hadoop-2.7.4+hbase-1.3.1+zookeeper-3.4.9搭建分布式集群环境
# 系统信息3台系统： centos6.8内核：4 内存：4G硬盘：50G
# 主机名称，ip地址
master: 192.168.1.110
slave1: 192.168.1.111
slave2: 192.168.1.112
######################## 软件下载地址 ########################
链接：https://pan.baidu.com/s/1dFuBnKt 密码：rhwu
######################## 基础初始配置 ########################
# 版本选择
jdk-8u77-linux-x64.rpm
zookeeper-3.4.9.tar.gz
hbase-1.3.1-bin.tar.gz
hadoop-2.7.4.tar.gz
# 配置hosts文件，三台机器都需要
# cat /etc/hosts
192.168.1.110 master
192.168.1.111 slave1
192.168.1.112 slave2
# 配置用户
1
2
groupadd -g 4000 hadoop
useradd -g 4000 -u 4001 hadoop
# 所有的主机 hbase,zookeeper 安装目录都在此处
1
2
mkdir /opt/hadoop
chown hadoop.hadoop /opt/hadoop/ -R
########################时间配置 ########################
# 双机互信
主要有三步：
①生成公钥和私钥
②导入公钥到认证文件
③更改权限
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
# ssh-keygen -t rsa
Generating public/private rsa key pair.
Enter file in which to save the key (/root/.ssh/id_rsa):
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /root/.ssh/id_rsa.
Your public key has been saved in /root/.ssh/id_rsa.pub.
The key fingerprint is:
ee:15:03:c7:3a:a2:8e:6a:c1:0c:74:d3:97:34:77:04 root@master
The key's randomart image is:
+--[ RSA 2048]----+
| . .o.Eoo |
| . o . oo.. |
|. . . . . o |
|. + |
|+ . S o |
| + . o . o |
|. . . . |
| . o . . |
|o.. . . |
+-----------------+
# cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys
# chmod 700 ~/.ssh && chmod 600 ~/.ssh/*
# 主机与从机之间必须可以双向无密码登陆，从机与从机之间无限制
1
2
scp ~/.ssh/authorized_keys slave1:/root/.ssh/
scp ~/.ssh/authorized_keys slave2:/root/.ssh/
# 同步时间
1
2
# ansible hbase -m cron -a "name='ntpdate' hour='*/1' job='/usr/sbin/ntpdate 192.168.1.110 &> /dev/null'"
# ansible hbase -m shell -a "crontab -l"
#时间一定要保持一致
########################防火墙配置 ########################
# 防火墙配置所有的主机上都得配置，或者开放 (2181，2888:3888端口,这部分端口是zookeeper端口)
1
2
3
# iptables -I INPUT -s 192.168.1.0/24 -j ACCEPT
# service iptables save
# service iptables restart
########################JDK配置 ########################
# 安装jdk，并配置环境变量，三台机器都需要安装
# 设置环境变量
1
2
3
# cat /etc/profile.d/java.sh
export JAVA_HOME=/usr/java/default
export PATH=$JAVA_HOME/bin:$PATH
# 重新加载配置文件使之生效
# source /etc/profile.d/java.sh
# 查看是否配置完成，3台机器都需要测试
# java -version
java version "1.8.0_77"
Java(TM) SE Runtime Environment (build 1.8.0_77-b03)
Java HotSpot(TM) 64-Bit Server VM (build 25.77-b03, mixed mode)
######################## zookeeper集群配置 ########################
# 参考文档： http://blog.csdn.net/reblue520/article/details/52279486
# 注意：zookeeper因为有主节点和从节点的关系，所以部署的集群台数最好为奇数个，否则可能出现脑裂导致服务异常
# 下载地址： http://archive.apache.org/dist/zookeeper/zookeeper-3.4.9/zookeeper-3.4.9.tar.gz
# 注意三台机器都需要安装，如果对ansible熟悉的话 可以直接使用它
1
2
mkdir /opt/hadoop
chown hadoop.hadoop /opt/hadoop/ -R
# 安装zookeeper
1
2
3
4
