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