As title.
We have stage/prod emr clusters and we may need to run emrfs delete s3_path command on both clusters via the jenkins jobs.
However, I can run the emrfs delete successfully on stage emr one, but failed on prod. Below are the log:
22:54:51 Clear meta-store before loading into DW table.
22:54:51 ----------------------------------------------------------
22:54:51
22:54:51 Pseudo-terminal will not be allocated because stdin is not a terminal.
22:54:52 19/03/23 02:54:52 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
22:54:53 EmrFsApplication.scala(91): dynamoDB endPoint = dynamodb.us-east-1.amazonaws.com
22:54:53 EmrFsApplication.scala(99): s3 endPoint = s3.amazonaws.com
22:54:53 EmrFsApplication.scala(107): sqs endPoint = sqs.us-east-1.amazonaws.com
22:54:54 Metadata 'EmrFSMetadata' does not exist
I don't know why EmrFSMetadata not exist in my prod emr? or is it something special settings need to apply to the prod one?
Thanks.
Ah.. I think I got the answer. Our prod emr doesnt enable EMRFS consistent view. That's why...
Question can be closed.