#
# Uptime
#

- id: "41x0300aaaaa9982"
  name: "[Host] Непредвиденная перезагрузка хоста {{.Host}}"
  description: "Непредвиденная перезагрузка хоста {{.Host}}"
  query: "node_time_seconds - node_boot_time_seconds"
  labels:
    integration: general
  object_tag: "hostname"
  aggregator: sum
  group_by:
    - "hostname"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "No data"
  rules:
    - id: "42x0300aaaaa9982"
      labels_required:
      labels_absent:
      type: "threshold_1level_sym"
      parameters:
        level: "600"
        critical: "true"
        reverse: "true"
      priority: 1


#
# CPU Utilization
#


- id: "41x0300aaaaa9991"
  name: "[Host] Высокая нагрузка CPU хоста {{.Host}} ({{.Value0}}%)"
  description: "Высокая нагрузка CPU хоста {{.Host}} ({{.Value0}}%)"
  query: '100 - avg by (hostname, instance, job, group) (irate(node_cpu_seconds_total{mode="idle"}[5m]) * 100)'
  labels:
    integration: general
  aggregator: sum
  object_tag: "hostname"
  group_by:
    - "hostname"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "No data"
  rules:
    - id: "42x0300aaaaa9991"
      labels_required:
      labels_absent:
      type: "threshold_2levels_sym"
      parameters:
          warning: "60"
          critical: "85"
          reverse: "false"
      priority: 1

- id: "41x0300aaaaa9992"
  name: "[Host] Высокая загрузка CPU ядром на хосте {{.Host}} ({{.Value2}}%)"
  description: "Высокая загрузка CPU ядром на хосте {{.Host}} ({{.Value2}}%)"
  query: 'avg by (hostname, instance, job, group) (irate(node_cpu_seconds_total{mode="system"}[5m]) * 100)'
  labels:
    integration: general
  aggregator: sum
  object_tag: "hostname"
  group_by:
    - "hostname"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "No data"
  rules:
    - id: "42x0300aaaaa9992"
      labels_required:
      labels_absent:
      type: "threshold_1level_sym"
      parameters:
        level: "50"
        critical: "true"
        reverse: "false"
      priority: 1

- id: "41x0300aaaaa9993"
  name: "[Host] Высокий IOWAIT на хосте {{.Host}} ({{.Value2}}%)"
  description: "Высокий IOWAIT на хосте {{.Host}} ({{.Value2}}%)"
  query: '(avg by (instance, hostname, job, group) (rate(node_cpu_seconds_total{mode="iowait"}[5m])) * 100) * on(instance, hostname, job, group) group_left (nodename) node_uname_info{nodename=~".+"}'
  labels:
    integration: general
  aggregator: sum
  object_tag: "hostname"
  group_by:
    - "hostname"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "No data"
  rules:
    - id: "42x0300aaaaa9993"
      labels_required:
      labels_absent:
      type: "threshold_2levels_sym"
      parameters:
        warning: "10"
        critical: "30"
        reverse: "false"
      priority: 1

#
# Memory utilization
#

- id: "41x0300aaaaa9994"
  name: "[Host] Высокая утилизация RAM на хосте {{.Host}} ({{.Value2}}%)"
  description: "Высокая утилизация RAM на хосте {{.Host}} ({{.Value2}}%)"
  query: '(100 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes * 100))'
  labels:
    integration: general
  aggregator: sum
  object_tag: "hostname"
  group_by:
    - "hostname"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "No data"
  rules:
    - id: "42x0300aaaaa9994"
      labels_required:
      labels_absent:
      type: "threshold_2levels_sym"
      parameters:
        warning: "80"
        critical: "95"
        reverse: "false"
      priority: 1

