0-1 伪分布hadoop和hbase搭建

时间:2021-01-20 08:27:24

Hadoop-2.2.0伪分布式和hbase-0.96.0-hadoop2伪分布式搭建

1. 部署

(1)主机名

gedit /etc/hostname  

master1

sudo gedit /etc/hosts 

127.0.0.1 master1

(2)SSH

sudo apt-get install ssh

mkdir /home/hadoop/.ssh

ssh-keygen -t rsa  

cat id_rsa.pub >> authorized_keys

ssh localhost

ssh master1

(3)环境变量

exportANT_HOME=/opt/apache-ant-1.9.3

exportMAVEN_HOME=/opt/apache-maven-3.0.5

exportFINDBUGS_HOME=/opt/findbugs-2.0.2

exportPATH=${ANT_HOME}/bin:${MAVEN_HOME}/bin:${FINDBUGS_HOME}/bin:$PATH

#jdk

exportJAVA_HOME=/opt/jdk1.7.0_45

exportJRE_HOME=/opt/jdk1.7.0_45

exportJRE_HOME=/opt/jdk1.7.0_45/jre

exportCLASSPATH=.:$JAVA_HOME/lib:$JRE_HOME/lib:$CLASSPATH

exportPATH=$JAVA_HOME/bin:$JRE_HOME/bin:$PATH

#hadoop

exportHADOOP_HOME=/opt/hadoop-2.2.0

exportHADOOP_CONF_DIR=${HADOOP_HOME}/etc/hadoop

exportPATH=$PATH:$HADOOP_HOME/bin:$HADOOP_HOME/sbin

#hbase

exportHBASE_HOME=/opt/hbase-0.96.0-hadoop2

exportPATH=$PATH:$HBASE_HOME/bin

(4)放置

/opt/jdk1.7.0_45

/opt/hadoop-2.2.0

/opt/hbase-0.96.0-hadoop2

(5)本地目录创建(根据下面的配置文件)

mkdir -p /data/tmp_hadoop /data/hdfs/nn   /data/hdfs/dn   /data/tmp_hbase   /data/yarn/local /data/yarn/logs/data/log/hadoop-hdfs /data/log/hadoop-yarn /data/log/hadoop-mapred

