基于Linux的Spark安装与环境配置

2023-12-28 19:16:49

基于Linux的Spark安装与环境配置

1、Hadoop测试

因为Spark是基于Hadoop上工作的,所以当我们使用Spark框架时,必须要确保Hadoop能够正常运行:

1.1 启动hadoop

cd /usr/local/hadoop
./sbin/start-all.sh

BUG,内容如下:

SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".
SLF4J: Defaulting to no-operation (NOP) logger implementation
SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details.

解决方法:SLF4J:Failed to load class org.slf4j.impl.StaticLoggerBinder.

1.2 再次启动hadoop

cd /usr/local/hadoop/sbin
./start-all.sh

1.3 查看是否成功

jps

2、Scala安装与配置

2.1 下载scala

官方网址:https://www.scala-lang.org/download/2.13.10.html

使用wget命令下载scala:

wget https://downloads.lightbend.com/scala/2.13.10/scala-2.13.10.tgz

2.2 解压并重命名

sudo tar zxvf ~/下载/scala-2.13.10.tgz -C /usr/local/ # 解压

cd /usr/local
sudo mv scala-2.13.10 scala # 重命名

2.3 配置环境

# 1.编辑环境变量
sudo vi ~/.bashrc

# 2.使其生效
source ~/.bashrc

2.4 测试

scala -version

3、Spark安装与配置

3.1 下载Spark

下载网址:https://archive.apache.org/dist/spark/spark-3.2.2/

使用wget命令进行下载:

wget https://archive.apache.org/dist/spark/spark-3.2.2/spark-3.2.2-bin-hadoop3.2.tgz

3.2 解压并重命名

# 1.解压
sudo tar zxvf ~/下载/spark-3.2.2-bin-hadoop3.2.tgz  -C /usr/local

# 2.重命名
cd /usr/local
sudo mv spark-3.2.2-bin-hadoop3.2 spark

3.3 配置环境

# 1.编辑环境变量
sudo vi ~/.bashrc

# 2.使其生效
source ~/.bashrc

3.4 配置spark-env.sh

进入到配置目录并打开spark-env.sh文件:

cd /usr/local/spark/conf
sudo cp spark-env.sh.template spark-env.sh
sudo vi spark-env.sh

添加以下内容:

export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-amd64
export HADOOP_HOME=/usr/local/hadoop
export HADOOP_CONF_DIR=/usr/local/hadoop/etc/hadoop
export SCALA_HOME=/usr/local/scala
export SPARK_HOME=/usr/local/spark
export SPARK_MASTER_IP=192.168.3.134
export SPARK_MASTER_PORT=7077
export SPARK_MASTER_WEBUI_PORT=8099
export SPARK_WORKER_CORES=3
export SPARK_WORKER_INSTANCES=1
export SPARK_WORKER_MEMORY=5G
export SPARK_WORKER_WEBUI_PORT=8081
export SPARK_EXECUTOR_CORES=1
export SPARK_EXECUTOR_MEMORY=1G
export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:$HADOOP_HOME/lib/native

3.5 配置slaves(好像不需要)

cd /usr/local/spark/conf
sudo vi workers.template

发现slaves文件里为localhost即本机地址,当前为伪分布式,因此不用修改!但要执行以下:

sudo cp workers.template  slaves

3.6 启动(报错)

启动sbin目录下的start-master.sh以及start-slaves.sh前提是hadoop已启动):

cd /usr/local/spark
sudo ./sbin/start-master.sh
sudo ./sbin/start-slaves.sh

报错!!!

3.7 测试

通过运行Spark自带的示例,验证Spark是否安装成功:

cd /usr/local/spark
./bin/run-example SparkPi

报错信息如下:

2022-11-01 20:49:24,377 WARN util.Utils: Your hostname, leoatliang-virtual-machine resolves to a loopback address: 127.0.1.1; using 192.168.3.134 instead (on interface ens33)

参考博客:Spark启动:WARN util.Utils: Your hostname, … resolves to a loopback address: …; using … instead

修改配置文件,配置 SPARK_LOCAL_IP 变量即可:

cd /usr/local/spark
sudo vim conf/spark-env.sh

# 添加以下内容:
export SPARK_LOCAL_IP=192.168.3.134  # 自己输出对应的IP

再次测试:

BUG解决!!!

执行时会输出非常多的运行信息,输出结果不容易找到,可以通过 grep 命令进行过滤:

./bin/run-example SparkPi 2>&1 | grep "Pi is"

3.8 查看Spark版本

cd /usr/local/spark
./bin/spark-shell

文章来源:https://blog.csdn.net/m0_70885101/article/details/127641702
本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。