[
https://issues.apache.org/jira/browse/HIVE-27657?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17761980#comment-17761980
]
Mayank Kunwar commented on HIVE-27657:
--------------------------------------
Thanks [~lvegh] for the review.
> Change hive.fetch.task.conversion.threshold default value
> ---------------------------------------------------------
>
> Key: HIVE-27657
> URL: https://issues.apache.org/jira/browse/HIVE-27657
> Project: Hive
> Issue Type: Bug
> Reporter: Mayank Kunwar
> Assignee: Mayank Kunwar
> Priority: Major
> Labels: pull-request-available
>
> With the introduction of [fetch task
> caching|https://github.com/apache/hive/blob/d0a06239b09396d1b7a6414d85011f9a20f8486a/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java#L3532-L3535],
> HS2 may quickly run out of memory if there are multiple parallel queries
> with fetch task.
> We should be reducing the default value for
> hive.fetch.task.conversion.threshold to 209715200 (200M) from current
> 1073741824 (1G)
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
