Azure Event Hubs Listener

Most recent version: v4.0.0

Learn more about this Listener type in this article.

v4.0.0

Released on 05/07/2026

• New Message Format parameter to control how incoming messages are parsed.

• Then Split Strategy and Custom Delimiter fields are removed and are replaced with Message Format and Format Options.

• Parse failures now route to the Error port instead of being silently dropped.

• Each emitted event now carries the original partition, offset, and sequence metadata.

• Messages with an empty body after decompression are now skipped instead of emitting an event with an empty field.

v3.1.0

Released on 04/21/2026

• Improved connection strings validation and error handling.

• Resolved an issue preventing the Listener from reading events without batch mode.

v3.0.0

Released on 03/11/2026

• To enhance performance and resilience, we highly recommend enabling the Client Checkpointing option in the Listener configuration now. If it's not enabled, you must set the environment variable in your distributor with Docker Compose using SINGLETON_LISTENER_EXECUTOR=true. For Listener versions before v3.0.0, you must use the AZURE_EVENTHUB_LISTENER_EXECUTION_ENABLED environment variable, even if the Client Checkpointing option is enabled.

• Fixed the event routing to the error port under certain conditions.

v2.0.0

Released on 11/06/2025

Added Starting Position options. The starting position defines where event processing begins and can be set to a specific timestamp, sequence number, or predefined position such as the beginning or end of the stream. This setting is crucial for controlling event replay and recovery scenarios.

v1.0.1

Released on 10/09/2025

In order to use this Listener, you must activate the Environment Variable in your distributor using docker compose (AZURE_EVENTHUB_LISTENER_EXECUTION_ENABLED)

v1.0.0

Released on 10/02/2025

• New consumer group.

• Added reserved fields: _timestamp, _error, and _worker_reception_timestamp.

• Added the checkpoint store and partition load balancing functionalities. Checkpoint Store: A persistent storage mechanism that records processing progress in data streams, enabling applications to resume from the last successful point after failures, preventing data loss and duplicate processing.

  Partition Load Balancing: The automated distribution of data partitions across multiple processing nodes to ensure even workload distribution, maximize throughput, and maintain system performance when nodes join, leave, or fail.

v0.0.2

Released on 07/11/2025

• The maximum number of receivers allowed is now limited to 1 per partition in your Event Hub.

• There's a new variable (AZURE_EVENTHUB_LISTENER_EXECUTION_ENABLED) to indicate the distributors in which the Listener can be executed.

v0.0.1

Released on 09/20/2024

Initial version