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: count(fastly_healthcheck_status == 0) > 0
 119         for: 15m
 120         labels:
 121           alertgroup: fastly
 122       - alert: fastly healthcheck failing
 123         expr: count(fastly_healthcheck_status == 0) > 4
 124         for: 5m
 125         labels:
 126           alertgroup: fastly
 127   - name: filesystem
 128     rules:
 129       - alert: readonly filesystem
 130         expr: node_filesystem_readonly == 1
 131         for: 0m
 132         labels:
 133           alertgroup: "{{ $labels.instance }}"
 134       - alert: filesystem low on space
 135         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 136         for: 5m
 137         labels:
 138           alertgroup: "{{ $labels.instance }}"
 139         annotations:
 140           percentage_free: "{{ $value | humanizePercentage }}"
 141           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 142           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 143       - alert: filesystem low on inodes
 144         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 145         for: 5m
 146         labels:
 147           alertgroup: "{{ $labels.instance }}"
 148         annotations:
 149           percentage_free: "{{ $value | humanizePercentage }}"
 150           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 151           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 152   - name: hwmon
 153     rules:
 154       - alert: hwmon fan alarm
 155         expr: node_hwmon_fan_alarm == 1
 156         for: 5m
 157         labels:
 158           alertgroup: "{{ $labels.instance }}"
 159         annotations:
 160           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}}"
 161           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}}"
 162       - alert: hwmon temperature alarm
 163         expr: node_hwmon_temp_alarm == 1
 164         for: 5m
 165         labels:
 166           alertgroup: "{{ $labels.instance }}"
 167         annotations:
 168           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}}"
 169           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}}"
 170           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}}"
 171       - alert: hwmon voltage alarm
 172         expr: node_hwmon_in_alarm == 1
 173         for: 5m
 174         labels:
 175           alertgroup: "{{ $labels.instance }}"
 176         annotations:
 177           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}}"
 178           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}}"
 179           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}}"
 180   - name: io
 181     rules:
 182       - alert: io pressure
 183         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 184         for: 60m
 185         labels:
 186           alertgroup: "{{ $labels.instance }}"
 187         annotations:
 188           pressure: "{{ $value | humanizePercentage }}"
 189   - name: ipmi
 190     rules:
 191       - alert: ipmi fan alarm
 192         expr: ipmi_fan_speed_state > 0
 193         for: 5m
 194         labels:
 195           alertgroup: "{{ $labels.instance }}"
 196         annotations:
 197           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 198       - alert: ipmi temperature alarm
 199         expr: ipmi_temperature_state > 0
 200         for: 5m
 201         labels:
 202           alertgroup: "{{ $labels.instance }}"
 203         annotations:
 204           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 205       - alert: ipmi voltage alarm
 206         expr: ipmi_voltage_state > 0
 207         for: 5m
 208         labels:
 209           alertgroup: "{{ $labels.instance }}"
 210         annotations:
 211           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 212       - alert: ipmi power alarm
 213         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 214         for: 5m
 215         labels:
 216           alertgroup: "{{ $labels.instance }}"
 217   - name: juniper
 218     rules:
 219       - alert: juniper fan alarm
 220         expr: jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!="running"} > 0
 221         for: 5m
 222         labels:
 223           alertgroup: "{{ $labels.site }}"
 224       - alert: juniper power alarm
 225         expr: jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!="running"} > 0
 226         for: 5m
 227         labels:
 228           alertgroup: "{{ $labels.site }}"
 229   - name: mail
 230     rules:
 231       - alert: exim queue length
 232         expr: exim_queue > exim_queue_limit
 233         for: 60m
 234         labels:
 235           alertgroup: mail
 236         annotations:
 237           queue_length: "{{ $value }}"
 238       - alert: mailman queue length
 239         expr: mailman_queue_length > 200
 240         for: 60m
 241         labels:
 242           alertgroup: mail
 243         annotations:
 244           queue_length: "{{ $value }}"
 245   - name: mdadm
 246     rules:
 247       - alert: mdadm array inactive
 248         expr: node_md_state{state="inactive"} > 0
 249         for: 0m
 250         labels:
 251           alertgroup: "{{ $labels.instance }}"
 252         annotations:
 253           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 254           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 255           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 256           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 257       - alert: mdadm array degraded
 258         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 259         for: 0m
 260         labels:
 261           alertgroup: "{{ $labels.instance }}"
 262         annotations:
 263           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 264           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 265           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 266           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 267       - alert: mdadm disk failed
