好久没有写过博客了,趁着年假还有一天,把去年项目所运用的读写分离在这里概述一下及其注意点,以防以后项目再有使用到;
准备工作
1 开发环境:window,idea,maven,spring boot,mybatis,druid(淘宝数据库连接池)
2 数据库服务器:linux,mysql master(192.168.203.135),mysql salve(192.168.203.139)
3 读写分离之前必须先做好数据库的主从复制,关于主从复制不是该篇幅的主要叙述重点,关于主从复制读者可以自行google或者百度,教程基本都是一样,可行
注意以下几点:
a:做主从复制时,首先确定两台服务器的mysql没任何自定义库(否则只可以配置完后之前的东西没法同步,或者两个库都有完全相同的库应该也是可以同步)
b:server_id必须配置不一样
c:防火墙不能把mysql服务端口给拦截了(默认3306)
d:确保两台mysql可以相互访问
e:重置master,slave。Reset master;reset slave;开启关闭slave,start slave;stop slave;
f:主DB server和从DB server数据库的版本一致
4 读写分离方式:
4-1 基于程序代码内部实现: 在代码中根据select 、insert进行路由分类,这类方法也是目前生产环境下应用最广泛的。优点是性能较好,因为程序在代码中实现,不需要增加额外的硬件开支,缺点是需要开发人员来实现,运维人员无从下手。
4-2 基于中间代理层实现: 代理一般介于应用服务器和数据库服务器之间,代理数据库服务器接收到应用服务器的请求后根据判断后转发到,后端数据库,有以下代表性的程序。
本文基于两种方式的叙述:
基于应用层代码实现方式(内容都是通过代码体现,必要的说明存在代码中)
1 配置pom.xml,导入需要的jar包
<?xml version="1.0" encoding="UTF-8"?> <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd"> <modelVersion>4.0.0</modelVersion> <groupId>com.lishun</groupId> <artifactId>mysql_master_salve</artifactId> <version>0.0.1-SNAPSHOT</version> <packaging>jar</packaging> <name>mysql_master_salve</name> <description>Demo project for Spring Boot</description> <parent> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-parent</artifactId> <version>1.5.10.RELEASE</version> <relativePath/> <!-- lookup parent from repository --> </parent> <properties> <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding> <project.reporting.outputEncoding>UTF-8</project.reporting.outputEncoding> <java.version>1.8</java.version> </properties> <dependencies> <dependency> <groupId>org.mybatis.spring.boot</groupId> <artifactId>mybatis-spring-boot-starter</artifactId> <version>1.3.1</version> </dependency> <dependency> <groupId>mysql</groupId> <artifactId>mysql-connector-java</artifactId> <scope>runtime</scope> </dependency> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-test</artifactId> <scope>test</scope> </dependency> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-web</artifactId> <version>RELEASE</version> </dependency> <dependency> <groupId>com.alibaba</groupId> <artifactId>druid</artifactId> <version>1.0.18</version> </dependency> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-aop</artifactId> </dependency> </dependencies> <build> <plugins> <plugin> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-maven-plugin</artifactId> </plugin> <plugin> <groupId>org.mybatis.generator</groupId> <artifactId>mybatis-generator-maven-plugin</artifactId> <version>1.3.2</version> <dependencies> <dependency> <groupId>mysql</groupId> <artifactId>mysql-connector-java</artifactId> <version>5.1.43</version> </dependency> </dependencies> <configuration> <overwrite>true</overwrite> </configuration> </plugin> </plugins> </build> </project>
2 配置application.properties
server.port=9022 #mybatis配置*mapper.xml文件和实体别名 mybatis.mapper-locations=classpath:mapper/*.xml mybatis.type-aliases-package=com.lishun.entity spring.datasource.driver-class-name=com.mysql.jdbc.Driver spring.datasource.password=123456 spring.datasource.username=root #写节点 spring.datasource.master.url=jdbc:mysql://192.168.203.135:3306/worldmap #两个个读节点(为了方便测试这里用的是同一个服务器数据库,生产环境应该不使用) spring.datasource.salve1.url=jdbc:mysql://192.168.203.139:3306/worldmap spring.datasource.salve2.url=jdbc:mysql://192.168.203.139:3306/worldmap # druid 连接池 Setting # 初始化大小,最小,最大 spring.datasource.type=com.alibaba.druid.pool.DruidDataSource spring.datasource.initialSize=5 spring.datasource.minIdle=5 spring.datasource.maxActive=20 # 配置获取连接等待超时的时间 spring.datasource.maxWait=60000 # 配置间隔多久才进行一次检测,检测需要关闭的空闲连接,单位是毫秒 spring.datasource.timeBetweenEvictionRunsMillis=60000 # 配置一个连接在池中最小生存的时间,单位是毫秒 spring.datasource.minEvictableIdleTimeMillis=300000 