热门IT资讯网

请输入关键字词

热门搜索排行

最新搜索排行

导航： 首页 > 互联网科技 >

centos7 hadoop-2.7.3完全分布式安装配置

发表于：2024-11-25 作者：热门IT资讯网编辑
编辑最后更新 2024年11月25日，操作系统版本：centos7 64位hadoop版本：hadoop-2.7.3hbase版本：hbase-1.2.4机器：192.168.11.131 master1 Namenode Resourc

操作系统版本：

centos7 64位


hadoop版本：

hadoop-2.7.3


hbase版本：

hbase-1.2.4


机器：

192.168.11.131 master1 Namenode ResourceManager QuorumPeerMain Jobhistory HMaster DFSZKFailoverController

192.168.11.132 master2 Namenode HMaster DFSZKFailoverController

192.168.11.133 slave1 Datanode HRegionServer NodeManager JournalNode

192.168.11.134 slave2 Datanode HRegionServer NodeManager JournalNode

192.168.11.135 slave3 Datanode HRegionServer NodeManager JournalNode


所有节点关闭防火墙及selinux

# firewall-cmd --state

running


# systemctl stop firewalld.service


# systemctl disable firewalld.service


# setenforce 0


# vi /etc/sysconfig/selinux

SELINUX=enforcing --> disabled


所有节点配置yum源

# cd


# mkdir apps


http://mirrors.163.com/centos/7/os/x86_64/Packages/wget-1.14-15.el7.x86_64.rpm


# rpm -i wget-1.14-15.el7.x86_64.rpm


# cd /etc/yum.repos.d


# wget http://mirrors.aliyun.com/repo/Centos-7.repo


# mv Centos-7.repo CentOS-Base.repo


# scp CentOS-Base.repo [email protected]:/etc/yum.repos.d/

# scp CentOS-Base.repo [email protected]:/etc/yum.repos.d/

# scp CentOS-Base.repo [email protected]:/etc/yum.repos.d/

# scp CentOS-Base.repo [email protected]:/etc/yum.repos.d/


# yum clean all


# yum makecache


# yum update


配置ntp时间同步

所有节点安装ntp

# yum install -y ntp


ntp server端：

# date -s "2018-05-27 23:03:30"


# vi /etc/ntp.conf

在注释下添加两行

#restrict 192.168.1.0 mask 255.255.255.0 nomodify notrap

server 127.127.1.0

fudge 127.127.1.0 stratum 11


注释下面

#server 0.centos.pool.ntp.org iburst

#server 1.centos.pool.ntp.org iburst

#server 2.centos.pool.ntp.org iburst

#server 3.centos.pool.ntp.org iburst


# systemctl start ntpd.service


# systemctl enable ntpd.service


ntp客户端（其余四台都为ntp客户端）：

# vi /etc/ntp.conf

同样注释下添加两行

#restrict 192.168.1.0 mask 255.255.255.0 nomodify notrap

server 192.168.11.131

fudge 127.127.1.0 stratum 11


四行添加注释

#server 0.centos.pool.ntp.org iburst

#server 1.centos.pool.ntp.org iburst

#server 2.centos.pool.ntp.org iburst

#server 3.centos.pool.ntp.org iburst


# systemctl start ntpd.service


# systemctl enable ntpd.service


# ntpdate 192.168.11.131

28 May 07:04:50 ntpdate[1714]: the NTP socket is in use, exiting


# lsof -i:123

-bash: lsof: command not found


