Skip to content

Latest commit

 

History

History
97 lines (66 loc) · 3.62 KB

README.md

File metadata and controls

97 lines (66 loc) · 3.62 KB

Module 2: Install Kubecost on AKS

This section helps you install Kubecost on the cluster you've successfully created in Section 1. Optionally, you'll also expose your Kubecost UI via a Load Balancer.

Prerequisites

  • You have completed Section 1 of this lab
  • You can access your cluster via kubectl
  • You have installed helm3 CLI

Step 1: Install Kubecost

These instructions are also available in kubecost.com/install.

  1. Open the terminal, and create a kubecost namespace in your cluster.

    $ kubectl create namespace kubecost

  2. Add the official Kubecost helm repository.

    $ helm repo add kubecost https://kubecost.github.io/cost-analyzer/

  3. Run the install command. A standard Kubecost installation will also install the Prometheus and Grafana dependencies within the kubecost namespace.

    $ helm install kubecost kubecost/cost-analyzer --namespace kubecost

    This step will take upto a couple of minutes to complete. Once it's done, verify that all the containers have been created and are running without issues:

    $ kubectl get pods -n kubecost

Step 2: View Kubecost data

You can expose the Kubecost UI dashboards locally using Port Forwarding.

  1. Run the port-forward command.

    $ kubectl port-forward --namespace kubecost deployment/kubecost-cost-analyzer 9090

  2. In your browser, open localhost:9090 to see the data. It should look like this:

Kubecost UI screenshot

  1. The Kubecost API depends on its bundled Prometheus installation. Verify that the Kubecost API is in full force. Settings -> Prometheus Status.

Prometheus status screenshot

(optional) Step 3: Expose Kubecost UI deployment outside of the cluster

One of the available options to open your Kubecost deployment to the world is by exposing port 9090 via Azure's Load Balancer.

  1. Create a file called kubecost-cost-analyzer.yml with the following contents. This file is also available within this section's files.

    apiVersion: v1
kind: Service
metadata:
    name: public-svc
spec:
    type: LoadBalancer
    ports:
    - port: 9090
    selector:
        app: cost-analyzer

    The app should be the service name of the Kubecost app minus the kubecost- prefic. You can view your services by running kubectl get services -n kubecost.

  2. Apply to cluster.

    $ kubectl apply -f kubecost-cost-analyzer.yml -n kubecost

  3. Verify that the public-svc service has been created as a Load Balancer, exposing port 9090.

NAME                                TYPE           CLUSTER-IP     EXTERNAL-IP     PORT(S)                      AGE
kubecost-cost-analyzer              ClusterIP      10.0.208.169   <none>          9001/TCP,9003/TCP,9090/TCP   1h
kubecost-grafana                    ClusterIP      10.0.181.47    <none>          80/TCP                       1h
kubecost-kube-state-metrics         ClusterIP      10.0.2.211     <none>          8080/TCP                     1h
kubecost-prometheus-node-exporter   ClusterIP      None           <none>          9100/TCP                     1h
kubecost-prometheus-server          ClusterIP      10.0.228.40    <none>          80/TCP                       1h
public-svc                          LoadBalancer   10.0.104.222   20.85.139.189   9090:32660/TCP               1h

  1. View Kubecost data by visiting the public-svc address: http://<EXTERNAL-IP>:9090

  2. Party time! 💃💃💃