hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Aihua Xu (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-16607) ColumnStatsAutoGatherContext regenerates HiveConf.HIVEQUERYID
Date Tue, 09 May 2017 14:58:04 GMT

    [ https://issues.apache.org/jira/browse/HIVE-16607?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16002823#comment-16002823
] 

Aihua Xu commented on HIVE-16607:
---------------------------------

[~pvary] Can you create a code review for the patch? Some basic questions: I thought we assign
queryId before the query starts the compilation. Can you point out where we reset the new
queryId? The Builder pattern seems to be a good idea. Can we just call it Builder (so it will
be QueryState.Builder. Looks like typically it's named like this?)?

> ColumnStatsAutoGatherContext regenerates HiveConf.HIVEQUERYID
> -------------------------------------------------------------
>
>                 Key: HIVE-16607
>                 URL: https://issues.apache.org/jira/browse/HIVE-16607
>             Project: Hive
>          Issue Type: Bug
>          Components: Beeline, Logging
>            Reporter: Peter Vary
>            Assignee: Peter Vary
>         Attachments: HIVE-16607.02.patch, HIVE-16607.patch
>
>
> Creating a new {{QueryState}} object regenerates the HIVEQUERYID stored in the {{HiveConf}}.
> In HiveServer logs it makes hard to follow the life of the query since a new queryid
is assigned to the query during the execution.
> Since BeeLine is showing the operation logs based on the queryid, only the first several
line of the logs is showed in BeeLine.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message