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 > 10
   8         for: 5m
   9         labels:
  10           alertgroup: "amsterdam"
  11         annotations:
  12           current: "{{ $value | humanize }}A"
  13       - alert: site current draw
  14         expr: sum(rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10) > 13
  15         for: 5m
  16         labels:
  17           alertgroup: "amsterdam"
  18         annotations:
  19           current: "{{ $value | humanize }}A"
  20       - alert: site temperature
  21         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 25
  22         for: 5m
  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: 5m
  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       - alert: apache low request rate
  49         expr: rate(apache_accesses_total[5m]) / rate(apache_accesses_total[1h] offset 1w) < 0.25 and rate(apache_accesses_total[1h] offset 1w) > 2
  50         for: 15m
  51         labels:
  52           alertgroup: "{{ $labels.instance }}"
  53         annotations:
  54           request_rate: "{{ $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]) > 0.6
  94         for: 15m
  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 > 5
 103         for: 5m
 104         labels:
 105           alertgroup: database
 106         annotations:
 107           delay: "{{ $value | humanizeDuration }}"
 108   - name: fastly
 109     rules:
 110       - alert: fastly error rate
 111         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
 112         for: 15m
 113         labels:
 114           alertgroup: fastly
 115         annotations:
 116           error_rate: "{{ $value | humanizePercentage }}"
 117       - alert: fastly healthcheck failing
 118         expr: fastly_healthcheck_status == 0
 119         for: 5m
 120         labels:
 121           alertgroup: fastly
 122   - name: filesystem
 123     rules:
 124       - alert: readonly filesystem
 125         expr: node_filesystem_readonly == 1
 126         for: 0m
 127         labels:
 128           alertgroup: "{{ $labels.instance }}"
 129       - alert: filesystem low on space
 130         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 131         for: 5m
 132         labels:
 133           alertgroup: "{{ $labels.instance }}"
 134         annotations:
 135           percentage_free: "{{ $value | humanizePercentage }}"
 136           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 137           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 138       - alert: filesystem low on inodes
 139         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 140         for: 5m
 141         labels:
 142           alertgroup: "{{ $labels.instance }}"
 143         annotations:
 144           percentage_free: "{{ $value | humanizePercentage }}"
 145           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 146           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 147   - name: hwmon
 148     rules:
 149       - alert: hwmon fan alarm
 150         expr: node_hwmon_fan_alarm == 1
 151         for: 5m
 152         labels:
 153           alertgroup: "{{ $labels.instance }}"
 154         annotations:
 155           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}}"
 156           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}}"
 157       - alert: hwmon temperature alarm
 158         expr: node_hwmon_temp_alarm == 1
 159         for: 5m
 160         labels:
 161           alertgroup: "{{ $labels.instance }}"
 162         annotations:
 163           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}}"
 164           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}}"
 165           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}}"
 166       - alert: hwmon voltage alarm
 167         expr: node_hwmon_in_alarm == 1
 168         for: 5m
 169         labels:
 170           alertgroup: "{{ $labels.instance }}"
 171         annotations:
 172           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}}"
 173           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}}"
 174           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}}"
 175   - name: io
 176     rules:
 177       - alert: io pressure
 178         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 179         for: 60m
 180         labels:
 181           alertgroup: "{{ $labels.instance }}"
 182         annotations:
 183           pressure: "{{ $value | humanizePercentage }}"
 184   - name: ipmi
 185     rules:
 186       - alert: ipmi fan alarm
 187         expr: ipmi_fan_speed_state > 0
 188         for: 5m
 189         labels:
 190           alertgroup: "{{ $labels.instance }}"
 191         annotations:
 192           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 193       - alert: ipmi temperature alarm
 194         expr: ipmi_temperature_state > 0
 195         for: 5m
 196         labels:
 197           alertgroup: "{{ $labels.instance }}"
 198         annotations:
 199           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 200       - alert: ipmi voltage alarm
 201         expr: ipmi_voltage_state > 0
 202         for: 5m
 203         labels:
 204           alertgroup: "{{ $labels.instance }}"
 205         annotations:
 206           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 207       - alert: ipmi power alarm
 208         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 209         for: 5m
 210         labels:
 211           alertgroup: "{{ $labels.instance }}"
 212   - name: juniper
 213     rules:
 214       - alert: juniper fan alarm
 215         expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!="running"} > 0
 216         for: 5m
 217         labels:
 218           alertgroup: "{{ $labels.site }}"
 219       - alert: juniper power alarm
 220         expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!="running"} > 0
