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

   1 # DO NOT EDIT - This file is being maintained by Chef
   2
   3 groups:
   4   - name: amsterdam
   5     rules:
   6       - alert: pdu current draw
   7         expr: rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10 > 10
   8         for: 5m
   9         labels:
  10           alertgroup: "amsterdam"
  11         annotations:
  12           current: "{{ $value | humanize }}A"
  13       - alert: site current draw
  14         expr: sum(rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10) > 13
  15         for: 5m
  16         labels:
  17           alertgroup: "amsterdam"
  18         annotations:
  19           current: "{{ $value | humanize }}A"
  20       - alert: site temperature
  21         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 25
  22         for: 5m
  23         labels:
  24           alertgroup: "amsterdam"
  25         annotations:
  26           temperature: "{{ $value | humanize }}C"
  27       - alert: site humidity
  28         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 > 0.65
  29         for: 5m
  30         labels:
  31           alertgroup: "amsterdam"
  32         annotations:
  33           humidity: "{{ $value | humanizePercentage }}"
  34   - name: apache
  35     rules:
  36       - alert: apache down
  37         expr: apache_up == 0
  38         for: 5m
  39         labels:
  40           alertgroup: "{{ $labels.instance }}"
  41       - alert: apache workers busy
  42         expr: sum(apache_workers{state="busy"}) by (instance) / sum(apache_scoreboard) by (instance) > 0.8
  43         for: 5m
  44         labels:
  45           alertgroup: "{{ $labels.instance }}"
  46         annotations:
  47           busy_workers: "{{ $value | humanizePercentage }}"
  48       - alert: apache low request rate
  49         expr: rate(apache_accesses_total[5m]) / rate(apache_accesses_total[1h] offset 1w) < 0.25 and rate(apache_accesses_total[1h] offset 1w) > 2
  50         for: 15m
  51         labels:
  52           alertgroup: "{{ $labels.instance }}"
  53         annotations:
  54           request_rate: "{{ $value | humanizePercentage }}"
  55   - name: chef
  56     rules:
  57       - alert: chef client not running
  58         expr: time() - node_systemd_timer_last_trigger_seconds{name="chef-client.timer"} > 3600
  59         for: 12h
  60         labels:
  61           alertgroup: "{{ $labels.instance }}"
  62         annotations:
  63           down_time: "{{ $value | humanizeDuration }}"
  64   - name: cisco
  65     rules:
  66       - alert: cisco fan alarm
  67         expr: rlPhdUnitEnvParamFan1Status{rlPhdUnitEnvParamFan1Status!="normal"} > 0 or rlPhdUnitEnvParamFan2Status{rlPhdUnitEnvParamFan2Status!="normal"} > 0
  68         for: 5m
  69         labels:
  70           alertgroup: "{{ $labels.site }}"
  71         annotations:
  72           fan_rpm: "{{ with printf \"rlPhdUnitEnvParamFan1Speed{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}rpm{{end}}"
  73       - alert: cisco temperature alarm
  74         expr: rlPhdUnitEnvParamTempSensorStatus{rlPhdUnitEnvParamTempSensorStatus!="ok"} > 0
  75         for: 5m
  76         labels:
  77           alertgroup: "{{ $labels.site }}"
  78         annotations:
  79           temp_celsius: "{{ with printf \"rlPhdUnitEnvParamTempSensorValue{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}C{{end}}"
  80       - alert: cisco main power alarm
  81         expr: rlPhdUnitEnvParamMainPSStatus{rlPhdUnitEnvParamMainPSStatus!="normal"} > 0
  82         for: 5m
  83         labels:
  84           alertgroup: "{{ $labels.site }}"
  85       - alert: cisco redundant power alarm
  86         expr: rlPhdUnitEnvParamRedundantPSStatus{rlPhdUnitEnvParamRedundantPSStatus!="normal"} > 0
  87         for: 5m
  88         labels:
  89           alertgroup: "{{ $labels.site }}"
  90   - name: cpu
  91     rules:
  92       - alert: cpu pressure
  93         expr: rate(node_pressure_cpu_waiting_seconds_total[5m]) > 0.6
  94         for: 15m
  95         labels:
  96           alertgroup: "{{ $labels.instance }}"
  97         annotations:
  98           pressure: "{{ $value | humanizePercentage }}"
  99   - name: database
 100     rules:
 101       - alert: postgres replication delay
 102         expr: pg_replication_lag_seconds > 5
 103         for: 5m
 104         labels:
 105           alertgroup: database
 106         annotations:
 107           delay: "{{ $value | humanizeDuration }}"
 108   - name: dublin
 109     rules:
 110       - alert: pdu current draw
 111         expr: rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10 > 13
 112         for: 5m
 113         labels:
 114           alertgroup: "dublin"
 115         annotations:
 116           current: "{{ $value | humanize }}A"
 117       - alert: site current draw
 118         expr: sum(rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10) > 17
 119         for: 5m
 120         labels:
 121           alertgroup: "dublin"
 122         annotations:
