[
https://issues.apache.org/jira/browse/HIVE-12947?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15122118#comment-15122118
]
Sergey Shelukhin commented on HIVE-12947:
-----------------------------------------
Would it be possible to not cache the plan under these conditions, or avoid
modifying it, rather than nuking the entire cache? The assumption behind plan
cache is that plan should not be modified.
> SMB join in tez has ClassCastException when container reuse is on
> -----------------------------------------------------------------
>
> Key: HIVE-12947
> URL: https://issues.apache.org/jira/browse/HIVE-12947
> Project: Hive
> Issue Type: Bug
> Components: Tez
> Affects Versions: 2.0.0
> Reporter: Vikram Dixit K
> Assignee: Vikram Dixit K
> Priority: Critical
> Attachments: HIVE-12947.1.patch
>
>
> SMB join in tez has multiple work items that are connected based on input tag
> followed by input initialization etc. In case of container re-use, what ends
> up happening is that we try to reconnect the work items and fail. If we try
> to work around that issue by recognizing somehow that the cache was in play,
> we will run into other initialization issues with respect to record readers.
> So the plan is to disable caching of the SMB work items by clearing out
> during the close phase.
> {code}
> java.lang.RuntimeException: Map operator initialization failed
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.init(MapRecordProcessor.java:247)
> at
> org.apache.hadoop.hive.ql.exec.tez.TezProcessor.initializeAndRunProcessor(TezProcessor.java:147)
> at
> org.apache.hadoop.hive.ql.exec.tez.TezProcessor.run(TezProcessor.java:137)
> at
> org.apache.tez.runtime.LogicalIOProcessorRuntimeTask.run(LogicalIOProcessorRuntimeTask.java:344)
> at
> org.apache.tez.runtime.task.TezTaskRunner$TaskRunnerCallable$1.run(TezTaskRunner.java:179)
> at
> org.apache.tez.runtime.task.TezTaskRunner$TaskRunnerCallable$1.run(TezTaskRunner.java:171)
> at java.security.AccessController.doPrivileged(Native Method)
> at javax.security.auth.Subject.doAs(Subject.java:422)
> at
> org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1657)
> at
> org.apache.tez.runtime.task.TezTaskRunner$TaskRunnerCallable.callInternal(TezTaskRunner.java:171)
> at
> org.apache.tez.runtime.task.TezTaskRunner$TaskRunnerCallable.callInternal(TezTaskRunner.java:167)
> at org.apache.tez.common.CallableWithNdc.call(CallableWithNdc.java:36)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> at java.lang.Thread.run(Thread.java:745)
> Caused by: java.lang.ClassCastException:
> org.apache.hadoop.hive.ql.exec.FileSinkOperator cannot be cast to
> org.apache.hadoop.hive.ql.exec.DummyStoreOperator
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.getJoinParentOp(MapRecordProcessor.java:300)
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.getJoinParentOp(MapRecordProcessor.java:302)
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.getJoinParentOp(MapRecordProcessor.java:302)
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.getJoinParentOp(MapRecordProcessor.java:302)
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.getJoinParentOp(MapRecordProcessor.java:302)
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.init(MapRecordProcessor.java:189)
> ... 15 more
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
