zookeeper和hbase安裝

一、zookeeper

1、/etc/profile

HADOOP_PREFIX=/opt/hadoop

JAVA_HOME=/opt/jdk18

ZOOKEEPER_HOME=/opt/zookeeper

HBASE_HOME=/opt/hbase

PATH=$PATH:$JAVA_HOME/bin:$HADOOP_PREFIX/bin:$HADOOP_PREFIX/sbin:$ZOOKEEPER_HOME/bin:$HBASE_HOME/bin

export HADOOP_PREFIX PATH JAVA_HOME ZOOKEEPER_HOME HBASE_HOME USER LOGNAME MAIL HOSTNAME HISTSIZE HISTCONTROL

2、創(chuàng)建目錄:

mkdir -p /opt/zookeeper/data

mkdir -p /opt/zookeeper/logs

3、修改zoo.cfg配置文件，在在文件末尾添加。

server.1=NameNode34:2888:3888

server.2=DataNode35:2888:3888

server.3=DataNode37:2888:3888

server.4=DataNode38:2888:3888

dataDir=/opt/zookeeper/data

dataLogDir=/opt/zookeeper/logs

4、創(chuàng)建myid文件

在dataDir目錄下創(chuàng)建myid文件，該文件的內(nèi)容根據(jù)server定義的不同而不同，如server.1 該文件的內(nèi)容是1，server.2 該文件內(nèi)容是 2，以此類推.

節(jié)點(diǎn)1

echo "1">myid

節(jié)點(diǎn)2

echo "2">myid

節(jié)點(diǎn)3

echo "3">myid

節(jié)點(diǎn)4

echo "4">myid

5、拷貝至其他節(jié)點(diǎn)

scp /etc/profile root@DataNode35:/etc/profile

scp /etc/profile root@DataNode37:/etc/profile

scp /etc/profile root@DataNode38:/etc/profile

scp -r /opt/zookeeper root@DataNode35:/opt

scp -r /opt/zookeeper root@DataNode37:/opt

scp -r /opt/zookeeper root@DataNode38:/opt

6、在每個(gè)節(jié)點(diǎn)上

source /etc/profile

7、在每個(gè)節(jié)點(diǎn)上執(zhí)行 zkServer.sh start

JMX enabled by default

Using config: /usr1/zookeeper/bin/../conf/zoo.cfg

Starting zookeeper ... STARTED

1.8 驗(yàn)證

[hadoop1@node4 bin]$ ./zkServer.sh status

JMX enabled by default

Using config: /usr1/zookeeper/bin/../conf/zoo.cfg

Mode: leader

[hadoop1@node1 bin]$ ./zkServer.sh status

JMX enabled by default

Using config: /usr1/zookeeper/bin/../conf/zoo.cfg

Mode: follower

[hadoop1@node2 bin]$ ./zkServer.sh status

JMX enabled by default

Using config: /usr1/zookeeper/bin/../conf/zoo.cfg

Mode: follower

[hadoop1@node3 bin]$ ./zkServer.sh status

JMX enabled by default

Using config: /usr1/zookeeper/bin/../conf/zoo.cfg

Mode: follower

-- The End --

二、hbase

1、下載hbase

2、/etc/profile

HADOOP_PREFIX=/opt/hadoop

JAVA_HOME=/opt/jdk18

ZOOKEEPER_HOME=/opt/zookeeper

HBASE_HOME=/opt/hbase

PATH=$PATH:$JAVA_HOME/bin:$HADOOP_PREFIX/bin:$HADOOP_PREFIX/sbin:$ZOOKEEPER_HOME/bin:$HBASE_HOME/bin

export HADOOP_PREFIX PATH JAVA_HOME ZOOKEEPER_HOME HBASE_HOME USER LOGNAME MAIL HOSTNAME HISTSIZE HISTCONTROL

3、修改配置文件/opt/hbase/conf/hbase-env.sh

export JAVA_HOME=/opt/jdk18/

export HBASE_CLASSPATH=/opt/hbase/conf

export HBASE_MANAGES_ZK=true

4、在hbase安裝目錄下建立tmp文件夾

mkdir tmp

5、修改hbase-site.xml

<configuration>

<property>

    <name>hbase.rootdir</name>

    <value>hdfs://NameNode34:9000/hbase</value>

</property>

  <property>

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

    <value>true</value>

</property>

<property>

    <name>hbase.tmp.dir</name>

    <value>/opt/hbase/tmp</value>

</property>

<property>

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

    <value>NameNode34,DataNode35,DataNode37,DataNode38</value>

</property>

<property>

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

    <value>/opt/hbase/tmp/zookeeper</value>

</property>

</configuration>

6、編輯/opt/hbase/conf/regionservers，將所有的slavenode添加到這個(gè)文件

NameNode34

DataNode35

DataNode37

DataNode38

7、將Hbase拷貝到其他節(jié)點(diǎn)機(jī)并對(duì)其他節(jié)點(diǎn)機(jī)配置環(huán)境變量

scp /etc/profile root@DataNode35:/etc/profile

scp /etc/profile root@DataNode37:/etc/profile

scp /etc/profile root@DataNode38:/etc/profile

scp -r /opt/hbase root@DataNode35:/opt

scp -r /opt/hbase root@DataNode37:/opt

scp -r /opt/hbase root@DataNode38:/opt

8、在每個(gè)節(jié)點(diǎn)上

source /etc/profile

9、在主節(jié)點(diǎn)啟動(dòng)hbase（只在主節(jié)點(diǎn)啟動(dòng)就OK） 

start-hbase.sh

10、驗(yàn)證hbase

jps

11、測(cè)試hbase shell

hbase shell

Create  ‘test’,’data’

Disable ‘test’

Dorp ‘test’

參考：

zookeeper: 

http://blog.chinaunix.net/uid-77311-id-4580114.html

hbase: 

http://blog.itpub.net/28929558/viewspace-1204409