="inactive"} > 0
  89         for: 0m
  90         labels:
  91           alertgroup: "{{ $labels.instance }}"
  92         annotations:
  93           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  94           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  95           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  96           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  97       - alert: mdadm disk failed
  98         expr: node_md_disks{state="failed"} > 0
  99         for: 0m
 100         labels:
 101           alertgroup: "{{ $labels.instance }}"
 102         annotations:
 103           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 104           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 105           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 106           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 107   - name: memory
 108     rules:
 109       - alert: low memory
 110         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 111         for: 5m
 112         labels:
 113           alertgroup: "{{ $labels.instance }}"
 114         annotations:
 115           memory_free: "{{ $value | humanizePercentage }}"
 116       - alert: memory pressure
 117         expr: rate(node_vmstat_pgmajfault[1m]) > 1000
 118         for: 5m
 119         labels:
 120           alertgroup: "{{ $labels.instance }}"
 121         annotations:
 122           major_page_faults: "{{ $value }} faults/s"
 123       - alert: oom kill detected
 124         expr: increase(node_vmstat_oom_kill[1m]) > 0
 125         for: 0m
 126         labels:
 127           alertgroup: "{{ $labels.instance }}"
 128         annotations:
 129           new_oom_kills: "{{ $value }}"
 130   - name: network
 131     rules:
 132       - alert: interface transmit rate
 133         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 134         for: 5m
 135         labels:
 136           alertgroup: "{{ $labels.instance }}"
 137         annotations:
 138           bandwidth_used: "{{ $value | humanizePercentage }}"
 139       - alert: interface receive rate
 140         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.98
 141         for: 5m
 142         labels:
 143           alertgroup: "{{ $labels.instance }}"
 144         annotations:
 145           bandwidth_used: "{{ $value | humanizePercentage }}"
 146       - alert: interface transmit errors
 147         expr: rate(node_network_transmit_errs_total[1m]) / rate(node_network_transmit_packets_total[1m]) > 0.01
 148         for: 5m
 149         labels:
 150           alertgroup: "{{ $labels.instance }}"
 151         annotations:
 152           error_rate: "{{ $value | humanizePercentage }}"
 153       - alert: interface receive errors
 154         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 155         for: 5m
 156         labels:
 157           alertgroup: "{{ $labels.instance }}"
 158         annotations:
 159           error_rate: "{{ $value | humanizePercentage }}"
 160       - alert: conntrack entries
 161         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 162         for: 5m
 163         labels:
 164           alertgroup: "{{ $labels.instance }}"
 165         annotations:
 166           entries_used: "{{ $value | humanizePercentage }}"
 167   - name: postgresql
 168     rules:
 169       - alert: postgresql down
 170         expr: pg_up == 0
 171         for: 1m
 172         labels:
 173           alertgroup: "{{ $labels.instance }}"
 174       - alert: postgresql replication delay
 175         expr: pg_replication_lag_seconds > 5
 176         for: 1m
 177         labels:
 178           alertgroup: "{{ $labels.instance }}"
 179         annotations:
 180           delay: "{{ $value | humanizeDuration }}"
 181       - alert: postgresql connection limit
 182         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 183         for: 1m
 184         labels:
 185           alertgroup: "{{ $labels.instance }}"
 186         annotations:
 187           connections_used: "{{ $value | humanizePercentage }}"
 188       - alert: postgresql deadlocks
 189         expr: increase(pg_stat_database_deadlocks[1m]) > 5
 190         for: 0m
 191         labels:
 192           alertgroup: "{{ $labels.instance }}"
 193         annotations:
 194           new_deadlocks: "{{ $value }}"
 195       - alert: postgresql slow queries
 196         expr: pg_slow_queries > 0
 197         for: 5m
 198         labels:
 199           alertgroup: "{{ $labels.instance }}"
 200         annotations:
 201           queries: "{{ $value }}"
 202   - name: ssl
 203     rules:
 204       - alert: ssl certificate probe failed
 205         expr: ssl_probe_success == 0
 206         for: 60m
 207         labels:
 208           alertgroup: ssl
 209       - alert: ssl certificate expiry
 210         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 211         for: 0m
 212         labels:
 213           alertgroup: ssl
 214         annotations:
 215           expires_in: "{{ $value | humanizeDuration }}"
 216       - alert: ssl certificate revoked
 217         expr: ssl_ocsp_response_status == 1
 218         for: 0m
 219         labels:
 220           alertgroup: ssl
 221       - alert: ocsp status unknown
 222         expr: ssl_ocsp_response_status == 1
 223         for: 0m
 224         labels:
 225           alertgroup: ssl
 226   - name: tile
 227     rules:
 228       - alert: renderd replication delay
 229         expr: renderd_replication_delay > 120
 230         for: 5m
 231         labels:
 232           alertgroup: tile
 233         annotations:
 234           delay: "{{ $value | humanizeDuration }}"
 235       - alert: missed tile rate
 236         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 237         for: 5m
 238         labels:
 239           alertgroup: tile
 240         annotations:
 241           miss_rate: "{{ $value | humanizePercentage }}"
 242   - name: time
 243     rules:
 244       - alert: clock not synchronising
 245         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 246         for: 5m
 247         labels:
 248           alertgroup: "{{ $labels.instance }}"
 249       - alert: clock skew detected
 250         expr: (node_timex_offset_seconds > 0.05 and deriv(node_timex_offset_seconds[5m]) >= 0) or (node_timex_offset_seconds < -0.05 and deriv(node_timex_offset_seconds[5m]) <= 0)
 251         for: 5m
 252         labels:
 253           alertgroup: "{{ $labels.instance }}"
 254         annotations:
 255           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"