cluster

Infrastructure files for Nordgedanken and Midnightthoughts.
git clone git://archive.git.mtrnord.blog/MTRNord/cluster.git
Log | Files | Refs | README

custom-rules.yaml (1576B)

      1 apiVersion: monitoring.coreos.com/v1
      2 kind: PrometheusRule
      3 metadata:
      4     name: custom-node-rules
      5     namespace: monitoring
      6     labels:
      7         release: prometheus-stack
      8 spec:
      9     groups:
     10         - name: custom.node.rules
     11           rules:
     12               # NodeMemoryMajorPagesFaults re-added with control-plane nodes excluded.
     13               # Control plane nodes are intentionally undersized and expected to have
     14               # occasional major page faults; alerting on them causes noise.
     15               - alert: NodeMemoryMajorPagesFaults
     16                 expr: |
     17                     (
     18                       rate(node_vmstat_pgmajfault{job="node-exporter"}[5m]) > 500
     19                     )
     20                     unless on(instance)
     21                     label_replace(
     22                       kube_node_status_addresses{type="InternalIP"} * on(node) group_left() kube_node_role{role="control-plane"},
     23                       "instance", "$1:9100", "address", "(.*)"
     24                     )
     25                 for: 15m
     26                 labels:
     27                     severity: warning
     28                 annotations:
     29                     description: 'Memory major pages are occurring at very high rate at {{ $labels.instance }}, 500 major page faults per second for the last 15 minutes, is currently at {{ $value | humanize }}. Please check that there is enough memory available at this instance.'
     30                     runbook_url: https://runbooks.prometheus-operator.dev/runbooks/node/nodememorymajorpagesfaults
     31                     summary: Memory major page faults are occurring at very high rate.