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: he 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: equinix uplink
  14         expr: junos_interface_up{site="amsterdam",name=~"xe-[01]/2/0"} != 1
  15         for: 6m
  16         labels:
  17           alertgroup: "amsterdam"
  18         annotations:
  19           status: "{{ $value }}"
  20       - alert: pdu current draw
  21         expr: rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10 > 28
  22         for: 6m
  23         labels:
  24           alertgroup: "amsterdam"
  25         annotations:
  26           current: "{{ $value | humanize }}A"
  27       - alert: site power
  28         expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="amsterdam",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 3.5
  29         for: 6m
  30         labels:
  31           alertgroup: "amsterdam"
  32         annotations:
  33           current: "{{ $value | humanize }}kVA"
  34       - alert: site temperature
  35         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 15 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 32
  36         for: 6m
  37         labels:
  38           alertgroup: "amsterdam"
  39         annotations:
  40           temperature: "{{ $value | humanize }}C"
  41       - alert: site humidity
  42         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 < 0.08 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 > 0.8
  43         for: 6m
  44         labels:
  45           alertgroup: "amsterdam"
  46         annotations:
  47           humidity: "{{ $value | humanizePercentage }}"
  48   - name: apache
  49     rules:
  50       - alert: apache down
  51         expr: apache_up == 0
  52         for: 5m
  53         labels:
  54           alertgroup: "{{ $labels.instance }}"
  55       - alert: apache workers busy
  56         expr: sum(apache_workers{state="busy"}) by (instance) / sum(apache_scoreboard) by (instance) > 0.8
  57         for: 5m
  58         labels:
  59           alertgroup: "{{ $labels.instance }}"
  60         annotations:
  61           busy_workers: "{{ $value | humanizePercentage }}"
  62       - alert: apache connection limit
  63         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
  64         for: 5m
  65         labels:
  66           alertgroup: "{{ $labels.instance }}"
  67         annotations:
  68           connections: "{{ $value | humanizePercentage }}"
  69   - name: chef
  70     rules:
  71       - alert: chef client not running
  72         expr: time() - node_systemd_timer_last_trigger_seconds{name="chef-client.timer"} > 3600
  73         for: 12h
  74         labels:
  75           alertgroup: "{{ $labels.instance }}"
  76         annotations:
  77           down_time: "{{ $value | humanizeDuration }}"
  78   - name: cisco
  79     rules:
  80       - alert: cisco fan alarm
  81         expr: rlPhdUnitEnvParamFan1Status{rlPhdUnitEnvParamFan1Status!="normal"} > 0 or rlPhdUnitEnvParamFan2Status{rlPhdUnitEnvParamFan2Status!="normal"} > 0
  82         for: 5m
  83         labels:
  84           alertgroup: "{{ $labels.site }}"
  85         annotations:
  86           fan_rpm: "{{ with printf \"rlPhdUnitEnvParamFan1Speed{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}rpm{{end}}"
  87       - alert: cisco temperature alarm
  88         expr: rlPhdUnitEnvParamTempSensorStatus{rlPhdUnitEnvParamTempSensorStatus!="ok"} > 0
  89         for: 5m
  90         labels:
  91           alertgroup: "{{ $labels.site }}"
  92         annotations:
  93           temp_celsius: "{{ with printf \"rlPhdUnitEnvParamTempSensorValue{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}C{{end}}"
  94       - alert: cisco main power alarm
  95         expr: rlPhdUnitEnvParamMainPSStatus{rlPhdUnitEnvParamMainPSStatus!="normal"} > 0
  96         for: 5m
  97         labels:
  98           alertgroup: "{{ $labels.site }}"
  99       - alert: cisco redundant power alarm
 100         expr: rlPhdUnitEnvParamRedundantPSStatus{rlPhdUnitEnvParamRedundantPSStatus!="normal"} > 0
 101         for: 5m
 102         labels:
 103           alertgroup: "{{ $labels.site }}"
 104   - name: cpu
 105     rules:
 106       - alert: cpu pressure
 107         expr: rate(node_pressure_cpu_waiting_seconds_total[5m]) > 0.75
 108         for: 60m
 109         labels:
 110           alertgroup: "{{ $labels.instance }}"
 111         annotations:
 112           pressure: "{{ $value | humanizePercentage }}"
 113   - name: database
 114     rules:
 115       - alert: active rails queries
 116         expr: sum(pg_stat_activity_count{datname="openstreetmap",usename="rails",state="active"}) by (instance) > 50 and on (instance) chef_role{name="db-master"}
 117         for: 5m
 118         labels:
 119           alertgroup: database
