I'm writing a Map-Reduce job on Hadoop and i've run into something
pretty peculiar that i cant seem to figure out. For some reason the
OutputCollector object is not what i'm expecting it to be in the
reducer. I declared the reducer as such:
public void reduce(Text key, Iterator<MapWritable> values,
OutputCollector<Text, Text> output, Reporter reporter)
However, when stepping through the code, the actual OutputCollector
object i'm given is of type OutputCollector<Text, MapWritable>. It
throws an error saying:
java.io.IOException: wrong value class: org.apache.hadoop.io.Text is
not class org.apache.hadoop.io.MapWritable
Does anyone know why this would happen? The signature for my mapper
is below which may give some hints:
public void map(Text key, ThriftWritable value, OutputCollector<Text,
MapWritable> output, Reporter reporter)
ThriftWritable is an input format we've created in house.
Any help would be greatly appreciated.
- Manish
Co-Founder Rapleaf.com
http://www.rapleaf.com/pub/Manish-Shah
