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: uplink
   7         expr: junos_interface_up{site="amsterdam",name=~"ge-[01]/2/2"} != 1
   8         for: 6m
   9         labels:
  10           alertgroup: "amsterdam"
  11         annotations:
  12           status: "{{ $value }}"
  13       - alert: pdu current draw
  14         expr: rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10 > 28
  15         for: 6m
  16         labels:
  17           alertgroup: "amsterdam"
  18         annotations:
  19           current: "{{ $value | humanize }}A"
  20       - alert: site power
  21         expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="amsterdam",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 3.5
  22         for: 6m
  23         labels:
  24           alertgroup: "amsterdam"
  25         annotations:
  26           current: "{{ $value | humanize }}kVA"
  27       - alert: site temperature
  28         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 26
  29         for: 6m
  30         labels:
  31           alertgroup: "amsterdam"
  32         annotations:
  33           temperature: "{{ $value | humanize }}C"
  34       - alert: site humidity
  35         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 > 0.65
  36         for: 6m
  37         labels:
  38           alertgroup: "amsterdam"
  39         annotations:
  40           humidity: "{{ $value | humanizePercentage }}"
  41   - name: apache
  42     rules:
  43       - alert: apache down
  44         expr: apache_up == 0
  45         for: 5m
  46         labels:
  47           alertgroup: "{{ $labels.instance }}"
  48       - alert: apache workers busy
  49         expr: sum(apache_workers{state="busy"}) by (instance) / sum(apache_scoreboard) by (instance) > 0.8
  50         for: 5m
  51         labels:
  52           alertgroup: "{{ $labels.instance }}"
  53         annotations:
  54           busy_workers: "{{ $value | humanizePercentage }}"
  55       - alert: apache connection limit
  56         expr: (apache_connections{state="total"} - on (instance) apache_connections{state="closing"}) / on (instance) (apache_server_limit * on (instance) (apache_threads_per_child + on (instance) (apache_async_request_worker_factor * on (instance) apache_workers{state="idle"} / on(instance) apache_processes{state="all"}))) > 0.8
  57         for: 5m
  58         labels:
  59           alertgroup: "{{ $labels.instance }}"
  60         annotations:
  61           connections: "{{ $value | humanizePercentage }}"
  62   - name: chef
  63     rules:
  64       - alert: chef client not running
  65         expr: time() - node_systemd_timer_last_trigger_seconds{name="chef-client.timer"} > 3600
  66         for: 12h
  67         labels:
  68           alertgroup: "{{ $labels.instance }}"
  69         annotations:
  70           down_time: "{{ $value | humanizeDuration }}"
  71   - name: cisco
  72     rules:
  73       - alert: cisco fan alarm
  74         expr: rlPhdUnitEnvParamFan1Status{rlPhdUnitEnvParamFan1Status!="normal"} > 0 or rlPhdUnitEnvParamFan2Status{rlPhdUnitEnvParamFan2Status!="normal"} > 0
  75         for: 5m
  76         labels:
  77           alertgroup: "{{ $labels.site }}"
  78         annotations:
  79           fan_rpm: "{{ with printf \"rlPhdUnitEnvParamFan1Speed{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}rpm{{end}}"
  80       - alert: cisco temperature alarm
  81         expr: rlPhdUnitEnvParamTempSensorStatus{rlPhdUnitEnvParamTempSensorStatus!="ok"} > 0
  82         for: 5m
  83         labels:
  84           alertgroup: "{{ $labels.site }}"
  85         annotations:
  86           temp_celsius: "{{ with printf \"rlPhdUnitEnvParamTempSensorValue{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}C{{end}}"
  87       - alert: cisco main power alarm
  88         expr: rlPhdUnitEnvParamMainPSStatus{rlPhdUnitEnvParamMainPSStatus!="normal"} > 0
  89         for: 5m
  90         labels:
  91           alertgroup: "{{ $labels.site }}"
  92       - alert: cisco redundant power alarm
  93         expr: rlPhdUnitEnvParamRedundantPSStatus{rlPhdUnitEnvParamRedundantPSStatus!="normal"} > 0
  94         for: 5m
  95         labels:
  96           alertgroup: "{{ $labels.site }}"
  97   - name: cpu
  98     rules:
  99       - alert: cpu pressure
 100         expr: rate(node_pressure_cpu_waiting_seconds_total[5m]) > 0.75
