[jira] [Created] (TEZ-4071) shuffle throws exceptions with an external table with multiple hdfs files

["Tomomichi Hirano (JIRA)" <ji...@apache.org>]

Tomomichi Hirano created TEZ-4071:
-------------------------------------

             Summary: shuffle throws exceptions with an external table with multiple hdfs files
                 Key: TEZ-4071
                 URL: https://issues.apache.org/jira/browse/TEZ-4071
             Project: Apache Tez
          Issue Type: Bug
    Affects Versions: 0.9.1
            Reporter: Tomomichi Hirano


{noformat}
2019-05-17 10:08:10,317 [INFO] [Fetcher_B \{Map_1} #1] |shuffle.Fetcher|: Failed to read data to memory for InputAttemptIdentifier [inputIdentifier=1, attemptNumber=0, pathComponent=attempt_1557383221332_0289_1_00_000001_0_10003, spillType=0, spillId=-1]. len=25, decomp=11. ExceptionMessage=Not a valid ifile header
2019-05-17 10:08:10,317 [WARN] [Fetcher_B \{Map_1} #1] |shuffle.Fetcher|: Failed to shuffle output of InputAttemptIdentifier [inputIdentifier=1, attemptNumber=0, pathComponent=attempt_1557383221332_0289_1_00_000001_0_10003, spillType=0, spillId=-1] from XXXXX
java.io.IOException: Not a valid ifile header
 at org.apache.tez.runtime.library.common.sort.impl.IFile$Reader.verifyHeaderMagic(IFile.java:859)
 at org.apache.tez.runtime.library.common.sort.impl.IFile$Reader.isCompressedFlagEnabled(IFile.java:866)
 at org.apache.tez.runtime.library.common.sort.impl.IFile$Reader.readToMemory(IFile.java:616)
 at org.apache.tez.runtime.library.common.shuffle.ShuffleUtils.shuffleToMemory(ShuffleUtils.java:121)
 at org.apache.tez.runtime.library.common.shuffle.Fetcher.fetchInputs(Fetcher.java:950)
 at org.apache.tez.runtime.library.common.shuffle.Fetcher.doHttpFetch(Fetcher.java:599)
 at org.apache.tez.runtime.library.common.shuffle.Fetcher.doHttpFetch(Fetcher.java:486)
 at org.apache.tez.runtime.library.common.shuffle.Fetcher.callInternal(Fetcher.java:284)
 at org.apache.tez.runtime.library.common.shuffle.Fetcher.callInternal(Fetcher.java:76)
 at org.apache.tez.common.CallableWithNdc.call(CallableWithNdc.java:36)
 at com.google.common.util.concurrent.TrustedListenableFutureTask$TrustedFutureInterruptibleTask.runInterruptibly(TrustedListenableFutureTask.java:108)
 at com.google.common.util.concurrent.InterruptibleTask.run(InterruptibleTask.java:41)
 at com.google.common.util.concurrent.TrustedListenableFutureTask.run(TrustedListenableFutureTask.java:77)
 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)
{noformat}

How to reproduce:
{noformat}
1) create two files - file1.csv and file2.csv
2) add two fields to the csv files as below
one,two
one,two
one,two
3) In Hive:
use testdb;
create external table test1(s1 string, s2 string) row format delimited fields terminated by ',' stored as textfile location '/user/usera/test1';
4) Copy one csv file to hdfs - /user/usera/test1
hdfs dfs -put ./file1.csv /user/usera/test1/
5) select count(*) from testdb.test1;
=> works fine.
6) copy the second csv file to HDFS
hdfs dfs -put ./file2.csv /user/usera/test1/
7) select * from testdb.test1;
=> Can see all data from 2 hdfs files.
8) select count(*) from testdb.test1;
=> get this issue.
{noformat}

 

Similar ticket: https://issues.apache.org/jira/browse/TEZ-3699 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)