[
https://issues.apache.org/jira/browse/HBASE-2373?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12849542#action_12849542
]
Jean-Daniel Cryans commented on HBASE-2373:
-------------------------------------------
This happens because scanners don't respect row locks so Stack at some point
added this fix where we do a get after every scan in the .META. scanner to make
sure we have the correct data. If not, it prints that out. Upon inspection of
your master log you should see exactly what happened with that region.
> BaseScanner GET got different address/startcode than SCAN
> ---------------------------------------------------------
>
> Key: HBASE-2373
> URL: https://issues.apache.org/jira/browse/HBASE-2373
> Project: Hadoop HBase
> Issue Type: Bug
> Affects Versions: 0.20.3
> Reporter: Kannan Muthukkaruppan
>
> Seeing some of these errors in the HBase master's logs:
> {code}
> 2010-03-24 16:52:17,003 DEBUG org.apache.hadoop.hbase.master.BaseScanner: GET
> on test1,8947030000,1269474720493 got different address than SCAN:
> sa=10.18.34.217:60020, serverAddress=
> 2010-03-24 16:52:17,003 DEBUG org.apache.hadoop.hbase.master.BaseScanner: GET
> on test1,8947030000,1269474720493 got different startcode than SCAN:
> sc=1269456397807, serverAddress=0
> 2010-03-24 16:52:17,018 DEBUG org.apache.hadoop.hbase.master.BaseScanner: GET
> on test1,8953040000,1269474720493 got different address than SCAN:
> sa=10.18.34.215:60020, serverAddress=
> 2010-03-24 16:52:17,018 DEBUG org.apache.hadoop.hbase.master.BaseScanner: GET
> on test1,8953040000,1269474720493 got different startcode than SCAN:
> sc=1269456397735, serverAddress=0
> {code}
> Ideas?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
