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

   1 # DO NOT EDIT - This file is being maintained by Chef
   2
   3 groups:
   4   - name: amsterdam
   5     rules:
   6       - alert: uplink
   7         expr: junos_interface_up{site="amsterdam",name=~"ge-[01]/2/2"} != 1
   8         for: 6m
   9         labels:
  10           alertgroup: "amsterdam"
  11         annotations:
  12           status: "{{ $value }}"
  13       - alert: pdu current draw
  14         expr: rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10 > 28
  15         for: 6m
  16         labels:
  17           alertgroup: "amsterdam"
  18         annotations:
  19           current: "{{ $value | humanize }}A"
  20       - alert: site power
  21         expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="amsterdam",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 3.5
  22         for: 6m
  23         labels:
  24           alertgroup: "amsterdam"
  25         annotations:
  26           current: "{{ $value | humanize }}kVA"
  27       - alert: site temperature
  28         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 15 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 32
  29         for: 6m
  30         labels:
  31           alertgroup: "amsterdam"
  32         annotations:
  33           temperature: "{{ $value | humanize }}C"
  34       - alert: site humidity
  35         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 < 0.08 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 > 0.8
  36         for: 6m
  37         labels:
  38           alertgroup: "amsterdam"
  39         annotations:
  40           humidity: "{{ $value | humanizePercentage }}"
  41   - name: apache
  42     rules:
  43       - alert: apache down
  44         expr: apache_up == 0
  45         for: 5m
  46         labels:
  47           alertgroup: "{{ $labels.instance }}"
  48       - alert: apache workers busy
  49         expr: sum(apache_workers{state="busy"}) by (instance) / sum(apache_scoreboard) by (instance) > 0.8
  50         for: 5m
  51         labels:
  52           alertgroup: "{{ $labels.instance }}"
  53         annotations:
  54           busy_workers: "{{ $value | humanizePercentage }}"
  55       - alert: apache connection limit
  56         expr: (apache_connections{state="total"} - on (instance) apache_connections{state="closing"}) / on (instance) (apache_server_limit * on (instance) (apache_threads_per_child + on (instance) (apache_async_request_worker_factor * on (instance) apache_workers{state="idle"} / on(instance) apache_processes{state="all"}))) > 0.8
  57         for: 5m
  58         labels:
  59           alertgroup: "{{ $labels.instance }}"
  60         annotations:
  61           connections: "{{ $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]) > 0.75
 101         for: 60m
 102         labels:
 103           alertgroup: "{{ $labels.instance }}"
 104         annotations:
 105           pressure: "{{ $value | humanizePercentage }}"
 106   - name: database
 107     rules:
 108       - alert: active rails queries
 109         expr: sum(pg_stat_activity_count{datname="openstreetmap",usename="rails",state="active"}) by (instance) > 50 and on (instance) chef_role{name="db-master"}
 110         for: 5m
 111         labels:
 112           alertgroup: database
 113         annotations:
 114           queries: "{{ $value }}"
 115       - alert: active cgimap queries
 116         expr: sum(pg_stat_activity_count{datname="openstreetmap",usename="cgimap",state="active"}) by (instance) > 30 and on (instance) chef_role{name="db-master"}
 117         for: 5m
 118         labels:
 119           alertgroup: database
 120         annotations:
 121           delay: "{{ $value }}"
 122   - name: discourse
 123     rules:
 124       - alert: discourse job failure rate
 125         expr: rate(discourse_job_failures[5m]) > 0
 126         for: 5m
 127         labels:
 128           alertgroup: discourse
 129         annotations:
 130           failure_rate: "{{ $value }} jobs/s"
 131   - name: dublin
 132     rules:
 133       - alert: uplink
 134         expr: junos_interface_up{site="dublin",name=~"ge-[01]/2/2"} != 1
 135         for: 6m
 136         labels:
 137           alertgroup: "dublin"
 138         annotations:
 139           status: "{{ $value }}"
 140       - alert: pdu current draw
 141         expr: rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10 > 28
