prom-alerts

alert-rules.yaml

apiVersion: alerting.cloud.weave.works/v1
kind: PrometheusRule
metadata:
  name: node-rules
spec:
  groups:
  - name: node-alerts
    rules:
    - alert: NodeDown
      expr: up{job="kubernetes-nodes"} == 0
      for: 5m
      labels:
        severity: critical
      annotations:
        summary: "Kubernetes node is not available"
        impact: "The cluster will be operating at a reduced capacity for scheduling workloads."
        detail: "Node: {{$labels.instance}}"
  - name: clock-alerts
    rules:
    - alert: ClockSkewiff
      expr: abs(node_ntp_drift_seconds) > 15
      for: 5m
      labels:
        severity: warning
      annotations:
        summary: "Clock is out of sync with NTP."
        impact: "Random things are about to break for our users"
        detail: "Node: {{$labels.node}} is {{$value}} seconds off"
    - alert: ClockSyncBroken
      expr: node_timex_sync_status != 1
      for: 5m
      labels:
        severity: warning
      annotations:
        summary: "The clock is not being synced."
        impact: "Random things are about to break for our users"
        detail: "Node: {{$labels.node}}"