{"id":1692,"date":"2019-05-13T05:09:13","date_gmt":"2019-05-13T05:09:13","guid":{"rendered":"http:\/\/kusuaks7\/?p=1297"},"modified":"2023-07-13T08:48:49","modified_gmt":"2023-07-13T08:48:49","slug":"key-kubernetes-commands","status":"publish","type":"post","link":"https:\/\/www.experfy.com\/blog\/bigdata-cloud\/key-kubernetes-commands\/","title":{"rendered":"Key Kubernetes Commands"},"content":{"rendered":"

Up and running with\u00a0K8s<\/h2>\n

Kubernetes is the premier technology for deploying and managing large apps. In this article, we\u2019ll get up and running with K8s on your local machine. Then you\u2019ll Deploy your first app. Finally, you\u2019ll see the top K8s commands to know.<\/p>\n

If you\u2019re new to Kubernetes, check out my previous article<\/a>\u00a0to learn fundamental K8s concepts.<\/p>\n

Let\u2019s set sail!<\/p>\n

<\/canvas><\/figure>\n

Lots of containers on a\u00a0ship.<\/p>\n

Running K8s\u00a0Locally<\/strong><\/h3>\n

Until recently, Minikube<\/a>\u00a0was the primary means of creating a local, single-node Kubernetes cluster. It works on Mac, Windows, and Linux. Minikube runs inside a virtual environment with drivers for many different virtual environments. It also has an extensive API. However, when using Minikube with VirtualBox, I found it to be slow and buggy.<\/p>\n

If you are playing with K8s locally on a Mac or Windows machine, I recommend you use the K8s version that comes with Docker Desktop. It uses HyperKit as its virtual environment. I found it to be a breath of fresh air.<\/p>\n

If you\u2019re using a Linux machine with K8s locally, check out\u00a0MicroK8s<\/a>. It\u2019s lightweight and uses a\u00a0snap<\/em>\u00a0instead of a virtual environment.<\/p>\n

For this article, I\u2019m running K8s v1.13.0 from Docker Desktop CE,\u00a0Edge<\/em>version in April 2019.<\/p>\n

<\/canvas><\/figure>\n

To install and start K8s on a Mac, first, download Docker Desktop with K8s\u00a0here<\/a>.<\/p>\n

Once installed, go to your apps and start Docker. Install and start K8s by selecting\u00a0Preferences<\/em>\u00a0from the Docker\u00a0<\/em>icon at the top of your screen.<\/p>\n

<\/p>\n

Select the\u00a0Kubernetes<\/em>\u00a0tab and check the\u00a0Enable Kubernetes<\/em>\u00a0box.<\/p>\n

<\/canvas><\/figure>\n

Agree to install Kubernetes. It might take a minute or two to fire up. Click on the Docker dropdown and you should see\u00a0Kubernetes is starting<\/em>\u00a0and then eventually,\u00a0Kubernetes is running<\/em>.<\/p>\n

<\/figure>\n

If you see two green circles, then you\u2019re set! If you don\u2019t see two green circles, Google and\u00a0Stack Overflow<\/a>\u00a0are a developer\u2019s best friends.<\/p>\n

Your First K8s\u00a0App<\/h3>\n

Let\u2019s make a\u00a0hello world<\/em>\u00a0Deployment that will pull a Docker image for a node.js app. Our app will display\u00a0Hello World<\/em>\u00a0in the browser. In production, you\u2019ll do things differently. This exercise is the first few steps to get your feet wet.<\/p>\n

With K8s running, enter the following command in your terminal:<\/p>\n

kubectl create deployment hello-node --image=gcr.io\/hello-minikube-zero-install\/hello-node<\/code><\/p>\n

We\u2019re using the\u00a0kubectl<\/code>\u00a0API to create a deployment named\u00a0hello-node\u00a0<\/em>that uses a Docker image stored in the\u00a0Google Container Registry<\/a>. We\u2019re using the Docker container from\u00a0the official Kubernetes\u00a0Hello Minikube<\/em>\u00a0tutoria<\/a>l.<\/p>\n

Here are the contents of the Dockerfile:<\/p>\n

FROM node:6.14.2
\nEXPOSE 8080
\nCOPY server.js .
\nCMD node server.js<\/span><\/div>\n

We\u2019re using an official node.js image, copying the contents of\u00a0server.js,\u00a0<\/em>exposing port 8080, and running the\u00a0server.js<\/em>\u00a0program with node. For more on Docker files, check out my article\u00a0here<\/a>.<\/p>\n

Here\u2019s the\u00a0server.js<\/em>\u00a0app code:<\/p>\n

var http = require('http');<\/strong><\/code><\/span><\/p>\n

var handleRequest = function(request, response) {
\nconsole.log(‘Received request for URL: ‘ + request.url);
\nresponse.writeHead(200);
\nresponse.end(‘Hello World!’);<\/span>
\n};
\nvar www = http.createServer(handleRequest);
\nwww.listen(8080);<\/span><\/div>\n

This code will create a server that listens on port 8080. When the server receives a request, it writes a helpful message to the console and returns \u201cHello World!\u201d.<\/p>\n

After running the\u00a0kubectl create<\/em>\u00a0command, you should see a message in your terminal that your Deployment has been created. Your app won\u2019t be ready to use right away, though. K8s needs to pull the Docker image and spin up your container.<\/p>\n

You can view the status of all K8s resources with\u00a0kubectl get all<\/code>. It will take a few minutes to pull the image and spin up the container, so you\u2019ll probably enter this command a few times. You can peak behind the curtain and see what K8s is doing with\u00a0kubectl get events<\/code>.<\/p>\n

When you see\u00a01<\/em>\u00a0Pod is showing\u00a0Available<\/em>, you\u2019re good to go.<\/p>\n

<\/figure>\n

To make the Pod accessible to the world outside the K8s cluster, you need to expose the Pod as a Service. Create a Service with the command\u00a0kubectl expose deployment hello-node --type=LoadBalancer --port=8080<\/code>.<\/p>\n

You specified the name of the Cluster, the type of Service, and the port for it to use. Not all Service types create an IP that can be accessed from the outside world, but\u00a0LoadBalancer\u00a0<\/em>does.<\/p>\n

You should be able to see that the Service has been created by entering\u00a0kubectl get services<\/code>\u00a0in your shell.<\/p>\n

Assuming you\u2019re running K8s locally, navigate your browser to\u00a0http:\/\/www.experfy.com:8080\/<\/a>\u00a0and you should see the message \u201cHello World!\u201d.<\/p>\n

<\/canvas><\/figure>\n

Hello World!<\/p>\n

Instead of creating and specifying your whole Deployment and Service via the command line, you could have specified them in a YAML file. Using a file has several advantages, including reusability and documentation. We\u2019ll cover YAML specification files in depth in future articles, so follow\u00a0me<\/a>\u00a0to make sure you don\u2019t miss them.<\/p>\n

Now let\u2019s check out popular K8s commands.<\/p>\n

kubectl<\/h3>\n
<\/canvas><\/figure>\n

Cubes<\/p>\n

kubectl<\/em><\/a>\u200a<\/em>\u2014\u200apronounced \u201ccube control\u201d\u200a\u2014\u200a<\/em><\/strong>is the\u00a0<\/em>command line tool for K8s. Here\u2019s how it works:<\/p>\n