Created 05-30-2018 10:42 PM
I have a new HDP on prem cluster and want make hive internal storage location to s3 instead of hdfs. Is this supported or is this only supported with cloudbreak/HDC?
Created 06-02-2018 01:23 AM
Per the docs, Hive on S3 should be using external tables with the Cloud Data Connectors: https://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.6.1/bk_cloud-data-access/content/hive.html
Created 06-04-2018 07:50 PM
@Michael Young Thanks for that. I am aware of external tables on s3. What specifically I am looking for is internally managed tables. If you come across any information please share.
Created 06-05-2018 02:26 PM
@sunile.manjee I don't believe internal managed tables are supported on S3 currently.
Created 06-08-2018 02:20 PM
@Michael Young I found the answer here: https://wiki.apache.org/hadoop/AmazonS3
“You cannot use any of the S3 filesystem clients as a drop-in replacement for HDFS.” So not only is this not recommended and not supported.