Etude : Différence entre versions

De TARENTINO
Aller à la navigationAller à la recherche
Ligne 184 : Ligne 184 :
 
*** dans /etc/ansible/ansible.cfg
 
*** dans /etc/ansible/ansible.cfg
 
** Contenu
 
** Contenu
 +
<center>
 +
<div style="width:95%; height:50px; top=0; right=10; overflow:auto; border:solid 2px black;">
 +
<syntaxhighlight lang="python" line='line'>
 +
# Example config file for ansible -- https://ansible.com/
 +
# =======================================================
 +
 +
# Nearly all parameters can be overridden in ansible-playbook
 +
# or with command line flags. Ansible will read ANSIBLE_CONFIG,
 +
# ansible.cfg in the current working directory, .ansible.cfg in
 +
# the home directory, or /etc/ansible/ansible.cfg, whichever it
 +
# finds first
 +
 +
# For a full list of available options, run ansible-config list or see the
 +
# documentation: https://docs.ansible.com/ansible/latest/reference_appendices/config.html.
 +
 +
[defaults]
 +
#inventory      = /etc/ansible/hosts
 +
#library        = ~/.ansible/plugins/modules:/usr/share/ansible/plugins/modules
 +
#module_utils    = ~/.ansible/plugins/module_utils:/usr/share/ansible/plugins/module_utils
 +
#remote_tmp      = ~/.ansible/tmp
 +
#local_tmp      = ~/.ansible/tmp
 +
#forks          = 5
 +
#poll_interval  = 0.001
 +
#ask_pass        = False
 +
#transport      = smart
 +
 +
# Plays will gather facts by default, which contain information about
 +
# the remote system.
 +
#
 +
# smart - gather by default, but don't regather if already gathered
 +
# implicit - gather by default, turn off with gather_facts: False
 +
# explicit - do not gather by default, must say gather_facts: True
 +
#gathering = implicit
 +
 +
# This only affects the gathering done by a play's gather_facts directive,
 +
# by default gathering retrieves all facts subsets
 +
# all - gather all subsets
 +
# network - gather min and network facts
 +
# hardware - gather hardware facts (longest facts to retrieve)
 +
# virtual - gather min and virtual facts
 +
# facter - import facts from facter
 +
# ohai - import facts from ohai
 +
# You can combine them using comma (ex: network,virtual)
 +
# You can negate them using ! (ex: !hardware,!facter,!ohai)
 +
# A minimal set of facts is always gathered.
 +
#
 +
#gather_subset = all
 +
 +
# some hardware related facts are collected
 +
# with a maximum timeout of 10 seconds. This
 +
# option lets you increase or decrease that
 +
# timeout to something more suitable for the
 +
# environment.
 +
#
 +
#gather_timeout = 10
 +
 +
# Ansible facts are available inside the ansible_facts.* dictionary
 +
# namespace. This setting maintains the behaviour which was the default prior
 +
# to 2.5, duplicating these variables into the main namespace, each with a
 +
# prefix of 'ansible_'.
 +
# This variable is set to True by default for backwards compatibility. It
 +
# will be changed to a default of 'False' in a future release.
 +
#
 +
#inject_facts_as_vars = True
 +
 +
# Paths to search for collections, colon separated
 +
# collections_paths = ~/.ansible/collections:/usr/share/ansible/collections
 +
 +
# Paths to search for roles, colon separated
 +
#roles_path = ~/.ansible/roles:/usr/share/ansible/roles:/etc/ansible/roles
 +
 +
# Host key checking is enabled by default
 +
#host_key_checking = True
 +
 +
# You can only have one 'stdout' callback type enabled at a time. The default
 +
# is 'default'. The 'yaml' or 'debug' stdout callback plugins are easier to read.
 +
#
 +
#stdout_callback = default
 +
#stdout_callback = yaml
 +
#stdout_callback = debug
 +
 +
 +
# Ansible ships with some plugins that require enabling
 +
# this is done to avoid running all of a type by default.
 +
# These setting lists those that you want enabled for your system.
 +
# Custom plugins should not need this unless plugin author disables them
 +
# by default.
 +
#
 +
# Enable callback plugins, they can output to stdout but cannot be 'stdout' type.
 +
#callback_enabled = timer, mail
 +
 +
# Determine whether includes in tasks and handlers are "static" by
 +
# default. As of 2.0, includes are dynamic by default. Setting these
 +
# values to True will make includes behave more like they did in the
 +
# 1.x versions.
 +
#
 +
#task_includes_static = False
 +
#handler_includes_static = False
 +
 +
# Controls if a missing handler for a notification event is an error or a warning
 +
#error_on_missing_handler = True
 +
 +
# Default timeout for connection plugins
 +
#timeout = 10
 +
 +
# Default user to use for playbooks if user is not specified
 +
# Uses the connection plugin's default, normally the user currently executing Ansible,
 +
# unless a different user is specified here.
 +
#
 +
#remote_user = root
 +
 +
# Logging is off by default unless this path is defined.
 +
#log_path = /var/log/ansible.log
 +
 +
# Default module to use when running ad-hoc commands
 +
#module_name = command
 +
 +
# Use this shell for commands executed under sudo.
 +
# you may need to change this to /bin/bash in rare instances
 +
# if sudo is constrained.
 +
#
 +
#executable = /bin/sh
 +
 +
# By default, variables from roles will be visible in the global variable
 +
# scope. To prevent this, set the following option to True, and only
 +
# tasks and handlers within the role will see the variables there
 +
#
 +
#private_role_vars = False
 +
 +
# List any Jinja2 extensions to enable here.
 +
#jinja2_extensions = jinja2.ext.do,jinja2.ext.i18n
 +
 +
# If set, always use this private key file for authentication, same as
 +
# if passing --private-key to ansible or ansible-playbook
 +
#
 +
#private_key_file = /path/to/file
 +
 +
# If set, configures the path to the Vault password file as an alternative to
 +
# specifying --vault-password-file on the command line. This can also be
 +
# an executable script that returns the vault password to stdout.
 +
#
 +
#vault_password_file = /path/to/vault_password_file
 +
 +
# Format of string {{ ansible_managed }} available within Jinja2
 +
# templates indicates to users editing templates files will be replaced.
 +
# replacing {file}, {host} and {uid} and strftime codes with proper values.
 +
#
 +
#ansible_managed = Ansible managed: {file} modified on %Y-%m-%d %H:%M:%S by {uid} on {host}
 +
 +
# {file}, {host}, {uid}, and the timestamp can all interfere with idempotence
 +
# in some situations so the default is a static string:
 +
#
 +
#ansible_managed = Ansible managed
 +
 +
# By default, ansible-playbook will display "Skipping [host]" if it determines a task
 +
# should not be run on a host. Set this to "False" if you don't want to see these "Skipping"
 +
# messages. NOTE: the task header will still be shown regardless of whether or not the
 +
# task is skipped.
 +
#
 +
#display_skipped_hosts = True
 +
 +
# By default, if a task in a playbook does not include a name: field then
 +
# ansible-playbook will construct a header that includes the task's action but
 +
# not the task's args. This is a security feature because ansible cannot know
 +
# if the *module* considers an argument to be no_log at the time that the
 +
# header is printed. If your environment doesn't have a problem securing
 +
# stdout from ansible-playbook (or you have manually specified no_log in your
 +
# playbook on all of the tasks where you have secret information) then you can
 +
# safely set this to True to get more informative messages.
 +
#
 +
