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: 6m
   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: 6m
  16         labels:
  17           alertgroup: "amsterdam"
  18         annotations:
  19           current: "{{ $value | humanize }}A"
  20       - alert: site power
  21         expr: sum(rPDU2PhaseStatusApparentPower{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 100) > 3
  22         for: 0m
  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 low request rate
  56         expr: rate(apache_accesses_total[5m]) / rate(apache_accesses_total[1h] offset 1w) < 0.25 and rate(apache_accesses_total[1h] offset 1w) > 2
  57         for: 15m
  58         labels:
  59           alertgroup: "{{ $labels.instance }}"
  60         annotations:
  61           request_rate: "{{ $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: pdu current draw
 127         expr: rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10 > 13
 128         for: 6m
 129         labels:
 130           alertgroup: "dublin"
 131         annotations:
 132           current: "{{ $value | humanize }}A"
 133       - alert: site current draw
 134         expr: sum(rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10) > 17
 135         for: 6m
 136         labels:
 137           alertgroup: "dublin"
 138         annotations:
 139           current: "{{ $value | humanize }}A"
 140       - alert: site power
 141         expr: sum(rPDU2PhaseStatusApparentPower{site="dublin",rPDU2PhaseStatusIndex="1"} / 100) > 4
 142         for: 0m
 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 healthcheck failing
 171         expr: count(fastly_healthcheck_status == 0) by (service) > 0
 172         for: 15m
 173         labels:
 174           alertgroup: fastly
 175       - alert: multiple fastly healthchecks failing
 176         expr: count(fastly_healthcheck_status == 0) by (service) > 4
 177         for: 5m
 178         labels:
 179           alertgroup: fastly
 180   - name: filesystem
 181     rules:
 182       - alert: readonly filesystem
 183         expr: node_filesystem_readonly > min_over_time(node_filesystem_readonly[7d])
 184         for: 0m
 185         labels:
 186           alertgroup: "{{ $labels.instance }}"
 187       - alert: filesystem low on space
 188         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 189         for: 5m
 190         labels:
 191           alertgroup: "{{ $labels.instance }}"
 192         annotations:
 193           percentage_free: "{{ $value | humanizePercentage }}"
 194           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 195           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 196       - alert: filesystem low on inodes
 197         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 198         for: 5m
 199         labels:
 200           alertgroup: "{{ $labels.instance }}"
 201         annotations:
 202           percentage_free: "{{ $value | humanizePercentage }}"
 203           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 204           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 205   - name: hwmon
 206     rules:
 207       - alert: hwmon fan alarm
 208         expr: node_hwmon_fan_alarm == 1
 209         for: 5m
 210         labels:
 211           alertgroup: "{{ $labels.instance }}"
 212         annotations:
 213           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}}"
 214           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}}"
 215       - alert: hwmon temperature alarm
 216         expr: node_hwmon_temp_alarm == 1
 217         for: 5m
 218         labels:
 219           alertgroup: "{{ $labels.instance }}"
 220         annotations:
 221           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}}"
 222           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}}"
 223           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}}"
 224       - alert: hwmon voltage alarm
 225         expr: node_hwmon_in_alarm == 1
 226         for: 5m
 227         labels:
 228           alertgroup: "{{ $labels.instance }}"
 229         annotations:
 230           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}}"
 231           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}}"
 232           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}}"
 233   - name: io
 234     rules:
 235       - alert: io pressure
 236         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 237         for: 60m
 238         labels:
 239           alertgroup: "{{ $labels.instance }}"
 240         annotations:
 241           pressure: "{{ $value | humanizePercentage }}"
 242   - name: ipmi
 243     rules:
 244       - alert: ipmi fan alarm
 245         expr: ipmi_fan_speed_state > 0
 246         for: 5m
 247         labels:
 248           alertgroup: "{{ $labels.instance }}"
 249         annotations:
 250           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 251       - alert: ipmi temperature alarm
 252         expr: ipmi_temperature_state > 0
 253         for: 5m
 254         labels:
 255           alertgroup: "{{ $labels.instance }}"
 256         annotations:
 257           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 258       - alert: ipmi voltage alarm
 259         expr: ipmi_voltage_state > 0
 260         for: 5m
 261         labels:
 262           alertgroup: "{{ $labels.instance }}"
 263         annotations:
 264           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 265       - alert: ipmi power alarm
 266         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 267         for: 5m
