A reader recently sent in this feedback:
I always found it odd that, by default, I had to add the Inventory to /etc/ansible/hosts. When I am running on a hosted environment I usually do not have access to the root-tree and furthermore, when focusing on infrastructure-as-code my inventories belong inside a scm anyways.
You mention how to pass a inventory via cli in the book, but at a rather late stage. I remember that when I started with Ansible (before your book) it was the first thing I tried to figure out.
So I would suggest to add this information in the beginning of the book, maybe as a note just where you explain the initial hosts-file.
As I've been working on the Ansible 101 video series, I noticed the same thing and was a little bothered by it. Nowadays, I always store my inventory (and, quite often, an ansible.cfg file with the path to it configured) inside my project directory.
I think I should default to that in all the early examples, and just mention that you can also store inventory globally in /etc/ansible/hosts.
Thanks Marc!
A reader recently sent in this feedback:
As I've been working on the Ansible 101 video series, I noticed the same thing and was a little bothered by it. Nowadays, I always store my inventory (and, quite often, an
ansible.cfgfile with the path to it configured) inside my project directory.I think I should default to that in all the early examples, and just mention that you can also store inventory globally in
/etc/ansible/hosts.Thanks Marc!