#display_args_to_stdout = False
 +
 +
# Ansible will raise errors when attempting to dereference
 +
# Jinja2 variables that are not set in templates or action lines. Uncomment this line
 +
# to change this behavior.
 +
#
 +
#error_on_undefined_vars = False
 +
 +
# Ansible may display warnings based on the configuration of the
 +
# system running ansible itself. This may include warnings about 3rd party packages or
 +
# other conditions that should be resolved if possible.
 +
# To disable these warnings, set the following value to False:
 +
#
 +
#system_warnings = True
 +
 +
# Ansible may display deprecation warnings for language
 +
# features that should no longer be used and will be removed in future versions.
 +
# To disable these warnings, set the following value to False:
 +
#
 +
#deprecation_warnings = True
 +
 +
# Ansible can optionally warn when usage of the shell and
 +
# command module appear to be simplified by using a default Ansible module
 +
# instead. These warnings can be silenced by adjusting the following
 +
# setting or adding warn=yes or warn=no to the end of the command line
 +
# parameter string. This will for example suggest using the git module
 +
# instead of shelling out to the git command.
 +
#
 +
#command_warnings = False
 +
 +
 +
# set plugin path directories here, separate with colons
 +
#action_plugins    = /usr/share/ansible/plugins/action
 +
#become_plugins    = /usr/share/ansible/plugins/become
 +
#cache_plugins      = /usr/share/ansible/plugins/cache
 +
#callback_plugins  = /usr/share/ansible/plugins/callback
 +
#connection_plugins = /usr/share/ansible/plugins/connection
 +
#lookup_plugins    = /usr/share/ansible/plugins/lookup
 +
#inventory_plugins  = /usr/share/ansible/plugins/inventory
 +
#vars_plugins      = /usr/share/ansible/plugins/vars
 +
#filter_plugins    = /usr/share/ansible/plugins/filter
 +
#test_plugins      = /usr/share/ansible/plugins/test
 +
#terminal_plugins  = /usr/share/ansible/plugins/terminal
 +
#strategy_plugins  = /usr/share/ansible/plugins/strategy
 +
 +
 +
# Ansible will use the 'linear' strategy but you may want to try another one.
 +
#strategy = linear
 +
 +
# By default, callbacks are not loaded for /bin/ansible. Enable this if you
 +
# want, for example, a notification or logging callback to also apply to
 +
# /bin/ansible runs
 +
#
 +
#bin_ansible_callbacks = False
 +
 +
 +
# Don't like cows?  that's unfortunate.
 +
# set to 1 if you don't want cowsay support or export ANSIBLE_NOCOWS=1
 +
#nocows = 1
 +
 +
# Set which cowsay stencil you'd like to use by default. When set to 'random',
 +
# a random stencil will be selected for each task. The selection will be filtered
 +
# against the `cow_enabled` option below.
 +
#
 +
#cow_selection = default
 +
#cow_selection = random
 +
 +
# When using the 'random' option for cowsay, stencils will be restricted to this list.
 +
# it should be formatted as a comma-separated list with no spaces between names.
 +
# NOTE: line continuations here are for formatting purposes only, as the INI parser
 +
#      in python does not support them.
 +
#
 +
#cowsay_enabled_stencils=bud-frogs,bunny,cheese,daemon,default,dragon,elephant-in-snake,elephant,eyes,\
 +
#              hellokitty,kitty,luke-koala,meow,milk,moofasa,moose,ren,sheep,small,stegosaurus,\
 +
#              stimpy,supermilker,three-eyes,turkey,turtle,tux,udder,vader-koala,vader,www
 +
 +
# Don't like colors either?
 +
# set to 1 if you don't want colors, or export ANSIBLE_NOCOLOR=1
 +
#
 +
#nocolor = 1
 +
 +
# If set to a persistent type (not 'memory', for example 'redis') fact values
 +
# from previous runs in Ansible will be stored. This may be useful when
 +
# wanting to use, for example, IP information from one group of servers
 +
# without having to talk to them in the same playbook run to get their
 +
# current IP information.
 +
#
 +
#fact_caching = memory
 +
 +
# This option tells Ansible where to cache facts. The value is plugin dependent.
 +
# For the jsonfile plugin, it should be a path to a local directory.
 +
# For the redis plugin, the value is a host:port:database triplet: fact_caching_connection = localhost:6379:0
 +
#
 +
#fact_caching_connection=/tmp
 +
 +
# retry files
 +
# When a playbook fails a .retry file can be created that will be placed in ~/
 +
# You can enable this feature by setting retry_files_enabled to True
 +
# and you can change the location of the files by setting retry_files_save_path
 +
#
 +
#retry_files_enabled = False
 +
#retry_files_save_path = ~/.ansible-retry
 +
 +
# prevents logging of task data, off by default
 +
#no_log = False
 +
 +
# prevents logging of tasks, but only on the targets, data is still logged on the master/controller
 +
#no_target_syslog = False
 +
 +
# Controls whether Ansible will raise an error or warning if a task has no
 +
# choice but to create world readable temporary files to execute a module on
 +
# the remote machine. This option is False by default for security. Users may
 +
# turn this on to have behaviour more like Ansible prior to 2.1.x. See
 +
# https://docs.ansible.com/ansible/latest/user_guide/become.html#becoming-an-unprivileged-user
 +
# for more secure ways to fix this than enabling this option.
 +
#
 +
#allow_world_readable_tmpfiles = False
 +
 +
# Controls what compression method is used for new-style ansible modules when
 +
# they are sent to the remote system. The compression types depend on having
 +
# support compiled into both the controller's python and the client's python.
 +
# The names should match with the python Zipfile compression types:
 +
# * ZIP_STORED (no compression. available everywhere)
 +
# * ZIP_DEFLATED (uses zlib, the default)
 +
# These values may be set per host via the ansible_module_compression inventory variable.
 +
#
 +
#module_compression = 'ZIP_DEFLATED'
 +
 +
# This controls the cutoff point (in bytes) on --diff for files
 +
# set to 0 for unlimited (RAM may suffer!).
 +
#
 +
#max_diff_size = 104448
 +
 +
# Controls showing custom stats at the end, off by default
 +
#show_custom_stats = False
 +
 +
# Controls which files to ignore when using a directory as inventory with
 +
# possibly multiple sources (both static and dynamic)
 +
#
 +
#inventory_ignore_extensions = ~, .orig, .bak, .ini, .cfg, .retry, .pyc, .pyo
 +
 +
# This family of modules use an alternative execution path optimized for network appliances
 +
# only update this setting if you know how this works, otherwise it can break module execution
 +
#
 +
#network_group_modules=eos, nxos, ios, iosxr, junos, vyos
 +
 +
# When enabled, this option allows lookups (via variables like {{lookup('foo')}} or when used as
 +
# a loop with `with_foo`) to return data that is not marked "unsafe". This means the data may contain
 +
# jinja2 templating language which will be run through the templating engine.
 +
# ENABLING THIS COULD BE A SECURITY RISK
 +
#
 +
#allow_unsafe_lookups = False
 +
 +
# set default errors for all plays
 +
#any_errors_fatal = False
 +
 +
 +
[inventory]
 +
# List of enabled inventory plugins and the order in which they are used.
 +
#enable_plugins = host_list, script, auto, yaml, ini, toml
 +
 +
# Ignore these extensions when parsing a directory as inventory source
 +
#ignore_extensions = .pyc, .pyo, .swp, .bak, ~, .rpm, .md, .txt, ~, .orig, .ini, .cfg, .retry
 +
 +
# ignore files matching these patterns when parsing a directory as inventory source
 +