 142         for: 6m
 143         labels:
 144           alertgroup: "dublin"
 145         annotations:
 146           current: "{{ $value | humanize }}A"
 147       - alert: site power
 148         expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="dublin",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 4
 149         for: 6m
 150         labels:
 151           alertgroup: "dublin"
 152         annotations:
 153           current: "{{ $value | humanize }}kVA"
 154       - alert: site temperature
 155         expr: min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 > 26
 156         for: 6m
 157         labels:
 158           alertgroup: "dublin"
 159         annotations:
 160           temperature: "{{ $value | humanize }}C"
 161       - alert: site humidity
 162         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 > 0.65
 163         for: 6m
 164         labels:
 165           alertgroup: "dublin"
 166         annotations:
 167           humidity: "{{ $value | humanizePercentage }}"
 168   - name: fastly
 169     rules:
 170       - alert: fastly error rate
 171         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
 172         for: 15m
 173         labels:
 174           alertgroup: fastly
 175         annotations:
 176           error_rate: "{{ $value | humanizePercentage }}"
 177       - alert: fastly frontend healthcheck warning
 178         expr: count(fastly_healthcheck_status == 0) by (service, datacenter) > 2
 179         for: 15m
 180         labels:
 181           alertgroup: fastly
 182       - alert: fastly frontend healthcheck critical
 183         expr: count(fastly_healthcheck_status == 0) by (service, datacenter) == count(fastly_healthcheck_status) by (service, datacenter)
 184         for: 5m
 185         labels:
 186           alertgroup: fastly
 187       - alert: fastly backend healthcheck warning
 188         expr: count(fastly_healthcheck_status == 0) by (service, backend) > 10
 189         for: 15m
 190         labels:
 191           alertgroup: fastly
 192       - alert: fastly backend healthcheck critical
 193         expr: count(fastly_healthcheck_status == 0) by (service, backend) == count(fastly_healthcheck_status) by (service, backend)
 194         for: 5m
 195         labels:
 196           alertgroup: fastly
 197   - name: filesystem
 198     rules:
 199       - alert: readonly filesystem
 200         expr: node_filesystem_readonly > min_over_time(node_filesystem_readonly[7d])
 201         for: 0m
 202         labels:
 203           alertgroup: "{{ $labels.instance }}"
 204       - alert: filesystem low on space
 205         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 206         for: 5m
 207         labels:
 208           alertgroup: "{{ $labels.instance }}"
 209         annotations:
 210           percentage_free: "{{ $value | humanizePercentage }}"
 211           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 212           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 213       - alert: filesystem low on inodes
 214         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 215         for: 5m
 216         labels:
 217           alertgroup: "{{ $labels.instance }}"
 218         annotations:
 219           percentage_free: "{{ $value | humanizePercentage }}"
 220           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 221           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 222   - name: hwmon
 223     rules:
 224       - alert: hwmon fan alarm
 225         expr: node_hwmon_fan_alarm == 1
 226         for: 5m
 227         labels:
 228           alertgroup: "{{ $labels.instance }}"
 229         annotations:
 230           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}}"
 231           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}}"
 232       - alert: hwmon temperature alarm
 233         expr: node_hwmon_temp_alarm == 1
 234         for: 5m
 235         labels:
 236           alertgroup: "{{ $labels.instance }}"
 237         annotations:
 238           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}}"
 239           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}}"
 240           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}}"
 241       - alert: hwmon voltage alarm
 242         expr: node_hwmon_in_alarm == 1
 243         for: 5m
 244         labels:
