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       - alert: ipmi power alarm
  86         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
  87         for: 5m
  88         labels:
  89           alertgroup: "{{ $labels.instance }}"
  90   - name: mdadm
  91     rules:
  92       - alert: mdadm array inactive
  93         expr: node_md_state{state="inactive"} > 0
  94         for: 0m
  95         labels:
  96           alertgroup: "{{ $labels.instance }}"
  97         annotations:
  98           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
  99           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 100           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 101           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 102       - alert: mdadm disk failed
 103         expr: node_md_disks{state="failed"} > 0
 104         for: 0m
 105         labels:
 106           alertgroup: "{{ $labels.instance }}"
 107         annotations:
 108           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 109           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 110           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 111           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 112   - name: memory
 113     rules:
 114       - alert: low memory
 115         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 116         for: 5m
 117         labels:
 118           alertgroup: "{{ $labels.instance }}"
 119         annotations:
 120           memory_free: "{{ $value | humanizePercentage }}"
 121       - alert: memory pressure
 122         expr: rate(node_vmstat_pgmajfault[1m]) > 1000
 123         for: 5m
 124         labels:
 125           alertgroup: "{{ $labels.instance }}"
 126         annotations:
 127           major_page_faults: "{{ $value }} faults/s"
 128       - alert: oom kill detected
 129         expr: increase(node_vmstat_oom_kill[1m]) > 0
 130         for: 0m
 131         labels:
 132           alertgroup: "{{ $labels.instance }}"
 133         annotations:
 134           new_oom_kills: "{{ $value }}"
 135   - name: network
 136     rules:
 137       - alert: interface transmit rate
 138         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 139         for: 5m
 140         labels:
 141           alertgroup: "{{ $labels.instance }}"
 142         annotations:
 143           bandwidth_used: "{{ $value | humanizePercentage }}"
 144       - alert: interface receive rate
 145         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.98
 146         for: 5m
 147         labels:
 148           alertgroup: "{{ $labels.instance }}"
 149         annotations:
 150           bandwidth_used: "{{ $value | humanizePercentage }}"
 151       - alert: interface transmit errors
 152         expr: rate(node_network_transmit_errs_total[1m]) / rate(node_network_transmit_packets_total[1m]) > 0.01
 153         for: 5m
 154         labels:
 155           alertgroup: "{{ $labels.instance }}"
 156         annotations:
 157           error_rate: "{{ $value | humanizePercentage }}"
 158       - alert: interface receive errors
 159         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 160         for: 5m
 161         labels:
 162           alertgroup: "{{ $labels.instance }}"
 163         annotations:
 164           error_rate: "{{ $value | humanizePercentage }}"
 165       - alert: conntrack entries
 166         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 167         for: 5m
 168         labels:
 169           alertgroup: "{{ $labels.instance }}"
 170         annotations:
 171           entries_used: "{{ $value | humanizePercentage }}"
 172   - name: postgresql
 173     rules:
 174       - alert: postgresql down
 175         expr: pg_up == 0
 176         for: 1m
 177         labels:
 178           alertgroup: "{{ $labels.instance }}"
 179       - alert: postgresql replication delay
 180         expr: pg_replication_lag_seconds > 5
 181         for: 1m
 182         labels:
 183           alertgroup: "{{ $labels.instance }}"
 184         annotations:
 185           delay: "{{ $value | humanizeDuration }}"
 186       - alert: postgresql connection limit
 187         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 188         for: 1m
 189         labels:
 190           alertgroup: "{{ $labels.instance }}"
 191         annotations:
 192           connections_used: "{{ $value | humanizePercentage }}"
 193       - alert: postgresql deadlocks
 194         expr: increase(pg_stat_database_deadlocks[1m]) > 5
 195         for: 0m
 196         labels:
 197           alertgroup: "{{ $labels.instance }}"
 198         annotations:
 199           new_deadlocks: "{{ $value }}"
 200       - alert: postgresql slow queries
 201         expr: pg_slow_queries > 0
 202         for: 5m
 203         labels:
 204           alertgroup: "{{ $labels.instance }}"
 205         annotations:
 206           queries: "{{ $value }}"
 207   - name: ssl
 208     rules:
 209       - alert: ssl certificate probe failed
 210         expr: ssl_probe_success == 0
 211         for: 60m
 212         labels:
 213           alertgroup: ssl
 214       - alert: ssl certificate expiry
 215         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 216         for: 0m
 217         labels:
 218           alertgroup: ssl
 219         annotations:
 220           expires_in: "{{ $value | humanizeDuration }}"
 221       - alert: ssl certificate revoked
 222         expr: ssl_ocsp_response_status == 1
 223         for: 0m
 224         labels:
 225           alertgroup: ssl
 226       - alert: ocsp status unknown
 227         expr: ssl_ocsp_response_status == 1
 228         for: 0m
 229         labels:
 230           alertgroup: ssl
 231   - name: tile
 232     rules:
 233       - alert: renderd replication delay
 234         expr: renderd_replication_delay > 120
 235         for: 5m
 236         labels:
 237           alertgroup: tile
 238         annotations:
 239           delay: "{{ $value | humanizeDuration }}"
 240       - alert: missed tile rate
 241         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 242         for: 5m
 243         labels:
 244           alertgroup: tile
 245         annotations:
 246           miss_rate: "{{ $value | humanizePercentage }}"
 247   - name: time
 248     rules:
 249       - alert: clock not synchronising
 250         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 251         for: 5m
 252         labels:
 253           alertgroup: "{{ $labels.instance }}"
 254       - alert: clock skew detected
 255         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)
 256         for: 5m
 257         labels:
 258           alertgroup: "{{ $labels.instance }}"
 259         annotations:
 260           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"