[jira] [Commented] (HIVE-19326) union_fast_stats MiniLlapLocal golden file has incorrect "accurate" stats (incorrect query results possible)

["Zoltan Haindrich (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HIVE-19326?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16463877#comment-16463877 ] 

Zoltan Haindrich commented on HIVE-19326:
-----------------------------------------

actually this seems to be 2 issues:

* there was an issue which arised from the fact that all the rs-es have overwritten eachothers output - I've fixed this..and now ctas for unions work as expected
* if {{hive.merge.tezfiles}} is enabled ; somehow the counters also get merged or something...I'm not sure if this is enabled in production environments or not...I'll take another look

I think {{hive.optimize.metadataonly}} should be enabled for all tests...or at least check that all tests would pass with it...

> union_fast_stats MiniLlapLocal golden file has incorrect "accurate" stats (incorrect query results possible)
> ------------------------------------------------------------------------------------------------------------
>
>                 Key: HIVE-19326
>                 URL: https://issues.apache.org/jira/browse/HIVE-19326
>             Project: Hive
>          Issue Type: Bug
>          Components: Statistics
>            Reporter: Sergey Shelukhin
>            Assignee: Zoltan Haindrich
>            Priority: Blocker
>             Fix For: 3.0.0
>
>
> Found when investigating the results change after converting tables to MM, turns out the MM result is correct but the current one is not.
> The test ends like so:
> {noformat}
> desc formatted small_alltypesorc_a;
> ANALYZE TABLE small_alltypesorc_a COMPUTE STATISTICS;
> desc formatted small_alltypesorc_a;
> insert into table small_alltypesorc_a select * from small_alltypesorc1a;
> desc formatted small_alltypesorc_a;
> {noformat}
> The results from the descs in the golden file are:
> {noformat}
> 	COLUMN_STATS_ACCURATE	{\"BASIC_STATS\":\"true\"}
> 	numFiles            	1                   
> 	numRows             	5                               
> ...
> 	COLUMN_STATS_ACCURATE	{\"BASIC_STATS\":\"true\"}
> 	numFiles            	1                   
> 	numRows             	15                                
> ...
> 	COLUMN_STATS_ACCURATE	{\"BASIC_STATS\":\"true\"}
> 	numFiles            	2                   
> 	numRows             	20                              
> {noformat}
> Note the result change after analyze - the original nomRows is inaccurate, but  BASIC_STATS is set to true.
> I am assuming with metadata only optimization this can produce incorrect results.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)