Win7 MyEclipse remote connection to Hadoop cluster in Mac/linux

Win7 myeclipse remote connection to Hadoop cluster in Mac/linux
(You can also visit this page to view: http://tn.51cto.com/article/562)
Required Software:
(1) Download Hadoop2.5.1 to Win7 system, and unzip
hadoop2.5.1:indexof/dist/hadoop/core/hadoop-2.5.1
Http://archive.apache.org/dist/hadoop/core/hadoop-2.5.1/

Description: The version of Hadoop that is downloaded in Win7 is Hadoop2.5, the version is inconsistent, but when tested, the result is no error, but some people say that the version is the best when tested, so you can keep an eye on it.
(2) Hadoop runs under a Linux system, running Hadoop programs under Windows Eclipse, first installing the support plug-in hadoop2.2-common-bin.zip running under Windows, The following: Https://github.com/srccodes/hadoop-common-2.2.0-bin
Description: Hadoop2.2-common-bin.zip is used here, Inconsistent with the Hadoop version, but the results are not error-free and are not reported as abnormal. Not configuring the plugin to run Hadoop will report an exception:
Could not locate executable null\bin\winutils.exe in the Hadoop binaries.
(3) Configuring the Hadoop environment variable (somewhat similar to Java's environment configuration)
Windows environment variable configuration: Right-click My Computer –> properties, advanced environment variable Configuration –> Advanced tab, environment variable, Click New hadoop_home-> configuration to download the Hadoop path.

(4) Download the plug-in required to develop Hadoop in Hadoop-eclipse-plugin-2.2.0.jar,myeclipse/eclipse. Download and copy it to the following location on the MyEclipse installation path;

(5) Restarting MyEclipse or eclipse will appear the icon on the right "blue elephant", indicating that the plugin Hadoop plugin was successfully installed.

(6) Eclipse Environment Configuration: Click "Menu Window"-->preference-->hadoop map/reduce to configure, as shown in:

(7) The view shown appears, indicating a successful configuration.

(8) switch to configuration file and modify
①CD Hadoop/hadoop2.5/etc/hadoop
②cat Core-site.xml
When you configure Hadoop pseudo-distribution, you typically use localhost as the Namenode identity.

It needs to be modified to the hostname or IP address of Linux, otherwise the Hadoop program under Windows can't connect to Linux or Hadoop on Mac to run and error.

, indicating that the connection failed:

Description: If you are connecting using a host name, you need to configure the hostname-to-IP mapping.
：

(9) Start Hadoop in your Mac or Linux and use the JPS command to see if it is running successfully.


(10) Configuring HDFs, creating new Hadooop location or edit Hadoop location

Specific configuration instructions:

(11) Test: see if you can access HDFs in Mac or Linux. As shown in:

, which shows a file in HDFs.

Win7 MyEclipse remote connection to Hadoop cluster in Mac/linux