 221         for: 5m
 222         labels:
 223           alertgroup: "{{ $labels.site }}"
 224   - name: mail
 225     rules:
 226       - alert: exim queue length
 227         expr: exim_queue > exim_queue_limit
 228         for: 60m
 229         labels:
 230           alertgroup: mail
 231         annotations:
 232           queue_length: "{{ $value }}"
 233       - alert: mailman queue length
 234         expr: mailman_queue_length > 200
 235         for: 60m
 236         labels:
 237           alertgroup: mail
 238         annotations:
 239           queue_length: "{{ $value }}"
 240   - name: mdadm
 241     rules:
 242       - alert: mdadm array inactive
 243         expr: node_md_state{state="inactive"} > 0
 244         for: 0m
 245         labels:
 246           alertgroup: "{{ $labels.instance }}"
 247         annotations:
 248           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 249           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 250           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 251           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 252       - alert: mdadm array degraded
 253         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 254         for: 0m
 255         labels:
 256           alertgroup: "{{ $labels.instance }}"
 257         annotations:
 258           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 259           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 260           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 261           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 262       - alert: mdadm disk failed
 263         expr: node_md_disks{state="failed"} > 0
 264         for: 0m
 265         labels:
 266           alertgroup: "{{ $labels.instance }}"
 267         annotations:
 268           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 269           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 270           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 271           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 272   - name: memory
 273     rules:
 274       - alert: low memory
 275         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 276         for: 15m
 277         labels:
 278           alertgroup: "{{ $labels.instance }}"
 279         annotations:
 280           memory_free: "{{ $value | humanizePercentage }}"
 281       - alert: memory pressure
 282         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 283         for: 60m
 284         labels:
 285           alertgroup: "{{ $labels.instance }}"
 286         annotations:
 287           pressure: "{{ $value | humanizePercentage }}"
 288       - alert: oom kill detected
 289         expr: increase(node_vmstat_oom_kill[1m]) > 0
 290         for: 0m
 291         labels:
 292           alertgroup: "{{ $labels.instance }}"
 293         annotations:
 294           new_oom_kills: "{{ $value }}"
 295   - name: network
 296     rules:
 297       - alert: interface transmit rate
 298         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 299         for: 5m
 300         labels:
 301           alertgroup: "{{ $labels.instance }}"
 302         annotations:
 303           bandwidth_used: "{{ $value | humanizePercentage }}"
 304       - alert: interface receive rate
 305         expr: rate(node_network_receive_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 transmit errors
 312         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 313         for: 5m
 314         labels:
 315           alertgroup: "{{ $labels.instance }}"
 316         annotations:
 317           error_rate: "{{ $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]) > 0.05
 320         for: 1h
 321         labels:
 322           alertgroup: "{{ $labels.instance }}"
 323         annotations:
 324           error_rate: "{{ $value | humanizePercentage }}"
 325       - alert: interface receive errors
 326         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 327         for: 5m
 328         labels:
 329           alertgroup: "{{ $labels.instance }}"
 330         annotations:
 331           error_rate: "{{ $value | humanizePercentage }}"
 332       - alert: conntrack entries
 333         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 334         for: 5m
 335         labels:
 336           alertgroup: "{{ $labels.instance }}"
 337         annotations:
 338           entries_used: "{{ $value | humanizePercentage }}"
 339   - name: planet
 340     rules:
 341       - alert: planet dump overdue
 342         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 343         for: 24h
 344         labels:
 345           alertgroup: planet
 346         annotations:
 347           overdue_by: "{{ $value | humanizeDuration }}"
 348       - alert: notes dump overdue
 349         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 350         for: 6h
 351         labels:
 352           alertgroup: planet
 353         annotations:
 354           overdue_by: "{{ $value | humanizeDuration }}"
 355       - alert: daily replication feed delayed
 356         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 357         for: 3h
 358         labels:
 359           alertgroup: planet
 360         annotations:
 361           delayed_by: "{{ $value | humanizeDuration }}"
 362       - alert: hourly replication feed delayed
 363         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 364         for: 30m
 365         labels:
 366           alertgroup: planet
 367         annotations:
 368           delayed_by: "{{ $value | humanizeDuration }}"
 369       - alert: minutely replication feed delayed
 370         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 371         for: 5m
 372         labels:
 373           alertgroup: planet
 374         annotations:
 375           delayed_by: "{{ $value | humanizeDuration }}"
