Skip to content

Chapter 2: Automation Architecture Overview

This chapter describes the hub-spoke automation architecture that orchestrates all configuration changes across Abhavtech's enterprise infrastructure, including the tool responsibilities matrix and role separation model.

What You'll Learn

High-Level Architecture

The automation architecture follows a hub-spoke model with:

  • Central Automation Control Node: Ubuntu WSL/Linux VM running Terraform, Ansible, Python
  • Git Repository: Source of truth for all infrastructure code
  • HashiCorp Vault: On-premises secrets management
  • Target Platforms: DNAC cluster, ISE cluster, SD-WAN controllers, Fabric nodes, WAN edge devices

Tool Responsibilities Matrix

Clear separation of concerns:

  • Terraform: Provisions infrastructure (DNAC site hierarchy, ISE SGTs, vManage templates)
  • Ansible: Configures devices (IS-IS, LISP, BGP, 802.1X)
  • Python: Custom API integration (SD-WAN feature templates, Webex automation)

Role Separation

Three distinct roles with clear boundaries:

  • Network Architect: Design authority, approves pull requests
  • Automation Engineer: Writes Terraform/Ansible code, submits pull requests
  • Network Operations: Executes approved automation during change windows

Chapter Navigation

Key Concepts

Control Node Requirements

The automation control node must have network reachability to: - DNAC cluster management IP (10.252.10.x) - ISE PAN/MNT nodes (10.252.30.x) - vManage cluster (10.252.50.x) - All fabric node management interfaces - HashiCorp Vault (10.252.200.10)

Why Not Containers/Kubernetes?

This framework deliberately avoids cloud-native orchestrators. It uses enterprise on-premises tools that network engineers already understand - no Docker, no Kubernetes, no cloud CI/CD platforms.

Previous: ← Executive Summary
Next: Development Environment Setup →