alerts.yaml

# used for generate prometheusrule and alertmanagerconfig
- name: mysql-down
  alertType: monitor
  expr: mysql_up{service="__fullname__",namespace="__namespace__"} == 0
  message: MySQL [namespace {{ $labels.namespace }}] [service {{ $labels.service }}] [instance {{ $labels.instance }}] down
  for: 0m
  alertLevels:
    - severity: critical
  receivers:
    - alertChannelID: 1
      interval: 10m
- name: mysql-too-many-connections
  alertType: monitor
  expr: max_over_time(mysql_global_status_threads_connected{service="__fullname__",namespace="__namespace__"}[1m]) / mysql_global_variables_max_connections{service="__fullname__",namespace="__namespace__"} * 100 > 80
  message: MySQL [namespace {{ $labels.namespace }}] [service {{ $labels.service }}] [instance {{ $labels.instance }}] too many connections > 80%
  for: 2m
  alertLevels:
    - severity: error
  receivers:
    - alertChannelID: 1
      interval: 10m
- name: mysql-slow-queries
  alertType: monitor
  expr: increase(mysql_global_status_slow_queries{service="__fullname__",namespace="__namespace__"}[1m]) > 0
  message: MySQL [namespace {{ $labels.namespace }}] [service {{ $labels.service }}] [instance {{ $labels.instance }}] slow queries
  for: 2m
  alertLevels:
    - severity: error
  receivers:
    - alertChannelID: 1
      interval: 10m
- name: mysql-restarted
  alertType: monitor
  expr: mysql_global_status_uptime{service="__fullname__",namespace="__namespace__"} < 60
  message: MySQL [namespace {{ $labels.namespace }}] [service {{ $labels.service }}] [instance {{ $labels.instance }}] restarted
  for: 0m
  alertLevels:
    - severity: error
  receivers:
    - alertChannelID: 1
      interval: 10m