[
https://issues.apache.org/jira/browse/HIVE-17496?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16170718#comment-16170718
]
Tao Li commented on HIVE-17496:
-------------------------------
The cause to the deletion failure is because the table dir was deleted on
target db for some reason. We did not try to drop the table so not sure why the
table dir was deleted. See log below. This issue cannot be reproduced locally.
{noformat}
2017-09-18T12:49:20,642 ERROR [main] parse.TestReplicationScenarios: Error
verifying the staging dir deletion
java.io.FileNotFoundException: File
file:/home/hiveptest/130.211.162.127-hiveptest-0/apache-github-source-source/itests/hive-unit/target/warehouse/deleteStagingDir_org_apache_hadoop_hive_ql_parse_testreplicationscenarios_1505763609715_dupe.db/unptned
does not exist
{noformat}
So for now the fix is to add a existence check.
> Bootstrap repl is not cleaning up staging dirs
> ----------------------------------------------
>
> Key: HIVE-17496
> URL: https://issues.apache.org/jira/browse/HIVE-17496
> Project: Hive
> Issue Type: Bug
> Components: repl
> Reporter: Tao Li
> Assignee: Tao Li
> Attachments: HIVE-17496.1.patch, HIVE-17496.2.patch,
> HIVE-17496.3.patch, HIVE-17496.4.patch, HIVE-17496.5.patch, HIVE-17496.6.patch
>
>
> This will put more pressure on the HDFS file limit.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