2. Hadoop-2.2.0配置

 cd  /opt/hadoop-2.2.0/etc/hadoop

(1)        core-site.xml

<!--- global properties -->

<property>

      <name>hadoop.tmp.dir</name>

      <value>/data/tmp_hadoop</value>

         <!-- /tmp/hadoop-${user.name} -->

         <!--A base for other temporarydirectories-->   

</property>

<!-- file system properties -->

<property>                                           

 <name>fs.defaultFS</name>                          

 <value>hdfs://localhost:9000</value>

</property>

<property>

     <name>fs.trash.interval</name>

     <value>1440</value>

        <!--24h,0代表关闭-->

</property>

<property>

     <name>fs.trash.checkpoint.interval</name>

     <value>1440</value>

         <!--一般小于等于fs.trash.interval-->

</property>

<!-- i/o properties -->

<property>

         <name>io.file.buffer.size</name>

         <value>131072</value>

         <!--读写序列化文件缓冲区大小-->

</property>

<!-- Local file system -->

<property>

       <name>dfs.blocksize</name>

       <value>67108864</value>

       <!--268435456 256M-->

</property>

(2)hadoop-env.sh

export JAVA_HOME=/opt/jdk1.7.0_45

export HADOOP_LOG_DIR=/data/log/hadoop-hdfs

export YARN_LOG_DIR=/data/log/hadoop-yarn

exportHADOOP_MAPRED_LOG_DIR=/data/log/hadoop-mapred

(3) hdfs-site.xml

<!--需要配置本地目录的地方-->

<property>                                           

      <name>dfs.namenode.name.dir</name>                          

       <value>/data/hdfs/nn</value>

              <!--file://${hadoop.tmp.dir}/dfs/name-->

              <!--命名空间和事务处理日志-->

</property>

<property>                                         

       <name>dfs.datanode.data.dir</name>                   

        <value>/data/hdfs/dn</value>

              <!--file://${hadoop.tmp.dir}/dfs/data-->

              <!--DataNode本地文件存储的路径-->

</property>

    <!--Snn单机必须配置-->

<property>

   <name>dfs.namenode.secondary.http-address</name>

   <value>localhost:9001</value>            

</property>

<property> 

       <name>dfs.replication</name> 

       <value>1</value> 

       <!--冗余备份数目,一般为3-->

</property>

(4) mapred-site.xml

<property>

        <name>mapreduce.framework.name</name>   

        <value>yarn</value>

          <!--运行框架设置为 Hadoop YARN-->

</property>

<property>

       <name>mapreduce.shuffle.port</name>

       <value>13562</value>

</property>

<!--JHS-->

<property>

    <name>mapreduce.jobhistory.address</name>   

    <value>localhost:10020</value>                       

</property>

<property>

    <name>mapreduce.jobhistory.webapp.address</name>

    <value>localhost:19888</value>                                        

</property>

(5) masters

localhost

(6) slaves

localhost

(7) yarn-env.sh

export JAVA_HOME=/opt/jdk1.7.0_45

(8) yarn-site.xml

<!-- Resource Manager Configs -->

<property>

      <name>yarn.resourcemanager.webapp.address</name>

       <value>localhost:8088</value>

              <!--RMweb-->

</property>

<property>

      <name>yarn.resourcemanager.admin.address</name>

      <value>localhost:8033</value>

          <!--RM admin interface.-->

</property>

<property>

       <name>yarn.resourcemanager.address</name>

       <value>localhost:8032</value>

              <!--RM设置客户端提交job-->

</property>

<property>

       <name>yarn.resourcemanager.scheduler.address</name>

       <value>localhost:8030</value>

              <!--schedulerinterface -->

</property>

<property>

      <name>yarn.resourcemanager.resource-tracker.address</name>

     <value>localhost:8031</value>

</property>

<!--NodeManager-->

<property>

       <name>yarn.nodemanager.aux-services</name>

       <value>mapreduce_shuffle</value>

</property>

<property>

      <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>

      <value>org.apache.hadoop.mapred.ShuffleHandler</value>

</property>

(9) 启动hadoop

sudo chown -R hadoop:hadoop hadoop-2.2.0/

hadoop namenode -format

start-dfs.sh

jps一下

NameNode

DataNode

SecondaryNameNode

localhost:50070

satrt-yarn.sh

jps一下:

ResourceManager

NodeManager

localhost:8088

mr-jobhistory-daemon.sh  start historyserver

jps一下:

JobHistoryServer

localhost:19888

3. Hbase-0.96.0-hadoop2配置

(1)hbase-env.sh

export JAVA_HOME=/opt/jdk1.7.0_45

exportPATH=$JAVA_HOME/bin:$JAVA_HOME/jre:$PATH  

export HBASE_MANAGES_ZK=true

export HBASE_CLASSPATH=/opt/hadoop-2.2.0/etc/hadoop

export  HBASE_LOG_DIR=/data/hbase/logs

(2)hbase-site.xml

<property>

    <name>hbase.rootdir</name>

    <value>hdfs://localhost:9000/hbase</value>

</property>

<property>

   <name>hbase.cluster.distributed</name>

   <value>true</value>

</property>

<property>

   <name>hbase.tmp.dir</name>

   <value>/data/tmp_hbase</value>

</property>

<property>

    <name>hbase.zookeeper.property.dataDir</name>

    <value>/data/hbase/zookeeper</value>

    <!—使用内置zookeeper的快照存放-->

</property>

<property>

    <name>hbase.zookeeper.property.clientPort</name>

     <value>2181</value>

</property>

<property>

   <name>hbase.zookeeper.quorum</name>

    <value>localhost</value>

</property>

(3)regionservers

localhost

(4)启动hbase

sudo chown -R hadoop:hadoop hbase-0.96.0-hadoop2/

start-hbase.sh

jps一下:

HQuorumPeer

HRegionServer

HMaster

 访问:localhost:60010