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

   1 # DO NOT EDIT - This file is being maintained by Chef
   2
   3 groups:
   4   - name: amsterdam
   5     rules:
   6       - alert: uplink
   7         expr: ifOperStatus{site="amsterdam",ifName=~"ge-[01]/2/2"} != 1
   8         for: 6m
   9         labels:
  10           alertgroup: "amsterdam"
  11         annotations:
  12           status: "{{ $value }}"
  13       - alert: pdu current draw
  14         expr: rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10 > 28
  15         for: 6m
  16         labels:
  17           alertgroup: "amsterdam"
  18         annotations:
  19           current: "{{ $value | humanize }}A"
  20       - alert: site power
  21         expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="amsterdam",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 3.5
  22         for: 6m
  23         labels:
  24           alertgroup: "amsterdam"
  25         annotations:
  26           current: "{{ $value | humanize }}kVA"
  27       - alert: site temperature
  28         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 26
  29         for: 6m
  30         labels:
  31           alertgroup: "amsterdam"
  32         annotations:
  33           temperature: "{{ $value | humanize }}C"
  34       - alert: site humidity
  35         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 > 0.65
  36         for: 6m
  37         labels:
  38           alertgroup: "amsterdam"
  39         annotations:
  40           humidity: "{{ $value | humanizePercentage }}"
  41   - name: apache
  42     rules:
  43       - alert: apache down
  44         expr: apache_up == 0
  45         for: 5m
  46         labels:
  47           alertgroup: "{{ $labels.instance }}"
  48       - alert: apache workers busy
  49         expr: sum(apache_workers{state="busy"}) by (instance) / sum(apache_scoreboard) by (instance) > 0.8
  50         for: 5m
  51         labels:
  52           alertgroup: "{{ $labels.instance }}"
  53         annotations:
  54           busy_workers: "{{ $value | humanizePercentage }}"
  55   - name: chef
  56     rules:
  57       - alert: chef client not running
  58         expr: time() - node_systemd_timer_last_trigger_seconds{name="chef-client.timer"} > 3600
  59         for: 12h
  60         labels:
  61           alertgroup: "{{ $labels.instance }}"
  62         annotations:
  63           down_time: "{{ $value | humanizeDuration }}"
  64   - name: cisco
  65     rules:
  66       - alert: cisco fan alarm
  67         expr: rlPhdUnitEnvParamFan1Status{rlPhdUnitEnvParamFan1Status!="normal"} > 0 or rlPhdUnitEnvParamFan2Status{rlPhdUnitEnvParamFan2Status!="normal"} > 0
  68         for: 5m
  69         labels:
  70           alertgroup: "{{ $labels.site }}"
  71         annotations:
  72           fan_rpm: "{{ with printf \"rlPhdUnitEnvParamFan1Speed{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}rpm{{end}}"
  73       - alert: cisco temperature alarm
  74         expr: rlPhdUnitEnvParamTempSensorStatus{rlPhdUnitEnvParamTempSensorStatus!="ok"} > 0
  75         for: 5m
  76         labels:
  77           alertgroup: "{{ $labels.site }}"
  78         annotations:
  79           temp_celsius: "{{ with printf \"rlPhdUnitEnvParamTempSensorValue{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}C{{end}}"
  80       - alert: cisco main power alarm
  81         expr: rlPhdUnitEnvParamMainPSStatus{rlPhdUnitEnvParamMainPSStatus!="normal"} > 0
  82         for: 5m
  83         labels:
  84           alertgroup: "{{ $labels.site }}"
  85       - alert: cisco redundant power alarm
  86         expr: rlPhdUnitEnvParamRedundantPSStatus{rlPhdUnitEnvParamRedundantPSStatus!="normal"} > 0
  87         for: 5m
  88         labels:
  89           alertgroup: "{{ $labels.site }}"
  90   - name: cpu
  91     rules:
  92       - alert: cpu pressure
  93         expr: rate(node_pressure_cpu_waiting_seconds_total[5m]) > 0.75
  94         for: 60m
  95         labels:
  96           alertgroup: "{{ $labels.instance }}"
  97         annotations:
  98           pressure: "{{ $value | humanizePercentage }}"
  99   - name: database
 100     rules:
 101       - alert: postgres replication delay
 102         expr: pg_replication_lag_seconds > 30
 103         for: 15m
 104         labels:
 105           alertgroup: database
 106         annotations:
 107           delay: "{{ $value | humanizeDuration }}"
 108   - name: discourse
 109     rules:
 110       - alert: discourse job failure rate
 111         expr: rate(discourse_job_failures[5m]) > 0
 112         for: 5m
 113         labels:
 114           alertgroup: discourse
 115         annotations:
 116           failure_rate: "{{ $value }} jobs/s"
 117   - name: dublin
 118     rules:
 119       - alert: uplink
 120         expr: ifOperStatus{site="dublin",ifName=~"ge-[01]/2/2"} != 1
 121         for: 6m
