General

• YAML files - All yaml files should use 2 space indents and end with .yml
• Variables - Use jinja variable syntax over deprecated variable syntax. {{ var }} not $var
• Use spaces around jinja variable names. {{ var }} not {{var}}
• Variables that are environment specific and that need to be overridden should be in ALL CAPS.
• Variables that are internal to the role should be lowercase.
• Prefix all variables defined in a role with the name of the role. Example: EDXAPP_FOO
• Keep roles self contained - Roles should avoid including tasks from other roles when possible
• Plays should do nothing more than include a list of roles except where pre_tasks and post_tasks are required (to manage a load balancer for example)
• Plays/Playbooks that apply to the general community should be copied to configuration/playbooks
• Plays/Playbooks that apply only to a specific organization (edx-east, edx-west) should be copied to a sub-directory under configuration/playbooks
• Tests - For automated integration tests use an included play named test.yml is included and run when a run_tests var is set to validate the role installation on a single instance.
• Deploys - For application updates include a deploy.yml with a list of tasks that start with service stop and end with a service start that is run to deploy an application update. Every task in deploy.yml should be tagged with deploy
• Nothing outside of tasks in deploy.yml should affect the application state
• All tasks in deploy.yml should be able to run as a user with limited sudo rule and not require root access.
• Handlers - Every role should have one or more handlers for restarting the service(s), for tasks that run in main.yml that require service restarts they/it will be notified.
• Separators - Use underscores (e.g. my_role) not dashes (my-role).
• Paths - When defining paths, do not include trailing slashes (e.g. my_path: /foo not my_path: /foo/. When concatenating paths, follow the same convention (e.g. {{ my_path }}/bar not {{ my_path }}bar)

Conditionals and Return Status

• Always use when: for conditionals - To check if a variable is defined when: my_var is defined or when: my_var is not defined

• To verify return status (see conditionals)

  - command: /bin/false register: my_result ignore_errors: True - debug: msg="task failed" when: my_result|failed

Formatting

Break long lines using yaml line continuation.

Reference: http://docs.ansible.com/playbooks_intro.html

  - file: dest="{{ test }}" src="./foo.txt" mode=0077 state=present user="root" group="wheel"

- file: >
	dest="{{ test }}"
	src="./foo.txt"
	mode=0077
	state=present
	user="root"
	group="wheel"

  - file:
      dest: "{{ test }}"
	  src: "./foo.txt"
	  mode: 0077 
	  state: present
	  user: "root"
	  group: "wheel"

Roles

Role Variables

• group_vars/all - Contains variable definitions that apply to all roles.
• "common" role - Contains variables and tasks that apply to all roles that are edX specific.
• Roles variables - Variables specific to a role should be defined in /vars/main.yml. All variables should be prefixed with the role name.
• Role defaults - Default variables should configure a role to install edx in such away that all services can run on a single server
• Variables that are environment specific and that need to be overridden should be in all caps.

• Every role should have a standard set of role directories, example that includes a python and ruby virtualenv:

  edxapp_data_dir: "{{ COMMON_DATA_DIR }}/edxapp" edxapp_app_dir: "{{ COMMON_APP_DIR }}/edxapp" edxapp_log_dir: "{{ COMMON_LOG_DIR }}/edxapp" edxapp_venvs_dir: "{{ edxapp_app_dir }}/venvs" edxapp_venv_dir: "{{ edxapp_venvs_dir }}/edxapp" edxapp_venv_bin: "{{ edxapp_venv_dir }}/bin" edxapp_rbenv_dir: "{{ edxapp_app_dir }}" edxapp_rbenv_root: "{{ edxapp_rbenv_dir }}/.rbenv" edxapp_rbenv_shims: "{{ edxapp_rbenv_root }}/shims" edxapp_rbenv_bin: "{{ edxapp_rbenv_root }}/bin" edxapp_gem_root: "{{ edxapp_rbenv_dir }}/.gem" edxapp_gem_bin: "{{ edxapp_gem_root }}/bin"

Role Naming Conventions

• Role names - Terse, one word if possible, use underscores if necessary.
• Role task names - Terse, descriptive, spaces are OK and should be prefixed with the role name.
• Role handlers - Terse, descriptive, spaces are OK and should be prefixed with the role name

Secure vs. Insecure data

As a general policy we want to protect the following data:

• Usernames
• Public keys (keys are OK to be public, but can be used to figure out usernames)
• Hostnames
• Passwords, API keys

Directory structure for the secure repository:

ansible
├── files
├── keys
└── vars

The default secure_dir is set in group_vars/all and can be overridden by adding another file in group_vars that corresponds to a deploy group name.

For templates or files that are secure use first_available_file, example:

- name: xserver | install read-only ssh key for the content repo that is required for grading
  copy: src={{ item }} dest=/etc/git-identity force=yes owner=ubuntu group=adm mode=60
  first_available_file:
    - "{{ secure_dir }}/files/git-identity"
    - "git-identity-example"