amoeba + mysql-mmm 实现读写分离

news/2024/7/4 1:41:26 标签: server, mysql, java, class, 数据库, schema
class="baidu_pl">
class="article_content clearfix">
class="htmledit_views">

class="tags" href="/tags/MYSQL.html" title=mysql>mysql-mmm配置请查看：http://blog.csdn.net/xyz846/article/details/7451774

class="tags" href="/tags/MYSQL.html" title=mysql>mysql-mmm只是把数据库的读写功能分离出来，对于web服务器不可能说我要执行读的功能就连接哪个IP，要执行写的功能就连接哪个IP，这是不可能的，解决这一问题我使用amoeba（变形虫）数据库透明代理，根据执行的sql可自动路由至相应的Ip，很好地结合class="tags" href="/tags/MYSQL.html" title=mysql>mysql-mmm真正实现读写分离。

整个拓扑如下：

amoeba服务器我也安装在class="tags" href="/tags/SERVER.html" title=server>server4上，下面是我的安装步骤：

一、安装class="tags" href="/tags/JAVA.html" title=java>java

wget http://download.oracle.com/otn-pub/class="tags" href="/tags/JAVA.html" title=java>java/jdk/6u25-b06/jdk-6u25-linux-i586.bin 
chmod 755 jdk-6u25-linux-i586.bin 
./jdk-6u25-linux-i586.bin 
mv jdk1.6.0_25/ /usr/local/jdk

配置环境变量vi /etc/profile

export JAVA_HOME=/usr/local/jdk/
export CLASSPATH=.:$JAVA_HOME/jre/lib/rt.jar:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar
export PATH=$JAVA_HOME/bin:$PATH

使文件即时生效

 source /etc/profile

检查

[root@localhost opt]# class="tags" href="/tags/JAVA.html" title=java>java -version
class="tags" href="/tags/JAVA.html" title=java>java version "1.6.0_25"
Java(TM) SE Runtime Environment (build 1.6.0_25-b06)
Java HotSpot(TM) Client VM (build 20.0-b11, mixed mode, sharing)

二、安装amoeba

 mkdir /usr/local/amoeba
mv amoeba-class="tags" href="/tags/MYSQL.html" title=mysql>mysql-binary-2.1.0-RC5.tar.gz  /usr/local/amoeba/
cd /usr/local/amoeba/
tar zxvf amoeba-class="tags" href="/tags/MYSQL.html" title=mysql>mysql-binary-2.1.0-RC5.tar.gz
chmod 755 bin/*
cd bin/
./amoeba

运行amoeba会显现如下信息

 [root@localhost bin]# ./amoeba 
amoeba start|stop

三、配置读写分离

Amoeba的配置分成几个文件：

amoeba.xml      需要修改 主配置文件，配置所有数据源及Amoeba自身的参数
dbServer.xml     需要修改
access_list.conf    需要修改
rule.xml         数据切分时使用
log4j.xml         无需修改
functionMap.xml 无需修改        配置用于解析Query中的函数所对应的Java实现类nullFunctionMap.xml 无需修改     配置路由规则中需要使用到的特定函数的实现类。

 dbServer.xml

<factoryConfig class="com.meidusa.amoeba.class="tags" href="/tags/MYSQL.html" title=mysql>mysql.net.MysqlServerConnectionFactory">
                        <property name="manager">${defaultManager}</property>
                        <property name="sendBufferSize">64</property>
                        <property name="receiveBufferSize">128</property>

                        <!-- class="tags" href="/tags/MYSQL.html" title=mysql>mysql port -->
                        <property name="port">3306</property>

                        <!-- class="tags" href="/tags/MYSQL.html" title=mysql>mysql schema -->
                        <property name="schema">test</property>

                        <!-- class="tags" href="/tags/MYSQL.html" title=mysql>mysql user -->
                        <property name="user">lihuipeng</property>

                        <!--  class="tags" href="/tags/MYSQL.html" title=mysql>mysql password-->
                        <property name="password">lihuipeng</property>
                        
                </factoryConfig>

……

        <dbServer name="class="tags" href="/tags/SERVER.html" title=server>server1"  parent="abstractServer">
                <factoryConfig>
                        <!-- class="tags" href="/tags/MYSQL.html" title=mysql>mysql ip -->
                        <property name="ipAddress">192.168.93.141</property>
                </factoryConfig>
        </dbServer>

