CentOS7 安装 Hbase-0.98.24

0. 准备安装环境

继CentOS7安装hadoop-1.2.1准备安装环境

1. 安装Hbase

1.1 安装到指定目录

下载Hbase安装包到 /usr/local/src 目录下。

https://archive.apache.org/dist/hbase/0.98.24/hbase-0.98.24-hadoop1-bin.tar.gz

[root@master /usr/local/src]# ls hbase-0.98.24-hadoop1-bin.tar.gz
hbase-0.98.24-hadoop1-bin.tar.gz

解压到/usr/local下

[root@master /usr/local]# tar zxf src/hbase-0.98.24-hadoop1-bin.tar.gz
[root@master /usr/local]# ls
bin  etc  games  hadoop-1.2.1  hbase-0.98.24-hadoop1  include  jdk1.8.0_212  lib  lib64  libexec  sbin  share  src  zookeeper-3.4.5

1.2 配置服务

进入 /usr/local/hbase-0.98.24-hadoop1/conf 目录

配置 hbase-site.xml 文件

[root@master /usr/local/hbase-0.98.24-hadoop1/conf]# vim hbase-site.xml
...
<configuration>
    <!--用来指定使用hbase时产生文件的存放目录-->
    <property>
        <name>hbase.tmp.dir</name>
        <value>/var/hbase</value>
    </property>
    <!--指定hbase在hdfs里的根目录-->
    <property>
        <name>hbase.rootdir</name>
        <value>hdfs://master:9000/hbase</value>
    </property>
    <!--指定是否采用分布式模式-->
    <property>
        <name>hbase.cluster.distributed</name>
        <value>true</value>
    </property>
    <!--配置zookeeper集群节点地址-->
    <property>
        <name>hbase.zookeeper.quorum</name>
        <value>master,slave1,slave2</value>
    </property>
    <!--指定存储zookeeper快照信息的目录-->
    <property>
        <name>hbase.zookeeper.property.dataDir</name>
        <value>/usr/local/hbase-0.98.24-hadoop1/zookeeper</value>
    </property>
</configuration>

配置 hbase-env.sh

[root@master /usr/local/hbase-0.98.24-hadoop1/conf]# vim hbase-env.sh
...
# The java implementation to use.  Java 1.6 required.
export JAVA_HOME=/usr/local/jdk1.6.0_45
export CLASSPATH=.:$CLASSPATH:$JAVA_HOME/lib
...
# Tell HBase whether it should manage it's own instance of Zookeeper or not.
export HBASE_MANAGES_ZK=true

配置 regionservers 文件

[root@master /usr/local/hbase-0.98.24-hadoop1/conf]# cat regionservers
master
slave1
slave2

同步/usr/local/hbase-0.98.24-hadoop1到slave1、slavd2

rsync -avzP /usr/local/hbase-0.98.24-hadoop1 root@slave1:/usr/local/
rsync -avzP /usr/local/hbase-0.98.24-hadoop1 root@slave2:/usr/local/

编辑 /etc/bashrc 文件配置环境变量

三个节点都要配置

[root@master ~]# tail -n18 /etc/bashrc

# java conf
export JAVA_HOME=/usr/local/jdk1.6.0_45

# hadoop conf
export HADOOP_HOME=/usr/local/hadoop-1.2.1

# zookeeper conf
export ZOOKEEPER_HOME=/usr/local/zookeeper-3.4.5

# Hbase conf
export HBASE_HOME=/usr/local/hbase-0.98.24-hadoop1
export HBASE_CLASSPATH=$HBASE_HOME/conf
export HBASE_LOG_DIR=$HBASE_HOME/logs

export CLASSPATH=.:$CLASSPATH:$JAVA_HOME/lib
export PATH=$JAVA_HOME/bin:/usr/local/hadoop-1.2.1/bin:$ZOOKEEPER_HOME/bin:$HBASE_HOME/bin:$PATH

