wangliujing commented 3 years ago

当前使用版本(必填,否则不予处理)

3.2.0

该问题是如何引起的？(确定最新版也有问题再提!!!)

业务库与quartz库是分开的，项目需要采用多数据源的方式开发，则集成了mybatispluse提供的多数据源功能，后发现切换数据源的过程频繁出错

重现步骤(如果有就写完整)

quartz配置类如下 @SpringBootConfiguration public class CustomQuartzAutoConfiguration {

/**
 * 设置schedulerFactory初始化属性
 * @author wangLiuJing
 * Created on 2020/11/30
 *
 * @param dataSource of type DynamicRoutingDataSource
 * @param dataSourceTransactionManager of type DataSourceTransactionManager
 * @return SchedulerFactoryBeanCustomizer
 */
@Order(1)
@Bean
public SchedulerFactoryBeanCustomizer schedulerFactoryBeanCustomizer(DynamicRoutingDataSource dataSource,
                                                                     DataSourceTransactionManager dataSourceTransactionManager) {
    return schedulerFactoryBean -> {
        schedulerFactoryBean.setDataSource(dataSource);
        schedulerFactoryBean.setTransactionManager(dataSourceTransactionManager);
        // 延时5秒启动
        schedulerFactoryBean.setStartupDelay(5);
        schedulerFactoryBean.setAutoStartup(true);
        schedulerFactoryBean.setOverwriteExistingJobs(true);
        // 设置全局的定时任务日志处理
        schedulerFactoryBean.setGlobalJobListeners(new QuartzJobLogListener());
    };
}

/**
 * 通过SchedulerFactoryBean获取Scheduler的实例
 */
@Bean
public Scheduler scheduler(SchedulerFactoryBean schedulerFactoryBean) {
    return schedulerFactoryBean.getScheduler();
}

}

相关核心配置如下 spring: datasource: type: com.baomidou.dynamic.datasource.DynamicRoutingDataSource dynamic: primary: main datasource: main: driverClassName: com.mysql.cj.jdbc.Driver url: jdbc:mysql://top.intellzens.com.cn:3306/honghu_datacenter_quartz?characterEncoding=utf8&connectTimeout=1000&socketTimeout=300000&autoReconnect=true&serverTimezone=Asia/Shanghai username: root password: IntellZens1104 kettle: driverClassName: com.mysql.cj.jdbc.Driver url: jdbc:mysql://top.intellzens.com.cn:3306/honghu_datacenter_kettle?characterEncoding=utf8&connectTimeout=1000&socketTimeout=300000&autoReconnect=true&serverTimezone=Asia/Shanghai username: root password: IntellZens1104

quartz: job-store-type: jdbc properties: org: quartz: threadPool: class: org.quartz.simpl.SimpleThreadPool threadCount: 100 threadPriority: 5 threadsInheritContextClassLoaderOfInitializingThread: true jobStore: class: org.quartz.impl.jdbcjobstore.JobStoreTX driverDelegateClass: org.quartz.impl.jdbcjobstore.StdJDBCDelegate dataSource: quartzDataSource tablePrefix: qrtz_ isClustered: true clusterCheckinInterval: 2000 misfireThreshold: 3000 useProperties: false scheduler: instanceName: clusteredScheduler instanceId: AUTO