 123           current: "{{ $value | humanize }}A"
 124       - alert: site temperature
 125         expr: min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 > 25
 126         for: 5m
 127         labels:
 128           alertgroup: "dublin"
 129         annotations:
 130           temperature: "{{ $value | humanize }}C"
 131       - alert: site humidity
 132         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 > 0.65
 133         for: 5m
 134         labels:
 135           alertgroup: "dublin"
 136         annotations:
 137           humidity: "{{ $value | humanizePercentage }}"
 138   - name: fastly
 139     rules:
 140       - alert: fastly error rate
 141         expr: sum(rate(fastly_rt_status_group_total{status_group="5xx"}[5m])) by (service_name, datacenter) / sum(rate(fastly_rt_status_group_total[5m])) by (service_name, datacenter) > 0.005
 142         for: 15m
 143         labels:
 144           alertgroup: fastly
 145         annotations:
 146           error_rate: "{{ $value | humanizePercentage }}"
 147       - alert: fastly healthcheck failing
 148         expr: count(fastly_healthcheck_status == 0) by (service) > 0
 149         for: 15m
 150         labels:
 151           alertgroup: fastly
 152       - alert: multiple fastly healthchecks failing
 153         expr: count(fastly_healthcheck_status == 0) by (service) > 4
 154         for: 5m
 155         labels:
 156           alertgroup: fastly
 157   - name: filesystem
 158     rules:
 159       - alert: readonly filesystem
 160         expr: node_filesystem_readonly == 1
 161         for: 0m
 162         labels:
 163           alertgroup: "{{ $labels.instance }}"
 164       - alert: filesystem low on space
 165         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 166         for: 5m
 167         labels:
 168           alertgroup: "{{ $labels.instance }}"
 169         annotations:
 170           percentage_free: "{{ $value | humanizePercentage }}"
 171           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 172           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 173       - alert: filesystem low on inodes
 174         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 175         for: 5m
 176         labels:
 177           alertgroup: "{{ $labels.instance }}"
 178         annotations:
 179           percentage_free: "{{ $value | humanizePercentage }}"
 180           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 181           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 182   - name: hwmon
 183     rules:
 184       - alert: hwmon fan alarm
 185         expr: node_hwmon_fan_alarm == 1
 186         for: 5m
 187         labels:
 188           alertgroup: "{{ $labels.instance }}"
 189         annotations:
 190           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}}"
 191           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}}"
 192       - alert: hwmon temperature alarm
 193         expr: node_hwmon_temp_alarm == 1
 194         for: 5m
 195         labels:
 196           alertgroup: "{{ $labels.instance }}"
 197         annotations:
 198           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}}"
 199           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}}"
 200           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}}"
 201       - alert: hwmon voltage alarm
 202         expr: node_hwmon_in_alarm == 1
 203         for: 5m
 204         labels:
 205           alertgroup: "{{ $labels.instance }}"
 206         annotations:
 207           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}}"
 208           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}}"
 209           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}}"
 210   - name: io
 211     rules:
 212       - alert: io pressure
 213         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 214         for: 60m
 215         labels:
 216           alertgroup: "{{ $labels.instance }}"
 217         annotations:
 218           pressure: "{{ $value | humanizePercentage }}"
 219   - name: ipmi
 220     rules:
 221       - alert: ipmi fan alarm
 222         expr: ipmi_fan_speed_state > 0
 223         for: 5m
 224         labels:
 225           alertgroup: "{{ $labels.instance }}"
 226         annotations:
 227           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 228       - alert: ipmi temperature alarm
 229         expr: ipmi_temperature_state > 0
 230         for: 5m
 231         labels:
 232           alertgroup: "{{ $labels.instance }}"
 233         annotations:
 234           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 235       - alert: ipmi voltage alarm
 236         expr: ipmi_voltage_state > 0
 237         for: 5m
 238         labels:
 239           alertgroup: "{{ $labels.instance }}"
 240         annotations:
 241           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 242       - alert: ipmi power alarm
 243         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 244         for: 5m
 245         labels:
 246           alertgroup: "{{ $labels.instance }}"