 101         for: 60m
 102         labels:
 103           alertgroup: "{{ $labels.instance }}"
 104         annotations:
 105           pressure: "{{ $value | humanizePercentage }}"
 106   - name: database
 107     rules:
 108       - alert: postgres replication delay
 109         expr: pg_replication_lag_seconds > 30
 110         for: 15m
 111         labels:
 112           alertgroup: database
 113         annotations:
 114           delay: "{{ $value | humanizeDuration }}"
 115   - name: discourse
 116     rules:
 117       - alert: discourse job failure rate
 118         expr: rate(discourse_job_failures[5m]) > 0
 119         for: 5m
 120         labels:
 121           alertgroup: discourse
 122         annotations:
 123           failure_rate: "{{ $value }} jobs/s"
 124   - name: dublin
 125     rules:
 126       - alert: uplink
 127         expr: junos_interface_up{site="dublin",name=~"ge-[01]/2/2"} != 1
 128         for: 6m
 129         labels:
 130           alertgroup: "dublin"
 131         annotations:
 132           status: "{{ $value }}"
 133       - alert: pdu current draw
 134         expr: rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10 > 28
 135         for: 6m
 136         labels:
 137           alertgroup: "dublin"
 138         annotations:
 139           current: "{{ $value | humanize }}A"
 140       - alert: site power
 141         expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="dublin",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 4
 142         for: 6m
 143         labels:
 144           alertgroup: "dublin"
 145         annotations:
 146           current: "{{ $value | humanize }}kVA"
 147       - alert: site temperature
 148         expr: min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 > 26
 149         for: 6m
 150         labels:
 151           alertgroup: "dublin"
 152         annotations:
 153           temperature: "{{ $value | humanize }}C"
 154       - alert: site humidity
 155         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 > 0.65
 156         for: 6m
 157         labels:
 158           alertgroup: "dublin"
 159         annotations:
 160           humidity: "{{ $value | humanizePercentage }}"
 161   - name: fastly
 162     rules:
 163       - alert: fastly error rate
 164         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
 165         for: 15m
 166         labels:
 167           alertgroup: fastly
 168         annotations:
 169           error_rate: "{{ $value | humanizePercentage }}"
 170       - alert: fastly frontend healthcheck warning
 171         expr: count(fastly_healthcheck_status == 0) by (service, datacenter) > 2
 172         for: 15m
 173         labels:
 174           alertgroup: fastly
 175       - alert: fastly frontend healthcheck critical
 176         expr: count(fastly_healthcheck_status == 0) by (service, datacenter) == count(fastly_healthcheck_status) by (service, datacenter)
 177         for: 5m
 178         labels:
 179           alertgroup: fastly
 180       - alert: fastly backend healthcheck warning
 181         expr: count(fastly_healthcheck_status == 0) by (service, backend) > 10
 182         for: 15m
 183         labels:
 184           alertgroup: fastly
 185       - alert: fastly backend healthcheck critical
 186         expr: count(fastly_healthcheck_status == 0) by (service, backend) == count(fastly_healthcheck_status) by (service, backend)
 187         for: 5m
 188         labels:
 189           alertgroup: fastly
 190   - name: filesystem
 191     rules:
 192       - alert: readonly filesystem
 193         expr: node_filesystem_readonly > min_over_time(node_filesystem_readonly[7d])
 194         for: 0m
 195         labels:
 196           alertgroup: "{{ $labels.instance }}"
 197       - alert: filesystem low on space
 198         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 199         for: 5m
 200         labels:
 201           alertgroup: "{{ $labels.instance }}"
 202         annotations:
 203           percentage_free: "{{ $value | humanizePercentage }}"
 204           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 205           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 206       - alert: filesystem low on inodes
 207         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 208         for: 5m
 209         labels:
 210           alertgroup: "{{ $labels.instance }}"
