In this section, we provide general information about the CDF output. Since our recommended approach is to output to Kafka, we provide JSON examples of the output in a dedicated Kafka section here.

Topics

A topic is a collection of fields from one of the CDF’s datastreams. The CDF publishes messages to the data transport solution for the following topics:

CDF-I

  • Agent Events
    Correlated and uncorrelatable Agent events.

  • Intervals
    Timings between agent events.

  • Associations
    Links between correlated items.

  • Alerts
    Where a threshold is breached, e.g. latencies. Thresholds are configured in VMX-Explorer.

  • Timeseries
    Time series events.

  • Configuration
    Labels and descriptions for the above data types.

Items (the consolidated, correlated objects) are not sent via the CDF, because they can be inferred by grouping all Agent Events with a common correlation ID.

CDF-M

CDF-M topics are created to meet requirement since they depend on the type of Consumer subscribing to them.

CDF-T

CDF-T topics are created to meet requirement since they depend on the type of Consumer subscribing to them.

To discuss CDF-M and CDF-T topics, please contact your Beek's Account Manager.

Further information on CDF-T and CDF-M

CDF-T and CDF-M are streamed out of VMX-Capture and are extremely flexible and configurable. We include example message formats within this document. To gain a better understanding of how CDF-T and CDF-M can be configured within VMX-Capture, refer to the Configuration Guide for VMX-Capture.

  • CDF-T streams output from VMX-Capture Stat Collectors.

  • CDF-M streams output from VMX-Capture Message Collectors.

Output Message Bundling

 The Core Data Feed message bundles together zero or more of each of the Topics. This grouping is purely for transmission efficiency and does not indicate the entities are related in any way. See Output Format for Kafka section of this document for example output message formats for different topics.