Hadoop官网提供的编译好的hadoop-2.3.0.tar.gz二进制包是在32位系统上编译的，在64系统上运行会有一些错误，比如：

WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable

此时需要自行编译hadoop 2.30 源码。本人编译的hadoop 2.30的二进制包经实验证明可以正确安装并正确运行了Word Count程序。此处可以下载我编译的二进制包:

http://pan.baidu.com/s/1eQrgsWa

以下为编译步骤

1. 更新软件包列表

apt-get update

2. 安装编译所需要的软件: 为什么需要这些依赖包？ 我也不知道==

apt-get install -y openjdk-7-jdk libprotobuf-dev protobuf-compiler maven cmake build-essential pkg-config libssl-dev zlib1g-dev llvm-gcc automake autoconf make

3. 下载hadoop 2.30的源文件包

wget http://archive.apache.org/dist/hadoop/core/hadoop-2.3.0/hadoop-2.3.0-src.tar.gz

4. 解压hadoop 2.30 的源文件包

tar -xzvf hadoop-2.3.0-src.tar.gz

5. 进入hadoop 2.30 文件夹

cd hadoop-2.3.0-src

6. 编译hadoop 2.30 源文件

mvn package -Pdist,native -DskipTests –Dtar

正确执行的结果如下:

[INFO] ------------------------------------------------------------------------

[INFO] Reactor Summary:

[INFO]

[INFO] Apache Hadoop Main ................................ SUCCESS [1:11.968s]

[INFO] Apache Hadoop Project POM ......................... SUCCESS [30.393s]

[INFO] Apache Hadoop Annotations ......................... SUCCESS [18.398s]

[INFO] Apache Hadoop Assemblies .......................... SUCCESS [0.246s]

[INFO] Apache Hadoop Project Dist POM .................... SUCCESS [20.372s]

[INFO] Apache Hadoop Maven Plugins ....................... SUCCESS [23.721s]

[INFO] Apache Hadoop MiniKDC ............................. SUCCESS [1:41.836s]

[INFO] Apache Hadoop Auth ................................ SUCCESS [22.303s]

[INFO] Apache Hadoop Auth Examples ....................... SUCCESS [7.052s]

[INFO] Apache Hadoop Common .............................. SUCCESS [2:29.466s]

[INFO] Apache Hadoop NFS ................................. SUCCESS [11.604s]

[INFO] Apache Hadoop Common Project ...................... SUCCESS [0.073s]

[INFO] Apache Hadoop HDFS ................................ SUCCESS [1:30.230s]

[INFO] Apache Hadoop HttpFS .............................. SUCCESS [17.976s]

[INFO] Apache Hadoop HDFS BookKeeper Journal ............. SUCCESS [19.927s]

[INFO] Apache Hadoop HDFS-NFS ............................ SUCCESS [3.304s]

[INFO] Apache Hadoop HDFS Project ........................ SUCCESS [0.032s]

[INFO] hadoop-yarn ....................................... SUCCESS [0.033s]

[INFO] hadoop-yarn-api ................................... SUCCESS [36.284s]

[INFO] hadoop-yarn-common ................................ SUCCESS [33.912s]

[INFO] hadoop-yarn-server ................................ SUCCESS [0.213s]

[INFO] hadoop-yarn-server-common ......................... SUCCESS [8.193s]

[INFO] hadoop-yarn-server-nodemanager .................... SUCCESS [41.181s]

[INFO] hadoop-yarn-server-web-proxy ...................... SUCCESS [2.768s]

[INFO] hadoop-yarn-server-resourcemanager ................ SUCCESS [13.923s]

[INFO] hadoop-yarn-server-tests .......................... SUCCESS [0.904s]

[INFO] hadoop-yarn-client ................................ SUCCESS [4.363s]

[INFO] hadoop-yarn-applications .......................... SUCCESS [0.120s]

[INFO] hadoop-yarn-applications-distributedshell ......... SUCCESS [2.262s]

[INFO] hadoop-yarn-applications-unmanaged-am-launcher .... SUCCESS [1.615s]

[INFO] hadoop-yarn-site .................................. SUCCESS [0.086s]

[INFO] hadoop-yarn-project ............................... SUCCESS [2.703s]

[INFO] hadoop-mapreduce-client ........................... SUCCESS [0.132s]

[INFO] hadoop-mapreduce-client-core ...................... SUCCESS [18.951s]

[INFO] hadoop-mapreduce-client-common .................... SUCCESS [14.320s]

[INFO] hadoop-mapreduce-client-shuffle ................... SUCCESS [3.330s]

[INFO] hadoop-mapreduce-client-app ....................... SUCCESS [9.664s]

[INFO] hadoop-mapreduce-client-hs ........................ SUCCESS [7.678s]

[INFO] hadoop-mapreduce-client-jobclient ................. SUCCESS [9.263s]

[INFO] hadoop-mapreduce-client-hs-plugins ................ SUCCESS [1.549s]

[INFO] Apache Hadoop MapReduce Examples .................. SUCCESS [5.748s]

[INFO] hadoop-mapreduce .................................. SUCCESS [2.880s]

[INFO] Apache Hadoop MapReduce Streaming ................. SUCCESS [7.080s]

[INFO] Apache Hadoop Distributed Copy .................... SUCCESS [14.648s]

[INFO] Apache Hadoop Archives ............................ SUCCESS [2.602s]

[INFO] Apache Hadoop Rumen ............................... SUCCESS [5.706s]

[INFO] Apache Hadoop Gridmix ............................. SUCCESS [3.649s]

[INFO] Apache Hadoop Data Join ........................... SUCCESS [2.483s]

[INFO] Apache Hadoop Extras .............................. SUCCESS [2.678s]

[INFO] Apache Hadoop Pipes ............................... SUCCESS [6.359s]

[INFO] Apache Hadoop OpenStack support ................... SUCCESS [5.088s]

[INFO] Apache Hadoop Client .............................. SUCCESS [4.534s]

[INFO] Apache Hadoop Mini-Cluster ........................ SUCCESS [0.433s]

[INFO] Apache Hadoop Scheduler Load Simulator ............ SUCCESS [7.757s]

[INFO] Apache Hadoop Tools Dist .......................... SUCCESS [4.099s]

[INFO] Apache Hadoop Tools ............................... SUCCESS [0.428s]

[INFO] Apache Hadoop Distribution ........................ SUCCESS [18.045s]

[INFO] ------------------------------------------------------------------------

[INFO] BUILD SUCCESS

[INFO] ------------------------------------------------------------------------

[INFO] Total time: 14:59.240s

[INFO] Finished at: Thu Jan 15 18:51:59 JST 2015

[INFO] Final Memory: 168M/435M

[INFO] ------------------------------------------------------------------------

编译好的二进制文件包位于

hadoop-2.3.0-src/hadoop-dist/target/hadoop-2.3.0.tar.gz

PS: 使用自行编译的hadoop 2.30二进制包安装hadoop 2.30时需要注意删除 .bashrc文件与hadoop-env.sh文件中下面两行（默认不会有这两行，但是尝试解决报错时可能改写了）

export HADOOP_COMMON_LIB_NATIVE_DIR="~/hadoop/lib/"

export HADOOP_OPTS="$HADOOP_OPTS -Djava.library.path=~/hadoop/lib/"