 211         annotations:
 212           percentage_free: "{{ $value | humanizePercentage }}"
 213           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 214           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 215   - name: hwmon
 216     rules:
 217       - alert: hwmon fan alarm
 218         expr: node_hwmon_fan_alarm == 1
 219         for: 5m
 220         labels:
 221           alertgroup: "{{ $labels.instance }}"
 222         annotations:
 223           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}}"
 224           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}}"
 225       - alert: hwmon temperature alarm
 226         expr: node_hwmon_temp_alarm == 1
 227         for: 5m
 228         labels:
 229           alertgroup: "{{ $labels.instance }}"
 230         annotations:
 231           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}}"
 232           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}}"
 233           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}}"
 234       - alert: hwmon voltage alarm
 235         expr: node_hwmon_in_alarm == 1
 236         for: 5m
 237         labels:
 238           alertgroup: "{{ $labels.instance }}"
 239         annotations:
 240           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}}"
 241           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}}"
 242           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}}"
 243   - name: io
 244     rules:
 245       - alert: io pressure
 246         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 247         for: 60m
 248         labels:
 249           alertgroup: "{{ $labels.instance }}"
 250         annotations:
 251           pressure: "{{ $value | humanizePercentage }}"
 252   - name: ipmi
 253     rules:
 254       - alert: ipmi fan alarm
 255         expr: ipmi_fan_speed_state > 0
 256         for: 5m
 257         labels:
 258           alertgroup: "{{ $labels.instance }}"
 259         annotations:
 260           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 261       - alert: ipmi temperature alarm
 262         expr: ipmi_temperature_state > 0
 263         for: 5m
 264         labels:
 265           alertgroup: "{{ $labels.instance }}"
 266         annotations:
 267           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 268       - alert: ipmi voltage alarm
 269         expr: ipmi_voltage_state > 0
 270         for: 5m
 271         labels:
 272           alertgroup: "{{ $labels.instance }}"
 273         annotations:
 274           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 275       - alert: ipmi power alarm
 276         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 277         for: 5m
 278         labels:
 279           alertgroup: "{{ $labels.instance }}"
 280   - name: juniper
 281     rules:
 282       - alert: juniper red alarms
 283         expr: juniper_alarms_red_count > 0
 284         for: 5m
 285         labels:
 286           alertgroup: "{{ $labels.site }}"
 287         annotations:
 288           alarm_count: "{{ $value }} alarms"
 289       - alert: juniper yellow alarms
 290         expr: juniper_alarms_yellow_count > 0
 291         for: 5m
 292         labels:
 293           alertgroup: "{{ $labels.site }}"
 294         annotations:
 295           alarm_count: "{{ $value }} alarms"
 296       - alert: juniper cpu alarm
 297         expr: junos_route_engine_load_average_five / 2 > 0.5
 298         for: 5m
 299         labels:
 300           alertgroup: "{{ $labels.site }}"
 301         annotations:
 302           load_average: "{{ $value | humanizePercentage }}"
 303       - alert: juniper fan alarm
 304         expr: junos_environment_fan_up != 1
 305         for: 5m
 306         labels:
 307           alertgroup: "{{ $labels.site }}"
 308       - alert: juniper power alarm
 309         expr: junos_environment_power_up != 1
 310         for: 5m
 311         labels:
 312           alertgroup: "{{ $labels.site }}"
 313   - name: mail
 314     rules:
 315       - alert: exim down
 316         expr: exim_up == 0
 317         for: 5m
 318         labels:
 319           alertgroup: "{{ $labels.instance }}"
 320       - alert: exim queue length
 321         expr: exim_queue > ignoring(job) exim_queue_limit
 322         for: 60m
 323         labels:
 324           alertgroup: mail
