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

Posted to issues@hive.apache.org by "Dmitriy Fingerman (Jira)" <ji...@apache.org> on 2024/07/26 18:37:00 UTC

[jira] [Updated] (HIVE-28047) Iceberg: Major QB Compaction on unpartitioned tables with a single commit

     [ https://issues.apache.org/jira/browse/HIVE-28047?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Dmitriy Fingerman updated HIVE-28047:
-------------------------------------
    Summary: Iceberg: Major QB Compaction on unpartitioned tables with a single commit  (was: Iceberg: Major QB Compaction with a single commit)

> Iceberg: Major QB Compaction on unpartitioned tables with a single commit
> -------------------------------------------------------------------------
>
>                 Key: HIVE-28047
>                 URL: https://issues.apache.org/jira/browse/HIVE-28047
>             Project: Hive
>          Issue Type: Improvement
>            Reporter: Dmitriy Fingerman
>            Assignee: Dmitriy Fingerman
>            Priority: Major
>              Labels: pull-request-available
>
> Currently Hive Iceberg Major QB Compaction happens in 2 commits in a single transaction. First commit deletes all existing files and second commit inserts compacted data files. If a user queries the table using snapshot id of the first commit after the compaction, there will be no records returned which is invalid. Need to find a way to do it in one commit.



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