[jira] [Updated] (HBASE-16011) TableSnapshotScanner and TableSnapshotInputFormat can produce duplicate rows

["Youngjoon Kim (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HBASE-16011?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Youngjoon Kim updated HBASE-16011:
----------------------------------
    Attachment: snapshot_bug_test.patch

Attach a patch adding two tests. These tests throw RuntimeException which indicates duplicate rows were produced.

{panel}
$ mvn -Dtest=org.apache.hadoop.hbase.client.TestTableSnapshotScanner test
...
Running org.apache.hadoop.hbase.client.TestTableSnapshotScanner
Tests run: 4, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 59.814 sec <<< FAILURE! - in org.apache.hadoop.hbase.client.TestTableSnapshotScanner
testDuplicateResultsBug(org.apache.hadoop.hbase.client.TestTableSnapshotScanner)  Time elapsed: 12.473 sec  <<< ERROR!
java.lang.RuntimeException: Row:bbb has a seen count of 2 instead of 1
        at org.apache.hadoop.hbase.HBaseTestingUtility$SeenRowTracker.validate(HBaseTestingUtility.java:2048)
        at org.apache.hadoop.hbase.client.TestTableSnapshotScanner.verifyScanner(TestTableSnapshotScanner.java:223)
        at org.apache.hadoop.hbase.client.TestTableSnapshotScanner.testDuplicateResultsBug(TestTableSnapshotScanner.java:155)


Results :

Tests in error:
  TestTableSnapshotScanner.testDuplicateResultsBug:155->verifyScanner:223 » Runtime

Tests run: 4, Failures: 0, Errors: 1, Skipped: 0
...
{panel}


{panel}
$ mvn -Dtest=org.apache.hadoop.hbase.mapreduce.TestTableSnapshotInputFormat test
...
Running org.apache.hadoop.hbase.mapreduce.TestTableSnapshotInputFormat
Tests run: 9, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 134.654 sec <<< FAILURE! - in org.apache.hadoop.hbase.mapreduce.TestTableSnapshotInputFormat
testDuplicateResultsBug(org.apache.hadoop.hbase.mapreduce.TestTableSnapshotInputFormat)  Time elapsed: 13.552 sec  <<< ERROR!
java.lang.RuntimeException: Row:bbb has a seen count of 2 instead of 1
        at org.apache.hadoop.hbase.HBaseTestingUtility$SeenRowTracker.validate(HBaseTestingUtility.java:2048)
        at org.apache.hadoop.hbase.mapreduce.TestTableSnapshotInputFormat.verifyWithMockedMapReduce(TestTableSnapshotInputFormat.java:317)
        at org.apache.hadoop.hbase.mapreduce.TestTableSnapshotInputFormat.testDuplicateResultsBug(TestTableSnapshotInputFormat.java:273)


Results :

Tests in error:
  TestTableSnapshotInputFormat.testDuplicateResultsBug:273->verifyWithMockedMapReduce:317 » Runtime

Tests run: 9, Failures: 0, Errors: 1, Skipped: 0
...
{panel}

> TableSnapshotScanner and TableSnapshotInputFormat can produce duplicate rows
> ----------------------------------------------------------------------------
>
>                 Key: HBASE-16011
>                 URL: https://issues.apache.org/jira/browse/HBASE-16011
>             Project: HBase
>          Issue Type: Bug
>          Components: snapshots
>            Reporter: Youngjoon Kim
>         Attachments: snapshot_bug_test.patch
>
>
> A snapshot of (non-pre) split table can include both a parent region and daughter regions. If run TableSnapshotScanner or TableSnapshotInputFormat on the such snapshot, duplicate rows are produced.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)