 245           alertgroup: "{{ $labels.instance }}"
 246         annotations:
 247           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}}"
 248           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}}"
 249           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}}"
 250   - name: io
 251     rules:
 252       - alert: io pressure
 253         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 254         for: 60m
 255         labels:
 256           alertgroup: "{{ $labels.instance }}"
 257         annotations:
 258           pressure: "{{ $value | humanizePercentage }}"
 259   - name: ipmi
 260     rules:
 261       - alert: ipmi fan alarm
 262         expr: ipmi_fan_speed_state > 0
 263         for: 5m
 264         labels:
 265           alertgroup: "{{ $labels.instance }}"
 266         annotations:
 267           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 268       - alert: ipmi temperature alarm
 269         expr: ipmi_temperature_state > 0
 270         for: 5m
 271         labels:
 272           alertgroup: "{{ $labels.instance }}"
 273         annotations:
 274           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 275       - alert: ipmi voltage alarm
 276         expr: ipmi_voltage_state > 0
 277         for: 5m
 278         labels:
 279           alertgroup: "{{ $labels.instance }}"
 280         annotations:
 281           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 282       - alert: ipmi power alarm
 283         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 284         for: 5m
 285         labels:
 286           alertgroup: "{{ $labels.instance }}"
 287   - name: juniper
 288     rules:
 289       - alert: juniper red alarms
 290         expr: juniper_alarms_red_count > 0
 291         for: 5m
 292         labels:
 293           alertgroup: "{{ $labels.site }}"
 294         annotations:
 295           alarm_count: "{{ $value }} alarms"
 296       - alert: juniper yellow alarms
 297         expr: juniper_alarms_yellow_count > 0
 298         for: 5m
 299         labels:
 300           alertgroup: "{{ $labels.site }}"
 301         annotations:
 302           alarm_count: "{{ $value }} alarms"
 303       - alert: juniper cpu alarm
 304         expr: junos_route_engine_load_average_five / 2 > 0.5
 305         for: 5m
 306         labels:
 307           alertgroup: "{{ $labels.site }}"
 308         annotations:
 309           load_average: "{{ $value | humanizePercentage }}"
 310       - alert: juniper fan alarm
 311         expr: junos_environment_fan_up != 1
 312         for: 5m
 313         labels:
 314           alertgroup: "{{ $labels.site }}"
 315       - alert: juniper power alarm
 316         expr: junos_environment_power_up != 1
 317         for: 5m
 318         labels:
 319           alertgroup: "{{ $labels.site }}"
 320       - alert: juniper laser receive power
 321         expr: junos_interface_diagnostics_laser_rx_dbm < -12 and on (site, instance, name) junos_interface_admin_up == 1
 322         for: 5m
 323         labels:
 324           alertgroup: "{{ $labels.site }}"
 325         annotations:
 326           power: "{{ $value }} dBm"
 327       - alert: juniper laser transmit power
 328         expr: junos_interface_diagnostics_laser_output_dbm < -8 and on (site, instance, name) junos_interface_admin_up == 1
 329         for: 5m
 330         labels:
 331           alertgroup: "{{ $labels.site }}"
 332         annotations:
 333           power: "{{ $value }} dBm"
 334   - name: mail
 335     rules:
 336       - alert: exim down
 337         expr: exim_up == 0
 338         for: 5m
 339         labels:
 340           alertgroup: "{{ $labels.instance }}"
 341       - alert: exim queue length
 342         expr: exim_queue > ignoring(job) exim_queue_limit
 343         for: 60m
 344         labels:
 345           alertgroup: mail
 346         annotations:
 347           queue_length: "{{ $value }}"
 348       - alert: mailman queue length
 349         expr: mailman_queue_length > 200
 350         for: 60m
 351         labels:
 352           alertgroup: mail
 353         annotations:
 354           queue_length: "{{ $value }}"
 355   - name: mdadm
 356     rules:
 357       - alert: mdadm array inactive
 358         expr: node_md_state{state="inactive"} > 0
 359         for: 0m
 360         labels:
 361           alertgroup: "{{ $labels.instance }}"
 362         annotations:
 363           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 364           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 365           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 366           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 367       - alert: mdadm array degraded
 368         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 369         for: 0m
 370         labels:
 371           alertgroup: "{{ $labels.instance }}"
 372         annotations:
 373           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 374           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 375           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 376           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 377       - alert: mdadm disk failed
 378         expr: node_md_disks{state="failed"} > 0
 379         for: 0m
 380         labels:
 381           alertgroup: "{{ $labels.instance }}"
 382         annotations:
 383           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 384           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 385           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 386           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 387   - name: memory
 388     rules:
 389       - alert: low memory
 390         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 391         for: 15m
 392         labels:
 393           alertgroup: "{{ $labels.instance }}"
 394         annotations:
 395           memory_free: "{{ $value | humanizePercentage }}"
 396       - alert: memory pressure
 397         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 398         for: 60m
 399         labels:
 400           alertgroup: "{{ $labels.instance }}"
 401         annotations:
 402           pressure: "{{ $value | humanizePercentage }}"
 403       - alert: oom kill detected
 404         expr: increase(node_vmstat_oom_kill[1m]) > 0
 405         for: 0m
 406         labels:
 407           alertgroup: "{{ $labels.instance }}"
 408         annotations:
 409           new_oom_kills: "{{ $value }}"
 410   - name: mysql
 411     rules:
 412       - alert: mysql down
 413         expr: mysql_up == 0
 414         for: 1m
 415         labels:
 416           alertgroup: "{{ $labels.instance }}"
 417       - alert: mysql connection limit
 418         expr: mysql_global_status_max_used_connections / mysql_global_variables_max_connections > 0.8
 419         for: 1m
 420         labels:
 421           alertgroup: "{{ $labels.instance }}"
 422         annotations:
 423           connections_used: "{{ $value | humanizePercentage }}"
 424       - alert: mysql connection errors
 425         expr: increase(mysql_global_status_connection_errors_total[1m]) > 0
 426         for: 0m
 427         labels:
 428           alertgroup: "{{ $labels.instance }}"
 429         annotations:
 430           error_count: "{{ $value }}"
 431   - name: network
 432     rules:
 433       - alert: interface redundancy lost
 434         expr: node_bonding_active < 2 and on (instance, master) label_replace(chef_network_interface{bond_mode="802.3ad"}, "master", "$1", "name", "(.*)")
 435         for: 5m
 436         labels:
 437           alertgroup: "{{ $labels.instance }}"
 438         annotations:
 439           link_count: "{{ $value }}"
 440       - alert: interface transmit rate
 441         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.99
 442         for: 5m
 443         labels:
 444           alertgroup: "{{ $labels.instance }}"
 445         annotations:
 446           bandwidth_used: "{{ $value | humanizePercentage }}"
 447       - alert: interface receive rate
 448         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.99
 449         for: 5m
 450         labels:
 451           alertgroup: "{{ $labels.instance }}"
 452         annotations:
 453           bandwidth_used: "{{ $value | humanizePercentage }}"
 454       - alert: interface transmit errors
 455         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 456         for: 5m
 457         labels:
 458           alertgroup: "{{ $labels.instance }}"
 459         annotations:
 460           error_rate: "{{ $value | humanizePercentage }}"
 461       - alert: wireguard interface transmit errors
 462         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 463         for: 1h
 464         labels:
 465           alertgroup: "{{ $labels.instance }}"
 466         annotations:
 467           error_rate: "{{ $value | humanizePercentage }}"
 468       - alert: interface receive errors
 469         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 470         for: 5m
 471         labels:
 472           alertgroup: "{{ $labels.instance }}"
 473         annotations:
 474           error_rate: "{{ $value | humanizePercentage }}"
 475       - alert: conntrack entries
 476         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 477         for: 5m
 478         labels:
 479           alertgroup: "{{ $labels.instance }}"
 480         annotations:
 481           entries_used: "{{ $value | humanizePercentage }}"
 482   - name: nominatim
 483     rules:
 484       - alert: nominatim replication delay
 485         expr: nominatim_replication_delay > 10800
 486         for: 1h
 487         labels:
 488           alertgroup: nominatim
 489         annotations:
 490           delay: "{{ $value | humanizeDuration }}"
 491   - name: overpass
 492     rules:
 493       - alert: overpass osm database age
 494         expr: overpass_database_age_seconds{database="osm"} > 3600
 495         for: 1h
 496         labels:
 497           alertgroup: overpass
 498         annotations:
 499           age: "{{ $value | humanizeDuration }}"
 500       - alert: overpass area database age
 501         expr: overpass_database_age_seconds{database="area"} > 86400
 502         for: 1h
 503         labels:
 504           alertgroup: overpass
 505         annotations:
 506           age: "{{ $value | humanizeDuration }}"
 507   - name: passenger
 508     rules:
 509       - alert: passenger down
 510         expr: passenger_up == 0
 511         for: 5m
 512         labels:
 513           alertgroup: "{{ $labels.instance }}"
 514       - alert: passenger queuing
 515         expr: passenger_top_level_request_queue > 0
 516         for: 5m
 517         labels:
 518           alertgroup: "{{ $labels.instance }}"
 519       - alert: passenger application queuing
 520         expr: passenger_app_request_queue > 0
 521         for: 5m
 522         labels:
 523           alertgroup: "{{ $labels.instance }}"
 524   - name: planet
 525     rules:
 526       - alert: planet dump overdue
 527         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 528         for: 24h
 529         labels:
 530           alertgroup: planet
 531         annotations:
 532           overdue_by: "{{ $value | humanizeDuration }}"
 533       - alert: notes dump overdue
 534         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 535         for: 6h
 536         labels:
 537           alertgroup: planet
 538         annotations:
 539           overdue_by: "{{ $value | humanizeDuration }}"
 540       - alert: daily replication feed delayed
 541         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 542         for: 3h
 543         labels:
 544           alertgroup: planet
 545         annotations:
 546           delayed_by: "{{ $value | humanizeDuration }}"
 547       - alert: hourly replication feed delayed
 548         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 549         for: 30m
 550         labels:
 551           alertgroup: planet
 552         annotations:
 553           delayed_by: "{{ $value | humanizeDuration }}"
 554       - alert: minutely replication feed delayed
 555         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 556         for: 5m
 557         labels:
 558           alertgroup: planet
 559         annotations:
 560           delayed_by: "{{ $value | humanizeDuration }}"
 561       - alert: changeset replication feed delayed
 562         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 563         for: 5m
 564         labels:
 565           alertgroup: planet
 566         annotations:
 567           delayed_by: "{{ $value | humanizeDuration }}"
 568   - name: postgresql
 569     rules:
 570       - alert: postgresql down
 571         expr: pg_up == 0
 572         for: 1m
 573         labels:
 574           alertgroup: "{{ $labels.instance }}"
 575       - alert: postgresql replication delay
 576         expr: pg_replication_lag_seconds > 30
 577         for: 15m
 578         labels:
 579           alertgroup: "{{ $labels.instance }}"
 580         annotations:
 581           delay: "{{ $value | humanizeDuration }}"
