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

   1 # DO NOT EDIT - This file is being maintained by Chef
   2
   3 groups:
   4   - name: alertmanager
   5     rules:
   6       - alert: prometheus target missing
   7         expr: up == 0
   8         for: 10m
   9         labels:
  10           alertgroup: "prometheus"
  11   - name: amsterdam
  12     rules:
  13       - alert: pdu current draw
  14         expr: rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10 > 10
  15         for: 5m
  16         labels:
  17           alertgroup: "amsterdam"
  18         annotations:
  19           current: "{{ $value | humanize }}A"
  20       - alert: site current draw
  21         expr: sum(rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10) > 13
  22         for: 5m
  23         labels:
  24           alertgroup: "amsterdam"
  25         annotations:
  26           current: "{{ $value | humanize }}A"
  27       - alert: site temperature
  28         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 25
  29         for: 5m
  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: 5m
  37         labels:
  38           alertgroup: "amsterdam"
  39         annotations:
  40           humidity: "{{ $value | humanizePercentage }}"
  41   - name: apache
  42     rules:
  43       - alert: apache down
  44         expr: apache_up == 0
  45         for: 5m
  46         labels:
  47           alertgroup: "{{ $labels.instance }}"
  48       - alert: apache workers busy
  49         expr: sum(apache_workers{state="busy"}) by (instance) / sum(apache_scoreboard) by (instance) > 0.8
  50         for: 5m
  51         labels:
  52           alertgroup: "{{ $labels.instance }}"
  53         annotations:
  54           busy_workers: "{{ $value | humanizePercentage }}"
  55       - alert: apache low request rate
  56         expr: rate(apache_accesses_total[5m]) / rate(apache_accesses_total[1h] offset 1w) < 0.25 and rate(apache_accesses_total[1h] offset 1w) > 2
  57         for: 15m
  58         labels:
  59           alertgroup: "{{ $labels.instance }}"
  60         annotations:
  61           request_rate: "{{ $value | humanizePercentage }}"
  62   - name: chef
  63     rules:
  64       - alert: chef client not running
  65         expr: time() - node_systemd_timer_last_trigger_seconds{name="chef-client.timer"} > 3600
  66         for: 12h
  67         labels:
  68           alertgroup: "{{ $labels.instance }}"
  69         annotations:
  70           down_time: "{{ $value | humanizeDuration }}"
  71   - name: cisco
  72     rules:
  73       - alert: cisco fan alarm
  74         expr: rlPhdUnitEnvParamFan1Status{rlPhdUnitEnvParamFan1Status!="normal"} > 0 or rlPhdUnitEnvParamFan2Status{rlPhdUnitEnvParamFan2Status!="normal"} > 0
  75         for: 5m
  76         labels:
  77           alertgroup: "{{ $labels.site }}"
  78         annotations:
  79           fan_rpm: "{{ with printf \"rlPhdUnitEnvParamFan1Speed{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}rpm{{end}}"
  80       - alert: cisco temperature alarm
  81         expr: rlPhdUnitEnvParamTempSensorStatus{rlPhdUnitEnvParamTempSensorStatus!="ok"} > 0
  82         for: 5m
  83         labels:
  84           alertgroup: "{{ $labels.site }}"
  85         annotations:
  86           temp_celsius: "{{ with printf \"rlPhdUnitEnvParamTempSensorValue{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}C{{end}}"
  87       - alert: cisco main power alarm
  88         expr: rlPhdUnitEnvParamMainPSStatus{rlPhdUnitEnvParamMainPSStatus!="normal"} > 0
  89         for: 5m
  90         labels:
  91           alertgroup: "{{ $labels.site }}"
  92       - alert: cisco redundant power alarm
  93         expr: rlPhdUnitEnvParamRedundantPSStatus{rlPhdUnitEnvParamRedundantPSStatus!="normal"} > 0
  94         for: 5m
  95         labels:
  96           alertgroup: "{{ $labels.site }}"
  97   - name: cpu
  98     rules:
  99       - alert: cpu pressure
 100         expr: rate(node_pressure_cpu_waiting_seconds_total[5m]) > 0.6
 101         for: 15m
 102         labels:
 103           alertgroup: "{{ $labels.instance }}"
 104         annotations:
 105           pressure: "{{ $value | humanizePercentage }}"
 106   - name: database
 107     rules:
 108       - alert: postgres replication delay
 109         expr: pg_replication_lag_seconds > 5
 110         for: 5m
 111         labels:
 112           alertgroup: database
 113         annotations:
 114           delay: "{{ $value | humanizeDuration }}"
 115   - name: fastly
 116     rules:
 117       - alert: error rate
 118         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
 119         for: 15m
 120         labels:
 121           alertgroup: fastly
 122         annotations:
 123           error_rate: "{{ $value | humanizePercentage }}"
 124   - name: filesystem
 125     rules:
 126       - alert: readonly filesystem
 127         expr: node_filesystem_readonly == 1
 128         for: 0m
 129         labels:
 130           alertgroup: "{{ $labels.instance }}"
 131       - alert: filesystem low on space
 132         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 133         for: 5m
