Home

Kafka Event Collector is designed to expose events happening inside a Kafka Cluster as events in a Kafka Topics. The main purpose if for these events to be used by other applications that would like to use changes of Kafka components happen.

Once this applications is started Kafka Admin Client API is used to query state of different components: Kafka Topics, Cluster and Nodes.

Current State is stored in memory and events are published in a Kafka Topic (__collector by default). This events are the same used to recreate state once the application is restarted.

Each component has a different frequency to query the state, defined by a poll interval. Currently, this is defined for Topics and for Cluster details.

The first time the application is started, as state is empty, events of components created are published in the event topic. The same events are consumed by the application and used to define the state. If the query results are different from the state, events of updated/deleted components are published.