Th current behavior of hive metastore during a "drop table <table_name>" command is to delete the data from HDFS warehouse and put it into Trash.
Currently there is no way to provide a flag to tell the warehouse to skip trash while deleting table data.
This ticket is to add skipTrash configuration "hive.warehouse.data.skipTrash" , which when set to true, will skipTrash while dropping table data from hdfs warehouse. This will be set to false by default to keep current behavior.
This would be good feature to add, so that an admin of the cluster can specify when not to put data into the trash directory (eg. in a dev environment) and thus not to fill hdfs space instead of relying on trash interval and policy configuration to take care of disk filling issue.