Playbook Keywords

These are the keywords available on common playbook objects. Keywords are one of several sources for configuring Ansible behavior. See Controlling how Ansible behaves: precedence rules for details on the relative precedence of each source.

Note

Please note:

Aliases for the directives are not reflected here, nor are mutable one. For example, action in task can be substituted by the name of any Ansible module.
The keywords do not have version_added information at this time
Some keywords set defaults for the objects inside of them rather than for the objects themselves

Play
Role
Block
Task

Play

any_errors_fatal: Force any un-handled task errors on any host to propagate to all hosts and end the play.
become: Boolean that controls if privilege escalation is used or not on Task execution.
become_flags: A string of flag(s) to pass to the privilege escalation program when become is True.
become_method: Which method of privilege escalation to use (such as sudo or su).
become_user: User that you ‘become’ after using privilege escalation. The remote/login user must have permissions to become this user.
check_mode: A boolean that controls if a task is executed in ‘check’ mode

See also

Check Mode (“Dry Run”)
collections: UNDOCUMENTED!!
connection: Allows you to change the connection plugin used for tasks to execute on the target.

See also

Using connection plugins
debugger: Enable debugging tasks based on state of the task result. See Playbook Debugger
diff: Toggle to make tasks return ‘diff’ information or not.
environment: A dictionary that gets converted into environment vars to be provided for the task upon execution. This cannot affect Ansible itself nor its configuration, it just sets the variables for the code responsible for executing the task.
fact_path: Set the fact path option for the fact gathering plugin controlled by gather_facts.
force_handlers: Will force notified handler execution for hosts even if they failed during the play. Will not trigger if the play itself fails.
gather_facts: A boolean that controls if the play will automatically run the ‘setup’ task to gather facts for the hosts.
gather_subset: Allows you to pass subset options to the fact gathering plugin controlled by gather_facts.
gather_timeout: Allows you to set the timeout for the fact gathering plugin controlled by gather_facts.
handlers: A section with tasks that are treated as handlers, these won’t get executed normally, only when notified after each section of tasks is complete. A handler’s listen field is not templatable.
hosts: A list of groups, hosts or host pattern that translates into a list of hosts that are the play’s target.
ignore_errors: Boolean that allows you to ignore task failures and continue with play. It does not affect connection errors.
ignore_unreachable: Boolean that allows you to ignore unreachable hosts and continue with play. This does not affect other task errors (see ignore_errors) but is useful for groups of volatile/ephemeral hosts.
max_fail_percentage: can be used to abort the run after a given percentage of hosts in the current batch has failed.
module_defaults: Specifies default parameter values for modules.
name: Identifier. Can be used for documentation, in or tasks/handlers.
no_log: Boolean that controls information disclosure.
order: Controls the sorting of hosts as they are used for executing the play. Possible values are inventory (default), sorted, reverse_sorted, reverse_inventory and shuffle.
port: Used to override the default port used in a connection.
post_tasks: A list of tasks to execute after the tasks section.
pre_tasks: A list of tasks to execute before roles.
remote_user: User used to log into the target via the connection plugin.
roles: List of roles to be imported into the play
run_once: Boolean that will bypass the host loop, forcing the task to attempt to execute on the first host available and afterwards apply any results and facts to all active hosts in the same batch.
serial: Explicitly define how Ansible batches the execution of the current play on the play’s target

See also

Rolling Update Batch Size
strategy: Allows you to choose the connection plugin to use for the play.
tags: Tags applied to the task or included tasks, this allows selecting subsets of tasks from the command line.
tasks: Main list of tasks to execute in the play, they run after roles and before post_tasks.
vars: Dictionary/map of variables
vars_files: List of files that contain vars to include in the play.
vars_prompt: list of variables to prompt for.

Role

any_errors_fatal: Force any un-handled task errors on any host to propagate to all hosts and end the play.
become: Boolean that controls if privilege escalation is used or not on Task execution.
become_flags: A string of flag(s) to pass to the privilege escalation program when become is True.
become_method: Which method of privilege escalation to use (such as sudo or su).
become_user: User that you ‘become’ after using privilege escalation. The remote/login user must have permissions to become this user.
check_mode: A boolean that controls if a task is executed in ‘check’ mode

See also

Check Mode (“Dry Run”)
collections: UNDOCUMENTED!!
connection: Allows you to change the connection plugin used for tasks to execute on the target.

See also

Using connection plugins
debugger: Enable debugging tasks based on state of the task result. See Playbook Debugger
delegate_facts: Boolean that allows you to apply facts to a delegated host instead of inventory_hostname.
delegate_to: Host to execute task instead of the target (inventory_hostname). Connection vars from the delegated host will also be used for the task.
diff: Toggle to make tasks return ‘diff’ information or not.
environment: A dictionary that gets converted into environment vars to be provided for the task upon execution. This cannot affect Ansible itself nor its configuration, it just sets the variables for the code responsible for executing the task.
ignore_errors: Boolean that allows you to ignore task failures and continue with play. It does not affect connection errors.
ignore_unreachable: Boolean that allows you to ignore unreachable hosts and continue with play. This does not affect other task errors (see ignore_errors) but is useful for groups of volatile/ephemeral hosts.
module_defaults: Specifies default parameter values for modules.
name: Identifier. Can be used for documentation, in or tasks/handlers.
no_log: Boolean that controls information disclosure.
port: Used to override the default port used in a connection.
remote_user: User used to log into the target via the connection plugin.
run_once: Boolean that will bypass the host loop, forcing the task to attempt to execute on the first host available and afterwards apply any results and facts to all active hosts in the same batch.
tags: Tags applied to the task or included tasks, this allows selecting subsets of tasks from the command line.
vars: Dictionary/map of variables
when: Conditional expression, determines if an iteration of a task is run or not.

