首页 > 编程语言 > Java @Async注解导致spring启动失败解决方案详解
2022
09-20

Java @Async注解导致spring启动失败解决方案详解

前言

这篇文章里,最后总结处,我说了会讲讲循环依赖中,其中一个类添加@Async有可能会导致注入失败而抛异常的情况,今天就分析一下。

一、异常表现,抛出内容

1.1循环依赖的两个class

1.CycleService1

@Service
public class CycleService1 {

	@Autowired
	private CycleService2 cycleService2;

	@WangAnno
	@Async
	public void doThings() {
		System.out.println("it's a async move");
	}

}

2.CycleService2

@Service
public class CycleService2 {

	private CycleService1 cycleService1;

	public void init() {

	}

	@WangAnno
	public void alsoDo() {
		System.out.println("create cycleService2");
	}

}

1.2 启动报错

Bean with name ‘cycleService1' has been injected into other beans [cycleService2] in its raw version as part of a circular reference, but has eventually been wrapped. This means that said other beans do not use the final version of the bean.

警告: Exception encountered during context initialization - cancelling refresh attempt: org.springframework.beans.factory.BeanCurrentlyInCreationException: Error creating bean with name 'cycleService1': Bean with name 'cycleService1' has been injected into other beans [cycleService2] in its raw version as part of a circular reference, but has eventually been wrapped. This means that said other beans do not use the final version of the bean. This is often the result of over-eager type matching - consider using 'getBeanNamesOfType' with the 'allowEagerInit' flag turned off, for example.

Exception in thread "main" org.springframework.beans.factory.BeanCurrentlyInCreationException: Error creating bean with name 'cycleService1': Bean with name 'cycleService1' has been injected into other beans [cycleService2] in its raw version as part of a circular reference, but has eventually been wrapped. This means that said other beans do not use the final version of the bean. This is often the result of over-eager type matching - consider using 'getBeanNamesOfType' with the 'allowEagerInit' flag turned off, for example.

at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.doCreateBean(AbstractAutowireCapableBeanFactory.java:654)

at org.springframework.beans.factory.support.AbstractAutowireCapableBeanFactory.createBean(AbstractAutowireCapableBeanFactory.java:523)

at org.springframework.beans.factory.support.AbstractBeanFactory.lambda$doGetBean$0(AbstractBeanFactory.java:323)

at org.springframework.beans.factory.support.DefaultSingletonBeanRegistry.getSingleton(DefaultSingletonBeanRegistry.java:226)

at org.springframework.beans.factory.support.AbstractBeanFactory.doGetBean(AbstractBeanFactory.java:320)

at org.springframework.beans.factory.support.AbstractBeanFactory.getBean(AbstractBeanFactory.java:199)

at org.springframework.beans.factory.support.DefaultListableBeanFactory.preInstantiateSingletons(DefaultListableBeanFactory.java:851)

at org.springframework.context.support.AbstractApplicationContext.finishBeanFactoryInitialization(AbstractApplicationContext.java:884)

at org.springframework.context.support.AbstractApplicationContext.refresh(AbstractApplicationContext.java:552)

at com.wang.Test.main(Test.java:109)

二、原因分析

2.1 主要原因

想想我在这篇博客里的图解步骤:

  1. 当Spring在进行bean的实例化的时候,由于CycleService1和CycleService2是循环依赖的,
  2. 同时,由于CycleService1创建早于CycleService2。
  3. 所以,在CycleService1对CycleService2的initializeBean方法执行之后得到了exposedObject,要从二级缓存里获取CycleService1的earlySingletonReference不为null,就需要比较exposedObject和raw CycleService是否还是同一个对象,如果不再是同一个对象,那么就会报错。
  4. 为什么有这个逻辑呢?
  5. 其实是因为如果能从二级缓存里拿出的earlySingletonReference不为null,说明了在该对象再创建过程中被其他对象循环依赖了,且调用了三级工厂中该对象的ObjectFactory方法,基于raw bean生成了对象放入到了二级缓存。但是当raw bean执行完initializeBean之后生成了新的对象,那就出问题了。如下图:

在这里插入图片描述

也就是说基于raw bean,得到了两个基于该raw bean生成的proxy对象,Spring容器不知道最终该在容器里保存哪一个了。

2.2 循环依赖放入二级缓存处逻辑

1.每个bean在进行属性注入之前,默认都会往Spring容器中放入一个ObjectFactory进入三级工厂,以便自己在属性注入的时候被循环依赖时调用生成对象

if (earlySingletonExposure) {
	// 返回一个进行了aop处理的ObjectFactory,提前暴露
	// 但是只有当该实例在创建过程中还被其他实例引用(循环依赖),才会被调用getEarlyBeanReference
	// 此处是第四次调用beanPostProcessor,不一定会调用,只有当该类真的在创建过程中被其他类当做属性所依赖
	addSingletonFactory(beanName, () -> getEarlyBeanReference(beanName, mbd, bean));
}

2.所以在创建CycleService1过程中,CycleService2去注入CycleService2之前在三级工厂里放入了自己的ObjectFactory对象,然后在CycleService2创建过程中,要注入CycleService1的时候,就会调用Spring容器中的getEarlyBeanReference(beanName, mbd, bean)获取CycleService1,下面我们来看看该方法调用的具体逻辑

protected Object getEarlyBeanReference(String beanName, RootBeanDefinition mbd, Object bean) {
		Object exposedObject = bean;
		if (!mbd.isSynthetic() && hasInstantiationAwareBeanPostProcessors()) {
			for (BeanPostProcessor bp : getBeanPostProcessors()) {
				if (bp instanceof SmartInstantiationAwareBeanPostProcessor) {
					SmartInstantiationAwareBeanPostProcessor ibp = (SmartInstantiationAwareBeanPostProcessor) bp;
					exposedObject = ibp.getEarlyBeanReference(exposedObject, beanName);
				}
			}
		}
		return exposedObject;
	}

在这里插入图片描述

然后咱们debug发现,只有AbstractAutoProxyCreator#getEarlyBeanReference方法,有具体实现逻辑

public Object getEarlyBeanReference(Object bean, String beanName) {
		Object cacheKey = getCacheKey(bean.getClass(), beanName);
		this.earlyProxyReferences.put(cacheKey, bean);
		return wrapIfNecessary(bean, beanName, cacheKey);
	}

具体的细节,我们就不进入的,主要就是通过调用wrapIfNecessary生成了raw bean的aop proxy bean,后面放入了二级缓存。

2.3 initializeBean生成的对象

在initializeBean方法里会调用applyBeanPostProcessorsAfterInitialization方法

@Override
	public Object applyBeanPostProcessorsAfterInitialization(Object existingBean, String beanName)
			throws BeansException {

		Object result = existingBean;
		for (BeanPostProcessor processor : getBeanPostProcessors()) {
			Object current = processor.postProcessAfterInitialization(result, beanName);
			if (current == null) {
				return result;
			}
			result = current;
		}
		return result;
	}

在循环里面会调用postProcessAfterInitialization方法
重点关注AbstractAutoProxyCreator的该方法实现:

@Override
	public Object postProcessAfterInitialization(@Nullable Object bean, String beanName) {
		if (bean != null) {
			Object cacheKey = getCacheKey(bean.getClass(), beanName);
			if (this.earlyProxyReferences.remove(cacheKey) != bean) {
				// 对bean进行proxy操作
				return wrapIfNecessary(bean, beanName, cacheKey);
			}
		}
		return bean;
	}

会发现AbstractAutoProxyCreator#postProcessAfterInitialization里面的具体逻辑就是判断这个类有没有调用过wrapIfNecessary,如果调用过就不再调用,就是保证同一个raw bean不会被多次proxy,同时提前暴露注入到其他对象里的就是proxy bean。
但是由于该bean(CycleService1)上加了@Async注解,此次也会触发AsyncAnnotationBeanPostProcessor#postProcessAfterInitialization,而这个方法,我们在这篇文章里讲过了,正是@Async注解能生效的关键逻辑。所以此处生成了一个具有Async功能的新的async proxy bean