# yum install -y lsof


# lsof -i:123

COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME

ntpd 1693 ntp 16u IPv4 25565 0t0 UDP *:ntp

ntpd 1693 ntp 17u IPv6 25566 0t0 UDP *:ntp

ntpd 1693 ntp 18u IPv4 25572 0t0 UDP localhost:ntp

ntpd 1693 ntp 19u IPv4 25573 0t0 UDP localhost.localdomain:ntp

ntpd 1693 ntp 20u IPv6 25574 0t0 UDP localhost:ntp

ntpd 1693 ntp 21u IPv6 25575 0t0 UDP localhost.localdomain:ntp


# kill -9 1693


# ntpdate 192.168.11.131

27 May 23:06:14 ntpdate[1728]: step time server 192.168.11.131 offset -28808.035509 sec


# date

Sun May 27 23:06:17 CST 2018


所有节点修改主机名（永久）

# hostnamectl set-hostname master1~slave3

临时修改主机名

# hostname master1~slave3


主节点修改hosts文件

# vi /etc/hosts

192.168.11.131 master1

192.168.11.132 master2

192.168.11.133 slave1

192.168.11.134 slave2

192.168.11.135 slave3


把hosts文件覆盖到其他机器

# scp /etc/hosts [email protected]~135:/etc/


所有节点创建管理用户和组

创建组和用户

# groupadd hduser


# useradd -g hduser hduser


# passwd hduser


创建目录并赋权

每台机器上创建如下文件夹

# mkdir /data1

# mkdir /data2


修改权限

# chown hudser:hduser /data1

# chown hudser:hduser /data2


# su hduser

$ mkdir -p /data1/hadoop_data/hdfs/namenode

$ mkdir -p /data2/hadoop_data/hdfs/namenode

$ mkdir -p /data1/hadoop_data/hdfs/datanode(NameNode不要)

$ mkdir -p /data2/hadoop_data/hdfs/datanode(NameNode不要)

$ mkdir -p /data1/hadoop_data/pids

$ mkdir -p /data2/hadoop_data/pids

$ mkdir -p /data1/hadoop_data/hadoop_tmp

$ mkdir -p /data2/hadoop_data/hadoop_tmp


无密验证

master1和master2节点操作

# su - hduser


$ ssh-keygen -t rsa


$ cd ~/.ssh

$ cat id_rsa.pub >> authorized_keys


master1节点操作

$ ssh-copy-id -i ~/.ssh/id_rsa.pub hadoop@master2


master2节点操作

$ scp ~/.ssh/authorized_keys hduser@master1:~/.ssh/


slave1、slave2和slave3节点创建.ssh目录

# mkdir /home/hduser/.ssh


# chown hduser:hduser /home/hduser/.ssh


master1节点操作

$ scp ~/.ssh/authorized_keys hduser@slave1:~/.ssh

$ scp ~/.ssh/authorized_keys hduser@slave2:~/.ssh

$ scp ~/.ssh/authorized_keys hduser@slave3:~/.ssh


master1和master2节点验证

验证方法，分别在两个节点，ssh登陆本机（hdusser用户）及其他四个节点，看看是不是无密登陆。


如果未通过验证，所有机器执行下面命令

$ chmod 600 ~/.ssh/authorized_keys

$ chmod 700 ~/.ssh


所有节点配置java环境

$ mkdir -p /data1/usr/src


上传包到/data1/usr/src目录下


$ cd /data1/usr/src

$ tar xf jdk1.7.0_79.tar -C /data1/usr/


$ vi ~/.bashrc

export JAVA_HOME=/data1/usr/jdk1.7.0_79

export JRE_HOME=$JAVA_HOME/jre

export CLASSPATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar:$JRE_HOME/lib/rt.jar

