[
https://issues.apache.org/jira/browse/KAFKA-8120?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17096659#comment-17096659
]
Jun Wang edited comment on KAFKA-8120 at 4/30/20, 3:37 PM:
-----------------------------------------------------------
Hi
I think [~ChrisEgerton] was right. The current logic of expanding the buffer is
too aggressive. We could just expand the buffer only when the buffer does not
extract a line.
I was able to reproduce similar issue KAFKA-4335 on Kafka version 2.5, I am
working the fix/refactoring FileStreamSourceTask. please assign this to me.
was (Author: wj1918):
Hi
I think [~ChrisEgerton] was right. The current logic of expanding the buffer is
too aggressive. We could just expand the buffer only when the buffer does not
extract a line.
I was able to reproduce similar issue KAFKA-4335, I am working the
fix/refactoring FileStreamSourceTask. please assign this to me.
> Getting NegativeArraySizeException when using Kafka Connect to send data to
> Kafka
> ---------------------------------------------------------------------------------
>
> Key: KAFKA-8120
> URL: https://issues.apache.org/jira/browse/KAFKA-8120
> Project: Kafka
> Issue Type: Bug
> Components: KafkaConnect
> Affects Versions: 2.1.1
> Environment: Ubuntu 16.04 LTS
> Reporter: Prashant Shahi
> Priority: Major
>
>
> I have a large MongoDump JSON which I tried pushing to Kafka using Kafka
> Connect.
> I am getting the following Exception after around 16k messages been pushed.
> After the exception, the program doesn't get killed or exit, but now no more
> messages are pushed.
> {code:java}
> [2019-03-15 08:48:13,812] ERROR WorkerSourceTask{id=od-test18-0} Task threw
> an uncaught and unrecoverable exception
> (org.apache.kafka.connect.runtime.WorkerTask:177)
> java.lang.NegativeArraySizeException at
> org.apache.kafka.connect.file.FileStreamSourceTask.poll(FileStreamSourceTask.java:141)
> at
> org.apache.kafka.connect.runtime.WorkerSourceTask.poll(WorkerSourceTask.java:244)
> at
> org.apache.kafka.connect.runtime.WorkerSourceTask.execute(WorkerSourceTask.java:220)
> at org.apache.kafka.connect.runtime.WorkerTask.doRun(WorkerTask.java:175) at
> org.apache.kafka.connect.runtime.WorkerTask.run(WorkerTask.java:219) 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) [2019-03-15 08:48:13,814] ERROR
> WorkerSourceTask{id=od-test18-0} Task is being killed and will not recover
> until manually restarted
> (org.apache.kafka.connect.runtime.WorkerTask:178){code}
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
