[
https://issues.apache.org/jira/browse/IMPALA-9753?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17110451#comment-17110451
]
ASF subversion and git services commented on IMPALA-9753:
---------------------------------------------------------
Commit 7b8d2e1f78f30fe25e14bea2e5308e80786a4adc in impala's branch
refs/heads/master from Csaba Ringhofer
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=7b8d2e1 ]
IMPALA-9753: Fix TRUNCATE of ACID tables on S3
The use of HDFS API was incorrect when creating an empty file
in the new base dir during truncate. Simply calling Create(Path)
does create the file in HDFS, but it is only created on S3 when
the returned stream is closed.
Testing:
- Acid truncate tests are not running on S3 as they need a running
Hive server. Aded a regression test that will run on S3 too. It
would be nice to run all tests on S3, but this is out of the scope
of this change.
Change-Id: I96d315638b669c5c7198a8e47939cb2b236e35bb
Reviewed-on: http://gerrit.cloudera.org:8080/15940
Reviewed-by: Impala Public Jenkins <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>
> Possible bug in TRUNCATE of ACID tables on S3
> ---------------------------------------------
>
> Key: IMPALA-9753
> URL: https://issues.apache.org/jira/browse/IMPALA-9753
> Project: IMPALA
> Issue Type: Bug
> Components: Frontend
> Reporter: Csaba Ringhofer
> Assignee: Csaba Ringhofer
> Priority: Critical
>
> TRUNCATE TABLE doesn't seem to work on ACID tables in an S3 environment (with
> S3 guard)
> I couldn't reproduce it yet in Impala dev env, and I am not completely sure
> that the cause is actually S3, but found an operation that could possibly go
> wrong on S3:
> https://github.com/apache/impala/blob/master/fe/src/main/java/org/apache/impala/service/CatalogOpExecutor.java#L2115
> Simply calling Create(Path) does create the file in HDFS, but I think that it
> is only created on S3 when the returned stream is closed.
> All related tests are skipped if the environment is S3, so it seems possible
> that the bug could slip through Impala's component tests.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