spring.datasource.validationQuery=SELECT 1 FROM rscipc_sys_user spring.datasource.testWhileIdle=true spring.datasource.testOnBorrow=false spring.datasource.testOnReturn=false # 打开PSCache,并且指定每个连接上PSCache的大小 spring.datasource.poolPreparedStatements=true spring.datasource.maxPoolPreparedStatementPerConnectionSize=20 # 配置监控统计拦截的filters,去掉后监控界面sql无法统计,'wall'用于防火墙 spring.datasource.filters=stat,wall,log4j # 通过connectProperties属性来打开mergeSql功能;慢SQL记录 spring.datasource.connectionProperties=druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000 spring.datasource.logSlowSql=true #End
3 启动类(注意:其他需要spring管理的bean(service,config等)必须放在该启动类的子包下,不然会扫描不到bean,导致注入失败)
@SpringBootApplication @MapperScan("com.lishun.mapper") //!!!!!! 注意:扫描所有mapper public class MysqlMasterSalveApplication { public static void main(String[] args) { SpringApplication.run(MysqlMasterSalveApplication.class, args); } }
4 动态数据源 DynamicDataSource
/** * @author lishun * @Description:动态数据源, 继承AbstractRoutingDataSource * @date 2017/8/9 */ public class DynamicDataSource extends AbstractRoutingDataSource { public static final Logger log = LoggerFactory.getLogger(DynamicDataSource.class); /** * 默认数据源 */ public static final String DEFAULT_DS = "read_ds"; private static final ThreadLocal<String> contextHolder = new ThreadLocal<>(); public static void setDB(String dbType) {// 设置数据源名 log.info("切换到{}数据源", dbType); contextHolder.set(dbType); } public static void clearDB() { contextHolder.remove(); }// 清除数据源名 @Override protected Object determineCurrentLookupKey() { return contextHolder.get(); } }
5 线程池配置数据源
@Configuration public class DruidConfig { private Logger logger = LoggerFactory.getLogger(DruidConfig.class); @Value("${spring.datasource.master.url}") private String masterUrl; @Value("${spring.datasource.salve1.url}") private String salve1Url; @Value("${spring.datasource.salve2.url}") private String salve2Url; @Value("${spring.datasource.username}") private String username; @Value("${spring.datasource.password}") private String password; @Value("${spring.datasource.driver-class-name}") private String driverClassName; @Value("${spring.datasource.initialSize}") private int initialSize; @Value("${spring.datasource.minIdle}") private int minIdle; @Value("${spring.datasource.maxActive}") private int maxActive; @Value("${spring.datasource.maxWait}") private int maxWait; @Value("${spring.datasource.timeBetweenEvictionRunsMillis}") private int timeBetweenEvictionRunsMillis; @Value("${spring.datasource.minEvictableIdleTimeMillis}") private int minEvictableIdleTimeMillis; @Value("${spring.datasource.validationQuery}") private String validationQuery; @Value("${spring.datasource.testWhileIdle}") private boolean testWhileIdle; @Value("${spring.datasource.testOnBorrow}") private boolean testOnBorrow; @Value("${spring.datasource.testOnReturn}") private boolean testOnReturn; @Value("${spring.datasource.filters}") private String filters; @Value("${spring.datasource.logSlowSql}") private String logSlowSql; @Bean public ServletRegistrationBean druidServlet() { logger.info("init Druid Servlet Configuration "); ServletRegistrationBean reg = new ServletRegistrationBean(); reg.setServlet(new StatViewServlet()); reg.addUrlMappings("/druid/*"); reg.addInitParameter("loginUsername", username); reg.addInitParameter("loginPassword", password); reg.addInitParameter("logSlowSql", logSlowSql); return reg; } @Bean public FilterRegistrationBean filterRegistrationBean() { FilterRegistrationBean filterRegistrationBean = new FilterRegistrationBean(); filterRegistrationBean.setFilter(new