Hadoop-2.7.3下安装Hbase-1.2.6

1.环境信息

操作系统信息如下:

1

[hadoop@hadoop1 soft]$ uname -a

2

Linux hadoop1 2.6.32-358.el6.x86_64 #1 SMP Fri Feb 22 00:31:26 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

3

[hadoop@hadoop1 soft]$ cat /etc/redhat-release 

4

CentOS release 6.4 (Final)

IP地址和主机名如下:

1

[hadoop@hadoop1 soft]$ cat /etc/hosts

2

127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4

3

::1         localhost localhost.localdomain localhost6 localhost6.localdomain6

4

192.168.56.15 basic

5

192.168.56.1 PC-201306171517

6

192.168.56.21 hadoop1

7

192.168.56.22 hadoop2

8

192.168.56.23 hadoop3

2.下载hbase

hbase需要依赖hadoop和zookeeper,所以在安装之前需要先安装hadoop,详细安装步骤见:

http://blog.csdn.net/su377486/article/details/72085134

zookeeper可以使用hbase自带的或者自己先安装,我这里是事先安装好了zookeeper,详细安装步骤见:

http://blog.csdn.net/su377486/article/details/77844462

安装的hbase还需要和hadoop版本相兼容,具体hbase和hadoop版本对应如下:

X:表示不支持

NT:表示没有测试

S:表示支持

 HBase-1.1.xHBase-1.2.xHBase-1.3.xHBase-2.0.x

Hadoop-2.0.x-alpha

X

X

X

X

Hadoop-2.1.0-beta

X

X

X

X

Hadoop-2.2.0

NT

X

X

X

Hadoop-2.3.x

NT

X

X

X

Hadoop-2.4.x

S

S

S

X

Hadoop-2.5.x

S

S

S

X

Hadoop-2.6.0

X

X

X

X

Hadoop-2.6.1+

NT

S

S

S

Hadoop-2.7.0

X

X

X

X

Hadoop-2.7.1+

NT

S

S

S

Hadoop-2.8.0

X

X

X

X

Hadoop-3.0.0-alphax

NT

NT

NT

NT

下载地址如下:http://hbase.apache.org/

我这里选择1.2.6版本的hbase.

3.解压缩hbase

上传hbase-1.2.6-bin.tar.gz到hadoop1主机,解压缩,将解压缩后的目录拷贝到你需要的安装目录

1

[hadoop@hadoop1 soft]$ tar zxvf hbase-1.2.6-bin.tar.gz 

2

[hadoop@hadoop1 soft]$ mv hbase-1.2.6 /hadoop/hbase

为了运行方便,我们将hbase的可执行目录添加到PATH环境变量中:

1

PATH=/hadoop/hbase/bin:/hadoop/zookeeper/bin:$HIVE_HOME/bin:/hadoop/pig/bin:$JAVA_HOME/bin:$PATH:$HOME/bin:/hadoop/hadoop/bin:/hadoop/hadoop/sbin

4.配置hbase

一.修改hbase-env.sh

修改java_home环境变量:

1

[hadoop@hadoop1 conf]$ pwd

2

/hadoop/hbase/conf

3

[hadoop@hadoop1 conf]$ vi hbase-env.sh

4

export JAVA_HOME=/usr/local/jdk1.8.0_131

5

export HBASE_MANAGES_ZK=false

6

export HBASE_CLASSPATH=/hadoop/hadoop/etc/hadoop          #hadoop配置文件目录

7

export HBASE_OPTS="-XX:+UseConcMarkSweepGC"

8

export HBASE_MASTER_OPTS="$HBASE_MASTER_OPTS -XX:PermSize=128m -XX:MaxPermSize=128m"

9

export HBASE_REGIONSERVER_OPTS="$HBASE_REGIONSERVER_OPTS -XX:PermSize=128m -XX:MaxPermSize=128m"

二.修改hbase-site.xml

1

[hadoop@hadoop1 conf]$ vi hbase-site.xml

2

<configuration>

3

4

<property>

5

        <name>hbase.rootdir</name>

6

        <value>hdfs://hadoop1:9000/hbase</value>                                   <!-- 和hadoop core-site.xml 里面的设置保持一直 -->

7

</property>     

8

<property>      

9

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

10

        <value>true</value>

11

</property>

12

<property>

13

        <name>base.zookeeper.quorum</name>

14

        <value>hadoop1:2181,hadoop2:2181,hadoop3:2181</value>

15

</property>

16

<property>

17

        <name>hbase.tmp.dir</name>

18

        <value>/hadoop/hbase/tmp</value>                                              <!-- 需要手工创建 -->

19

</property>

20

21

</configuration>

22

三.修改regionservers

1

[hadoop@hadoop1 conf]$ vi regionservers 

2

3

hadoop1

4

hadoop2

5

hadoop3

6

4.复制hbase

将刚配置完成的hbase安装目录拷贝到其它两个节点:

1

[hadoop@hadoop3 ~]$ scp -r hadoop1:/hadoop/hbase /hadoop/

1

[hadoop@hadoop2 hadoop]$ scp -r hadoop1:/hadoop/hbase /hadoop/

5.启动hbase

在启动hbase之前需要先启动hadoop和zookeeper.

