Win7 The Eclipse Remote development Hadoop program is divided into two types:
(1) Run "Run as" Java application, the wrapper is jar, upload the cluster execution (not explained here)
(2) Running "Run as" run on Hadoop
The point is that run on Hadoop is a way to build the Eclipse remote development environment, run on Hadoop, the program succeeds, and the heart steals, but it looks like this:
I clearly set job.setnumreducetasks (6), the end should have 6 reduce output, how to become a?
Found here are localjobrunner local execution, and did not submit resources to the cluster execution, I think, then set it to cluster execution chant, so add:
execution, the following error has occurred:
Altogether can not find a custom mymapper, so, remember the cluster on the task, is to put the resources first packaged into HDFs, so,
So, the package Hadoop program is Ttt.jar, and the configuration settings are as follows:
Execute again, run successfully, as follows:
Summary:
1, Eclipse on the submission of tasks to the cluster execution, a lot of configuration or use of the local, preferably in code implementation
2, submit the task to the cluster execution, the first to submit resources to HDFs
Submit a task to the cluster execution on eclipse