WebStatFilter()); filterRegistrationBean.addUrlPatterns("/*"); filterRegistrationBean.addInitParameter("exclusions", "*.js,*.gif,*.jpg,*.png,*.css,*.ico,/druid/*"); filterRegistrationBean.addInitParameter("profileEnable", "true"); return filterRegistrationBean; } @Bean public DataSource druidDataSource() { DruidDataSource datasource = new DruidDataSource(); datasource.setUrl(masterUrl); datasource.setUsername(username); datasource.setPassword(password); datasource.setDriverClassName(driverClassName); datasource.setInitialSize(initialSize); datasource.setMinIdle(minIdle); datasource.setMaxActive(maxActive); datasource.setMaxWait(maxWait); datasource.setTimeBetweenEvictionRunsMillis(timeBetweenEvictionRunsMillis); datasource.setMinEvictableIdleTimeMillis(minEvictableIdleTimeMillis); datasource.setValidationQuery(validationQuery); datasource.setTestWhileIdle(testWhileIdle); datasource.setTestOnBorrow(testOnBorrow); datasource.setTestOnReturn(testOnReturn); try { datasource.setFilters(filters); } catch (SQLException e) { logger.error("druid configuration initialization filter", e); } Map<Object, Object> dsMap = new HashMap(); dsMap.put("read_ds_1", druidDataSource_read1()); dsMap.put("read_ds_2", druidDataSource_read2()); dsMap.put("write_ds", datasource); DynamicDataSource dynamicDataSource = new DynamicDataSource(); dynamicDataSource.setTargetDataSources(dsMap); return dynamicDataSource; } public DataSource druidDataSource_read1() { DruidDataSource datasource = new DruidDataSource(); datasource.setUrl(salve1Url); datasource.setUsername(username); datasource.setPassword(password); datasource.setDriverClassName(driverClassName); datasource.setInitialSize(initialSize); datasource.setMinIdle(minIdle); datasource.setMaxActive(maxActive); datasource.setMaxWait(maxWait); datasource.setTimeBetweenEvictionRunsMillis(timeBetweenEvictionRunsMillis); datasource.setMinEvictableIdleTimeMillis(minEvictableIdleTimeMillis); datasource.setValidationQuery(validationQuery); datasource.setTestWhileIdle(testWhileIdle); datasource.setTestOnBorrow(testOnBorrow); datasource.setTestOnReturn(testOnReturn); try { datasource.setFilters(filters); } catch (SQLException e) { logger.error("druid configuration initialization filter", e); } return datasource; } public DataSource druidDataSource_read2() { DruidDataSource datasource = new DruidDataSource(); datasource.setUrl(salve2Url); datasource.setUsername(username); datasource.setPassword(password); datasource.setDriverClassName(driverClassName); datasource.setInitialSize(initialSize); datasource.setMinIdle(minIdle); datasource.setMaxActive(maxActive); datasource.setMaxWait(maxWait); datasource.setTimeBetweenEvictionRunsMillis(timeBetweenEvictionRunsMillis); datasource.setMinEvictableIdleTimeMillis(minEvictableIdleTimeMillis); datasource.setValidationQuery(validationQuery); datasource.setTestWhileIdle(testWhileIdle); datasource.setTestOnBorrow(testOnBorrow); datasource.setTestOnReturn(testOnReturn); try { datasource.setFilters(filters); } catch (SQLException e) { logger.error("druid configuration initialization filter", e); } return datasource; } }
6 数据源注解:在service层通过数据源注解来指定数据源
/** * @author lishun * @Description: 读数据源注解 * @date 2017/8/9 */ @Target({ElementType.METHOD}) @Retention(RetentionPolicy.RUNTIME) public @interface ReadDataSource { String vlaue() default "read_ds"; } /** * @author lishun * @Description: 写数据源注解 * @date 2017/8/9 */ @Target({ElementType.METHOD}) @Retention(RetentionPolicy.RUNTIME) public @interface WriteDataSource { String value() default "write_ds"; }
7 service aop切面来切换数据源