 268         labels:
 269           alertgroup: "{{ $labels.instance }}"
 270   - name: juniper
 271     rules:
 272       - alert: juniper cpu alarm
 273         expr: jnxOperatingCPU{jnxOperatingContentsIndex="7"} > 30
 274         for: 5m
 275         labels:
 276           alertgroup: "{{ $labels.site }}"
 277       - alert: juniper fan alarm
 278         expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!~"running.*"} > 0
 279         for: 5m
 280         labels:
 281           alertgroup: "{{ $labels.site }}"
 282       - alert: juniper power alarm
 283         expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!~"running.*"} > 0
 284         for: 5m
 285         labels:
 286           alertgroup: "{{ $labels.site }}"
 287   - name: mail
 288     rules:
 289       - alert: exim down
 290         expr: exim_up == 0
 291         for: 5m
 292         labels:
 293           alertgroup: "{{ $labels.instance }}"
 294       - alert: exim queue length
 295         expr: exim_queue > exim_queue_limit
 296         for: 60m
 297         labels:
 298           alertgroup: mail
 299         annotations:
 300           queue_length: "{{ $value }}"
 301       - alert: mailman queue length
 302         expr: mailman_queue_length > 200
 303         for: 60m
 304         labels:
 305           alertgroup: mail
 306         annotations:
 307           queue_length: "{{ $value }}"
 308   - name: mdadm
 309     rules:
 310       - alert: mdadm array inactive
 311         expr: node_md_state{state="inactive"} > 0
 312         for: 0m
 313         labels:
 314           alertgroup: "{{ $labels.instance }}"
 315         annotations:
 316           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 317           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 318           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 319           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 320       - alert: mdadm array degraded
 321         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 322         for: 0m
 323         labels:
 324           alertgroup: "{{ $labels.instance }}"
 325         annotations:
 326           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 327           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 328           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 329           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 330       - alert: mdadm disk failed
 331         expr: node_md_disks{state="failed"} > 0
 332         for: 0m
 333         labels:
 334           alertgroup: "{{ $labels.instance }}"
 335         annotations:
 336           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 337           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 338           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 339           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 340   - name: memory
 341     rules:
 342       - alert: low memory
 343         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 344         for: 15m
 345         labels:
 346           alertgroup: "{{ $labels.instance }}"
 347         annotations:
 348           memory_free: "{{ $value | humanizePercentage }}"
 349       - alert: memory pressure
 350         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 351         for: 60m
 352         labels:
 353           alertgroup: "{{ $labels.instance }}"
 354         annotations:
 355           pressure: "{{ $value | humanizePercentage }}"
 356       - alert: oom kill detected
 357         expr: increase(node_vmstat_oom_kill[1m]) > 0
 358         for: 0m
 359         labels:
 360           alertgroup: "{{ $labels.instance }}"
 361         annotations:
 362           new_oom_kills: "{{ $value }}"
 363   - name: mysql
 364     rules:
 365       - alert: mysql down
 366         expr: mysql_up == 0
 367         for: 1m
 368         labels:
 369           alertgroup: "{{ $labels.instance }}"
 370       - alert: mysql connection limit
 371         expr: mysql_global_status_max_used_connections / mysql_global_variables_max_connections > 0.8
 372         for: 1m
 373         labels:
 374           alertgroup: "{{ $labels.instance }}"
 375         annotations:
 376           connections_used: "{{ $value | humanizePercentage }}"
 377   - name: network
 378     rules:
 379       - alert: interface transmit rate
 380         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 381         for: 5m
 382         labels:
 383           alertgroup: "{{ $labels.instance }}"
 384         annotations:
 385           bandwidth_used: "{{ $value | humanizePercentage }}"
 386       - alert: interface receive rate
 387         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.98
 388         for: 5m
 389         labels:
 390           alertgroup: "{{ $labels.instance }}"
 391         annotations:
 392           bandwidth_used: "{{ $value | humanizePercentage }}"
 393       - alert: interface transmit errors
 394         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 395         for: 5m
 396         labels:
 397           alertgroup: "{{ $labels.instance }}"
 398         annotations:
 399           error_rate: "{{ $value | humanizePercentage }}"
 400       - alert: wireguard interface transmit errors
 401         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 402         for: 1h
 403         labels:
 404           alertgroup: "{{ $labels.instance }}"
 405         annotations:
 406           error_rate: "{{ $value | humanizePercentage }}"
 407       - alert: interface receive errors