 122         labels:
 123           alertgroup: "dublin"
 124         annotations:
 125           status: "{{ $value }}"
 126       - alert: pdu current draw
 127         expr: rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10 > 28
 128         for: 6m
 129         labels:
 130           alertgroup: "dublin"
 131         annotations:
 132           current: "{{ $value | humanize }}A"
 133       - alert: site power
 134         expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="dublin",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 4
 135         for: 6m
 136         labels:
 137           alertgroup: "dublin"
 138         annotations:
 139           current: "{{ $value | humanize }}kVA"
 140       - alert: site temperature
 141         expr: min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 > 26
 142         for: 6m
 143         labels:
 144           alertgroup: "dublin"
 145         annotations:
 146           temperature: "{{ $value | humanize }}C"
 147       - alert: site humidity
 148         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 > 0.65
 149         for: 6m
 150         labels:
 151           alertgroup: "dublin"
 152         annotations:
 153           humidity: "{{ $value | humanizePercentage }}"
 154   - name: fastly
 155     rules:
 156       - alert: fastly error rate
 157         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
 158         for: 15m
 159         labels:
 160           alertgroup: fastly
 161         annotations:
 162           error_rate: "{{ $value | humanizePercentage }}"
 163       - alert: fastly frontend healthcheck warning
 164         expr: count(fastly_healthcheck_status == 0) by (service, datacenter) > 2
 165         for: 15m
 166         labels:
 167           alertgroup: fastly
 168       - alert: fastly frontend healthcheck critical
 169         expr: count(fastly_healthcheck_status == 0) by (service, datacenter) == count(fastly_healthcheck_status) by (service, datacenter)
 170         for: 5m
 171         labels:
 172           alertgroup: fastly
 173       - alert: fastly backend healthcheck warning
 174         expr: count(fastly_healthcheck_status == 0) by (service, backend) > 10
 175         for: 15m
 176         labels:
 177           alertgroup: fastly
 178       - alert: fastly backend healthcheck critical
 179         expr: count(fastly_healthcheck_status == 0) by (service, backend) == count(fastly_healthcheck_status) by (service, backend)
 180         for: 5m
 181         labels:
 182           alertgroup: fastly
 183   - name: filesystem
 184     rules:
 185       - alert: readonly filesystem
 186         expr: node_filesystem_readonly > min_over_time(node_filesystem_readonly[7d])
 187         for: 0m
 188         labels:
 189           alertgroup: "{{ $labels.instance }}"
 190       - alert: filesystem low on space
 191         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 192         for: 5m
 193         labels:
 194           alertgroup: "{{ $labels.instance }}"
 195         annotations:
 196           percentage_free: "{{ $value | humanizePercentage }}"
 197           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 198           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 199       - alert: filesystem low on inodes
 200         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 201         for: 5m
 202         labels:
 203           alertgroup: "{{ $labels.instance }}"
 204         annotations:
 205           percentage_free: "{{ $value | humanizePercentage }}"
 206           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 207           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 208   - name: hwmon
 209     rules:
 210       - alert: hwmon fan alarm
 211         expr: node_hwmon_fan_alarm == 1
 212         for: 5m
 213         labels:
 214           alertgroup: "{{ $labels.instance }}"
 215         annotations:
 216           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}}"
 217           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}}"
 218       - alert: hwmon temperature alarm
 219         expr: node_hwmon_temp_alarm == 1
 220         for: 5m
 221         labels:
 222           alertgroup: "{{ $labels.instance }}"
 223         annotations:
 224           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}}"
 225           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}}"
 226           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}}"
 227       - alert: hwmon voltage alarm
 228         expr: node_hwmon_in_alarm == 1
 229         for: 5m
 230         labels:
 231           alertgroup: "{{ $labels.instance }}"
 232         annotations:
 233           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}}"
 234           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}}"
 235           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}}"
