hadoop

HOME

  1. #查看主机名
  2. hostname
  3. #修改主机名
  4. sudo hostnamectl set-hostname linux01
  6. # 关闭防火墙，关闭防火墙开机自启
  7. systemctl stop firewalld
  8. systemctl disable firewalld.service
  10. #在/etc/profile.d文件下新建文件my_home.sh,把下面的写入文件中
  12. export HADOOP_HOME=/opt/hadoop-3.1.3
  13. export PATH=$PATH:$HADOOP_HOME/bin
  14. export PATH=$PATH:$HADOOP_HOME/sbin
  15. export JAVA_LIBRARY_PATH=$PATH:$HADOOP_HOME/lib/native
  17. export JAVA_HOME=/opt/jdk1.8.0_321
  18. export PATH=$PATH:$JAVA_HOME/bin
  20. export SCALA_HOME=/opt/scala-2.12.16
  21. export PATH=$PATH:$SCALA_HOME/bin
  23. export SBT_HOME=/opt/sbt
  24. export PATH=$PATH:$SBT_HOME/bin
  26. export SPARK_HOME=/opt/spark-yarn
  27. export H=$PATH:$SPARK_HOME/bin
  29. source /etc/profile
  1. #查看主机名
  2. hostname
  3. #修改主机名
  4. sudo hostnamectl set-hostname linux01
  8. cat  /etc/sysconfig/network-scripts/ifcfg-ens33
  9. cat > /etc/sysconfig/network-scripts/ifcfg-ens33
  11. DEVICE=ens33
  12. TYPE=Ethernet
  13. ONBOOT=yes
  14. BOOTPROTO=static
  15. NAME="ens33"
  16. IPADDR=192.168.10.100
  17. PREFIX=24
  18. GATEWAY=192.168.10.2
  19. DNS1=192.168.10.2
  22. # 重启
  23. reboot
  25. cat >/etc/hosts
  28. 192.168.10.100 linux00
  29. 192.168.10.101 linux01
  30. 192.168.10.102 linux02
  31. 192.168.10.103 linux03
  32. 192.168.10.104 linux04
  33. 192.168.10.105 linux05
  34. 192.168.10.106 linux06
  35. 192.168.10.107 linux07
  36. 192.168.10.108 linux08
  37. 192.168.10.109 linux09
  1. ifconfig ens33 up
  2. systemctl stop NetworkManager
  3. ifup ens33
  4. systemctl restart network.service
  5. ifconfig
  1. #!/bin/bash
  3. echo "=========================================="
  4. java -version
  6. echo "=========================================="
  7. hadoop version
  9. echo "=========================================="
  10. scala -version
  12. echo "=========================================="
  13. sbt --version
  16. echo "=========================================="
  17. $SPARK_HOME/bin/spark-shell

免密登录

  1. ssh-keygen -t rsa
  3. ssh-copy-id linux02
  4. ssh-copy-id linux03
  5. ssh-copy-id linux04

分发脚本

chmod 777 /bin/sync

  1. #!/bin/bash
  3. #1. 判断参数个数
  4. if [ $# -lt 1 ]
  5. then
  6.     echo Not Enough Arguement!
  7.     exit;
  8. fi
  10. #2. 遍历集群所有机器
  11. for host in linux02 linux03 linux04
  12. do
  13.     echo ====================  $host  ====================
  14.     #3. 遍历所有目录，挨个发送
  16.     for file in $@
  17.     do
  18.         #4. 判断文件是否存在
  19.         if [ -e $file ]
  20.             then
  21.                 #5. 获取父目录
  22.                 pdir=$(cd -P $(dirname $file); pwd)
  24.                 #6. 获取当前文件的名称
  25.                 fname=$(basename $file)
  26.                 ssh $host "mkdir -p $pdir"
  27.                 rsync -av $pdir/$fname $host:$pdir
  28.             else
  29.                 echo $file does not exists!
  30.         fi
  31.     done
  32. done

集群部署规划

NameNode和SecondaryNameNode不要安装在同一台服务器 ResourceManager也很消耗内存，不要和NameNode、SecondaryNameNode配置在同一台机器上。

linux02 linux03 linux04
HDFS
 NameNode
DataNode
JobHistoryServer

DataNode		 SecondaryNameNode
DataNode
YARN

NodeManager		 ResourceManager
NodeManager

NodeManager

conf

/opt/hadoop-3.1.3/etc/hadoop

  2. <configuration>
  3.  <!-- 指定 NameNode 的地址 -->
  4.  <property>
  5.    <name>fs.defaultFS</name>
  6.    <value>hdfs://linux02:8020</value>
  7.  </property>
  8.  <!-- 指定 hadoop 数据的存储目录 -->
  9.  <property>
  10.    <name>hadoop.tmp.dir</name>
  11.    <value>/opt/hadoop-3.1.3/data</value>
  12.  </property>
  13.  <!-- 配置 HDFS 网页登录使用的静态用户为 root -->
  14.  <property>
  15.    <name>hadoop.http.staticuser.user</name>
  16.    <value>root</value>
  17.  </property>
  18. </configuration>

<configuration>
<!-- nn web 端访问地址-->
<property>
 <name>dfs.namenode.http-address</name>
 <value>linux02:9870</value>
 </property>
<!-- 2nn web 端访问地址-->
 <property>
 <name>dfs.namenode.secondary.http-address</name>
 <value>linux04:9868</value>
 </property>
</configuration>

