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: fastly 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       - alert: fastly healthcheck failing
 125         expr: fastly_healthcheck_status = 0
 126         for: 5m
 127         labels:
 128           alertgroup: fastly
 129   - name: filesystem
 130     rules:
 131       - alert: readonly filesystem
 132         expr: node_filesystem_readonly == 1
 133         for: 0m
 134         labels:
 135           alertgroup: "{{ $labels.instance }}"
 136       - alert: filesystem low on space
 137         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 138         for: 5m
 139         labels:
 140           alertgroup: "{{ $labels.instance }}"
 141         annotations:
 142           percentage_free: "{{ $value | humanizePercentage }}"
 143           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 144           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 145       - alert: filesystem low on inodes
 146         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 147         for: 5m
 148         labels:
 149           alertgroup: "{{ $labels.instance }}"
 150         annotations:
 151           percentage_free: "{{ $value | humanizePercentage }}"
 152           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 153           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 154   - name: hwmon
 155     rules:
 156       - alert: hwmon fan alarm
 157         expr: node_hwmon_fan_alarm == 1
 158         for: 5m
 159         labels:
 160           alertgroup: "{{ $labels.instance }}"
 161         annotations:
 162           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}}"
 163           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}}"
 164       - alert: hwmon temperature alarm
 165         expr: node_hwmon_temp_alarm == 1
 166         for: 5m
 167         labels:
 168           alertgroup: "{{ $labels.instance }}"
 169         annotations:
 170           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}}"
 171           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}}"
 172           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}}"
 173       - alert: hwmon voltage alarm
 174         expr: node_hwmon_in_alarm == 1
 175         for: 5m
 176         labels:
 177           alertgroup: "{{ $labels.instance }}"
 178         annotations:
 179           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}}"
 180           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}}"
 181           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}}"
 182   - name: io
 183     rules:
 184       - alert: io pressure
 185         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 186         for: 60m
 187         labels:
 188           alertgroup: "{{ $labels.instance }}"
 189         annotations:
 190           pressure: "{{ $value | humanizePercentage }}"
 191   - name: ipmi
 192     rules:
 193       - alert: ipmi fan alarm
 194         expr: ipmi_fan_speed_state > 0
 195         for: 5m
 196         labels:
 197           alertgroup: "{{ $labels.instance }}"
 198         annotations:
 199           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 200       - alert: ipmi temperature alarm
 201         expr: ipmi_temperature_state > 0
 202         for: 5m
 203         labels:
 204           alertgroup: "{{ $labels.instance }}"
 205         annotations:
 206           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 207       - alert: ipmi voltage alarm
 208         expr: ipmi_voltage_state > 0
 209         for: 5m
 210         labels:
 211           alertgroup: "{{ $labels.instance }}"
 212         annotations:
 213           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 214       - alert: ipmi power alarm
 215         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 216         for: 5m
 217         labels:
 218           alertgroup: "{{ $labels.instance }}"
 219   - name: juniper
 220     rules:
 221       - alert: juniper fan alarm
 222         expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!="running"} > 0
 223         for: 5m
 224         labels:
 225           alertgroup: "{{ $labels.site }}"
 226       - alert: juniper power alarm
 227         expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!="running"} > 0
 228         for: 5m
 229         labels:
 230           alertgroup: "{{ $labels.site }}"
 231   - name: mail
 232     rules:
 233       - alert: exim queue length
 234         expr: exim_queue > exim_queue_limit
 235         for: 60m
 236         labels:
 237           alertgroup: mail
 238         annotations:
 239           queue_length: "{{ $value }}"
 240       - alert: mailman queue length
 241         expr: mailman_queue_length > 200
 242         for: 60m
 243         labels:
 244           alertgroup: mail
 245         annotations:
 246           queue_length: "{{ $value }}"
 247   - name: mdadm
 248     rules:
 249       - alert: mdadm array inactive
 250         expr: node_md_state{state="inactive"} > 0
 251         for: 0m
 252         labels:
 253           alertgroup: "{{ $labels.instance }}"
 254         annotations:
