[
https://issues.apache.org/jira/browse/HBASE-10867?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13954583#comment-13954583
]
Liu Shaohui commented on HBASE-10867:
-------------------------------------
[~yuzhih...@gmail.com]
The patch does fix the failed test. But, the root cause of the issue is that
the Master is also a RegionServer since HBASE-10569.
So the servers.length will be 11 and larger than the SLAVES:10 in the test.
{code}
killIndex = random.nextInt(servers.length);
{code}
What about making the MiniHBaseCluster adjust to HBASE-10569?
> TestRegionPlacement#testRegionPlacement occasionally fails
> ----------------------------------------------------------
>
> Key: HBASE-10867
> URL: https://issues.apache.org/jira/browse/HBASE-10867
> Project: HBase
> Issue Type: Test
> Reporter: Ted Yu
> Assignee: Ted Yu
> Priority: Minor
> Attachments: 10867-v1.txt
>
>
> From
> https://builds.apache.org/job/HBase-TRUNK/5047/testReport/org.apache.hadoop.hbase.master/TestRegionPlacement/testRegionPlacement/
> :
> {code}
> java.lang.ArrayIndexOutOfBoundsException: 10
> at
> java.util.concurrent.CopyOnWriteArrayList.get(CopyOnWriteArrayList.java:368)
> at
> java.util.concurrent.CopyOnWriteArrayList.get(CopyOnWriteArrayList.java:377)
> at
> org.apache.hadoop.hbase.LocalHBaseCluster.getRegionServer(LocalHBaseCluster.java:224)
> at
> org.apache.hadoop.hbase.MiniHBaseCluster.getRegionServer(MiniHBaseCluster.java:609)
> at
> org.apache.hadoop.hbase.master.TestRegionPlacement.killRandomServerAndVerifyAssignment(TestRegionPlacement.java:303)
> at
> org.apache.hadoop.hbase.master.TestRegionPlacement.testRegionPlacement(TestRegionPlacement.java:270)
> {code}
> In the setup:
> {code}
> TEST_UTIL.startMiniCluster(SLAVES);
> {code}
> where SLAVES is 10.
> So when 10 was used in
> TEST_UTIL.getHBaseCluster().getRegionServer(killIndex), we would get
> ArrayIndexOutOfBoundsException.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
