I use RabbitMQ and Spring Integration to handle incoming JSON messages.
The relevant part of the configuration looks something like this:
<amqp:inbound-channel-adapter channel="incomingChannel" queue-names="..."
message-converter="jsonConverter" error-handler="errorHandler"
error-channel="errorChannel" />
I'm using Jackson Databind as the JSON converter.
Sometimes the incoming JSON messages have an incorrect syntax. This results in the following (correct) exception:
org.springframework.amqp.rabbit.listener.ListenerExecutionFailedException: Listener threw exception
Caused by: org.springframework.amqp.support.converter.MessageConversionException: Failed to convert Message content
Caused by: com.fasterxml.jackson.databind.JsonMappingException: Can not deserialize instance of java.lang.String out of START_ARRAY token...
The exception is then handled by the errorHandler which simply is a MessagePublishingErrorHandler to errorChannel.
So far so good. The problem is, that the message is still being rejected by the AMQP client, although I am handling it with an error handler. RabbitMQ then redelivers the message over and over. Even configuring a dead letter queue did not help. Any ideas how to handle this scenario correctly?
Exceptions further down the processing (after successful deserialization) are handled just fine: AMQP message acknowledged and error message sent to errorChannel.
Any ideas?
Library versions:
Spring Integration: 3.0.1
Spring Framework: 4.0.2
Jackson Databind: 2.3.1
You can set defaultRequeueRejected=false on the listener container.
The attribute is not currently exposed on the channel adapter, but you can define the container as a <bean... class="...SimpleMessageListenerContainer"/> and inject it into adapter using the listener-container attribute.
Or, use a custom ErrorHandler that throws an AmqpRejectAndDontRequeueException.
Related
I use ApiKit to receive queries. Occasionally I get the following line in a log file:
WARN org.mule.module.http.internal.listener.grizzly.ResponseCompletionHandler - HTTP response sending task failed with error: Locally closed
It seems that in this case the integration has not sent a response to the party that called the integration. I thought that there might be a sort of timeout before ApiKit closes the connection to caller but based on timestamps that might not be the case as everything happens within a second.
In this case the payload is sent to Artemis queue before this warning and despite the warning the message is read from Artemis normally and the whole flow works otherwise just fine besides this warning and not sending the response.
So, am I correct when I think that this warning is an indication why the response is not sent? In addition to that what can be done to prevent this situation?
How to deal with error handling in spring-kafka version 1.1.x?
As I see it was introduced in version 2.0...
In 1.x you can add an ErrorHandler implementation to the listener container (or container factory if you are using #KafkaListener annotations).
The 2.0 feature was added to make it easier to configure if you want a different error handler for each listener when using annotations.
I'm trying to publish to a MQTT topic thanks to the appropriate Apache Camel MQTT component.
So in my spring context XML I have the following:
<camel:to uri="mqtt:test?host=ssl://myhost:8883&publishTopicName=test&userName=test&password=test"/>
But I'm getting the following error at startup :
Failed to create Producer for endpoint:
Endpoint[mqtt:test?host=ssl://myhost:8883&publishTopicName=test&userName=test&password=test]. Reason: java.lang.NullPointerException
Everything is working fine when not using ssl, the following configuration (regular tcp instead of ssl) is working well :
<camel:to uri="mqtt:test?host=tcp://myhost:1883&publishTopicName=test&userName=test&password=test"/>
I've added the javax.net.ssl.trustStore JVM property pointing to my certificates store but without any effect.
Does someone already met this issue ? Is there something to specifically add in the spring DSL configuration file when using the camel mqtt component with ssl ?
EDIT :
The stacktrace of the NPE :
Caused by: java.lang.NullPointerException at
org.fusesource.hawtdispatch.transport.SslTransport.connecting(SslTransport.java:194)
at
org.fusesource.mqtt.client.CallbackConnection.createTransport(CallbackConnection.java:285)
at
org.fusesource.mqtt.client.CallbackConnection.connect(CallbackConnection.java:138)
at
org.apache.camel.component.mqtt.MQTTEndpoint.connect(MQTTEndpoint.java:305)
at
org.apache.camel.component.mqtt.MQTTProducer.doStart(MQTTProducer.java:38)
at
org.apache.camel.support.ServiceSupport.start(ServiceSupport.java:61)
at
org.apache.camel.impl.DefaultCamelContext.startService(DefaultCamelContext.java:3219)
at
org.apache.camel.impl.DefaultCamelContext.doAddService(DefaultCamelContext.java:1209)
at
org.apache.camel.impl.DefaultCamelContext.addService(DefaultCamelContext.java:1170)
at
org.apache.camel.impl.ProducerCache.doGetProducer(ProducerCache.java:442)
... 33 more
Debugging through javax.net.debug=ssl was useful.
Actually there were an issue on the java.security where the security.provider property was not set properly. That was manually changed for testing purpose related to another application.
Since, everything is working fine. Sorry for the post related to a internal specific mistake.
Alex.
I am using bean validation in my application. When there is no constraint validation errors everything works nicely. Every time there is a validation error, Glassfish throws the following exception:
Caused by: java.lang.ClassNotFoundException: javax.validation.groups.Default: java.net.MalformedURLException: Unknown protocol: osgi
at com.sun.corba.ee.impl.util.JDKBridge.loadClassM(JDKBridge.java:325)
at com.sun.corba.ee.impl.util.JDKBridge.loadClass(JDKBridge.java:228)
at com.sun.corba.ee.impl.javax.rmi.CORBA.Util.loadClass(Util.java:640)
at com.sun.corba.ee.impl.util.RepositoryId.getClassFromType(RepositoryId.java:628)
at com.sun.corba.ee.impl.orbutil.RepIdDelegator.getClassFromType(RepIdDelegator.java:169)
at com.sun.corba.ee.impl.encoding.CDRInputStream_1_0.readClass(CDRInputStream_1_0.java:1439)
The bean-validation.jar is present in glassfish/modules folder. The startup doesn't throw any exceptions regarding validation.
PS. Note that we are using remote beans with CORBA.
Reproducable on GlassFish 3.1.2.2 and 3.1.1.
I traced this problem to serialization of the ConstraintValidationException through CORBA. Somehow the bean-validation module is not loaded properly with osgi and the javax.validation.groups.Default class is missing. I made a quick workaround so that the ConstraintValidationException is intercepted and never sent through CORBA. Instead the validation error information is gathered in a custom Exception class that can be actually serialized through the services.
facing some strange behavior in OSB, i have configured WLS with MQ in client mode, i am doing some minor test to check the connection, i have created a proxy service to read the message from Q1 and a Business Service(BS) to route it to Q2. The issue is the proxy is able to read the message but the BS is throwing this:
JMSPool BEA-169807 There was an error while making the initial connection to the JMS resource named ALSB_JMS_SessionPool_491704821 from within an EJB or a servlet. The server will attempt the connection again later. The error was javax.jms.JMSException: [JMSPool:169803]JNDI lookup of the JMS connection factory AKBConnFact failed: javax.naming.NoInitialContextException: Cannot instantiate class: com.sun.jndi.fscontext.RefFSContextFactory [Root exception is java.lang.ClassNotFoundException: com.sun.jndi.fscontext.RefFSContextFactory
Note: The classpath or the domain/lib folder contains the RefFSContextFactory class
Any ideas gang..? TIA
The answer is this is a bug in OSB which needs to be reported.
As a workaround you need to individually set the jars in the weblogic classpath in your domain/server/bin folder. just go through the link below for more details:
http://forums.oracle.com/forums/thread.jspa?threadID=2135523&start=0&tstart=0