[
https://issues.apache.org/jira/browse/SPARK-51638?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Attila Zsolt Piros resolved SPARK-51638.
----------------------------------------
Fix Version/s: 4.0.0
4.1.0
Resolution: Fixed
> Fetching the remote disk stored RDD blocks via the external shuffle service
> is not working
> ------------------------------------------------------------------------------------------
>
> Key: SPARK-51638
> URL: https://issues.apache.org/jira/browse/SPARK-51638
> Project: Spark
> Issue Type: Bug
> Components: Block Manager
> Affects Versions: 4.0.0, 4.1.0, 3.5.6
> Reporter: Attila Zsolt Piros
> Assignee: Attila Zsolt Piros
> Priority: Major
> Labels: pull-request-available
> Fix For: 4.0.0, 4.1.0
>
>
> After https://issues.apache.org/jira/browse/SPARK-43221, remote fetching of
> disk-stored RDD blocks is not working.
> The getLocationsAndStatus method does not use the external shuffle service
> when the blocks are not host-local. The fallback to extend the search to all
> executors only includes the active executors, so after downscaling, blocks
> stored on a killed executor will not be found there which leads to recompute.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
