ConsistencyException: Directory ‘path_name/20190916’ present in the metadata but not s3

Trouble Shooting

I got ConsistencyException this message during spark job,

ERROR yarn.ApplicationMaster: User class threw exception: com.amazon.ws.emr.hadoop.fs.consistency.exception.ConsistencyException: Directory 'path_name/datetime=20190916' present in the metadata but not s3

After searching google, some people resolved with emrfs command

emrfs delete s3://path

But I can’t do emrfs cli since EC2 key pair issue.

And I found another solution that trying to put ignore s3 consistent option to spark-submit.

spark.hadoop.fs.s3.consistent=false
spark.hadoop.fs.s3.consistent.throwExceptionOnInconsistency=false

This way is not a complete solution because this code is defence code of consistent check.
But if this spark job is a single job, and this spark job is not running at the same time, it is okay to turn off emrfs.