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

Posted to issues@hive.apache.org by "Antal Sinkovits (Jira)" <ji...@apache.org> on 2022/04/19 18:07:00 UTC

[jira] [Work started] (HIVE-26155) Create a new connection pool for compaction

     [ https://issues.apache.org/jira/browse/HIVE-26155?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Work on HIVE-26155 started by Antal Sinkovits.
----------------------------------------------
> Create a new connection pool for compaction
> -------------------------------------------
>
>                 Key: HIVE-26155
>                 URL: https://issues.apache.org/jira/browse/HIVE-26155
>             Project: Hive
>          Issue Type: Improvement
>          Components: Standalone Metastore
>            Reporter: Antal Sinkovits
>            Assignee: Antal Sinkovits
>            Priority: Major
>              Labels: compaction
>
> Currently the TxnHandler uses 2 connection pools to communicate with the HMS: the default one and one for mutexing. If compaction is configured incorrectly (e.g. too many Initiators are running on the same db) then compaction can use up all the connections in the default connection pool and all user queries can get stuck.
> We should have a separate connection pool (configurable size) just for compaction-related activities.



--
This message was sent by Atlassian Jira
(v8.20.7#820007)