[ 
https://issues.apache.org/jira/browse/HIVE-26584?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17612693#comment-17612693
 ] 

John Sherman commented on HIVE-26584:
-------------------------------------

Thanks [~ayushtkn] and [~zabetak] for helping to make the patch better.
I don't see any indication that the tests require external tables. But I agree 
that I could be overlooking a detail and I don't functionally need to change 
the test so significantly (even if I think the newer version is cleaner). I 
went with just rmr the created directories to clean up the directories at the 
end.

I did not add DROP IF EXISTS or rmr(s) before the creation since I find that 
practice typically hides problems and sometimes causes hidden dependencies 
between tests.

As for clearTablesCreatedDuringTests - it doesn't clean these files up because 
it only cleans up tables under the configured warehouse directory. This test 
case manually creates a location not under the warehouse directory so it 
doesn't end up cleaning up the files. I could modify the 
clearTablesCreatedDuringTests to clean up all directories mentioned in CREATE 
EXTERNAL TABLE location clauses, but that could be risky since it could lead to 
user files being accidentally removed with a misconfigured location clause and 
I am not sure I would be able to add all the possible checks to prevent that.

In the future I think tests like this should load the data via LOAD DATA if 
possible and not use a custom LOCATION clause so it gets cleaned up normally.

> compressed_skip_header_footer_aggr.q is flaky
> ---------------------------------------------
>
>                 Key: HIVE-26584
>                 URL: https://issues.apache.org/jira/browse/HIVE-26584
>             Project: Hive
>          Issue Type: Bug
>          Components: HiveServer2
>    Affects Versions: 4.0.0-alpha-2
>            Reporter: John Sherman
>            Assignee: John Sherman
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> One of my PRs compressed_skip_header_footer_aggr.q  was failing with 
> unexpected diff. Such as:
> {code:java}
>  TestMiniLlapLocalCliDriver.testCliDriver:62 Client Execution succeeded but 
> contained differences (error code = 1) after executing 
> compressed_skip_header_footer_aggr.q
> 69,71c69,70
> < 1 2019-12-31
> < 2 2018-12-31
> < 3 2017-12-31
> ---
> > 2 2019-12-31
> > 3 2019-12-31
> 89d87
> < NULL  NULL
> 91c89
> < 2 2018-12-31
> ---
> > 2 2019-12-31
> 100c98
> < 1
> ---
> > 2
> 109c107
> < 1 2019-12-31
> ---
> > 2 2019-12-31
> 127,128c125,126
> < 1 2019-12-31
> < 3 2017-12-31
> ---
> > 2 2019-12-31
> > 3 2019-12-31
> 146a145
> > 2 2019-12-31
> 155c154
> < 1
> ---
> > 2 {code}
> Investigating it, it did not seem to fail when executed locally. Since I 
> suspected test interference I searched for the tablenames/directories used 
> and discovered empty_skip_header_footer_aggr.q which uses the same table 
> names AND external directories.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to