Block

always: List of tasks, in a block, that execute no matter if there is an error in the block or not.
any_errors_fatal: Force any un-handled task errors on any host to propagate to all hosts and end the play.
become: Boolean that controls if privilege escalation is used or not on Task execution.
become_flags: A string of flag(s) to pass to the privilege escalation program when become is True.
become_method: Which method of privilege escalation to use (such as sudo or su).
become_user: User that you ‘become’ after using privilege escalation. The remote/login user must have permissions to become this user.
block: List of tasks in a block.
check_mode: A boolean that controls if a task is executed in ‘check’ mode

See also

Check Mode (“Dry Run”)
collections: UNDOCUMENTED!!
connection: Allows you to change the connection plugin used for tasks to execute on the target.

See also

Using connection plugins
debugger: Enable debugging tasks based on state of the task result. See Playbook Debugger
delegate_facts: Boolean that allows you to apply facts to a delegated host instead of inventory_hostname.
delegate_to: Host to execute task instead of the target (inventory_hostname). Connection vars from the delegated host will also be used for the task.
diff: Toggle to make tasks return ‘diff’ information or not.
environment: A dictionary that gets converted into environment vars to be provided for the task upon execution. This cannot affect Ansible itself nor its configuration, it just sets the variables for the code responsible for executing the task.
ignore_errors: Boolean that allows you to ignore task failures and continue with play. It does not affect connection errors.
ignore_unreachable: Boolean that allows you to ignore unreachable hosts and continue with play. This does not affect other task errors (see ignore_errors) but is useful for groups of volatile/ephemeral hosts.
module_defaults: Specifies default parameter values for modules.
name: Identifier. Can be used for documentation, in or tasks/handlers.
no_log: Boolean that controls information disclosure.
port: Used to override the default port used in a connection.
remote_user: User used to log into the target via the connection plugin.
rescue: List of tasks in a block that run if there is a task error in the main block list.
run_once: Boolean that will bypass the host loop, forcing the task to attempt to execute on the first host available and afterwards apply any results and facts to all active hosts in the same batch.
tags: Tags applied to the task or included tasks, this allows selecting subsets of tasks from the command line.
vars: Dictionary/map of variables
when: Conditional expression, determines if an iteration of a task is run or not.

Task

action: The ‘action’ to execute for a task, it normally translates into a C(module) or action plugin.
any_errors_fatal: Force any un-handled task errors on any host to propagate to all hosts and end the play.
args: A secondary way to add arguments into a task. Takes a dictionary in which keys map to options and values.
async: Run a task asynchronously if the C(action) supports this; value is maximum runtime in seconds.
become: Boolean that controls if privilege escalation is used or not on Task execution.
become_flags: A string of flag(s) to pass to the privilege escalation program when become is True.
become_method: Which method of privilege escalation to use (such as sudo or su).
become_user: User that you ‘become’ after using privilege escalation. The remote/login user must have permissions to become this user.
changed_when: Conditional expression that overrides the task’s normal ‘changed’ status.
check_mode: A boolean that controls if a task is executed in ‘check’ mode

See also

Check Mode (“Dry Run”)
collections: UNDOCUMENTED!!
connection: Allows you to change the connection plugin used for tasks to execute on the target.

See also

Using connection plugins
debugger: Enable debugging tasks based on state of the task result. See Playbook Debugger
delay: Number of seconds to delay between retries. This setting is only used in combination with until.
delegate_facts: Boolean that allows you to apply facts to a delegated host instead of inventory_hostname.
delegate_to: Host to execute task instead of the target (inventory_hostname). Connection vars from the delegated host will also be used for the task.
diff: Toggle to make tasks return ‘diff’ information or not.
environment: A dictionary that gets converted into environment vars to be provided for the task upon execution. This cannot affect Ansible itself nor its configuration, it just sets the variables for the code responsible for executing the task.
failed_when: Conditional expression that overrides the task’s normal ‘failed’ status.
ignore_errors: Boolean that allows you to ignore task failures and continue with play. It does not affect connection errors.
ignore_unreachable: Boolean that allows you to ignore unreachable hosts and continue with play. This does not affect other task errors (see ignore_errors) but is useful for groups of volatile/ephemeral hosts.
local_action: Same as action but also implies delegate_to: localhost
loop: Takes a list for the task to iterate over, saving each list element into the item variable (configurable via loop_control)
loop_control: Several keys here allow you to modify/set loop behaviour in a task.

See also

Adding controls to loops
module_defaults: Specifies default parameter values for modules.
name: Identifier. Can be used for documentation, in or tasks/handlers.
no_log: Boolean that controls information disclosure.
notify: List of handlers to notify when the task returns a ‘changed=True’ status.
poll: Sets the polling interval in seconds for async tasks (default 10s).
port: Used to override the default port used in a connection.
register: Name of variable that will contain task status and module return data.
remote_user: User used to log into the target via the connection plugin.
retries: Number of retries before giving up in a until loop. This setting is only used in combination with until.
run_once: Boolean that will bypass the host loop, forcing the task to attempt to execute on the first host available and afterwards apply any results and facts to all active hosts in the same batch.
tags: Tags applied to the task or included tasks, this allows selecting subsets of tasks from the command line.
until: This keyword implies a ‘retries loop’ that will go on until the condition supplied here is met or we hit the retries limit.
vars: Dictionary/map of variables
when: Conditional expression, determines if an iteration of a task is run or not.
with_<lookup_plugin>: The same as loop but magically adds the output of any lookup plugin to generate the item list.

© 2012–2018 Michael DeHaan
© 2018–2019 Red Hat, Inc.
Licensed under the GNU General Public License version 3.
https://docs.ansible.com/ansible/2.8/reference_appendices/playbooks_keywords.html