# cd /opt/hadoop/
# ls
zookeeper-3.4.9.tar.gz
# tar xf zookeeper-3.4.9.tar.gz
# 弄一个软链接，配置文件直接指向这个地址，未来方便更新版本
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
# ln -sv zookeeper-3.4.9 zookeeper
"zookeeper" -> "zookeeper-3.4.9"
# cd /opt/hadoop/zookeeper/conf
# cp zoo_sample.cfg zoo.cfg
# cat zoo.cfg
tickTime=2000
initLimit=10
syncLimit=5
dataDir=/opt/hadoop/zookeeper/data
dataLogDir=/opt/hadoop/zookeeper/logs
clientPort=2181
server.1=master:2888:3888
server.2=slave1:2888:3888
server.3=slave2:2888:3888
# 创建数据以及日志目录，将设置属主属组权限
1
2
# mkdir /opt/hadoop/zookeeper/data
# mkdir /opt/hadoop/zookeeper/logs
# 在zoo.cfg中的dataDir指定的目录下，新建myid文件。
# 例如：$ZK_INSTALL/data下，新建myid。在myid文件中输入1。表示为server.1。
echo "1" > data/myid 这里表示的是server.1如果是第二个机器那么表示server.2
启动：在集群中的每台主机上执行如下命令
bin/zkServer.sh start
查看状态，可以看到其中一台为主节点，其他两台为从节点：
bin/zkServer.sh status
# 启动zookeeper集群
1
2
3
4
# bin/zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /opt/hadoop/zookeeper/bin/../conf/zoo.cfg
Mode: leader
# 从节点
1
2
3
4
# bin/zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /opt/hadoop/zookeeper/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED
#启动报错说明没有配置myid文件，
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
2017-12-04 11:56:21,306 - INFO - Reading configuration from: /opt/hadoop/zookeeper/bin/../conf/zoo.cfg
2017-12-04 11:56:21,323 - INFO - Resolved hostname: slave2 to address: slave2/192.168.1.112
2017-12-04 11:56:21,324 - INFO - Resolved hostname: slave1 to address: slave1/192.168.1.111
2017-12-04 11:56:21,324 - INFO - Resolved hostname: master to address: master/192.168.1.110
2017-12-04 11:56:21,325 - INFO - Defaulting to majority quorums
2017-12-04 11:56:21,326 - ERROR - Invalid config, exiting abnormally
org.apache.zookeeper.server.quorum.QuorumPeerConfig$ConfigException: Error processing /opt/hadoop/zookeeper/bin/../conf/zoo.cfg
at org.apache.zookeeper.server.quorum.QuorumPeerConfig.parse(QuorumPeerConfig.java:144)
at org.apache.zookeeper.server.quorum.QuorumPeerMain.initializeAndRun(QuorumPeerMain.java:101)
at org.apache.zookeeper.server.quorum.QuorumPeerMain.main(QuorumPeerMain.java:78)
Caused by: java.lang.IllegalArgumentException: /opt/hadoop/zookeeper/data/myid file is missing
at org.apache.zookeeper.server.quorum.QuorumPeerConfig.parseProperties(QuorumPeerConfig.java:362)
at org.apache.zookeeper.server.quorum.QuorumPeerConfig.parse(QuorumPeerConfig.java:140)
... 2 more
Invalid config, exiting abnormally
# 这里是因为防火墙开着，没有开放端口的原因
1
2
3
4
5
2016-03-26 03:48:07,957 - WARN /0:0:0:0:0:0:0:0:2181:QuorumCnxManager@400] - Cannot open channel to 3 at election address S2/这里是地址
java.net.ConnectException: 主机无法连接
at java.net.PlainSocketImpl.socketConnect(Native Method)
at java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:339)
at java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:200)