 236   - name: io
 237     rules:
 238       - alert: io pressure
 239         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 240         for: 60m
 241         labels:
 242           alertgroup: "{{ $labels.instance }}"
 243         annotations:
 244           pressure: "{{ $value | humanizePercentage }}"
 245   - name: ipmi
 246     rules:
 247       - alert: ipmi fan alarm
 248         expr: ipmi_fan_speed_state > 0
 249         for: 5m
 250         labels:
 251           alertgroup: "{{ $labels.instance }}"
 252         annotations:
 253           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 254       - alert: ipmi temperature alarm
 255         expr: ipmi_temperature_state > 0
 256         for: 5m
 257         labels:
 258           alertgroup: "{{ $labels.instance }}"
 259         annotations:
 260           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 261       - alert: ipmi voltage alarm
 262         expr: ipmi_voltage_state > 0
 263         for: 5m
 264         labels:
 265           alertgroup: "{{ $labels.instance }}"
 266         annotations:
 267           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 268       - alert: ipmi power alarm
 269         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 270         for: 5m
 271         labels:
 272           alertgroup: "{{ $labels.instance }}"
 273   - name: juniper
 274     rules:
 275       - alert: juniper cpu alarm
 276         expr: jnxOperating5MinLoadAvg{jnxOperatingContentsIndex="9"} / 200 > 0.5
 277         for: 5m
 278         labels:
 279           alertgroup: "{{ $labels.site }}"
 280         annotations:
 281           load_average: "{{ $value | humanizePercentage }}"
 282       - alert: juniper fan alarm
 283         expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!~"running.*"} > 0
 284         for: 5m
 285         labels:
 286           alertgroup: "{{ $labels.site }}"
 287       - alert: juniper power alarm
 288         expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!~"running.*"} > 0
 289         for: 5m
 290         labels:
 291           alertgroup: "{{ $labels.site }}"
 292   - name: mail
 293     rules:
 294       - alert: exim down
 295         expr: exim_up == 0
 296         for: 5m
 297         labels:
 298           alertgroup: "{{ $labels.instance }}"
 299       - alert: exim queue length
 300         expr: exim_queue > ignoring(job) exim_queue_limit
 301         for: 60m
 302         labels:
 303           alertgroup: mail
 304         annotations:
 305           queue_length: "{{ $value }}"
 306       - alert: mailman queue length
 307         expr: mailman_queue_length > 200
 308         for: 60m
 309         labels:
 310           alertgroup: mail
 311         annotations:
 312           queue_length: "{{ $value }}"
 313   - name: mdadm
 314     rules:
 315       - alert: mdadm array inactive
 316         expr: node_md_state{state="inactive"} > 0
 317         for: 0m
 318         labels:
 319           alertgroup: "{{ $labels.instance }}"
 320         annotations:
 321           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 322           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 323           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 324           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 325       - alert: mdadm array degraded
 326         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 327         for: 0m
 328         labels:
 329           alertgroup: "{{ $labels.instance }}"
 330         annotations:
 331           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 332           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 333           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 334           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 335       - alert: mdadm disk failed
 336         expr: node_md_disks{state="failed"} > 0
 337         for: 0m
 338         labels:
 339           alertgroup: "{{ $labels.instance }}"
 340         annotations:
 341           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 342           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 343           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 344           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 345   - name: memory
 346     rules:
 347       - alert: low memory
 348         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 349         for: 15m
 350         labels:
 351           alertgroup: "{{ $labels.instance }}"
 352         annotations:
 353           memory_free: "{{ $value | humanizePercentage }}"
 354       - alert: memory pressure
 355         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 356         for: 60m
 357         labels:
 358           alertgroup: "{{ $labels.instance }}"
 359         annotations:
 360           pressure: "{{ $value | humanizePercentage }}"
