Consumption Analytics Documentation

Home > Documentation for older versions > Cloud Cruiser 4 > Collecting, transforming, and publishing > OpenStack via Ceilometer

OpenStack via Ceilometer

Table of contents

The ​OpenStack Ceilometer Collector, introduced in Cloud Cruiser 3.1, collects usage data from OpenStack’s Ceilometer metering API. It uses the Ceilometer V2 Web API, and is compatible with OpenStack versions Grizzly (2013.1) and Havana (2013.2). The collector also uses the Keystone API to collect contextual data unavailable from Ceilometer.

For more information about the Ceilometer V2 Web API, including a list of meters available for collection, see its documentation at

This section contains the following articles:

To set up regular collection and loading of Ceilometer data

  1. Ensure that resources for Ceilometer exist in Cloud Cruiser.
    These resources will be in the OpenStack group. For more information, see Included resources.
  2. Ensure that you have the two provided Ceilometer sample jobs saved as production jobs.
    If you selected the OpenStack Ceilometer collector when installing Cloud Cruiser, the installer created the production jobs ceilometer_default_collect and ceilometer_default_load for you. Otherwise you must save each sample job with a new name. For information about the sample jobs, see Sample jobs.
  3. Create an OpenStack data source.
    See Creating a data source.
  4. In the OpenStack Dashboard, edit the admin project. Add the user from your data source as a member, giving it the following roles:
    • admin
    • _member_
  5. Give the same user the same roles for the services project.
  6. Edit your usage collection job XML to perform the following tasks:
    • Set the usage time zone for the process.
    • If you don’t want to collect all Ceilometer meters, edit the feed configuration to name the specific meters you want to collect. For more information, see Resources.
    • If you want to collect the IDs of tenants, users, or resources, edit the feed configuration to map them. For more information, see Identifiers.
  7. Edit your load job XML to perform the following tasks:
    • Edit the aggregate step to perform the desired aggregation of resources.
    • Edit the transform step to perform the desired transformation of identifiers and resources, such as creating account identifiers compatible with your account structure.
  8. For each resource for which you will charge, set the desired rate.
  9. Schedule both jobs to run daily, with the load job starting thirty minutes after the collection job.
  10. Run the collection job, and then the load job, against all dates prior to the first day for which scheduled job execution will collect that have Ceilometer usage.

Sample jobs

Sample jobs for data collection and data loading are located in the <install_dir>/job_samples directory.

  • ceilometer_sample_collect.xml: Collects Ceilometer usage data to CC Records.
  • ceilometer_sample_load.xml: Processes the CC Records and loads charges into the database, then imports Ceilometer tenants, users, and resources as customer records.

Bean reference

This section lists the beans you need to write batch jobs that collect and process data from Ceilometer and provides pointers to their reference information.


API documentation




  • CeilometerCollector
  • CeilometerFeedConfig
  • ManagedOpenStackDataSource
  • OpenStackDataSourceConfig
Last modified



This page has no classifications.

 (c) Copyright 2017-2020 Hewlett Packard Enterprise Development LP