 325         annotations:
 326           queue_length: "{{ $value }}"
 327       - alert: mailman queue length
 328         expr: mailman_queue_length > 200
 329         for: 60m
 330         labels:
 331           alertgroup: mail
 332         annotations:
 333           queue_length: "{{ $value }}"
 334   - name: mdadm
 335     rules:
 336       - alert: mdadm array inactive
 337         expr: node_md_state{state="inactive"} > 0
 338         for: 0m
 339         labels:
 340           alertgroup: "{{ $labels.instance }}"
 341         annotations:
 342           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 343           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 344           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 345           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 346       - alert: mdadm array degraded
 347         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 348         for: 0m
 349         labels:
 350           alertgroup: "{{ $labels.instance }}"
 351         annotations:
 352           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 353           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 354           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 355           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 356       - alert: mdadm disk failed
 357         expr: node_md_disks{state="failed"} > 0
 358         for: 0m
 359         labels:
 360           alertgroup: "{{ $labels.instance }}"
 361         annotations:
 362           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 363           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 364           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 365           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 366   - name: memory
 367     rules:
 368       - alert: low memory
 369         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 370         for: 15m
 371         labels:
 372           alertgroup: "{{ $labels.instance }}"
 373         annotations:
 374           memory_free: "{{ $value | humanizePercentage }}"
 375       - alert: memory pressure
 376         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 377         for: 60m
 378         labels:
 379           alertgroup: "{{ $labels.instance }}"
 380         annotations:
 381           pressure: "{{ $value | humanizePercentage }}"
 382       - alert: oom kill detected
 383         expr: increase(node_vmstat_oom_kill[1m]) > 0
 384         for: 0m
 385         labels:
 386           alertgroup: "{{ $labels.instance }}"
 387         annotations:
 388           new_oom_kills: "{{ $value }}"
 389   - name: mysql
 390     rules:
 391       - alert: mysql down
 392         expr: mysql_up == 0
 393         for: 1m
 394         labels:
 395           alertgroup: "{{ $labels.instance }}"
 396       - alert: mysql connection limit
 397         expr: mysql_global_status_max_used_connections / mysql_global_variables_max_connections > 0.8
 398         for: 1m
 399         labels:
 400           alertgroup: "{{ $labels.instance }}"
 401         annotations:
 402           connections_used: "{{ $value | humanizePercentage }}"
 403   - name: network
 404     rules:
 405       - alert: interface transmit rate
 406         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.99
 407         for: 5m
 408         labels:
 409           alertgroup: "{{ $labels.instance }}"
 410         annotations:
 411           bandwidth_used: "{{ $value | humanizePercentage }}"
 412       - alert: interface receive rate
 413         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.99
 414         for: 5m
 415         labels:
 416           alertgroup: "{{ $labels.instance }}"
 417         annotations:
 418           bandwidth_used: "{{ $value | humanizePercentage }}"
 419       - alert: interface transmit errors
 420         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 421         for: 5m
 422         labels:
 423           alertgroup: "{{ $labels.instance }}"
 424         annotations:
 425           error_rate: "{{ $value | humanizePercentage }}"
 426       - alert: wireguard interface transmit errors
 427         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 428         for: 1h
 429         labels:
 430           alertgroup: "{{ $labels.instance }}"
 431         annotations:
