[
https://issues.apache.org/jira/browse/HBASE-15126?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15110799#comment-15110799
]
Ted Yu commented on HBASE-15126:
--------------------------------
{code}
Failed tests:
TestReplicationKillSlaveRS.killOneSlaveRS:34->TestReplicationKillRS.loadTableAndKillRS:88
Waited too much time for queueFailover replication. Waited 18130ms.
{code}
The above test failure was not related to patch.
TestQuotaThrottle failure was not related either.
> HBaseFsck's checkRegionBoundaries function set incorrect 'storesFirstKey'
> -------------------------------------------------------------------------
>
> Key: HBASE-15126
> URL: https://issues.apache.org/jira/browse/HBASE-15126
> Project: HBase
> Issue Type: Bug
> Components: hbck
> Affects Versions: 2.0.0
> Reporter: chenrongwei
> Assignee: chenrongwei
> Fix For: 2.0.0, 1.3.0
>
> Attachments: HBASE-15126-v001.patch, HBASE-15126-v002.patch,
> HBASE-15126-v003.patch, HBASE-15126-v004.patch, HBASE-15126-v005.patch,
> HBASE-15126-v006.patch
>
>
> HBaseFsck's checkRegionBoundaries function set the
> 'currentRegionBoundariesInformation.storesFirstKey' was incorrect.I think it
> should be set like below,
> currentRegionBoundariesInformation.storesFirstKey = keyOnly(storeFirstKey);
> but current the 'currentRegionBoundariesInformation.storesFirstKey ' is just
> set to 'storeFirstKey',which will cause to
> comparator.compare(currentRegionBoundariesInformation.storesFirstKey,
> currentRegionBoundariesInformation.metaFirstKey) get a wrong
> result. Because it just compared the rowkey's length.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