 134         labels:
 135           alertgroup: "{{ $labels.instance }}"
 136         annotations:
 137           percentage_free: "{{ $value | humanizePercentage }}"
 138           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 139           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 140       - alert: filesystem low on inodes
 141         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 142         for: 5m
 143         labels:
 144           alertgroup: "{{ $labels.instance }}"
 145         annotations:
 146           percentage_free: "{{ $value | humanizePercentage }}"
 147           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 148           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 149   - name: hwmon
 150     rules:
 151       - alert: hwmon fan alarm
 152         expr: node_hwmon_fan_alarm == 1
 153         for: 5m
 154         labels:
 155           alertgroup: "{{ $labels.instance }}"
 156         annotations:
 157           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}}"
 158           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}}"
 159       - alert: hwmon temperature alarm
 160         expr: node_hwmon_temp_alarm == 1
 161         for: 5m
 162         labels:
 163           alertgroup: "{{ $labels.instance }}"
 164         annotations:
 165           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}}"
 166           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}}"
 167           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}}"
 168       - alert: hwmon voltage alarm
 169         expr: node_hwmon_in_alarm == 1
 170         for: 5m
 171         labels:
 172           alertgroup: "{{ $labels.instance }}"
 173         annotations:
 174           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}}"
 175           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}}"
 176           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}}"
 177   - name: io
 178     rules:
 179       - alert: io pressure
 180         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 181         for: 60m
 182         labels:
 183           alertgroup: "{{ $labels.instance }}"
 184         annotations:
 185           pressure: "{{ $value | humanizePercentage }}"
 186   - name: ipmi
 187     rules:
 188       - alert: ipmi fan alarm
 189         expr: ipmi_fan_speed_state > 0
 190         for: 5m
 191         labels:
 192           alertgroup: "{{ $labels.instance }}"
 193         annotations:
 194           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 195       - alert: ipmi temperature alarm
 196         expr: ipmi_temperature_state > 0
 197         for: 5m
 198         labels:
 199           alertgroup: "{{ $labels.instance }}"
 200         annotations:
 201           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 202       - alert: ipmi voltage alarm
 203         expr: ipmi_voltage_state > 0
 204         for: 5m
 205         labels:
 206           alertgroup: "{{ $labels.instance }}"
 207         annotations:
 208           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 209       - alert: ipmi power alarm
 210         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 211         for: 5m
 212         labels:
 213           alertgroup: "{{ $labels.instance }}"
 214   - name: juniper
 215     rules:
 216       - alert: juniper fan alarm
 217         expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!="running"} > 0
 218         for: 5m
 219         labels:
 220           alertgroup: "{{ $labels.site }}"
 221       - alert: juniper power alarm
 222         expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!="running"} > 0
 223         for: 5m
 224         labels:
 225           alertgroup: "{{ $labels.site }}"
 226   - name: mail
 227     rules:
 228       - alert: exim queue length
 229         expr: exim_queue > exim_queue_limit
 230         for: 60m
 231         labels:
 232           alertgroup: mail
 233         annotations:
 234           queue_length: "{{ $value }}"
 235       - alert: mailman queue length
 236         expr: mailman_queue_length > 200
 237         for: 60m
 238         labels:
 239           alertgroup: mail
 240         annotations:
 241           queue_length: "{{ $value }}"
 242   - name: mdadm
 243     rules:
 244       - alert: mdadm array inactive
 245         expr: node_md_state{state="inactive"} > 0
 246         for: 0m
 247         labels:
 248           alertgroup: "{{ $labels.instance }}"
 249         annotations:
 250           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 251           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 252           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 253           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 254       - alert: mdadm array degraded
 255         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 256         for: 0m
 257         labels:
 258           alertgroup: "{{ $labels.instance }}"
 259         annotations:
 260           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 261           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 262           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 263           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 264       - alert: mdadm disk failed
 265         expr: node_md_disks{state="failed"} > 0
 266         for: 0m
 267         labels:
 268           alertgroup: "{{ $labels.instance }}"
 269         annotations:
 270           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 271           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 272           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 273           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 274   - name: memory
 275     rules:
 276       - alert: low memory
 277         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 278         for: 15m
 279         labels:
 280           alertgroup: "{{ $labels.instance }}"
 281         annotations:
 282           memory_free: "{{ $value | humanizePercentage }}"
 283       - alert: memory pressure
 284         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 285         for: 60m
 286         labels:
 287           alertgroup: "{{ $labels.instance }}"
 288         annotations:
 289           pressure: "{{ $value | humanizePercentage }}"
 290       - alert: oom kill detected
 291         expr: increase(node_vmstat_oom_kill[1m]) > 0
 292         for: 0m
 293         labels:
 294           alertgroup: "{{ $labels.instance }}"
 295         annotations:
 296           new_oom_kills: "{{ $value }}"
 297   - name: network
 298     rules:
 299       - alert: interface transmit rate
 300         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 301         for: 5m
 302         labels:
 303           alertgroup: "{{ $labels.instance }}"
 304         annotations:
 305           bandwidth_used: "{{ $value | humanizePercentage }}"
 306       - alert: interface receive rate
 307         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.98
 308         for: 5m
 309         labels:
 310           alertgroup: "{{ $labels.instance }}"
 311         annotations:
 312           bandwidth_used: "{{ $value | humanizePercentage }}"
 313       - alert: interface transmit errors
 314         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 315         for: 5m
 316         labels:
 317           alertgroup: "{{ $labels.instance }}"
 318         annotations:
 319           error_rate: "{{ $value | humanizePercentage }}"
 320       - alert: interface transmit errors
 321         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 322         for: 1h
 323         labels:
 324           alertgroup: "{{ $labels.instance }}"
 325         annotations:
 326           error_rate: "{{ $value | humanizePercentage }}"
 327       - alert: interface receive errors
 328         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 329         for: 5m
 330         labels:
 331           alertgroup: "{{ $labels.instance }}"
 332         annotations:
 333           error_rate: "{{ $value | humanizePercentage }}"
 334       - alert: conntrack entries
 335         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 336         for: 5m
 337         labels:
 338           alertgroup: "{{ $labels.instance }}"
 339         annotations:
 340           entries_used: "{{ $value | humanizePercentage }}"
 341   - name: planet
 342     rules:
 343       - alert: planet dump overdue
 344         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 345         for: 24h
 346         labels:
 347           alertgroup: planet
 348         annotations:
 349           overdue_by: "{{ $value | humanizeDuration }}"
 350       - alert: notes dump overdue
 351         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 352         for: 6h
 353         labels:
 354           alertgroup: planet
 355         annotations:
 356           overdue_by: "{{ $value | humanizeDuration }}"
 357       - alert: daily replication feed delayed
 358         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 359         for: 3h
 360         labels:
 361           alertgroup: planet
 362         annotations:
 363           delayed_by: "{{ $value | humanizeDuration }}"
 364       - alert: hourly replication feed delayed
 365         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 366         for: 30m
 367         labels:
 368           alertgroup: planet
 369         annotations:
 370           delayed_by: "{{ $value | humanizeDuration }}"
 371       - alert: minutely replication feed delayed
 372         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 373         for: 5m
 374         labels:
 375           alertgroup: planet
 376         annotations:
 377           delayed_by: "{{ $value | humanizeDuration }}"
 378       - alert: changeset replication feed delayed
 379         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 380         for: 5m
 381         labels:
 382           alertgroup: planet
 383         annotations:
 384           delayed_by: "{{ $value | humanizeDuration }}"
 385   - name: postgresql
 386     rules:
 387       - alert: postgresql down
 388         expr: pg_up == 0
 389         for: 1m
 390         labels:
 391           alertgroup: "{{ $labels.instance }}"
 392       - alert: postgresql replication delay
 393         expr: pg_replication_lag_seconds > 5
 394         for: 1m
 395         labels:
 396           alertgroup: "{{ $labels.instance }}"
 397         annotations:
 398           delay: "{{ $value | humanizeDuration }}"
 399       - alert: postgresql connection limit
 400         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 401         for: 1m
 402         labels:
 403           alertgroup: "{{ $labels.instance }}"
 404         annotations:
 405           connections_used: "{{ $value | humanizePercentage }}"
 406       - alert: postgresql deadlocks
 407         expr: increase(pg_stat_database_deadlocks[1m]) > 5
 408         for: 0m
 409         labels:
 410           alertgroup: "{{ $labels.instance }}"
 411         annotations:
 412           new_deadlocks: "{{ $value }}"
 413       - alert: postgresql slow queries
 414         expr: pg_slow_queries > 0
 415         for: 5m
 416         labels:
 417           alertgroup: "{{ $labels.instance }}"
 418         annotations:
 419           queries: "{{ $value }}"
 420   - name: smart
 421     rules:
 422       - alert: smart failure
 423         expr: smart_health_status == 0
 424         for: 60m
 425         labels:
 426           alertgroup: "{{ $labels.instance }}"
 427       - alert: smart ssd wearout approaching
 428         expr: smart_percentage_used >= 90
 429         for: 60m
 430         labels:
 431           alertgroup: "{{ $labels.instance }}"
 432         annotations:
 433           percentage_used: "{{ $value | humanizePercentage }}"
 434   - name: ssl
 435     rules:
 436       - alert: ssl certificate probe failed
 437         expr: ssl_probe_success == 0
 438         for: 60m
 439         labels:
 440           alertgroup: ssl
 441       - alert: ssl certificate expiry
 442         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 443         for: 0m
 444         labels:
 445           alertgroup: ssl
 446         annotations:
 447           expires_in: "{{ $value | humanizeDuration }}"
 448       - alert: ssl certificate revoked
 449         expr: ssl_ocsp_response_status == 1
 450         for: 0m
 451         labels:
 452           alertgroup: ssl
 453       - alert: ocsp status unknown
 454         expr: ssl_ocsp_response_status == 1
 455         for: 0m
 456         labels:
 457           alertgroup: ssl
 458   - name: statuscake
 459     rules:
 460       - alert: statuscake uptime check failing
 461         expr: statuscake_uptime{status="down",paused="false"} > 0
 462         for: 10m
 463         labels:
 464           alertgroup: statuscake
 465   - name: systemd
 466     rules:
 467       - alert: systemd failed service
 468         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 469         for: 5m
 470         labels:
 471           alertgroup: "{{ $labels.instance }}"
 472       - alert: systemd failed service
 473         expr: node_systemd_unit_state{state="failed",name="chef-client.service"} == 1
 474         for: 6h
 475         labels:
 476           alertgroup: "{{ $labels.instance }}"
 477   - name: tile
 478     rules:
 479       - alert: renderd replication delay
 480         expr: renderd_replication_delay > 120
 481         for: 15m
 482         labels:
 483           alertgroup: tile
 484         annotations:
 485           delay: "{{ $value | humanizeDuration }}"
 486       - alert: missed tile rate
 487         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 488         for: 5m
 489         labels:
 490           alertgroup: tile
 491         annotations:
 492           miss_rate: "{{ $value | humanizePercentage }}"
 493   - name: time
 494     rules:
 495       - alert: clock not synchronising
 496         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 497         for: 5m
 498         labels:
 499           alertgroup: "{{ $labels.instance }}"
 500       - alert: clock skew detected
 501         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)
 502         for: 5m
 503         labels:
 504           alertgroup: "{{ $labels.instance }}"
 505         annotations:
 506           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 507   - name: web
 508     rules:
 509       - alert: web error rate
 510         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
 511         for: 5m
 512         labels:
 513           alertgroup: web
 514         annotations:
 515           error_rate: "{{ $value | humanizePercentage }}"
 516       - alert: job processing rate
 517         expr: rate(pg_stat_user_tables_n_tup_del{datname="openstreetmap",relname="delayed_jobs"}[5m]) / rate(pg_stat_user_tables_n_tup_ins{datname="openstreetmap",relname="delayed_jobs"}[5m]) < 0.9 and ignoring(job, name, datname, relname, schemaname, server) chef_role{name="db-master"} == 1
 518         for: 15m
 519         labels:
 520           alertgroup: web
 521         annotations:
 522           job_processing_rate: "{{ $value | humanizePercentage }}"