以前在32位linux机器上编译过hadoop2.6.0,这次在redhat7 64bit上再次编译hadoop2.6.0，除必须的jdk,maven,protobuf需要安装之外，还需要安装系统依赖库gcc,gcc-c++,ncurses-devel,openssl-devel,cmake,这些依赖库的安装可以直接通过yum命令一下安装: yum install -y gcc gcc-c++ cmake openssl-devel ncurses-devel,依赖库之间通过空格分割。
这次的安装还是遇到了只能使用jdk1.7编译的问题。如果使用jdk1.8，会报如下图所示的错误。[ERROR] Exit code:1 - /home/hadoop/hadoop-2.6.0-src/hadoop-common-project/hadoop-annotations/src/main/java/org/apache/hadoop/classification/InterfaceStability.java:27: error: unexcepted end tag: </ul>
遇到这个错误，我们只能将jdk换成1.7，之后顺利编译通过。
编译命令：mvn package -DskipTests -Pdist,native -Dtar
build success
INFO] ------------------------------------------------------------------------
[INFO] Reactor Summary:
[INFO]
[INFO] Apache Hadoop Main ................................. SUCCESS [ 1.041 s]
[INFO] Apache Hadoop Project POM .......................... SUCCESS [ 1.018 s]
[INFO] Apache Hadoop Annotations .......................... SUCCESS [ 1.937 s]
[INFO] Apache Hadoop Assemblies ........................... SUCCESS [ 0.255 s]
[INFO] Apache Hadoop Project Dist POM ..................... SUCCESS [ 15.730 s]
[INFO] Apache Hadoop Maven Plugins ........................ SUCCESS [ 14.448 s]
[INFO] Apache Hadoop MiniKDC .............................. SUCCESS [ 25.414 s]
[INFO] Apache Hadoop Auth ................................. SUCCESS [ 15.027 s]
[INFO] Apache Hadoop Auth Examples ........................ SUCCESS [ 7.052 s]
[INFO] Apache Hadoop Common ............................... SUCCESS [02:01 min]
[INFO] Apache Hadoop NFS .................................. SUCCESS [ 8.556 s]
[INFO] Apache Hadoop KMS .................................. SUCCESS [02:31 min]
[INFO] Apache Hadoop Common Project ....................... SUCCESS [ 0.029 s]
[INFO] Apache Hadoop HDFS ................................. SUCCESS [02:46 min]
[INFO] Apache Hadoop HttpFS ............................... SUCCESS [02:42 min]
[INFO] Apache Hadoop HDFS BookKeeper Journal .............. SUCCESS [ 14.788 s]
[INFO] Apache Hadoop HDFS-NFS ............................. SUCCESS [ 3.519 s]
[INFO] Apache Hadoop HDFS Project ......................... SUCCESS [ 0.029 s]
[INFO] hadoop-yarn ........................................ SUCCESS [ 0.037 s]
[INFO] hadoop-yarn-api .................................... SUCCESS [01:12 min]
[INFO] hadoop-yarn-common ................................. SUCCESS [ 25.486 s]
[INFO] hadoop-yarn-server ................................. SUCCESS [ 0.044 s]
[INFO] hadoop-yarn-server-common .......................... SUCCESS [ 11.236 s]
[INFO] hadoop-yarn-server-nodemanager ..................... SUCCESS [ 19.452 s]
[INFO] hadoop-yarn-server-web-proxy ....................... SUCCESS [ 2.385 s]
[INFO] hadoop-yarn-server-applicationhistoryservice ....... SUCCESS [ 5.028 s]
[INFO] hadoop-yarn-server-resourcemanager ................. SUCCESS [ 15.531 s]
[INFO] hadoop-yarn-server-tests ........................... SUCCESS [ 4.019 s]
[INFO] hadoop-yarn-client ................................. SUCCESS [ 5.003 s]
[INFO] hadoop-yarn-applications ........................... SUCCESS [ 0.044 s]
[INFO] hadoop-yarn-applications-distributedshell .......... SUCCESS [ 2.060 s]
[INFO] hadoop-yarn-applications-unmanaged-am-launcher ..... SUCCESS [ 1.525 s]
[INFO] hadoop-yarn-site ................................... SUCCESS [ 0.047 s]
[INFO] hadoop-yarn-registry ............................... SUCCESS [ 4.122 s]
[INFO] hadoop-yarn-project ................................ SUCCESS [ 5.211 s]
[INFO] hadoop-mapreduce-client ............................ SUCCESS [ 0.051 s]
[INFO] hadoop-mapreduce-client-core ....................... SUCCESS [ 16.098 s]
[INFO] hadoop-mapreduce-client-common ..................... SUCCESS [ 12.579 s]
[INFO] hadoop-mapreduce-client-shuffle .................... SUCCESS [ 3.310 s]
[INFO] hadoop-mapreduce-client-app ........................ SUCCESS [ 6.972 s]
[INFO] hadoop-mapreduce-client-hs ......................... SUCCESS [ 5.790 s]
[INFO] hadoop-mapreduce-client-jobclient .................. SUCCESS [ 7.923 s]
[INFO] hadoop-mapreduce-client-hs-plugins ................. SUCCESS [ 1.437 s]
[INFO] Apache Hadoop MapReduce Examples ................... SUCCESS [ 4.272 s]
[INFO] hadoop-mapreduce ................................... SUCCESS [ 4.270 s]
[INFO] Apache Hadoop MapReduce Streaming .................. SUCCESS [ 5.100 s]
[INFO] Apache Hadoop Distributed Copy ..................... SUCCESS [ 12.644 s]
[INFO] Apache Hadoop Archives ............................. SUCCESS [ 1.797 s]
[INFO] Apache Hadoop Rumen ................................ SUCCESS [ 4.551 s]
[INFO] Apache Hadoop Gridmix .............................. SUCCESS [ 3.380 s]
[INFO] Apache Hadoop Data Join ............................ SUCCESS [ 2.232 s]
[INFO] Apache Hadoop Ant Tasks ............................ SUCCESS [ 2.054 s]
[INFO] Apache Hadoop Extras ............................... SUCCESS [ 2.312 s]
[INFO] Apache Hadoop Pipes ................................ SUCCESS [ 8.269 s]
[INFO] Apache Hadoop OpenStack support .................... SUCCESS [ 4.217 s]
[INFO] Apache Hadoop Amazon Web Services support .......... SUCCESS [ 29.522 s]
[INFO] Apache Hadoop Client ............................... SUCCESS [ 7.187 s]
[INFO] Apache Hadoop Mini-Cluster ......................... SUCCESS [ 0.106 s]
[INFO] Apache Hadoop Scheduler Load Simulator ............. SUCCESS [ 3.483 s]
[INFO] Apache Hadoop Tools Dist ........................... SUCCESS [ 8.224 s]
[INFO] Apache Hadoop Tools ................................ SUCCESS [ 0.025 s]
[INFO] Apache Hadoop Distribution ......................... SUCCESS [ 49.890 s]
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 18:13 min
[INFO] Finished at: 2017-06-23T03:56:38+08:00
[INFO] Final Memory: 157M/350M
[INFO] ------------------------------------------------------------------------
截图如下：
如果不通过源码编译安装，在hadoop,spark,flink等启动时经常会提示一个警告:
2017-06-23 21:57:49,105 WARN org.apache.hadoop.util.NativeCodeLoader - Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