 582       - alert: postgresql connection limit
 583         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 584         for: 1m
 585         labels:
 586           alertgroup: "{{ $labels.instance }}"
 587         annotations:
 588           connections_used: "{{ $value | humanizePercentage }}"
 589       - alert: postgresql deadlocks
 590         expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
 591         for: 0m
 592         labels:
 593           alertgroup: "{{ $labels.instance }}"
 594         annotations:
 595           new_deadlocks: "{{ $value }}"
 596       - alert: postgresql idle transactions
 597         expr: sum(pg_process_idle_seconds_count{state="idle in transaction"}) by (instance, server) > sum(pg_process_idle_seconds_bucket{state="idle in transaction",le="300"}) by (instance, server)
 598         for: 5m
 599         labels:
 600           alertgroup: "{{ $labels.instance }}"
 601         annotations:
 602           queries: "{{ $value }}"
 603   - name: prometheus
 604     rules:
 605       - alert: prometheus configuration error
 606         expr: prometheus_config_last_reload_successful == 0
 607         for: 10m
 608         labels:
 609           alertgroup: "prometheus"
 610       - alert: prometheus target missing
 611         expr: up == 0
 612         for: 10m
 613         labels:
 614           alertgroup: "prometheus"
 615       - alert: node exporter text file scrape error
 616         expr: node_textfile_scrape_error > 0
 617         for: 10m
 618         labels:
 619           alertgroup: "prometheus"
 620   - name: raid
 621     rules:
 622       - alert: raid controller battery failed
 623         expr: ohai_controller_info{battery_status="failed"} > 0
 624         for: 5m
 625         labels:
 626           alertgroup: "{{ $labels.instance }}"
 627       - alert: raid controller battery recharging
 628         expr: ohai_controller_info{battery_status="recharging"} > 0
 629         for: 4h
 630         labels:
 631           alertgroup: "{{ $labels.instance }}"
 632       - alert: raid array degraded
 633         expr: ohai_array_info{status="degraded"} > 0
 634         for: 5m
 635         labels:
 636           alertgroup: "{{ $labels.instance }}"
 637       - alert: raid disk failed
 638         expr: ohai_disk_info{status="failed"} > 0
 639         for: 5m
 640         labels:
 641           alertgroup: "{{ $labels.instance }}"
 642   - name: rasdaemon
 643     rules:
 644       - alert: memory controller errors
 645         expr: increase(rasdaemon_mc_events_total[1m]) > 0
 646         for: 0m
 647         labels:
 648           alertgroup: "{{ $labels.instance }}"
 649         annotations:
 650           new_errors: "{{ $value }}"
 651       - alert: pcie aer errors
 652         expr: increase(rasdaemon_aer_events_total[1m]) > 0
 653         for: 0m
 654         labels:
 655           alertgroup: "{{ $labels.instance }}"
 656         annotations:
 657           new_errors: "{{ $value }}"
 658   - name: smart
 659     rules:
 660       - alert: smart failure
 661         expr: smart_health_status == 0
 662         for: 60m
 663         labels:
 664           alertgroup: "{{ $labels.instance }}"
 665       - alert: smart ssd wearout approaching
 666         expr: smart_percentage_used / 100 >= 0.8
 667         for: 60m
 668         labels:
 669           alertgroup: "{{ $labels.instance }}"
 670         annotations:
 671           percentage_used: "{{ $value | humanizePercentage }}"
 672   - name: smokeping
 673     rules:
 674       - alert: packet loss
 675         expr: 1 - (rate(smokeping_response_duration_seconds_count[5m]) / rate(smokeping_requests_total[5m])) > 0.02
 676         for: 10m
 677         labels:
 678           alertgroup: smokeping
 679         annotations:
 680           loss_rate: "{{ $value | humanizePercentage }}"
 681   - name: snmp
 682     rules:
 683       - alert: snmp pdus missing
 684         expr: max_over_time(snmp_scrape_pdus_returned[1d]) - snmp_scrape_pdus_returned > 0
 685         for: 15m
 686         labels:
 687           alertgroup: snmp
 688         annotations:
 689           missing_pdus: "{{ $value }}"
 690   - name: ssl
 691     rules:
 692       - alert: ssl certificate probe failed
 693         expr: ssl_probe_success == 0
 694         for: 60m
 695         labels:
 696           alertgroup: ssl
 697       - alert: ssl certificate expiry
