Guided scenarios
Step-by-step exercises — send your first message, observe key-based partitioning, compare acks modes, trigger a rebalance — with objectives, explanations and knowledge checks.
A visual Kafka cluster you can safely break. Stop brokers and watch leader election happen. Send keyed messages and see exactly which partition they land on. Slow a consumer down and watch lag grow. Every significant event comes with a cause-and-effect explanation: what you changed, how Kafka responded, why, and what it would mean in production.
Step-by-step exercises — send your first message, observe key-based partitioning, compare acks modes, trigger a rebalance — with objectives, explanations and knowledge checks.
The unrestricted cluster: full producer configuration, broker controls, consumer group management, offset inspection, unclean election toggles and an X-ray view of the wire protocol.
Incident-style drills: broker crashes, network partitions, disk full, poison messages, rebalance storms — with symptoms, root causes, fixes and prevention. Available now.
Use the Cluster tab to stop or isolate brokers and watch leader election. Use Consumer to trigger a rebalance, or run a guided scenario above for a narrated failure sequence.
Stop a broker to trigger leader election, isolate one to see a network partition, drag a lag slider past 5s to shrink the ISR, or send a message with acks=all while ISR is degraded. Every action explains itself here.