[KAFKA-7692] updateFirstUnstableOffset NPE due to sequenceId overflow in ProducerStateManager.append

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.0.0
• Fix Version/s: 1.1.2, 2.2.0, 2.1.1, 2.0.2
• Component/s: core
• Labels: easyfix
• Assignee: Unassigned
• Reporter: Ming Liu
• Reviewer: Jason Gustafson
• Created: 02/Dec/18 20:25
• Updated: 25/Jan/19 01:01
• Resolved: 25/Jan/19 00:39

Description

When operating Kafka, we frequently saw this exception on Kafka server log, 

ERROR [KafkaServer id=19] Fatal error during KafkaServer startup. Prepare to shutdown (kafka.server.KafkaServer)
 java.lang.NullPointerException
     at kafka.log.Log.kafka$log$Log$$updateFirstUnstableOffset(Log.scala:792)
     at kafka.log.Log.kafka$log$Log$$loadProducerState(Log.scala:518)
     at kafka.log.Log.<init>(Log.scala:228)
     at kafka.log.Log$.apply(Log.scala:1747)
     at kafka.log.LogManager.kafka$log$LogManager$$loadLog(LogManager.scala:255)
     at kafka.log.LogManager$$anonfun$loadLogs$2$$anonfun$11$$anonfun$apply$15$$anonfun$apply$2.apply$mcV$sp(LogManager.scala:335)
     at kafka.utils.CoreUtils$$anon$1.run(CoreUtils.scala:62)
     at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
     at java.util.concurrent.FutureTask.run(FutureTask.java:266)
     at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
     at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
     at java.lang.Thread.run(Thread.java:748)

 

 

Issue Links

• links to: GitHub Pull Request #5990