Spring全局Executor与内置一些功能的Executor冲突问题

1、问题描述

自己配置的Executor（自己创建的线程池）

目录

1、问题描述

2、原因探索

3、问题解决：

（1）方法1：直接用EnableScheduling注解里头自己定义的全局线程池：

（2）方法2：自己开自己的线程池

（3）方法3：重新配置注解所需线程池

---

比方说：

@Configuration
public class ThreadPoolConfig {

    @Bean(value = "selfExecutor")
    public Executor createDefaultIOExecutor() {
        int coreCount = Runtime.getRuntime().availableProcessors() * 2;

        NamedThreadFactory namedThreadFactory = new NamedThreadFactory("statistics-IO-thread-pool");
        Executor executor = new ThreadPoolExecutor(
                coreCount
                , coreCount
                , 60
                , TimeUnit.SECONDS
                , new LinkedBlockingQueue<>(1024)
                , namedThreadFactory
                , new ThreadPoolExecutor.CallerRunsPolicy()
        );
        return executor;
    }
}

会和定时任务的Executor的冲突，如果在启动类上添加了@EnableScheduled 注解的话。

启动类：

@EnableScheduling
@EnableCaching
public class StoreStatisticsServerApplication {

    public static void main(String[] args) {
        SpringApplication.run(StoreStatisticsServerApplication.class, args);
    }

}

报错：Caused by: org.springframework.beans.factory.NoUniqueBeanDefinitionException: No qualifying bean of type 'java.util.concurrent.Executor' available: expected single matching bean but found 2: selfExecutor,taskScheduler

Caused by: org.springframework.beans.factory.NoUniqueBeanDefinitionException: No qualifying bean of type 'java.util.concurrent.Executor' available: expected single matching bean but found 2: selfExecutor,taskScheduler
	at org.springframework.beans.factory.config.DependencyDescriptor.resolveNotUnique(DependencyDescriptor.java:220)
	at org.springframework.beans.factory.support.DefaultListableBeanFactory.doResolveDependency(DefaultListableBeanFactory.java:1284)
	at org.springframework.beans.factory.support.DefaultListableBeanFactory.resolveDependency(DefaultListableBeanFactory.java:1226)
	at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.autowireResource(CommonAnnotationBeanPostProcessor.java:521)
	at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.getResource(CommonAnnotationBeanPostProcessor.java:497)
	at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor$ResourceElement.getResourceToInject(CommonAnnotationBeanPostProcessor.java:650)
	at org.springframework.beans.factory.annotation.InjectionMetadata$InjectedElement.inject(InjectionMetadata.java:239)
	at org.springframework.beans.factory.annotation.InjectionMetadata.inject(InjectionMetadata.java:130)
	at org.springframework.context.annotation.CommonAnnotationBeanPostProcessor.postProcessProperties(CommonAnnotationBeanPostProcessor.java:318)
	... 34 common frames omitted

2、原因探索

实际上就是spring容器自己管理了全局的线程池，和自己配置的冲突了。可以去看看上述spring项目自己做的定时任务的这个线程池启动的类。

org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor

还有spring自动配置里面的：

org.springframework.boot.autoconfigure.task.TaskExecutionAutoConfiguration#applicationTaskExecutor

因此自己配置了线程池的话，还需要重新配置这个threadpooltaskexecutor，否则：

（1）用重命名bean的方式注入这种方式解决，则可能因为类型不对，注入失败；

（2）修改自己配置的bean的名字为人家写死的bean的名字（“applicationTaskExecutor”），依然有类型不对问题；

3、问题解决：

（1）方法1：直接用EnableScheduling注解里头自己定义的全局线程池：

org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor;

使用案例

@Resource
private ThreadPoolTaskExecutor threadPoolTaskExecutor;

/**
* 获取按照状态区分的订单信息
* @param statParamRequest
* @return
*/
@Override
public PageableResponse<List<StatBillStatusVO>> getStatusBillInfoList(StatParamRequest statParamRequest, OrderStatTagEnum orderStatTagEnum) {
    // 业务处理代码1XXX

    // 计算单据的占比和设值
    CompletableFuture<Void> billInfo = CompletableFuture.runAsync(() -> {
        this.getBillInfo(statBillStatusVO, orderAgg, finalQuantitySum, finalAmountSum);
    }, threadPoolTaskExecutor).exceptionally(throwable -> {
        throw new IllegalStateException("计算单据的占比和设值出错，request="+ JsonUtils.toJson(statParamRequest));
    });

    // 按照状态或来源聚合
    CompletableFuture<Void> statusSourceInfo = CompletableFuture.runAsync(() -> {
        this.setStatusOrSource(orderAgg, orderStatTagEnum, statBillStatusVO);
    }, threadPoolTaskExecutor).exceptionally(throwable -> {
        throw new IllegalStateException("按照状态或来源聚合设值出错，request="+ JsonUtils.toJson(statParamRequest));
    });

    CompletableFuture.allOf(billInfo, statusSourceInfo).join();

    // 业务处理代码2XXX
}

（2）方法2：自己开自己的线程池

/**
 * @Author: Zeng Yanru
 * @Description:
 * @DateTime: 2021/7/8 11:05
 */
public class ThreadPoolInit {

    public static Executor executor = createDefaultIOExecutor();

    public static Executor createDefaultIOExecutor() {
        int coreCount = Runtime.getRuntime().availableProcessors() * 2;
        NamedThreadFactory namedThreadFactory = new NamedThreadFactory("statistics-IO-thread-pool");
        return new ThreadPoolExecutor(
                coreCount
                , coreCount
                , 60
                , TimeUnit.SECONDS
                , new LinkedBlockingQueue<>(1024)
                , namedThreadFactory
                , new ThreadPoolExecutor.CallerRunsPolicy()
        );
    }
}

然后在外部调用即可。最好做成spring容器全部能用的，否则不断创建线程池开销也很大。

（3）方法3：重新配置注解所需线程池

/**
 * @Author: Zeng Yanru
 * @Description:
 * @DateTime: 2021/7/9 9:46
 */
@Configuration
public class DefaultThreadPool {

    @Bean("aaa")
    public ListenableThreadPoolExecutor createDefaultIOExecutor() {
        int coreCount = Runtime.getRuntime().availableProcessors() * 2;
        NamedThreadFactory namedThreadFactory = new NamedThreadFactory("statistics-IO-thread-pool1");
        return new ListenableThreadPoolExecutor(
                coreCount,
                coreCount,
                60,
                TimeUnit.SECONDS,
                new LinkedBlockingQueue<>(),
                namedThreadFactory);
    }
    
    // @EnableScheduling注解所需要的线程池，配置则自定义或者按照原本的配置
    @Bean(name = "applicationTaskExecutor")
    public ThreadPoolTaskExecutor createTreadPoolTaskExecutor() {
        int coreCount = Runtime.getRuntime().availableProcessors() * 2;
        NamedThreadFactory namedThreadFactory = new NamedThreadFactory("statistics-IO-thread-pool2");
        ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
        executor.setCorePoolSize(coreCount);
        executor.setMaxPoolSize(coreCount);
        executor.setKeepAliveSeconds(60);
        // 这个executor的写死的blockingqueue就是linkedblockingqueue
        // 其他值不用设置
        return executor;
    }
}