看如下日志，分析问题出现原因：

2020-12-01 15:56:41.004 ERROR 12676 --- [dulerExecutor-2] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【257】, 绑定数据源【kettle】, 线程【Thread[TaskSchedulerExecutor-2,5,main]】, 获取ArrayDeque【[kettle]】,ArrayDeque【824736048】 2020-12-01 15:56:41.098 ERROR 12676 --- [dulerExecutor-2] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【257】, 获取ArrayDeque【[kettle]】,ArrayDeque【824736048】 2020-12-01 15:56:41.098 ERROR 12676 --- [dulerExecutor-1] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【256】, 获取ArrayDeque【[kettle]】,ArrayDeque【824736048】 2020-12-01 15:56:41.098 ERROR 12676 --- [dulerExecutor-2] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【257】, 获取数据源【kettle】, 线程【Thread[TaskSchedulerExecutor-2,5,main]】 2020-12-01 15:56:41.098 ERROR 12676 --- [dulerExecutor-1] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【256】, 获取数据源【kettle】, 线程【Thread[TaskSchedulerExecutor-1,5,main]】 2020-12-01 15:56:41.098 ERROR 12676 --- [dulerExecutor-2] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【257】, 获取ArrayDeque【[kettle]】,ArrayDeque【824736048】 2020-12-01 15:56:41.098 ERROR 12676 --- [dulerExecutor-1] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【256】, 获取ArrayDeque【[kettle]】,ArrayDeque【824736048】 2020-12-01 15:56:41.098 ERROR 12676 --- [dulerExecutor-2] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【257】, 获取数据源【kettle】, 线程【Thread[TaskSchedulerExecutor-2,5,main]】 2020-12-01 15:56:41.098 ERROR 12676 --- [dulerExecutor-1] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【256】, 获取数据源【kettle】, 线程【Thread[TaskSchedulerExecutor-1,5,main]】

线程256没有绑定数据源，但是获取到数据源key=kettle，很明显看第一行日志数据源kettle首先是有线程257绑定，那257绑定的key为什么被256获取到了，再看栈容器ArrayDeque的hash值可以发现257跟256绑定了同一个栈容器，我们再看如下，前面的日志分析ArrayDeque的容器是被谁创建的

2020-12-01 15:56:37.056 ERROR 12676 --- [ main] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【1】, 创建ArrayDeque【[]】,ArrayDeque【824736048】==================== 2020-12-01 15:56:37.056 ERROR 12676 --- [ main] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【1】, 获取ArrayDeque【[]】,ArrayDeque【824736048】 2020-12-01 15:56:37.057 ERROR 12676 --- [ main] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【1】, 获取数据源【null】, 线程【Thread[main,5,main]】 2020-12-01 15:56:37.261 INFO 12676 --- [ main] org.quartz.impl.StdSchedulerFactory : Using default implementation for ThreadExecutor 2020-12-01 15:56:37.265 INFO 12676 --- [ main] org.quartz.simpl.SimpleThreadPool : Job execution threads will use class loader of thread: main 2020-12-01 15:56:37.294 INFO 12676 --- [ main] org.quartz.core.SchedulerSignalerImpl : Initialized Scheduler Signaller of type: class org.quartz.core.SchedulerSignalerImpl 2020-12-01 15:56:37.295 INFO 12676 --- [ main] org.quartz.core.QuartzScheduler : Quartz Scheduler v.2.3.1 created. 2020-12-01 15:56:37.303 ERROR 12676 --- [ main] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【1】, 获取ArrayDeque【[]】,ArrayDeque【824736048】 2020-12-01 15:56:37.303 ERROR 12676 --- [ main] c.b.d.d.t.DynamicDataSourceContextHolder : 线程ID【1】, 获取数据源【null】, 线程【Thread[main,5,main]】

上面日志可以看出ArrayDeque是被线程1创建，这里可以发现后面几步中ThreadLocal执行get的时候并未重新初始化ArrayDeque

修改相关代码，问题解决： `/**

Copyright © 2018 organization baomidou
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

*/
package com.baomidou.dynamic.datasource.toolkit;

import java.util.ArrayDeque; import java.util.ArrayList; import java.util.Deque; import java.util.List;

import lombok.extern.slf4j.Slf4j; import org.springframework.core.NamedInheritableThreadLocal; import org.springframework.util.StringUtils;

/**

核心基于ThreadLocal的切换数据源工具类
@author TaoYu Kanyuxia
@since 1.0.0 */ @Slf4j public final class DynamicDataSourceContextHolder {