        <dbServer name="class="tags" href="/tags/SERVER.html" title=server>server2"  parent="abstractServer">
                <factoryConfig>
                        <!-- class="tags" href="/tags/MYSQL.html" title=mysql>mysql ip -->
                        <property name="ipAddress">192.168.93.142</property>
                </factoryConfig>
        </dbServer>

        <dbServer name="class="tags" href="/tags/SERVER.html" title=server>server3"  parent="abstractServer">
                <factoryConfig>
                        <!-- class="tags" href="/tags/MYSQL.html" title=mysql>mysql ip -->
                        <property name="ipAddress">192.168.93.144</property>
                </factoryConfig>
        </dbServer>

        <dbServer name="ReadPool" virtual="true">
                <poolConfig class="com.meidusa.amoeba.class="tags" href="/tags/SERVER.html" title=server>server.MultipleServerPool">
                        <!-- Load balancing strategy: 1=ROUNDROBIN , 2=WEIGHTBASED , 3=HA-->
                        <property name="loadbalance">1</property>

                        <!-- Separated by commas,such as: class="tags" href="/tags/SERVER.html" title=server>server1,class="tags" href="/tags/SERVER.html" title=server>server2,class="tags" href="/tags/SERVER.html" title=server>server1 -->
                        <property name="poolNames">class="tags" href="/tags/SERVER.html" title=server>server2,class="tags" href="/tags/SERVER.html" title=server>server3</property>
                </poolConfig>
        </dbServer>

定义dbclass="tags" href="/tags/SERVER.html" title=server>server的时候，记得参考mmm_control show的信息，关键是write角色的机器一定要用write的ip

 [root@localhost ~]# mmm_control show
  db1(192.168.93.137) master/ONLINE. Roles: reader(192.168.93.143), writer(192.168.93.141)
  db2(192.168.93.138) master/ONLINE. Roles: reader(192.168.93.142)
  db3(192.168.93.139) slave/ONLINE. Roles: reader(192.168.93.144)

amoeba.xml

<property name="authenticator">
          <bean class="com.meidusa.amoeba.class="tags" href="/tags/MYSQL.html" title=mysql>mysql.class="tags" href="/tags/SERVER.html" title=server>server.MysqlClientAuthenticator">

                                        <property name="user">amoeba</property>

                                        <property name="password">amoeba</property>

                                        <property name="filter">
                                                <bean class="com.meidusa.amoeba.class="tags" href="/tags/SERVER.html" title=server>server.IPAccessController">
                                                        <property name="ipFile">${amoeba.home}/conf/access_list.conf</property>
                                                </bean>
                                        </property>
                                </bean>
                        </property>
 

……

      <queryRouter class="com.meidusa.amoeba.class="tags" href="/tags/MYSQL.html" title=mysql>mysql.parser.MysqlQueryRouter">
                <property name="ruleLoader">
                        <bean class="com.meidusa.amoeba.route.TableRuleFileLoader">
                                <property name="ruleFile">${amoeba.home}/conf/rule.xml</property>
                                <property name="functionFile">${amoeba.home}/conf/ruleFunctionMap.xml</property>
                        </bean>
                </property>
                <property name="sqlFunctionFile">${amoeba.home}/conf/functionMap.xml</property>
                <property name="LRUMapSize">1500</property>
                <property name="defaultPool">class="tags" href="/tags/SERVER.html" title=server>server1</property>


                <property name="writePool">class="tags" href="/tags/SERVER.html" title=server>server1</property>
                <property name="readPool">ReadPool</property>

                <property name="needParse">true</property>
        </queryRouter>

第一处修改web数据库可以连接使用的帐号密码，第二处修改amoeba读 写对应的class="tags" href="/tags/SERVER.html" title=server>server，该class="tags" href="/tags/SERVER.html" title=server>server就是我们之前定义的。

启动amoeba

 nohup /usr/local/amoeba/bin/amoeba start &

查看进程

 [root@localhost conf]# netstat -tunlp | grep class="tags" href="/tags/JAVA.html" title=java>java
tcp        0      0 :::8066                     :::*                        LISTEN      28426/class="tags" href="/tags/JAVA.html" title=java>java          
tcp        0      0 ::ffff:127.0.0.1:19356      :::*                        LISTEN      28426/class="tags" href="/tags/JAVA.html" title=java>java

测试登陆

 [root@localhost conf]# class="tags" href="/tags/MYSQL.html" title=mysql>mysql -uamoeba -pamoeba -P8066 -h192.168.93.140
Welcome to the MySQL monitor.  Commands end with ; or \g.
Your MySQL connection id is 17244592
Server version: 5.1.45-class="tags" href="/tags/MYSQL.html" title=mysql>mysql-amoeba-proxy-2.1.0-RC5 Source distribution

Copyright (c) 2000, 2011, Oracle and/or its affiliates. All rights reserved.

Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

class="tags" href="/tags/MYSQL.html" title=mysql>mysql> show databases;
+--------------------+
| Database           |
+--------------------+
| information_schema |
| test               |
+--------------------+
2 rows in set (0.02 sec)
 

配置完了！想测试读写分离效果的童鞋，可以将三台db的主从先关掉，然后在各个库插入一条不同的数据，然后登陆amoeba进行多次搜索就可以看到结果了！


http://www.niftyadmin.cn/n/1552221.html

相关文章

nutch源码分析---2

nutch源码分析---2

nutch源码分析—generate 根据上一章的分析&#xff0c;“bin/nutch generate crawl/crawldb crawl/segments”这条命令最终会调用org.apache.nutch.crawl.Generator的main函数。 Generator::main public static void main(String args[]) throws Exception {int res ToolRu…
阅读更多...
通过Oracle Stream实现数据库之间的同步

通过Oracle Stream实现数据库之间的同步

Oracle Stream功能是为提高数据库的高可用性而设计的&#xff0c;在Oracle 9i及之前的版本这个功能被称为Advance Replication。Oracle Stream利用高级队列技术&#xff0c;通过解析归档日志&#xff0c;将归档日志解析成DDL及DML语句&#xff0c;从而实现数据库之间的同步。这…
阅读更多...
lucene源码分析---10

lucene源码分析---10

lucene源码分析—倒排索引的读过程 上一章中分析了lucene倒排索引的写过程&#xff0c;本章开始分析其读过程&#xff0c;重点分析SegmentTermsEnum的seekExact函数。 首先看几个构造函数&#xff0c;先看SegmentCoreReaders的构造函数&#xff0c;在Lucene50PostingFormat的…
阅读更多...
V$FLASHBACK_DATABASE_LOG和V$FLASHBACK_DATABASE_STAT

V$FLASHBACK_DATABASE_LOG和V$FLASHBACK_DATABASE_STAT

V$FLASHBACK_DATABASE_LOG显示闪回数据的信息&#xff0c;用来评估当前负载下需要的闪回空间v$flashback_database_logName Description------------------------------ -------------------------- OLDEST_F…
阅读更多...
nutch源码分析---3

nutch源码分析---3

nutch源码分析—fetch 根据上一章的分析&#xff0c;“bin/nutch fetch crawl/segments/*”这条命令最终会调用org.apache.nutch.fetcher.Fetcher的main函数。 public static void main(String[] args) throws Exception {int res ToolRunner.run(NutchConfiguration.create(…
阅读更多...
利用LVS+Keepalived 实现高性能高可用负载均衡

利用LVS+Keepalived 实现高性能高可用负载均衡

背景: 随着你的网站业务量的增长你网站的服务器压力越来越大&#xff1f;需要负载均衡方案&#xff01;商业的硬件如F5又太贵&#xff0c;你们又是创业型互联公司如何有效节约成本&#xff0c;节省不必要的浪费&#xff1f;同时实现商业硬件一样的高性能高可用的功能&#xff1…
阅读更多...
nutch源码分析---4

nutch源码分析---4

nutch源码分析—parse “bin/nutch parse crawl/segments/*”这条命令最终会调用org.apache.nutch.parse.ParseSegment的main函数。 ParseSegment::main public static void main(String[] args) throws Exception {int res ToolRunner.run(NutchConfiguration.create(), ne…
阅读更多...
nutch源码分析---5

nutch源码分析---5

nutch源码分析—updatedb “bin/nutch updatedb crawl/crawldb crawl/segments/2*”这条命令最终会执行org.apache.nutch.crawl.CrawlDb的main函数。 public static void main(String[] args) throws Exception {int res ToolRunner.run(NutchConfiguration.create(), new Cr…
阅读更多...
最新文章

Copyright @ 2022~2023