#ignore_patterns=
 +
 +
# If 'True' unparsed inventory sources become fatal errors, otherwise they are warnings.
 +
#unparsed_is_failed = False
 +
 +
 +
[privilege_escalation]
 +
#become = False
 +
#become_method = sudo
 +
#become_ask_pass = False
 +
 +
 +
## Connection Plugins ##
 +
 +
# Settings for each connection plugin go under a section titled '[[plugin_name]_connection]'
 +
# To view available connection plugins, run ansible-doc -t connection -l
 +
# To view available options for a connection plugin, run ansible-doc -t connection [plugin_name]
 +
# https://docs.ansible.com/ansible/latest/plugins/connection.html
 +
 +
[paramiko_connection]
 +
# uncomment this line to cause the paramiko connection plugin to not record new host
 +
# keys encountered. Increases performance on new host additions. Setting works independently of the
 +
# host key checking setting above.
 +
#record_host_keys=False
 +
 +
# by default, Ansible requests a pseudo-terminal for commands executed under sudo. Uncomment this
 +
# line to disable this behaviour.
 +
#pty = False
 +
 +
# paramiko will default to looking for SSH keys initially when trying to
 +
# authenticate to remote devices. This is a problem for some network devices
 +
# that close the connection after a key failure. Uncomment this line to
 +
# disable the Paramiko look for keys function
 +
#look_for_keys = False
 +
 +
# When using persistent connections with Paramiko, the connection runs in a
 +
# background process. If the host doesn't already have a valid SSH key, by
 +
# default Ansible will prompt to add the host key. This will cause connections
 +
# running in background processes to fail. Uncomment this line to have
 +
# Paramiko automatically add host keys.
 +
#host_key_auto_add = True
 +
 +
 +
[ssh_connection]
 +
# ssh arguments to use
 +
# Leaving off ControlPersist will result in poor performance, so use
 +
# paramiko on older platforms rather than removing it, -C controls compression use
 +
#ssh_args = -C -o ControlMaster=auto -o ControlPersist=60s
 +
 +
# The base directory for the ControlPath sockets.
 +
# This is the "%(directory)s" in the control_path option
 +
#
 +
# Example:
 +
# control_path_dir = /tmp/.ansible/cp
 +
#control_path_dir = ~/.ansible/cp
 +
 +
# The path to use for the ControlPath sockets. This defaults to a hashed string of the hostname,
 +
# port and username (empty string in the config). The hash mitigates a common problem users
 +
# found with long hostnames and the conventional %(directory)s/ansible-ssh-%%h-%%p-%%r format.
 +
# In those cases, a "too long for Unix domain socket" ssh error would occur.
 +
#
 +
# Example:
 +
# control_path = %(directory)s/%%C
 +
#control_path =
 +
 +
# Enabling pipelining reduces the number of SSH operations required to
 +
# execute a module on the remote server. This can result in a significant
 +
# performance improvement when enabled, however when using "sudo:" you must
 +
# first disable 'requiretty' in /etc/sudoers
 +
#
 +
# By default, this option is disabled to preserve compatibility with
 +
# sudoers configurations that have requiretty (the default on many distros).
 +
#
 +
#pipelining = False
 +
 +
# Control the mechanism for transferring files (old)
 +
#  * smart = try sftp and then try scp [default]
 +
#  * True = use scp only
 +
#  * False = use sftp only
 +
#scp_if_ssh = smart
 +
 +
# Control the mechanism for transferring files (new)
 +
# If set, this will override the scp_if_ssh option
 +
#  * sftp  = use sftp to transfer files
 +
#  * scp  = use scp to transfer files
 +
#  * piped = use 'dd' over SSH to transfer files
 +
#  * smart = try sftp, scp, and piped, in that order [default]
 +
#transfer_method = smart
 +
 +
# If False, sftp will not use batch mode to transfer files. This may cause some
 +
# types of file transfer failures impossible to catch however, and should
 +
# only be disabled if your sftp version has problems with batch mode
 +
#sftp_batch_mode = False
 +
 +
# The -tt argument is passed to ssh when pipelining is not enabled because sudo
 +
# requires a tty by default.
 +
#usetty = True
 +
 +
# Number of times to retry an SSH connection to a host, in case of UNREACHABLE.
 +
# For each retry attempt, there is an exponential backoff,
 +
# so after the first attempt there is 1s wait, then 2s, 4s etc. up to 30s (max).
 +
#retries = 3
 +
 +
 +
[persistent_connection]
 +
# Configures the persistent connection timeout value in seconds. This value is
 +
# how long the persistent connection will remain idle before it is destroyed.
 +
# If the connection doesn't receive a request before the timeout value
 +
# expires, the connection is shutdown. The default value is 30 seconds.
 +
#connect_timeout = 30
 +
 +
# The command timeout value defines the amount of time to wait for a command
 +
# or RPC call before timing out. The value for the command timeout must
 +
# be less than the value of the persistent connection idle timeout (connect_timeout)
 +
# The default value is 30 second.
 +
#command_timeout = 30
 +
 +
 +
## Become Plugins ##
 +
 +
# Settings for become plugins go under a section named '[[plugin_name]_become_plugin]'
 +
# To view available become plugins, run ansible-doc -t become -l
 +
# To view available options for a specific plugin, run ansible-doc -t become [plugin_name]
 +
# https://docs.ansible.com/ansible/latest/plugins/become.html
 +
 +
[sudo_become_plugin]
 +
#flags = -H -S -n
 +
#user = root
 +
 +
 +
[selinux]
 +
# file systems that require special treatment when dealing with security context
 +
# the default behaviour that copies the existing context or uses the user default
 +
# needs to be changed to use the file system dependent context.
 +
#special_context_filesystems=fuse,nfs,vboxsf,ramfs,9p,vfat
 +
 +
# Set this to True to allow libvirt_lxc connections to work without SELinux.
 +
#libvirt_lxc_noseclabel = False
 +
 +
 +
[colors]
 +
#highlight = white
 +
#verbose = blue
 +
#warn = bright purple
 +
#error = red
 +
#debug = dark gray
 +
#deprecate = purple
 +
#skip = cyan
 +
#unreachable = red
 +
#ok = green
 +
#changed = yellow
 +
#diff_add = green
 +
#diff_remove = red
 +
#diff_lines = cyan
 +
 +
 +
[diff]
 +
# Always print diff when running ( same as always running with -D/--diff )
 +
#always = False
 +
 +
# Set how many context lines to show in diff
 +
#context = 3
 +
 +
[galaxy]
 +
# Controls whether the display wheel is shown or not
 +
#display_progress=
 +
 +
# Validate TLS certificates for Galaxy server
 +
#ignore_certs = False
 +
 +
# Role or collection skeleton directory to use as a template for
 +
# the init action in ansible-galaxy command
 +
#role_skeleton=
 +
 +
# Patterns of files to ignore inside a Galaxy role or collection
 +
# skeleton directory
 +
#role_skeleton_ignore="^.git$", "^.*/.git_keep$"
 +
 +
# Galaxy Server URL
 +
#server=https://galaxy.ansible.com
 +
 +
# A list of Galaxy servers to use when installing a collection.
 +
#server_list=automation_hub, release_galaxy
 +
 +
# Server specific details which are mentioned in server_list
 +
#[galaxy_server.automation_hub]
 +
#url=https://cloud.redhat.com/api/automation-hub/
 +
#auth_url=https://sso.redhat.com/auth/realms/redhat-external/protocol/openid-connect/token
 +
#token=my_ah_token
 +
#
 +
#[galaxy_server.release_galaxy]
 +
#url=https://galaxy.ansible.com/
 +
#token=my_token
 +
</syntaxhighlight>
 +
</div>
 +
</center>
 
