π₯ Consumer Groups & Rebalancing
How Kafka Distributes Work β and What Happens When Things Change
A consumer group is a set of consumers that cooperate to consume a topic.
Each partition is assigned to exactly one consumer in the group β this is how Kafka parallelizes consumption.
Consumers track their progress via committed offsets stored in the internal
__consumer_offsets topic. When consumers join, leave, or crash, Kafka triggers a
rebalance to redistribute partitions.
π Partition β Consumer Assignment
Adjust partitions, consumers, and strategy to see how assignments change in real-time.
Range Assignor (default): Partitions are divided by range per topic. Each consumer gets a contiguous block. Uneven splits give extra partitions to earlier consumers.
π Offset Tracking
Each consumer commits its position per partition to __consumer_offsets.
Click "Consume Messages" to advance consumer positions
π Rebalancing Deep Dive
Step through a rebalance event to see eager vs. cooperative protocols.
Ready Choose a trigger and click "Start Rebalance"
Consumer Lag During Rebalance
0 msgs
πͺοΈ The Rebalance Storm
When a slow consumer triggers cascading rebalances.