Telemetry best practices

Telemetry best practices

The following are some suggested best practices to follow when deploying and configuring the Telemetry service. The best practices are divided into data collection and storage.

Data collection

  1. The Telemetry service collects a continuously growing set of data. Not all the data will be relevant for an administrator to monitor.

    • Based on your needs, you can edit the pipeline.yaml configuration file to include a selected number of meters while disregarding the rest. Similarly, in Ocata, you will need to edit polling.yaml to define which meters to generate.

    • By default, Telemetry service polls the service APIs every 10 minutes. You can change the polling interval on a per meter basis by editing the polling.yaml configuration file.

      Note

      Prior to Ocata, the polling configuration was handled by pipeline.yaml

      Warning

      If the polling interval is too short, it will likely increase the stress on the service APIs.

    • Expand the configuration to have greater control over different meter intervals. For more information, see the Pipeline configuration.

  2. You can delay or adjust polling requests by enabling the jitter support. This adds a random delay on how the polling agents send requests to the service APIs. To enable jitter, set shuffle_time_before_polling_task in the ceilometer.conf configuration file to an integer greater than 0.

  3. If polling many resources or at a high frequency, you can add additional central and compute agents as necessary. The agents are designed to scale horizontally. For more information see, Support for HA deployment.

Data storage

Note

As of Newton, data storage is not recommended in ceilometer. Alarm, metric, and event data should be stored in aodh, gnocchi, and panko respectively. The following details only relate to ceilometer’s legacy API.

  1. We recommend that you avoid open-ended queries. In order to get better performance you can use reasonable time ranges and/or other query constraints for retrieving measurements.

    For example, this open-ended query might return an unpredictable amount of data:

    $ ceilometer sample-list --meter cpu -q 'resource_id=INSTANCE_ID_1'
    

    Whereas, this well-formed query returns a more reasonable amount of data, hence better performance:

    $ ceilometer sample-list --meter cpu -q 'resource_id=INSTANCE_ID_1;timestamp > 2015-05-01T00:00:00;timestamp < 2015-06-01T00:00:00'
    

    Note

    The number of items returned will be restricted to the value defined by default_api_return_limit in the ceilometer.conf configuration file. Alternatively, the value can be set per query by passing limit option in request.

  2. We recommend that you install the API behind mod_wsgi, as it provides more settings to tweak, like threads and processes in case of WSGIDaemon.

    Note

    For more information on how to configure mod_wsgi, see the Telemetry Install Documentation.

  3. The collection service provided by the Telemetry project is not intended to be an archival service. Set a Time to Live (TTL) value to expire data and minimize the database size. If you would like to keep your data for longer time period, you may consider storing it in a data warehouse outside of Telemetry.

    Note

    For more information on how to set the TTL, see Storing samples.

  4. We recommend that you do not run MongoDB on the same node as the controller. Keep it on a separate node optimized for fast storage for better performance. Also it is advisable for the MongoDB node to have a lot of memory.

    Note

    For more information on how much memory you need, see MongoDB FAQ.

  5. Use replica sets in MongoDB. Replica sets provide high availability through automatic failover. If your primary node fails, MongoDB will elect a secondary node to replace the primary node, and your cluster will remain functional.

    For more information on replica sets, see the MongoDB replica sets docs.

  6. Use sharding in MongoDB. Sharding helps in storing data records across multiple machines and is the MongoDB’s approach to meet the demands of data growth.

    For more information on sharding, see the MongoDB sharding docs.

Creative Commons Attribution 3.0 License

Except where otherwise noted, this document is licensed under Creative Commons Attribution 3.0 License. See all OpenStack Legal Documents.