 120         annotations:
 121           queries: "{{ $value }}"
 122       - alert: active cgimap queries
 123         expr: sum(pg_stat_activity_count{datname="openstreetmap",usename="cgimap",state="active"}) by (instance) > 30 and on (instance) chef_role{name="db-master"}
 124         for: 5m
 125         labels:
 126           alertgroup: database
 127         annotations:
 128           queries: "{{ $value }}"
 129   - name: discourse
 130     rules:
 131       - alert: discourse job failure rate
 132         expr: rate(discourse_job_failures[5m]) > 0
 133         for: 5m
 134         labels:
 135           alertgroup: discourse
 136         annotations:
 137           failure_rate: "{{ $value }} jobs/s"
 138   - name: dublin
 139     rules:
 140       - alert: he uplink
 141         expr: junos_interface_up{site="dublin",name=~"ge-[01]/2/2"} != 1
 142         for: 6m
 143         labels:
 144           alertgroup: "dublin"
 145         annotations:
 146           status: "{{ $value }}"
 147       - alert: equinix uplink
 148         expr: junos_interface_up{site="dublin",name=~"xe-[01]/2/0"} != 1
 149         for: 6m
 150         labels:
 151           alertgroup: "dublin"
 152         annotations:
 153           status: "{{ $value }}"
 154       - alert: pdu current draw
 155         expr: rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10 > 28
 156         for: 6m
 157         labels:
 158           alertgroup: "dublin"
 159         annotations:
 160           current: "{{ $value | humanize }}A"
 161       - alert: site power
 162         expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="dublin",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 4
 163         for: 6m
 164         labels:
 165           alertgroup: "dublin"
 166         annotations:
 167           current: "{{ $value | humanize }}kVA"
 168       - alert: site temperature
 169         expr: min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 > 26
 170         for: 6m
 171         labels:
 172           alertgroup: "dublin"
 173         annotations:
 174           temperature: "{{ $value | humanize }}C"
 175       - alert: site humidity
 176         expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 > 0.65
 177         for: 6m
 178         labels:
 179           alertgroup: "dublin"
 180         annotations:
 181           humidity: "{{ $value | humanizePercentage }}"
 182   - name: fastly
 183     rules:
 184       - alert: fastly error rate
 185         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
 186         for: 15m
 187         labels:
 188           alertgroup: fastly
 189         annotations:
 190           error_rate: "{{ $value | humanizePercentage }}"
 191       - alert: fastly frontend healthcheck warning
 192         expr: count(fastly_healthcheck_status == 0) by (service, datacenter) > 2
 193         for: 15m
 194         labels:
 195           alertgroup: fastly
 196       - alert: fastly frontend healthcheck critical
 197         expr: count(fastly_healthcheck_status == 0) by (service, datacenter) == count(fastly_healthcheck_status) by (service, datacenter)
 198         for: 5m
 199         labels:
 200           alertgroup: fastly
 201       - alert: fastly backend healthcheck warning
 202         expr: count(fastly_healthcheck_status == 0) by (service, backend) > 10
 203         for: 15m
 204         labels:
 205           alertgroup: fastly
 206       - alert: fastly backend healthcheck critical
 207         expr: count(fastly_healthcheck_status == 0) by (service, backend) == count(fastly_healthcheck_status) by (service, backend)
 208         for: 5m
 209         labels:
 210           alertgroup: fastly
 211   - name: filesystem
 212     rules:
 213       - alert: readonly filesystem
 214         expr: node_filesystem_readonly > min_over_time(node_filesystem_readonly[7d])
 215         for: 0m
 216         labels:
 217           alertgroup: "{{ $labels.instance }}"
 218       - alert: filesystem low on space
 219         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
 220         for: 5m
 221         labels:
 222           alertgroup: "{{ $labels.instance }}"
 223         annotations:
 224           percentage_free: "{{ $value | humanizePercentage }}"
 225           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 226           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
 227       - alert: filesystem low on inodes
 228         expr: node_filesystem_files_free / node_filesystem_files < 0.1
 229         for: 5m
 230         labels:
 231           alertgroup: "{{ $labels.instance }}"
 232         annotations:
 233           percentage_free: "{{ $value | humanizePercentage }}"
 234           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 235           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
