For example, the arp collector described above exposes a metric called **node_arp_entries** that contains the number of ARP entries in the ARP table for each network interface on a node.Prometheus, Grafana, and Node Exporters are commonly used together in Kubernetes to monitor system-level application insights. In Prometheus, you will see that the majority of metrics exposed by the Node Exporter are prefixed with **node_**. For example, the arp collector exposes the metrics available in /proc/net/arp on Linux.

If you refer back to the Node Exporter documenation you can see the method that each of these collectors uses to acquire metrics. Time = '' level =info msg = ' - zfs' source = 'node_exporter.go:104' Time = '' level =info msg = ' - xfs' source = 'node_exporter.go:104' Time = '' level =info msg = ' - vmstat' source = 'node_exporter.go:104' Time = '' level =info msg = ' - uname' source = 'node_exporter.go:104' Time = '' level =info msg = ' - timex' source = 'node_exporter.go:104' Time = '' level =info msg = ' - time' source = 'node_exporter.go:104' Time = '' level =info msg = ' - textfile' source = 'node_exporter.go:104' Time = '' level =info msg = ' - stat' source = 'node_exporter.go:104' Time = '' level =info msg = ' - sockstat' source = 'node_exporter.go:104' Time = '' level =info msg = ' - pressure' source = 'node_exporter.go:104' Time = '' level =info msg = ' - nfsd' source = 'node_exporter.go:104' Time = '' level =info msg = ' - nfs' source = 'node_exporter.go:104' Time = '' level =info msg = ' - netstat' source = 'node_exporter.go:104' Time = '' level =info msg = ' - netdev' source = 'node_exporter.go:104' Time = '' level =info msg = ' - netclass' source = 'node_exporter.go:104' Time = '' level =info msg = ' - meminfo' source = 'node_exporter.go:104'

Time = '' level =info msg = ' - mdadm' source = 'node_exporter.go:104' Time = '' level =info msg = ' - loadavg' source = 'node_exporter.go:104' Time = '' level =info msg = ' - ipvs' source = 'node_exporter.go:104' Time = '' level =info msg = ' - infiniband' source = 'node_exporter.go:104' Time = '' level =info msg = ' - hwmon' source = 'node_exporter.go:104' Time = '' level =info msg = ' - filesystem' source = 'node_exporter.go:104' Time = '' level =info msg = ' - filefd' source = 'node_exporter.go:104' Time = '' level =info msg = ' - entropy' source = 'node_exporter.go:104' Time = '' level =info msg = ' - edac' source = 'node_exporter.go:104' Time = '' level =info msg = ' - diskstats' source = 'node_exporter.go:104' Time = '' level =info msg = ' - cpufreq' source = 'node_exporter.go:104' Time = '' level =info msg = ' - cpu' source = 'node_exporter.go:104' Time = '' level =info msg = ' - conntrack' source = 'node_exporter.go:104' Time = '' level =info msg = ' - bonding' source = 'node_exporter.go:104' Time = '' level =info msg = ' - bcache' source = 'node_exporter.go:104' Time = '' level =info msg = ' - arp' source = 'node_exporter.go:104' Time = '' level =info msg = 'Enabled collectors:' source = 'node_exporter.go:97' $kubectl logs node-exporter-c8cwp -namespace prometheus If you check the log output from a Node Exporter Pod using kubectl logs you can see the collectors that are active: Node Exporter has numerous collectors designed to gather OS and hardware metrics from various sources on a node. After a brief moment you can then check the configured Targets in Prometheus and you will see that node-exporter is now being successfully scraped. You can then use kubectl get pods -namespace prometheus to see the node-exporter Pod(s) being created by Kubernetes. Go ahead and install Node Exporter into your Kubernetes cluster by executing kubectl apply -f node-exporter.yaml. It includes a Kubernetes Service and ServiceMonitor to scrape metrics from all instances of Node Exporter. The above YAML will create a DaemonSet that launches the Node Exporter on each node in the Kubernetes cluster. honorLabels: true interval: 30s path: /metrics targetPort: 9100 jobLabel: node-exporter namespaceSelector: name: node-exporter port: 9100 protocol: TCP targetPort: 9100 selector:Īpp: node-exporter sessionAffinity: None type: ClusterIPĪpiVersion: /v1 kind: ServiceMonitor metadata:Īpp: node-exporter serviceMonitorSelector: prometheus name: node-exporter namespace: prometheus spec: effect: NoExecute operator: Exists volumes: mountPath: /host/sys name: sys readOnly: true hostNetwork: true hostPID: true restartPolicy: Always tolerations: mountPath: /host/proc name: proc readOnly: true

containerPort: 9100 hostPort: 9100 name: metrics protocol: TCP resources: path.sysfs=/host/sys image: quay.io/prometheus/node-exporter:v0.18.1 imagePullPolicy: IfNotPresent name: node-exporter ports:

ApiVersion: extensions/v1beta1 kind: DaemonSet metadata:Īpp: node-exporter name: node-exporter namespace: prometheus spec:Ĭ/safe-to-evict: 'true' labels:

Prometheus node exporter