######################## hbase 与hadoop的版本需要对应 ########################
http://blog.csdn.net/shuaigexiaobo/article/details/78114221 低版本与高版本会安不上，还需要注意jdk版本
######################## hadoop 集群配置 ########################
# 软件放置路径为初级配置的路径 /opt/hadoop
1
2
3
# tar xf hadoop-2.7.4.tar.gz
# ln -sv hadoop-2.7.4 hadoop
"hadoop" -> "hadoop-2.7.4"
# 配置属主属组权限
1
# chown hadoop.hadoop /opt/hadoop/hadoop-2.7.4 -R
# 环境变量设置
1
2
3
4
5
6
7
8
9
10
vim /etc/profile.d/hadoop.sh
export HADOOP_HOME=/opt/hadoop/hadoop
export HADOOP_INSTALL=$HADOOP_HOME
export HADOOP_MAPRED_HOME=$HADOOP_HOME
export HADOOP_COMMON_HOME=$HADOOP_HOME
export HADOOP_HDFS_HOME=$HADOOP_HOME
export YARN_HOME=$HADOOP_HOME
export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_HOME/lib/native
export PATH=$PATH:$HADOOP_HOME/sbin:$HADOOP_HOME/bin
# export HADOOP_SSH_OPTS="-p 22"
# 复制到其它主机中
1
2
# scp /etc/profile.d/hadoop.sh slave1:/etc/profile.d/
# scp /etc/profile.d/hadoop.sh slave2:/etc/profile.d/
# 加载环境变量
1
# soure /etc/profile.d/hadoop.sh
# 查看是否生效
1
2
3
# hadoop version
Hadoop 2.7.4
Subversion https://shv@git-wip-us.apache.org/repos/asf/hadoop.git -r cd915e1e8d9d0131462a0b7301586c175728a282
# hadoop配置文件在放置于/opt/hadoop/hadoop/etc/hadoop
1
2
3
4
5
6
7
vimcore-site.xml # 添加如下内容
<configuration>
<property>
<name>fs.default.name</name>
<value>hdfs://master:9000</value>
</property>
</configuration>
1
2
3
vim hadoop-env.sh
#export JAVA_HOME=${JAVA_HOME}
export JAVA_HOME=/usr/java/default
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
vim hdfs-site.xml # 配置hdfs文件数据节点以及名称节点
<configuration>
<property>
<name>dfs.name.dir</name>
<value>/opt/hadoop/hadoop/name</value>
</property>
<property>
<name>dfs.data.dir</name>
<value>/opt/hadoop/hadoop/data</value>
</property>
<property>
<name>dfs.replication</name>
<value>3</value>
</property>
</configuration>
mkdir /opt/hadoop/hadoop/name
mkdir /opt/hadoop/hadoop/data
1
2
3
4
5
6
7
8
# cp mapred-site.xml.template mapred-site.xml
# vim !$
<configuration>
<property>
<name>mapred.job.tracker</name>
<value>master:9001</value>
</property>
</configuration>
# 配置从节点 先删除localhost
1
2
3
/opt/hadoop/hadoop/etc/hadoop/slaves
slave1
slave2
# 三台机器都是一样的配置，放置相同的路径
1
2
# scp -r hadoop-2.7.4 slave1:/opt/hadoop/
# scp -r hadoop-2.7.4 slave2:/opt/hadoop/
# 使用ansible或者手动直接软链接过去就行
1
# ansible hbase -m shell -a 'ln -sv /opt/hadoop/hadoop-2.7.4 /opt/hadoop/hadoop'
# 配置属主属组文件
1
# ansible hbase -m shell -a 'chown hadoop.hadoop /opt/hadoop/hadoop -R'
# 进入master的/opt/hadoop/hadoop目录，执行以下操作
1
# bin/hadoop namenode -format # 格式化namenode，第一次启动服务前执行的操作,以后不需要执行
# 启动hadoop服务
1
2
3
4
5
# sbin/start-all.sh
This script is Deprecated. Instead use start-dfs.sh and start-yarn.sh
17/12/04 15:56:51 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Starting namenodes on
