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

   1 # DO NOT EDIT - This file is being maintained by Chef
   2
   3 groups:
   4   - name: alertmanager
   5     rules:
   6       - alert: prometheus target missing
   7         expr: up == 0
   8         for: 5m
   9         labels:
  10           alertgroup: "prometheus"
  11   - name: apache
  12     rules:
  13       - alert: apache down
  14         expr: apache_up == 0
  15         for: 5m
  16         labels:
  17           alertgroup: "{{ $labels.instance }}"
  18       - alert: apache workers busy
  19         expr: sum(apache_workers{state="busy"}) by (instance) / sum(apache_scoreboard) by (instance) > 0.8
  20         for: 5m
  21         labels:
  22           alertgroup: "{{ $labels.instance }}"
  23         annotations:
  24           busy_workers: "{{ $value | humanizePercentage }}"
  25   - name: database
  26     rules:
  27       - alert: postgres replication delay
  28         expr: pg_replication_lag_seconds > 5
  29         for: 5m
  30         labels:
  31           alertgroup: database
  32         annotations:
  33           delay: "{{ $value | humanizeDuration }}"
  34   - name: filesystem
  35     rules:
  36       - alert: filesystem low on space
  37         expr: node_filesystem_avail_bytes / node_filesystem_size_bytes < 0.05
  38         for: 5m
  39         labels:
  40           alertgroup: "{{ $labels.instance }}"
  41         annotations:
  42           percentage_free: "{{ $value | humanizePercentage }}"
  43           free_bytes: "{{ with printf \"node_filesystem_avail_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
  44           total_total: "{{ with printf \"node_filesystem_size_bytes{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value | humanize1024 }}bytes{{end}}"
  45       - alert: filesystem low on inodes
  46         expr: node_filesystem_files_free / node_filesystem_files < 0.1
  47         for: 5m
  48         labels:
  49           alertgroup: "{{ $labels.instance }}"
  50         annotations:
  51           percentage_free: "{{ $value | humanizePercentage }}"
  52           free_inodes: "{{ with printf \"node_filesystem_files_free{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
  53           total_inodes: "{{ with printf \"node_filesystem_files{instance='%s',mountpoint='%s'}\" $labels.instance $labels.mountpoint | query }}{{ . | first | value }}{{end}}"
  54   - name: hwmon
  55     rules:
  56       - alert: hwmon fan alarm
  57         expr: node_hwmon_fan_alarm == 1
  58         for: 5m
  59         labels:
  60           alertgroup: "{{ $labels.instance }}"
  61         annotations:
  62           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}}"
  63           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}}"
  64       - alert: hwmon temperature alarm
  65         expr: node_hwmon_temp_alarm == 1
  66         for: 5m
  67         labels:
  68           alertgroup: "{{ $labels.instance }}"
  69         annotations:
  70           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}}"
  71           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}}"
  72           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}}"
  73       - alert: hwmon voltage alarm
  74         expr: node_hwmon_in_alarm == 1
  75         for: 5m
  76         labels:
  77           alertgroup: "{{ $labels.instance }}"
  78         annotations:
  79           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}}"
  80           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}}"
  81           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}}"
  82   - name: ipmi
  83     rules:
  84       - alert: ipmi fan alarm
  85         expr: ipmi_fan_speed_state > 0
  86         for: 5m
  87         labels:
  88           alertgroup: "{{ $labels.instance }}"
  89         annotations:
  90           fan_speed_rpm: "{{ with printf \"ipmi_fan_speed_rpm{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}rpm{{end}}"
  91       - alert: ipmi temperature alarm
  92         expr: ipmi_temperature_state > 0
  93         for: 5m
  94         labels:
  95           alertgroup: "{{ $labels.instance }}"
  96         annotations:
