Major Releases

This page notes all enhancements and features that we have introduced to our major releases of the Axon Kafka Extension.

Release 4.0

Release 4.0 has seen several release candidates:

Release 4.0 - Release Candidate 3

We solidified the API of the Kafka extension with the following main points:

• The original implementation only allowed users to use this extension as a streamable message source. However, as Kafka provides a lot of internal logic to segment, start and stop a stream, making it a subscribing solution is feasible. Issue #17 thus introduces a SubscribingKafkaMessageSource. This adjustment makes the Kafka extension a viable solution for the TrackingEventProcessor and the SubscribingEventProcessor. Furthermore, it allows the user to choose Axon's logic of partitioning or Kafka's.

• Contributor zambrovski did a tremendous job enhancing this extension's configuration in pull request #11. On top of that, he included a sample application showing how you can use the Kafka extension.

• When using this extension as a streaming source in Axon, it automatically constructed KafkaTrackingTokens. These tokens were, however, unaware of the topic they stored the progress for. Issue #28 addressed the problem, which we resolved after this in pull request #29.

For a complete list of all the adjustments, we refer to the release notes.

Note that this is still a release candidate. As such, users should consider we might introduce API changes in future releases.

Release 4.0 - Release Candidate 2

We introduced several minor API changes in this version. We released it to provide users a window of opportunity to further verify the current implementation

Note that this is still a release candidate. As such, users should consider we might introduce API changes in future releases.

Release 4.0 - Release Candidate 1

We split off the Kafka logic from Axon Framework core into a dedicated repository. Next to that, it complies with Axon Framework's 4.0 release.

Note that we adjusted the Kafka package into a release candidate. As such, users should consider we might introduce API changes in future releases.