apache 网站上面已经有了已经构建好了的版本，我这里还是自己利用午休时间重新构建一下（jdk,python,scala的安装就省略了，自己可以去安装）

http://www.apache.org/dist/spark/spark-1.0.1/ 具体官网的下载链接可以去这里

我下载的是http://www.apache.org/dist/spark/spark-1.0.1/spark-1.0.1.tgz源码包

下载对应的linux服务器上面，然后解压

wget http://www.apache.org/dist/spark/spark-1.0.1/spark-1.0.1.tgztar zxf spark-1.0.1.tgz
cd spark-1.0.1

我们可以大概的ls浏览一下目录中有哪些文件，心中有个数

如果想要构建对应的安装包，apache已经为我们提供了一个构建脚本make-distribution.sh（也可以自己去参考官方提供的文档http://spark.apache.org/docs/latest/building-with-maven.html 不过脚本里面都给我们封装好了）

可以查看下脚本，里面肯定有介绍怎么用，看到下面这一段

# Optional Arguments
#      --tgz: Additionally creates spark-$VERSION-bin.tar.gz
#      --hadoop VERSION: Builds against specified version of Hadoop.
#      --with-yarn: Enables support for Hadoop YARN.
#      --with-hive: Enable support for reading Hive tables.
#      --name: A moniker for the release target. Defaults to the Hadoop verison.

下面我们开始执行构建命令

sh make-distribution.sh --hadoop 2.2.0 --with-yarn --with-hive --tgz

漫长的等待......

公司网络真慢，maven下载一些依赖要下老半天，人家午休都睡醒了，我还在呆呆的看着屏幕在滚...

不是一般的坑爹，中间网络断了几次导致maven下载jar失败，重试了几次，最后终于成功了，容易么

[INFO] ------------------------------------------------------------------------
[INFO] Reactor Summary:
[INFO]
[INFO] Spark Project Parent POM .......................... SUCCESS [8.214s]
[INFO] Spark Project Core ................................ SUCCESS [17:02.599s]
[INFO] Spark Project Bagel ............................... SUCCESS [3:59.860s]
[INFO] Spark Project GraphX .............................. SUCCESS [3:19.409s]
[INFO] Spark Project ML Library .......................... SUCCESS [4:58.597s]
[INFO] Spark Project Streaming ........................... SUCCESS [5:04.713s]
[INFO] Spark Project Tools ............................... SUCCESS [1:15.190s]
[INFO] Spark Project Catalyst ............................ SUCCESS [3:44.361s]
[INFO] Spark Project SQL ................................. SUCCESS [3:29.240s]
[INFO] Spark Project Hive ................................ SUCCESS [8:23.878s]
[INFO] Spark Project REPL ................................ SUCCESS [3:39.937s]
[INFO] Spark Project YARN Parent POM ..................... SUCCESS [14.723s]
[INFO] Spark Project YARN Stable API ..................... SUCCESS [1:54.831s]
[INFO] Spark Project Assembly ............................ SUCCESS [2:10.078s]
[INFO] Spark Project External Twitter .................... SUCCESS [3:02.286s]
[INFO] Spark Project External Kafka ...................... SUCCESS [1:49.284s]
[INFO] Spark Project External Flume ...................... SUCCESS [1:24.960s]
[INFO] Spark Project External ZeroMQ ..................... SUCCESS [1:08.075s]
[INFO] Spark Project External MQTT ....................... SUCCESS [58.880s]
[INFO] Spark Project Examples ............................ SUCCESS [9:48.584s]
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 1:17:40.063s
[INFO] Finished at: Wed Jul 23 03:22:57 PDT 2014
[INFO] Final Memory: 43M/393M
[INFO] ------------------------------------------------------------------------

花了一个多小时，真受不了，下面再来看下构建成功后多出了什么

在spark根目录下面会多出一个 spark-1.0.1-bin-2.2.0.tgz的文件，这也就是我们需要部署到其他机器的文件，

完成。