Deploying CNGFW in Azure - VNet

This project will work us through deploying Palo Alto in Azure. We’ll be deploying this using the Hub and Spoke model and also managing policies via Panorama.

We’ll look at the VWan deployment in another project as this is another deployment model in Azure.

Pre-requisite

Certain things needs to be in place to enable you complete this project.

  • A Palo Alto Customer Support Account.
  • An Azure account
  • Some knowledge of Terraform
  • Intermediate knowledge of Azure, especially around networking
  • A GitHub account and some knowledge of GitHub Actions
  • Azure CLI

Some concepts

Hub and Spoke Architecture

Hub and Spoke Architecture is a design model often used in network systems, data management, logistics, and enterprise architecture. It’s structured around a central “hub” that connects to multiple “spokes” (peripheral systems, locations, or processes).

Advantages ✅ Disadvantages 🛑
Centralised control and management Single point of failure at the hub
Simplifies security and monitoring Potential bottlenecks
Easier maintenance and upgrades in the hub Higher latency for inter-spoke communication (must go through hub)

Table 1: Some pros and cons of Hub and Spoke architecture. See What is a Hub-and-Spoke Network and xx blog for some more info on the subject.

What we will build?

CNGFW Hub & Spoke Architecture

Img 1: Final project architecture

Summary

At the end of this project we would have deployed the following

  • 2 resource groups
  • 3 virtual networks
  • 2 virtual machines
  • 1 Panorama server
  • 1 Jumpbox virtual machine
  • 3 Public IPs
  • 2 Route tables

We will refer to information in these docs from time to time

Deployment Phases

  • Setup CI/CD pipeline in Git/GitHub
  • Deploy the networking components
  • Deploy the management components
    • Register the management component
  • Deploy CNGFW resource
  • Update networking components (network path)
  • Configure CNGFW for inbound/outbound flow
  • Wrap up and clear down resources

Setup CI/CD pipeline

ℹ️ Watch out for more updates coming soon