Specifying Delivery Delay on Messages¶
In a normal message flow, JMS messages that are sent by the JMS producer to the JMS broker are forwarded to the respective JMS consumer without any delay.
With the delivery delay messaging feature introduced with JMS 2.0, you can specify a delivery delay time value in each JMS message so that the JMS broker will not deliver the message until after the specified delivery delay has elapsed. Specifying a delivery delay is useful if there is a scenario where you do not want a message consumer to receive a message that is sent until a specified time duration has elapsed. To implement this, you need to add a delivery delay to the JMS producer so that the publisher does not deliver a message until the specified delivery delay time interval is elapsed.
The following diagram illustrates how you can use ESB Micro Integrator as a JMS producer and specify a delivery delay on messages when you do not want the message consumer to receive a message until a specified time duration has elapsed.
Synapse configuration¶
Given below are the synapse configurations that are required for mediating the above use case.
See the instructions on how to build and run this example.
<proxy name="JMSDelivery" startOnLoad="true" trace="disable" transports="https http">
<description/>
<target>
<inSequence>
<property name="OUT_ONLY" value="true"/>
<property name="FORCE_SC_ACCEPTED" scope="axis2" value="true"/>
<property action="remove" name="Content-Length" scope="transport"/>
<property action="remove" name="MIME-Version" scope="transport"/>
<property action="remove" name="Transfer-Encoding" scope="transport"/>
<property action="remove" name="User-Agent" scope="transport"/>
<property action="remove" name="Accept-Encoding" scope="transport"/>
<property name="messageType" scope="axis2" value="application/xml"/>
<property action="remove" name="Content-Type" scope="transport"/>
<log level="full"/>
<send>
<endpoint>
<address uri="jms:transport.jms.ConnectionFactory=myQueueConnectionFactory"/>
</endpoint>
</send>
</inSequence>
<outSequence>
<send/>
</outSequence>
</target>
<publishWSDL uri="file:repository/samples/resources/proxy/sample_proxy_1.wsdl"/>
</proxy>
<proxy name="JMSDeliveryDelayed" startOnLoad="true" trace="disable" transports="https http">
<description/>
<target>
<inSequence>
<property name="OUT_ONLY" value="true"/>
<property name="FORCE_SC_ACCEPTED" scope="axis2" value="true"/>
<property action="remove" name="Content-Length" scope="transport"/>
<property action="remove" name="MIME-Version" scope="transport"/>
<property action="remove" name="Transfer-Encoding" scope="transport"/>
<property action="remove" name="User-Agent" scope="transport"/>
<property action="remove" name="Accept-Encoding" scope="transport"/>
<property name="messageType" scope="axis2" value="application/xml"/>
<property action="remove" name="Content-Type" scope="transport"/>
<property name="JMS_MESSAGE_DELAY" scope="axis2" value="10000"/>
<log level="full"/>
<send>
<endpoint>
<address uri="jms:/transport.jms.ConnectionFactory=myQueueConnectionFactory"/>
</endpoint>
</send>
</inSequence>
<outSequence>
<send/>
</outSequence>
</target>
<publishWSDL uri="file:repository/samples/resources/proxy/sample_proxy_1.wsdl"/>
</proxy>
<sequence name="main">
<in>
<!-- Log all messages passing through -->
<log level="full"/>
<!-- ensure that the default configuration only sends if it is one of samples -->
<!-- Otherwise Synapse would be an open proxy by default (BAD!) -->
<filter regex="http://localhost:9000.*" source="get-property('To')">
<!-- Send the messages where they have been sent (i.e. implicit "To" EPR) -->
<send/>
</filter>
</in>
<out>
<send/>
</out>
<description>The main sequence for the message mediation</description>
</sequence>
<sequence name="fault">
<!-- Log the message at the full log level with the ERROR_MESSAGE and the ERROR_CODE-->
<log level="full">
<property name="MESSAGE" value="Executing default 'fault' sequence"/>
<property expression="get-property('ERROR_CODE')" name="ERROR_CODE"/>
<property expression="get-property('ERROR_MESSAGE')" name="ERROR_MESSAGE"/>
</log>
<!-- Drops the messages by default if there is a fault -->
<drop/>
</sequence>
<registry provider="org.wso2.micro.integrator.registry.MicroIntegratorRegistry">
<parameter name="cachableDuration">15000</parameter>
</registry>
<taskManager provider="org.wso2.micro.integrator.mediation.ntask.NTaskTaskManager"/>
See the descriptions of the above configurations:
Artifact | Description |
---|---|
Proxy Service 1 |
The JMSDeliveryDelayed proxy service sets a delivery delay of 10 seconds on the message that it forwards
|
Proxy Service 2 | The JMSDelivery proxy service does not set a delivery delay on the message. |
Build and run¶
Create the artifacts:
- Set up ESB Integration Studio.
