Ubuntu 11.4 安装 Cloudera CDH3

Part I Hadoop

Click one of the following:
this link for a Maverick system. 安装一个最近的吧
Install the package. Do one of the following:
Choose Open with in the download window to use the package manager, or
Choose Save File, save the package to a directory to which you have write access (it can be your home directory) and install it from the command line, for example:

sudo dpkg -i Downloads/cdh3-repository_1.0_all.deb

3. 查看/etc/apt/sources.list.d$ ls
cloudera-cdh3.list 已经有了，就不需要再设置了

4. (Optionally) add a repository key. Add the Cloudera Public GPG Key to your repository by executing the following command:

$ curl -s http://archive.cloudera.com/debian/archive.key | sudo apt-key add -

This key enables you to verify that you are downloading genuine packages.

sudo apt-get update
sudo apt-get dist-upgrade

6. apt-cache search hadoop

通过这个可以查看所能安装的源

7. 安装pseudo

sudo apt-get install hadoop-0.20-conf-pseudo

8. View how the hadoop-0.20-conf-pseudo packages changes your system.

To view the files on Ubuntu systems:

$ dpkg -L hadoop-0.20-conf-pseudo

9. 启动一个node上的所有deamon（datanode，namenode, secondnamenode, jobtracker, tasktracker):

$forx in /etc/init.d/hadoop-*;dosudo$x start; done

10. Verify the installation

hadoop jar /usr/lib/hadoop/hadoop-*-examples.jar pi 2 100000

以上步骤都在 user1的权限下进行的，你可以改成你自己的。

Part II , Zookeeper

There are two ZooKeeper server packages:

The hadoop-zookeeper base package provides the basic libraries and scripts that are necessary to run ZooKeeper servers and clients. The documentation is also included in this package.
The hadoop-zookeeper-server package contains the init.d scripts necessary to run ZooKeeper as a daemon process. Because hadoop-zookeeper-server depends on hadoop-zookeeper, installing the server package automatically installs the base package.

1. sudo apt-get install hadoop-zookeeper-server

2. To check that zookeeper is started， ps -ef|grep zookeeper

3. if not, please start zookeeper

To start ZooKeeper