community.general.deploy_helper module – Manages some of the steps common in deploying projects
Note
This module is part of the community.general collection (version 10.1.0).
You might already have this collection installed if you are using the ansible
package.
It is not included in ansible-core
.
To check whether it is installed, run ansible-galaxy collection list
.
To install it, use: ansible-galaxy collection install community.general
.
To use it in a playbook, specify: community.general.deploy_helper
.
Synopsis
The Deploy Helper manages some of the steps common in deploying software. It creates a folder structure, manages a symlink for the current release and cleans up old releases.
Running it with the
state=query
orstate=present
will return thedeploy_helper
fact.project_path
, whatever you set in thepath
parameter,current_path
, the path to the symlink that points to the active release,releases_path
, the path to the folder to keep releases in,shared_path
, the path to the folder to keep shared resources in,unfinished_filename
, the file to check for to recognize unfinished builds,previous_release
, the release the ‘current’ symlink is pointing to,previous_release_path
, the full path to the ‘current’ symlink target,new_release
, either the ‘release’ parameter or a generated timestamp,new_release_path
, the path to the new release folder (not created by the module).
Parameters
Parameter |
Comments |
---|---|
The attributes the resulting filesystem object should have. To get supported flags look at the man page for chattr on the target system. This string should contain the attributes in the same order as the one displayed by lsattr. The |
|
The name of the symlink that is created when the deploy is finalized. Used in Default: |
|
Name of the group that should own the filesystem object, as would be fed to chown. When left unspecified, it uses the current group of the current user unless you are root, in which case it can preserve the previous ownership. |
|
The number of old releases to keep when cleaning. Used in Default: |
|
The permissions the resulting filesystem object should have. For those used to /usr/bin/chmod remember that modes are actually octal numbers. You must give Ansible enough information to parse them correctly. For consistent results, quote octal numbers (for example, Giving Ansible a number without following either of these rules will end up with a decimal number which will have unexpected results. As of Ansible 1.8, the mode may be specified as a symbolic mode (for example, If If Specifying |
|
Name of the user that should own the filesystem object, as would be fed to chown. When left unspecified, it uses the current user unless you are root, in which case it can preserve the previous ownership. Specifying a numeric username will be assumed to be a user ID and not a username. Avoid numeric usernames to avoid this confusion. |
|
The root path of the project. Returned in the |
|
The release version that is being deployed. Defaults to a timestamp format |
|
The name of the folder that will hold the releases. This can be relative to Default: |
|
The level part of the SELinux filesystem object context. This is the MLS/MCS attribute, sometimes known as the When set to |
|
The role part of the SELinux filesystem object context. When set to |
|
The type part of the SELinux filesystem object context. When set to |
|
The user part of the SELinux filesystem object context. By default it uses the When set to |
|
The name of the folder that will hold the shared resources. This can be relative to Default: |
|
The state of the project.
Choices:
|
|
The name of the file that indicates a deploy has not finished. All folders in the Default: |
|
Influence when to use atomic operation to prevent data corruption or inconsistent reads from the target filesystem object. By default this module uses atomic operations to prevent data corruption or inconsistent reads from the target filesystem objects, but sometimes systems are configured or just broken in ways that prevent this. One example is docker mounted filesystem objects, which cannot be updated atomically from inside the container and can only be written in an unsafe manner. This option allows Ansible to fall back to unsafe methods of updating filesystem objects when atomic operations fail (however, it doesn’t force Ansible to perform unsafe writes). IMPORTANT! Unsafe writes are subject to race conditions and can lead to data corruption. Choices:
|
Attributes
Attribute |
Support |
Description |
---|---|---|
Support: full |
Can run in |
|
Support: none |
Will return details on what has changed (or possibly needs changing in |
Notes
Note
Facts are only returned for
state=query
andstate=present
. If you use both, you should pass any overridden parameters to both calls, otherwise the second call will overwrite the facts of the first one.When using
state=clean
, the releases are ordered by creation date. You should be able to switch to a new naming strategy without problems.Because of the default behaviour of generating the
new_release
fact, this module will not be idempotent unless you pass your own release name withrelease
. Due to the nature of deploying software, this should not be much of a problem.