 247   - name: juniper
 248     rules:
 249       - alert: juniper cpu alarm
 250         expr: jnxOperatingCPU{jnxOperatingContentsIndex="7"} > 30
 251         for: 5m
 252         labels:
 253           alertgroup: "{{ $labels.site }}"
 254       - alert: juniper fan alarm
 255         expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!~"running.*"} > 0
 256         for: 5m
 257         labels:
 258           alertgroup: "{{ $labels.site }}"
 259       - alert: juniper power alarm
 260         expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!~"running.*"} > 0
 261         for: 5m
 262         labels:
 263           alertgroup: "{{ $labels.site }}"
 264   - name: mail
 265     rules:
 266       - alert: exim queue length
 267         expr: exim_queue > exim_queue_limit
 268         for: 60m
 269         labels:
 270           alertgroup: mail
 271         annotations:
 272           queue_length: "{{ $value }}"
 273       - alert: mailman queue length
 274         expr: mailman_queue_length > 200
 275         for: 60m
 276         labels:
 277           alertgroup: mail
 278         annotations:
 279           queue_length: "{{ $value }}"
 280   - name: mdadm
 281     rules:
 282       - alert: mdadm array inactive
 283         expr: node_md_state{state="inactive"} > 0
 284         for: 0m
 285         labels:
 286           alertgroup: "{{ $labels.instance }}"
 287         annotations:
 288           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 289           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 290           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 291           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 292       - alert: mdadm array degraded
 293         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 294         for: 0m
 295         labels:
 296           alertgroup: "{{ $labels.instance }}"
 297         annotations:
 298           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 299           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 300           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 301           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 302       - alert: mdadm disk failed
 303         expr: node_md_disks{state="failed"} > 0
 304         for: 0m
 305         labels:
 306           alertgroup: "{{ $labels.instance }}"
 307         annotations:
 308           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 309           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 310           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 311           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 312   - name: memory
 313     rules:
 314       - alert: low memory
 315         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 316         for: 15m
 317         labels:
 318           alertgroup: "{{ $labels.instance }}"
 319         annotations:
 320           memory_free: "{{ $value | humanizePercentage }}"
 321       - alert: memory pressure
 322         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 323         for: 60m
 324         labels:
 325           alertgroup: "{{ $labels.instance }}"
 326         annotations:
 327           pressure: "{{ $value | humanizePercentage }}"
 328       - alert: oom kill detected
 329         expr: increase(node_vmstat_oom_kill[1m]) > 0
 330         for: 0m
 331         labels:
 332           alertgroup: "{{ $labels.instance }}"
 333         annotations:
 334           new_oom_kills: "{{ $value }}"
 335   - name: network
 336     rules:
 337       - alert: interface transmit rate
 338         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 339         for: 5m
 340         labels:
 341           alertgroup: "{{ $labels.instance }}"
 342         annotations:
 343           bandwidth_used: "{{ $value | humanizePercentage }}"
 344       - alert: interface receive rate
 345         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.98
 346         for: 5m
 347         labels:
 348           alertgroup: "{{ $labels.instance }}"
 349         annotations:
 350           bandwidth_used: "{{ $value | humanizePercentage }}"
 351       - alert: interface transmit errors
 352         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 353         for: 5m
 354         labels:
 355           alertgroup: "{{ $labels.instance }}"
 356         annotations:
 357           error_rate: "{{ $value | humanizePercentage }}"
 358       - alert: wireguard interface transmit errors
 359         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 360         for: 1h
 361         labels:
 362           alertgroup: "{{ $labels.instance }}"
 363         annotations:
 364           error_rate: "{{ $value | humanizePercentage }}"
 365       - alert: interface receive errors
 366         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 367         for: 5m
 368         labels:
 369           alertgroup: "{{ $labels.instance }}"
 370         annotations:
 371           error_rate: "{{ $value | humanizePercentage }}"
