Skip to main content
In this lesson, you will learn how to monitor your Kubernetes cluster by inspecting deployed Pods and deploying a metrics server to track resource usage.

​
Inspecting Deployed Pods

Begin by verifying the running Pods in your cluster. Execute the following command: 
kubectl get pods
You should see output similar to the following: 
NAME       READY   STATUS              RESTARTS   AGE
elephant   1/1     Running             0          20s
lion       1/1     Running             0          20s
rabbit     0/1     ContainerCreating   0          20s
At this point, the “elephant” and “lion” Pods are running, while “rabbit” is still initializing.

​
Deploying the Metrics Server

To monitor resource usage, you’ll deploy the metrics server. This lab uses a preconfigured Git repository containing all the necessary configurations.
Do not use these configurations in a production environment. Always refer to the official documentation before deploying to production.

​
Step 1: Clone the Repository

Clone the repository by running: 
git clone https://github.com/kodekloudhub/kubernetes-metrics-server.git
After cloning, you should see output similar to this: 
Cloning into 'kubernetes-metrics-server'...
remote: Enumerating objects: 24, done.
remote: Counting objects: 100% (12/12), done.
remote: Compressing objects: 100% (12/12), done.
remote: Total 24 (delta 4), reused 0 (delta 0), pack-reused 12
Unpacking objects: 100% (24/24), done.

​
Step 2: Review the Repository Contents

First, check the directory structure: 
ls
You should see a directory named kubernetes-metrics-server and possibly a file like sample.yaml. Next, navigate into the repository directory and list its files: 
cd kubernetes-metrics-server/
ls
You will find key configuration files such as:
  • README.md
  • aggregated-metrics-reader.yaml
  • auth-delegator.yaml
  • metrics-server-deployment.yaml
  • metrics-server-service.yaml
  • resource-reader.yaml (if present)

​
Step 3: Deploy the Metrics Server

Deploy the metrics server by creating all required objects with the following command: 
kubectl create -f .
You should see output indicating that roles, role bindings, deployments, and services have been created successfully: 
clusterrole.rbac.authorization.k8s.io/system:aggregated-metrics-reader created
clusterrole.rbac.authorization.k8s.io/system:metrics-server:system:auth-delegator created
clusterrole.rbac.authorization.k8s.io/system:metrics-server:auth-reader created
clusterrole.apiresources.k8s.io created
deployment.apps/metrics-server created
service/metrics-server created
clusterrolebinding.rbac.authorization.k8s.io/system:metrics-server created
The image shows a GitHub repository for "kubernetes-metrics-server" with files, commit history, and contributor information, intended for development and learning environments.
It may take a few minutes for the metrics server to start gathering and reporting data. Verify this by listing the Pods again: 
kubectl get pods

​
Verifying Metrics Collection

After deployment, you can view resource usage by using the kubectl top commands.

​
Check Node Resource Utilization

Run the following command: 
kubectl top node
Expected output: 
NAME         CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%
controlplane 470m         1%     1252Mi          0%
node01       57m          0%     349Mi           0%

​
Check Pod Resource Utilization

To view the resource allocation for each Pod, execute: 
kubectl top pod
Expected output: 
NAME       CPU(cores)   MEMORY(bytes)
elephant   20m          32Mi
lion       1m           18Mi
rabbit     131m         252Mi

​
Analyzing Resource Consumption

Use the collected metrics to identify resource usage by both nodes and Pods.

​
Node Resource Analysis

  1. Identifying the Node with the Highest CPU Consumption From the kubectl top node output, the “controlplane” node shows a CPU usage of approximately 470 milli cores, which is significantly higher than node01’s 57 milli cores. Answer: The controlplane node consumes the most CPU.
  2. Identifying the Node with the Highest Memory Consumption The controlplane node also consumes 1252Mi of memory compared to node01’s 349Mi. Answer: The controlplane node consumes the most memory.

​
Pod Resource Analysis

  1. Identifying the Pod with the Highest Memory Consumption The “rabbit” Pod consumes 252Mi of memory, which is higher than the other two Pods. Answer: The “rabbit” Pod consumes the most memory.
  2. Identifying the Pod with the Lowest CPU Consumption The “lion” Pod uses only 1m of CPU, making it the least resource-intensive among the Pods. Answer: The “lion” Pod consumes the least CPU.

​
Recap

In this lesson, you performed the following steps:
  • Inspected running Pods using the kubectl get pods command.
  • Cloned and explored a preconfigured metrics server repository.
  • Deployed the metrics server using kubectl create -f ..
  • Verified node and Pod resource usage utilizing the kubectl top commands.
  • Analyzed resource consumption to identify which nodes and Pods were consuming the most or least CPU and memory.

​
Final Verification

Run the following commands one more time to confirm metrics collection: 
kubectl top node
Expected output: 
NAME           CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%
controlplane   470m         1%     1252Mi          0%
node01         57m          0%     349Mi           0%

kubectl top pod
Expected output: 
NAME        CPU(cores)   MEMORY(bytes)
elephant    20m          32Mi
lion        1m           18Mi
rabbit      131m         252Mi
This concludes the lesson on monitoring cluster components using the metrics server. For additional Kubernetes monitoring resources, consider visiting the Kubernetes Documentation. Happy monitoring!

Watch Video