[
https://issues.apache.org/jira/browse/SPARK-40582?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17610161#comment-17610161
]
Yang Jie commented on SPARK-40582:
----------------------------------
Do you use Scala 2.13? [~garretwilson]
> NullPointerException: Cannot invoke
> invalidateSerializedMapOutputStatusCache() because "shuffleStatus" is null
> --------------------------------------------------------------------------------------------------------------
>
> Key: SPARK-40582
> URL: https://issues.apache.org/jira/browse/SPARK-40582
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
> Affects Versions: 3.3.0
> Reporter: Garret Wilson
> Priority: Critical
>
> I'm running a simple little Spark 3.3.0 pipeline on Windows 10 using Java 17
> and UDFs. I hardly do anything interesting, and now when I run the pipeline
> on only 30,000 records I'm getting this:
> {noformat}
> [ERROR] Error in removing shuffle 2
> java.lang.NullPointerException: Cannot invoke
> "org.apache.spark.ShuffleStatus.invalidateSerializedMapOutputStatusCache()"
> because "shuffleStatus" is null
> at
> org.apache.spark.MapOutputTrackerMaster.$anonfun$unregisterShuffle$1(MapOutputTracker.scala:882)
> at
> org.apache.spark.MapOutputTrackerMaster.$anonfun$unregisterShuffle$1$adapted(MapOutputTracker.scala:881)
> at scala.Option.foreach(Option.scala:437)
> at
> org.apache.spark.MapOutputTrackerMaster.unregisterShuffle(MapOutputTracker.scala:881)
> at
> org.apache.spark.storage.BlockManagerStorageEndpoint$$anonfun$receiveAndReply$1.$anonfun$applyOrElse$3(BlockManagerStorageEndpoint.scala:59)
> at
> scala.runtime.java8.JFunction0$mcZ$sp.apply(JFunction0$mcZ$sp.scala:17)
> at
> org.apache.spark.storage.BlockManagerStorageEndpoint.$anonfun$doAsync$1(BlockManagerStorageEndpoint.scala:89)
> at scala.concurrent.Future$.$anonfun$apply$1(Future.scala:678)
> at scala.concurrent.impl.Promise$Transformation.run(Promise.scala:467)
> at
> java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
> at
> java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
> at java.base/java.lang.Thread.run(Thread.java:833)
> {noformat}
> I searched and couldn't find any of the principal terms in the error message.
> Disconcerting that Spark is breaking at what seems to be a fundamental part
> of processing, and with a {{NullPointerException}} at that.
> I have already asked this question on [Stack
> Overflow|https://stackoverflow.com/q/73732970], and even posted a bounty,
> with no solutions. (The only answer so far is from someone who doesn't even
> use Spark and just posted links.)
> _Update:_ Now it just happened with only 1000 records. But then I reran the
> pipeline immediately with no changes, and it succeeded. So this
> {{NullPointerException}} bug is nondeterministic. Not good at all.
> _Update:_ Now it just happened with only 10 records. But then as before I
> reran the pipeline immediately with no changes, and it succeeded.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