[root@master ~]# source /etc/bashrc

2. 启动Hbase

因此处使用hbase自带zookeeper，所以需要停掉之前安装的zookeeper

在主节点启动Hbase

[root@master ~]# start-hbase.sh
slave1: starting zookeeper, logging to /usr/local/hbase-0.98.24-hadoop1/logs/hbase-root-zookeeper-slave1.out
master: starting zookeeper, logging to /usr/local/hbase-0.98.24-hadoop1/logs/hbase-root-zookeeper-master.out
slave2: starting zookeeper, logging to /usr/local/hbase-0.98.24-hadoop1/logs/hbase-root-zookeeper-slave2.out
slave1: SLF4J: Class path contains multiple SLF4J bindings.
slave1: SLF4J: Found binding in [jar:file:/usr/local/hbase-0.98.24-hadoop1/lib/slf4j-log4j12-1.6.4.jar!/org/slf4j/impl/StaticLoggerBinder.class]
slave1: SLF4J: Found binding in [jar:file:/usr/local/hadoop-1.2.1/lib/slf4j-log4j12-1.4.3.jar!/org/slf4j/impl/StaticLoggerBinder.class]
slave1: SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
master: SLF4J: Class path contains multiple SLF4J bindings.
master: SLF4J: Found binding in [jar:file:/usr/local/hbase-0.98.24-hadoop1/lib/slf4j-log4j12-1.6.4.jar!/org/slf4j/impl/StaticLoggerBinder.class]
master: SLF4J: Found binding in [jar:file:/usr/local/hadoop-1.2.1/lib/slf4j-log4j12-1.4.3.jar!/org/slf4j/impl/StaticLoggerBinder.class]
master: SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
slave2: SLF4J: Class path contains multiple SLF4J bindings.
slave2: SLF4J: Found binding in [jar:file:/usr/local/hbase-0.98.24-hadoop1/lib/slf4j-log4j12-1.6.4.jar!/org/slf4j/impl/StaticLoggerBinder.class]
slave2: SLF4J: Found binding in [jar:file:/usr/local/hadoop-1.2.1/lib/slf4j-log4j12-1.4.3.jar!/org/slf4j/impl/StaticLoggerBinder.class]
slave2: SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
starting master, logging to /usr/local/hbase-0.98.24-hadoop1/logs/hbase-root-master-master.out
slave2: starting regionserver, logging to /usr/local/hbase-0.98.24-hadoop1/logs/hbase-root-regionserver-slave2.out
slave1: starting regionserver, logging to /usr/local/hbase-0.98.24-hadoop1/logs/hbase-root-regionserver-slave1.out
master: starting regionserver, logging to /usr/local/hbase-0.98.24-hadoop1/logs/hbase-root-regionserver-master.out

[root@master ~]# jps
9101 JobTracker
9010 SecondaryNameNode
9519 HQuorumPeer
8839 NameNode
10125 Jps
9640 HMaster
9797 HRegionServer

[root@slave1 ~]# jps
9074 HRegionServer
9293 Jps
8818 TaskTracker
8720 DataNode
8950 HQuorumPeer

[root@slave2 ~]# jps
8759 DataNode
8857 TaskTracker
9348 Jps
8990 HQuorumPeer
9120 HRegionServer

3. hbase shell检查服务状态

进入hbase shell

[root@master ~]# hbase shell
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 0.98.24-hadoop1, r9c13a1c3d8cf999014f30104d1aa9d79e74ca3d6, Thu Dec 22 02:28:55 UTC 2016

hbase(main):001:0>

检查状态

hbase(main):001:0> status
...
1 active master, 0 backup masters, 3 servers, 0 dead, 0.6667 average load

4. 安装Thrift Server

下载thrift安装包到/usr/local/src目录下。

https://archive.apache.org/dist/thrift/0.8.0/thrift-0.8.0.tar.gz

[root@master /usr/local/src]# ls thrift-0.8.0.tar.gz
thrift-0.8.0.tar.gz

准备编译环境

[root@master ~]# yum -y install automake libtool flex bison pkgconfig gcc-c++ boost-devel libevent-devel zlib-devel python-devel ruby-devel openssl-devel

安装Thrift

[root@master /usr/local/src]# tar zxf thrift-0.8.0.tar.gz
[root@master /usr/local/src]# cd thrift-0.8.0
[root@master /usr/local/src/thrift-0.8.0]# ./configure --with-cpp=no --with-ruby=no
...
thrift 0.8.0

Building code generators ..... :

Building C++ Library ......... : no
Building C (GLib) Library .... : no
Building Java Library ........ : no
Building C# Library .......... : no
Building Python Library ...... : yes
Building Ruby Library ........ : no
Building Haskell Library ..... : no
Building Perl Library ........ : no
Building PHP Library ......... : no
Building Erlang Library ...... : no
Building Go Library .......... : no