/** * @author lishun * @Description: TODO * @date 2017/8/9 */ @Component @Aspect public class ServiceAspect implements PriorityOrdered { @Pointcut("execution(public * com.lishun.service.*.*(..))") public void dataSource(){}; @Before("dataSource()") public void before(JoinPoint joinPoint){ Class<?> className = joinPoint.getTarget().getClass();//获得当前访问的class String methodName = joinPoint.getSignature().getName();//获得访问的方法名 Class[] argClass = ((MethodSignature)joinPoint.getSignature()).getParameterTypes();//得到方法的参数的类型 String dataSource = DynamicDataSource.DEFAULT_DS; try { Method method = className.getMethod(methodName, argClass);// 得到访问的方法对象 if (method.isAnnotationPresent(ReadDataSource.class)) { ReadDataSource annotation = method.getAnnotation(ReadDataSource.class); dataSource = annotation.vlaue(); int i = new Random().nextInt(2) + 1; /* 简单的负载均衡 */ dataSource = dataSource + "_" + i; }else if (method.isAnnotationPresent(WriteDataSource.class)){ WriteDataSource annotation = method.getAnnotation(WriteDataSource.class); dataSource = annotation.value(); } } catch (Exception e) { e.printStackTrace(); } DynamicDataSource.setDB(dataSource);// 切换数据源 } /* 基于方法名 @Before("execution(public * com.lishun.service.*.find*(..)) || execution(public * com.lishun.service.*.query*(..))") public void read(JoinPoint joinPoint){ DynamicDataSource.setDB("read_ds");// 切换数据源 } @Before("execution(public * com.lishun.service.*.insert*(..)) || execution(public * com.lishun.service.*.add*(..))") public void write(JoinPoint joinPoint){ DynamicDataSource.setDB("write_ds");// 切换数据源 } */ @After("dataSource()") public void after(JoinPoint joinPoint){ DynamicDataSource.clearDB();// 切换数据源 } @AfterThrowing("dataSource()") public void AfterThrowing(){ System.out.println("AfterThrowing---------------" ); } @Override public int getOrder() { return 1;//数值越小该切面先被执行,先选择数据源(防止事务aop使用数据源出现空异常) } }
8 测试 mapper的代码就不贴了,主要是service和controller
service
@Service @Transactional public class WmIpInfoServiceImpl implements WmIpInfoService { @Autowired public WmIpInfoMapper wmIpInfoMapper; @Override @ReadDataSource public WmIpInfo findOneById(String id) { //wmIpInfoMapper.selectByPrimaryKey(id); return wmIpInfoMapper.selectByPrimaryKey(id); } @Override @WriteDataSource public int insert(WmIpInfo wmIpInfo) { int result = wmIpInfoMapper.insert(wmIpInfo); return result; } }
contrlloer
@RestController public class IndexController { @Autowired public WmIpInfoService wmIpInfoService; @GetMapping("/index/{id}") public WmIpInfo index(@PathVariable(value = "id") String id){ WmIpInfo wmIpInfo = new WmIpInfo(); wmIpInfo.setId(UUID.randomUUID().toString()); wmIpInfoService.insert(wmIpInfo); wmIpInfoService.findOneById(id); return null; } }
运行spring boot 在浏览器输入http://localhost:9022/index/123456
查看日志
基于中间件方式实现读写分离(mycat:主要是mycat安装使用及其注意事项)
3-1 下载 http://dl.mycat.io/
3-2 解压,配置MYCAT_HOME;
3-3 修改文件 vim conf/schema.xml
<?xml version="1.0"?> <!DOCTYPE mycat:schema SYSTEM "schema.dtd"> <mycat:schema xmlns:mycat="http://io.mycat/"> <schema name="worldmap" checkSQLschema="false" sqlMaxLimit="100" dataNode="worldmap_node"></schema> <dataNode name="worldmap_node" dataHost="worldmap_host" database="worldmap" /> <!-- database:数据库名称 --> <dataHost name="worldmap_host" maxCon="1000" minCon="10" balance="1" writeType="0" dbType="mysql" dbDriver="native" switchType="2" slaveThreshold="100"> <heartbeat>select user()</heartbeat> <writeHost host="hostM1" url="192.168.203.135:3306" user="root" password="123456"><!--读写分离模式,写库:192.168.203.135,读库192.168.203.139--> <readHost host="hostR1" url="192.168.203.139:3306" user="root" password="123456" /> </writeHost> <writeHost host="hostM2" url="192.168.203.135:3306" user="root" password="123456"> <!--主从切换模式,当hostM1宕机,读写操作在hostM2服务器数据库执行--> </dataHost> </mycat:schema>
配置说明:
name:属性唯一标识dataHost标签,供上层的标签使用。
maxCon:最大连接数
minCon:最先连接数
balance
1、balance=0 不开启读写分离机制,所有读操作都发送到当前可用的writehost了 .
