Metrics-server Addon
Introduction
The Kubernetes cluster Metrics Server addon is an aggregator of resource usage data in your cluster.
Description
It is an open-source, scalable, and effective cluster addon that provides a means of collecting resource data (like memory and CPU usage) from the kubelet on each worker node. Additionally, it provides data for the Kubernetes Horizontal Pod Autoscaler when used for auto-scaling workloads. Once obtained, that info is exposed inside the Kubernetes API server via the Metrics API. It is an integral part of the core metrics pipeline.
Note: The Metrics Server addon is not meant for non-autoscaling purposes like Grafana or Prometheus monitoring.
Settings
To access and modify the settings, go to the Infrastructure page, check the radio button next to the cluster you wish to modify, then click edit.

Once the new page is open, scroll down the where the cluster addon can be enabled.

CLI
Once installed and activated, the Metrics API can then be accessed using the kubectl top
command, making for easier debugging of various issues.
Syntax
kubectl top [flags] [options]
All Available Commands and Flags
worker1.ubuntu@ip-10-0-1-140:~$ kubectl top
Display Resource (CPU/Memory) usage.
The top command allows you to see the resource consumption for nodes or pods.
This command requires Metrics Server to be correctly configured and working on
the server.
Available Commands:
node Display Resource (CPU/Memory) usage of nodes
pod Display Resource (CPU/Memory) usage of pods
Usage:
kubectl top [flags] [options]
Use "kubectl <command> --help" for more information about a given command.
Use "kubectl options" for a list of global command-line options (applies to all
commands).
Example
kubectl top pod --namespace web-app
NAME CPU (CORES) MEMORY (BYTES)
nginx-84ac2948db-12bce 14m 1Mi
nginx-84ac2948db-bc9ae 14m 1Mi
YAML Specifications
apiVersion sunpike.platform9.com/v1alpha2
kind ClusterAddon
metadata
labels
sunpike.pf9.io/cluster'<clsuuid>'
type metrics-server
name <clsuuid>-metrics-server
namespace default
spec
clusterID'<clsuuid>'
override
type metrics-server
version0.5.0
watchtrue
Conclusion
Addons are updated along with the pf9-kube
version and can be referenced in the Support Matrix article. For additional assistance, contact our support team.