Lab 05 - Introduction to Kubernetes
On this lab you’ll have a first hands-on experience with Kubernetes.
On this lab
- Prepare your machine
- Set context
- Run your first pod
- Using manifests
- Pod advanced configuration
- Work with namespaces
Prepare your machine
Before you start this lab you need to make sure you have your machine ready to use a Kubernetes dev cluster.
First, enable minikube cluster on your machine.
minikube start --extra-config=kubelet.housekeeping-interval="10s"
When you start minikube cluster, kubectl is automatically configured. Let’s confirm it’s working properly.
kubectl version
You may have an output similar with this.
Client Version: version.Info{Major:"1", Minor:"22", GitVersion:"v1.22.0", GitCommit:"c2b5237ccd9c0f1d600d3072634ca66cefdf272f", GitTreeState:"clean", BuildDate:"2021-08-04T18:03:20Z", GoVersion:"go1.16.6", Compiler:"gc", Platform:"windows/amd64"}
Server Version: version.Info{Major:"1", Minor:"21", GitVersion:"v1.21.9", GitCommit:"88ccc7584c4460cf46e4180be83e7c4ab7a8ffb4", GitTreeState:"clean", BuildDate:"2022-01-30T02:57:18Z", GoVersion:"go1.16.12", Compiler:"gc", Platform:"linux/amd64"}
Set Context
Now you need to configure the context of kubectl tool to be set for Kubernetes cluster configured by Docker Desktop.
Let’s list all available contexts.
kubectl config get-contexts
On the list you get, you’ll have a line with a * on CURRENT column. This represent the current context set for kubectltool.
To have direct access to current context you may run the following command.
kubectl config current-context
If the output of this command is not minikube, you need to change your current context.
kubectl config use-context minikube
kubectl config current-context
Now you should get minikube as the output of last commands.
Run your first pod
Getting a first pod running on Kubernetes can be done using an imperative command.
We’ll use a previously created image during containers labs.
kubectl run my-simple-site --image=tasb/hello-static-site --port=80
Let’s list all running pods to check if your pod is running properly.
kubectl get pods
You should get the following output.
NAME READY STATUS RESTARTS AGE
my-simple-site 1/1 Running 0 42s
On STATUS column you need to have RUNNING value.
Now let’s interact with our website.
kubectl port-forward my-simple-site 8080:80
Navigate to http://localhost:8080 to check that your website is working properly.
To exit port-forward you can user Ctrl+C on the terminal.
Let’s delete our pod to keep Kubernetes clean.
kubectl delete pod my-simple-site
Using manifests
Using a declarative configuration is the recommended way to interact with Kubernetes.
To create the same pod using this approach, create a file called pod.yaml with the following content. You need to be careful when add content to the file since whitespace have semantic value on YAML.
apiVersion: v1
kind: Pod
metadata:
name: my-simple-website
spec:
containers:
- image: tasb/hello-static-site
name: my-simple-site
ports:
- containerPort: 80
To send this manifest to Kubernetes use the next command.
kubectl apply -f pod.yaml
And let’s check if the pod is running properly using the following command.
kubectl get pods
You should see Running on Status column on returned list.
Now it’s time to test our pod.
kubectl port-forward my-simple-website 8080:80
Navigate to http://localhost:8080 to check that your website is working properly.
To exit port-forward you can user Ctrl+C on the terminal.
Now you will deploy a different version of that pod using the same file only changing image name.
On pod.yaml file change image to tasb/hello-static-site:k8s and run apply command again.
kubectl apply -f pod.yaml
And let’s check if the pod is running properly.
kubectl get pods
You should see Running on Status column on returned list.
Now let’s interact with our new website.
kubectl port-forward my-simple-website 8080:80
Navigate to http://localhost:8080 to check that your website is working properly.
To exit port-forward you can user Ctrl+C on the terminal.
Let’s delete our pod to keep Kubernetes clean.
kubectl delete pod my-simple-website
Pod advanced configuration
Now you’ll create a more complex pod with additional configuration.
First, if you need to enable metrics server on your cluster.
minikube addons enable metrics-server
You should get this output.
