Ubuntu and Rocky in one playbook
Handling OS differences with facts and variable files was very well explained.
Browse
Turn manual server setup into idempotent playbooks and roles: users, packages, configuration, services, secrets and tests you can trust
Created by Hana Novak
If you have ever set up a server by following a wiki page of commands, you know the problems: steps get skipped, servers drift apart and nobody remembers why a setting is there. Ansible replaces that wiki page with code that runs the same way every time, over plain SSH, with no agent to install on the servers.
We automate the setup of a small fleet of Ubuntu and Rocky Linux servers:
Section one is fully written and gets you from installation to a working playbook. Later sections build the fleet in small, testable steps. Every task is written to be idempotent: running a playbook twice changes nothing the second time, and we check this explicitly.
We use ansible-core with the community collections you need, fully qualified module names and ansible-lint from the start, so your playbooks follow current conventions. The lab uses local virtual machines created with Multipass or Vagrant, and everything also works against cloud servers.
Linux automation engineer teaching Bash scripting and Ansible for people who manage real servers
I look after fleets of Linux servers and I automate anything I have to do twice. Bash covers the small jobs and Ansible covers everything that needs to be repeatable across many machines.
My courses are practical and a little strict. You will learn to write scripts that fail safely, quote variables properly and log what they did, and playbooks that can run a second time without breaking anything.
I build every lab on virtual machines you can create for free, and I explain each command line flag the first time it appears. If you have ever copied a script from the internet and hoped it worked, my courses are for you.
Handling OS differences with facts and variable files was very well explained.
Running every playbook twice and expecting zero changes is now my habit. It caught several sloppy shell tasks.
Validate plus testing on one host with a second session open prevented a very bad afternoon.
Turning a long playbook into small roles made our automation much easier to share. Molecule took some setup.
Practical, not theoretical. The rolling update lesson matched exactly what we needed for our web servers.
Content is good. Getting the lab virtual machines running on my laptop took a while.