Introduction

Monitors broker-level cluster changes during an experiment. This check observes whether specific broker events occur — such as a controller election or a broker going offline — and evaluates them against an expected change list.

This action targets brokers, not topics or consumer groups. For topic partition changes (leader election, replica changes), use Check Partitions instead. For consumer group state monitoring, use Check Consumer State.

Use Cases

• Verify that a controller election happens when the active controller broker is taken down.
• Confirm that broker downtime is detected when a broker pod is killed or network-isolated.
• Assert that no unexpected broker-level disruptions occur during a topic-level fault injection.

Parameters

Parameter	Description	Default
Duration	How long the check runs. The check continuously polls broker state for this duration.	30s
Expected Changes	(optional) Which broker-level changes to watch for. Valid values: New Controller Elected (kafka controller changed), Broker downtime (kafka broker with downtime). If left empty, any broker change triggers.
Change Check Mode	How the expected changes are evaluated. All the time (allTheTime): every poll must detect the change. At least once (atLeastOnce): the change must be observed at least once during the duration.	allTheTime