<configuration>
 <!-- 指定 MR 走 shuffle -->
 <property>
 <name>yarn.nodemanager.aux-services</name>
 <value>mapreduce_shuffle</value>
 </property>
 <!-- 指定 ResourceManager 的地址-->
 <property>
 <name>yarn.resourcemanager.hostname</name>
 <value>linux03</value>
 </property>
 <!-- 环境变量的继承 -->
 <property>
 <name>yarn.nodemanager.env-whitelist</name>

<value>JAVA_HOME,HADOOP_COMMON_HOME,HADOOP_HDFS_HOME,HADOOP_CO
NF_DIR,CLASSPATH_PREPEND_DISTCACHE,HADOOP_YARN_HOME,HADOOP_MAP
RED_HOME</value>
 </property>


 <!-- 开启日志聚集功能 -->
<property>
    <name>yarn.log-aggregation-enable</name>
    <value>true</value>
</property>
<!-- 设置日志聚集服务器地址 -->
<property>  
    <name>yarn.log.server.url</name>  
    <value>http://linux02:19888/jobhistory/logs</value>
</property>
<!-- 设置日志保留时间为7天 -->
<property>
    <name>yarn.log-aggregation.retain-seconds</name>
    <value>604800</value>
</property>

</configuration>

<configuration>
<!-- 指定 MapReduce 程序运行在 Yarn 上 -->
 <property>
   <name>mapreduce.framework.name</name>
   <value>yarn</value>
 </property>

<!-- 历史服务器端地址 -->
<property>
   <name>mapreduce.jobhistory.address</name>
   <value>linux02:10020</value>
</property>  

<!-- 历史服务器web端地址 -->
<property>
    <name>mapreduce.jobhistory.webapp.address</name>
    <value>linux02:19888</value>
</property>


   <property>
        <name>yarn.app.mapreduce.am.env</name>
        <value>HADOOP_MAPRED_HOME=/opt/hadoop-3.1.3</value>
    </property>
    <property>
        <name>mapreduce.map.env</name>
        <value>HADOOP_MAPRED_HOME=/opt/hadoop-3.1.3</value>
    </property>
    <property>
        <name>mapreduce.reduce.env</name>
        <value>HADOOP_MAPRED_HOME=/opt/hadoop-3.1.3</value>
    </property>


</configuration>

linux02
linux03
linux04

初始化

hdfs namenode -format

启动脚本

#!/bin/bash

if [ $# -lt 1 ]
then
    echo "No Args Input..."
    exit ;
fi

case $1 in
"start")
        echo " =================== 启动 hadoop集群 ==================="

        echo " --------------- 启动 hdfs ---------------"
        ssh linux02 "/opt/hadoop-3.1.3/sbin/start-dfs.sh"
        echo " --------------- 启动 yarn ---------------"
        ssh linux03 "/opt/hadoop-3.1.3/sbin/start-yarn.sh"
        echo " --------------- 启动 historyserver ---------------"
        ssh linux02 "/opt/hadoop-3.1.3/bin/mapred --daemon start historyserver"
;;
"stop")
        echo " =================== 关闭 hadoop集群 ==================="

        echo " --------------- 关闭 historyserver ---------------"
        ssh linux02 "/opt/hadoop-3.1.3/bin/mapred --daemon stop historyserver"
        echo " --------------- 关闭 yarn ---------------"
        ssh linux03 "/opt/hadoop-3.1.3/sbin/stop-yarn.sh"
        echo " --------------- 关闭 hdfs ---------------"
        ssh linux02 "/opt/hadoop-3.1.3/sbin/stop-dfs.sh"
;;
*)
    echo "Input Args Error..."
;;
esac

查看所有进程脚本

#!/bin/bash

for host in linux02 linux03 linux04
do
        echo =============== $host ===============
        ssh $host jps 
done

格式化Hadoop

先停进程
在删除data logs
再格式化 hdfs namenode -format
在启动集群

test hadoop

# 创建文件夹
hadoop fs -mkdir /myinput

hadoop fs -put wcinput/word.txt /myinput

hadoop fs -put 文件路径 hdfs中的路径

hadoop fs -put jdk-8u301-linux-x64.rpm /myinput

hadoop jar /opt/hadoop-3.1.3/share/hadoop/mapreduce/hadoop-mapreduce-examples-3.1.3.jar wordcount /myinput /output1

spark

local env

sc.textFile("data/word.txt").flatMap(_.split(" ")).map((_,1)).reduceByKey(_+_).collect


bin/spark-submit \
--class org.apache.spark.examples.SparkPi \
--master local[*] \
./examples/jars/spark-examples_2.12-3.0.0.jar \
10

yarn env

./bin/spark-submit \
--class org.apache.spark.examples.SparkPi \
--master yarn \
--deploy-mode client \
./examples/jars/spark-examples_2.12-3.0.0.jar 10


# 启动Spark 集群
sbin/start-all.sh
# 启动Spark 历史服务器
sbin/start-history-server.sh

$SPARK_HOME/bin/spark-submit \
--class org.apache.spark.examples.SparkPi \
--master yarn \
--properties-file /home/job1/conf.conf \
$SPARK_HOME/examples/jars/spark-examples_2.12-3.0.0.jar 10