* cli (ligne de commande)
 
* cli (ligne de commande)
  

Version du 14 novembre 2020 à 09:11

Ansible

Introduction

Qu'est ce que Ansible?

  • Outil basé sur du Push
  • Pas d'agent à installer
  • Utilisation de modules (Ansible Galaxie)
  • Inventory / Playbook / Roles
    • Inventory (Vocation à avoir l'inventaire de toutes les machines)
    • Playbook (Faire l'articulation entre Inventory et Roles)
    • Roles (Pour une installation spécifique)
  • Type d'installations
    • Sources
    • Paquets
    • PIP (Librarie Python)
  • Sytème
    • Templating

Notions et définitions

Concepts de base de l'outil

  • Control Node
    • Noeud central contenant Ansible
    • Prérequis: Accès SSH aux autres machines
  • Managed Node
    • Serveurs cibles ayant SSH et Python installés
    • Dont un User avec privilèges Sudo
  • Inventory
    • Inventaire des machines (IP, ou DNS) sous forme de fichier
    • Format INI ou YAML
    • Egalement fichiers de variable (host_vars/group_vars)
    • Inventaire statique (fichiers) ou dynamique (API via scripts)
    • Utilisation de patterns façon Regex
  • Groupes
    • Machines regroupées et typées
      • Création de "boites" dans lesquelles seront classés les serveurs
      • Les boites peuvent être imbriquées et définir des groupes parent et enfant
      • Le groupe Racine (All) est le père de tous
    • Groupes VARS (9:28)
      • Répertoire contenant les variables d'un même groupe contenant un fichier YAML
      • Soit dans le fichier central d'inventory
    • Hosts VARS (10:51)
      • Variables spécifiques à un serveur en particulier
      • Surcharge d'autres variables
  • Exemple d'inventory (12:05)
  • Tasks (13:00)
    • Faire des actions ponctuelles.
      • Créer des users
      • Créer des groupes
      • Passer des commandes
      • Utiliser des modules
      • Copier des fichiers
      • Check de variables
      • etc....
    • Format YAML
  • Modules (13:49)
    • Permet de décrire des actions
    • Modules de base (natifs)
      • Templates
      • Copie de fichiers
      • Création de users
      • etc....
    • Chaque action de module est utilisable par des Tasks
      • Une Task = Une action
      • Une action appelle un module
    • Les actions des modules peuvent fournir des retours
  • Rôles (16:00)
    • Ensemble d'actions
    • L'idéal est de segmenter les rôles
    • Minimiser les rôles
    • Ils sont organisés avec une structure
      • Ils contiennent:
        • Tasks (actions à réaliser)
        • Templates (Fichiers mis à disposition dans les roles)
        • Handlers (Trigger qui va écouter les actions)
        • Variables
        • Meta (Pour le partage des Rôles)
  • Playbooks (19:13)
    • C'est un fichier
Inventory (liste des machines) >> playbook << Rôles à jouer
    • Le playbook fera la correspondance entre quelle machine sur quel groupe avec quelle variable, tel rôle s'appliquera
    • Peut contenir des variables mais à éviter.
    • Peut contenir des tasks mais à éviter.
    • Peut contenir des conditions mais à éviter.
  • Plugins (20:50)
    • Augmenter les capacités de Ansible

