Hi all,
I have a set of Ansible configuration that I intend to use as a Generic
template for setting up a Django webserver and related services. This
project contains a multitude of roles for configuring various aspects
optional services (Nginx, Django, Redis, Postgres, etc.).
I'm trying to find a good way to specify good global default, overridable
variables that can be used throughout all of my roles. For example, I'd
like to define variables like "app_user" or "app_name" (and many others)
that I can use in all of my roles.
Right now, these variables are placed at the top level of my project inside
of env_vars/base.yml. I'm thinking that this would be my global default
configuration file that would define all of the defaults throughout the
project and inform users of what settings are available to override.
Also, the role defaults that I have defined in each role read in variables
set in the env_vars/base.yml (my global defaults file), and set local
copies of those variables for use in that role's tasks.
# contents of roles/postgres/defaults/main.yml
---
postgres__db_user: "{{ db_user }}"
In that example, that global "db_user" might be used in several roles. I
like this approach because it's very DRY and modular; the global 'db_user'
variable sets the namespaced local 'postgres__db_user' variable inside of
the postgres role, but can be namespaced differently in other roles.
What I'd love to be able to do is use the inventory file to override the
global defaults. The inventory file seems like a good place to put
variables that would customize the provisioning of specific hosts.
However, the problem is that the inventory file is one of the things that
takes least precedence in the variable precedence hierarchy.
http://docs.ansible.com/ansible/playbooks_variables.html#variable-precedence-where-should-i-put-a-variable
(I'm a bit confused by this because. I'd think the inventory file, being
the thing that is most likely be used to create host-specific
configurations, would take a high precedence.)
I'm wondering if there is a better way to organize my project such that I
can set some global defaults that can be the defaults for all roles, retain
modular roles with the namespaced variables, but be able to cleanly
configure any non-defaults in the inventory.
Does anyone know how I might achieve this or know of any other best
practices for situations like this?
Note: I'm currently using the --extra-vars command line switch to pass in
variables that I want to override, but this makes for really long command
line executions.
Here is my project thus far to see a working example:
https://github.com/JoeJasinski/ansible-django-stack
Thanks for any advice,
Joe
--
You received this message because you are subscribed to the Google Groups
"Ansible Project" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/ansible-project/29b46d5d-c57d-4482-a711-61d5928c96e8%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