  97           temperature_celsius: "{{ with printf \"ipmi_temperature_celsius{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}C{{end}}"
  98       - alert: ipmi voltage alarm
  99         expr: ipmi_voltage_state > 0
 100         for: 5m
 101         labels:
 102           alertgroup: "{{ $labels.instance }}"
 103         annotations:
 104           voltage_volts: "{{ with printf \"ipmi_voltage_volts{instance='%s',id='%s'}\" $labels.instance $labels.id | query }}{{ . | first | value | humanize }}V{{end}}"
 105       - alert: ipmi power alarm
 106         expr: ipmi_power_state > 0 or ipmi_sensor_state{type=~"Power .*"} > 0
 107         for: 5m
 108         labels:
 109           alertgroup: "{{ $labels.instance }}"
 110   - name: mdadm
 111     rules:
 112       - alert: mdadm array inactive
 113         expr: node_md_state{state="inactive"} > 0
 114         for: 0m
 115         labels:
 116           alertgroup: "{{ $labels.instance }}"
 117         annotations:
 118           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 119           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 120           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 121           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 122       - alert: mdadm disk failed
 123         expr: node_md_disks{state="failed"} > 0
 124         for: 0m
 125         labels:
 126           alertgroup: "{{ $labels.instance }}"
 127         annotations:
 128           required: "{{ with printf \"node_md_disks_required{instance='%s',device='%s'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 129           active: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='active'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 130           failed: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='failed'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 131           spare: "{{ with printf \"node_md_disks{instance='%s',device='%s',state='spare'}\" $labels.instance $labels.device | query }}{{ . | first | value | humanize }} disks{{end}}"
 132   - name: memory
 133     rules:
 134       - alert: low memory
 135         expr: node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes < 0.1
 136         for: 5m
 137         labels:
 138           alertgroup: "{{ $labels.instance }}"
 139         annotations:
 140           memory_free: "{{ $value | humanizePercentage }}"
 141       - alert: memory pressure
 142         expr: rate(node_vmstat_pgmajfault[1m]) > 1000
 143         for: 5m
 144         labels:
 145           alertgroup: "{{ $labels.instance }}"
 146         annotations:
 147           major_page_faults: "{{ $value }} faults/s"
 148       - alert: oom kill detected
 149         expr: increase(node_vmstat_oom_kill[1m]) > 0
 150         for: 0m
 151         labels:
 152           alertgroup: "{{ $labels.instance }}"
 153         annotations:
 154           new_oom_kills: "{{ $value }}"
 155   - name: network
 156     rules:
 157       - alert: interface transmit rate
 158         expr: rate(node_network_transmit_bytes_total[1m]) / node_network_speed_bytes > 0.98
 159         for: 5m
 160         labels:
 161           alertgroup: "{{ $labels.instance }}"
 162         annotations:
 163           bandwidth_used: "{{ $value | humanizePercentage }}"
 164       - alert: interface receive rate
 165         expr: rate(node_network_receive_bytes_total[1m]) / node_network_speed_bytes > 0.98
 166         for: 5m
 167         labels:
 168           alertgroup: "{{ $labels.instance }}"
 169         annotations:
 170           bandwidth_used: "{{ $value | humanizePercentage }}"
 171       - alert: interface transmit errors
 172         expr: rate(node_network_transmit_errs_total[1m]) / rate(node_network_transmit_packets_total[1m]) > 0.01
 173         for: 5m
 174         labels:
 175           alertgroup: "{{ $labels.instance }}"
 176         annotations:
 177           error_rate: "{{ $value | humanizePercentage }}"
 178       - alert: interface receive errors
 179         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
 180         for: 5m
 181         labels:
 182           alertgroup: "{{ $labels.instance }}"
 183         annotations:
 184           error_rate: "{{ $value | humanizePercentage }}"