/**
- 为什么要用链表存储(准确的是栈)
- 为了支持嵌套切换，如ABC三个service都是不同的数据源
- 其中A的某个业务要调B的方法，B的方法需要调用C的方法。一级一级调用切换，形成了链。
- 传统的只设置当前线程的方式不能满足此业务需求，必须模拟栈，后进先出。
- */ @SuppressWarnings("unchecked") private static final ThreadLocal<Deque> LOOKUP_KEY_HOLDER = new NamedInheritableThreadLocal("dynamic-datasource");
private static List list = new ArrayList();

private DynamicDataSourceContextHolder() { }

/**
- 获得当前线程数据源
- @return 数据源名称 */ public static String peek() { list.add(Thread.currentThread()); Deque deque = LOOKUP_KEY_HOLDER.get(); if(deque == null){ return null; } log.error("线程ID【{}】, 获取ArrayDeque【{}】,ArrayDeque【{}】", Thread.currentThread().getId(), deque.toString(), deque.hashCode()); log.error("线程ID【{}】, 获取数据源【{}】, 线程【{}】", Thread.currentThread().getId(), LOOKUP_KEY_HOLDER.get().peek(), Thread.currentThread().toString());
 
 return deque.peek(); }
/**
- 设置当前线程数据源
- 如非必要不要手动调用，调用后确保最终清除
- @param ds 数据源名称 */ public static void push(String ds) { Deque deque = LOOKUP_KEY_HOLDER.get(); if(deque == null){ deque = new ArrayDeque<>(); } deque.push(StringUtils.isEmpty(ds) ? "" : ds); LOOKUP_KEY_HOLDER.set(deque); log.error("线程ID【{}】, 绑定数据源【{}】, 线程【{}】, 获取ArrayDeque【{}】,ArrayDeque【{}】", Thread.currentThread().getId(), ds, Thread.currentThread().toString(), deque.toString(), deque.hashCode()); }
/**
- 清空当前线程数据源
- 如果当前线程是连续切换数据源只会移除掉当前线程的数据源名称
- */ public static void poll() { log.error("线程ID【{}】, 清空数据源【{}】", Thread.currentThread().getId(), LOOKUP_KEY_HOLDER.get().peek()); Deque deque = LOOKUP_KEY_HOLDER.get(); if(deque != null){ deque.poll(); if (deque.isEmpty()) { LOOKUP_KEY_HOLDER.remove(); } }
}

/**
- 强制清空本地线程
- 防止内存泄漏，如手动调用了push可调用此方法确保清除
- */ public static void clear() { log.error("线程ID【{}】, 强制清空数据源【{}】", Thread.currentThread().getId(), LOOKUP_KEY_HOLDER.get().peek()); LOOKUP_KEY_HOLDER.remove(); } } `

