[
https://issues.apache.org/jira/browse/HADOOP-16838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17029835#comment-17029835
]
Steve Loughran commented on HADOOP-16838:
-----------------------------------------
It sounds like you're using spark
* the Hadoop documentation applies here, please consult before finding JIRAs
* the fact Apache spark ships withs hadoop 2.7 jars and so does not contain
fixes from five years ago is a topic to take up with spark team
* as it is, before you file requests/patches/bug reports against the hadoop-
modules, your first step must be to download and try the latest hadoop release
artifacts first.
We're five years ahead of the JARS which ship with Spark. There are a lots of
changes there -none of them are going to be back ported. Please upgrade
I considered whether to close this as WORKSFOR or INVALID; gone to duplicate
because it sounds like the cause is more Spark artifact related rather than
just favourite consult documentation.
> Support for `fs.s3a.endpoint.region`
> ------------------------------------
>
> Key: HADOOP-16838
> URL: https://issues.apache.org/jira/browse/HADOOP-16838
> Project: Hadoop Common
> Issue Type: New Feature
> Reporter: Nitish
> Priority: Major
>
> Currently it is not possible to connect S3 Compatible services like MinIO,
> Ceph, etc (running with a custom region) to Spark with s3a connector. For
> example, if MinIO is running on a Server with
> * IP Address: 192.168.0.100
> * Region: ap-southeast-1
> The s3a connector can't be configured to use the region `ap-southeast-1`.
> It would be great to have a configuration field like
> `fs.s3a.endpoint.region`. This will be very helpful for users deploying
> Private Cloud and who intend to use S3 like services on premises.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
