Video summary
Kubernetes End to End project on EKS | EKS Install and app deploy with Ingress | #abhishekveeramalla
Main summary
Key takeaways
Summary
Abhishek Veeramalla’s Day 22 tutorial explains AWS Elastic Kubernetes Service (EKS) and demonstrates deploying a public-facing 2048 game on an EKS cluster.
Key Concepts
- EKS manages the Kubernetes control plane: AWS operates highly available control-plane components, such as the API server and etcd. This reduces the effort required to install, maintain, and troubleshoot those components compared with running Kubernetes yourself on EC2.
- Worker-node options: The tutorial uses AWS Fargate to run pods without managing EC2 worker instances. EC2-based worker nodes are another option, but require more responsibility for scaling and availability. Fargate profiles are associated with namespaces, so a profile must cover a namespace before pods can run there.
- Ingress and external access: A Kubernetes
Servicealone does not provide the desired public entry point. The demo uses an Ingress resource to define routing and the AWS Load Balancer Controller to watch that resource and create and configure an Application Load Balancer (ALB). Traffic flows from the ALB through the service to the application pods. The presenter contrasts this approach with creating a separate load balancer service for each application, which can increase costs. - Workload identity and permissions: The controller needs AWS permissions to create and manage load-balancer resources. The tutorial configures an IAM OIDC provider, an IAM policy and role, and a Kubernetes service account for the controller.
Tutorial Steps
- Install and configure
kubectl,eksctl, and the AWS CLI. - Use
eksctlto create an EKS cluster inus-east-1. The command also provisions networking, including public and private subnets. Cluster creation takes several minutes. - Configure local
kubectlaccess withaws eks update-kubeconfig. - Create a Fargate profile for the application’s namespace.
- Apply Kubernetes manifests for the 2048 app’s namespace, deployment, service, and Ingress. The deployment uses five replicas.
- Associate the cluster with an IAM OIDC provider, create the controller’s permissions and service account, and install the AWS Load Balancer Controller using Helm.
- Verify that the controller is running, confirm that it created an ALB for the Ingress, and open the ALB address to access the game.
Troubleshooting and Resources
The presenter demonstrates troubleshooting a controller deployment that remained unavailable because of a service-account or CloudFormation-stack issue. He deletes or replaces the stale resources and retries the setup.
The video points viewers to a GitHub repository containing commands and manifests, and mentions a Kubernetes playlist for background concepts. Kubernetes and AWS EKS documentation are also referenced for installation instructions and AWS Load Balancer Controller setup.
Practical note: The presenter uses a root AWS account for convenience during the demo. This is a demonstration choice, not a general security recommendation. Real deployments should use appropriately scoped IAM permissions.
Although the presenter says he will discuss EKS drawbacks, the subtitles do not include a substantial drawbacks analysis beyond noting operational choices and load-balancer cost considerations.
Main Speakers and Sources
- Abhishek Veeramalla — presenter and instructor.
- GitHub repository for the course — commands and Kubernetes manifests used in the demo.
- Kubernetes and AWS EKS documentation — installation instructions and AWS Load Balancer Controller setup.
Rate this summary
Your feedback will help improve summaries.
Improve this summary
Reprocess with a stronger model when the summary feels incomplete or inaccurate.
Translate summary in another language
Ask questions to this video
Chat for follow-up questions, clarifications, and source-backed answers.