Spark大数据实战之五：Spark安装

Apache官网下载Spark，拖拽到master节点上lala用户目录下，解压
https://archive.apache.org/dist/spark/spark-1.4.0/

执行如下语句进行解压缩

tar -zvxf spark-1.4.0-bin-hadoop2.4.tgz

(1) 进入spark-1.4.0-bin-hadoop2.4/conf，复制其中的spark-env.sh.template并另存为该目录下的spark-env.sh

cd spark-1.4.0-bin-hadoop2.4/conf
cp spark-env.sh.template spark-env.sh

(2)打开spark-env.sh

gedit spark-env.sh

(3) 将如下代码添加到spark-env.sh中

export HADDOP_CONF_DIR=/home/lala/hadoop-2.5.2/
export JAVA_HOME=/usr/java/jdk1.7.0_71/
export SCALA_HOME=/usr/scala-2.10.4
export SPARK_MASTER_IP=192.168.149.132
export SPARK_MASTER_PORT=7077
export SPARK_MASTER_WEBUI_PORT=8080
export SPARK_WORKER_PORT=7078
export SPARK_WORKER_WEBUI_PORT=8081
export SPARK_WORKER_CORES=1
export SPARK_WORKER_INSTRANCES=1
export SPARK_WORKER_MEMORY=2g
export SPARK_JAR=/home/lala/spark-1.4.0-bin-hadoop2.4/lib/spark-assembly-1.4.0-hadoop2.4.0.jar

(1)进入spark-1.4.0-bin-hadoop2.4/conf，复制其中的spark-defaults.conf.template并另存为该目录下的spark-defaults.conf

cd spark-1.4.0-bin-hadoop2.4/conf
cp spark-defaults.conf.template spark-defaults.conf

(2)打开spark-defaults.conf

gedit spark-defaults.conf

(3)向spark-defaults.conf添加如下代码，根据用户ip进行修改

spark.master=spark://192.168.109.134:7077

(1)进入spark-1.4.0-bin-hadoop2.4/conf，复制其中的slaves.template并另存为该目录下的slaves

cp slaves.template slaves

(2)打开spark-defaults.conf

gedit slaves

(3) 在slaves中添加子节点机器名或者IP地址

192.168.109.133
192.168.109.134