Project Overview

This project establishes a robust, repeatable pipeline for deploying K3s Kubernetes clusters. By leveraging a modular "Separation of Concerns" strategy, we utilize Terraform for infrastructure provisioning and Ansible for software configuration, creating a scalable GitOps-lite workflow.

The Architectural Strategy

To ensure maintainability and portability, the architecture is divided into two distinct layers:

• Provisioning (Terraform): Focused solely on the lifecycle of the guest VMs, including disk creation, network attachment, and SSH key injection via cloud-init.
• Configuration (Ansible): Automates the OS state and K3s installation once VMs reach a "Ready" state, ensuring a clean separation from infrastructure code.

Innovative Pipeline Features

• Dynamic Inventory: Instead of static host files, the pipeline utilizes dynamic inventory plugins to automatically discover VMs immediately after creation.
• Post-Provision Hooks: Utilizes Terraform null_resource with local-exec to trigger Ansible playbooks automatically, removing manual intervention.
• Modular Roles: Infrastructure logic is organized into discrete roles (e.g., k3s_prereqs, k3s_cluster_join) for easier debugging and scaling.
• Automated Validation: Post-deployment sanity checks are integrated directly into the Ansible workflow to validate cluster health via kubectl before marking the deployment as complete.

Strategic Value

This "Infrastructure as Code" approach provides high portability across different hypervisors. By treating Ansible roles as OS-level configurations rather than hypervisor-specific scripts, the pipeline can be adapted from on-prem KVM/VMware labs to cloud environments with minimal changes.