2.4 再次分析原因

基于2.3和2.4,我们基于raw bean得到了二级缓存里的aop proxy bean和async proxy bean。
让我们再回忆一下判断逻辑:

//此处是从二级缓存里面根据beanName拿出对象,因为二级缓存里放入的是因为循环依赖给其他bean注入的代理对象
	Object earlySingletonReference = getSingleton(beanName, false);
	if (earlySingletonReference != null) {
		if (exposedObject == bean) {
			exposedObject = earlySingletonReference;
		}
		// 我们之前早期暴露出去的Bean跟现在最后要放到容器中的Bean不是同一个
		// allowRawInjectionDespiteWrapping为false
		// 并且当前Bean被当成依赖注入到了别的Bean中
		else if (!this.allowRawInjectionDespiteWrapping && hasDependentBean(beanName)) {
			// 获取到当前Bean依赖的Bean
			String[] dependentBeans = getDependentBeans(beanName);
			// 要得到真实的依赖的Bean
			Set<String> actualDependentBeans = new LinkedHashSet<>(dependentBeans.length);
			for (String dependentBean : dependentBeans) {
				// 移除那些仅仅为了类型检查而创建出来
				if (!removeSingletonIfCreatedForTypeCheckOnly(dependentBean)) {
					actualDependentBeans.add(dependentBean);
				}
			}
			if (!actualDependentBeans.isEmpty()) {
				throw new BeanCurrentlyInCreationException(beanName,
						"Bean with name '" + beanName + "' has been injected into other beans [" +
						StringUtils.collectionToCommaDelimitedString(actualDependentBeans) +
						"] in its raw version as part of a circular reference, but has eventually been " +
						"wrapped. This means that said other beans do not use the final version of the " +
						"bean. This is often the result of over-eager type matching - consider using " +
						"'getBeanNamesOfType' with the 'allowEagerInit' flag turned off, for example.");
			}
		}
	}

简而言之,也就是此时从二级缓存里拿到了aop proxy bean,同时了执行完initializeBean之后,raw bean变为了async proxybean,Spring容器基于raw bean得到了两个proxy bean,无法处理了。所以在使用@Async注解时,尽量不要在被循环依赖的Class上添加

解决方案

打破循环依赖

目前我能想到的方法就是打破循环依赖,因为循环依赖发生在bean生命周期的–属性注入阶段,所以我们需要做的就是打破这种循环依赖

1.延迟注入(使用@Lazy注解)

@Service
public class CycleService1 {

	@Lazy
	@Autowired
	private CycleService2 cycleService2;

	@WangAnno
	@Async
	public void doThings() {
		cycleService2.alsoDo();
		System.out.println("it's a async move");
	}

}

看过这篇文章的都知道原理了,此处不再累赘

2. 手动延迟注入(使用applicationContext.getBean)

@Service
public class CycleService1 {

	@Autowired
	private ApplicationContext applicationContext;

	private CycleService2 cycleService2;

	@WangAnno
	@Async
	public void doThings() {
		if (Objects.isNull(cycleService2)) {
			cycleService2 = applicationContext.getBean(CycleService2.class);
		}
		cycleService2.alsoDo();
		System.out.println("it's a async move");
	}

}

其实效果是上面加了@Lazy效果是一样的,不过是我们自己在方法执行的过程中手动进行延迟注入而已。

总结

在这里插入图片描述

从二级缓存里拿到earlySingletonReference(aop proxy bean),同时了执行完initializeBean之后,raw bean变为了exposedObject(async proxy bean),Spring容器基于raw bean得到了两个proxy bean,无法处理了。
所以在使用@Async注解时,尽量不要在被循环依赖的Class上添加
实在非要添加,可以看看我给出的解决方法。

到此这篇关于Java @Async注解导致spring启动失败解决方案详解的文章就介绍到这了,更多相关Java @Async注解导致spring启动失败解决方案内容请搜索自学编程网以前的文章或继续浏览下面的相关文章希望大家以后多多支持自学编程网!

编程技巧