 372       - alert: conntrack entries
 373         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 374         for: 5m
 375         labels:
 376           alertgroup: "{{ $labels.instance }}"
 377         annotations:
 378           entries_used: "{{ $value | humanizePercentage }}"
 379   - name: nominatim
 380     rules:
 381       - alert: nominatim replication delay
 382         expr: nominatim_replication_delay > 10800
 383         for: 1h
 384         labels:
 385           alertgroup: nominatim
 386         annotations:
 387           delay: "{{ $value | humanizeDuration }}"
 388   - name: overpass
 389     rules:
 390       - alert: overpass osm database age
 391         expr: overpass_database_age_seconds{database="osm"} > 3600
 392         for: 1h
 393         labels:
 394           alertgroup: overpass
 395         annotations:
 396           age: "{{ $value | humanizeDuration }}"
 397       - alert: overpass area database age
 398         expr: overpass_database_age_seconds{database="area"} > 86400
 399         for: 1h
 400         labels:
 401           alertgroup: overpass
 402         annotations:
 403           age: "{{ $value | humanizeDuration }}"
 404   - name: passenger
 405     rules:
 406       - alert: passenger down
 407         expr: passenger_up == 0
 408         for: 5m
 409         labels:
 410           alertgroup: "{{ $labels.instance }}"
 411       - alert: passenger queuing
 412         expr: passenger_top_level_request_queue > 0
 413         for: 5m
 414         labels:
 415           alertgroup: "{{ $labels.instance }}"
 416       - alert: passenger application queuing
 417         expr: passenger_app_request_queue > 0
 418         for: 5m
 419         labels:
 420           alertgroup: "{{ $labels.instance }}"
 421   - name: planet
 422     rules:
 423       - alert: planet dump overdue
 424         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 425         for: 24h
 426         labels:
 427           alertgroup: planet
 428         annotations:
 429           overdue_by: "{{ $value | humanizeDuration }}"
 430       - alert: notes dump overdue
 431         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 432         for: 6h
 433         labels:
 434           alertgroup: planet
 435         annotations:
 436           overdue_by: "{{ $value | humanizeDuration }}"
 437       - alert: daily replication feed delayed
 438         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 439         for: 3h
 440         labels:
 441           alertgroup: planet
 442         annotations:
 443           delayed_by: "{{ $value | humanizeDuration }}"
 444       - alert: hourly replication feed delayed
 445         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 446         for: 30m
 447         labels:
 448           alertgroup: planet
 449         annotations:
 450           delayed_by: "{{ $value | humanizeDuration }}"
 451       - alert: minutely replication feed delayed
 452         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 453         for: 5m
 454         labels:
 455           alertgroup: planet
 456         annotations:
 457           delayed_by: "{{ $value | humanizeDuration }}"
 458       - alert: changeset replication feed delayed
 459         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 460         for: 5m
 461         labels:
 462           alertgroup: planet
 463         annotations:
 464           delayed_by: "{{ $value | humanizeDuration }}"
 465   - name: postgresql
 466     rules:
 467       - alert: postgresql down
 468         expr: pg_up == 0
 469         for: 1m
 470         labels:
 471           alertgroup: "{{ $labels.instance }}"
 472       - alert: postgresql replication delay
 473         expr: pg_replication_lag_seconds > 5
 474         for: 1m
 475         labels:
 476           alertgroup: "{{ $labels.instance }}"
 477         annotations:
 478           delay: "{{ $value | humanizeDuration }}"
 479       - alert: postgresql connection limit
 480         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 481         for: 1m
 482         labels:
 483           alertgroup: "{{ $labels.instance }}"
 484         annotations:
 485           connections_used: "{{ $value | humanizePercentage }}"
 486       - alert: postgresql deadlocks
 487         expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
 488         for: 0m
 489         labels:
 490           alertgroup: "{{ $labels.instance }}"
 491         annotations:
 492           new_deadlocks: "{{ $value }}"
 493       - alert: postgresql slow queries
 494         expr: pg_slow_queries > 0
 495         for: 5m
 496         labels:
 497           alertgroup: "{{ $labels.instance }}"
 498         annotations:
 499           queries: "{{ $value }}"
 500   - name: prometheus
 501     rules:
 502       - alert: prometheus configuration error
 503         expr: prometheus_config_last_reload_successful == 0
 504         for: 10m
 505         labels:
 506           alertgroup: "prometheus"
 507       - alert: prometheus target missing
 508         expr: up == 0
