[
https://issues.apache.org/jira/browse/HIVE-6699?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Owen O'Malley updated HIVE-6699:
--------------------------------
Attachment: HIVE-6699.patch
Ok, the fix is a little more complicated. Hadoop 1.x needs sync and Hadoop 2.x
need hflush, so I added a shim for hflush and make it use sync in Hadoop 1.x.
> Fix OrcRecordUpdater to use sync instead of flush
> -------------------------------------------------
>
> Key: HIVE-6699
> URL: https://issues.apache.org/jira/browse/HIVE-6699
> Project: Hive
> Issue Type: Sub-task
> Reporter: Owen O'Malley
> Assignee: Owen O'Malley
> Attachments: HIVE-6699.patch, HIVE-6699.patch
>
>
> OrcRecordUpdater needs to use sync so that changes are visible before the
> client closes the file.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