 236   - name: hwmon
 237     rules:
 238       - alert: hwmon fan alarm
 239         expr: node_hwmon_fan_alarm == 1
 240         for: 5m
 241         labels:
 242           alertgroup: "{{ $labels.instance }}"
 243         annotations:
 244           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}}"
 245           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}}"
 246       - alert: hwmon temperature alarm
 247         expr: node_hwmon_temp_alarm == 1
 248         for: 5m
 249         labels:
 250           alertgroup: "{{ $labels.instance }}"
 251         annotations:
 252           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}}"
 253           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}}"
 254           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}}"
 255       - alert: hwmon voltage alarm
 256         expr: node_hwmon_in_alarm == 1
 257         for: 5m
 258         labels:
 259           alertgroup: "{{ $labels.instance }}"
 260         annotations:
 261           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}}"
 262           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}}"
 263           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}}"
 264   - name: io
 265     rules:
 266       - alert: io pressure
 267         expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6
 268         for: 60m
 269         labels:
 270           alertgroup: "{{ $labels.instance }}"
 271         annotations:
 272           pressure: "{{ $value | humanizePercentage }}"
 273   - name: ipmi
 274     rules:
 275       - alert: ipmi fan alarm
 276         expr: ipmi_fan_speed_state > 0
 277         for: 5m
 278         labels:
 279           alertgroup: "{{ $labels.instance }}"
 280         annotations:
 281           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
 282       - alert: ipmi temperature alarm
 283         expr: ipmi_temperature_state > 0
 284         for: 5m
 285         labels:
 286           alertgroup: "{{ $labels.instance }}"
 287         annotations:
 288           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
 289       - alert: ipmi voltage alarm
 290         expr: ipmi_voltage_state > 0
 291         for: 5m
 292         labels:
 293           alertgroup: "{{ $labels.instance }}"
 294         annotations:
 295           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 296       - alert: ipmi power alarm
 297         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 298         for: 5m
 299         labels:
 300           alertgroup: "{{ $labels.instance }}"
 301   - name: juniper
 302     rules:
 303       - alert: juniper red alarms
 304         expr: juniper_alarms_red_count > 0
 305         for: 5m
 306         labels:
 307           alertgroup: "{{ $labels.site }}"
 308         annotations:
 309           alarm_count: "{{ $value }} alarms"
 310       - alert: juniper yellow alarms
 311         expr: juniper_alarms_yellow_count > 0
 312         for: 5m
 313         labels:
 314           alertgroup: "{{ $labels.site }}"
 315         annotations:
 316           alarm_count: "{{ $value }} alarms"
 317       - alert: juniper cpu alarm
 318         expr: junos_route_engine_load_average_five / 2 > 0.5
 319         for: 5m
 320         labels:
 321           alertgroup: "{{ $labels.site }}"
 322         annotations:
 323           load_average: "{{ $value | humanizePercentage }}"
 324       - alert: juniper fan alarm
 325         expr: junos_environment_fan_up != 1
 326         for: 5m
 327         labels:
 328           alertgroup: "{{ $labels.site }}"
 329       - alert: juniper power alarm
 330         expr: junos_environment_power_up != 1
 331         for: 5m
 332         labels:
 333           alertgroup: "{{ $labels.site }}"
 334       - alert: juniper laser receive power
 335         expr: junos_interface_diagnostics_laser_rx_dbm < -12 and on (site, instance, name) junos_interface_admin_up == 1
 336         for: 5m
 337         labels:
 338           alertgroup: "{{ $labels.site }}"
 339         annotations:
 340           power: "{{ $value }} dBm"
 341       - alert: juniper laser transmit power
 342         expr: junos_interface_diagnostics_laser_output_dbm < -8 and on (site, instance, name) junos_interface_admin_up == 1
 343         for: 5m
 344         labels:
 345           alertgroup: "{{ $labels.site }}"
 346         annotations:
 347           power: "{{ $value }} dBm"
 348   - name: load
 349     rules:
 350       - alert: load average
 351         expr: sum(node_load5) by (instance) / count(node_cpu_frequency_max_hertz) by (instance) > 2
 352         for: 5m
 353         labels:
 354           alertgroup: "{{ $labels.instance }}"
 355         annotations:
 356           load: "{{ $value | humanizePercentage }}"
