[jira] [Commented] (HIVE-17306) Support MySQL InnoDB Cluster

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

    [ https://issues.apache.org/jira/browse/HIVE-17306?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17892103#comment-17892103 ] 

Kiran Velumuri commented on HIVE-17306:
---------------------------------------

[~steveyeom2017] Can I pick this up?

> Support MySQL InnoDB Cluster
> ----------------------------
>
>                 Key: HIVE-17306
>                 URL: https://issues.apache.org/jira/browse/HIVE-17306
>             Project: Hive
>          Issue Type: Improvement
>          Components: Transactions
>            Reporter: Shawn Weeks
>            Assignee: Steve Yeom
>            Priority: Minor
>
> To support high availability of the Hive Metastore using a highly available database is required. To support the MySQL InnoDB Cluster it looks like we're just missing a couple primary keys as we were already using InnoDB tables for the metastore. It looks like it's primarily the transaction tables that don't have primary keys like TXN_COMPONENTS and COMPLETED_TXN_COMPONENTS. The primary keys can be surrogate sequences if there really is no unique identifier in these tables.



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