 361       - alert: oom kill detected
 362         expr: increase(node_vmstat_oom_kill[1m]) > 0
 363         for: 0m
 364         labels:
 365           alertgroup: "{{ $labels.instance }}"
 366         annotations:
 367           new_oom_kills: "{{ $value }}"
 368   - name: mysql
 369     rules:
 370       - alert: mysql down
 371         expr: mysql_up == 0
 372         for: 1m
 373         labels:
 374           alertgroup: "{{ $labels.instance }}"
 375       - alert: mysql connection limit
 376         expr: mysql_global_status_max_used_connections / mysql_global_variables_max_connections > 0.8
 377         for: 1m
 378         labels:
 379           alertgroup: "{{ $labels.instance }}"
 380         annotations:
 381           connections_used: "{{ $value | humanizePercentage }}"
 382   - name: network
 383     rules:
 384       - alert: interface transmit rate
 385         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.99
 386         for: 5m
 387         labels:
 388           alertgroup: "{{ $labels.instance }}"
 389         annotations:
 390           bandwidth_used: "{{ $value | humanizePercentage }}"
 391       - alert: interface receive rate
 392         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.99
 393         for: 5m
 394         labels:
 395           alertgroup: "{{ $labels.instance }}"
 396         annotations:
 397           bandwidth_used: "{{ $value | humanizePercentage }}"
 398       - alert: interface transmit errors
 399         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 400         for: 5m
 401         labels:
 402           alertgroup: "{{ $labels.instance }}"
 403         annotations:
 404           error_rate: "{{ $value | humanizePercentage }}"
 405       - alert: wireguard interface transmit errors
 406         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 407         for: 1h
 408         labels:
 409           alertgroup: "{{ $labels.instance }}"
 410         annotations:
 411           error_rate: "{{ $value | humanizePercentage }}"
 412       - alert: interface receive errors
 413         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 414         for: 5m
 415         labels:
 416           alertgroup: "{{ $labels.instance }}"
 417         annotations:
 418           error_rate: "{{ $value | humanizePercentage }}"
 419       - alert: conntrack entries
 420         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 421         for: 5m
 422         labels:
 423           alertgroup: "{{ $labels.instance }}"
 424         annotations:
 425           entries_used: "{{ $value | humanizePercentage }}"
 426   - name: nominatim
 427     rules:
 428       - alert: nominatim replication delay
 429         expr: nominatim_replication_delay > 10800
 430         for: 1h
 431         labels:
 432           alertgroup: nominatim
 433         annotations:
 434           delay: "{{ $value | humanizeDuration }}"
 435   - name: overpass
 436     rules:
 437       - alert: overpass osm database age
 438         expr: overpass_database_age_seconds{database="osm"} > 3600
 439         for: 1h
 440         labels:
 441           alertgroup: overpass
 442         annotations:
 443           age: "{{ $value | humanizeDuration }}"
 444       - alert: overpass area database age
 445         expr: overpass_database_age_seconds{database="area"} > 86400
 446         for: 1h
 447         labels:
 448           alertgroup: overpass
 449         annotations:
 450           age: "{{ $value | humanizeDuration }}"
 451   - name: passenger
 452     rules:
 453       - alert: passenger down
 454         expr: passenger_up == 0
 455         for: 5m
 456         labels:
 457           alertgroup: "{{ $labels.instance }}"
 458       - alert: passenger queuing
 459         expr: passenger_top_level_request_queue > 0
 460         for: 5m
 461         labels:
 462           alertgroup: "{{ $labels.instance }}"
 463       - alert: passenger application queuing
 464         expr: passenger_app_request_queue > 0
 465         for: 5m
 466         labels:
 467           alertgroup: "{{ $labels.instance }}"
 468   - name: planet
 469     rules:
 470       - alert: planet dump overdue
 471         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 472         for: 24h
 473         labels:
 474           alertgroup: planet
 475         annotations:
 476           overdue_by: "{{ $value | humanizeDuration }}"
 477       - alert: notes dump overdue
 478         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 479         for: 6h
 480         labels:
 481           alertgroup: planet
 482         annotations:
 483           overdue_by: "{{ $value | humanizeDuration }}"
 484       - alert: daily replication feed delayed
 485         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 486         for: 3h
 487         labels:
 488           alertgroup: planet
 489         annotations:
 490           delayed_by: "{{ $value | humanizeDuration }}"
 491       - alert: hourly replication feed delayed
 492         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 493         for: 30m
 494         labels:
 495           alertgroup: planet
 496         annotations:
 497           delayed_by: "{{ $value | humanizeDuration }}"
 498       - alert: minutely replication feed delayed
 499         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 500         for: 5m
 501         labels:
 502           alertgroup: planet
 503         annotations:
 504           delayed_by: "{{ $value | humanizeDuration }}"
 505       - alert: changeset replication feed delayed
 506         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 507         for: 5m
 508         labels:
 509           alertgroup: planet
 510         annotations:
 511           delayed_by: "{{ $value | humanizeDuration }}"
 512   - name: postgresql
 513     rules:
 514       - alert: postgresql down
 515         expr: pg_up == 0
 516         for: 1m
 517         labels:
 518           alertgroup: "{{ $labels.instance }}"
 519       - alert: postgresql replication delay
 520         expr: pg_replication_lag_seconds > 30
 521         for: 15m
 522         labels:
 523           alertgroup: "{{ $labels.instance }}"
 524         annotations:
 525           delay: "{{ $value | humanizeDuration }}"
 526       - alert: postgresql connection limit
 527         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 528         for: 1m
 529         labels:
 530           alertgroup: "{{ $labels.instance }}"
 531         annotations:
 532           connections_used: "{{ $value | humanizePercentage }}"
 533       - alert: postgresql deadlocks
 534         expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
 535         for: 0m
 536         labels:
 537           alertgroup: "{{ $labels.instance }}"
 538         annotations:
 539           new_deadlocks: "{{ $value }}"
 540       - alert: postgresql slow queries
 541         expr: pg_slow_queries > 0
 542         for: 5m
 543         labels:
 544           alertgroup: "{{ $labels.instance }}"
 545         annotations:
 546           queries: "{{ $value }}"
 547       - alert: postgresql idle transactions
 548         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)
 549         for: 5m
 550         labels:
 551           alertgroup: "{{ $labels.instance }}"
 552         annotations:
 553           queries: "{{ $value }}"
 554   - name: prometheus
 555     rules:
 556       - alert: prometheus configuration error
 557         expr: prometheus_config_last_reload_successful == 0
 558         for: 10m
 559         labels:
 560           alertgroup: "prometheus"
 561       - alert: prometheus target missing
 562         expr: up == 0
 563         for: 10m
 564         labels:
 565           alertgroup: "prometheus"
 566   - name: raid
 567     rules:
 568       - alert: raid controller battery failed
 569         expr: ohai_controller_info{battery_status="failed"} > 0
 570         for: 5m
 571         labels:
 572           alertgroup: "{{ $labels.instance }}"
 573       - alert: raid controller battery recharging
 574         expr: ohai_controller_info{battery_status="recharging"} > 0
 575         for: 4h
 576         labels:
 577           alertgroup: "{{ $labels.instance }}"
 578       - alert: raid array degraded
 579         expr: ohai_array_info{status="degraded"} > 0
 580         for: 5m
 581         labels:
 582           alertgroup: "{{ $labels.instance }}"
 583       - alert: raid disk failed
 584         expr: ohai_disk_info{status="failed"} > 0
 585         for: 5m
 586         labels:
 587           alertgroup: "{{ $labels.instance }}"
 588   - name: rasdaemon
 589     rules:
 590       - alert: memory controller errors
 591         expr: increase(rasdaemon_mc_events_total[1m]) > 0
 592         for: 0m
 593         labels:
 594           alertgroup: "{{ $labels.instance }}"
 595         annotations:
 596           new_errors: "{{ $value }}"
 597       - alert: pcie aer errors
 598         expr: increase(rasdaemon_aer_events_total[1m]) > 0
 599         for: 0m
 600         labels:
 601           alertgroup: "{{ $labels.instance }}"
 602         annotations:
 603           new_errors: "{{ $value }}"
 604   - name: smart
 605     rules:
 606       - alert: smart failure
 607         expr: smart_health_status == 0
 608         for: 60m
 609         labels:
 610           alertgroup: "{{ $labels.instance }}"
 611       - alert: smart ssd wearout approaching
 612         expr: smart_percentage_used / 100 >= 0.8
 613         for: 60m
 614         labels:
 615           alertgroup: "{{ $labels.instance }}"
 616         annotations:
 617           percentage_used: "{{ $value | humanizePercentage }}"
 618   - name: smokeping
 619     rules:
 620       - alert: packet loss
