You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Martijn Visser (Jira)" <ji...@apache.org> on 2022/07/06 11:30:00 UTC

[jira] [Commented] (FLINK-28422) Fall back to lastest supported HiveShim when user use a higher hive version

    [ https://issues.apache.org/jira/browse/FLINK-28422?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17563163#comment-17563163 ] 

Martijn Visser commented on FLINK-28422:
----------------------------------------

I'm wondering if this is a good idea. If we do this, then we'll easily forget to update the documentation with the versions of Hive that are supported. I think it's better to explicitly add support for specific versions. 

> Fall back to lastest supported HiveShim when user use a higher hive version
> ---------------------------------------------------------------------------
>
>                 Key: FLINK-28422
>                 URL: https://issues.apache.org/jira/browse/FLINK-28422
>             Project: Flink
>          Issue Type: Improvement
>          Components: Connectors / Hive
>            Reporter: luoyuxia
>            Priority: Major
>             Fix For: 1.16.0
>
>
> Flink provides flink-connector-hive.jar, so that user can bundle it with a specific hive-exec.jar to make Flink read / write their Hive.
> The flink-connector-hive mantains the Hive versions that Flink supports.  But when Hive releases a higher version, which may well be not in the maintain list of Flink for the delay of Flink's release .
> Then user will find the unsupported version exception with such higher version hive-exec.jar.
> The better way is to fall back to a nearest Hive version directly, with a warning message.  It usually works fine unless some big changes in Hive.



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