 357   - name: mail
 358     rules:
 359       - alert: exim down
 360         expr: exim_up == 0
 361         for: 5m
 362         labels:
 363           alertgroup: "{{ $labels.instance }}"
 364       - alert: exim queue length
 365         expr: exim_queue > ignoring(job) exim_queue_limit
 366         for: 60m
 367         labels:
 368           alertgroup: mail
 369         annotations:
 370           queue_length: "{{ $value }}"
 371       - alert: mailman queue length
 372         expr: mailman_queue_length > 200
 373         for: 60m
 374         labels:
 375           alertgroup: mail
 376         annotations:
 377           queue_length: "{{ $value }}"
 378   - name: mdadm
 379     rules:
 380       - alert: mdadm array inactive
 381         expr: node_md_state{state="inactive"} > 0
 382         for: 0m
 383         labels:
 384           alertgroup: "{{ $labels.instance }}"
 385         annotations:
 386           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 387           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 388           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 389           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 390       - alert: mdadm array degraded
 391         expr: sum (node_md_disks{state="active"}) without (state) < node_md_disks_required
 392         for: 0m
 393         labels:
 394           alertgroup: "{{ $labels.instance }}"
 395         annotations:
 396           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 397           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 398           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 399           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 400       - alert: mdadm disk failed
 401         expr: node_md_disks{state="failed"} > 0
 402         for: 0m
 403         labels:
 404           alertgroup: "{{ $labels.instance }}"
 405         annotations:
 406           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 407           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 408           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 409           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 410   - name: memory
 411     rules:
 412       - alert: low memory
 413         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 414         for: 15m
 415         labels:
 416           alertgroup: "{{ $labels.instance }}"
 417         annotations:
 418           memory_free: "{{ $value | humanizePercentage }}"
 419       - alert: memory pressure
 420         expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.6
 421         for: 60m
 422         labels:
 423           alertgroup: "{{ $labels.instance }}"
 424         annotations:
 425           pressure: "{{ $value | humanizePercentage }}"
 426       - alert: oom kill detected
 427         expr: increase(node_vmstat_oom_kill[1m]) > 0
 428         for: 0m
 429         labels:
 430           alertgroup: "{{ $labels.instance }}"
 431         annotations:
 432           new_oom_kills: "{{ $value }}"
 433   - name: mysql
 434     rules:
 435       - alert: mysql down
 436         expr: mysql_up == 0
 437         for: 1m
 438         labels:
 439           alertgroup: "{{ $labels.instance }}"
 440       - alert: mysql connection limit
 441         expr: mysql_global_status_max_used_connections / mysql_global_variables_max_connections > 0.8
 442         for: 1m
 443         labels:
 444           alertgroup: "{{ $labels.instance }}"
 445         annotations:
 446           connections_used: "{{ $value | humanizePercentage }}"
 447       - alert: mysql connection errors
 448         expr: increase(mysql_global_status_connection_errors_total[1m]) > 0
 449         for: 0m
 450         labels:
 451           alertgroup: "{{ $labels.instance }}"
 452         annotations:
 453           error_count: "{{ $value }}"
 454   - name: network
 455     rules:
 456       - alert: interface redundancy lost
 457         expr: node_bonding_active < 2 and on (instance, master) label_replace(chef_network_interface{bond_mode="802.3ad"}, "master", "$1", "name", "(.*)")
 458         for: 5m
 459         labels:
 460           alertgroup: "{{ $labels.instance }}"
 461         annotations:
 462           link_count: "{{ $value }}"
 463       - alert: interface transmit rate
 464         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.99
 465         for: 5m
 466         labels:
 467           alertgroup: "{{ $labels.instance }}"
 468         annotations:
 469           bandwidth_used: "{{ $value | humanizePercentage }}"
 470       - alert: interface receive rate
 471         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.99
 472         for: 5m
 473         labels:
 474           alertgroup: "{{ $labels.instance }}"
 475         annotations:
 476           bandwidth_used: "{{ $value | humanizePercentage }}"
 477       - alert: interface transmit errors
 478         expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
 479         for: 5m
 480         labels:
 481           alertgroup: "{{ $labels.instance }}"
 482         annotations:
 483           error_rate: "{{ $value | humanizePercentage }}"
 484       - alert: wireguard interface transmit errors
 485         expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
 486         for: 1h
 487         labels:
 488           alertgroup: "{{ $labels.instance }}"
 489         annotations:
 490           error_rate: "{{ $value | humanizePercentage }}"
 491       - alert: interface receive errors
 492         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 493         for: 5m
 494         labels:
 495           alertgroup: "{{ $labels.instance }}"
 496         annotations:
 497           error_rate: "{{ $value | humanizePercentage }}"
 498       - alert: conntrack entries
 499         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 500         for: 5m
 501         labels:
 502           alertgroup: "{{ $labels.instance }}"
 503         annotations:
 504           entries_used: "{{ $value | humanizePercentage }}"
 505   - name: nominatim
 506     rules:
 507       - alert: nominatim replication delay
 508         expr: nominatim_replication_delay > 10800
 509         for: 1h
 510         labels:
 511           alertgroup: nominatim
 512         annotations:
 513           delay: "{{ $value | humanizeDuration }}"
 514       - alert: nominatim connections
 515         expr: sum(nginx_connections_writing and on (instance) chef_role{name="nominatim"}) > 2500
 516         for: 15m
 517         labels:
 518           alertgroup: nominatim
 519   - name: overpass
 520     rules:
 521       - alert: overpass osm database age
 522         expr: overpass_database_age_seconds{database="osm"} > 3600
 523         for: 1h
 524         labels:
 525           alertgroup: overpass
 526         annotations:
 527           age: "{{ $value | humanizeDuration }}"
 528       - alert: overpass area database age
 529         expr: overpass_database_age_seconds{database="area"} > 86400
 530         for: 1h
 531         labels:
 532           alertgroup: overpass
 533         annotations:
 534           age: "{{ $value | humanizeDuration }}"
 535   - name: passenger
 536     rules:
 537       - alert: passenger down
 538         expr: passenger_up == 0
 539         for: 5m
 540         labels:
 541           alertgroup: "{{ $labels.instance }}"
 542       - alert: passenger queuing
 543         expr: passenger_top_level_request_queue > 0
 544         for: 5m
 545         labels:
 546           alertgroup: "{{ $labels.instance }}"
 547       - alert: passenger application queuing
 548         expr: passenger_app_request_queue > 0
 549         for: 5m
 550         labels:
 551           alertgroup: "{{ $labels.instance }}"
 552   - name: planet
 553     rules:
 554       - alert: planet dump overdue
 555         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/(pbf|planet)/.*"} > 7 * 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 556         for: 24h
 557         labels:
 558           alertgroup: planet
 559         annotations:
 560           overdue_by: "{{ $value | humanizeDuration }}"
 561       - alert: notes dump overdue
 562         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/notes/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 563         for: 6h
 564         labels:
 565           alertgroup: planet
 566         annotations:
 567           overdue_by: "{{ $value | humanizeDuration }}"
 568       - alert: daily replication feed delayed
 569         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/day/.*"} > 86400 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 570         for: 3h
 571         labels:
 572           alertgroup: planet
 573         annotations:
 574           delayed_by: "{{ $value | humanizeDuration }}"
 575       - alert: hourly replication feed delayed
 576         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/hour/.*"} > 3600 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 577         for: 30m
 578         labels:
 579           alertgroup: planet
 580         annotations:
 581           delayed_by: "{{ $value | humanizeDuration }}"
 582       - alert: minutely replication feed delayed
 583         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/minute/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 584         for: 5m
 585         labels:
 586           alertgroup: planet
 587         annotations:
 588           delayed_by: "{{ $value | humanizeDuration }}"
 589       - alert: changeset replication feed delayed