LES INSTALLATIONS (PIP, DEPOTS ET A SAVOIR)

Installation de Ansible

  • Machines
    • Controller Node
      • Python >= 2.7
      • Pas de Windows
      • SSH
    • Managed Node
      • Python >= 2.6
  • Installations (2:00)
    • Binaire
      • git clone https://github.com/ansible/ansible.git
      • cd ansible
      • source ./hacking/env-setup
      • sudo apt install python
      • pip install --user -r ./requirements.txt
      • echo "127.0.0.1" > ~/ansible_hosts
      • export ANSIBLE_INVENTORY=~/ansible_hosts
      • ansible all -m ping --ask-pass
    • PIP
      • sudo apt install python3-pip
      • pip3 install ansible (ou pip3 install ansible==2.2.9) (4:22)
    • Par la distribution (5:30)
      • sudo apt install
  • Utilisation (5:00)
    • Se reconnecter
    • Versions
      • ansible --version
      • cat /etc/*ea* (6:32)
    • Précisions et configuration(6:54)
      • sudo launchctl limit maxfiles unlimited (sur MacOS)
      • Indiquer l'endroit de l'interpréteur Python3 (/usr/bin/python3)
      • Installer Python sur la machine distante
        • ansible myhost --become -m raw -a "yum install -y python3"

SSH : CLEFS ET ASTUCES

Permettre aux Nodes de communiquer

  • Clé privée et clé publique
    • La clé privée ne doit pas être diffusée
    • La clé publique a vocation à être déployée sur les serveurs distants
  • Générer les clefs
    • ssh-keygen -t ecdsa -b 521
      • va générer une clef privée et publiquesans extension ou .pem
      • droits en 600
  • Transférer la clé
    • ssh-copy-id $USER@$SERVEUR
    • Sécuriser éventuellement (5:45)
    • Création d'un .ssh/config (8:40)

