keepalived/meta/prometheus.yml

{%- from "keepalived/map.jinja" import cluster with context %}
{%- if cluster.get('enabled', False) %}
server:
  alert:
    KeepalivedProcessDown:
      if: >-
        procstat_running{process_name="keepalived"} == 0
      {% raw %}
      for: 2m
      labels:
        severity: major
        service: keepalived
      annotations:
        summary: "Keepalived process is down"
        description: "The Keepalived process on the {{ $labels.host }} node is down."
      {% endraw %}
    KeepalivedProcessNotResponsive:
      if: >-
        keepalived_up == 0
      {% raw %}
      labels:
        severity: major
        service: keepalived
      annotations:
        summary: "Keepalived process is not responding"
        description: "The Keepalived process on the {{ $labels.host }} node is not responding."
      {% endraw %}
    KeepalivedFailedState:
      if: >-
        keepalived_state == 0
      {% raw %}
      labels:
        severity: minor
        service: keepalived
      annotations:
        summary: "Keepalived VRRP state is FAILED"
        description: "The Keepalived VRRP {{ $labels.name }} is in the FAILED state on the {{ $labels.host }} node."
      {% endraw %}
    KeepalivedUnknownState:
      if: >-
        keepalived_state == -1
      {% raw %}
      labels:
        severity: minor
        service: keepalived
      annotations:
        summary: "Keepalived VRRP state is UNKNOWN"
        description: "The Keepalived VRRP {{ $labels.name }} is in the UNKNOWN state on the {{ $labels.host }} node."
      {% endraw %}
    KeepalivedMultipleIPAddr:
      if: >-
        count(ipcheck_assigned) by (ip) > 1
      {% raw %}
      for: 2m
      labels:
        severity: major
        service: keepalived
      annotations:
        summary: "Keepalived VIP is assigned more than once"
        description: "The Keepalived {{ $labels.ip }} virtual IP is assigned more than once."
      {% endraw %}
    KeepalivedServiceOutage:
      if: >-
        count(label_replace(procstat_running{process_name="keepalived"}, "cluster", "$1", "host", "([^0-9]+).+")) by (cluster) == count(label_replace(procstat_running{process_name="keepalived"} == 0, "cluster", "$1", "host", "([^0-9]+).+")) by (cluster)
      {% raw %}
      for: 2m
      labels:
        severity: critical
        service: keepalived
      annotations:
        summary: "Keepalived service outage"
        description: "All Keepalived processes within the {{ $labels.cluster }} cluster are down."
      {% endraw %}
{%- endif %}