#!/usr/bin/python # -*- coding: utf-8 -*- # (c) 2015, Sebastian Kornehl # GNU General Public License v3.0+ (see COPYING or https://www.gnu.org/licenses/gpl-3.0.txt) from __future__ import absolute_import, division, print_function __metaclass__ = type ANSIBLE_METADATA = {'metadata_version': '1.1', 'status': ['preview'], 'supported_by': 'community'} DOCUMENTATION = ''' --- module: datadog_monitor short_description: Manages Datadog monitors description: - "Manages monitors within Datadog" - "Options like described on http://docs.datadoghq.com/api/" version_added: "2.0" author: "Sebastian Kornehl (@skornehl)" requirements: [datadog] options: api_key: description: ["Your DataDog API key."] required: true app_key: description: ["Your DataDog app key."] required: true state: description: ["The designated state of the monitor."] required: true choices: ['present', 'absent', 'mute', 'unmute'] tags: description: ["A list of tags to associate with your monitor when creating or updating. This can help you categorize and filter monitors."] version_added: "2.2" type: description: - "The type of the monitor." - The 'event alert'is available starting at Ansible 2.1 choices: ['metric alert', 'service check', 'event alert'] query: description: ["The monitor query to notify on with syntax varying depending on what type of monitor you are creating."] name: description: ["The name of the alert."] required: true message: description: - A message to include with notifications for this monitor. Email notifications can be sent to specific users by using the same '@username' notation as events. Monitor message template variables can be accessed by using double square brackets, i.e '[[' and ']]'. silenced: description: ["Dictionary of scopes to timestamps or None. Each scope will be muted until the given POSIX timestamp or forever if the value is None. "] default: "" notify_no_data: description: ["A boolean indicating whether this monitor will notify when data stops reporting.."] type: bool default: 'no' no_data_timeframe: description: - The number of minutes before a monitor will notify when data stops reporting. Must be at least 2x the monitor timeframe for metric alerts or 2 minutes for service checks. required: false default: 2x timeframe for metric, 2 minutes for service timeout_h: description: ["The number of hours of the monitor not reporting data before it will automatically resolve from a triggered state."] renotify_interval: description: - The number of minutes after the last notification before a monitor will re-notify on the current status. It will only re-notify if it's not resolved. escalation_message: description: - A message to include with a re-notification. Supports the '@username' notification we allow elsewhere. Not applicable if renotify_interval is None notify_audit: description: ["A boolean indicating whether tagged users will be notified on changes to this monitor."] type: bool default: 'no' thresholds: description: - A dictionary of thresholds by status. This option is only available for service checks and metric alerts. Because each of them can have multiple thresholds, we don't define them directly in the query."] default: {'ok': 1, 'critical': 1, 'warning': 1} locked: description: ["A boolean indicating whether changes to this monitor should be restricted to the creator or admins."] type: bool default: 'no' version_added: "2.2" require_full_window: description: - A boolean indicating whether this monitor needs a full window of data before it's evaluated. We highly recommend you set this to False for sparse metrics, otherwise some evaluations will be skipped. version_added: "2.3" new_host_delay: description: ["A positive integer representing the number of seconds to wait before evaluating the monitor for new hosts. This gives the host time to fully initialize."] version_added: "2.4" evaluation_delay: description: ["Time to delay evaluation (in seconds). It is effective for sparse values."] required: false default: null version_added: "2.7" id: description: ["The id of the alert. If set, will be used instead of the name to locate the alert."] version_added: "2.3" ''' EXAMPLES = ''' # Create a metric monitor - datadog_monitor: type: "metric alert" name: "Test monitor" state: "present" query: "datadog.agent.up.over('host:host1').last(2).count_by_status()" message: "Host [[host.name]] with IP [[host.ip]] is failing to report to datadog." api_key: "9775a026f1ca7d1c6c5af9d94d9595a4" app_key: "87ce4a24b5553d2e482ea8a8500e71b8ad4554ff" # Deletes a monitor - datadog_monitor: name: "Test monitor" state: "absent" api_key: "9775a026f1ca7d1c6c5af9d94d9595a4" app_key: "87ce4a24b5553d2e482ea8a8500e71b8ad4554ff" # Mutes a monitor - datadog_monitor: name: "Test monitor" state: "mute" silenced: '{"*":None}' api_key: "9775a026f1ca7d1c6c5af9d94d9595a4" app_key: "87ce4a24b5553d2e482ea8a8500e71b8ad4554ff" # Unmutes a monitor - datadog_monitor: name: "Test monitor" state: "unmute" api_key: "9775a026f1ca7d1c6c5af9d94d9595a4" app_key: "87ce4a24b5553d2e482ea8a8500e71b8ad4554ff" ''' import traceback # Import Datadog try: from datadog import initialize, api HAS_DATADOG = True except: HAS_DATADOG = False from ansible.module_utils.basic import AnsibleModule from ansible.module_utils._text import to_native def main(): module = AnsibleModule( argument_spec=dict( api_key=dict(required=True, no_log=True), app_key=dict(required=True, no_log=True), state=dict(required=True, choises=['present', 'absent', 'mute', 'unmute']), type=dict(required=False, choises=['metric alert', 'service check', 'event