[
https://issues.apache.org/jira/browse/HADOOP-18447?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17601976#comment-17601976
]
Mukund Thakur commented on HADOOP-18447:
----------------------------------------
Currently the threadpool is shared unbounded one but will be moved bounded one.
But we are terminating the running vectored IO operations when the stream is
closed
[https://github.com/apache/hadoop/blob/trunk/hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AInputStream.java#L114]
https://github.com/apache/hadoop/blob/trunk/hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AInputStream.java#L611
> Vectored IO: Threadpool should be closed on interrupts or during close calls
> ----------------------------------------------------------------------------
>
> Key: HADOOP-18447
> URL: https://issues.apache.org/jira/browse/HADOOP-18447
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: common, fs, fs/adl, fs/s3
> Reporter: Rajesh Balamohan
> Priority: Major
> Labels: performance, stability
> Attachments: Screenshot 2022-09-08 at 9.22.07 AM.png
>
>
> Vectored IO threadpool should be closed on any interrupts or during
> S3AFileSystem/S3AInputStream close() calls.
> E.g Query which got cancelled in the middle of the run. However, in
> background (e.g LLAP) vectored IO threads continued to run.
>
> !Screenshot 2022-09-08 at 9.22.07 AM.png|width=537,height=164!
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