 255           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 256           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 257           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 258           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 259       - alert: mdadm array degraded
 260         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 261         for: 0m
 262         labels:
 263           alertgroup: "{{ $labels.instance }}"
 264         annotations:
 265           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 266           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 267           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 268           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 269       - alert: mdadm disk failed
 270         expr: node_md_disks{state="failed"} > 0
 271         for: 0m
 272         labels:
 273           alertgroup: "{{ $labels.instance }}"
 274         annotations:
 275           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 276           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 277           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 278           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 279   - name: memory
 280     rules:
 281       - alert: low memory
 282         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 283         for: 15m
 284         labels:
 285           alertgroup: "{{ $labels.instance }}"
 286         annotations:
 287           memory_free: "{{ $value | humanizePercentage }}"
 288       - alert: memory pressure
 289         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 290         for: 60m
 291         labels:
 292           alertgroup: "{{ $labels.instance }}"
 293         annotations:
 294           pressure: "{{ $value | humanizePercentage }}"
 295       - alert: oom kill detected
 296         expr: increase(node_vmstat_oom_kill[1m]) > 0
 297         for: 0m
 298         labels:
 299           alertgroup: "{{ $labels.instance }}"
 300         annotations:
 301           new_oom_kills: "{{ $value }}"
 302   - name: network
 303     rules:
 304       - alert: interface transmit rate
 305         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 306         for: 5m
 307         labels:
 308           alertgroup: "{{ $labels.instance }}"
 309         annotations:
 310           bandwidth_used: "{{ $value | humanizePercentage }}"
 311       - alert: interface receive rate
 312         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.98
 313         for: 5m
 314         labels:
 315           alertgroup: "{{ $labels.instance }}"
 316         annotations:
 317           bandwidth_used: "{{ $value | humanizePercentage }}"
 318       - alert: interface transmit errors
 319         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 320         for: 5m
 321         labels:
 322           alertgroup: "{{ $labels.instance }}"
 323         annotations:
 324           error_rate: "{{ $value | humanizePercentage }}"
 325       - alert: interface transmit errors
 326         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 327         for: 1h
 328         labels:
 329           alertgroup: "{{ $labels.instance }}"
 330         annotations:
 331           error_rate: "{{ $value | humanizePercentage }}"
 332       - alert: interface receive errors
 333         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 334         for: 5m
 335         labels:
 336           alertgroup: "{{ $labels.instance }}"
 337         annotations:
 338           error_rate: "{{ $value | humanizePercentage }}"
 339       - alert: conntrack entries
 340         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 341         for: 5m
 342         labels:
 343           alertgroup: "{{ $labels.instance }}"
 344         annotations:
 345           entries_used: "{{ $value | humanizePercentage }}"
 346   - name: planet
 347     rules:
 348       - alert: planet dump overdue
 349         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 350         for: 24h
 351         labels:
 352           alertgroup: planet
 353         annotations:
 354           overdue_by: "{{ $value | humanizeDuration }}"
 355       - alert: notes dump overdue
 356         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 357         for: 6h
 358         labels:
 359           alertgroup: planet
 360         annotations:
 361           overdue_by: "{{ $value | humanizeDuration }}"
 362       - alert: daily replication feed delayed
 363         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 364         for: 3h
 365         labels:
 366           alertgroup: planet
 367         annotations:
 368           delayed_by: "{{ $value | humanizeDuration }}"
 369       - alert: hourly replication feed delayed
 370         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 371         for: 30m
 372         labels:
 373           alertgroup: planet
 374         annotations:
 375           delayed_by: "{{ $value | humanizeDuration }}"
 376       - alert: minutely replication feed delayed
 377         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 378         for: 5m
 379         labels:
 380           alertgroup: planet
 381         annotations:
 382           delayed_by: "{{ $value | humanizeDuration }}"