 185       - alert: conntrack entries
 186         expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
 187         for: 5m
 188         labels:
 189           alertgroup: "{{ $labels.instance }}"
 190         annotations:
 191           entries_used: "{{ $value | humanizePercentage }}"
 192   - name: postgresql
 193     rules:
 194       - alert: postgresql down
 195         expr: pg_up == 0
 196         for: 1m
 197         labels:
 198           alertgroup: "{{ $labels.instance }}"
 199       - alert: postgresql replication delay
 200         expr: pg_replication_lag_seconds > 5
 201         for: 1m
 202         labels:
 203           alertgroup: "{{ $labels.instance }}"
 204         annotations:
 205           delay: "{{ $value | humanizeDuration }}"
 206       - alert: postgresql connection limit
 207         expr: sum (pg_stat_activity_count) by (instance, server) / sum (pg_settings_max_connections) by (instance, server) > 0.8
 208         for: 1m
 209         labels:
 210           alertgroup: "{{ $labels.instance }}"
 211         annotations:
 212           connections_used: "{{ $value | humanizePercentage }}"
 213       - alert: postgresql deadlocks
 214         expr: increase(pg_stat_database_deadlocks[1m]) > 5
 215         for: 0m
 216         labels:
 217           alertgroup: "{{ $labels.instance }}"
 218         annotations:
 219           new_deadlocks: "{{ $value }}"
 220       - alert: postgresql slow queries
 221         expr: pg_slow_queries > 0
 222         for: 5m
 223         labels:
 224           alertgroup: "{{ $labels.instance }}"
 225         annotations:
 226           queries: "{{ $value }}"
 227   - name: smart
 228     rules:
 229       - alert: smart failure
 230         expr: smart_health_status == 0
 231         for: 60m
 232         labels:
 233           alertgroup: "{{ $labels.instance }}"
 234       - alert: smart ssd wearout approaching
 235         expr: smart_percentage_used >= 90
 236         for: 60m
 237         labels:
 238           alertgroup: "{{ $labels.instance }}"
 239         annotations:
 240           percentage_used: "{{ $value | humanizePercentage }}"
 241   - name: ssl
 242     rules:
 243       - alert: ssl certificate probe failed
 244         expr: ssl_probe_success == 0
 245         for: 60m
 246         labels:
 247           alertgroup: ssl
 248       - alert: ssl certificate expiry
 249         expr: ssl_verified_cert_not_after{chain_no="0"} - time() < 86400 * 14
 250         for: 0m
 251         labels:
 252           alertgroup: ssl
 253         annotations:
 254           expires_in: "{{ $value | humanizeDuration }}"
 255       - alert: ssl certificate revoked
 256         expr: ssl_ocsp_response_status == 1
 257         for: 0m
 258         labels:
 259           alertgroup: ssl
 260       - alert: ocsp status unknown
 261         expr: ssl_ocsp_response_status == 1
 262         for: 0m
 263         labels:
 264           alertgroup: ssl
 265   - name: tile
 266     rules:
 267       - alert: renderd replication delay
 268         expr: renderd_replication_delay > 120
 269         for: 5m
 270         labels:
 271           alertgroup: tile
 272         annotations:
 273           delay: "{{ $value | humanizeDuration }}"
 274       - alert: missed tile rate
 275         expr: sum(rate(modtile_http_response_total{code="404"}[5m])) by (instance) / sum(rate(modtile_http_response_total[5m])) by (instance) > 0.05
 276         for: 5m
 277         labels:
 278           alertgroup: tile
 279         annotations:
 280           miss_rate: "{{ $value | humanizePercentage }}"
 281   - name: time
 282     rules:
 283       - alert: clock not synchronising
 284         expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
 285         for: 5m
 286         labels:
 287           alertgroup: "{{ $labels.instance }}"
 288       - alert: clock skew detected
 289         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)
 290         for: 5m
 291         labels:
 292           alertgroup: "{{ $labels.instance }}"
 293         annotations:
 294           skew: "{{ with printf \"node_timex_offset_seconds{instance='%s'}\" $labels.instance | query }} {{ . | humanizeDuration }}{{ end }}"