 432           error_rate: "{{ $value | humanizePercentage }}"
 433       - alert: interface receive errors
 434         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 435         for: 5m
 436         labels:
 437           alertgroup: "{{ $labels.instance }}"
 438         annotations:
 439           error_rate: "{{ $value | humanizePercentage }}"
 440       - alert: conntrack entries
 441         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 442         for: 5m
 443         labels:
 444           alertgroup: "{{ $labels.instance }}"
 445         annotations:
 446           entries_used: "{{ $value | humanizePercentage }}"
 447   - name: nominatim
 448     rules:
 449       - alert: nominatim replication delay
 450         expr: nominatim_replication_delay > 10800
 451         for: 1h
 452         labels:
 453           alertgroup: nominatim
 454         annotations:
 455           delay: "{{ $value | humanizeDuration }}"
 456   - name: overpass
 457     rules:
 458       - alert: overpass osm database age
 459         expr: overpass_database_age_seconds{database="osm"} > 3600
 460         for: 1h
 461         labels:
 462           alertgroup: overpass
 463         annotations:
 464           age: "{{ $value | humanizeDuration }}"
 465       - alert: overpass area database age
 466         expr: overpass_database_age_seconds{database="area"} > 86400
 467         for: 1h
 468         labels:
 469           alertgroup: overpass
 470         annotations:
 471           age: "{{ $value | humanizeDuration }}"
 472   - name: passenger
 473     rules:
 474       - alert: passenger down
 475         expr: passenger_up == 0
 476         for: 5m
 477         labels:
 478           alertgroup: "{{ $labels.instance }}"
 479       - alert: passenger queuing
 480         expr: passenger_top_level_request_queue > 0
 481         for: 5m
 482         labels:
 483           alertgroup: "{{ $labels.instance }}"
 484       - alert: passenger application queuing
 485         expr: passenger_app_request_queue > 0
 486         for: 5m
 487         labels:
 488           alertgroup: "{{ $labels.instance }}"
 489   - name: planet
 490     rules:
 491       - alert: planet dump overdue
 492         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 493         for: 24h
 494         labels:
 495           alertgroup: planet
 496         annotations:
 497           overdue_by: "{{ $value | humanizeDuration }}"
 498       - alert: notes dump overdue
 499         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 500         for: 6h
 501         labels:
 502           alertgroup: planet
 503         annotations:
 504           overdue_by: "{{ $value | humanizeDuration }}"
 505       - alert: daily replication feed delayed
 506         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 507         for: 3h
 508         labels:
 509           alertgroup: planet
 510         annotations:
 511           delayed_by: "{{ $value | humanizeDuration }}"
 512       - alert: hourly replication feed delayed
 513         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 514         for: 30m
 515         labels:
 516           alertgroup: planet
 517         annotations:
 518           delayed_by: "{{ $value | humanizeDuration }}"
 519       - alert: minutely replication feed delayed
 520         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 521         for: 5m
 522         labels:
 523           alertgroup: planet
 524         annotations:
 525           delayed_by: "{{ $value | humanizeDuration }}"
 526       - alert: changeset replication feed delayed
 527         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 528         for: 5m
 529         labels:
 530           alertgroup: planet
 531         annotations:
 532           delayed_by: "{{ $value | humanizeDuration }}"
 533   - name: postgresql
 534     rules:
 535       - alert: postgresql down
 536         expr: pg_up == 0
 537         for: 1m
 538         labels:
 539           alertgroup: "{{ $labels.instance }}"
 540       - alert: postgresql replication delay
 541         expr: pg_replication_lag_seconds > 30
 542         for: 15m
 543         labels:
 544           alertgroup: "{{ $labels.instance }}"
 545         annotations:
 546           delay: "{{ $value | humanizeDuration }}"
 547       - alert: postgresql connection limit
 548         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 549         for: 1m
