In this tutorial, you’ll learn about
- Installing kubectl and eksctl
- Creating and configuring a EKS cluster
Installing Kubectl
Instructions. Here’s a TLDR for Linux. For other OS, check the link.
Create a folder where you’ll keep it. For example, ~/bin
Go there, download the kubectl binary:
curl -LO
https://storage.googleapis.com/kubernetes-release/release/v1.20.0/bin/linux/amd64/kubectl
Make it executable:
chmod +x ./kubectl
Add this folder to PATH:
export PATH="~/bin:${PATH}"
Add this line to your .bashrc
Install eksctl
Eksctl is a command line tool for creating and managing EKS clusters. More info
Let’s install it to the same ~/.bin directory where we installed
kubectl:
curl --silent --location
"https://github.com/weaveworks/eksctl/releases/latest/download/eksctl\_$(uname
-s)_amd64.tar.gz" | tar xz -C ~/bin/
Create a EKS cluster
We’ll use eksctl for creating a cluster. More info
eksctl create cluster \
--name ml-bookcamp-eks \
--region eu-west-1
Note: if you want to use Fargate, check this tutorial. Fargate might be better, but the setup process is more complex.
It should also create a config file in ~/.kube/config. You should be
able to use kubectl now to connect to the cluster
Check that the config works:
kubectl get service
It should return the list of services currently running on a cluster:
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.100.0.1 <none> 443/TCP 6m17s
If you have an error like that:
[✖] unable to use kubectl with the EKS cluster (check 'kubectl version'):
Unable to connect to the server: getting credentials: exec: fork/exec
/usr/local/bin/aws-iam-authenticator: exec format error
You need to generate the config using aws cli (Instruction).
This is how you do it:
aws eks --region eu-west-1 update-kubeconfig --name ml-bookcamp-eks
It will create a config located at ~/.kube/config
Check that the config works:
kubectl get service
It should return the list of services currently running on a cluster:
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.100.0.1 <none> 443/TCP 6m17s
Deleting the cluster
Don’t forget to delete your cluster when you finish your experiments. Use eksctl for that:
eksctl delete cluster --name ml-bookcamp-eks