FICHIER CFG : CONFIGURATION ET TUNING

Fichier de configuration de Ansible

  • ansible.cfg
    • Permet de configurer la manière dont ansible va se jouer
    • Si le fichier ansible.cfg doit se trouver à un autre endroit, il faut l'indiquer dans la variable ANSIBLE_CONFIG
      • A l'endroit du playbook (4:16)
      • dans le HOME/.ansible/ansible.cfg
      • dans /etc/ansible/ansible.cfg
    • Contenu
  1 # Example config file for ansible -- https://ansible.com/
  2 # =======================================================
  3 
  4 # Nearly all parameters can be overridden in ansible-playbook
  5 # or with command line flags. Ansible will read ANSIBLE_CONFIG,
  6 # ansible.cfg in the current working directory, .ansible.cfg in
  7 # the home directory, or /etc/ansible/ansible.cfg, whichever it
  8 # finds first
  9 
 10 # For a full list of available options, run ansible-config list or see the
 11 # documentation: https://docs.ansible.com/ansible/latest/reference_appendices/config.html.
 12 
 13 [defaults]
 14 #inventory       = /etc/ansible/hosts
 15 #library         = ~/.ansible/plugins/modules:/usr/share/ansible/plugins/modules
 16 #module_utils    = ~/.ansible/plugins/module_utils:/usr/share/ansible/plugins/module_utils
 17 #remote_tmp      = ~/.ansible/tmp
 18 #local_tmp       = ~/.ansible/tmp
 19 #forks           = 5
 20 #poll_interval   = 0.001
 21 #ask_pass        = False
 22 #transport       = smart
 23 
 24 # Plays will gather facts by default, which contain information about
 25 # the remote system.
 26 #
 27 # smart - gather by default, but don't regather if already gathered
 28 # implicit - gather by default, turn off with gather_facts: False
 29 # explicit - do not gather by default, must say gather_facts: True
 30 #gathering = implicit
 31 
 32 # This only affects the gathering done by a play's gather_facts directive,
 33 # by default gathering retrieves all facts subsets
 34 # all - gather all subsets
 35 # network - gather min and network facts
 36 # hardware - gather hardware facts (longest facts to retrieve)
 37 # virtual - gather min and virtual facts
 38 # facter - import facts from facter
 39 # ohai - import facts from ohai
 40 # You can combine them using comma (ex: network,virtual)
 41 # You can negate them using ! (ex: !hardware,!facter,!ohai)
 42 # A minimal set of facts is always gathered.
 43 #
 44 #gather_subset = all
 45 
 46 # some hardware related facts are collected
 47 # with a maximum timeout of 10 seconds. This
 48 # option lets you increase or decrease that
 49 # timeout to something more suitable for the
 50 # environment.
 51 #
 52 #gather_timeout = 10
 53 
 54 # Ansible facts are available inside the ansible_facts.* dictionary
 55 # namespace. This setting maintains the behaviour which was the default prior
 56 # to 2.5, duplicating these variables into the main namespace, each with a
 57 # prefix of 'ansible_'.
 58 # This variable is set to True by default for backwards compatibility. It
 59 # will be changed to a default of 'False' in a future release.
 60 #
 61 #inject_facts_as_vars = True
 62 
 63 # Paths to search for collections, colon separated
 64 # collections_paths = ~/.ansible/collections:/usr/share/ansible/collections
 65 
 66 # Paths to search for roles, colon separated
 67 #roles_path = ~/.ansible/roles:/usr/share/ansible/roles:/etc/ansible/roles
 68 
 69 # Host key checking is enabled by default
 70 #host_key_checking = True
 71 
 72 # You can only have one 'stdout' callback type enabled at a time. The default
 73 # is 'default'. The 'yaml' or 'debug' stdout callback plugins are easier to read.
 74 #
 75 #stdout_callback = default
 76 #stdout_callback = yaml
 77 #stdout_callback = debug
 78 
 79 
 80 # Ansible ships with some plugins that require enabling
 81 # this is done to avoid running all of a type by default.
 82 # These setting lists those that you want enabled for your system.
 83 # Custom plugins should not need this unless plugin author disables them
 84 # by default.
 85 #
 86 # Enable callback plugins, they can output to stdout but cannot be 'stdout' type.
 87 #callback_enabled = timer, mail
 88 
 89 # Determine whether includes in tasks and handlers are "static" by
 90 # default. As of 2.0, includes are dynamic by default. Setting these
 91 # values to True will make includes behave more like they did in the
 92 # 1.x versions.
 93 #
 94 #task_includes_static = False
 95 #handler_includes_static = False
 96 
 97 # Controls if a missing handler for a notification event is an error or a warning
 98 #error_on_missing_handler = True
 99 
100 # Default timeout for connection plugins
101 #timeout = 10
102 
103 # Default user to use for playbooks if user is not specified
104 # Uses the connection plugin's default, normally the user currently executing Ansible,
105 # unless a different user is specified here.
106 #
107 #remote_user = root
108 
109 # Logging is off by default unless this path is defined.
110 #log_path = /var/log/ansible.log
111 
112 # Default module to use when running ad-hoc commands
113 #module_name = command
114 
115 # Use this shell for commands executed under sudo.
116 # you may need to change this to /bin/bash in rare instances
117 # if sudo is constrained.
118 #
119 #executable = /bin/sh
120 
121 # By default, variables from roles will be visible in the global variable
122 # scope. To prevent this, set the following option to True, and only
123 # tasks and handlers within the role will see the variables there
124 #
125 #private_role_vars = False
126 
127 # List any Jinja2 extensions to enable here.
128 #jinja2_extensions = jinja2.ext.do,jinja2.ext.i18n
129 
130 # If set, always use this private key file for authentication, same as
131 # if passing --private-key to ansible or ansible-playbook
132 #
133 #private_key_file = /path/to/file
134 
135 # If set, configures the path to the Vault password file as an alternative to
136 # specifying --vault-password-file on the command line. This can also be
137 # an executable script that returns the vault password to stdout.
138 #
139 #vault_password_file = /path/to/vault_password_file
140 
141 # Format of string {{ ansible_managed }} available within Jinja2
142 # templates indicates to users editing templates files will be replaced.
143 # replacing {file}, {host} and {uid} and strftime codes with proper values.
144 #
145 #ansible_managed = Ansible managed: {file} modified on %Y-%m-%d %H:%M:%S by {uid} on {host}
146 
147 # {file}, {host}, {uid}, and the timestamp can all interfere with idempotence
148 # in some situations so the default is a static string:
149 #
150 #ansible_managed = Ansible managed
151 
152 # By default, ansible-playbook will display "Skipping [host]" if it determines a task
153 # should not be run on a host. Set this to "False" if you don't want to see these "Skipping"
154 # messages. NOTE: the task header will still be shown regardless of whether or not the
155 # task is skipped.
156 #
157 #display_skipped_hosts = True
158 
159 # By default, if a task in a playbook does not include a name: field then
160 # ansible-playbook will construct a header that includes the task's action but
161 # not the task's args. This is a security feature because ansible cannot know
162 # if the *module* considers an argument to be no_log at the time that the
163 # header is printed. If your environment doesn't have a problem securing
164 # stdout from ansible-playbook (or you have manually specified no_log in your
165 # playbook on all of the tasks where you have secret information) then you can
166 # safely set this to True to get more informative messages.
167 #