报错信息

Caused by: java.sql.SQLSyntaxErrorException: Table 'ckettle.dc_kettle_task' doesn't exist at com.mysql.cj.jdbc.exceptions.SQLError.createSQLException(SQLError.java:120) ~[mysql-connector-java-8.0.17.jar:8.0.17] at com.mysql.cj.jdbc.exceptions.SQLError.createSQLException(SQLError.java:97) ~[mysql-connector-java-8.0.17.jar:8.0.17] at com.mysql.cj.jdbc.exceptions.SQLExceptionsMapping.translateException(SQLExceptionsMapping.java:122) ~[mysql-connector-java-8.0.17.jar:8.0.17] at com.mysql.cj.jdbc.ClientPreparedStatement.executeInternal(ClientPreparedStatement.java:953) ~[mysql-connector-java-8.0.17.jar:8.0.17] at com.mysql.cj.jdbc.ClientPreparedStatement.execute(ClientPreparedStatement.java:370) ~[mysql-connector-java-8.0.17.jar:8.0.17] at com.alibaba.druid.filter.FilterChainImpl.preparedStatement_execute(FilterChainImpl.java:3461) ~[druid-1.2.3.jar:1.2.3] at com.alibaba.druid.wall.WallFilter.preparedStatement_execute(WallFilter.java:660) ~[druid-1.2.3.jar:1.2.3] at com.alibaba.druid.filter.FilterChainImpl.preparedStatement_execute(FilterChainImpl.java:3459) ~[druid-1.2.3.jar:1.2.3] at com.alibaba.druid.filter.FilterEventAdapter.preparedStatement_execute(FilterEventAdapter.java:440) ~[druid-1.2.3.jar:1.2.3] at com.alibaba.druid.filter.FilterChainImpl.preparedStatement_execute(FilterChainImpl.java:3459) ~[druid-1.2.3.jar:1.2.3] at com.alibaba.druid.proxy.jdbc.PreparedStatementProxyImpl.execute(PreparedStatementProxyImpl.java:167) ~[druid-1.2.3.jar:1.2.3] at com.alibaba.druid.pool.DruidPooledPreparedStatement.execute(DruidPooledPreparedStatement.java:497) ~[druid-1.2.3.jar:1.2.3] at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ~[na:1.8.0_171] at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) ~[na:1.8.0_171] at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) ~[na:1.8.0_171] at java.lang.reflect.Method.invoke(Method.java:498) ~[na:1.8.0_171] at org.apache.ibatis.logging.jdbc.PreparedStatementLogger.invoke(PreparedStatementLogger.java:59) ~[mybatis-3.5.2.jar:3.5.2] at com.sun.proxy.$Proxy241.execute(Unknown Source) ~[na:na] at org.apache.ibatis.executor.statement.PreparedStatementHandler.query(PreparedStatementHandler.java:64) ~[mybatis-3.5.2.jar:3.5.2] at org.apache.ibatis.executor.statement.RoutingStatementHandler.query(RoutingStatementHandler.java:79) ~[mybatis-3.5.2.jar:3.5.2] at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ~[na:1.8.0_171] at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) ~[na:1.8.0_171] at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) ~[na:1.8.0_171] at java.lang.reflect.Method.invoke(Method.java:498) ~[na:1.8.0_171] at org.apache.ibatis.plugin.Plugin.invoke(Plugin.java:63) ~[mybatis-3.5.2.jar:3.5.2] at com.sun.proxy.$Proxy239.query(Unknown Source) ~[na:na] at com.baomidou.mybatisplus.core.executor.MybatisSimpleExecutor.doQuery(MybatisSimpleExecutor.java:67) ~[mybatis-plus-core-3.2.0.jar:3.2.0] at org.apache.ibatis.executor.BaseExecutor.queryFromDatabase(BaseExecutor.java:324) ~[mybatis-3.5.2.jar:3.5.2] at org.apache.ibatis.executor.BaseExecutor.query(BaseExecutor.java:156) ~[mybatis-3.5.2.jar:3.5.2] at org.apache.ibatis.executor.CachingExecutor.query(CachingExecutor.java:109) ~[mybatis-3.5.2.jar:3.5.2] at org.apache.ibatis.executor.CachingExecutor.query(CachingExecutor.java:83) ~[mybatis-3.5.2.jar:3.5.2] at org.apache.ibatis.session.defaults.DefaultSqlSession.selectList(DefaultSqlSession.java:147) ~[mybatis-3.5.2.jar:3.5.2] at org.apache.ibatis.session.defaults.DefaultSqlSession.selectList(DefaultSqlSession.java:140) ~[mybatis-3.5.2.jar:3.5.2] at org.apache.ibatis.session.defaults.DefaultSqlSession.selectOne(DefaultSqlSession.java:76) ~[mybatis-3.5.2.jar:3.5.2] at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ~[na:1.8.0_171] at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) ~[na:1.8.0_171] at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) ~[na:1.8.0_171] at java.lang.reflect.Method.invoke(Method.java:498) ~[na:1.8.0_171] at org.mybatis.spring.SqlSessionTemplate$SqlSessionInterceptor.invoke(SqlSessionTemplate.java:426) ~[mybatis-spring-2.0.2.jar:2.0.2] ... 30 common frames omitted

huayanYu commented 3 years ago

@wangliujing 动态数据源的问题请在那边提，我很少会来看这边的ISSUE。~ 请查看最新文档和示例项目，如仍有问题。建议fork示例项目，给出复现步骤在那边提ISSUE。

wangliujing commented 3 years ago

/**

Copyright © 2018 organization baomidou
<pre>
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
<pre/> */ package com.baomidou.dynamic.datasource.toolkit;

import lombok.extern.slf4j.Slf4j; import org.springframework.core.NamedThreadLocal; import org.springframework.util.StringUtils;

import java.util.ArrayDeque; import java.util.Deque;

