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: jnxOperating5MinLoadAvg{jnxOperatingContentsIndex="9"} / 200 > 0.5
 253         for: 5m
 254         labels:
 255           alertgroup: "{{ $labels.site }}"
 256         annotations:
 257           load_average: "{{ $value | humanizePercentage }}"
 258       - alert: juniper fan alarm
 259         expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!~"running.*"} > 0
 260         for: 5m
 261         labels:
 262           alertgroup: "{{ $labels.site }}"
 263       - alert: juniper power alarm
 264         expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!~"running.*"} > 0
 265         for: 5m
 266         labels:
 267           alertgroup: "{{ $labels.site }}"
 268   - name: mail
 269     rules:
 270       - alert: exim down
 271         expr: exim_up == 0
 272         for: 5m
 273         labels:
 274           alertgroup: "{{ $labels.instance }}"
 275       - alert: exim queue length
 276         expr: exim_queue > ignoring(job) exim_queue_limit
 277         for: 60m
 278         labels:
 279           alertgroup: mail
 280         annotations:
 281           queue_length: "{{ $value }}"
 282       - alert: mailman queue length
 283         expr: mailman_queue_length > 200
 284         for: 60m
 285         labels:
 286           alertgroup: mail
 287         annotations:
 288           queue_length: "{{ $value }}"
 289   - name: mdadm
 290     rules:
 291       - alert: mdadm array inactive
 292         expr: node_md_state{state="inactive"} > 0
 293         for: 0m
 294         labels:
 295           alertgroup: "{{ $labels.instance }}"
 296         annotations:
 297           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 298           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 299           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 300           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 301       - alert: mdadm array degraded
 302         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 303         for: 0m
 304         labels:
 305           alertgroup: "{{ $labels.instance }}"
 306         annotations:
 307           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 308           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 309           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 310           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 311       - alert: mdadm disk failed
 312         expr: node_md_disks{state="failed"} > 0
 313         for: 0m
 314         labels:
 315           alertgroup: "{{ $labels.instance }}"
 316         annotations:
 317           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 318           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 319           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 320           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 321   - name: memory
 322     rules:
 323       - alert: low memory
 324         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 325         for: 15m
 326         labels:
 327           alertgroup: "{{ $labels.instance }}"
 328         annotations:
 329           memory_free: "{{ $value | humanizePercentage }}"
 330       - alert: memory pressure
 331         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 332         for: 60m
 333         labels:
 334           alertgroup: "{{ $labels.instance }}"
 335         annotations:
 336           pressure: "{{ $value | humanizePercentage }}"
 337       - alert: oom kill detected
 338         expr: increase(node_vmstat_oom_kill[1m]) > 0
 339         for: 0m
 340         labels:
 341           alertgroup: "{{ $labels.instance }}"
 342         annotations:
 343           new_oom_kills: "{{ $value }}"
 344   - name: mysql
 345     rules:
 346       - alert: mysql down
 347         expr: mysql_up == 0
 348         for: 1m
 349         labels:
 350           alertgroup: "{{ $labels.instance }}"
 351       - alert: mysql connection limit
 352         expr: mysql_global_status_max_used_connections / mysql_global_variables_max_connections > 0.8
 353         for: 1m
 354         labels:
 355           alertgroup: "{{ $labels.instance }}"
 356         annotations:
 357           connections_used: "{{ $value | humanizePercentage }}"
 358   - name: network
 359     rules:
 360       - alert: interface transmit rate
 361         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.99
 362         for: 5m
 363         labels:
 364           alertgroup: "{{ $labels.instance }}"
 365         annotations:
 366           bandwidth_used: "{{ $value | humanizePercentage }}"
 367       - alert: interface receive rate
 368         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.99
 369         for: 5m
 370         labels:
 371           alertgroup: "{{ $labels.instance }}"
 372         annotations:
 373           bandwidth_used: "{{ $value | humanizePercentage }}"
 374       - alert: interface transmit errors
 375         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 376         for: 5m
 377         labels:
 378           alertgroup: "{{ $labels.instance }}"
 379         annotations:
 380           error_rate: "{{ $value | humanizePercentage }}"
 381       - alert: wireguard interface transmit errors
 382         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 383         for: 1h
 384         labels:
 385           alertgroup: "{{ $labels.instance }}"
 386         annotations:
 387           error_rate: "{{ $value | humanizePercentage }}"
 388       - alert: interface receive errors
 389         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 390         for: 5m
 391         labels:
 392           alertgroup: "{{ $labels.instance }}"
 393         annotations:
 394           error_rate: "{{ $value | humanizePercentage }}"
 395       - alert: conntrack entries
 396         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 397         for: 5m
 398         labels:
 399           alertgroup: "{{ $labels.instance }}"
 400         annotations:
 401           entries_used: "{{ $value | humanizePercentage }}"
 402   - name: nominatim
 403     rules:
 404       - alert: nominatim replication delay
 405         expr: nominatim_replication_delay > 10800
 406         for: 1h
 407         labels:
 408           alertgroup: nominatim
 409         annotations:
 410           delay: "{{ $value | humanizeDuration }}"
 411   - name: overpass
 412     rules:
 413       - alert: overpass osm database age
 414         expr: overpass_database_age_seconds{database="osm"} > 3600
 415         for: 1h
 416         labels:
 417           alertgroup: overpass
 418         annotations:
 419           age: "{{ $value | humanizeDuration }}"
 420       - alert: overpass area database age
 421         expr: overpass_database_age_seconds{database="area"} > 86400
 422         for: 1h
 423         labels:
 424           alertgroup: overpass
 425         annotations:
 426           age: "{{ $value | humanizeDuration }}"
 427   - name: passenger
 428     rules:
 429       - alert: passenger down
 430         expr: passenger_up == 0
 431         for: 5m
 432         labels:
 433           alertgroup: "{{ $labels.instance }}"
 434       - alert: passenger queuing
 435         expr: passenger_top_level_request_queue > 0
 436         for: 5m
 437         labels:
 438           alertgroup: "{{ $labels.instance }}"
 439       - alert: passenger application queuing
 440         expr: passenger_app_request_queue > 0
 441         for: 5m
 442         labels:
 443           alertgroup: "{{ $labels.instance }}"
 444   - name: planet
 445     rules:
 446       - alert: planet dump overdue
 447         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 448         for: 24h
 449         labels:
 450           alertgroup: planet
 451         annotations:
 452           overdue_by: "{{ $value | humanizeDuration }}"
 453       - alert: notes dump overdue
 454         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 455         for: 6h
 456         labels:
 457           alertgroup: planet
 458         annotations:
 459           overdue_by: "{{ $value | humanizeDuration }}"
 460       - alert: daily replication feed delayed
 461         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 462         for: 3h
 463         labels:
 464           alertgroup: planet
 465         annotations:
 466           delayed_by: "{{ $value | humanizeDuration }}"
 467       - alert: hourly replication feed delayed
 468         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 469         for: 30m
 470         labels:
 471           alertgroup: planet
 472         annotations:
 473           delayed_by: "{{ $value | humanizeDuration }}"
 474       - alert: minutely replication feed delayed
 475         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 476         for: 5m
 477         labels:
 478           alertgroup: planet
 479         annotations:
 480           delayed_by: "{{ $value | humanizeDuration }}"
 481       - alert: changeset replication feed delayed
 482         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 483         for: 5m
 484         labels:
 485           alertgroup: planet
 486         annotations:
 487           delayed_by: "{{ $value | humanizeDuration }}"
 488   - name: postgresql
 489     rules:
 490       - alert: postgresql down
 491         expr: pg_up == 0
 492         for: 1m
 493         labels:
 494           alertgroup: "{{ $labels.instance }}"
 495       - alert: postgresql replication delay
 496         expr: pg_replication_lag_seconds > 30
 497         for: 15m
 498         labels:
 499           alertgroup: "{{ $labels.instance }}"
 500         annotations:
 501           delay: "{{ $value | humanizeDuration }}"
 502       - alert: postgresql connection limit
 503         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 504         for: 1m
 505         labels:
 506           alertgroup: "{{ $labels.instance }}"
 507         annotations:
 508           connections_used: "{{ $value | humanizePercentage }}"
 509       - alert: postgresql deadlocks
 510         expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
 511         for: 0m
 512         labels:
 513           alertgroup: "{{ $labels.instance }}"
 514         annotations:
 515           new_deadlocks: "{{ $value }}"
 516       - alert: postgresql slow queries
 517         expr: pg_slow_queries > 0
 518         for: 5m
 519         labels:
 520           alertgroup: "{{ $labels.instance }}"
 521         annotations:
 522           queries: "{{ $value }}"
 523       - alert: postgresql idle transactions
 524         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)
 525         for: 5m
 526         labels:
 527           alertgroup: "{{ $labels.instance }}"
 528         annotations:
 529           queries: "{{ $value }}"
 530   - name: prometheus
 531     rules:
 532       - alert: prometheus configuration error
 533         expr: prometheus_config_last_reload_successful == 0
 534         for: 10m
 535         labels:
 536           alertgroup: "prometheus"
 537       - alert: prometheus target missing
 538         expr: up == 0
 539         for: 10m
 540         labels:
 541           alertgroup: "prometheus"
 542   - name: raid
 543     rules:
 544       - alert: raid controller battery failed
 545         expr: ohai_controller_info{battery_status="failed"} > 0
 546         for: 5m
 547         labels:
 548           alertgroup: "{{ $labels.instance }}"
 549       - alert: raid controller battery recharging
 550         expr: ohai_controller_info{battery_status="recharging"} > 0
 551         for: 4h
 552         labels:
 553           alertgroup: "{{ $labels.instance }}"
 554       - alert: raid array degraded
 555         expr: ohai_array_info{status="degraded"} > 0
 556         for: 5m
 557         labels:
 558           alertgroup: "{{ $labels.instance }}"
 559       - alert: raid disk failed
 560         expr: ohai_disk_info{status="failed"} > 0
 561         for: 5m
 562         labels:
 563           alertgroup: "{{ $labels.instance }}"
 564   - name: rasdaemon
 565     rules:
 566       - alert: memory controller errors
 567         expr: increase(rasdaemon_mc_events_total[1m]) > 0
 568         for: 0m
 569         labels:
 570           alertgroup: "{{ $labels.instance }}"
 571         annotations:
 572           new_errors: "{{ $value }}"
 573       - alert: pcie aer errors
 574         expr: increase(rasdaemon_aer_events_total[1m]) > 0
 575         for: 0m
 576         labels:
 577           alertgroup: "{{ $labels.instance }}"
 578         annotations:
 579           new_errors: "{{ $value }}"
 580   - name: smart
 581     rules:
 582       - alert: smart failure
 583         expr: smart_health_status == 0
 584         for: 60m
 585         labels:
 586           alertgroup: "{{ $labels.instance }}"
 587       - alert: smart ssd wearout approaching
 588         expr: smart_percentage_used / 100 >= 0.8
 589         for: 60m
 590         labels:
 591           alertgroup: "{{ $labels.instance }}"
 592         annotations:
 593           percentage_used: "{{ $value | humanizePercentage }}"
 594   - name: smokeping
 595     rules:
 596       - alert: packet loss
 597         expr: 1 - (rate(smokeping_response_duration_seconds_count[5m]) / rate(smokeping_requests_total[5m])) > 0.02
 598         for: 10m
 599         labels:
 600           alertgroup: smokeping
 601         annotations:
 602           loss_rate: "{{ $value | humanizePercentage }}"
 603   - name: snmp
 604     rules:
 605       - alert: snmp pdus missing
 606         expr: max_over_time(snmp_scrape_pdus_returned[1d]) - snmp_scrape_pdus_returned > 0
 607         for: 15m
 608         labels:
 609           alertgroup: snmp
 610         annotations:
 611           missing_pdus: "{{ $value }}"
 612   - name: ssl
 613     rules:
 614       - alert: ssl certificate probe failed
 615         expr: ssl_probe_success == 0
 616         for: 60m
 617         labels:
 618           alertgroup: ssl
 619       - alert: ssl certificate expiry
 620         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 621         for: 0m
 622         labels:
 623           alertgroup: ssl
 624         annotations:
 625           expires_in: "{{ $value | humanizeDuration }}"
 626       - alert: ssl certificate revoked
 627         expr: ssl_ocsp_response_status == 1
 628         for: 0m
 629         labels:
 630           alertgroup: ssl
 631       - alert: ocsp status unknown
 632         expr: ssl_ocsp_response_status == 1
 633         for: 0m
 634         labels:
 635           alertgroup: ssl
 636   - name: statuscake
 637     rules:
 638       - alert: statuscake uptime check failing
 639         expr: statuscake_paused == 0 and statuscake_up == 0
 640         for: 10m
 641         labels:
 642           alertgroup: statuscake
 643   - name: systemd
 644     rules:
 645       - alert: systemd failed service
 646         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 647         for: 5m
 648         labels:
 649           alertgroup: "{{ $labels.instance }}"
 650       - alert: systemd failed chef client service
 651         expr: sum_over_time(node_systemd_unit_state{state="inactive",name="chef-client.service"}[6h]) == 0
 652         for: 0m
 653         labels:
 654           alertgroup: "{{ $labels.instance }}"
 655   - name: taginfo
 656     rules:
 657       - alert: taginfo planet age
 658         expr: time() - taginfo_data_from_seconds > 129600
 659         for: 0m
 660         labels:
 661           alertgroup: taginfo
 662         annotations:
 663           age: "{{ $value | humanizeDuration }}"
 664       - alert: taginfo database age
 665         expr: time() - taginfo_database_update_finish_seconds > 129600
 666         for: 0m
 667         labels:
 668           alertgroup: taginfo
 669         annotations:
 670           age: "{{ $value | humanizeDuration }}"
 671       - alert: taginfo database size
 672         expr: abs(delta(taginfo_database_size_bytes[30m])) / taginfo_database_size_bytes > 0.1
 673         for: 30m
 674         labels:
 675           alertgroup: taginfo
 676         annotations:
 677           size_change: "{{ $value | humanizePercentage }}"
 678   - name: tile
 679     rules:
 680       - alert: renderd replication delay
 681         expr: renderd_replication_delay > 120
 682         for: 15m
 683         labels:
 684           alertgroup: tile
 685         annotations:
 686           delay: "{{ $value | humanizeDuration }}"
 687       - alert: missed tile rate
 688         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 689         for: 5m
 690         labels:
 691           alertgroup: tile
 692         annotations:
 693           miss_rate: "{{ $value | humanizePercentage }}"
 694       - alert: tile render rate
 695         expr: sum(rate(renderd_zoom_metatiles_total[5m])) by (instance) == 0
 696         for: 15m
 697         labels:
 698           alertgroup: tile
 699         annotations:
 700           render_rate: "{{ $value }} tiles/s"
 701   - name: time
 702     rules:
 703       - alert: clock not synchronising
 704         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 705         for: 5m
 706         labels:
 707           alertgroup: "{{ $labels.instance }}"
 708       - alert: clock skew detected
 709         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)
 710         for: 5m
 711         labels:
 712           alertgroup: "{{ $labels.instance }}"
 713         annotations:
 714           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 715   - name: web
 716     rules:
 717       - alert: web error rate
 718         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
 719         for: 5m
 720         labels:
 721           alertgroup: web
 722         annotations:
 723           error_rate: "{{ $value | humanizePercentage }}"
 724       - alert: job processing rate
 725         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
 726         for: 1h
 727         labels:
 728           alertgroup: web
 729         annotations:
 730           job_processing_rate: "{{ $value | humanizePercentage }}"