Why RabbitMQ is triggered or running while building the spring-boot jar.While running the Application.java or pom.xml.
i am able to see the following loggers
2016-07-01 16:40:04.334 INFO 7004 --- [ main] com.rabbit.App : No active profile set, falling back to default profiles: default
2016-07-01 16:40:04.391 INFO 7004 --- [ main] ationConfigEmbeddedWebApplicationContext : Refreshing org.springframework.boot.context.embedded.AnnotationConfigEmbeddedWebApplicationContext#1da51a35: startup date [Fri Jul 01 16:40:04 CEST 2016]; root of context hierarchy
2016-07-01 16:40:05.331 INFO 7004 --- [ main] o.s.b.f.s.DefaultListableBeanFactory : Overriding bean definition for bean 'rabbitListenerContainerFactory' with a different definition: replacing [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=rabbitMqConfiguration; factoryMethodName=rabbitListenerContainerFactory; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [com/rabbit/messaging/configuration/RabbitMqConfiguration.class]] with [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=taskConsumerConfiguration; factoryMethodName=rabbitListenerContainerFactory; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [com/rabbit/messaging/configuration/TaskConsumerConfiguration.class]]
2016-07-01 16:40:05.334 INFO 7004 --- [ main] o.s.b.f.s.DefaultListableBeanFactory : Overriding bean definition for bean 'connectionFactory' with a different definition: replacing [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=rabbitMqConfiguration; factoryMethodName=connectionFactory; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [com/rabbit/messaging/configuration/RabbitMqConfiguration.class]] with [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=taskConsumerConfiguration; factoryMethodName=connectionFactory; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [com/rabbit/messaging/configuration/TaskConsumerConfiguration.class]]
2016-07-01 16:40:05.334 INFO 7004 --- [ main] o.s.b.f.s.DefaultListableBeanFactory : Overriding bean definition for bean 'jsonMessageConverter' with a different definition: replacing [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=rabbitMqConfiguration; factoryMethodName=jsonMessageConverter; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [com/rabbit/messaging/configuration/RabbitMqConfiguration.class]] with [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=taskConsumerConfiguration; factoryMethodName=jsonMessageConverter; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [com/rabbit/messaging/configuration/TaskConsumerConfiguration.class]]
2016-07-01 16:40:05.868 INFO 7004 --- [ main] trationDelegate$BeanPostProcessorChecker : Bean 'org.springframework.amqp.rabbit.annotation.RabbitBootstrapConfiguration' of type [class org.springframework.amqp.rabbit.annotation.RabbitBootstrapConfiguration$$EnhancerBySpringCGLIB$$ad9295b0] is not eligible for getting processed by all BeanPostProcessors (for example: not eligible for auto-proxying)
2016-07-01 16:40:06.657 INFO 7004 --- [ main] s.b.c.e.t.TomcatEmbeddedServletContainer : Tomcat initialized with port(s): 8080 (http)
Is there any option to disable this?
While building the jar during test phase it throws following exception.
LifecycleProcessor with name 'lifecycleProcessor': using default [org.springframework.context.support.DefaultLifecycleProcessor#76dc36e5]
2016-06-30 15:10:32.989 WARN 21614 --- [cTaskExecutor-1] o.s.a.r.l.SimpleMessageListenerContainer : Consumer raised exception, processing can restart if the connection factory supports it
org.springframework.amqp.AmqpIOException: java.net.SocketTimeoutException: connect timed out
at org.springframework.amqp.rabbit.support.RabbitExceptionTranslator.convertRabbitAccessException(RabbitExceptionTranslator.java:67)
at
You get the log entry 'Overriding bean definition..' when you define a bean with that name more than once.
This leads to the question, which of the, at least two, beans do you use and how are they configured?
Set breakpoint(s) on the constructor(s) of the rabbitListenerContainerFactory bean and see which instances you are creating.
Looking up the stack trace may give you information about why this bean is created, so you can remove the duplicate from your spring configuration.
You can also add breakpoints, where the rabbit connection properties are set.
"connect timed out" also appears, when the given ip is wrong.
Anyway I would not recommend to set the properties in the build script, instead point to the property files on the command line, when you start the application. That allows you to use your build in different environments.
See howto-properties-and-configuration
Why it is throwing while building the jar?
You wrote you are setting the properties with a script during build.
This means that the tests are using those properties and now the tests are failing, may be because the rabbit mq server is not reachable with the provided properties or is not up.
As long as you are overriding beans, you do not really know which instance you use, so you dont really know which properties you use.
Try to log it out, than I am sure you will see....
You can't "lazy load" beans that implement SmartLifecycle (such as the listener container) because the context has to load the bean to call isAutoStartup() to see whether or not the bean should be started.
If the context is being loaded while "building the jar", you must have test cases that are using the beans. You can set autoStartup to false - but your tests will probably fail then.
You can skip running the tests with -DskipTests=true (maven) or -x test (gradle).
I have the same problem in my tests. I don't use rabbit and because the timeouts occur the tests duration increases.