/**

核心基于ThreadLocal的切换数据源工具类
@author TaoYu Kanyuxia
@since 1.0.0 */ @Slf4j public final class DynamicDataSourceContextHolder {

/**
- 为什么要用链表存储(准确的是栈)
- <pre>
- 为了支持嵌套切换，如ABC三个service都是不同的数据源
- 其中A的某个业务要调B的方法，B的方法需要调用C的方法。一级一级调用切换，形成了链。
- 传统的只设置当前线程的方式不能满足此业务需求，必须使用栈，后进先出。
- </pre> */ private static final ThreadLocal<Deque<String>> LOOKUP_KEY_HOLDER = new NamedThreadLocal<Deque<String>>("dynamic-datasource");
private DynamicDataSourceContextHolder() { }

/**
- 获得当前线程数据源
- @return 数据源名称 */ public static String peek() { Deque<String> deque = LOOKUP_KEY_HOLDER.get(); if(deque == null){ log.debug("当前线程【{}】获取数据源标识【null】", Thread.currentThread().hashCode()); return null; } log.debug("当前线程【{}】获取数据源标识【{}】", Thread.currentThread().hashCode(), deque.peek()); return deque.peek(); }
/**
- 设置当前线程数据源
- 
- 如非必要不要手动调用，调用后确保最终清除
- 
- @param ds 数据源名称 */ public static void push(String ds) { log.debug("当前线程【{}】设置数据源标识【{}】", Thread.currentThread().hashCode(), ds); Deque<String> deque = LOOKUP_KEY_HOLDER.get(); if(deque == null){ deque = new ArrayDeque<>(); LOOKUP_KEY_HOLDER.set(deque); } deque.push(StringUtils.isEmpty(ds) ? "" : ds); }
/**
- 清空当前线程数据源
- 
- 如果当前线程是连续切换数据源只会移除掉当前线程的数据源名称
- */ public static void poll() { Deque<String> deque = LOOKUP_KEY_HOLDER.get(); if(deque != null){ String ds = deque.poll(); log.debug("当前线程【{}】清空数据源标识【{}】", Thread.currentThread().hashCode(), ds); if (deque.isEmpty()) { LOOKUP_KEY_HOLDER.remove(); } }
}

/**
- 强制清空本地线程
- 
- 防止内存泄漏，如手动调用了push可调用此方法确保清除
- */ public static void clear() { log.debug("当前线程【{}】强制清空", Thread.currentThread().hashCode()); LOOKUP_KEY_HOLDER.remove(); } }

这块又出现bug了，我已经确定，如果并发情况下去切换数据源很容易出现我说的问题，把源码修改成上面这种就好了，我也不知道为什么，看源码感觉跟这种是一样的，只不过变换了一种写法，但是测试的确我这种写法不会有问题，源码的写法会导致数据库切换混乱，比如本来是应该获取A数据源，但是实际获取的是B，最新的我看过了，还是有问题 ------------------ 原始邮件 ------------------ 发件人: "baomidou/mybatis-plus" <notifications@github.com>; 发送时间: 2021年1月15日(星期五) 下午3:21 收件人: "baomidou/mybatis-plus"<mybatis-plus@noreply.github.com>; 抄送: "王柳敬"<273833883@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [baomidou/mybatis-plus] 多数据源集成spring-boot-starter-quartz导致数据源切换错误 (#3120)

@wangliujing 动态数据源的问题请在那边提，我很少会来看这边的ISSUE。~ 请查看最新文档和示例项目，如仍有问题。建议fork示例项目，给出复现步骤在那边提ISSUE。

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

wangliujing commented 3 years ago

------------------ 原始邮件 ------------------ 发件人: "王柳敬" <273833883@qq.com>; 发送时间: 2021年2月1日(星期一) 下午4:49 收件人: "baomidou/mybatis-plus"<reply@reply.github.com>;

主题: 回复： [baomidou/mybatis-plus] 多数据源集成spring-boot-starter-quartz导致数据源切换错误 (#3120)

