[jira] [Updated] (KYLIN-4159) The first step of build cube job will fail and throw "Column 'xx' in where clause is ambiguous" in jdbc datasource.

["luguosheng (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/KYLIN-4159?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

luguosheng updated KYLIN-4159:
------------------------------
    Summary: The first step of build cube job will fail and throw "Column 'xx' in where clause is ambiguous" in jdbc datasource.  (was: The first step of job will fail and throw "Column 'xx' in where clause is ambiguous" in jdbc datasource.)

> The first step of build cube job will fail and throw "Column 'xx' in where clause is ambiguous" in jdbc datasource.
> -------------------------------------------------------------------------------------------------------------------
>
>                 Key: KYLIN-4159
>                 URL: https://issues.apache.org/jira/browse/KYLIN-4159
>             Project: Kylin
>          Issue Type: Bug
>            Reporter: luguosheng
>            Assignee: luguosheng
>            Priority: Major
>
> Reappear:
> In the model of kylin, If we have a column used as a partition column in the fact table, and there is a column in lookup table has the same name. It can be reappeared。
> RootCause:
>  In jdbc datasource , we use Sqoop option "--boundary-query sql"  to split data,  the sql like "select max(partition_column), min(partition_column) from flat_table where flat_table.partition_column> 20120101 and flat_table.partition.column < 20190101" and  there is more than one column which have the same name of the "partition_column" in the flat_table. the sql driver can not distinguish the table that partition_column belong to。
> How to resolve:
>  Alter the sql like: "select max(t.partition_column),min(t.partition_column) from flat_table where flat_table.partition_column> 20120101 and flat_table.partition.column < 20190101"
>  
>  



--
This message was sent by Atlassian Jira
(v8.3.2#803003)