 376       - alert: changeset replication feed delayed
 377         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 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   - name: postgresql
 384     rules:
 385       - alert: postgresql down
 386         expr: pg_up == 0
 387         for: 1m
 388         labels:
 389           alertgroup: "{{ $labels.instance }}"
 390       - alert: postgresql replication delay
 391         expr: pg_replication_lag_seconds > 5
 392         for: 1m
 393         labels:
 394           alertgroup: "{{ $labels.instance }}"
 395         annotations:
 396           delay: "{{ $value | humanizeDuration }}"
 397       - alert: postgresql connection limit
 398         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 399         for: 1m
 400         labels:
 401           alertgroup: "{{ $labels.instance }}"
 402         annotations:
 403           connections_used: "{{ $value | humanizePercentage }}"
 404       - alert: postgresql deadlocks
 405         expr: increase(pg_stat_database_deadlocks[1m]) > 5
 406         for: 0m
 407         labels:
 408           alertgroup: "{{ $labels.instance }}"
 409         annotations:
 410           new_deadlocks: "{{ $value }}"
 411       - alert: postgresql slow queries
 412         expr: pg_slow_queries > 0
 413         for: 5m
 414         labels:
 415           alertgroup: "{{ $labels.instance }}"
 416         annotations:
 417           queries: "{{ $value }}"
 418   - name: prometheus
 419     rules:
 420       - alert: prometheus configuration error
 421         expr: prometheus_config_last_reload_successful == 0
 422         for: 10m
 423         labels:
 424           alertgroup: "prometheus"
 425       - alert: prometheus target missing
 426         expr: up == 0
 427         for: 10m
 428         labels:
 429           alertgroup: "prometheus"
 430   - name: raid
 431     rules:
 432       - alert: raid array degraded
 433         expr: ohai_array_info{status="degraded"} > 0
 434         for: 5m
 435         labels:
 436           alertgroup: "{{ $labels.instance }}"
 437       - alert: raid disk failed
 438         expr: ohai_disk_info{status="failed"} > 0
 439         for: 5m
 440         labels:
 441           alertgroup: "{{ $labels.instance }}"
 442   - name: smart
 443     rules:
 444       - alert: smart failure
 445         expr: smart_health_status == 0
 446         for: 60m
 447         labels:
 448           alertgroup: "{{ $labels.instance }}"
 449       - alert: smart ssd wearout approaching
 450         expr: smart_percentage_used >= 90
 451         for: 60m
 452         labels:
 453           alertgroup: "{{ $labels.instance }}"
 454         annotations:
 455           percentage_used: "{{ $value | humanizePercentage }}"
 456   - name: ssl
 457     rules:
 458       - alert: ssl certificate probe failed
 459         expr: ssl_probe_success == 0
 460         for: 60m
 461         labels:
 462           alertgroup: ssl
 463       - alert: ssl certificate expiry
 464         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 465         for: 0m
 466         labels:
 467           alertgroup: ssl
 468         annotations:
 469           expires_in: "{{ $value | humanizeDuration }}"
 470       - alert: ssl certificate revoked
 471         expr: ssl_ocsp_response_status == 1
 472         for: 0m
 473         labels:
 474           alertgroup: ssl
 475       - alert: ocsp status unknown
 476         expr: ssl_ocsp_response_status == 1
 477         for: 0m
 478         labels:
 479           alertgroup: ssl
 480   - name: statuscake
 481     rules:
 482       - alert: statuscake uptime check failing
 483         expr: statuscake_uptime{status="down",paused="false"} > 0
 484         for: 10m
 485         labels:
 486           alertgroup: statuscake
 487   - name: systemd
 488     rules:
 489       - alert: systemd failed service
 490         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 491         for: 5m
 492         labels:
 493           alertgroup: "{{ $labels.instance }}"
 494       - alert: systemd failed service
 495         expr: node_systemd_unit_state{state="failed",name="chef-client.service"} == 1
 496         for: 6h
 497         labels:
 498           alertgroup: "{{ $labels.instance }}"
 499   - name: tile
 500     rules:
 501       - alert: renderd replication delay
 502         expr: renderd_replication_delay > 120
 503         for: 15m
 504         labels:
 505           alertgroup: tile
 506         annotations:
 507           delay: "{{ $value | humanizeDuration }}"
 508       - alert: missed tile rate
 509         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 510         for: 5m
 511         labels:
 512           alertgroup: tile
 513         annotations:
 514           miss_rate: "{{ $value | humanizePercentage }}"
 515   - name: time
 516     rules:
 517       - alert: clock not synchronising
 518         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 519         for: 5m
 520         labels:
 521           alertgroup: "{{ $labels.instance }}"
 522       - alert: clock skew detected
 523         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)
 524         for: 5m
 525         labels:
 526           alertgroup: "{{ $labels.instance }}"
 527         annotations:
 528           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 529   - name: web
 530     rules:
 531       - alert: web error rate
 532         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
 533         for: 5m
 534         labels:
 535           alertgroup: web
 536         annotations:
 537           error_rate: "{{ $value | humanizePercentage }}"
 538       - alert: job processing rate
 539         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
 540         for: 15m
 541         labels:
 542           alertgroup: web
 543         annotations:
 544           job_processing_rate: "{{ $value | humanizePercentage }}"