pypingdom
pypingdom copied to clipboard
Python library for interact with Pingdom services (REST API and maintenance windows).
Pypingdom
.. image:: https://img.shields.io/pypi/v/pypingdom.svg :target: https://pypi.python.org/pypi/pypingdom
.. image:: https://img.shields.io/pypi/l/pypingdom.svg :target: https://pypi.python.org/pypi/pypingdom
.. image:: https://img.shields.io/pypi/pyversions/pypingdom.svg :target: https://pypi.python.org/pypi/pypingdom
.. image:: https://travis-ci.org/sekipaolo/pypingdom.svg?branch=master :target: https://travis-ci.org/sekipaolo/pypingdom
Python library for interacting with Pingdom services (REST API and maintenance windows).
Features
- Support for
Multi-User Authentication <https://www.pingdom.com/resources/api#multi-user+authentication>
_ - Check management: create, delete, update, list
- Maintenance windows: create, delete, list
- Fetching outage summaries
.. warning::
Since the Pingdom REST API don't support maintenance windows, we interact
with the Website for it. Therefore this feature is highly fragile and can
*break* at any moment due to frontend changes on the Pingdom website.
Requirements
- Pingdom account
- requests (0.10.8 or newer)
Installation
.. code-block:: python
pip install pypingdom
Usage
The client
object will allow you to interact both with the REST API and the
GUI (for maintenance windows).
.. code-block:: python
>>> import pypingdom
>>> client = pypingdom.Client(username="[email protected]",
password="your_password",
apikey="your_api_key",
email="your_email")
the email
parameter is required for Multiuser Authentication <https://www.pingdom.com/resources/api#multi-user+authentication>
_.
Checks
Since Pingdom does not treat the check name as identifier (as we probably want
to do) the client object will retrieve the check list from the API and cache it
as a dictionary ( check_name => check_instance). You can access it through the
checks
attribute:
.. code-block:: python
>>> client.checks["my awesome check"]
pingdom.Check <1895866>
autoresolve: 0
alert_policy: 2118909
name: example_com
created: 1448565930
lasterrortime: 1489325292
resolution: 1
lastresponsetime: 558
lasttesttime: 1489847772
alert_policy_name: Production Systems
paused: False
host: hostname.example.com
acktimeout: 0
ipv6: False
use_legacy_notifications: False
type: http
tags: []
a better way to retrieve a check would be:
.. code-block:: python
>>> client.get_check("my awesome check")
that will return None if the check doesn't exists
List checks with production
and frontend
tags:
.. code-block:: python
>>> client.get_checks(filters={"tags": ["production", "frontend"]})
Create a check:
.. code-block:: python
>>> check_definition = {
"name": "My awesome check",
"paused": True,
"alert_policy": 201745,
"type": "http",
"host": "www.google.com",
"url": "/",
"requestheaders": {
'XCustom': 'my header value'
},
"tags": [{"name": "pypingdom-test"}, {"name": "custom-tag"}],
"encryption": False
}
>>> client.create_check(check_definition)
Refers to this page <https://www.pingdom.com/resources/api#MethodCreate+New+Check>
_ for the list of options.
When you create or modify a check some related entity need to be referenced by id:
Integrations
To enable/disable an integration plugins (like webhooks) use the field integrationids
(array with integer ids to set or "null" tring to remove it)
Alert policies
To bind an alerting policy use the field alert_policy
(numeric id to set it or string "null" to disable alerts)
Update a check:
.. code-block:: python
>>> client.update_check(check, {"paused": True})
this will return True if an effective change was sent to the API and False otherwise (useful for idempotent usage, like ansible modules)
Delete a check:
.. code-block:: python
>>> client.delete_check(check)
Maintenance windows
Retreive maintenance windows for production websites in the last 7 days:
.. code-block:: python
>>> import datetime
>>> checks = client.get_checks(filters={"tags": ["production", "frontend"]})
>>> start = datetime.datetime.now() - datetime.timedelta(days=7)
>>> client.get_maintenances(filters={"checks": checks, "after": start})
Create a 1 hour maintenance window for production websites:
.. code-block:: python
>>> start = datetime.datetime.now() + datetime.timedelta(minutes=10)
>>> end = start + datetime.timedelta(hours=1)
>>> window = client.create_maintenance({"checks": checks, "name": "pypingdom test maintenance", "start": start, "stop": end})
Delete future maintenance windows:
.. code-block:: python
>>> windows = client.get_maintenances(filters={"checks": checks, "after": datetime.datetime.now()}):
>>> for m in maintenances:
client.delete_maintenance(m)
Reporting/summary
Retrieve average response time and uptime summaries:
.. code-block:: python
>>> checkid = client.get_check("my awesome check")._id
>>> start = int(time.time()) - 30*24*60*60 # 30 days back
>>> end = time.time()
>>> client.get_summary_average(checkid, start, end, include_uptime="true")