Mailing-List: contact users-help@activemq.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@activemq.apache.org
Received-SPF: neutral (nike.apache.org: local policy)
Date: Tue, 5 Jul 2011 08:37:09 -0700 (PDT)
From: boday <ben.oday@initekconsulting.com>
To: users@activemq.apache.org
Message-ID: <1309880229528-3646309.post@n4.nabble.com>
In-Reply-To: <1309879496588-3646271.post@n4.nabble.com>
References: <1309264808605-3630225.post@n4.nabble.com>
 <1309526156876-3638168.post@n4.nabble.com>
 <1309879496588-3646271.post@n4.nabble.com>
Subject: Re: Master/Slave configuring clients
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

are you using Windows?  If so, you might see this bug 
https://issues.apache.org/jira/browse/AMQ-3273 AMQ-3273 ...it caused me a
lot of confusion while trying to test this scenario.


Vijay wrote:
> 
> I am rephrasing my question for get rid of confusion.
> 
> We have a Converter app (Standalone JMS consumer client based on Spring
> DefaultMessageListenerContainer) which will consume messages from AMQ
> Broker hosted on same server. An upstream system will publish messages on
> to "TEST.FOO"  Q on this broker. Whole purpose of convert app is to
> transform the message to another format and send it to downstream system
> for further processing.
> 
> 
> Server1 (working fine)
> -----------
> Consumer App
> Regular AMQ Broker
> 
> We verified this to be working fine with a consumer, AMQ broker on the
> same server.
> 
> For redundancy we want to  use Master/Slave Broker config + deploy one
> more Consumer app on server2.  This is the new config we have.  But this
> seem to be causing message loss. I am attaching a Junit test case to
> reproduce this.
> 
> Server1 
> -----------
> Consumer App 1 listening to tcp://server1:61616
> Master AMQ Broker (tcp://server1:61616)
> 
> Server2 
> -----------
> Consumer App 2  listening to tcp://server2:61617
> Slave AMQ Broker (tcp://server2:61617)
> 
> 
> 
> Attached is the JUnit test case:
> Steps
> 1. start master on 61616
> 2. Start slave on 61617
> 3. start consumer 1  to listen to tcp://host:61616
> 4. Start consumer2 to listen to tcp://host:61617
> 5. produce 30 msgs 
> Notice only part of message being consumed by listener 1
> 6. kill master and listener1
> 7. produce 50 msgs 
> Notice only part of messages being consumed by listener 1
> 7.kill slave
> 
> May I know what's wrong with this test case?
> 
> 
> http://activemq.2283324.n4.nabble.com/file/n3646271/MasterSlaveFailoverTestcase.zip
> MasterSlaveFailoverTestcase.zip 
> 


--
View this message in context: http://activemq.2283324.n4.nabble.com/Master-Slave-configuring-clients-tp3630225p3646309.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.