The Data Integration Service truncates the target files and reject files before writing the data. To append output data to HDFS target files and reject files, choose to append data if the HDFS target exists.
Data is appended to reject files only if the reject file directory is on the Data Integration Service machine. If the directory is in the Hadoop environment, rejected rows are overwritten.
When you choose to append data if the HDFS target exists, the Data Integration Service appends the mapping execution ID to the names of the target files and reject files.
When you use a HDFS flat file target in a mapping, you must specify the full path that includes the output file directory and file name. The Data Integration Service might generate multiple output files in the output directory when you run the mapping in a Hadoop environment.
An HDFS target cannot reside on a remote cluster. A remote cluster is a cluster that is remote from the machine that the Hadoop connection references in the mapping.