 550         labels:
 551           alertgroup: "{{ $labels.instance }}"
 552         annotations:
 553           connections_used: "{{ $value | humanizePercentage }}"
 554       - alert: postgresql deadlocks
 555         expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
 556         for: 0m
 557         labels:
 558           alertgroup: "{{ $labels.instance }}"
 559         annotations:
 560           new_deadlocks: "{{ $value }}"
 561       - alert: postgresql slow queries
 562         expr: pg_slow_queries > 0
 563         for: 5m
 564         labels:
 565           alertgroup: "{{ $labels.instance }}"
 566         annotations:
 567           queries: "{{ $value }}"
 568       - alert: postgresql idle transactions
 569         expr: sum(pg_process_idle_seconds_count{state="idle in transaction"}) by (instance, server) > sum(pg_process_idle_seconds_bucket{state="idle in transaction",le="300"}) by (instance, server)
 570         for: 5m
 571         labels:
 572           alertgroup: "{{ $labels.instance }}"
 573         annotations:
 574           queries: "{{ $value }}"
 575   - name: prometheus
 576     rules:
 577       - alert: prometheus configuration error
 578         expr: prometheus_config_last_reload_successful == 0
 579         for: 10m
 580         labels:
 581           alertgroup: "prometheus"
 582       - alert: prometheus target missing
 583         expr: up == 0
 584         for: 10m
 585         labels:
 586           alertgroup: "prometheus"
 587   - name: raid
 588     rules:
 589       - alert: raid controller battery failed
 590         expr: ohai_controller_info{battery_status="failed"} > 0
 591         for: 5m
 592         labels:
 593           alertgroup: "{{ $labels.instance }}"
 594       - alert: raid controller battery recharging
 595         expr: ohai_controller_info{battery_status="recharging"} > 0
 596         for: 4h
 597         labels:
 598           alertgroup: "{{ $labels.instance }}"
 599       - alert: raid array degraded
 600         expr: ohai_array_info{status="degraded"} > 0
 601         for: 5m
 602         labels:
 603           alertgroup: "{{ $labels.instance }}"
 604       - alert: raid disk failed
 605         expr: ohai_disk_info{status="failed"} > 0
 606         for: 5m
 607         labels:
 608           alertgroup: "{{ $labels.instance }}"
 609   - name: rasdaemon
 610     rules:
 611       - alert: memory controller errors
 612         expr: increase(rasdaemon_mc_events_total[1m]) > 0
 613         for: 0m
 614         labels:
 615           alertgroup: "{{ $labels.instance }}"
 616         annotations:
 617           new_errors: "{{ $value }}"
 618       - alert: pcie aer errors
 619         expr: increase(rasdaemon_aer_events_total[1m]) > 0
 620         for: 0m
 621         labels:
 622           alertgroup: "{{ $labels.instance }}"
 623         annotations:
 624           new_errors: "{{ $value }}"
 625   - name: smart
 626     rules:
 627       - alert: smart failure
 628         expr: smart_health_status == 0
 629         for: 60m
 630         labels:
 631           alertgroup: "{{ $labels.instance }}"
 632       - alert: smart ssd wearout approaching
 633         expr: smart_percentage_used / 100 >= 0.8
 634         for: 60m
 635         labels:
 636           alertgroup: "{{ $labels.instance }}"
 637         annotations:
 638           percentage_used: "{{ $value | humanizePercentage }}"
 639   - name: smokeping
 640     rules:
 641       - alert: packet loss
 642         expr: 1 - (rate(smokeping_response_duration_seconds_count[5m]) / rate(smokeping_requests_total[5m])) > 0.02
 643         for: 10m
 644         labels:
 645           alertgroup: smokeping
 646         annotations:
 647           loss_rate: "{{ $value | humanizePercentage }}"
 648   - name: snmp
 649     rules:
 650       - alert: snmp pdus missing
 651         expr: max_over_time(snmp_scrape_pdus_returned[1d]) - snmp_scrape_pdus_returned > 0
 652         for: 15m
 653         labels:
 654           alertgroup: snmp
 655         annotations:
 656           missing_pdus: "{{ $value }}"
 657   - name: ssl
 658     rules:
 659       - alert: ssl certificate probe failed
 660         expr: ssl_probe_success == 0
 661         for: 60m
 662         labels:
 663           alertgroup: ssl
 664       - alert: ssl certificate expiry