 408         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 409         for: 5m
 410         labels:
 411           alertgroup: "{{ $labels.instance }}"
 412         annotations:
 413           error_rate: "{{ $value | humanizePercentage }}"
 414       - alert: conntrack entries
 415         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 416         for: 5m
 417         labels:
 418           alertgroup: "{{ $labels.instance }}"
 419         annotations:
 420           entries_used: "{{ $value | humanizePercentage }}"
 421   - name: nominatim
 422     rules:
 423       - alert: nominatim replication delay
 424         expr: nominatim_replication_delay > 10800
 425         for: 1h
 426         labels:
 427           alertgroup: nominatim
 428         annotations:
 429           delay: "{{ $value | humanizeDuration }}"
 430   - name: overpass
 431     rules:
 432       - alert: overpass osm database age
 433         expr: overpass_database_age_seconds{database="osm"} > 3600
 434         for: 1h
 435         labels:
 436           alertgroup: overpass
 437         annotations:
 438           age: "{{ $value | humanizeDuration }}"
 439       - alert: overpass area database age
 440         expr: overpass_database_age_seconds{database="area"} > 86400
 441         for: 1h
 442         labels:
 443           alertgroup: overpass
 444         annotations:
 445           age: "{{ $value | humanizeDuration }}"
 446   - name: passenger
 447     rules:
 448       - alert: passenger down
 449         expr: passenger_up == 0
 450         for: 5m
 451         labels:
 452           alertgroup: "{{ $labels.instance }}"
 453       - alert: passenger queuing
 454         expr: passenger_top_level_request_queue > 0
 455         for: 5m
 456         labels:
 457           alertgroup: "{{ $labels.instance }}"
 458       - alert: passenger application queuing
 459         expr: passenger_app_request_queue > 0
 460         for: 5m
 461         labels:
 462           alertgroup: "{{ $labels.instance }}"
 463   - name: planet
 464     rules:
 465       - alert: planet dump overdue
 466         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 467         for: 24h
 468         labels:
 469           alertgroup: planet
 470         annotations:
 471           overdue_by: "{{ $value | humanizeDuration }}"
 472       - alert: notes dump overdue
 473         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 474         for: 6h
 475         labels:
 476           alertgroup: planet
 477         annotations:
 478           overdue_by: "{{ $value | humanizeDuration }}"
 479       - alert: daily replication feed delayed
 480         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 481         for: 3h
 482         labels:
 483           alertgroup: planet
 484         annotations:
 485           delayed_by: "{{ $value | humanizeDuration }}"
 486       - alert: hourly replication feed delayed
 487         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 488         for: 30m
 489         labels:
 490           alertgroup: planet
 491         annotations:
 492           delayed_by: "{{ $value | humanizeDuration }}"
 493       - alert: minutely replication feed delayed
 494         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 495         for: 5m
 496         labels:
 497           alertgroup: planet
 498         annotations:
 499           delayed_by: "{{ $value | humanizeDuration }}"
 500       - alert: changeset replication feed delayed
 501         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 502         for: 5m
 503         labels:
 504           alertgroup: planet
 505         annotations:
 506           delayed_by: "{{ $value | humanizeDuration }}"
 507   - name: postgresql
 508     rules:
 509       - alert: postgresql down
 510         expr: pg_up == 0
 511         for: 1m
 512         labels:
 513           alertgroup: "{{ $labels.instance }}"
 514       - alert: postgresql replication delay
 515         expr: pg_replication_lag_seconds > 30
 516         for: 15m
 517         labels:
 518           alertgroup: "{{ $labels.instance }}"
 519         annotations:
 520           delay: "{{ $value | humanizeDuration }}"
 521       - alert: postgresql connection limit
 522         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 523         for: 1m
 524         labels:
 525           alertgroup: "{{ $labels.instance }}"
 526         annotations:
 527           connections_used: "{{ $value | humanizePercentage }}"
 528       - alert: postgresql deadlocks
 529         expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
 530         for: 0m
 531         labels:
 532           alertgroup: "{{ $labels.instance }}"
 533         annotations:
 534           new_deadlocks: "{{ $value }}"
 535       - alert: postgresql slow queries
 536         expr: pg_slow_queries > 0
 537         for: 5m
 538         labels:
 539           alertgroup: "{{ $labels.instance }}"
 540         annotations:
 541           queries: "{{ $value }}"
 542   - name: prometheus
 543     rules:
 544       - alert: prometheus configuration error
 545         expr: prometheus_config_last_reload_successful == 0
 546         for: 10m
 547         labels:
 548           alertgroup: "prometheus"
 549       - alert: prometheus target missing
 550         expr: up == 0
 551         for: 10m
 552         labels:
 553           alertgroup: "prometheus"
 554   - name: raid
 555     rules:
 556       - alert: raid controller battery failed
 557         expr: ohai_controller_info{battery_status="failed"} > 0
 558         for: 5m
 559         labels:
 560           alertgroup: "{{ $labels.instance }}"
