Processing All Unread Data

When you start a pipeline for the first time, the MapR Streams Consumer becomes a new consumer group for the topic. It reads only incoming data, processing data from all partitions, and ignores any existing data in the topic by default.

To read all unread data in the topic, add the auto.offset.reset Kafka configuration property and set it to earliest. For more information about this property, see the MapR Streams documentation.