You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kylin.apache.org by "Shaofeng SHI (JIRA)" <ji...@apache.org> on 2015/07/13 07:25:06 UTC

[jira] [Commented] (KYLIN-772) Continue cube job when hive query return empty resultset

    [ https://issues.apache.org/jira/browse/KYLIN-772?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14624210#comment-14624210 ] 

Shaofeng SHI commented on KYLIN-772:
------------------------------------

This has been implemented in v0.8, will back port to 0.7;

> Continue cube job when hive query return empty resultset
> --------------------------------------------------------
>
>                 Key: KYLIN-772
>                 URL: https://issues.apache.org/jira/browse/KYLIN-772
>             Project: Kylin
>          Issue Type: Improvement
>          Components: Job Engine
>    Affects Versions: v0.7.1
>            Reporter: Luke Han
>            Assignee: Shaofeng SHI
>             Fix For: v0.8.1, v0.7.4
>
>
> When there's no data in Hive table or in incremental query result set, the #3 "Step Name: Build Dimension Dictionary" will fail due to zero record. 
> Error and block job is not ideal solution since many cases require empty segments to be created as placeholder for further refresh and merge, and continue to build with future data.
> Requirement:
> Add checkpoint about empty result set and continue entire cube build job.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)