 |
Consumption Analytics Documentation |
|
Consumption Analytics Documentation |
The feed configuration for Ceilometer has just two properties: meterNames , which lists the Ceilometer meters to collect as Cloud Cruiser resources, and the outputs property common to all collectors that maps data to resources and identifiers.
To collect all meters, set the meterNames property to an asterisk (*). To collect only specified meters, provide their names in a comma-delimited list. For example:
<property name="meterNames" value="instance:m1.nano,disk.read.bytes,cpu_util"/>
The Ceilometer documentation has a list of available meters at http://docs.openstack.org/developer/ceilometer/measurements.html. You do not need to add any mappings to the outputs property of the feed configuration for meters you specify. The sample job contains a mapping that handles all resources.
NOTE: If OpenStack adds meters to Ceilometer that did not exist at the time Cloud Cruiser 3.1 was released, they are collected when you set meterNames to *. However, you must create corresponding resources in Cloud Cruiser to be able to load charges against these meters.
Each usage record collected has common elements that provide context about the usage. The sample collect job maps some of these as identifiers, but you can add others to suit your requirements. The following table shows all the available elements, including identifier names for those mapped by the sample job:
| Name in Ceilometer API | Name as identifier in sample collect job | Description |
|---|---|---|
|
|
| The name of the project, also known as a tenant , associated with the usage |
|
| Not mapped | The ID of the project associated with the usage |
|
|
| The name of the user associated with the usage |
|
| Not mapped | The ID of the user associated with the usage |
|
|
| The unique name of the resource associated with the usage |
|
|
| The human-readable display name for the resource associated with the usage |
|
| Not mapped | The ID of the resource associated with the usage |