alert']), name=dict(required=True), query=dict(required=False), message=dict(required=False, default=None), silenced=dict(required=False, default=None, type='dict'), notify_no_data=dict(required=False, default=False, type='bool'), no_data_timeframe=dict(required=False, default=None), timeout_h=dict(required=False, default=None), renotify_interval=dict(required=False, default=None), escalation_message=dict(required=False, default=None), notify_audit=dict(required=False, default=False, type='bool'), thresholds=dict(required=False, type='dict', default=None), tags=dict(required=False, type='list', default=None), locked=dict(required=False, default=False, type='bool'), require_full_window=dict(required=False, default=None, type='bool'), new_host_delay=dict(required=False, default=None), evaluation_delay=dict(required=False, default=None), id=dict(required=False) ) ) # Prepare Datadog if not HAS_DATADOG: module.fail_json(msg='datadogpy required for this module') options = { 'api_key': module.params['api_key'], 'app_key': module.params['app_key'] } initialize(**options) # Check if api_key and app_key is correct or not # if not, then fail here. response = api.Monitor.get_all() if isinstance(response, dict): msg = response.get('errors', None) if msg: module.fail_json(msg="Failed to connect Datadog server using given app_key and api_key : {0}".format(msg[0])) if module.params['state'] == 'present': install_monitor(module) elif module.params['state'] == 'absent': delete_monitor(module) elif module.params['state'] == 'mute': mute_monitor(module) elif module.params['state'] == 'unmute': unmute_monitor(module) def _fix_template_vars(message): if message: return message.replace('[[', '{{').replace(']]', '}}') return message def _get_monitor(module): if module.params['id'] is not None: monitor = api.Monitor.get(module.params['id']) if 'errors' in monitor: module.fail_json(msg="Failed to retrieve monitor with id %s, errors are %s" % (module.params['id'], str(monitor['errors']))) return monitor else: monitors = api.Monitor.get_all() for monitor in monitors: if monitor['name'] == module.params['name']: return monitor return {} def _post_monitor(module, options): try: kwargs = dict(type=module.params['type'], query=module.params['query'], name=module.params['name'], message=_fix_template_vars(module.params['message']), options=options) if module.params['tags'] is not None: kwargs['tags'] = module.params['tags'] msg = api.Monitor.create(**kwargs) if 'errors' in msg: module.fail_json(msg=str(msg['errors'])) else: module.exit_json(changed=True, msg=msg) except Exception as e: module.fail_json(msg=to_native(e), exception=traceback.format_exc()) def _equal_dicts(a, b, ignore_keys): ka = set(a).difference(ignore_keys) kb = set(b).difference(ignore_keys) return ka == kb and all(a[k] == b[k] for k in ka) def _update_monitor(module, monitor, options): try: kwargs = dict(id=monitor['id'], query=module.params['query'], name=module.params['name'], message=_fix_template_vars(module.params['message']), options=options) if module.params['tags'] is not None: kwargs['tags'] = module.params['tags'] msg = api.Monitor.update(**kwargs) if 'errors' in msg: module.fail_json(msg=str(msg['errors'])) elif _equal_dicts(msg, monitor, ['creator', 'overall_state', 'modified', 'matching_downtimes', 'overall_state_modified']): module.exit_json(changed=False, msg=msg) else: module.exit_json(changed=True, msg=msg) except Exception as e: module.fail_json(msg=to_native(e), exception=traceback.format_exc()) def install_monitor(module): options = { "silenced": module.params['silenced'], "notify_no_data": module.boolean(module.params['notify_no_data']), "no_data_timeframe": module.params['no_data_timeframe'], "timeout_h": module.params['timeout_h'], "renotify_interval": module.params['renotify_interval'], "escalation_message": module.params['escalation_message'], "notify_audit": module.boolean(module.params['notify_audit']), "locked": module.boolean(module.params['locked']), "require_full_window": module.params['require_full_window'], "new_host_delay": module.params['new_host_delay'], "evaluation_delay": module.params['evaluation_delay'] } if module.params['type'] == "service check": options["thresholds"] = module.params['thresholds'] or {'ok': 1, 'critical': 1, 'warning': 1} if module.params['type'] == "metric alert" and module.params['thresholds'] is not None: options["thresholds"] = module.params['thresholds'] monitor = _get_monitor(module) if not monitor: _post_monitor(module, options) else: _update_monitor(module, monitor, options) def delete_monitor(module): monitor = _get_monitor(module) if not monitor: module.exit_json(changed=False) try: msg = api.Monitor.delete(monitor['id']) module.exit_json(changed=True, msg=msg) except Exception as e: module.fail_json(msg=to_native(e), exception=traceback.format_exc()) def mute_monitor(module): monitor = _get_monitor(module) if not monitor: module.fail_json(msg="Monitor %s not found!" % module.params['name']) elif monitor['options']['silenced']: module.fail_json(msg="Monitor is already muted. Datadog does not allow to modify muted alerts, consider unmuting it first.") elif (module.params['silenced'] is not None and len(set(monitor['options']['silenced']) ^ set(module.params['silenced'])) == 0): module.exit_json(changed=False) try: if module.params['silenced'] is None or module.params['silenced'] == "": msg = api.Monitor.mute(id=monitor['id']) else: msg = api.Monitor.mute(id=monitor['id'], silenced=module.params['silenced']) module.exit_json(changed=True, msg=msg) except Exception as e: module.fail_json(msg=to_native(e), exception=traceback.format_exc()) def unmute_monitor(module): monitor = _get_monitor(module) if not monitor: module.fail_json(msg="Monitor %s not found!" % module.params['name']) elif not monitor['options']['silenced']: module.exit_json(changed=False) try: msg = api.Monitor.unmute(monitor['id']) module.exit_json(changed=True, msg=msg) except Exception as e: module.fail_json(msg=to_native(e), exception=traceback.format_exc()) if __name__ == '__main__': main()