 621         expr: 1 - (rate(smokeping_response_duration_seconds_count[5m]) / rate(smokeping_requests_total[5m])) > 0.02
 622         for: 10m
 623         labels:
 624           alertgroup: smokeping
 625         annotations:
 626           loss_rate: "{{ $value | humanizePercentage }}"
 627   - name: snmp
 628     rules:
 629       - alert: snmp pdus missing
 630         expr: max_over_time(snmp_scrape_pdus_returned[1d]) - snmp_scrape_pdus_returned > 0
 631         for: 15m
 632         labels:
 633           alertgroup: snmp
 634         annotations:
 635           missing_pdus: "{{ $value }}"
 636   - name: ssl
 637     rules:
 638       - alert: ssl certificate probe failed
 639         expr: ssl_probe_success == 0
 640         for: 60m
 641         labels:
 642           alertgroup: ssl
 643       - alert: ssl certificate expiry
 644         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 645         for: 0m
 646         labels:
 647           alertgroup: ssl
 648         annotations:
 649           expires_in: "{{ $value | humanizeDuration }}"
 650       - alert: ssl certificate revoked
 651         expr: ssl_ocsp_response_status == 1
 652         for: 0m
 653         labels:
 654           alertgroup: ssl
 655       - alert: ocsp status unknown
 656         expr: ssl_ocsp_response_status == 1
 657         for: 0m
 658         labels:
 659           alertgroup: ssl
 660   - name: statuscake
 661     rules:
 662       - alert: statuscake uptime check failing
 663         expr: statuscake_paused == 0 and statuscake_up == 0
 664         for: 10m
 665         labels:
 666           alertgroup: statuscake
 667   - name: systemd
 668     rules:
 669       - alert: systemd failed service
 670         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 671         for: 5m
 672         labels:
 673           alertgroup: "{{ $labels.instance }}"
 674       - alert: systemd failed chef client service
 675         expr: sum_over_time(node_systemd_unit_state{state="inactive",name="chef-client.service"}[6h]) == 0
 676         for: 0m
 677         labels:
 678           alertgroup: "{{ $labels.instance }}"
 679   - name: taginfo
 680     rules:
 681       - alert: taginfo planet age
 682         expr: time() - taginfo_data_from_seconds > 129600
 683         for: 0m
 684         labels:
 685           alertgroup: taginfo
 686         annotations:
 687           age: "{{ $value | humanizeDuration }}"
 688       - alert: taginfo database age
 689         expr: time() - taginfo_database_update_finish_seconds > 129600
 690         for: 0m
 691         labels:
 692           alertgroup: taginfo
 693         annotations:
 694           age: "{{ $value | humanizeDuration }}"
 695       - alert: taginfo database size
 696         expr: abs(delta(taginfo_database_size_bytes[30m])) / taginfo_database_size_bytes > 0.1
 697         for: 30m
 698         labels:
 699           alertgroup: taginfo
 700         annotations:
 701           size_change: "{{ $value | humanizePercentage }}"
 702   - name: tile
 703     rules:
 704       - alert: renderd replication delay
 705         expr: renderd_replication_delay > 120
 706         for: 15m
 707         labels:
 708           alertgroup: tile
 709         annotations:
 710           delay: "{{ $value | humanizeDuration }}"
 711       - alert: missed tile rate
 712         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 713         for: 5m
 714         labels:
 715           alertgroup: tile
 716         annotations:
 717           miss_rate: "{{ $value | humanizePercentage }}"
 718       - alert: tile render rate
 719         expr: sum(rate(renderd_zoom_metatiles_total[5m])) by (instance) == 0
 720         for: 15m
 721         labels:
 722           alertgroup: tile
 723         annotations:
 724           render_rate: "{{ $value }} tiles/s"
 725   - name: time
 726     rules:
 727       - alert: clock not synchronising
 728         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 729         for: 5m
 730         labels:
 731           alertgroup: "{{ $labels.instance }}"
 732       - alert: clock skew detected
 733         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)
 734         for: 5m
 735         labels:
 736           alertgroup: "{{ $labels.instance }}"
 737         annotations:
 738           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 739   - name: web
 740     rules:
 741       - alert: web error rate
 742         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
 743         for: 5m
 744         labels:
 745           alertgroup: web
 746         annotations:
 747           error_rate: "{{ $value | humanizePercentage }}"
 748       - alert: job processing rate
 749         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
 750         for: 1h
 751         labels:
 752           alertgroup: web
 753         annotations:
 754           job_processing_rate: "{{ $value | humanizePercentage }}"