When running the examples, there is no separate starting of the nodes that hold data. The examples seem like they themselves start up the datanode meaning I have to write a main on each server? Is that true, or is there a way to start up all the datanodes first and then run the examples against my small little 3 node cluster?
Another thing that really wasn't nice was that the first example has localhost hardcoded in the code so it will not work on more than one node if someone wanted to play with the clustering(which is the first thing I want to do with a clustered environment like this). Any examples that run against 3 datanodes where I start the datanodes first and THEN run the example?