Search

What will happen if we run a Mapreduce job with an output directory that already existing ?



Job will fail with org.apache.hadoop.mapred.FileAlreadyExistsException. In this case, delete the output directory and re-execute the job.