 590         expr: time() - file_stat_modif_time_seconds{path=~"/store/planet/replication/changesets/.*"} > 60 and ignoring (job, name, path) chef_role{name="planetdump"} == 1
 591         for: 5m
 592         labels:
 593           alertgroup: planet
 594         annotations:
 595           delayed_by: "{{ $value | humanizeDuration }}"
 596   - name: postgresql
 597     rules:
 598       - alert: postgresql down
 599         expr: pg_up == 0
 600         for: 1m
 601         labels:
 602           alertgroup: "{{ $labels.instance }}"
 603       - alert: postgresql replication delay
 604         expr: pg_replication_lag_seconds > 30
 605         for: 15m
 606         labels:
 607           alertgroup: "{{ $labels.instance }}"
 608         annotations:
 609           delay: "{{ $value | humanizeDuration }}"
 610       - alert: postgresql connection limit
 611         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 612         for: 1m
 613         labels:
 614           alertgroup: "{{ $labels.instance }}"
 615         annotations:
 616           connections_used: "{{ $value | humanizePercentage }}"
 617       - alert: postgresql deadlocks
 618         expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
 619         for: 0m
 620         labels:
 621           alertgroup: "{{ $labels.instance }}"
 622         annotations:
 623           new_deadlocks: "{{ $value }}"
 624       - alert: postgresql idle transactions
 625         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)
 626         for: 5m
 627         labels:
 628           alertgroup: "{{ $labels.instance }}"
 629         annotations:
 630           queries: "{{ $value }}"
 631   - name: prometheus
 632     rules:
 633       - alert: prometheus configuration error
 634         expr: prometheus_config_last_reload_successful == 0
 635         for: 10m
 636         labels:
 637           alertgroup: "prometheus"
 638       - alert: prometheus target missing
 639         expr: up == 0
 640         for: 10m
 641         labels:
 642           alertgroup: "prometheus"
 643       - alert: node exporter text file scrape error
 644         expr: node_textfile_scrape_error > 0
 645         for: 10m
 646         labels:
 647           alertgroup: "prometheus"
 648   - name: raid
 649     rules:
 650       - alert: raid controller battery failed
 651         expr: ohai_controller_info{battery_status="failed"} > 0
 652         for: 5m
 653         labels:
 654           alertgroup: "{{ $labels.instance }}"
 655       - alert: raid controller battery recharging
 656         expr: ohai_controller_info{battery_status="recharging"} > 0
 657         for: 4h
 658         labels:
 659           alertgroup: "{{ $labels.instance }}"
 660       - alert: raid array degraded
 661         expr: ohai_array_info{status="degraded"} > 0
 662         for: 5m
 663         labels:
 664           alertgroup: "{{ $labels.instance }}"
 665       - alert: raid disk failed
 666         expr: ohai_disk_info{status="failed"} > 0
 667         for: 5m
 668         labels:
 669           alertgroup: "{{ $labels.instance }}"
 670   - name: rasdaemon
 671     rules:
 672       - alert: memory controller errors
 673         expr: increase(rasdaemon_mc_events_total[1m]) > 0
 674         for: 0m
 675         labels:
 676           alertgroup: "{{ $labels.instance }}"
 677         annotations:
 678           new_errors: "{{ $value }}"
 679       - alert: pcie aer errors
 680         expr: increase(rasdaemon_aer_events_total[1m]) > 0
 681         for: 0m
 682         labels:
 683           alertgroup: "{{ $labels.instance }}"
 684         annotations:
 685           new_errors: "{{ $value }}"
 686   - name: resolved
 687     rules:
 688       - alert: dnssec validation failures
 689         expr: rate(resolved_dnssec_verdicts_total{result="bogus"}[1m]) > 1
 690         for: 5m
 691         labels:
 692           alertgroup: "{{ $labels.instance }}"
 693   - name: smart
 694     rules:
 695       - alert: smart failure
 696         expr: smart_health_status == 0
 697         for: 60m
 698         labels:
 699           alertgroup: "{{ $labels.instance }}"
 700       - alert: smart ssd wearout approaching
 701         expr: smart_percentage_used / 100 >= 0.8
 702         for: 60m
 703         labels:
 704           alertgroup: "{{ $labels.instance }}"
 705         annotations:
 706           percentage_used: "{{ $value | humanizePercentage }}"
 707   - name: smokeping
 708     rules:
 709       - alert: packet loss
 710         expr: 1 - (rate(smokeping_response_duration_seconds_count[5m]) / rate(smokeping_requests_total[5m])) > 0.02
 711         for: 10m
 712         labels:
 713           alertgroup: smokeping
 714         annotations:
 715           loss_rate: "{{ $value | humanizePercentage }}"
 716   - name: snmp
 717     rules:
 718       - alert: snmp pdus missing
 719         expr: max_over_time(snmp_scrape_pdus_returned[1d]) - snmp_scrape_pdus_returned > 0
 720         for: 15m
 721         labels:
 722           alertgroup: snmp
 723         annotations:
 724           missing_pdus: "{{ $value }}"
 725   - name: ssl
 726     rules:
 727       - alert: ssl certificate probe failed
 728         expr: ssl_probe_success == 0
 729         for: 60m
 730         labels:
 731           alertgroup: ssl
 732       - alert: ssl certificate expiry
 733         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 734         for: 0m
 735         labels:
 736           alertgroup: ssl
 737         annotations:
 738           expires_in: "{{ $value | humanizeDuration }}"
 739       - alert: ssl certificate revoked
 740         expr: ssl_ocsp_response_status == 1
 741         for: 0m
 742         labels:
 743           alertgroup: ssl
 744       - alert: ocsp status unknown
 745         expr: ssl_ocsp_response_status == 1
 746         for: 0m
 747         labels:
 748           alertgroup: ssl
 749   - name: statuscake
 750     rules:
 751       - alert: statuscake uptime check failing
 752         expr: statuscake_paused == 0 and statuscake_up == 0
 753         for: 10m
 754         labels:
 755           alertgroup: statuscake
 756   - name: systemd
 757     rules:
 758       - alert: systemd failed service
 759         expr: node_systemd_unit_state{state="failed",name!="chef-client.service"} == 1
 760         for: 5m
 761         labels:
 762           alertgroup: "{{ $labels.instance }}"
 763       - alert: systemd failed chef client service
 764         expr: sum_over_time(node_systemd_unit_state{state="inactive",name="chef-client.service"}[6h]) == 0
 765         for: 0m
 766         labels:
 767           alertgroup: "{{ $labels.instance }}"
 768   - name: taginfo
 769     rules:
 770       - alert: taginfo planet age
 771         expr: time() - taginfo_data_from_seconds > 129600 and on (instance) chef_role{name="taginfo"}
 772         for: 0m
 773         labels:
 774           alertgroup: taginfo
 775         annotations:
 776           age: "{{ $value | humanizeDuration }}"
 777       - alert: taginfo database age
 778         expr: time() - taginfo_database_update_finish_seconds > 129600 and on (instance) chef_role{name="taginfo"}
 779         for: 0m
 780         labels:
 781           alertgroup: taginfo
 782         annotations:
 783           age: "{{ $value | humanizeDuration }}"
 784       - alert: taginfo database size
 785         expr: abs(delta(taginfo_database_size_bytes[30m])) / taginfo_database_size_bytes > 0.1
 786         for: 30m
 787         labels:
 788           alertgroup: taginfo
 789         annotations:
 790           size_change: "{{ $value | humanizePercentage }}"
 791   - name: tile
 792     rules:
 793       - alert: renderd replication delay
 794         expr: renderd_replication_delay > 120
 795         for: 15m
 796         labels:
 797           alertgroup: tile
 798         annotations:
 799           delay: "{{ $value | humanizeDuration }}"
 800       - alert: missed tile rate
 801         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 802         for: 5m
 803         labels:
 804           alertgroup: tile
 805         annotations:
 806           miss_rate: "{{ $value | humanizePercentage }}"
 807       - alert: tile render rate
 808         expr: sum(rate(renderd_zoom_metatiles_total[5m])) by (instance) == 0
 809         for: 15m
 810         labels:
 811           alertgroup: tile
 812         annotations:
 813           render_rate: "{{ $value }} tiles/s"
 814   - name: time
 815     rules:
 816       - alert: clock not synchronising
 817         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 818         for: 5m
 819         labels:
 820           alertgroup: "{{ $labels.instance }}"
 821       - alert: clock skew detected
 822         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)
 823         for: 5m
 824         labels:
 825           alertgroup: "{{ $labels.instance }}"
 826         annotations:
 827           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"
 828   - name: web
 829     rules:
 830       - alert: web error rate
 831         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 and sum(rate(api_call_count_total{status=~"50[0-8]|5[1-9][0-9]"}[5m])) by (instance) > 0.01
 832         for: 5m
 833         labels:
 834           alertgroup: web
 835         annotations:
 836           error_rate: "{{ $value | humanizePercentage }}"
 837       - alert: job processing rate
 838         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
 839         for: 1h
 840         labels:
 841           alertgroup: web
 842         annotations:
 843           job_processing_rate: "{{ $value | humanizePercentage }}"