Sorry, what I meant is to transfer jms requests from server1 to server2 so that server1 only focuses on basic business request handling and server2 focuses on jms request handling. In this way, server1 can obtain better performance since it will not handle any jms request on it.

Is this clear enough?

Thanks, Anders. We could do that, but a switch is preferred so that we can still use server1 for JMS if we do not want to use server2 in the future.

And can you show me a little more details about how to setup the JCA adapter? Or is there any document about this guide? Much appreciated.

Do you mean setup hornetq HA with a mod_cluster? I have tried that but met with some problems, which is why I am now asking for this in this thread.

If you can share with me how to setup the JCA adapter and how to configure the producers, that will be much helpful.

I have just had a quick view on the hornetq quickstarts

Chapter 7. Running the Examples

but seems no more useful information is found about the JCA adapter. Can you advise?

the JMS producers are running within server1.

And they are not running in a cluster.

So it is possible to do the JMS transfer without configuring mod_cluster or cluster at all? If it is this case, then that will be much better because what I am trying to find is only how to transfer the JMS requests to another server at lowest cost.

Thanks so much, Justin.

I have followed the user guide on hornetq bridge

http://docs.jboss.org/hornetq/2.3.0.CR1/docs/user-manual/html/core-bridges.html

but it only describes simple examples on the divert and bridge settings, as below.

<diverts>
  <divert name="test-divert">
        <address>queue/com.demo.test1</address>
        <forwarding-address>queue/com.demo.test2</forwarding-address>
        <exclusive>true</exclusive>
    </divert>
</diverts>
<bridges>
  <bridge name="test-bridge">
        <queue-name>queue/com.demo.test1</queue-name>
        <forwarding-address>queue/com.demo.test2</forwarding-address>
        <retry-interval>1000</retry-interval>
        <retry-interval-multiplier>1.0</retry-interval-multiplier>
        <reconnect-attempts>-1</reconnect-attempts>
        <use-duplicate-detection>true</use-duplicate-detection>
        <confirmation-window-size>10000000</confirmation-window-size>
        <static-connectors>
            <connector-ref>netty</connector-ref>
        </static-connectors>
  </bridge>
</bridges>

Are these correct and enough? As I unders queue "queue/com.demo.test1" will be forwarded to queue "queue/com.demo.test2" locally, and I have confirmed this indeed works. but I have no idea how to forward this queue to another server. I know it is done by bridge and I have added the bridge configuration in my standalone.xml in server1, but I do not know what is next? How will the connector-ref "netty" do for the bridge? Anything needs to be done in my code? Anything needs to configue in server2 standalone.xml? Please advise.

Thanks so much, Justin. I have just made it work! but only once..Then I modified some setting and it will not work forever.. I can not see where I have made it wrong. Can you help do a check for me?

This is configuration for server1 which wants to connect to server2:

  <diverts>
     <divert name="test-divert">
                        <address>queue/com.demo.testQueue</address>
                        <forwarding-address>queue/com.demo.testQueue1</forwarding-address>
                        <exclusive>true</exclusive>
                    </divert>
  </diverts>
  <bridges>
     <bridge name="test-bridge">
                        <queue-name>queue/com.demo.testQueue1</queue-name>
                        <forwarding-address>queue/com.demo.testQueue2</forwarding-address>
                        <retry-interval>1000</retry-interval>
                        <retry-interval-multiplier>1.0</retry-interval-multiplier>
                        <reconnect-attempts>-1</reconnect-attempts>
                        <use-duplicate-detection>true</use-duplicate-detection>
                        <confirmation-window-size>10000000</confirmation-window-size>
                        <static-connectors>
                            <connector-ref>remote-connector</connector-ref>
                        </static-connectors>
                    </bridge>
                </bridges>
                <connectors>
                    <netty-connector name="netty" socket-binding="messaging"/>
                    <netty-connector name="netty-throughput" socket-binding="messaging-throughput">
                        <param key="batch-delay" value="50"/>
                    </netty-connector>
                    <in-vm-connector name="in-vm" server-id="0"/>
                    <connector name="remote-connector">
                        <factory-class>
                              org.hornetq.core.remoting.impl.netty.NettyConnectorFactory
                        </factory-class>
                         <param key="host" value="localhost"/>
                        <param key="port" value="5446"/>
                    </connector>
                </connectors>

Configuration for server2 which is accepting requests from server1:

<connectors>
                    <netty-connector name="netty" socket-binding="messaging"/>
                    <netty-connector name="netty-throughput" socket-binding="messaging-throughput">
                        <param key="batch-delay" value="50"/>
                    </netty-connector>
                    <in-vm-connector name="in-vm" server-id="0"/>
                 <connector name="remote-connector">
                        <factory-class>
                              org.hornetq.core.remoting.impl.netty.NettyConnectorFactory
                        </factory-class>
                    </connector>
                </connectors>
                <acceptors>
                    <netty-acceptor name="netty" socket-binding="messaging"/>
                    <netty-acceptor name="netty-throughput" socket-binding="messaging-throughput">
                        <param key="batch-delay" value="50"/>
                        <param key="direct-deliver" value="false"/>
                    </netty-acceptor>
                    <in-vm-acceptor name="in-vm" server-id="0"/>
                    <acceptor name="remote-connector">
                        <factory-class>
                             org.hornetq.core.remoting.impl.netty.NettyAcceptorFactory
                        </factory-class>
                        <param key="host" value="localhost"/>
                        <param key="port" value="5446"/>
                    </acceptor>
                </acceptors>

And the queue name on server2 has been configured to testQueue2 which is defined in server1's bridge forwarding address.

<jms-queue name="queue/com.demo.testQueue2">
     <entry name="queue/com.demo.testQueue2"/>
</jms-queue>

And the connection factory used on server2 has been changed to the "remote-connector"

<pooled-connection-factory name="hornetq-ra">
                        <transaction mode="xa"/>
                        <connectors>
                            <connector-ref connector-name="remote-connector"/>
                        </connectors>
                        <entries>
                            <entry name="java:/JmsXA"/>
                        </entries>
                    </pooled-connection-factory>
</jms-connection-factories>

I can confirm I have made it work once after following the hornetq user manual, but then it never worked again after I messed up with testing different changes. I have tried to rollback to which version I have made it work but strangely could not.

If you have a chance, can you help check where could be the problem? I may be too dull to debug configuration stuff...

Thanks, Justin.

Correct, I am running the two servers on same host. I will try with your suggestion.

But I am a little confused about the queue name. Why is it working fine for each server without the prefix "jms.queue." when I did not add configuration of divert and bridge?