 509         for: 10m
 510         labels:
 511           alertgroup: "prometheus"
 512   - name: raid
 513     rules:
 514       - alert: raid array degraded
 515         expr: ohai_array_info{status="degraded"} > 0
 516         for: 5m
 517         labels:
 518           alertgroup: "{{ $labels.instance }}"
 519       - alert: raid disk failed
 520         expr: ohai_disk_info{status="failed"} > 0
 521         for: 5m
 522         labels:
 523           alertgroup: "{{ $labels.instance }}"
 524   - name: rasdaemon
 525     rules:
 526       - alert: memory controller errors
 527         expr: increase(rasdaemon_mc_events_total[1m]) > 0
 528         for: 0m
 529         labels:
 530           alertgroup: "{{ $labels.instance }}"
 531         annotations:
 532           new_errors: "{{ $value }}"
 533       - alert: pcie aer errors
 534         expr: increase(rasdaemon_aer_events_total[1m]) > 0
 535         for: 0m
 536         labels:
 537           alertgroup: "{{ $labels.instance }}"
 538         annotations:
 539           new_ercrors: "{{ $value }}"
 540   - name: smart
 541     rules:
 542       - alert: smart failure
 543         expr: smart_health_status == 0
 544         for: 60m
 545         labels:
 546           alertgroup: "{{ $labels.instance }}"
 547       - alert: smart ssd wearout approaching
 548         expr: smart_percentage_used >= 80
 549         for: 60m
 550         labels:
 551           alertgroup: "{{ $labels.instance }}"
 552         annotations:
 553           percentage_used: "{{ $value | humanizePercentage }}"
 554   - name: ssl
 555     rules:
 556       - alert: ssl certificate probe failed
 557         expr: ssl_probe_success == 0
 558         for: 60m
 559         labels:
 560           alertgroup: ssl
 561       - alert: ssl certificate expiry
 562         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 563         for: 0m
 564         labels:
 565           alertgroup: ssl
 566         annotations:
 567           expires_in: "{{ $value | humanizeDuration }}"
 568       - alert: ssl certificate revoked
 569         expr: ssl_ocsp_response_status == 1
 570         for: 0m
 571         labels:
 572           alertgroup: ssl
 573       - alert: ocsp status unknown
 574         expr: ssl_ocsp_response_status == 1
 575         for: 0m
 576         labels:
 577           alertgroup: ssl
 578   - name: statuscake
 579     rules:
 580       - alert: statuscake uptime check failing
 581         expr: statuscake_uptime{status="down",paused="false"} > 0
 582         for: 10m
 583         labels:
 584           alertgroup: statuscake
 585   - name: systemd
 586     rules:
 587       - alert: systemd failed service
 588         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 589         for: 5m
 590         labels:
 591           alertgroup: "{{ $labels.instance }}"
 592       - alert: systemd failed chef client service
 593         expr: node_systemd_unit_state{state="failed",name="chef-client.service"} == 1
 594         for: 6h
 595         labels:
 596           alertgroup: "{{ $labels.instance }}"
 597   - name: tile
 598     rules:
 599       - alert: renderd replication delay
 600         expr: renderd_replication_delay > 120
 601         for: 15m
 602         labels:
 603           alertgroup: tile
 604         annotations:
 605           delay: "{{ $value | humanizeDuration }}"
 606       - alert: missed tile rate
 607         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 608         for: 5m
 609         labels:
 610           alertgroup: tile
 611         annotations:
 612           miss_rate: "{{ $value | humanizePercentage }}"
 613   - name: time
 614     rules:
 615       - alert: clock not synchronising
 616         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 617         for: 5m
 618         labels:
 619           alertgroup: "{{ $labels.instance }}"
 620       - alert: clock skew detected
 621         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)
 622         for: 5m
 623         labels:
 624           alertgroup: "{{ $labels.instance }}"
 625         annotations:
 626           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 627   - name: web
 628     rules:
 629       - alert: web error rate
 630         expr: sum(rate(api_call_count_total{status=~"50[0-8]|5[1-9][0-9]"}[5m])) by (instance) / sum(rate(api_call_count_total[5m])) by (instance) > 0.002
 631         for: 5m
 632         labels:
 633           alertgroup: web
 634         annotations:
 635           error_rate: "{{ $value | humanizePercentage }}"
 636       - alert: job processing rate
 637         expr: rate(pg_stat_user_tables_n_tup_del{datname="openstreetmap",relname="delayed_jobs"}[5m]) / rate(pg_stat_user_tables_n_tup_ins{datname="openstreetmap",relname="delayed_jobs"}[5m]) < 0.9 and ignoring(job, name, datname, relname, schemaname, server) chef_role{name="db-master"} == 1
 638         for: 15m
 639         labels:
 640           alertgroup: web
 641         annotations:
 642           job_processing_rate: "{{ $value | humanizePercentage }}"