- Create an integration project with an ESB Configs module and an Composite Exporter.
- Create the proxy services, registry artifact, scheduled task, and sequences with the configurations given above.
- Deploy the artifacts in your Micro Integrator.
Set up the broker:
-
Configure a broker with your Micro Integrator instance. Let's use HornetQ for this example.
- On Windows: HORNETQ_HOME\bin\run.bat --run
- On MacOS/Linux/Solaris: sh HORNETQ_HOME/bin/run.sh
-
Start the broker.
- Start the Micro Integrator.
Follow the steps given below to run the example:
-
Run the following java file (QueueConsumer.java), which acts as the JMS consumer that consumes messages from the queue:
package DeliveryDelay; import java.sql.Timestamp; import java.util.Date; import java.util.Properties; import javax.jms.Connection; import javax.jms.ConnectionFactory; import javax.jms.MessageConsumer; import javax.jms.Session; import javax.jms.TextMessage; import javax.jms.Queue; import javax.naming.Context; import javax.naming.InitialContext; /** * Sample consumer to demonstrate JMS 2.0 feature : * Message Delivery Delay * Classic API is used */ public class QueueConsumer { private static final String DEFAULT_CONNECTION_FACTORY = "QueueConnectionFactory"; private static final String DEFAULT_DESTINATION = "queue/mySampleQueue"; private static final String INITIAL_CONTEXT_FACTORY = "org.jnp.interfaces.NamingContextFactory"; private static final String PROVIDER_URL = "jnp://localhost:1099"; public static void main(final String[] args) { try { runExample(); } catch (Exception e) { e.printStackTrace(); } } public static void runExample() throws Exception { Connection connection = null; Context initialContext = null; try { // Step 1. Create an initial context to perform the JNDI lookup. final Properties env = new Properties(); env.put(Context.INITIAL_CONTEXT_FACTORY, INITIAL_CONTEXT_FACTORY); env.put(Context.PROVIDER_URL, System.getProperty(Context.PROVIDER_URL, PROVIDER_URL)); initialContext = new InitialContext(env); // Step 2. perform a lookup on the Queue Queue queue = (Queue) initialContext.lookup(DEFAULT_DESTINATION); // Step 3. perform a lookup on the Connection Factory ConnectionFactory cf = (ConnectionFactory) initialContext.lookup(DEFAULT_CONNECTION_FACTORY); // Step 4. Create a JMS Connection connection = cf.createConnection(); // Step 5. Create a JMS Session Session session = connection.createSession(false, Session.AUTO_ACKNOWLEDGE); // Step 6. Create a JMS Message Consumer MessageConsumer messageConsumer = session.createConsumer(queue); // Step 7. Start the Connection connection.start(); System.out.println("JMS consumer stated on the queue " + DEFAULT_DESTINATION + "\n"); //Clear the queue, if there is any previous messages in the queue TextMessage tempMessage; do{ tempMessage = (TextMessage) messageConsumer.receive(1); } while(tempMessage != null); // Step 8.1. Receive the message one TextMessage firstMessage = (TextMessage) messageConsumer.receive(); long first = System.currentTimeMillis(); System.out.println("Consumer received message: [ "+new Timestamp(new Date(first).getTime())+" ] " + firstMessage.getText() + "\n"); // Step 8.2. Receive delayed TextMessage secondMessage = (TextMessage) messageConsumer.receive(); long second = System.currentTimeMillis(); System.out.println("Consumer received dealyed message: [ "+new Timestamp(new Date(second).getTime())+" ] " + secondMessage.getText() + "\n"); System.out.println("Time difference between two messages : "+(second-first)/1000+"s"); } finally { // Step 9. Close JMS resources if (connection != null) { connection.close(); } // Also the initialContext if (initialContext != null) { initialContext.close(); } } } }
-
Invoke the two proxy services (http://localhost:8290/services/JMSDelivery, http://localhost:8290/services/JMSDeliveryDelayed) with the following payload:
<soapenv:Envelope xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/" xmlns:ser="http://services.samples" xmlns:xsd="http://services.samples/xsd"> <soapenv:Header/> <soapenv:Body> <ser:getQuote> <ser:request> <xsd:symbol>IBM</xsd:symbol> </ser:request> </ser:getQuote> </soapenv:Body> </soapenv:Envelope>
You will see that two messages are received by the Java consumer with a time difference of more than 10 seconds.
This is because the JMSDeliveryDelayed
proxy service sets a delivery delay of 10 seconds on the message that it forwards, whereas the JMSDelivery
proxy service does not set a delivery delay on the message.