#
# Filesystem
#

- id: "41x0300aaaaa9995"
  name: '[Host] Мало места на диске {{index .Labels "mountpoint"}} на хосте {{.Host}} свободно {{.Value0}}%'
  description: 'Мало места на диске {{index .Labels "mountpoint"}} на хосте {{.Host}} свободно {{.Value0}}%'
# этот запрос выводит информацию по файлам, которые примонтированы с хост машины считаются файловыми системами, и их отображение "зашумляет" вывод
#  query: '((node_filesystem_avail_bytes{device!~"rootfs"} * 100) / node_filesystem_size_bytes{device!~"rootfs"} and ON (instance, device, mountpoint) node_filesystem_readonly == 0) * on(instance, hostname, job, group) group_left (nodename) node_uname_info'
# этот запрос фильтрует часть файловых систем, примонтированных с хост машины
  query: '((node_filesystem_avail_bytes{device!~"rootfs|tmpfs"} * 100) / node_filesystem_size_bytes{device!~"rootfs|tmpfs"} and ON (instance, device, mountpoint) node_filesystem_readonly == 0) * on(instance, hostname, job, group) group_left (nodename) node_uname_info'
  labels:
    integration: general
  aggregator: sum
  object_tag: "hostname"
  group_by:
    - "hostname"
    - "mountpoint"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "No data"
  rules:
    - id: "42x0300aaaaa9995"
      labels_required:
      labels_absent:
      type: "threshold_2levels_sym"
      parameters:
        warning: "20"
        critical: "10"
        reverse: "true"
      priority: 1

- id: "41x0300aaaaa9996"
  name: '[Host] Мало свободных inodes на хосте {{.Host}} ({{.Value2}}%)'
  description: 'Мало свободных inodes на хосте {{.Host}} ({{.Value2}}%)'
  query: '(node_filesystem_files_free{fstype!="msdosfs",fstype!="tmpfs",mountpoint!~"/etc/hostname",mountpoint!~"/etc/resolv.conf",mountpoint!~"/etc/hosts"} / node_filesystem_files{fstype!="msdosfs",fstype!="tmpfs",mountpoint!~"/etc/hostname",mountpoint!~"/etc/resolv.conf",mountpoint!~"/etc/hosts"} * 100 and ON (instance, device, mountpoint) node_filesystem_readonly == 0) * on(instance, hostname, job, group) group_left (nodename) node_uname_info{nodename=~".+"}'
  labels:
    integration: general
  aggregator: sum
  object_tag: "hostname"
  group_by:
    - "hostname"
    - "mountpoint"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "No data"
  rules:
    - id: "42x0300aaaaa9996"
      labels_required:
      labels_absent:
      type: "threshold_1level_sym"
      parameters:
        level: "10"
        critical: "true"
        reverse: "true"
      priority: 1

- id: "41x0300aaaaa9997"
  name: '[Host] Ошибка устройства файловой системы хоста {{.Host}} на устройстве {{index .Labels "device"}}'
  description: 'Ошибка устройства файловой системы хоста {{.Host}} на устройстве {{index .Labels "device"}}'
  query: 'node_filesystem_device_error{fstype!~"fuse.xrdp-chansrv|parsecfs|tmpfs|fuse.*"}'
  labels:
    integration: general
  aggregator: sum
  object_tag: "hostname"
  group_by:
    - "hostname"
    - "mountpoint"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "No data"
  rules:
    - id: "42x0300aaaaa9997"
      labels_required:
      labels_absent:
      type: "threshold_1level_sym"
      parameters:
        level: "1"
        critical: "true"
        reverse: "false"
      priority: 1

#
# Network
#

# Из-за особенностей выражения этот монитор не будет возвращать группы метрик в состоянии OK,
# а только в состоянии critical. Если метрика перешла в состояние NO_DATA, то проблема закрывается
# так как сетевой интерфейс вышел из состояния operstate="down", то есть - включился

- id: "41x0300aaaaa9998"
  name: '[Host] Отключился сетевой интерфейс {{index .Labels "device"}} на хосте {{.Host}}'
  description: 'Отключился сетевой интерфейс {{index .Labels "device"}} на хосте {{.Host}}'
  query: 'sum by(hostname, device) ( node_network_info{device!~"^(eno[0-9]+|docker[0-9]*|veth[0-9a-f]+|br-[0-9a-f]+|lo|tun[0-9]*|cali[0-9a-f]+|flannel\\.\\d+)$", operstate="up"} * 0) or sum by(hostname, device) (node_network_info{device!~"^(eno[0-9]+|docker[0-9]*|veth[0-9a-f]+|br-[0-9a-f]+|lo|tun[0-9]*|cali[0-9a-f]+|flannel\\.\\d+)$", operstate="down"} * 1)'
  object_tag: "hostname"
  labels:
    integration: general
  aggregator: sum
  group_by:
    - "device"
    - "hostname"
  step: 5m
  rate: 60s
  metric_ttl: 30m
  no_data_mode: "close problem"
  rules:
    - id: "42x0300aaaaa9998"
      labels_required:
      labels_absent:
      type: "threshold_1level_sym"
      parameters:
        level: "0.5"
        critical: "false"
        reverse: "false"
      priority: 1