[KAFKA-1120] Controller could miss a broker state change

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.8.1
• Fix Version/s: 2.2.0
• Component/s: core
• Labels: reliability
• Assignee: Zhanxiang (Patrick) Huang
• Reporter: Jun Rao
• Created: 05/Nov/13 18:23
• Updated: 03/Dec/18 06:06
• Resolved: 03/Dec/18 06:06

Description

When the controller is in the middle of processing a task (e.g., preferred leader election, broker change), it holds a controller lock. During this time, a broker could have de-registered and re-registered itself in ZK. After the controller finishes processing the current task, it will start processing the logic in the broker change listener. However, it will see no broker change and therefore won't do anything to the restarted broker. This broker will be in a weird state since the controller doesn't inform it to become the leader of any partition. Yet, the cached metadata in other brokers could still list that broker as the leader for some partitions. Client requests routed to that broker will then get a TopicOrPartitionNotExistException. This broker will continue to be in this bad state until it's restarted again.

Issue Links

• duplicates: KAFKA-1155 Kafka server can miss zookeeper watches during long zkclient callbacks
• duplicates: KAFKA-7235 Use brokerZkNodeVersion to prevent broker from processing outdated controller request
• is related to: KAFKA-5027 Kafka Controller Redesign
• mentioned in: Page Loading...