export PATH=$PATH:$JAVA_HOME/bin


$ source ~/.bashrc


mastar1节点配置hadoop（hdsuer用户）

下载hadoop-2.7.3.tar.gz，上传到/data1/usr/src

http://mirrors.cnnic.cn/apache/hadoop/common/hadoop-2.7.3/hadoop-2.7.3.tar.gz


$ cd /data1/usr/src

$ tar -zxf hadoop-2.7.3.tar.gz -C /data1/usr/


$ vi /data1/usr/hadoop-2.7.3/etc/hadoop/hadoop-env.sh

export JAVA_HOME=/data1/usr/jdk1.7.0_79

export HADOOP_PID_DIR=/data1/hadoop_data/pids

export HADOOP_PID_DIR=/data2/hadoop_data/pids

export HADOOP_MAPRED_PID_DIR=/data1/hadoop_data/pids


$ vi /data1/usr/hadoop-2.7.3/etc/hadoop/mapred-env.sh

export HADOOP_MAPRED_PID_DIR=/data2/hadoop_data/pids


$ vi /data1/usr/hadoop-2.7.3/etc/hadoop/yarn-env.sh

export YARN_PID_DIR=/data2/hadoop_data/pids


$ vi /data1/usr/hadoop-2.7.3/etc/hadoop/core-site.xml

fs.defaultFS

hdfs://masters

hadoop.tmp.dir

/data2/hadoop_data/hadoop_tmp

ha.zookeeper.quorum

master1:2181,master2:2181,slave1:2181,slave2:2181,slave3:2181


$ vi /data1/usr/hadoop-2.7.3/etc/hadoop/hdfs-site.xml

dfs.nameservices

masters

dfs.ha.namenodes.masters

master1,master2

dfs.namenode.rpc-address.masters.master1

master1:9000

dfs.namenode.http-address.masters.master1

master1:50070

dfs.namenode.rpc-address.masters.master2

master2:9000

dfs.namenode.http-address.masters.master2

master2:50070

dfs.namenode.name.dir

file:///data2/hadoop_data/hdfs/namenode

dfs.datanode.data.dir

file:///data1/hadoop_data/hdfs/datanode,data2/hadoop_data/hdfs/datanode

dfs.namenode.shared.edits.dir

qjournal://slave1:8485;slave2:8485;slave3:8485/masters

dfs.journalnode.edits.dir

/data2/hadoop_data/journal

dfs.ha.automatic-failover.enabled

true

dfs.client.failover.proxy.provider.masters

org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider

dfs.ha.fencing.methods

sshfence

shell(/bin/true)

dfs.ha.fencing.ssh.private-key-files

/home/hduser/.ssh/id_rsa

dfs.ha.fencing.ssh.connect-timeout

30000

dfs.datanode.max.xcievers

8192

dfs.qjournal.write-txns.timeout.ms

60000


$ vi /data1/usr/hadoop-2.7.3/etc/hadoop/yarn-site.xml


yarn.resourcemanager.ha.enabled

true

yarn.resourcemanager.cluster-id

RM_HA_ID

yarn.resourcemanager.ha.rm-ids

rm1,rm2

yarn.resourcemanager.hostname.rm1

master1

yarn.resourcemanager.hostname.rm2

master2

yarn.resourcemanager.recovery.enabled

true

yarn.resourcemanager.store.class

org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore

yarn.resourcemanager.zk-address

master1:2181,master2:2181,slave1:2181,slave2:2181,slave3:2181

yarn.nodemanager.aux-services

mapreduce_shuffle

yarn.nodemanager.aux-services.mapreduce.shuffle.class

org.apache.hadoop.mapred.ShuffleHandler


$ cp /data1/usr/hadoop-2.7.3/etc/hadoop/mapred-site.xml.template /data1/usr/hadoop-2.7.3/etc/hadoop/mapred-site.xml

$ vi /data1/usr/hadoop-2.7.3/etc/hadoop/mapred-site.xml

mapreduce.framework.name

yarn


$ vi /data1/usr/hadoop-2.7.3/etc/hadoop/slaves

slave3

slave4

slave5


$ for ip in `seq 2 5`;do scp -rpq /data1/usr/hadoop-2.7.3 192.168.11.13$ip:/data1/usr;done


各节点zookeeper配置

http://archive.apache.org/dist/zookeeper/zookeeper-3.4.6/zookeeper-3.4.6.tar.gz

上传包到/data1/usr/src目录下

创建目录

$ mkdir -p /home/hduser/storage/zookeeper


$ cd /data1/usr/src

$ tar -zxf zookeeper-3.4.6.tar.gz -C /data1/usr

$ cp /data1/usr/zookeeper-3.4.6/conf/zoo_sample.cfg /data1/usr/zookeeper-3.4.6/conf/zoo.cfg

$ vi /data1/usr/zookeeper-3.4.6/conf/zoo.cfg

dataDir=/home/hduser/storage/zookeeper

server.1=master1:2888:3888

server.2=master2:2888:3888

server.3=slave1:2888:3888

server.4=slave2:2888:3888

server.5=slave3:2888:3888


master1-slave3各节点依次做操作

$ echo "1" > /home/hduser/storage/zookeeper/myid


$ echo "2" > /home/hduser/storage/zookeeper/myid


$ echo "3" > /home/hduser/storage/zookeeper/myid


$ echo "4" > /home/hduser/storage/zookeeper/myid


$ echo "5" > /home/hduser/storage/zookeeper/myid


$ cd /data1/usr/zookeeper-3.4.6/bin

$ ./zkServer.sh start


slave1、slave2和slave3启动journalnode

$ cd /data1/usr/hadoop-2.7.3/sbin

$ ./sbin/hadoop-daemon.sh start journalnode

用jps确认启动结果


在master1上格式化zookeeper节点格式化(第一次)

