[
https://issues.apache.org/jira/browse/HADOOP-15751?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16612274#comment-16612274
]
Steve Loughran commented on HADOOP-15751:
-----------------------------------------
stream at fault here looks like its actually the input stream, the data for the
_SUCCESS file. Not something remote.
possibilities
* transient failure in PUT -> retry -> inability of stream to reset
* somehow source data being deleted?
FWIW, block output buffering in {{fs.s3a.fast.upload.buffer}} explicitly set to
disk
> AWS Data read stack trace in S3a putObjectDirect
> ------------------------------------------------
>
> Key: HADOOP-15751
> URL: https://issues.apache.org/jira/browse/HADOOP-15751
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3, test
> Affects Versions: 3.3.0
> Reporter: Steve Loughran
> Priority: Minor
>
> (transient) failure in a put to AWS london from UK Bristol; 120+ miles away
> (I.e not that far).
> Error "Data read has a different length than the expected:"
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