启动hadoop:

1

[hadoop@hadoop1 hadoop]$ start-all.sh

2

This script is Deprecated. Instead use start-dfs.sh and start-yarn.sh

3

Starting namenodes on [hadoop1]

4

hadoop1: starting namenode, logging to /hadoop/hadoop/logs/hadoop-hadoop-namenode-hadoop1.out

5

hadoop3: starting datanode, logging to /hadoop/hadoop/logs/hadoop-hadoop-datanode-hadoop3.out

6

hadoop2: starting datanode, logging to /hadoop/hadoop/logs/hadoop-hadoop-datanode-hadoop2.out

7

Starting secondary namenodes [hadoop1]

8

hadoop1: starting secondarynamenode, logging to /hadoop/hadoop/logs/hadoop-hadoop-secondarynamenode-hadoop1.out

9

starting yarn daemons

10

starting resourcemanager, logging to /hadoop/hadoop/logs/yarn-hadoop-resourcemanager-hadoop1.out

11

hadoop3: starting nodemanager, logging to /hadoop/hadoop/logs/yarn-hadoop-nodemanager-hadoop3.out

12

hadoop2: starting nodemanager, logging to /hadoop/hadoop/logs/yarn-hadoop-nodemanager-hadoop2.out

检查namenode节点hadoop状态:

1

[hadoop@hadoop1 hadoop]$ jps

2

2497 NameNode

3

2852 ResourceManager

4

2698 SecondaryNameNode

5

3148 Jps

其它节点状态:

1

[hadoop@hadoop2 hadoop]$ jps

2

2292 NodeManager

3

2391 Jps

4

2190 DataNode

启动zookeeper:

三个节点上分别执行zkserver.sh start

1

[hadoop@hadoop2 hadoop]$ zkServer.sh start

2

ZooKeeper JMX enabled by default

3

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

4

Starting zookeeper ... STARTED

检查zookeeper进程状态:

1

[hadoop@hadoop1 bin]$ jps

2

3184 QuorumPeerMain

3

2497 NameNode

4

2852 ResourceManager

5

2698 SecondaryNameNode

6

3215 Jps

1

[hadoop@hadoop2 hadoop]$ jps

2

2497 Jps

3

2292 NodeManager

4

2472 QuorumPeerMain

5

2190 DataNode

启动hbase:

在namenode节点上执行start-hbase.sh即可

1

[hadoop@hadoop1 bin]$ ./start-hbase.sh

2

starting master, logging to /hadoop/hbase/bin/../logs/hbase-hadoop-master-hadoop1.out

3

Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0

4

Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0

5

hadoop3: starting regionserver, logging to /hadoop/hbase/bin/../logs/hbase-hadoop-regionserver-hadoop3.out

6

hadoop2: starting regionserver, logging to /hadoop/hbase/bin/../logs/hbase-hadoop-regionserver-hadoop2.out

7

hadoop1: starting regionserver, logging to /hadoop/hbase/bin/../logs/hbase-hadoop-regionserver-hadoop1.out

8

hadoop3: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0

9

hadoop3: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0

10

hadoop2: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0

11

hadoop2: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0

12

hadoop1: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0

13

hadoop1: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0

检查进程状态:

1

[hadoop@hadoop1 bin]$ jps

2

3184 QuorumPeerMain

3

2497 NameNode

4

3779 Jps

5

3524 HRegionServer

6

2852 ResourceManager

7

2698 SecondaryNameNode

8

3390 HMaster

其它节点进程状态:

1

[hadoop@hadoop2 hadoop]$ jps

2

2292 NodeManager

3

2472 QuorumPeerMain

4

2556 HRegionServer

5

2190 DataNode

6

2718 Jps

6.测试连接hbase

使用hbase shell来登录连接hbase:

1

[hadoop@hadoop1 bin]$ hbase shell

2

SLF4J: Class path contains multiple SLF4J bindings.

3

SLF4J: Found binding in [jar:file:/hadoop/hbase/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]

4

SLF4J: Found binding in [jar:file:/hadoop/hadoop/share/hadoop/common/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class]

5

SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.

6

SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]

7

HBase Shell; enter 'help<RETURN>' for list of supported commands.

8

Type "exit<RETURN>" to leave the HBase Shell

9

Version 1.2.6, rUnknown, Mon May 29 02:25:32 CDT 2017

10

11

hbase(main):002:0> list

12

TABLE                                                                                                                                                                                                                                 

13

0 row(s) in 0.5330 seconds

14

15

=> []

16

hbase(main):003:0> create 'test','data'

17

0 row(s) in 2.5520 seconds

18

19

=> Hbase::Table - test

20

hbase(main):004:0> list

21

TABLE                                                                                                                                                                                                                                 

22

test   

在其它节点上连接测试:

1

[hadoop@hadoop2 etc]$ hbase shell

2

SLF4J: Class path contains multiple SLF4J bindings.

3

SLF4J: Found binding in [jar:file:/hadoop/hbase/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]

4

SLF4J: Found binding in [jar:file:/hadoop/hadoop/share/hadoop/common/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class]

5

SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.

6

SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]

7

HBase Shell; enter 'help<RETURN>' for list of supported commands.

8

Type "exit<RETURN>" to leave the HBase Shell

9

Version 1.2.6, rUnknown, Mon May 29 02:25:32 CDT 2017

10

11

hbase(main):001:0> status

12

1 active master, 0 backup masters, 3 servers, 0 dead, 1.0000 average load

13

还可以用网页查看hbase状态:

1

http://192.168.56.21:16010/master-status