master: starting namenode, logging to /opt/hadoop/hadoop-2.7.4/logs/hadoop-root-namenode-master.out
# 查看进程 会发现多了资源名称节点以及namanode
1
2
3
4
5
6
# jps
5057 ResourceManager
4900 SecondaryNameNode
4709 NameNode
5208 Jps
2734 QuorumPeerMain
# 登陆其它节点会发现多了一个数据节点
1
2
3
4
5
# jps
2624 QuorumPeerMain
3489 NodeManager
3378 DataNode
3603 Jps
######################## hbase集群配置 ########################
# 软件放置路径为初级配置的路径 /opt/hadoop
1
2
3
# tar xf hbase-1.3.1-bin.tar.gz
# ln -sv hbase-1.3.1 hbase
"hbase" -> "hbase-1.3.1"
# 配置文件目录 /opt/hadoop/hbase/conf
vim hbase-env.sh
1
2
3
export JAVA_HOME=/usr/java/default/
export HBASE_CLASSPATH=/opt/hadoop/hadoop/etc/hadoop
export HBASE_MANAGES_ZK=false # 不使用自带的zk，使用独立的zookeeper
vim hbase-site.xml # 配置站点信息
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
<configuration>
<property>
<name>hbase.rootdir</name>
<value>hdfs://master:9000/hbase</value>
</property>
<property>
<name>hbase.master</name>
<value>master</value>
</property>
<property>
<name>hbase.cluster.distributed</name>
<value>true</value>
</property>
<property>
<name>hbase.zookeeper.property.clientPort</name>
<value>2181</value> # 这里指的是zook的端口
</property>
<property>
<name>hbase.zookeeper.quorum</name> # 主机名一定要对应上
<value>master,slave1,slave2</value>
</property>
<property>
<name>zookeeper.session.timeout</name> # zook的session超时时长
<value>60000000</value>
</property>
<property>
<name>dfs.support.append</name>
<value>true</value>
</property>
</configuration>
vim regionservers # 配置从节点 一定要对应上
1
2
slave1
slave2
# 设置软链接，方便未来升级
# ansible hbase -m shell -a "ln -sv /opt/hadoop/hbase-1.3.1 /opt/hadoop/hbase"
# 设置属主属组权限
# ansible hbase -m shell -a "chown hadoop.hadoop /opt/hadoop/hbase-1.3.1 -R"
# 启动三台机器上的 hbase服务
# ansible hbase -m shell -a "/opt/hadoop/hbase-1.3.1/bin/start-hbase.sh"
# 只需要启动master上的，其它机器上会自动启动
# /opt/hadoop/hbase/bin/start-hbase.sh
# 查看master上的服务
1
2
3
4
5
6
7
# jps
5057 ResourceManager
4900 SecondaryNameNode
6516 HMaster
4709 NameNode
6809 Jps
2734 QuorumPeerMain
# 查看slave上的从节点服务
1
2
3
4
5
6
7
# jps
3510 NodeManager
3399 DataNode
2680 QuorumPeerMain
5464 Jps
5049 HMaster
4730 HRegionServer
# 进入hbase shell进行验证
/opt/hadoop/hbase/bin/hbase shell
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2017-12-04 16:20:28,690 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in
SLF4J: Found binding in
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 1.3.1, r930b9a55528fe45d8edce7af42fef2d35e77677a, Thu Apr6 19:36:54 PDT 2017
hbase(main):001:0>
hbase(main):002:0* list
TABLE
0 row(s) in 0.2350 seconds
=> []
hbase(main):003:0> create 'scores', 'grade', 'course'
0 row(s) in 2.4310 seconds
=> Hbase::Table - scores
hbase(main):004:0> list
TABLE
scores
1 row(s) in 0.0080 seconds
=> ["scores"]
####此处打开的地址都是 master 的IP ，192.168.1.110
页:
[1]