 665         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 666         for: 0m
 667         labels:
 668           alertgroup: ssl
 669         annotations:
 670           expires_in: "{{ $value | humanizeDuration }}"
 671       - alert: ssl certificate revoked
 672         expr: ssl_ocsp_response_status == 1
 673         for: 0m
 674         labels:
 675           alertgroup: ssl
 676       - alert: ocsp status unknown
 677         expr: ssl_ocsp_response_status == 1
 678         for: 0m
 679         labels:
 680           alertgroup: ssl
 681   - name: statuscake
 682     rules:
 683       - alert: statuscake uptime check failing
 684         expr: statuscake_paused == 0 and statuscake_up == 0
 685         for: 10m
 686         labels:
 687           alertgroup: statuscake
 688   - name: systemd
 689     rules:
 690       - alert: systemd failed service
 691         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 692         for: 5m
 693         labels:
 694           alertgroup: "{{ $labels.instance }}"
 695       - alert: systemd failed chef client service
 696         expr: sum_over_time(node_systemd_unit_state{state="inactive",name="chef-client.service"}[6h]) == 0
 697         for: 0m
 698         labels:
 699           alertgroup: "{{ $labels.instance }}"
 700   - name: taginfo
 701     rules:
 702       - alert: taginfo planet age
 703         expr: time() - taginfo_data_from_seconds > 129600 and on (instance) chef_role{name="taginfo"}
 704         for: 0m
 705         labels:
 706           alertgroup: taginfo
 707         annotations:
 708           age: "{{ $value | humanizeDuration }}"
 709       - alert: taginfo database age
 710         expr: time() - taginfo_database_update_finish_seconds > 129600 and on (instance) chef_role{name="taginfo"}
 711         for: 0m
 712         labels:
 713           alertgroup: taginfo
 714         annotations:
 715           age: "{{ $value | humanizeDuration }}"
 716       - alert: taginfo database size
 717         expr: abs(delta(taginfo_database_size_bytes[30m])) / taginfo_database_size_bytes > 0.1
 718         for: 30m
 719         labels:
 720           alertgroup: taginfo
 721         annotations:
 722           size_change: "{{ $value | humanizePercentage }}"
 723   - name: tile
 724     rules:
 725       - alert: renderd replication delay
 726         expr: renderd_replication_delay > 120
 727         for: 15m
 728         labels:
 729           alertgroup: tile
 730         annotations:
 731           delay: "{{ $value | humanizeDuration }}"
 732       - alert: missed tile rate
 733         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 734         for: 5m
 735         labels:
 736           alertgroup: tile
 737         annotations:
 738           miss_rate: "{{ $value | humanizePercentage }}"
 739       - alert: tile render rate
 740         expr: sum(rate(renderd_zoom_metatiles_total[5m])) by (instance) == 0
 741         for: 15m
 742         labels:
 743           alertgroup: tile
 744         annotations:
 745           render_rate: "{{ $value }} tiles/s"
 746   - name: time
 747     rules:
 748       - alert: clock not synchronising
 749         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 750         for: 5m
 751         labels:
 752           alertgroup: "{{ $labels.instance }}"
 753       - alert: clock skew detected
 754         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)
 755         for: 5m
 756         labels:
 757           alertgroup: "{{ $labels.instance }}"
 758         annotations:
 759           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 760   - name: web
 761     rules:
 762       - alert: web error rate
 763         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
 764         for: 5m
 765         labels:
 766           alertgroup: web
 767         annotations:
 768           error_rate: "{{ $value | humanizePercentage }}"
 769       - alert: job processing rate
 770         expr: rate(pg_stat_user_tables_n_tup_del{datname="openstreetmap",relname="delayed_jobs"}[1h]) / rate(pg_stat_user_tables_n_tup_ins{datname="openstreetmap",relname="delayed_jobs"}[1h]) < 0.9 and ignoring(job, name, datname, relname, schemaname, server) chef_role{name="db-master"} == 1
 771         for: 1h
 772         labels:
 773           alertgroup: web
 774         annotations:
 775           job_processing_rate: "{{ $value | humanizePercentage }}"