I am using the --clone option correctly? hadoop jar filecrush.jar crush.Crush --clone --threshold 1 --compress org.apache.hadoop.io.compress.GzipCodec --input-format org.apache.hadoop.mapred.TextInputFormat --output-format org.apache.hadoop.mapred.SequenceFileOutputFormat --info --verbose /user/hive/warehouse/partner_breakdown/ds=2012-10-30 /user/gtevelde/test
my understanding is the small "non-crushed" files will get copied to /user/gtevelde/test and the new crushed files will be in the original folder. However, when I run the command above, the small files stay in the input folder and the larger files are place in the output directory. Any help would be great! Thanks,George
