After you have installed influxDB, Heapster, and Grafana, let's learn how to get the status of the resource. First, you may use kubectl top now. Check the utilization of nodes and Pods, as well as verifying the functionality of monitoring:
// check the status of nodes
$ kubectl top node
NAME CPU(cores) CPU% MEMORY(bytes) MEMORY%
minikube 236m 11% 672Mi 35%
// check the status of Pods in Namespace kube-system
$ kubectl top pod -n kube-system
NAME CPU(cores) MEMORY(bytes)
heapster-5c448886d-k9wt7 1m 18Mi
kube-addon-manager-minikube 36m 32Mi
kube-dns-54cccfbdf8-j65x6 3m 22Mi
kubernetes-dashboard-77d8b98585-z8hch 0m 12Mi
monitoring-grafana-65757b9656-8cl6d 0m 13Mi
monitoring-influxdb-66946c9f58-hwv8g 1m 26Mi
Currently, kubectl top only covers nodes and Pods, and just shows their CPU and RAM usage.
According to the output of kubectl top, what does the m mean in terms of the quantity of CPU usage? It means "milli", as in millisecond and millimeter. Millicpu is regarded as 10-3 CPU. For example, if the Heapster Pod uses 1 m CPU, it only takes 0.1% CPU computation power at this moment.