1. 多数据源配置的核心场景与价值在Java企业级应用开发中多数据源配置是一个高频出现的需求场景。我经历过多个需要同时连接不同数据库的项目比如主业务库报表库分离的电商系统生产库历史库的金融业务系统本地库第三方系统库的政务平台这种架构设计的核心价值在于性能隔离将OLTP和OLAP查询分流到不同数据库实例安全隔离核心业务数据与日志/报表数据权限分离技术异构不同业务模块使用最适合的数据库类型如MySQLRedisMongoDB组合2. Spring Boot多数据源实现方案2.1 基础环境准备以Spring Boot 2.7 MySQL为例需在pom.xml中声明dependency groupIdorg.springframework.boot/groupId artifactIdspring-boot-starter-data-jpa/artifactId /dependency dependency groupIdmysql/groupId artifactIdmysql-connector-java/artifactId scoperuntime/scope /dependency2.2 多数据源配置类创建主数据源配置示例使用HikariCP连接池Configuration EnableTransactionManagement EnableJpaRepositories( basePackages com.example.primary, entityManagerFactoryRef primaryEntityManager, transactionManagerRef primaryTransactionManager ) public class PrimaryDataSourceConfig { Primary Bean(name primaryDataSource) ConfigurationProperties(prefix spring.datasource.primary) public DataSource primaryDataSource() { return DataSourceBuilder.create().build(); } Primary Bean(name primaryEntityManager) public LocalContainerEntityManagerFactoryBean primaryEntityManager( EntityManagerFactoryBuilder builder, Qualifier(primaryDataSource) DataSource dataSource) { return builder .dataSource(dataSource) .packages(com.example.primary) .persistenceUnit(primaryPU) .build(); } Primary Bean(name primaryTransactionManager) public PlatformTransactionManager primaryTransactionManager( Qualifier(primaryEntityManager) EntityManagerFactory entityManagerFactory) { return new JpaTransactionManager(entityManagerFactory); } }2.3 从数据源配置Configuration EnableTransactionManagement EnableJpaRepositories( basePackages com.example.secondary, entityManagerFactoryRef secondaryEntityManager, transactionManagerRef secondaryTransactionManager ) public class SecondaryDataSourceConfig { Bean(name secondaryDataSource) ConfigurationProperties(prefix spring.datasource.secondary) public DataSource secondaryDataSource() { return DataSourceBuilder.create().build(); } Bean(name secondaryEntityManager) public LocalContainerEntityManagerFactoryBean secondaryEntityManager( EntityManagerFactoryBuilder builder, Qualifier(secondaryDataSource) DataSource dataSource) { return builder .dataSource(dataSource) .packages(com.example.secondary) .persistenceUnit(secondaryPU) .build(); } Bean(name secondaryTransactionManager) public PlatformTransactionManager secondaryTransactionManager( Qualifier(secondaryEntityManager) EntityManagerFactory entityManagerFactory) { return new JpaTransactionManager(entityManagerFactory); } }3. 配置参数详解application.yml中的典型配置spring: datasource: primary: url: jdbc:mysql://localhost:3306/primary_db username: root password: 123456 driver-class-name: com.mysql.cj.jdbc.Driver hikari: maximum-pool-size: 10 minimum-idle: 5 secondary: url: jdbc:mysql://localhost:3306/secondary_db username: report password: report123 driver-class-name: com.mysql.cj.jdbc.Driver hikari: maximum-pool-size: 20 minimum-idle: 10关键参数说明maximum-pool-size根据QPS设置建议OLTP库设置较小10-20报表库可适当增大minimum-idle保持最小连接数避免突发请求时建立连接的开销connection-timeout获取连接超时时间默认30秒4. 事务管理实践4.1 单数据源事务Service public class OrderService { Transactional(primaryTransactionManager) public void createOrder(Order order) { // 使用主数据源执行操作 } }4.2 跨数据源事务需要引入JTA实现如AtomikosBean public JtaTransactionManager transactionManager() { return new JtaTransactionManager(); } // 使用示例 Transactional public void crossDatabaseOperation() { // 操作主库 primaryRepository.save(entity); // 操作从库 secondaryRepository.save(entity); }警告分布式事务性能损耗较大应尽量避免高频使用5. 动态数据源切换方案对于需要运行时动态切换的场景可采用AbstractRoutingDataSourcepublic class DynamicDataSource extends AbstractRoutingDataSource { Override protected Object determineCurrentLookupKey() { return DataSourceContextHolder.getDataSourceType(); } } // 使用ThreadLocal保存数据源标识 public class DataSourceContextHolder { private static final ThreadLocalString context new ThreadLocal(); public static void setDataSourceType(String type) { context.set(type); } public static String getDataSourceType() { return context.get(); } public static void clear() { context.remove(); } } // 切面实现 Aspect Component public class DataSourceAspect { Before(annotation(dataSource)) public void switchDataSource(DataSource dataSource) { DataSourceContextHolder.setDataSourceType(dataSource.value()); } After(annotation(dataSource)) public void restoreDataSource(DataSource dataSource) { DataSourceContextHolder.clear(); } }6. 性能优化建议连接池监控通过Spring Boot Actuator监控各数据源状态management: endpoint: health: show-details: always endpoints: web: exposure: include: health,info,metricsSQL日志分离为不同数据源配置独立的日志输出logging.level.com.example.primaryDEBUG logging.level.com.example.secondaryINFO连接泄漏检测添加以下配置spring: datasource: hikari: leak-detection-threshold: 5000 # 5秒7. 常见问题排查问题现象可能原因解决方案启动时报Bean创建冲突未正确指定Primary确保主数据源相关Bean标记Primary事务不生效使用了错误的事务管理器检查Transactional注解指定的manager名称连接泄漏未正确关闭连接使用try-with-resources或确保Transactional覆盖所有操作性能下降连接池配置不合理根据监控调整max-pool-size和min-idle8. 高级应用场景8.1 读写分离实现结合AbstractRoutingDataSourcepublic class ReadWriteDataSource extends AbstractRoutingDataSource { Override protected Object determineCurrentLookupKey() { return TransactionSynchronizationManager.isCurrentTransactionReadOnly() ? read : write; } }8.2 多租户数据隔离通过租户ID路由数据源public class TenantDataSourceRouter { public static String determineDataSource(String tenantId) { // 根据租户ID返回对应数据源key return tenant_ tenantId; } }在实际项目中多数据源配置需要根据具体业务场景灵活调整。我建议在开发初期就规划好数据源策略避免后期重构带来的风险。对于新项目可以考虑使用ShardingSphere等中间件来简化多数据源管理。