Putting It All Together

Building systems from small components usually works better than tackling a gigantic project, but in the end still need to combine those components into a bigger system and provide its users with the expected user experience, be it a simple CLI or GUI.

This section describes how you can combine Ansible playbooks with front-end orchestrations systems, external databases or IPAM, and how you can use Ansible Tower (or AWX) to provide a basic GUI or API to your Ansible environment.

Ryan Heffernan and Mat Wood described how Facebook uses network automation concepts in their enterprise network.

UBS AG started using network automation to speed up the data center deployment process. The project was a huge success and resulted in UBS embracing automation in campus, WAN and remote office network deployments.

In this section Thomas Wacker described the UBS automation journey and how far they got in 2017.

Damien Garros explained how to build a network automation framework from the ground up:

• How and why they organized the project around a source of truth (NetBox & GitHub);
• How to integrate an automation framework (Ansible) to use the source of truth to manage the network devices;
• How to leverage Docker to package Ansible and create a consistent environment;
• How to properly import data into the source of truth and how to manage many roles and variations of the network design;
• How to migrate from a traditional SNMP based monitoring to a modern architecture using Influxdb & Grafana.

Sander Steffann built a complete network automation solution including integration with NetBox, BGP Peering Manager, and full-blown CI/CD pipeline.

In his presentation he described the tool selection, business processes he had to support, and the evolution of his solution from simple configuration templating to full-blown GitLab-based CI/CD pipeline with automated configuration deployment.

In this presentation Dirk Feldhaus described an Ansible-based solution to Cisco ACI tenant provisioning he developed as a student of this course. The solution is used for production deployments within his company.

The structure of Dirk's presentation mirrors the approach you should take when developing any network automation solution:

• Figure out the problem and create a pilot solution;
• Define the service you plan to automate;
• Describe the service with a data model;
• Develop automation scripts that provision the target service;
• Develop support scripts to monitor and operate the service.

Ireland Internet Exchange (INEX) uses IXP Manager (an open-source tool developed by INEX and used by over 60 IXPs worldwide) as the orchestration layer on top of SaltStack to provision and operate multi-vendor INEX infrastructure.

In his presentation, Nick Hilliard described how to take a practical, hands-on approach to network automation for live networks, starting at business analysis, product abstraction and data modeling, and ending up with practical automation of live configurations.

Mitja Robas and a team of software developers created a fully-automated disaster recovery solution for a customer who wanted to migrate from an existing stretched VLAN architecture to a truly redundant multi-data-center design.

In his presentation Mitja described the underlying network architecture, automation workflows, and the overall architecture of the automation solution.

Anton Karneliuk focused on network automation in very large service provider networks (with examples from mobile operators). His presentation covered:

• The typical vendor promises;
• Service Provider-specific scenarios and challenges;
• Using Ansible to automate Service Provider networks;
• How to start your automation journey

Kurt Wauters and Wim de Hul described a typical automation journey in a large organization:

• From early enthusiasm to disappointments;
• Finding a challenge that is simple enough to solve while resulting in immediate wins;
• Slowly growing the solution, and moving from read-only queries to service deployment automation and orchestration

During the Q&A part of the live sessions we addressed these questions:

• Managing test and production environments;
• Integrating Ansible with external systems like IPAMs, NMS, or CMDBs;
• Executing Ansible plays on dynamically-created hosts
• Inputting data;
• Writing your own Ansible module

After completing the self-study materials in this section, listen to these Software Gone Wild episodes.