Hiive EKS Deployment with Terraform

Local machine requirements: Terraform, AWS CLI, kubectl. This project provisions a complete EKS cluster in AWS using Terraform and deploys a sample NGINX application.

• Provisions an AWS EKS cluster in ca-central-1
• Deploy nginx using raw yaml
• Set up required VPC, subnets, IAM roles, and security configuration
• Deploys NGINX using kubectl and a simple YAML file
• Uses Terraform modules for VPC, IAM, and EKS
• Minimal setup of 1-2 nodes (t3.small). Terraform destroy to not incur further costs.
• Creator access granted via enable_cluster_creator_admin_permissions = truefor easier take home project access to the nodes.

Prerequisites

• Terraform ≥ 1.5
• AWS CLI with valid credentials - I used an IAMRole instead of root account for best practises
• GitHub (public repo required)
• I used raw yaml for deployment and for the purpose of simplicity but for production or complex deployments, HELM would be a better choice.

Modules:

• vpc: VPC with two public subnets and internet gateway
• security: IAM roles for the EKS control plane and nodes
• eks: EKS cluster(1–2 t3.small nodes) and managed node group

How to Deploy

Clone the repo

git clone https://github.com/yourname/hiive cd hiive

Set up AWS credentials

Setup the aws credentials and initialize it on local machine via aws configure Make sure you are in the right scope via aws sts get-caller-identity

Initialize Terraform

terraform init

Review the plan

terraform plan

Apply the infrastructure

terraform apply

Nginx deployment

Run a simple yaml.

• Update context to make sure kubectl is in the right scope for the cluster - hiive % aws eks update-kubeconfig
  --name hiive-simple-eks-cluster
  --region ca-central-1

• Run deploy raw yaml for nginx - kubectl apply -f deploy-nginx.yaml

• Check your pods - kubectl get pods

To minimize cost, please run terraform destroy after testing.

Screenshots and short explanations

• cluster-running-on-aws.png - shows cluster running in ca-central-1
• terraform-successfully-deployed.png - shows full terraform output
• deployed-nginx-running-kubectl.png - shows deployed nginx as pods

Key hardships and debuggings

• Initial change from HELM deployment for raw YAML for simplicity
• Terraform syntax errors and module rewiring (such as order of module execution, right terraform versions, etc)
• Configuration issues such as installing and updating requirements on local, setting up aws cli, iamrole permissions, vpc route tables, kubectl initial configurations (e.g. aws-auth, enable_cluster_creator_admin_permissions, scoping)
• Remembering to destroy the infra to not incur AWS costs after verification (t3 small, while a small setup, incurs costs daily.)