 383       - alert: changeset replication feed delayed
 384         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 385         for: 5m
 386         labels:
 387           alertgroup: planet
 388         annotations:
 389           delayed_by: "{{ $value | humanizeDuration }}"
 390   - name: postgresql
 391     rules:
 392       - alert: postgresql down
 393         expr: pg_up == 0
 394         for: 1m
 395         labels:
 396           alertgroup: "{{ $labels.instance }}"
 397       - alert: postgresql replication delay
 398         expr: pg_replication_lag_seconds > 5
 399         for: 1m
 400         labels:
 401           alertgroup: "{{ $labels.instance }}"
 402         annotations:
 403           delay: "{{ $value | humanizeDuration }}"
 404       - alert: postgresql connection limit
 405         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 406         for: 1m
 407         labels:
 408           alertgroup: "{{ $labels.instance }}"
 409         annotations:
 410           connections_used: "{{ $value | humanizePercentage }}"
 411       - alert: postgresql deadlocks
 412         expr: increase(pg_stat_database_deadlocks[1m]) > 5
 413         for: 0m
 414         labels:
 415           alertgroup: "{{ $labels.instance }}"
 416         annotations:
 417           new_deadlocks: "{{ $value }}"
 418       - alert: postgresql slow queries
 419         expr: pg_slow_queries > 0
 420         for: 5m
 421         labels:
 422           alertgroup: "{{ $labels.instance }}"
 423         annotations:
 424           queries: "{{ $value }}"
 425   - name: smart
 426     rules:
 427       - alert: smart failure
 428         expr: smart_health_status == 0
 429         for: 60m
 430         labels:
 431           alertgroup: "{{ $labels.instance }}"
 432       - alert: smart ssd wearout approaching
 433         expr: smart_percentage_used >= 90
 434         for: 60m
 435         labels:
 436           alertgroup: "{{ $labels.instance }}"
 437         annotations:
 438           percentage_used: "{{ $value | humanizePercentage }}"
 439   - name: ssl
 440     rules:
 441       - alert: ssl certificate probe failed
 442         expr: ssl_probe_success == 0
 443         for: 60m
 444         labels:
 445           alertgroup: ssl
 446       - alert: ssl certificate expiry
 447         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 448         for: 0m
 449         labels:
 450           alertgroup: ssl
 451         annotations:
 452           expires_in: "{{ $value | humanizeDuration }}"
 453       - alert: ssl certificate revoked
 454         expr: ssl_ocsp_response_status == 1
 455         for: 0m
 456         labels:
 457           alertgroup: ssl
 458       - alert: ocsp status unknown
 459         expr: ssl_ocsp_response_status == 1
 460         for: 0m
 461         labels:
 462           alertgroup: ssl
 463   - name: statuscake
 464     rules:
 465       - alert: statuscake uptime check failing
 466         expr: statuscake_uptime{status="down",paused="false"} > 0
 467         for: 10m
 468         labels:
 469           alertgroup: statuscake
 470   - name: systemd
 471     rules:
 472       - alert: systemd failed service
 473         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 474         for: 5m
 475         labels:
 476           alertgroup: "{{ $labels.instance }}"
 477       - alert: systemd failed service
 478         expr: node_systemd_unit_state{state="failed",name="chef-client.service"} == 1
 479         for: 6h
 480         labels:
 481           alertgroup: "{{ $labels.instance }}"
 482   - name: tile
 483     rules:
 484       - alert: renderd replication delay
 485         expr: renderd_replication_delay > 120
 486         for: 15m
 487         labels:
 488           alertgroup: tile
 489         annotations:
 490           delay: "{{ $value | humanizeDuration }}"
 491       - alert: missed tile rate
 492         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 493         for: 5m
 494         labels:
 495           alertgroup: tile
 496         annotations:
 497           miss_rate: "{{ $value | humanizePercentage }}"
 498   - name: time
 499     rules:
 500       - alert: clock not synchronising
 501         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 502         for: 5m
 503         labels:
 504           alertgroup: "{{ $labels.instance }}"
 505       - alert: clock skew detected
 506         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)
 507         for: 5m
 508         labels:
 509           alertgroup: "{{ $labels.instance }}"
 510         annotations:
 511           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 512   - name: web
 513     rules:
 514       - alert: web error rate
 515         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
 516         for: 5m
 517         labels:
 518           alertgroup: web
 519         annotations:
 520           error_rate: "{{ $value | humanizePercentage }}"
 521       - alert: job processing rate
 522         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
 523         for: 15m
 524         labels:
 525           alertgroup: web
 526         annotations:
 527           job_processing_rate: "{{ $value | humanizePercentage }}"