Kafka (Read) block

The kafka_topic_read block pulls data from a Kafka topic. It acts as the source block in a Connect+ dataflow, connecting to a Kafka server and consuming messages from one or more predefined topics. The block supports configurable consumer group IDs, header extraction as data attributes for routing and tracking, and a unique ID for end-to-end message tracing across the dataflow.

When to use this block

Use this block when your dataflow sources data from a Kafka topic, such as streaming event data or real-time transaction feeds.

Prerequisites

Before configuring this block, make sure you have:

The name of the Kafka topic or topics to consume from
Access to the appropriate Kafka client service

Standard properties

Field name	Required	Description
Block name	Yes	A name for the block instance. The name must be alphanumeric. There is no character limit.
Kafka topics comma separated	Yes	The Kafka topic names to consume messages from. Enter multiple topics separated by commas.
Kafka client service to connect to server	Yes	The Kafka client service used to connect to the server. Select `Event-notification-kafka` or `connectplus-kafka service`. Default value: `Event-notification-kafka`.

Advanced properties

⚠️ Make changes to advanced properties only if you know what you are doing.

Field name	Required	Description
Consumer Group Id	No	The consumer group ID used to identify this block's consumer within the Kafka cluster. If you do not provide a value, the system uses a default value derived from the expression `${neoDataflowId}`.
Headers as Attribute	No	Specifies which Kafka message headers are extracted as data attributes and made available to downstream blocks. Select `All` to extract all headers, or `Traceable Id` to extract only the traceable identifier. Default value: `All`.
Unique Id for tracing	No	The unique identifier used to trace a message through the dataflow. If you do not provide a value, the system uses a default value derived from the expression `${kafka.key}_${kafka.offset}_${kafka.partition}_${kafka.topic}`.