Roles let you automatically load related vars_files, tasks, handlers, and other Ansible artifacts based on a known file structure. Once you group your content in roles, you can easily reuse them and share them with other users.
An Ansible role has a defined directory structure with seven main standard directories. You must include at least one of these directories in each role. You can omit any directories the role does not use. For example:
-``defaults/main.yml`` - default variables for the role (see :ref:`playbooks_variables` for more information). These variables have the lowest priority of any variables available, and can be easily overridden by any other variable, including inventory variables.
-``vars/main.yml`` - other variables for the role (see :ref:`playbooks_variables` for more information).
-``files/main.yml`` - files that the role deploys.
-``templates/main.yml`` - templates that the role deploys.
-``meta/main.yml`` - metadata for the role, including role dependencies.
You can add other YAML files in some directories. For example, you can place platform-specific tasks in separate files and refer to them in the ``tasks/main.yml`` file:
Roles may also include modules and other plugin types in a directory called ``library``. For more information, please refer to :ref:`embedding_modules_and_plugins_in_roles` below.
.._role_search_path:
Storing and finding roles
=========================
By default, Ansible looks for roles in two locations:
- in a directory called ``roles/``, relative to the playbook file
If you store your roles in a different location, set the :ref:`roles_path <DEFAULT_ROLES_PATH>` configuration option so Ansible can find your roles. Checking shared roles into a single location makes them easier to use in multiple playbooks. See :ref:`intro_configuration` for details about managing settings in ansible.cfg.
- If roles/x/tasks/main.yml exists, Ansible adds the tasks in that file to the play.
- If roles/x/handlers/main.yml exists, Ansible adds the handlers in that file to the play.
- If roles/x/vars/main.yml exists, Ansible adds the variables in that file to the play.
- If roles/x/defaults/main.yml exists, Ansible adds the variables in that file to the play.
- If roles/x/meta/main.yml exists, Ansible adds any role dependencies in that file to the list of roles.
- Any copy, script, template or include tasks (in the role) can reference files in roles/x/{files,templates,tasks}/ (dir depends on task) without having to path them relatively or absolutely.
When you use the ``roles`` option at the play level, Ansible treats the roles as static imports and processes them during playbook parsing. Ansible executes your playbook in this order:
- Each role listed in ``roles:``, in the order listed. Any role dependencies defined in the role's ``meta/main.yml`` run first, subject to tag filtering and conditionals. See :ref:`role_dependencies` for more details.
If using tags with tasks in a role, be sure to also tag your pre_tasks, post_tasks, and role dependencies and pass those along as well, especially if the pre/post tasks and role dependencies are used for monitoring outage window control or load balancing. See :ref:`tags` for details on adding and using tags.
When using ``vars:`` within the ``roles:`` section of a playbook, the variables are added to the play variables, making them available to all tasks within the play before and after the role. This behavior can be changed by :ref:`DEFAULT_PRIVATE_ROLE_VARS`.
You can reuse roles dynamically anywhere in the ``tasks`` section of a play using ``include_role``. While roles added in a ``roles`` section run before any other tasks in a playbook, included roles run in the order they are defined. If there are other tasks before an ``include_role`` task, the other tasks will run first.
When you add a :ref:`tag <tags>` to an ``include_role`` task, Ansible applies the tag `only` to the include itself. This means you can pass ``--tags`` to run only selected tasks from the role, if those tasks themselves have the same tag as the include statement. See :ref:`selective_reuse` for details.
You can reuse roles statically anywhere in the ``tasks`` section of a play using ``import_role``. The behavior is the same as using the ``roles`` keyword. For example:
Ansible only executes each role once, even if you define it multiple times, unless the parameters defined on the role are different for each definition. For example, Ansible only runs the role ``foo`` once in a play like this:
Role dependencies let you automatically pull in other roles when using a role. Ansible does not execute role dependencies when you include or import a role. You must use the ``roles`` keyword if you want Ansible to execute role dependencies.
Role dependencies are stored in the ``meta/main.yml`` file within the role directory. This file should contain a list of roles and parameters to insert before the specified role. For example:
Ansible always executes role dependencies before the role that includes them. Ansible executes recursive role dependencies as well. If one role depends on a second role, and the second role depends on a third role, Ansible executes the third role, then the second role, then the first role.
Ansible treats duplicate role dependencies like duplicate roles listed under ``roles:``: Ansible only executes role dependencies once, even if defined multiple times, unless the parameters defined on the role are different for each definition. If two roles in a playbook both list a third role as a dependency, Ansible only runs that role dependency once, unless you pass different parameters or use ``allow_duplicates: true`` in the dependent (third) role. See :ref:`Galaxy role dependencies <galaxy_dependencies>` for more details.
To use ``allow_duplicates: true`` with role dependencies, you must specify it for the dependent role, not for the parent role. In the example above, ``allow_duplicates: true`` appears in the ``meta/main.yml`` of the ``tire`` and ``brake`` roles. The ``wheel`` role does not require ``allow_duplicates: true``, because each instance defined by ``car`` uses different parameter values.
If you write a custom module (see :ref:`developing_modules`) or a plugin (see :ref:`developing_plugins`), you might wish to distribute it as part of a role. For example, if you write a module that helps configure your company's internal software, and you want other people in your organization to use this module, but you do not want to tell everyone how to configure their Ansible library path, you can include the module in your internal_config role.
Alongside the 'tasks' and 'handlers' structure of a role, add a directory named 'library' and then include the module directly inside the 'library' directory.
If necessary, you can also embed a module in a role to modify a module in Ansible's core distribution. For example, you can use the development version of a particular module before it is released in production releases by copying the module and embedding the copy in a role. Use this approach with caution, as API signatures may change in core components, and this workaround is not guaranteed to work.
The same mechanism can be used to embed and distribute plugins in a role, using the same schema. For example, for a filter plugin:
`Ansible Galaxy <https://galaxy.ansible.com>`_ is a free site for finding, downloading, rating, and reviewing all kinds of community-developed Ansible roles and can be a great way to get a jumpstart on your automation projects.
The client ``ansible-galaxy`` is included in Ansible. The Galaxy client allows you to download roles from Ansible Galaxy, and also provides an excellent default framework for creating your own roles.