Using Python ................. : /usr/bin/python

If something is missing that you think should be present,
please skim the output of configure to find the missing
component.  Details are present in config.log.
[root@master /usr/local/src/thrift-0.8.0]# echo $?
0
[root@master /usr/local/src/thrift-0.8.0]# make
make  all-recursive
make[1]: 进入目录“/usr/local/src/thrift-0.8.0”
Making all in compiler/cpp
make[2]: 进入目录“/usr/local/src/thrift-0.8.0/compiler/cpp”
make  all-am
make[3]: 进入目录“/usr/local/src/thrift-0.8.0/compiler/cpp”
...
make[3]: 离开目录“/usr/local/src/thrift-0.8.0/test”
make[2]: 离开目录“/usr/local/src/thrift-0.8.0/test”
make[2]: 进入目录“/usr/local/src/thrift-0.8.0”
make[2]: 离开目录“/usr/local/src/thrift-0.8.0”
make[1]: 离开目录“/usr/local/src/thrift-0.8.0”
[root@master /usr/local/src/thrift-0.8.0]# echo $?
0
[root@master /usr/local/src/thrift-0.8.0]# make install
...
make[2]: 对“install-exec-am”无需做任何事。
make[2]: 对“install-data-am”无需做任何事。
make[2]: 离开目录“/usr/local/src/thrift-0.8.0”
make[1]: 离开目录“/usr/local/src/thrift-0.8.0”
[root@master /usr/local/src/thrift-0.8.0]# echo $?
0

启动Thrift server

[root@master ~]# hbase-daemon.sh start thrift
starting thrift, logging to /usr/local/hbase-0.98.24-hadoop1/logs/hbase-root-thrift-master.out
[root@master ~]# jps
9101 JobTracker
21649 ThriftServer
9010 SecondaryNameNode
9519 HQuorumPeer
8839 NameNode
9640 HMaster
21743 Jps
9797 HRegionServer

5. 安装Thrift api for python

产生针对Python的Hbase的API

下载hbase源码：

https://archive.apache.org/dist/hbase/0.98.24/hbase-0.98.24-src.tar.gz

[root@master /usr/local/src]# ls hbase-0.98.24-src.tar.gz
hbase-0.98.24-src.tar.gz

解压寻找Hbase.thrift文件

[root@master /usr/local/src]# tar zxf hbase-0.98.24-src.tar.gz
[root@master /usr/local/src]# cd hbase-0.98.24
[root@master /usr/local/src/hbase-0.98.24]# find . -name Hbase.thrift
./hbase-thrift/src/main/resources/org/apache/hadoop/hbase/thrift/Hbase.thrift

生成python的hbase模块

[root@master /usr/local/src/hbase-0.98.24]# cd ./hbase-thrift/src/main/resources/org/apache/hadoop/hbase/thrift
[root@master /usr/local/src/hbase-0.98.24/hbase-thrift/src/main/resources/org/apache/hadoop/hbase/thrift]# ls
Hbase.thrift
[root@master /usr/local/src/hbase-0.98.24/hbase-thrift/src/main/resources/org/apache/hadoop/hbase/thrift]# thrift -gen py Hbase.thrift
[root@master /usr/local/src/hbase-0.98.24/hbase-thrift/src/main/resources/org/apache/hadoop/hbase/thrift]# ls gen-py
hbase  __init__.py

安装hbase模块

[root@master /usr/local/src/hbase-0.98.24/hbase-thrift/src/main/resources/org/apache/hadoop/hbase/thrift]# cp -raf gen-py/hbase /usr/lib64/python2.7/site-packages/

Hbase的Python操作

实例化连接

#!/usr/bin/env python
# -*- coding: utf-8 -*-
from thrift import Thrift
from thrift.transport import TSocket
from thrift.transport import TTransport
from thrift.protocol import TBinaryProtocol

from hbase import Hbase
from hbase.ttypes import *

transport = TSocket.TSocket('master', 9090)
transport = TTransport.TBufferedTransport(transport)

protocol = TBinaryProtocol.TBinaryProtocol(transport)

client = Hbase.Client(protocol)

transport.open()

#========

实例1：创建表

#========

base_info_contents = ColumnDescriptor(name='meta_data:', maxVersions=1)
other_info_contents = ColumnDescriptor(name='flags:', maxVersions=1)

client.createTable('new_music_table', [base_info_contents, other_info_contents])

print client.getTableNames()

[root@master ~/hbase_test]# python create_table.py
['music_table', 'new_music_table']

实例2：插入数据

#========
tableName = 'new_music_table'
rowKey = '1100'

mutations = [Mutation(column="meta_data:name", value="wangqingshui"), \
        Mutation(column="meta_data:tag", value="pop"), \
        Mutation(column="flags:is_valid", value="TRUE")]

client.mutateRow(tableName, rowKey, mutations, None)

实例3：读取多条记录

#========
tableName = 'new_music_table'

scan = TScan()
id = client.scannerOpenWithScan(tableName, scan, None)
result = client.scannerGetList(id, 10)

for r in result:
    print '======'
    print 'the row is ' , r.row

    for k, v in r.columns.items():
        print "\t".join([k, v.value])

实例4：读取指定row key记录

#========
tableName = 'new_music_table'
rowKey = '1100'

result = client.getRow(tableName, rowKey, None)

for r in result:
    print 'the row is ' , r.row
    print 'the name is ' , r.columns.get('meta_data:name').value
    print 'the flag is ' , r.columns.get('flags:is_valid').value