Fractal Techware

Alert runbooks /

NodeExporterDown

Host or node_exporter is down. Free rule (MIT): the complete runbook and rule definition are below.

   
Severity critical
Pending (for:) 5m (critical)
Domain Hosts (node_exporter)
Requires node_exporter 1.x (default collectors; systemd and hwmon noted per alert)
Rule file rules/node-exporter.rules.yml (group ftw.node-exporter.alerts)

Meaning

Prometheus cannot scrape node_exporter on the host. Either the host is down, the exporter stopped, or the network path between Prometheus and the host is broken.

Impact

All host alerts for this instance are blind. If the host itself is down, everything running on it is unavailable.

Diagnosis

Mitigation

Escalation

Owned by whoever operates the host (platform team for Kubernetes nodes, service owner for standalone VMs).

Rule definition

From rules/node-exporter.rules.yml in the free repository (MIT). Unit tests for it are in tests/.

- alert: NodeExporterDown
  expr: up{job=~"node|node-exporter|node_exporter|prometheus-node-exporter"} == 0
  for: 5m
  labels:
    severity: critical
  annotations:
    summary: Host or node_exporter is down.
    description: node_exporter target {{ $labels.instance }} (job {{ $labels.job }}) has been unreachable for 5m.
    runbook_url: runbooks/node-exporter/NodeExporterDown.md