As a simple workaround, you can disable the configuration of queues, exchanges etc. during tests using spring profiles.
Example:
Having the configuration that is disabled if the spring profile "no_rabbit" is present.
#Configuration
#Profile("!no_rabbit")
public class RabbitMQConfig {
#Bean
public Queue queue() {
return QueueBuilder.durable("NAMW").withArgument("x-message-ttl", 3600).build();
}
// more definitions of bean, exchanges etc.
}
Then, in the test simply add the profile "no_rabbit".
#RunWith(SpringRunner.class)
#ActiveProfiles({ "no_rabbit" })
public class TestClass {
...
}
The drawbacks of this approach is that you have to add the profile in every single test that don't uses rabbit. But for my case it's ok until I have a better solution.
Related
I have configured Spring Batch, Boot (2.1.4) to retrieve the transaction manager from JBoss via JNDI and I still have conflicts with the transactionManager bean name, is it possible that every time it is always the same story but for different reasons?
Now the message is:
DEBUG [org.springframework.boot.diagnostics.LoggingFailureAnalysisReporter] (ServerService Thread Pool -- 76) Application failed to start due to an exception: org.springframework.beans.factory.support.BeanDefinitionOverrideException: Invalid bean definition with name 'transactionManager' defined in class path resource [org/springframework/batch/core/configuration/annotation/SimpleBatchConfiguration.class]: Cannot register bean definition [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=org.springframework.batch.core.configuration.annotation.SimpleBatchConfiguration; factoryMethodName=transactionManager; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [org/springframework/batch/core/configuration/annotation/SimpleBatchConfiguration.class]] for bean 'transactionManager': There is already [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=org.springframework.boot.autoconfigure.transaction.jta.JndiJtaConfiguration; factoryMethodName=transactionManager; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [org/springframework/boot/autoconfigure/transaction/jta/JndiJtaConfiguration.class]] bound.
Spring batch doesn't work with a transactional manager retrieved via JNDI?
Thanks
You need to give your transaction manager a name other than transactionManager because #EnableBatchProcessing already exposes a bean with that name (this is mentioned in the Javadoc of the annotation).
I have a spring-data-jpa repository called TagRepository. My spring-boot version is 2.1.2. I am using H2 dependency with runtime scope and I intend to use it for both the application and integration testing. I want to write a DataJpaTest for the TagRepository. I have written the following code:
#RunWith(SpringRunner.class)
#EnableAutoConfiguration
#DataJpaTest
#ContextConfiguration(classes={TagRepository.class})
public class TagRepositoryTest {
#Autowired
private TestEntityManager testEntityManager;
#Autowired
private TagRepository tagRepository;
#Test
public void findByTagTest() {
Tag tag = new Tag("java");
testEntityManager.persistAndFlush(tag);
Optional<Tag> optionalTag = tagRepository.findByTag(tag.getTag());
if(!optionalTag.isPresent()) {
fail("Tag not found hence test failed");
}
assertThat(optionalTag.get()).isEqualTo(tag);
}
}
However, when I execute the test it says Application failed to start and I get following error:
Invalid bean definition with name 'tagRepository' defined in null: Cannot register bean definition [Root bean: class [org.springframework.data.jpa.repository.support.JpaRepositoryFactoryBean]; scope=; abstract=false; lazyInit=false; autowireMode=0; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=null; factoryMethodName=null; initMethodName=null; destroyMethodName=null] for bean 'tagRepository': There is already [Generic bean: class [com.upday.task.repository.TagRepository]; scope=singleton; abstract=false; lazyInit=false; autowireMode=0; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=null; factoryMethodName=null; initMethodName=null; destroyMethodName=null] bound.
The bean 'tagRepository', defined in null, could not be registered. A bean with that name has already been defined in null and overriding is disabled.
Action:
Consider renaming one of the beans or enabling overriding by setting spring.main.allow-bean-definition-overriding=true
The test class itself resides in a different package than jpa repository.
Just remove
#ContextConfiguration(classes={TagRepository.class})
from your test.
It is meant for passing a custom #Configuration not a regular bean.
UPD:
When data jpa test and the repository under test resides in a different packages spring runner needed a little hint to scan classes from another place. Pointing out the application class via
#ContextConfiguration(classes={SpringBootApplication.class})
will extend the component scan scope and allow test environment to pick up additional beans.
This question already has answers here:
SpringBoot - BeanDefinitionOverrideException: Invalid bean definition
(8 answers)
Closed 3 years ago.
I'm having following config in my production files:
#Configuration
internal class Config {
#Bean
fun clock() = Clock.systemUTC()
}
In tests:
#Configuration
class ClockTestConfiguration {
#Bean
fun clock() = SetableClock()
}
My test annotations:
#SpringBootTest(
webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT,
classes = [
MyApplication::class,
ClockTestConfiguration::class
]
)
class MyTest {
...
When I was using Spring Boot 2.0.5.RELEASE it worked like a charm. After upgrading to 2.1.0.RELEASE it fails during bean registration.