/**

<pre>
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
<pre/> */ package com.baomidou.dynamic.datasource.toolkit;

import lombok.extern.slf4j.Slf4j; import org.springframework.core.NamedThreadLocal; import org.springframework.util.StringUtils;

import java.util.ArrayDeque; import java.util.Deque;

/**

核心基于ThreadLocal的切换数据源工具类
@author TaoYu Kanyuxia
@since 1.0.0 */ @Slf4j public final class DynamicDataSourceContextHolder {

/**
- 为什么要用链表存储(准确的是栈)
- <pre>
- 为了支持嵌套切换，如ABC三个service都是不同的数据源
- 其中A的某个业务要调B的方法，B的方法需要调用C的方法。一级一级调用切换，形成了链。
- 传统的只设置当前线程的方式不能满足此业务需求，必须使用栈，后进先出。
- </pre> */ private static final ThreadLocal<Deque<String>> LOOKUP_KEY_HOLDER = new NamedThreadLocal<Deque<String>>("dynamic-datasource");
private DynamicDataSourceContextHolder() { }

/**
- 获得当前线程数据源
- @return 数据源名称 */ public static String peek() { Deque<String> deque = LOOKUP_KEY_HOLDER.get(); if(deque == null){ log.debug("当前线程【{}】获取数据源标识【null】", Thread.currentThread().hashCode()); return null; } log.debug("当前线程【{}】获取数据源标识【{}】", Thread.currentThread().hashCode(), deque.peek()); return deque.peek(); }
/**
- 设置当前线程数据源
- 
- 如非必要不要手动调用，调用后确保最终清除
- 
- @param ds 数据源名称 */ public static void push(String ds) { log.debug("当前线程【{}】设置数据源标识【{}】", Thread.currentThread().hashCode(), ds); Deque<String> deque = LOOKUP_KEY_HOLDER.get(); if(deque == null){ deque = new ArrayDeque<>(); LOOKUP_KEY_HOLDER.set(deque); } deque.push(StringUtils.isEmpty(ds) ? "" : ds); }
/**
- 清空当前线程数据源
- 
- 如果当前线程是连续切换数据源只会移除掉当前线程的数据源名称
- */ public static void poll() { Deque<String> deque = LOOKUP_KEY_HOLDER.get(); if(deque != null){ String ds = deque.poll(); log.debug("当前线程【{}】清空数据源标识【{}】", Thread.currentThread().hashCode(), ds); if (deque.isEmpty()) { LOOKUP_KEY_HOLDER.remove(); } }
}

/**
- 强制清空本地线程
- 
- 防止内存泄漏，如手动调用了push可调用此方法确保清除
- */ public static void clear() { log.debug("当前线程【{}】强制清空", Thread.currentThread().hashCode()); LOOKUP_KEY_HOLDER.remove(); } }

这块又出现bug了，我已经确定，如果并发情况下去切换数据源很容易出现我说的问题，把源码修改成上面这种就好了，我也不知道为什么，看源码感觉跟这种是一样的，只不过变换了一种写法，但是测试的确我这种写法不会有问题，源码的写法会导致数据库切换混乱，比如本来是应该获取A数据源，但是实际获取的是B，最新的我看过了，还是有问题 ------------------ 原始邮件 ------------------ 发件人: "baomidou/mybatis-plus" <notifications@github.com>; 发送时间: 2021年1月15日(星期五) 下午3:21 收件人: "baomidou/mybatis-plus"<mybatis-plus@noreply.github.com>; 抄送: "王柳敬"<273833883@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [baomidou/mybatis-plus] 多数据源集成spring-boot-starter-quartz导致数据源切换错误 (#3120)

@wangliujing 动态数据源的问题请在那边提，我很少会来看这边的ISSUE。~ 请查看最新文档和示例项目，如仍有问题。建议fork示例项目，给出复现步骤在那边提ISSUE。

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

luohaoGit commented 7 months ago

我也遇到类似问题

baomidou / mybatis-plus

多数据源集成spring-boot-starter-quartz导致数据源切换错误 #3120

当前使用版本(必填,否则不予处理)

该问题是如何引起的？(确定最新版也有问题再提!!!)

重现步骤(如果有就写完整)

报错信息