cotea 1.3.14

Tool that provides Python API to run Ansible programmatically.

cotea

(COntrol Thread Execution Ansible)

cotea is:

Tool that provides Python API to run Ansible programmatically.

cotea allows:

• To control Ansible execution by iterating over the Ansible plays and tasks
• To embed Ansible into another system
• To debug Ansible execution by getting the values of Ansible variables and by retrieving the results of the execution of Ansible tasks/plays

Installation

pip install cotea

Quick start

from cotea.runner import runner
from cotea.arguments_maker import argument_maker

inv_path = "/path/to/inventory"
playbook_path = "/path/to/playbook"

arg_maker = argument_maker()

# setting ansible-playbook flags
arg_maker.add_arg("-i", inv_path)
arg_maker.add_arg("-vvv")

# setting extra-vars
extra_vars = {"SPARK_VERSION": "3.0.2"}
arg_maker.add_arg("--extra-vars", extra_vars)

r = runner(playbook_path, am)

while r.has_next_play():
    current_play = r.get_cur_play_name()
    print("PLAY:", current_play)

    while r.has_next_task():
        next_task = r.get_next_task_name()
        print("\tTASK:", next_task)
            
        r.run_next_task()

r.finish_ansible()

Any argument of the "ansible-playbook" command can be passed by using argument_maker objects. The launch and control of the Ansible is carried out using the runner object. With the help of cotea one can do certain things dynamically at specific Ansible execution points.

Debugging

# imports and object creation...

while r.has_next_play():
    while r.has_next_task():
        task_results = r.run_next_task()
    
        some_task_failed = False
        failed_task = None

        already_ignore_failed = r.get_already_ignore_failed()
        already_ignore_unrch = r.get_already_ignore_unrch()

        for task in task_results:
            task_uuid = task.original_task_uuid
            
            # checking that task didn't have ignore_errors: true
            if task.is_failed:    
                if not task_uuid in already_ignore_failed:
                    some_task_failed = True
                    failed_task = task.task_ansible_object
                    break
            
            # checking that task didn't have ignore_unreachable: true
            if task.is_unreachable:    
                if not task_uuid in already_ignore_unrch:
                    some_task_failed = True
                    failed_task = task.task_ansible_object
                    break
        
        # the task was failed and didn't have
        # ignore_errors or ignore_unreachable flags -
        # going to interactive debugging mod
        if some_task_failed:
            interactive_discotech(failed_task, r)

        r.ignore_errors_of_next_task()


r.finish_ansible()

if r.was_error():
    print("Ansible error was:", r.get_error_msg())

Cotea also contains interactive debugging mode. Using the cotea one can detects that the task was failed in runtime (including the check that task didn't have ignore_errors or ignore_unreachable flag) and call the interactive_discotech() method in that case as shown above. Supported commands:

• 'ft' - print info about the Failed Task
• 'msg' - print all ansible error MSGs (including the ignored ones)
• 'p' - print Progress bar
• 'h'/'help' - print help message
• 're' - RErun of the failed task
• 'v' - add new Variable as extra var
• 'c' - Continue ansible execution (go to the next task)
• 'nt' - add New Task
• 'w' - Watch ansible variable value

A detailed overview of all interfaces is provided in cotea documentation. The full review of cotea debugging mode will be added to documentation soon.