Caused by: org.springframework.beans.factory.support.BeanDefinitionOverrideException: Invalid bean definition with name 'clock' defined in com.foo.clock.ClockTestConfiguration:
Cannot register bean definition [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=clockTestConfiguration; factoryMethodName=clock; initMethodName=null; destroyMethodName=(inferred);
defined in com.foo.clock.ClockTestConfiguration] for bean 'clock': There is already [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0; autowireCandidate=true; primary=false; factoryBeanName=config; factoryMethodName=clock; initMethodName=null; destroyMethodName=(inferred); defined in class path resource [com/foo/clock/Config.class]] bound.
at org.springframework.beans.factory.support.DefaultListableBeanFactory.registerBeanDefinition(DefaultListableBeanFactory.java:894)
Is there a clean way to override such bean?
You could use the properties attribute of #SpringBootTest to set spring.main.allow-bean-definition-overriding=true.
I have defined ConfigServicePropertySourceLocator bean like following
#Primary
#Bean
public ConfigServicePropertySourceLocator configServicePropertySource(
ConfigClientProperties configClientProperties) {
ConfigServicePropertySourceLocator sourceLocator = new ConfigServicePropertySourceLocator(
configClientProperties);
sourceLocator.setRestTemplate(clientOnlyRestTemplate());
return sourceLocator;
}
but i get the following exception (This is how it is printed in docker) although my bean is marked as #Primary
WARN 1 --- [ main] s.c.a.AnnotationConfigApplicationContext : Exception encountered during context initialization - cancelling refresh attempt:
org.springframework.beans.factory.support.BeanDefinitionOverrideException:
Invalid bean definition with name 'configServicePropertySource' defined in de.ig.client.security.configuration.ConfigClientSecurityConfiguration:
Cannot register bean definition [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3;
dependencyCheck=0; autowireCandidate=true; primary=true; factoryBeanName=configClientSecurityConfiguration;
factoryMethodName=configServicePropertySource; initMethodName=null; destroyMethodName=(inferred);
defined in de.ig.client.security.configuration.ConfigClientSecurityConfiguration] for bean 'configServicePropertySource':
There is already [Root bean: class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3; dependencyCheck=0;
autowireCandidate=true; primary=false; factoryBeanName=configServiceBootstrapConfiguration;
factoryMethodName=configServicePropertySource; initMethodName=null; destroyMethodName=(inferred);
defined in org.springframework.cloud.config.client.ConfigServiceBootstrapConfiguration] bound.
Solved it by setting
spring.main.allow-bean-definition-overriding to true
New from boot version 2.1
Originally posted this as an issue in GitHub but might be better suited for this forum...
We are assigning some custom metadata values to our Spring Cloud services that are being registered in Eureka and we now need to have visibility to those values in the Eureka dashboard. I am trying to extend/override the /info endpoint so that our metadata is visible from the Eureka Dashboard which already provides hyperlinks to each registered service's /info endpoint. I had read that I could override the Boot autoconfigured InfoEndpoint by just adding my own version of that bean to the context. I am trying to test with the following configuration bean:
#Configuration
public class EndpointConfig {
#Bean
public InfoEndpoint infoEndpoint() throws Exception {
LinkedHashMap<String, Object> info = new LinkedHashMap<String, Object>();
info.put("name", "value");
return new InfoEndpoint(info);
}
}
When I run my service though and hit its /info endpoint I do not see this test value. I also see this in the log:
2015-04-17 14:54:23,910 main INFO DefaultListableBeanFactory - - - -
Overriding bean definition for bean 'infoEndpoint': replacing [Root
bean: class [null]; scope=; abstract=false; lazyInit=false;
autowireMode=3; dependencyCheck=0; autowireCandidate=true;
primary=false; factoryBeanName=endpointConfig;
factoryMethodName=infoEndpoint; initMethodName=null;
destroyMethodName=(inferred); defined in class path resource
[com/acme/ecom/items/config/EndpointConfig.class]] with [Root bean:
class [null]; scope=; abstract=false; lazyInit=false; autowireMode=3;
dependencyCheck=0; autowireCandidate=true; primary=false;
factoryBeanName=org.springframework.cloud.autoconfigure.RefreshAutoConfiguration$InfoEndpointRebinderConfiguration;
factoryMethodName=infoEndpoint; initMethodName=null;
destroyMethodName=(inferred); defined in class path resource
[org/springframework/cloud/autoconfigure/RefreshAutoConfiguration$InfoEndpointRebinderConfiguration.class]]
It appears that my version of the infoEndpoint bean is being replaced with another bean coming from Spring cloud (in org/springframework/cloud/autoconfigure/RefreshAutoConfiguration$InfoEndpointRebinderConfiguration.class).
Am I reading this correctly? And if so how do I prevent it?
Thanks,
Bill
If you put your metadata in info.* e.g. info.myfoo=${eureka.instance.metadataMap.myfoo:none} it will show up in the default /info endpoint.
Don't know if this will actually help but I have this in my application.yml which I can see when I do /info on the service.
info:
component: Service Name
so when i hit the service with /info this is what i see:-
{
"component": "Service Name"
}
Hope this helps.