$ cd /data1/usr/hadoop-2.7.3

$ ./bin/hdfs zkfc -formatZK


在master1上执行命令:

./bin/hadoop namenode -format


在master1上启动namenode

./sbin/hadoop-daemon.sh start namenode


需要在master2（备节点）上执行数据同步

./bin/hdfs namenode -bootstrapStandby

scp -r /data2/hadoop_data/hdfs/namenode hduser@mster2:/data2/hadoop_data/hdfs/


在master2上启动namenode

./sbin/hadoop-daemon.sh start namenode


设置master1为active

./bin/hdfs haadmin -transitionToActive master1

./bin/hdfs haadmin -getServiceState master1



在master1上启动datanode

./sbin/hadoop-daemons.sh start datanode


启动HDFS(第二次以后)

在master1上执行命令:

./sbin/start-dfs.sh

启动YARN

在master1上执行命令:

./sbin/start-yarn.sh


验证

验证namenode

http://master1:50070

Overview 'master1:9000' (active)

http://master2:50070

Overview 'master2:9000' (standby)


上传文件

./bin/hadoop fs -put /data1/usr/hadoop-2.7.3/etc/hadoop /test

./bin/hadoop fs -ls /test


namenode的备份验证

杀死master1，master2变为active


验证yarn

./bin/hadoop jar /data1/usr/hadoop-2.7.3/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.7.3.jar wordcount /test/hadoop /test/out



安装HBASE

master1节点操作：


下载hbase-1.2.4-bin.tar.gz，解压

$ cd /data1/usr/src

$ tar -zxvf hbase-1.2.4-bin.tar.gz -C /data1/usr/


$ mkdir -p /data1/hadoop_data/hbase_tmp

$ mkdir -p /data2/hadoop_data/hbase_tmp


配置master1的hbase环境

配置hbase-env.sh

$ vi /data1/usr/hbase-1.2.4/conf/hbase-env.sh

export JAVA_HOME=/data1/usr/jdk1.7.0_79

export HBASE_PID_DIR=/data2/hadoop_data/pids

export HBASE_MANAGES_ZK=false

export HADOOP_HOME=/data1/usr/hadoop-2.7.3


配置hbase-site.xml

$ vi /data1/usr/hbase-1.2.4/conf/hbase-site.xml

hbase.rootdir

hdfs://masters/hbase

hbase.cluster.distributed

true

hbase.master.port

60000

hbase.tmp.dir

/data2/hadoop_data/hbase_tmp

hbase.zookeeper.quorum

master1,master2,slave1,slave2,slave3


配置regionservers

$ vi /data1/usr/hbase-1.2.4/conf/regionservers

slave1

slave2

slave3


配置backup-masters

$ vi /data1/usr/hbase-1.2.4/conf/backup-masters


移除 HBase 里面的不必要 log4j 的 jar 包

cd ${HBASE_HOME}/lib

mv slf4j-log4j12-1.7.5.jar slf4j-log4j12-1.7.5.jar.bak


将master1的hbase环境传输到其他节点

$ for ip in `seq 2 5`;do scp -rpq /data1/usr/hbase-1.2.4 192.168.11.13$ip:/data1/usr;done


启动顺序

按hadoop集群的启动步骤，启动hadoop集群


master1上启动Hbase

$ cd /data1/usr/hbase-1.2.4/bin

$ ./start-hbase.sh


验证

$ /data1/usr/hadoop-2.7.3/bin/hadoop fs -ls / 查看hbase是否在HDFS文件系统创建成功


执行： bin/hbase shell 可以进入Hbase管理界面、

输入 status 查看状态


创建表

create 'test', 'cf'


显示表信息

list 'test'


表中插入数据

put 'test', 'row1', 'cf:a', 'value1'

put 'test', 'row2', 'cf:b', 'value2'

put 'test', 'row3', 'cf:c', 'value3'


查询表

scan 'test'


取一行数据

get 'test', 'row1'


失效表

disable 'test'


删除表

drop 'test'


浏览器输入http://master1:16010可以打开Hbase管理界面

http://192.168.11.131/master-status


启动thrift2

hbase-daemons.sh start thrift2


去datanode节点用jps确认


很赞哦！

相关文章

0