Win7+myeclipse10 developing Hadoop Application environment Building

1, replication Linux under the configured installed Hadoop directory into Windows,

Also modify the java_home of the hadoo-env.sh file to be the JDK directory under Windows.

2, copy Hadoop-eclipse-1.2.jar plug-in into the MyEclipse folder under Dropins

　　Download: I have a file under the Hadoop-eclipse1.2.rar file (compressed package), download down, the suffix is changed to jar can be

3. Restart MyEclipse

4, configuration: Windows-preference->windows under the Hadoop directory

　　

5. Edit the Hadoop location

　　

6. Set M/R and DFS see configuration of Mapred-site.xml and Core-site.xml

　　

7. Set user name, user name on Linux note

　　

8. Set TEMP directory TMP, see Configuration of Core-site.xml

　　

9,Hadoop service cluster shutdown authorization authentication, set Linux on the hadoop/conf under the Mapred-site.xml, but also modify Windows.

1. <property>
2. <name>dfs.permissions</name>
3. <value>false</value>
4. </property>

Restart the Hadoop cluster.

10, save the settings on the MyEclipse, restart the myeclipse, the configuration item will be re-assigned according to the settings of TMP.

11, still need to edit the Hadoop address, set Dfs.replication to 1 (pseudo-distribution) click Finish.

Note: The installation is at the beginning of the FS. Settings, there is no DFS configuration, this step can be run later in the program, Hadoop configuration items will be changed, and no later.

12, may appear in the left column

　　

13, the environment to build a successful, may not appear. (Since I was running my first program, this DFS locations is not working, but it doesn't affect the program.) ）

14, next to run the first sample program WordCount.