Kafka same consumer group for message consumers as well as stream topology

Question

We run a clustered worker application that relies upon Kafka to consume messages using High-level consumer API. All nodes in a cluster share the same Consumer group. Now what we want is to migrate part of that logic to Kafka stream processor API. What is the approach here? Is stream topology going to fight with existing consumers over messages if assigned the same groupId/clientId? Should we assign different groupId/clientId? to stream topology? say "group." + "background-worker-stream"?

props.put(StreamsConfig.CLIENT_ID_CONFIG, "background-worker");
props.put(ConsumerConfig.GROUP_ID_CONFIG, "group." + "background-worker");

Answer 1

Kafka Streams requires, uses & will insert the application.id property into the group ID. You cannot set the group.id explicitly for the Streams API, AFAIK