Question

我们有一个Spring引导应用程序，使用hikariCP作为连接池和多个数据源来处理长作业处理（基本上它使用相同的连接字符串，具有更长的超时和更少的连接数）

自1.5.2.RELEASE升级以来，只有第一个用@Primary注释的数据源才有其内务处理和池线程启动。虽然调试显示正在执行数据源初始化，但是简单地丢弃了第二个。

使用1.5.1.RELEASE这个工作正常，两组线程都会启动。

以下是我们的数据源定义

import java.util.Properties;

import javax.sql.DataSource;

import com.zaxxer.hikari.HikariDataSource;

import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.beans.factory.annotation.Qualifier;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Primary;
import org.springframework.context.annotation.Profile;
import org.springframework.stereotype.Component;
import org.springframework.validation.annotation.Validated;

import lombok.AllArgsConstructor;

@Profile("hikari")
@Validated
@AllArgsConstructor(onConstructor = @__(@Autowired))
@Component
class HikariDataSourceConfig
{
  private final DataSourceConfig dataSourceConfig;

  protected Properties setDataSourceProperties()
  {
    final Properties dataSourceProperties = new Properties();
    dataSourceProperties.put("driverType", "thin");
    dataSourceProperties.put("user", this.dataSourceConfig.getUsername());
    dataSourceProperties.put("password", this.dataSourceConfig.getPassword());
    dataSourceProperties.put("serverName", this.dataSourceConfig.getServer());
    dataSourceProperties.put("portNumber", String.valueOf(this.dataSourceConfig.getPort()));
    dataSourceProperties.put("databaseName", this.dataSourceConfig.getSid());
    return dataSourceProperties;
  }

  protected HikariDataSource initializeDataSource(final int maximumPoolSize)
  {
    final HikariDataSource dataSource = new HikariDataSource();
    dataSource.setMaximumPoolSize(maximumPoolSize);
    dataSource.setValidationTimeout(this.dataSourceConfig.getValidationTimeout() * 1000L);
    dataSource.setDataSourceClassName("oracle.jdbc.pool.OracleDataSource");
    dataSource.setConnectionTimeout(this.dataSourceConfig.getConnectionTimeout() * 1000L);
    dataSource.setMaxLifetime(this.dataSourceConfig.getMaxLifetime() * 1000L);
    dataSource.setIdleTimeout(this.dataSourceConfig.getIdleTimeout() * 1000L);
    dataSource.setAutoCommit(false);
    return dataSource;
  }

  @Bean(destroyMethod = "close")
  @Primary
  public DataSource dataSource()
  {
    final HikariDataSource dataSource = initializeDataSource(this.dataSourceConfig.getMaximumPoolSize());
    dataSource.setPoolName(DataSourceConfig.CONNECTION_POOL_DEFAULT_NAME);
    dataSource.setLeakDetectionThreshold(this.dataSourceConfig.getLeakDetectionThreshold() * 1000L);
    dataSource.setDataSourceProperties(setDataSourceProperties());
    return dataSource;
  }

  @Bean(name = DataSourceConfig.DATASOURCE_ALT_NAME, destroyMethod = "close")
  public DataSource slowDataSource()
  {
    final HikariDataSource dataSource = initializeDataSource(this.dataSourceConfig.getWorkingQueueSize());
    dataSource.setMinimumIdle(1);
    dataSource.setPoolName(DataSourceConfig.CONNECTION_POOL_ALT_NAME);
    dataSource.setLeakDetectionThreshold(this.dataSourceConfig.getSlowJobLeakDetectionThreshold() * 1000L);
    dataSource.setDataSourceProperties(setDataSourceProperties());
    return dataSource;
  }
}

有没有办法配置spring以再次允许克隆的数据源？

使用的版本：

HikariCP：2.6.1
Spring-boot：1.5.3.RELEASE
Java：1.8.121

- 修改

1.5.2.RELEASE和1.5.3.RELEASE

都失败了

- 修改2

我尝试将第二个数据源绑定到不同的数据库，但是出现同样的问题，辅助数据源池无法启动。

spring-boot 1.5.2 + .RELEASE：使用多个具有相同连接字符串

0 个答案: