The ceilometer.pipeline Module

class ceilometer.pipeline.Pipeline(cfg, publisher_manager, transformer_manager)[source]

Bases: object

Sample handling pipeline

Pipeline describes a chain of handlers. The chain starts with tranformer and ends with one or more publishers.

The first transformer in the chain gets counter from data collector, i.e. pollster or notification handler, takes some action like dropping, aggregation, changing field etc, then passes the updated counter to next step.

The subsequent transformers, if any, handle the data similarly.

In the end of the chain, publishers publish the data. The exact publishing method depends on publisher type, for example, pushing into data storage through message bus, sending to external CW software through CW API call.

If no transformer is included in the chain, the publishers get counters from data collector and publish them directly.

flush(ctxt, source)[source]

Flush data after all counter have been injected to pipeline.

get_interval()[source]
publish_counter(ctxt, counter, source)[source]
publish_counters(ctxt, counters, source)[source]
support_counter(counter_name)[source]
exception ceilometer.pipeline.PipelineException(message, pipeline_cfg)[source]

Bases: exceptions.Exception

class ceilometer.pipeline.PipelineManager(cfg, publisher_manager)[source]

Bases: object

Pipeline Manager

Pipeline manager sets up pipelines according to config file

Usually only one pipeline manager exists in the system.

publisher(context, source)[source]

Build a new Publisher for these manager pipelines.

Parameters:
  • context – The context.
  • source – Counter source.
class ceilometer.pipeline.PublishContext(context, source, pipelines=[])[source]

Bases: object

add_pipelines(pipelines)[source]
class ceilometer.pipeline.TransformerExtensionManager[source]

Bases: stevedore.extension.ExtensionManager

get_ext(name)[source]

Previous topic

The ceilometer.openstack.common.version Module

Next topic

The ceilometer.plugin Module

This Page