Building a Deploy¶

Layout¶

• *.py - files that describe deploys
• inventories/*.py - files that describe different inventories
• group_data/*.py - files that describe data for groups
• templates/*.jn2 - templates used in the deploys
• roles/*.py - files that describe role specific deploys
• files/* - files used in the deploys
• config.py - optional config and hooks

Inventory¶

Inventory files contain groups of hosts. Groups are defined as ALL_CAPS variables assigned to a list of hosts. For example, this inventory creates two groups, “app_servers” and “db_servers”:

# inventories/production.py

APP_SERVERS = [
    'app-1.net',
    'app-2.net'
]

DB_SERVERS = [
    'db-1.net',
    'db-2.net',
    'db-3.net'
]

In addition to the groups defined in the inventory, all the hosts are added to two more groups: “all” and the name of the inventory file, in this case “production”. Both can be overriden by defining them in the inventory.

Data¶

Data allows you to separate deploy variables from the deploy script. With data per host and per group, you can easily build deploys that satisfy multiple environments.

# inventories/production.py

APP_SERVERS = [
    'app-1.net',
    ('app-2.net', {'some_key': True})
]

# group_data/production.py

app_user = 'myuser'
app_dir = '/opt/myapp'

ssh_port = 22
ssh_user = 'ubuntu'
ssh_key = '~/.ssh/some_key'
ssh_key_password = 'password for key'
# ssh_password = 'password auth is bad'

# group_data/all.py

env = 'dev'
git_repo = 'https://github.com/Fizzadar/pyinfra'

# group_data/production.py

env = 'production'
git_branch = 'master'

# group_data/staging.py

env = 'staging'
git_branch = 'develop'

Operations¶

Now that you’ve got an inventory of hosts and know how to auth with them, you can start writing the deploy. This is described in a Python file normally situated in the top level of the deploy directory.

In this file, eg deploy.py, you import pyinfra modules. Each of these contains a number of operations. You call these operations inside the deploy file, with arguments describing remote state, and pyinfra uses this to run the deploy.

For example, this deploy will ensure that user “pyinfra” exists with home directory “/home/pyinfra”, and that the “/var/log/pyinfra.log” file exists and is owned by that user.

# deploy.py

# Import pyinfra modules, each containing operations to use
from pyinfra.modules import server, files

# Ensure the state of a user
server.user(
    'pyinfra',
    home='/home/pyinfra'
)

# Ensure the state of files
files.file(
    '/var/log/pyinfra.log',
    user='pyinfra',
    group='pyinfra',
    permissions='644',
    sudo=True
)

Uses the server module and files module. You can see all the modules in the modules index.

Naming operations:

Pass a set object as the first argument to name the operation, which will appear during a deploy. By default the operation module, name and arguments are shown:

server.user(
    {'Ensure user pyinfra'},  # the contents of the set will become the op name
    'pyinfra',
    home='/home/pyinfra'
)

from pyinfra import host
from pyinfra.modules import server

# Ensure the state of a user based on host/group data
server.user(
    host.data.app_user,
    home=host.data.app_dir
)

from pyinfra import host
from pyinfra.modules import server

server.user(
    host.data.app_user,
    '/opt/{{ host.data.app_dir }}'  # for multiple values of host.data.app_dir we still
                                    # generate a single operation
)

from pyinfra.modules import server

# Run an operation, collecting its meta output
meta = server.user(
    'myuser'
)

# If we added a user above, do something extra
if meta.commands:
    server.shell('# add server to sudo, etc...')

from pyinfra import host
from pyinfra.modules import apt, yum

if host.fact.linux_distribution == 'CentOS':
    yum.packages(
        'nano',
        sudo=True
    )
else:
    apt.packages(
        'nano',
        sudo=True
    )

from pyinfra import local, inventory

# Operations in this file will be added to all hosts
local.include('roles/my_role.py')

# Operations in this file will be added to the hosts in group "my_group"
local.include('roles/limited_role.py', hosts=inventory.my_group)

Config¶

There are a number of configuration options for how deploys are managed. These can be defined at the top of a deploy file, or in a config.py alongside the deploy file. See the full list of options & defaults.

# config.py or top of deploy.py

# SSH connect timeout
TIMEOUT = 1

# Fail the entire deploy after 10% of hosts fail
FAIL_PERCENT = 10

config.py advantage:

When added to config.py, these options will take affect when using pyinfra --fact or --run.

Hooks¶

Deploy hooks are executed by the CLI at various points during the deploy process. Like config, they can be defined in a config.py or at the top of the deploy file:

• before_connect
• before_facts
• before_deploy
• after_deploy

These can be used, for example, to check the right branch before connecting or to build some clientside assets locally before fact gathering. Hooks all take data, state as arguments:

# config.py or top of deploy.py

from pyinfra import hook

@hook.before_connect
def my_callback(data, state):
    print('Before connect hook!')

To abort a deploy, a hook can raise a hook.Error which the CLI will handle.

When executing commands locally inside a hook (ie webpack build), you should always use the pyinfra.local module:

@hook.before_connect
def my_callback(data, state):
    # Check something local is correct, etc
    branch = local.shell('git rev-parse --abbrev-ref HEAD')
    app_branch = data.app_branch

    if branch != app_branch:
        # Raise HookError for pyinfra to handle
        raise HookError('We\'re on the wrong branch (want {0}, got {1})!'.format(
            branch, app_branch
        ))