 561       - alert: raid array degraded
 562         expr: ohai_array_info{status="degraded"} > 0
 563         for: 5m
 564         labels:
 565           alertgroup: "{{ $labels.instance }}"
 566       - alert: raid disk failed
 567         expr: ohai_disk_info{status="failed"} > 0
 568         for: 5m
 569         labels:
 570           alertgroup: "{{ $labels.instance }}"
 571   - name: rasdaemon
 572     rules:
 573       - alert: memory controller errors
 574         expr: increase(rasdaemon_mc_events_total[1m]) > 0
 575         for: 0m
 576         labels:
 577           alertgroup: "{{ $labels.instance }}"
 578         annotations:
 579           new_errors: "{{ $value }}"
 580       - alert: pcie aer errors
 581         expr: increase(rasdaemon_aer_events_total[1m]) > 0
 582         for: 0m
 583         labels:
 584           alertgroup: "{{ $labels.instance }}"
 585         annotations:
 586           new_errors: "{{ $value }}"
 587   - name: smart
 588     rules:
 589       - alert: smart failure
 590         expr: smart_health_status == 0
 591         for: 60m
 592         labels:
 593           alertgroup: "{{ $labels.instance }}"
 594       - alert: smart ssd wearout approaching
 595         expr: smart_percentage_used / 100 >= 0.8
 596         for: 60m
 597         labels:
 598           alertgroup: "{{ $labels.instance }}"
 599         annotations:
 600           percentage_used: "{{ $value | humanizePercentage }}"
 601   - name: smokeping
 602     rules:
 603       - alert: packet loss
 604         expr: 1 - (rate(smokeping_response_duration_seconds_count[5m]) / rate(smokeping_requests_total[5m])) > 0.02
 605         for: 10m
 606         labels:
 607           alertgroup: smokeping
 608         annotations:
 609           loss_rate: "{{ $value | humanizePercentage }}"
 610   - name: snmp
 611     rules:
 612       - alert: snmp pdus missing
 613         expr: max_over_time(snmp_scrape_pdus_returned[1d]) - snmp_scrape_pdus_returned > 0
 614         for: 15m
 615         labels:
 616           alertgroup: snmp
 617         annotations:
 618           missing_pdus: "{{ $value }}"
 619   - name: ssl
 620     rules:
 621       - alert: ssl certificate probe failed
 622         expr: ssl_probe_success == 0
 623         for: 60m
 624         labels:
 625           alertgroup: ssl
 626       - alert: ssl certificate expiry
 627         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 628         for: 0m
 629         labels:
 630           alertgroup: ssl
 631         annotations:
 632           expires_in: "{{ $value | humanizeDuration }}"
 633       - alert: ssl certificate revoked
 634         expr: ssl_ocsp_response_status == 1
 635         for: 0m
 636         labels:
 637           alertgroup: ssl
 638       - alert: ocsp status unknown
 639         expr: ssl_ocsp_response_status == 1
 640         for: 0m
 641         labels:
 642           alertgroup: ssl
 643   - name: statuscake
 644     rules:
 645       - alert: statuscake uptime check failing
 646         expr: statuscake_paused == 0 and statuscake_up == 0
 647         for: 10m
 648         labels:
 649           alertgroup: statuscake
 650   - name: systemd
 651     rules:
 652       - alert: systemd failed service
 653         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 654         for: 5m
 655         labels:
 656           alertgroup: "{{ $labels.instance }}"
 657       - alert: systemd failed chef client service
 658         expr: sum_over_time(node_systemd_unit_state{state="inactive",name="chef-client.service"}[6h]) == 0
 659         for: 0m
 660         labels:
 661           alertgroup: "{{ $labels.instance }}"
 662   - name: tile
 663     rules:
 664       - alert: renderd replication delay
 665         expr: renderd_replication_delay > 120
 666         for: 15m
 667         labels:
 668           alertgroup: tile
 669         annotations:
 670           delay: "{{ $value | humanizeDuration }}"
 671       - alert: missed tile rate
 672         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 673         for: 5m
 674         labels:
 675           alertgroup: tile
 676         annotations:
 677           miss_rate: "{{ $value | humanizePercentage }}"
 678       - alert: tile render rate
 679         expr: sum(rate(renderd_zoom_metatiles_total[5m])) by (instance) == 0
 680         for: 15m
 681         labels:
 682           alertgroup: tile
 683         annotations:
 684           render_rate: "{{ $value }} tiles/s"
 685   - name: time
 686     rules:
 687       - alert: clock not synchronising
 688         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 689         for: 5m
 690         labels:
 691           alertgroup: "{{ $labels.instance }}"
 692       - alert: clock skew detected
 693         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)
 694         for: 5m
 695         labels:
 696           alertgroup: "{{ $labels.instance }}"
 697         annotations:
 698           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 699   - name: web
 700     rules:
 701       - alert: web error rate
 702         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
 703         for: 5m
 704         labels:
 705           alertgroup: web
 706         annotations:
 707           error_rate: "{{ $value | humanizePercentage }}"
 708       - alert: job processing rate
 709         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
 710         for: 15m
 711         labels:
 712           alertgroup: web
 713         annotations:
 714           job_processing_rate: "{{ $value | humanizePercentage }}"