How to Manage Kubernetes Clusters With kubectl

The kubectl command is how administrators interact with and manage a Kubernetes cluster on a Linux system. It is an essential command line tool that works with all Kubernetes cluster envrionments like Amazon Web Services, Google Cloud Platform, or a cluster on your own hardware (e.g., kubeadm). Managing Kubernetes clusters at scale can be a challenge at first, but mastering the kubectl command will make it much easier.

In this tutorial, we will learn how to manage Kubernetes clusters with the kubectl command. We will explore various command options and see how to effectively use kubectl, as well as some tips and tricks to make using kubectl easier.

In this tutorial you will learn:

• How to use kubectl commands against a Kubernetes cluster

Software Requirements and Linux Command Line Conventions

Category	Requirements, Conventions or Software Version Used
System	Any Linux distro
Software	Kubernetes, kubectl
Other	Privileged access to your Linux system as root or via the sudo command.
Conventions	# – requires given linux commands to be executed with root privileges either directly as a root user or by use of sudo command $ – requires given linux commands to be executed as a regular non-privileged user

Manage Kubernetes Clusters With kubectl

---

---

The examples below show some of the most common and essential kubectl commands to use against a Kubernetes cluster in Linux.

1. To apply a manifest file, which can also deploy an application in Kubernetes, we can use the following syntax:

   $ kubectl apply -f ./application.yaml 
   

   This can deploy our custom Docker containerized application or one that we retrieve from a central repository, such as Docker Hub.

2. We can also delete a deployment with the delete parameter. Here are some examples:

   $ kubectl delete deployment my-application
   

   Or specify the file:

   $ kubectl delete -f application.yaml
   

3. Retrieve information about all of the Kubernetes pods in your network with:

   $ kubectl get pods
   

   Or the more verbose version, which will show pods in all name spaces:

   $ kubectl get pods --all-namespaces
   

   

   Note that the command option --all-namespaces can also be shortened to -A:

   $ kubectl get pods -A
   

   ---

   ---

4. See information about all the configured nodes in your Kubernetes cluster:

   $ kubectl get nodes
   

5. You can see a running list of all available services running in your cluster with the following command, issued from the Kubernetes maser node:

   $ kubectl get svc
   

6. Get details about a specific pod that has been deployed (in this example, one named ‘apache’):

   $ kubectl get pods/apache
   

   And more detailed information with the describe argument:

   $ kubectl describe pods/apache
   

7. See the logs for a specific pod:

   $ kubectl logs my-pod
   

8. We can use the run parameter to begin running an image. Let’s use Nginx for example:

   $ kubectl run --image=nginx nginx-server --port=80 --env="DOMAIN=cluster"
   

   We can then expose port 80:

   $ kubectl expose deployment nginx-deployment --port=80 --name=nginx-http
   

9. The kubectl command is also essential when scaling resources. To scale deployment ‘mysql’ to 5 replicas:

   $ kubectl scale --replicas=5 deployment/mysql  
   

   Or to scale resources as specified in a YAML file:

   $ kubectl scale --replicas=5 -f mysql.yaml
   

10. To drain all pods from a node (e.g., to perform maintenance on the node without it causing interruptions to the Kubernetes cluster):

    $ kubectl drain worker-node
    

11. To see all information about your Kubernetes cluster:

    $ kubectl cluster-info
    

---

---

Closing Thoughts

In this tutorial, we saw how to use the kubectl command to manage a Kubernetes cluster on a Linux system. Learning how to use kubectl is essential for administrators to create new deployments, pods, and manage nodes from the command line. It also allows us to inspect logs and see a current status of how the cluster and its individual components are operating at any given time.