2、balance=1 全部的readhost与stand by writeHost 参与select语句的负载均衡。简单的说,双主双从模式(M1àS1,M2àS2,并且M1和M2互为主备),正常情况下,M1,S1,S2都参与select语句的复杂均衡。
3、balance=2 所有读操作都随机的在readhost和writehost上分发
writeType 负载均衡类型,目前的取值有3种:
1、writeType="0″, 所有写操作发送到配置的第一个writeHost。
2、writeType="1″,所有写操作都随机的发送到配置的writeHost。
3、writeType="2″,不执行写操作。
switchType
1、switchType=-1 表示不自动切换
2、switchType=1 默认值,自动切换
3、switchType=2 基于MySQL 主从同步的状态决定是否切换
dbType:数据库类型 mysql,postgresql,mongodb、oracle、spark等。
heartbeat:用于和后端数据库进行心跳检查的语句。例如,MYSQL可以使用select user(),Oracle可以使用select 1 from dual等。
这个标签还有一个connectionInitSql属性,主要是当使用Oracla数据库时,需要执行的初始化SQL语句就这个放到这里面来。例如:altersession set nls_date_format='yyyy-mm-dd hh24:mi:ss'
当switchType=2 主从切换的语句必须是:show slave status
writeHost、readHost:这两个标签都指定后端数据库的相关配置给mycat,用于实例化后端连接池。唯一不同的是,writeHost指定写实例、readHost指定读实例,
在一个dataHost内可以定义多个writeHost和readHost。但是,如果writeHost指定的后端数据库宕机,那么这个writeHost绑定的所有readHost都将不可用。
另一方面,由于这个writeHost宕机系统会自动的检测到,并切换到备用的writeHost上去。
3-4 修改文件 vim conf/server.xml
<!DOCTYPE mycat:server SYSTEM "server.dtd"> <mycat:server xmlns:mycat="http://io.mycat/"> <system> </system> <user name="root"> <property name="password">123456</property> <property name="schemas">worldmap</property><!--与schema.xml相对应--> <property name="readOnly">false</property> <!--readOnly是应用连接中间件逻辑库所具有的权限。true为只读,false为读写都有,默认为false。--> </user> </mycat:server>
3-5 启动 mycat start
查看启动日志:logs/wrapper.log;,正常启动成功后会有mycat.log日志,如果服务未启动成功不会有对应日志
3-6:对于开发人员mycat相当于一个新的数据库服务端(默认端口8066),开发人员增删改查不再是直接连接数据库,而是连接数据库中间件,中间件通过其自带的lua脚本进行sql判断,来路由到指定数据库(实质根据selet,insert,update,delete关键字)
3-7:测试读写分离
读数据路由到 192.168.203.139
写数据路由到192.168.203.135
当主库宕机,读写操作都在192.168.203.139
3-8:注意事项
一般使用框架都会用到事务,如果都要到事务那么就都会访问主服务器,达不到分离的效果,因此配置事务的时候要注意区分,比如只对包含增删改的进行事务配置