 698         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 699         for: 0m
 700         labels:
 701           alertgroup: ssl
 702         annotations:
 703           expires_in: "{{ $value | humanizeDuration }}"
 704       - alert: ssl certificate revoked
 705         expr: ssl_ocsp_response_status == 1
 706         for: 0m
 707         labels:
 708           alertgroup: ssl
 709       - alert: ocsp status unknown
 710         expr: ssl_ocsp_response_status == 1
 711         for: 0m
 712         labels:
 713           alertgroup: ssl
 714   - name: statuscake
 715     rules:
 716       - alert: statuscake uptime check failing
 717         expr: statuscake_paused == 0 and statuscake_up == 0
 718         for: 10m
 719         labels:
 720           alertgroup: statuscake
 721   - name: systemd
 722     rules:
 723       - alert: systemd failed service
 724         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 725         for: 5m
 726         labels:
 727           alertgroup: "{{ $labels.instance }}"
 728       - alert: systemd failed chef client service
 729         expr: sum_over_time(node_systemd_unit_state{state="inactive",name="chef-client.service"}[6h]) == 0
 730         for: 0m
 731         labels:
 732           alertgroup: "{{ $labels.instance }}"
 733   - name: taginfo
 734     rules:
 735       - alert: taginfo planet age
 736         expr: time() - taginfo_data_from_seconds > 129600 and on (instance) chef_role{name="taginfo"}
 737         for: 0m
 738         labels:
 739           alertgroup: taginfo
 740         annotations:
 741           age: "{{ $value | humanizeDuration }}"
 742       - alert: taginfo database age
 743         expr: time() - taginfo_database_update_finish_seconds > 129600 and on (instance) chef_role{name="taginfo"}
 744         for: 0m
 745         labels:
 746           alertgroup: taginfo
 747         annotations:
 748           age: "{{ $value | humanizeDuration }}"
 749       - alert: taginfo database size
 750         expr: abs(delta(taginfo_database_size_bytes[30m])) / taginfo_database_size_bytes > 0.1
 751         for: 30m
 752         labels:
 753           alertgroup: taginfo
 754         annotations:
 755           size_change: "{{ $value | humanizePercentage }}"
 756   - name: tile
 757     rules:
 758       - alert: renderd replication delay
 759         expr: renderd_replication_delay > 120
 760         for: 15m
 761         labels:
 762           alertgroup: tile
 763         annotations:
 764           delay: "{{ $value | humanizeDuration }}"
 765       - alert: missed tile rate
 766         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 767         for: 5m
 768         labels:
 769           alertgroup: tile
 770         annotations:
 771           miss_rate: "{{ $value | humanizePercentage }}"
 772       - alert: tile render rate
 773         expr: sum(rate(renderd_zoom_metatiles_total[5m])) by (instance) == 0
 774         for: 15m
 775         labels:
 776           alertgroup: tile
 777         annotations:
 778           render_rate: "{{ $value }} tiles/s"
 779   - name: time
 780     rules:
 781       - alert: clock not synchronising
 782         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 783         for: 5m
 784         labels:
 785           alertgroup: "{{ $labels.instance }}"
 786       - alert: clock skew detected
 787         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)
 788         for: 5m
 789         labels:
 790           alertgroup: "{{ $labels.instance }}"
 791         annotations:
 792           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 793   - name: web
 794     rules:
 795       - alert: web error rate
 796         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
 797         for: 5m
 798         labels:
 799           alertgroup: web
 800         annotations:
 801           error_rate: "{{ $value | humanizePercentage }}"
 802       - alert: job processing rate
 803         expr: rate(pg_stat_user_tables_n_tup_del{datname="openstreetmap",relname="delayed_jobs"}[1h]) / rate(pg_stat_user_tables_n_tup_ins{datname="openstreetmap",relname="delayed_jobs"}[1h]) < 0.9 and ignoring(job, name, datname, relname, schemaname, server) chef_role{name="db-master"} == 1
 804         for: 1h
 805         labels:
 806           alertgroup: web
 807         annotations:
 808           job_processing_rate: "{{ $value | humanizePercentage }}"