[
https://issues.apache.org/jira/browse/HADOOP-16570?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16933487#comment-16933487
]
Steve Loughran commented on HADOOP-16570:
-----------------------------------------
note: one other aspect of the issue is when you are generating files with many,
many blocks, the size of a SinglePendingCommit increases: one md5 checksum
string per block. So its a function of file size/blocksize as well as #of files
and #of tasks
> S3A committers leak threads/raises OOM on job/task commit at scale
> ------------------------------------------------------------------
>
> Key: HADOOP-16570
> URL: https://issues.apache.org/jira/browse/HADOOP-16570
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3
> Affects Versions: 3.2.0, 3.1.2
> Reporter: Steve Loughran
> Assignee: Steve Loughran
> Priority: Major
>
> The fixed size ThreadPool created in AbstractS3ACommitter doesn't get cleaned
> up at EOL; as a result you leak the no. of threads set in
> "fs.s3a.committer.threads"
> Not visible in MR/distcp jobs, but ultimately causes OOM on Spark
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
