AWS EKS Cluster Setup for Spring Boot Project
Budget: $30 – $250 USD
I am preparing to roll out a Spring Boot–based micro-service suite on Kubernetes and need a rock-solid Amazon EKS foundation first. The cluster does not exist yet, nor does the underlying network, so the engagement starts with designing and provisioning a brand-new VPC that follows AWS best practices (public/private subnets, NAT gateways, proper route tables and security groups).
Once the network is in place, the task is to stand up an EKS cluster that is ready for CI/CD-driven deployments later on. I want node groups sized for typical Java workloads, IAM roles mapped cleanly to Kubernetes RBAC, and cluster logging wired into CloudWatch. Spot capacity or managed node groups are fine as long as the solution is stable and cost-aware.
Deliverables
• Terraform, CloudFormation or eksctl scripts (your choice) that create the VPC and EKS cluster from scratch
• Documentation of every variable and command so the environment can be reproduced in another account
• kubeconfig and an initial namespace structure ready to receive Spring Boot services
• Verification that kubectl commands work from my workstation and that a simple test pod schedules successfully
Acceptance criteria
– A new AWS account can run the provided scripts and end up with a functional EKS cluster in its own VPC.
– Nodes join automatically and scale as defined.
– Cluster and audit logs appear in CloudWatch.
– No manual console clicks required after script execution.
Tools I am comfortable with include Terraform, eksctl, Helm, kubectl and GitHub Actions, so feel free to weave any of these into the solution. The follow-up phase will cover CI/CD and the actual Spring Boot deployment, but for this milestone I just need the infrastructure running cleanly.
Once the network is in place, the task is to stand up an EKS cluster that is ready for CI/CD-driven deployments later on. I want node groups sized for typical Java workloads, IAM roles mapped cleanly to Kubernetes RBAC, and cluster logging wired into CloudWatch. Spot capacity or managed node groups are fine as long as the solution is stable and cost-aware.
Deliverables
• Terraform, CloudFormation or eksctl scripts (your choice) that create the VPC and EKS cluster from scratch
• Documentation of every variable and command so the environment can be reproduced in another account
• kubeconfig and an initial namespace structure ready to receive Spring Boot services
• Verification that kubectl commands work from my workstation and that a simple test pod schedules successfully
Acceptance criteria
– A new AWS account can run the provided scripts and end up with a functional EKS cluster in its own VPC.
– Nodes join automatically and scale as defined.
– Cluster and audit logs appear in CloudWatch.
– No manual console clicks required after script execution.
Tools I am comfortable with include Terraform, eksctl, Helm, kubectl and GitHub Actions, so feel free to weave any of these into the solution. The follow-up phase will cover CI/CD and the actual Spring Boot deployment, but for this milestone I just need the infrastructure running cleanly.