Examples
# General explanation, starting with an example folder structure for a project:
# root:
# releases:
# - 20140415234508
# - 20140415235146
# - 20140416082818
#
# shared:
# - sessions
# - uploads
#
# current: releases/20140416082818
# The 'releases' folder holds all the available releases. A release is a complete build of the application being
# deployed. This can be a clone of a repository for example, or a sync of a local folder on your filesystem.
# Having timestamped folders is one way of having distinct releases, but you could choose your own strategy like
# git tags or commit hashes.
#
# During a deploy, a new folder should be created in the releases folder and any build steps required should be
# performed. Once the new build is ready, the deploy procedure is 'finalized' by replacing the 'current' symlink
# with a link to this build.
#
# The 'shared' folder holds any resource that is shared between releases. Examples of this are web-server
# session files, or files uploaded by users of your application. It's quite common to have symlinks from a release
# folder pointing to a shared/subfolder, and creating these links would be automated as part of the build steps.
#
# The 'current' symlink points to one of the releases. Probably the latest one, unless a deploy is in progress.
# The web-server's root for the project will go through this symlink, so the 'downtime' when switching to a new
# release is reduced to the time it takes to switch the link.
#
# To distinguish between successful builds and unfinished ones, a file can be placed in the folder of the release
# that is currently in progress. The existence of this file will mark it as unfinished, and allow an automated
# procedure to remove it during cleanup.
# Typical usage
- name: Initialize the deploy root and gather facts
community.general.deploy_helper:
path: /path/to/root
- name: Clone the project to the new release folder
ansible.builtin.git:
repo: ansible.builtin.git://foosball.example.org/path/to/repo.git
dest: '{{ deploy_helper.new_release_path }}'
version: v1.1.1
- name: Add an unfinished file, to allow cleanup on successful finalize
ansible.builtin.file:
path: '{{ deploy_helper.new_release_path }}/{{ deploy_helper.unfinished_filename }}'
state: touch
- name: Perform some build steps, like running your dependency manager for example
composer:
command: install
working_dir: '{{ deploy_helper.new_release_path }}'
- name: Create some folders in the shared folder
ansible.builtin.file:
path: '{{ deploy_helper.shared_path }}/{{ item }}'
state: directory
with_items:
- sessions
- uploads
- name: Add symlinks from the new release to the shared folder
ansible.builtin.file:
path: '{{ deploy_helper.new_release_path }}/{{ item.path }}'
src: '{{ deploy_helper.shared_path }}/{{ item.src }}'
state: link
with_items:
- path: app/sessions
src: sessions
- path: web/uploads
src: uploads
- name: Finalize the deploy, removing the unfinished file and switching the symlink
community.general.deploy_helper:
path: /path/to/root
release: '{{ deploy_helper.new_release }}'
state: finalize
# Retrieving facts before running a deploy
- name: Run 'state=query' to gather facts without changing anything
community.general.deploy_helper:
path: /path/to/root
state: query
# Remember to set the 'release' parameter when you actually call 'state=present' later
- name: Initialize the deploy root
community.general.deploy_helper:
path: /path/to/root
release: '{{ deploy_helper.new_release }}'
state: present
# all paths can be absolute or relative (to the 'path' parameter)
- community.general.deploy_helper:
path: /path/to/root
releases_path: /var/www/project/releases
shared_path: /var/www/shared
current_path: /var/www/active
# Using your own naming strategy for releases (a version tag in this case):
- community.general.deploy_helper:
path: /path/to/root
release: v1.1.1
state: present
- community.general.deploy_helper:
path: /path/to/root
release: '{{ deploy_helper.new_release }}'
state: finalize
# Using a different unfinished_filename:
- community.general.deploy_helper:
path: /path/to/root
unfinished_filename: README.md
release: '{{ deploy_helper.new_release }}'
state: finalize
# Postponing the cleanup of older builds:
- community.general.deploy_helper:
path: /path/to/root
release: '{{ deploy_helper.new_release }}'
state: finalize
clean: false
- community.general.deploy_helper:
path: /path/to/root
state: clean
# Or running the cleanup ahead of the new deploy
- community.general.deploy_helper:
path: /path/to/root
state: clean
- community.general.deploy_helper:
path: /path/to/root
state: present
# Keeping more old releases:
- community.general.deploy_helper:
path: /path/to/root
release: '{{ deploy_helper.new_release }}'
state: finalize
keep_releases: 10
# Or, if you use 'clean=false' on finalize:
- community.general.deploy_helper:
path: /path/to/root
state: clean
keep_releases: 10
# Removing the entire project root folder
- community.general.deploy_helper:
path: /path/to/root
state: absent
# Debugging the facts returned by the module
- community.general.deploy_helper:
path: /path/to/root
- ansible.builtin.debug:
var: deploy_helper