[jira] [Comment Edited] (HIVE-12947) SMB join in tez has ClassCastException when container reuse is on

["Vikram Dixit K (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HIVE-12947?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15122079#comment-15122079 ] 

Vikram Dixit K edited comment on HIVE-12947 at 1/28/16 6:48 PM:
----------------------------------------------------------------

It is the latter, the plan is modified in the container. It looks like we need to detect that the modification of the plan has already occurred in case of retrieving from cache and not try to re-modify the plan. Alternatively, I could just clear the specific work items from the cache in the same way we put this in the cache.


was (Author: vikram.dixit):
It is the latter, the plan is modified in the container. It looks like we need to detect that the modification of the plan has already occurred in case of retrieving from cache and not try to re-modify the plan.

> 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)