168 #display_args_to_stdout = False
169 
170 # Ansible will raise errors when attempting to dereference
171 # Jinja2 variables that are not set in templates or action lines. Uncomment this line
172 # to change this behavior.
173 #
174 #error_on_undefined_vars = False
175 
176 # Ansible may display warnings based on the configuration of the
177 # system running ansible itself. This may include warnings about 3rd party packages or
178 # other conditions that should be resolved if possible.
179 # To disable these warnings, set the following value to False:
180 #
181 #system_warnings = True
182 
183 # Ansible may display deprecation warnings for language
184 # features that should no longer be used and will be removed in future versions.
185 # To disable these warnings, set the following value to False:
186 #
187 #deprecation_warnings = True
188 
189 # Ansible can optionally warn when usage of the shell and
190 # command module appear to be simplified by using a default Ansible module
191 # instead. These warnings can be silenced by adjusting the following
192 # setting or adding warn=yes or warn=no to the end of the command line
193 # parameter string. This will for example suggest using the git module
194 # instead of shelling out to the git command.
195 #
196 #command_warnings = False
197 
198 
199 # set plugin path directories here, separate with colons
200 #action_plugins     = /usr/share/ansible/plugins/action
201 #become_plugins     = /usr/share/ansible/plugins/become
202 #cache_plugins      = /usr/share/ansible/plugins/cache
203 #callback_plugins   = /usr/share/ansible/plugins/callback
204 #connection_plugins = /usr/share/ansible/plugins/connection
205 #lookup_plugins     = /usr/share/ansible/plugins/lookup
206 #inventory_plugins  = /usr/share/ansible/plugins/inventory
207 #vars_plugins       = /usr/share/ansible/plugins/vars
208 #filter_plugins     = /usr/share/ansible/plugins/filter
209 #test_plugins       = /usr/share/ansible/plugins/test
210 #terminal_plugins   = /usr/share/ansible/plugins/terminal
211 #strategy_plugins   = /usr/share/ansible/plugins/strategy
212 
213 
214 # Ansible will use the 'linear' strategy but you may want to try another one.
215 #strategy = linear
216 
217 # By default, callbacks are not loaded for /bin/ansible. Enable this if you
218 # want, for example, a notification or logging callback to also apply to
219 # /bin/ansible runs
220 #
221 #bin_ansible_callbacks = False
222 
223 
224 # Don't like cows?  that's unfortunate.
225 # set to 1 if you don't want cowsay support or export ANSIBLE_NOCOWS=1
226 #nocows = 1
227 
228 # Set which cowsay stencil you'd like to use by default. When set to 'random',
229 # a random stencil will be selected for each task. The selection will be filtered
230 # against the `cow_enabled` option below.
231 #
232 #cow_selection = default
233 #cow_selection = random
234 
235 # When using the 'random' option for cowsay, stencils will be restricted to this list.
236 # it should be formatted as a comma-separated list with no spaces between names.
237 # NOTE: line continuations here are for formatting purposes only, as the INI parser
238 #       in python does not support them.
239 #
240 #cowsay_enabled_stencils=bud-frogs,bunny,cheese,daemon,default,dragon,elephant-in-snake,elephant,eyes,\
241 #              hellokitty,kitty,luke-koala,meow,milk,moofasa,moose,ren,sheep,small,stegosaurus,\
242 #              stimpy,supermilker,three-eyes,turkey,turtle,tux,udder,vader-koala,vader,www
243 
244 # Don't like colors either?
245 # set to 1 if you don't want colors, or export ANSIBLE_NOCOLOR=1
246 #
247 #nocolor = 1
248 
249 # If set to a persistent type (not 'memory', for example 'redis') fact values
250 # from previous runs in Ansible will be stored. This may be useful when
251 # wanting to use, for example, IP information from one group of servers
252 # without having to talk to them in the same playbook run to get their
253 # current IP information.
254 #
255 #fact_caching = memory
256 
257 # This option tells Ansible where to cache facts. The value is plugin dependent.
258 # For the jsonfile plugin, it should be a path to a local directory.
259 # For the redis plugin, the value is a host:port:database triplet: fact_caching_connection = localhost:6379:0
260 #
261 #fact_caching_connection=/tmp
262 
263 # retry files
264 # When a playbook fails a .retry file can be created that will be placed in ~/
265 # You can enable this feature by setting retry_files_enabled to True
266 # and you can change the location of the files by setting retry_files_save_path
267 #
268 #retry_files_enabled = False
269 #retry_files_save_path = ~/.ansible-retry
270 
271 # prevents logging of task data, off by default
272 #no_log = False
273 
274 # prevents logging of tasks, but only on the targets, data is still logged on the master/controller
275 #no_target_syslog = False
276 
277 # Controls whether Ansible will raise an error or warning if a task has no
278 # choice but to create world readable temporary files to execute a module on
279 # the remote machine. This option is False by default for security. Users may
280 # turn this on to have behaviour more like Ansible prior to 2.1.x. See
281 # https://docs.ansible.com/ansible/latest/user_guide/become.html#becoming-an-unprivileged-user
282 # for more secure ways to fix this than enabling this option.
283 #
284 #allow_world_readable_tmpfiles = False
285 
286 # Controls what compression method is used for new-style ansible modules when
287 # they are sent to the remote system. The compression types depend on having
288 # support compiled into both the controller's python and the client's python.
289 # The names should match with the python Zipfile compression types:
290 # * ZIP_STORED (no compression. available everywhere)
291 # * ZIP_DEFLATED (uses zlib, the default)
292 # These values may be set per host via the ansible_module_compression inventory variable.
293 #
294 #module_compression = 'ZIP_DEFLATED'
295 
296 # This controls the cutoff point (in bytes) on --diff for files
297 # set to 0 for unlimited (RAM may suffer!).
298 #
299 #max_diff_size = 104448
300 
301 # Controls showing custom stats at the end, off by default
302 #show_custom_stats = False
303 
304 # Controls which files to ignore when using a directory as inventory with
305 # possibly multiple sources (both static and dynamic)
306 #
307 #inventory_ignore_extensions = ~, .orig, .bak, .ini, .cfg, .retry, .pyc, .pyo
308 
309 # This family of modules use an alternative execution path optimized for network appliances
310 # only update this setting if you know how this works, otherwise it can break module execution
311 #
312 #network_group_modules=eos, nxos, ios, iosxr, junos, vyos
313 
314 # When enabled, this option allows lookups (via variables like {{lookup('foo')}} or when used as
315 # a loop with `with_foo`) to return data that is not marked "unsafe". This means the data may contain
316 # jinja2 templating language which will be run through the templating engine.
317 # ENABLING THIS COULD BE A SECURITY RISK
318 #
319 #allow_unsafe_lookups = False
320 
321 # set default errors for all plays
322 #any_errors_fatal = False
323 
324 
325 [inventory]
326 # List of enabled inventory plugins and the order in which they are used.
327 #enable_plugins = host_list, script, auto, yaml, ini, toml
328 
329 # Ignore these extensions when parsing a directory as inventory source
330 #ignore_extensions = .pyc, .pyo, .swp, .bak, ~, .rpm, .md, .txt, ~, .orig, .ini, .cfg, .retry
331 
332 # ignore files matching these patterns when parsing a directory as inventory source
333 #ignore_patterns=
334 
