大数据环境部署6：Spark环境部署



1、下载scala2.11.4版本下载地址为：http://www.scala-lang.org/download/2.11.4.html ，也可以使用wget http://downloads.typesafe.com/scala/2.11.4/scala-2.11.4.tgz?_ga=1.248348352.61371242.1418807768 

2、解压和安装：解压：[spark@LOCALHOST scala]$ tar -xvfscala-2.11.4.tgz  ,安装：[spark@LOCALHOST scala]$ mv scala-2.11.4 ~/opt/

3、编辑 ~/.bash_profile文件增加SCALA_HOME环境变量配置，

exportJAVA_HOME=/usr/java/jdk1.7.0_79
export CLASSPATH=.:$JAVA_HOME/jre/lib:$JAVA_HOME/lib:$JAVA_HOME/lib/tools.jar
export SCALA_HOME=/home/spark/opt/scala-2.11.4
export HADOOP_HOME=/home/spark/opt/hadoop-2.6.0
PATH=$PATH:$HOME/bin:$JAVA_HOME/bin:${SCALA_HOME}/bin

立即生效 bash_profile  ，[spark@LOCALHOST scala]$ source ~/.bash_profile 

4、验证scala：[spark@LOCALHOST scala]$ scala -version
Scala code runner version 2.11.4 -- Copyright 2002-2013, LAMP/EPFL

[spark@LOCALHOSTscala]$ scala
Welcome to Scala version 2.11.4 (Java HotSpot(TM) 64-Bit Server VM, Java1.6.0_37).
Type in expressions to have them evaluated.
Type :help for more information.

scala> varstr = "SB is"+"SB"
str: String = SB isSB

scala> 

 

5、copy到slave机器，

[spark@LOCALHOSTscala]$ scp  ~/.bash_profile  spark@172.16.107.8:~/.bash_profile 

[spark@LOCALHOSTscala]$ scp  ~/.bash_profile  spark@172.16.107.7:~/.bash_profile 

 

6、下载spark，wgethttp://d3kbcqa49mib13.cloudfront.net/spark-1.2.0-bin-hadoop2.4.tgz

7、在master主机配置spark：

将下载的spark-1.2.0-bin-hadoop2.4.tgz解压到 ~/opt/，即 ~/opt/spark-1.2.0-bin-hadoop2.4，配置环境变量SPARK_HOME

# set  javaenv
export JAVA_HOME=/usr/java/jdk1.7.0_79
export CLASSPATH=.:$JAVA_HOME/jre/lib:$JAVA_HOME/lib:$JAVA_HOME/lib/tools.jar
export SCALA_HOME=/home/spark/opt/scala-2.11.4
export HADOOP_HOME=/home/spark/opt/hadoop-2.6.0
export SPARK_HOME=/home/spark/opt/spark-1.2.0-bin-hadoop2.4
PATH=$PATH:$HOME/bin:$JAVA_HOME/bin:${SCALA_HOME}/bin:${SPARK_HOME}/bin:${HADOOP_HOME}/bin

配置完成后使用source命令使配置生效

进入 spark conf目录：

[spark@LOCALHOST opt]$ cd spark-1.2.0-bin-hadoop2.4/
[spark@LOCALHOST spark-1.2.0-bin-hadoop2.4]$ ls
bin  conf  data  ec2  examples  lib  LICENSE logs  NOTICE  python  README.md  RELEASE  sbin work
[spark@LOCALHOST spark-1.2.0-bin-hadoop2.4]$ cd conf/
[spark@LOCALHOST conf]$ ls
fairscheduler.xml.template  metrics.properties.template slaves.template              spark-env.sh
log4j.properties.template   slaves                      spark-defaults.conf.template spark-env.sh.template

first：修改slaves文件，增加三个slave节点172.16.107.9、172.16.107.8、172.16.107.7

[spark@LOCALHOSTconf]$ vi slaves
172.16.107.9

172.16.107.8

172.16.107.7

second：配置spark-env.sh

首先把spark-env.sh.template copy spark-env.sh

vi spark-env.sh文件在最下面增加：

exportJAVA_HOME=/usr/java/jdk1.7.0_79
export SCALA_HOME=/home/spark/opt/scala-2.11.4
export SPARK_MASTER_IP=172.16.107.9
export SPARK_WORKER_MEMORY=2g
export HADOOP_CONF_DIR=/home/spark/opt/hadoop-2.6.0/etc/hadoop

HADOOP_CONF_DIR是Hadoop配置文件目录，SPARK_MASTER_IP主机IP地址，SPARK_WORKER_MEMORY是worker使用的最大内存

完成配置后，将spark目录copy slave机器

scp -r~/opt/spark-1.2.0-bin-hadoop2.4  spark@172.16.107.8:~/opt/

scp -r~/opt/spark-1.2.0-bin-hadoop2.4  spark@172.16.107.7:~/opt/

 

8、启动spark分布式集群并查看信息

[spark@LOCALHOSTsbin]$ ./start-all.sh 

查看：

[spark@LOCALHOSTsbin]$ jps
31233 ResourceManager
27201 Jps
30498 NameNode
30733 SecondaryNameNode
5648 Worker
5399 Master
15888 JobHistoryServer
如果HDFS没有启动，请启动起来。

查看slave节点：

[spark@localhostscala]$ jps
20352 Bootstrap
30737 NodeManager
7219 Jps
30482 DataNode
29500 Bootstrap
757 Worker

9、页面查看集群状况：

进去spark集群的web管理页面，访问

http://172.16.107.9:8080/

 

进入spark的bin目录，启动spark-shell控制台

[spark@localhostbin]$ sh spark-shell

访问http:// 172.16.107.9:4040/，可以看到spark WEBUI页面

 

到目前为止，spark集群环境搭建成功了。

 

 

参考：

http://blog.csdn.net/stark_summer/article/details/42458081