Closed bm3780 closed 12 years ago
Can you explain the reason for doing this? Did you run into any problems with the current behavior?
Yes, the subsequent MapReduce job for the Pail consolidation was being run locally rather than against my Hadoop cluster. I'm a new Hadoop user so perhaps there's another way to accomplish this. Setting the "mapred.job.tracker" property for my FileSystem configuration seemed to resolve this issue, and this is why I was proposing that the JobConf inherit the FileSystem's configuration.
@nathanmarz, are you okay with this? I'm happy to merge this in.
I'm proposing that JobConf should inherit from the FileSystem's configuration for a Consolidation.