335 # If 'True' unparsed inventory sources become fatal errors, otherwise they are warnings.
336 #unparsed_is_failed = False
337 
338 
339 [privilege_escalation]
340 #become = False
341 #become_method = sudo
342 #become_ask_pass = False
343 
344 
345 ## Connection Plugins ##
346 
347 # Settings for each connection plugin go under a section titled '[[plugin_name]_connection]'
348 # To view available connection plugins, run ansible-doc -t connection -l
349 # To view available options for a connection plugin, run ansible-doc -t connection [plugin_name]
350 # https://docs.ansible.com/ansible/latest/plugins/connection.html
351 
352 [paramiko_connection]
353 # uncomment this line to cause the paramiko connection plugin to not record new host
354 # keys encountered. Increases performance on new host additions. Setting works independently of the
355 # host key checking setting above.
356 #record_host_keys=False
357 
358 # by default, Ansible requests a pseudo-terminal for commands executed under sudo. Uncomment this
359 # line to disable this behaviour.
360 #pty = False
361 
362 # paramiko will default to looking for SSH keys initially when trying to
363 # authenticate to remote devices. This is a problem for some network devices
364 # that close the connection after a key failure. Uncomment this line to
365 # disable the Paramiko look for keys function
366 #look_for_keys = False
367 
368 # When using persistent connections with Paramiko, the connection runs in a
369 # background process. If the host doesn't already have a valid SSH key, by
370 # default Ansible will prompt to add the host key. This will cause connections
371 # running in background processes to fail. Uncomment this line to have
372 # Paramiko automatically add host keys.
373 #host_key_auto_add = True
374 
375 
376 [ssh_connection]
377 # ssh arguments to use
378 # Leaving off ControlPersist will result in poor performance, so use
379 # paramiko on older platforms rather than removing it, -C controls compression use
380 #ssh_args = -C -o ControlMaster=auto -o ControlPersist=60s
381 
382 # The base directory for the ControlPath sockets.
383 # This is the "%(directory)s" in the control_path option
384 #
385 # Example:
386 # control_path_dir = /tmp/.ansible/cp
387 #control_path_dir = ~/.ansible/cp
388 
389 # The path to use for the ControlPath sockets. This defaults to a hashed string of the hostname,
390 # port and username (empty string in the config). The hash mitigates a common problem users
391 # found with long hostnames and the conventional %(directory)s/ansible-ssh-%%h-%%p-%%r format.
392 # In those cases, a "too long for Unix domain socket" ssh error would occur.
393 #
394 # Example:
395 # control_path = %(directory)s/%%C
396 #control_path =
397 
398 # Enabling pipelining reduces the number of SSH operations required to
399 # execute a module on the remote server. This can result in a significant
400 # performance improvement when enabled, however when using "sudo:" you must
401 # first disable 'requiretty' in /etc/sudoers
402 #
403 # By default, this option is disabled to preserve compatibility with
404 # sudoers configurations that have requiretty (the default on many distros).
405 #
406 #pipelining = False
407 
408 # Control the mechanism for transferring files (old)
409 #   * smart = try sftp and then try scp [default]
410 #   * True = use scp only
411 #   * False = use sftp only
412 #scp_if_ssh = smart
413 
414 # Control the mechanism for transferring files (new)
415 # If set, this will override the scp_if_ssh option
416 #   * sftp  = use sftp to transfer files
417 #   * scp   = use scp to transfer files
418 #   * piped = use 'dd' over SSH to transfer files
419 #   * smart = try sftp, scp, and piped, in that order [default]
420 #transfer_method = smart
421 
422 # If False, sftp will not use batch mode to transfer files. This may cause some
423 # types of file transfer failures impossible to catch however, and should
424 # only be disabled if your sftp version has problems with batch mode
425 #sftp_batch_mode = False
426 
427 # The -tt argument is passed to ssh when pipelining is not enabled because sudo
428 # requires a tty by default.
429 #usetty = True
430 
431 # Number of times to retry an SSH connection to a host, in case of UNREACHABLE.
432 # For each retry attempt, there is an exponential backoff,
433 # so after the first attempt there is 1s wait, then 2s, 4s etc. up to 30s (max).
434 #retries = 3
435 
436 
437 [persistent_connection]
438 # Configures the persistent connection timeout value in seconds. This value is
439 # how long the persistent connection will remain idle before it is destroyed.
440 # If the connection doesn't receive a request before the timeout value
441 # expires, the connection is shutdown. The default value is 30 seconds.
442 #connect_timeout = 30
443 
444 # The command timeout value defines the amount of time to wait for a command
445 # or RPC call before timing out. The value for the command timeout must
446 # be less than the value of the persistent connection idle timeout (connect_timeout)
447 # The default value is 30 second.
448 #command_timeout = 30
449 
450 
451 ## Become Plugins ##
452 
453 # Settings for become plugins go under a section named '[[plugin_name]_become_plugin]'
454 # To view available become plugins, run ansible-doc -t become -l
455 # To view available options for a specific plugin, run ansible-doc -t become [plugin_name]
456 # https://docs.ansible.com/ansible/latest/plugins/become.html
457 
458 [sudo_become_plugin]
459 #flags = -H -S -n
460 #user = root
461 
462 
463 [selinux]
464 # file systems that require special treatment when dealing with security context
465 # the default behaviour that copies the existing context or uses the user default
466 # needs to be changed to use the file system dependent context.
467 #special_context_filesystems=fuse,nfs,vboxsf,ramfs,9p,vfat
468 
469 # Set this to True to allow libvirt_lxc connections to work without SELinux.
470 #libvirt_lxc_noseclabel = False
471 
472 
473 [colors]
474 #highlight = white
475 #verbose = blue
476 #warn = bright purple
477 #error = red
478 #debug = dark gray
479 #deprecate = purple
480 #skip = cyan
481 #unreachable = red
482 #ok = green
483 #changed = yellow
484 #diff_add = green
485 #diff_remove = red
486 #diff_lines = cyan
487 
488 
489 [diff]
490 # Always print diff when running ( same as always running with -D/--diff )
491 #always = False
492 
493 # Set how many context lines to show in diff
494 #context = 3
495 
496 [galaxy]
497 # Controls whether the display wheel is shown or not
498 #display_progress=
499 
500 # Validate TLS certificates for Galaxy server
501 #ignore_certs = False
502 
503 # Role or collection skeleton directory to use as a template for
504 # the init action in ansible-galaxy command
505 #role_skeleton=
506 
507 # Patterns of files to ignore inside a Galaxy role or collection
508 # skeleton directory
509 #role_skeleton_ignore="^.git$", "^.*/.git_keep$"
510 
511 # Galaxy Server URL
512 #server=https://galaxy.ansible.com
513 
514 # A list of Galaxy servers to use when installing a collection.
515 #server_list=automation_hub, release_galaxy
516 
517 # Server specific details which are mentioned in server_list
518 #[galaxy_server.automation_hub]
519 #url=https://cloud.redhat.com/api/automation-hub/
520 #auth_url=https://sso.redhat.com/auth/realms/redhat-external/protocol/openid-connect/token
521 #token=my_ah_token
522 #
523 #[galaxy_server.release_galaxy]
524 #url=https://galaxy.ansible.com/
525 #token=my_token
  • cli (ligne de commande)
Récupérée de « https://tarentino.fr/index.php?title=Etude&oldid=2430 »