[
https://issues.apache.org/jira/browse/SPARK-30069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17172264#comment-17172264
]
Apache Spark commented on SPARK-30069:
--------------------------------------
User 'LantaoJin' has created a pull request for this issue:
https://github.com/apache/spark/pull/29378
> Clean up non-shuffle disk block manager files following executor exists on
> YARN
> -------------------------------------------------------------------------------
>
> Key: SPARK-30069
> URL: https://issues.apache.org/jira/browse/SPARK-30069
> Project: Spark
> Issue Type: Bug
> Components: Spark Core, YARN
> Affects Versions: 3.0.0
> Reporter: Lantao Jin
> Priority: Major
>
> Currently we only clean up the local directories on application removed.
> However, when executors die and restart repeatedly, many temp files are left
> untouched in the local directories, which is undesired behavior and could
> cause disk space used up gradually.
> SPARK-24340 had fixed this problem in standalone mode. But in YARN mode, this
> issue still exists. Especially, in long running service like Spark
> thrift-server with dynamic resource allocation disabled, it's very easy
> causes local disk full.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