▪ Using image k8s.gcr.io/metrics-server/metrics-server:v0.4.2
🌟 The 'metrics-server' addon is enabled
This component may take some time to be ready. To test if it’s working properly execute the following command.
kubectl top nodes
When everything is working properly you should get an output like this.
NAME CPU(cores) CPU% MEMORY(bytes) MEMORY%
minikube 255m 3% 4Mi 0%
Now you need to create a file called vote-app.yaml with the following content. You need to be careful when add content to the file since whitespace have semantic value on YAML.
apiVersion: v1
kind: Pod
metadata:
name: vote-app
spec:
containers:
- name: azure-vote-back
image: mcr.microsoft.com/oss/bitnami/redis:6.0.8
ports:
- containerPort: 6379
env:
- name: ALLOW_EMPTY_PASSWORD
value: "yes"
resources:
limits:
cpu: 1000m
memory: "1G"
requests:
cpu: 500m
memory: "500M"
- name: azure-vote-front
image: mcr.microsoft.com/azuredocs/azure-vote-front:v1
ports:
- containerPort: 80
env:
- name: REDIS
value: localhost
resources:
limits:
cpu: 1000m
memory: "1G"
requests:
cpu: 500m
memory: "500M"
On this manifest, you’re creating a single pod with two containers: redis database and voting website.
Let’s dig on specific parts of manifest.
env:
- name: REDIS
value: localhost
On this block you defined a environment variable to be set on container startup to define where website can find the Redis database.
You use localhost as value since the two containers communicate using IPC (Inter Process Communication).
resources:
limits:
cpu: 1000m
memory: "1G"
requests:
cpu: 500m
memory: "500M"
On this block you defined the resources requests and limits for each container. Requests needs to be met by scheduler when selecting the node. Limits represent the maximum amount of each resource that the container can consume.
Let’s run this pod.
kubectl apply -f vote-app.yaml
And let’s check if the pod is running properly.
kubectl get pods
You should get an output like this.
NAME READY STATUS RESTARTS AGE
vote-app 2/2 Running 0 79m
On READYcolumn you see 2/2 that means you have 2 containers running inside this pod.
To get more details about this pod you may try the following command.
kubectl describe pod vote-app
Take some time to look into all the details from this pod.
Then you need to make a port forwarding to have access to this website.
kubectl port-forward vote-app 8080:80
Now it’s time to navigate to http://localhost:8080 and have access to voting app website.
You should get an webpage like this.
Please make your votes clicking on “Cats” and “Dogs” to make changes on the values.
Now we’ll enter Redis container to check the changes on data. Open a new terminal window and execute the following command.
kubectl -it exec vote-app -c azure-vote-back -- bash
Inside the container use Redis client to make some data queries. To start Redis client, use the following command inside the container.
redis-cli
After getting access to Redis, execute the following commands to get data from database.
keys *
get Cats
get Dogs
You should get the same numbers that you see on website.
To return to your terminal you need to execute 2 times exit command.
Work with namespaces
On this final topic you will see how namespaces works.
First, create a new namespace.
kubectl create ns my-ns
You still have your voting app running on default namespace and now you’ll create a new pod on new namespace.
kubectl apply -f vote-app.yaml -n my-ns
On the command you added -n my-ns option to inform Kubernetes that you want to create this pod on namespace my-ns.
From now on, you need to add this option on all your commands to interact with resources available on this namespace.
Let’s check if new pod is running properly.
kubectl get pods -n my-ns
You can see that the name of the pod is exactly the same that the previous one that you created. Since they are in different namespace this is possible to happen.
Let’s try to have access to this webapp.
kubectl port-forward vote-app -n my-ns 9000:80
Now it’s time to navigate to http://localhost:9000 and have access to voting app website.
You should get an webpage like this with 0 votes.
Please make your votes clicking on “Cats” and “Dogs” to make changes on the values.
To stop port forwarding use Ctrl+c on the terminal.
It’s time for clean up.
All resources that exists on a namespace are delete when you delete the namespace.
kubectl delete ns my-ns
The first voting app is still running and to delete it use the following command.
kubectl delete pod vote-app