 268         expr: node_md_disks{state="failed"} > 0
 269         for: 0m
 270         labels:
 271           alertgroup: "{{ $labels.instance }}"
 272         annotations:
 273           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 274           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 275           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 276           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 277   - name: memory
 278     rules:
 279       - alert: low memory
 280         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 281         for: 15m
 282         labels:
 283           alertgroup: "{{ $labels.instance }}"
 284         annotations:
 285           memory_free: "{{ $value | humanizePercentage }}"
 286       - alert: memory pressure
 287         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 288         for: 60m
 289         labels:
 290           alertgroup: "{{ $labels.instance }}"
 291         annotations:
 292           pressure: "{{ $value | humanizePercentage }}"
 293       - alert: oom kill detected
 294         expr: increase(node_vmstat_oom_kill[1m]) > 0
 295         for: 0m
 296         labels:
 297           alertgroup: "{{ $labels.instance }}"
 298         annotations:
 299           new_oom_kills: "{{ $value }}"
 300   - name: network
 301     rules:
 302       - alert: interface transmit rate
 303         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 304         for: 5m
 305         labels:
 306           alertgroup: "{{ $labels.instance }}"
 307         annotations:
 308           bandwidth_used: "{{ $value | humanizePercentage }}"
 309       - alert: interface receive rate
 310         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.98
 311         for: 5m
 312         labels:
 313           alertgroup: "{{ $labels.instance }}"
 314         annotations:
 315           bandwidth_used: "{{ $value | humanizePercentage }}"
 316       - alert: interface transmit errors
 317         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 318         for: 5m
 319         labels:
 320           alertgroup: "{{ $labels.instance }}"
 321         annotations:
 322           error_rate: "{{ $value | humanizePercentage }}"
 323       - alert: interface transmit errors
 324         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 325         for: 1h
 326         labels:
 327           alertgroup: "{{ $labels.instance }}"
 328         annotations:
 329           error_rate: "{{ $value | humanizePercentage }}"
 330       - alert: interface receive errors
 331         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 332         for: 5m
 333         labels:
 334           alertgroup: "{{ $labels.instance }}"
 335         annotations:
 336           error_rate: "{{ $value | humanizePercentage }}"
 337       - alert: conntrack entries
 338         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 339         for: 5m
 340         labels:
 341           alertgroup: "{{ $labels.instance }}"
 342         annotations:
 343           entries_used: "{{ $value | humanizePercentage }}"
 344   - name: nominatim
 345     rules:
 346       - alert: nominatim replication delay
 347         expr: nominatim_replication_delay > 10800
 348         for: 1h
 349         labels:
 350           alertgroup: nominatim
 351         annotations:
 352           delay: "{{ $value | humanizeDuration }}"
 353   - name: overpass
 354     rules:
 355       - alert: overpass osm database age
 356         expr: overpass_database_age_seconds{database="osm"} > 300
 357         for: 5m
 358         labels:
 359           alertgroup: overpass
 360         annotations:
 361           age: "{{ $value | humanizeDuration }}"
 362       - alert: overpass area database age
 363         expr: overpass_database_age_seconds{database="area"} > 86400
 364         for: 1h
 365         labels:
 366           alertgroup: overpass
 367         annotations:
 368           age: "{{ $value | humanizeDuration }}"
 369   - name: planet
 370     rules:
 371       - alert: planet dump overdue
 372         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 373         for: 24h
 374         labels:
 375           alertgroup: planet
 376         annotations:
 377           overdue_by: "{{ $value | humanizeDuration }}"
 378       - alert: notes dump overdue
 379         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 380         for: 6h
 381         labels:
 382           alertgroup: planet
 383         annotations:
 384           overdue_by: "{{ $value | humanizeDuration }}"
 385       - alert: daily replication feed delayed
 386         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 387         for: 3h
 388         labels:
 389           alertgroup: planet
 390         annotations:
 391           delayed_by: "{{ $value | humanizeDuration }}"
 392       - alert: hourly replication feed delayed
 393         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 394         for: 30m
 395         labels:
 396           alertgroup: planet
 397         annotations:
 398           delayed_by: "{{ $value | humanizeDuration }}"
 399       - alert: minutely replication feed delayed
 400         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 401         for: 5m
 402         labels:
 403           alertgroup: planet
 404         annotations:
 405           delayed_by: "{{ $value | humanizeDuration }}"
 406       - alert: changeset replication feed delayed
 407         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 408         for: 5m
 409         labels:
 410           alertgroup: planet
 411         annotations:
 412           delayed_by: "{{ $value | humanizeDuration }}"
 413   - name: postgresql
 414     rules:
 415       - alert: postgresql down
 416         expr: pg_up == 0
 417         for: 1m
 418         labels:
 419           alertgroup: "{{ $labels.instance }}"
 420       - alert: postgresql replication delay
 421         expr: pg_replication_lag_seconds > 5
