Step 1: Install ClusterAPI on Management Cluster
We’ll use a Kind cluster as our management cluster. To begin, we need to deploy ClusterAPI with the AWS infrastructure provider.
export AWS_ACCESS_KEY_ID=
export AWS_SECRET_ACCESS_KEY=
clusterawsadm bootstrap iam create-cloudformation-stack --region us-east-1
export AWS_B64ENCODED_CREDENTIALS=$(clusterawsadm bootstrap credentials encode-as-profile)
Step 7: Removing user
Just as easily as we created a cluster by adding a user, we can remove a user and, consequently, delete their associated EKS cluster. This is achieved through the same GitOps workflow, but in reverse.
To remove a user, we simply submit a pull request (PR) that removes the user’s entry from the existing-users.yaml ConfigMap. For example, to remove user1, we would revert the changes we made in Step 4. Once this PR is merged and Argo CD synchronizes the changes to the management cluster, Sveltos detects the removal of the user from the ConfigMap.
Sveltos, upon detecting this change, proceeds to delete all the ClusterAPI resources it previously deployed for that user. This includes the Cluster, AWSManagedCluster, AWSManagedControlPlane, MachinePool, and AWSManagedMachinePool resources within the user’s namespace (in this case, “user1”).
Conclusion
In this tutorial, we’ve demonstrated how to build a fully automated, GitOps-driven pipeline for provisioning dedicated EKS clusters on demand. By leveraging the power of Argo CD, Sveltos, and ClusterAPI, we’ve established a robust and scalable solution that eliminates the manual effort and potential errors associated with traditional cluster management.
Contact Information
If you have some questions, would like to have a friendly chat or just network to not miss any topics, then don’t use the comment function at medium, just feel free to add me to your LinkedIn network!
Support this project
If you enjoyed this article, please check out the Projectsveltos GitHub repo. You can also star 🌟 the project if you found it helpful.
