1.9 KiB

Raw Blame History

Sample App

Containerized application that retrieves metrics server data

How does it work ?

The sample app is a Nodejs app that requests the metrics every second and prints that in the console. The deployment creates a service account, roles and role binding to allows this application communicate with the api, this token is stored by default on /var/run/secrets/kubernetes.io/serviceaccount/token, the application is reading that token and doing the requests on the endpoint /apis/metrics.k8s.io/v1beta1/pods that returns all pod metrics and log it on console.

Structure

sample-app
├── debian
│   └── sample-app.stable_docker_image
├── docker
│   ├── Dockerfile.debian
│   └── src
│       ├── package.json
│       └── sample-application.js
└── readme.md

Important files

src - Contains Nodejs application
Dockerfile - Application Dockerfile

Run application

Deploy tha sample app using system helm-override-update --reuse-values --set atribute=value <app name> <chart name> <namespace> follow the steps below:

Run system helm-override-update --reuse-values --set sampleApp.create=true metrics-server metrics-server metrics-server
Run system application-apply metrics-server to apply the override
Run kubectl get pods -n metrics-server to get the name of the pod.
Run kubectl logs -n metrics-server <pod-name> --tail 1 -f to see the logs and check if the sample application is requesting successfully the metrics server api

Endpoints

All of the following endpoints are GET endpoints and they are under the base path /apis/metrics.k8s.io/v1beta1:

/nodes - all node metrics
/nodes/{node} - metrics for a specified node
/namespaces/{namespace}/pods - all pod metrics within namespace with support for all-namespaces
/namespaces/{namespace}/pods/{pod} - metrics for a specified pod
/pods - all pod metrics of all namespaces

1.9 KiB Raw Blame History