Manual Triggers
##############################

Robusta automates tasks in Kubernetes clusters with a library of re-usable actions.

These actions can be triggered automatically (:ref:`when supported triggers occur <Triggers>`) or on-demand
(manually triggered) by running a CLI command. Most actions support both types of triggers.

We'll focus below on actions which are most useful when triggered manually.

Example Manual Triggers
------------------------------

Here are some high-level tasks included in Robusta's library of actions:

1. `Run a chaos engineering scenario <https://github.com/robusta-dev/robusta-chaos>`_
2. :ref:`Debug a python pod with VSCode <Python debugger>`
3. :ref:`See why a pod is using high CPU by running a profiler <Python profiler>`
4. :ref:`Find memory leaks in python applications <Python memory>`
5. Stress-test a pod over HTTP
6. Show recent OOM kills

All these tasks can be done without Robusta, but they're annoying to do. We automated them for you.

How to use manual triggers
----------------------------------
Let's manually run the ``python_profiler`` playbook:

.. code-block:: bash

    robusta playbooks trigger python_profiler name=robusta-runner-8f4558f9b-pcbj9 namespace=default seconds=5

The result will be sent to the sink. Here is an example result in Slack:

.. image:: /images/python-profiler.png
  :width: 600
  :align: center

How manual triggers work
----------------------------------
Like every Robusta action, manual playbooks are just re-usable Python functions.

The above ``python_profiler`` is a built-in Robusta action. It takes a Pod as input and can be run automatically:

.. code-block:: yaml

    customPlaybooks:
    - triggers:
        - on_prometheus_alert:
            alert_name:  "HighCPU"
      actions:
        - python_profiler:
            seconds: 5

Or it can be run manually like we did here.

Full reference
---------------------------------

Most playbooks can be triggered manually as follows:

.. code-block:: bash

    robusta playbooks trigger <action_name> name=<name> namespace=<namespace> kind=<kind> <key>=<value>

The parameters above are:

name
    The name of a Kubernetes resource

namespace
    The resource's namespace

kind
    ``pod``, ``deployment``, or any other resource the action supports. This can be left out for playbooks that support
    one input type.

<key>=<value>
    Any additional parameters the action needs