 422         for: 1m
 423         labels:
 424           alertgroup: "{{ $labels.instance }}"
 425         annotations:
 426           delay: "{{ $value | humanizeDuration }}"
 427       - alert: postgresql connection limit
 428         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 429         for: 1m
 430         labels:
 431           alertgroup: "{{ $labels.instance }}"
 432         annotations:
 433           connections_used: "{{ $value | humanizePercentage }}"
 434       - alert: postgresql deadlocks
 435         expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
 436         for: 0m
 437         labels:
 438           alertgroup: "{{ $labels.instance }}"
 439         annotations:
 440           new_deadlocks: "{{ $value }}"
 441       - alert: postgresql slow queries
 442         expr: pg_slow_queries > 0
 443         for: 5m
 444         labels:
 445           alertgroup: "{{ $labels.instance }}"
 446         annotations:
 447           queries: "{{ $value }}"
 448   - name: prometheus
 449     rules:
 450       - alert: prometheus configuration error
 451         expr: prometheus_config_last_reload_successful == 0
 452         for: 10m
 453         labels:
 454           alertgroup: "prometheus"
 455       - alert: prometheus target missing
 456         expr: up == 0
 457         for: 10m
 458         labels:
 459           alertgroup: "prometheus"
 460   - name: raid
 461     rules:
 462       - alert: raid array degraded
 463         expr: ohai_array_info{status="degraded"} > 0
 464         for: 5m
 465         labels:
 466           alertgroup: "{{ $labels.instance }}"
 467       - alert: raid disk failed
 468         expr: ohai_disk_info{status="failed"} > 0
 469         for: 5m
 470         labels:
 471           alertgroup: "{{ $labels.instance }}"
 472   - name: rasdaemon
 473     rules:
 474       - alert: memory controller errors
 475         expr: increase(rasdaemon_mc_events_total[1m]) > 0
 476         for: 0m
 477         labels:
 478           alertgroup: "{{ $labels.instance }}"
 479         annotations:
 480           new_errors: "{{ $value }}"
 481       - alert: pcie aer errors
 482         expr: increase(rasdaemon_aer_events_total[1m]) > 0
 483         for: 0m
 484         labels:
 485           alertgroup: "{{ $labels.instance }}"
 486         annotations:
 487           new_ercrors: "{{ $value }}"
 488   - name: smart
 489     rules:
 490       - alert: smart failure
 491         expr: smart_health_status == 0
 492         for: 60m
 493         labels:
 494           alertgroup: "{{ $labels.instance }}"
 495       - alert: smart ssd wearout approaching
 496         expr: smart_percentage_used >= 80
 497         for: 60m
 498         labels:
 499           alertgroup: "{{ $labels.instance }}"
 500         annotations:
 501           percentage_used: "{{ $value | humanizePercentage }}"
 502   - name: ssl
 503     rules:
 504       - alert: ssl certificate probe failed
 505         expr: ssl_probe_success == 0
 506         for: 60m
 507         labels:
 508           alertgroup: ssl
 509       - alert: ssl certificate expiry
 510         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 511         for: 0m
 512         labels:
 513           alertgroup: ssl
 514         annotations:
 515           expires_in: "{{ $value | humanizeDuration }}"
 516       - alert: ssl certificate revoked
 517         expr: ssl_ocsp_response_status == 1
 518         for: 0m
 519         labels:
 520           alertgroup: ssl
 521       - alert: ocsp status unknown
 522         expr: ssl_ocsp_response_status == 1
 523         for: 0m
 524         labels:
 525           alertgroup: ssl
 526   - name: statuscake
 527     rules:
 528       - alert: statuscake uptime check failing
 529         expr: statuscake_uptime{status="down",paused="false"} > 0
 530         for: 10m
 531         labels:
 532           alertgroup: statuscake
 533   - name: systemd
 534     rules:
 535       - alert: systemd failed service
 536         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 537         for: 5m
 538         labels:
 539           alertgroup: "{{ $labels.instance }}"
 540       - alert: systemd failed service
 541         expr: node_systemd_unit_state{state="failed",name="chef-client.service"} == 1
 542         for: 6h
 543         labels:
 544           alertgroup: "{{ $labels.instance }}"
 545   - name: tile
 546     rules:
 547       - alert: renderd replication delay
 548         expr: renderd_replication_delay > 120
 549         for: 15m
 550         labels:
 551           alertgroup: tile
 552         annotations:
 553           delay: "{{ $value | humanizeDuration }}"
 554       - alert: missed tile rate
 555         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 556         for: 5m
 557         labels:
 558           alertgroup: tile
 559         annotations:
 560           miss_rate: "{{ $value | humanizePercentage }}"
 561   - name: time
 562     rules:
 563       - alert: clock not synchronising
 564         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 565         for: 5m
 566         labels:
 567           alertgroup: "{{ $labels.instance }}"
 568       - alert: clock skew detected
 569         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)
 570         for: 5m
 571         labels:
 572           alertgroup: "{{ $labels.instance }}"
 573         annotations:
 574           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 575   - name: web
 576     rules:
 577       - alert: web error rate
 578         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
 579         for: 5m
 580         labels:
 581           alertgroup: web
 582         annotations:
 583           error_rate: "{{ $value | humanizePercentage }}"
 584       - alert: job processing rate
 585         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
 586         for: 15m
 587         labels:
 588           alertgroup: web
 589         annotations:
 590           job_processing_rate: "{{ $value | humanizePercentage }}"