You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by KenCoe <Ke...@fnis.com> on 2009/01/06 20:58:22 UTC

IBM MQSeries user seeking advice on ActiveMQ usage

Hello, 
I'm an old user of IBM MQSeries, and, for whatever reason, the ActiveMQ
terms/architecture is confusing my old brain.  I'm used to communicating
between Queue Managers using local, remote and xmit queues.  

My requirement is fairly simple:  
An application on Server A needs to send message to another application on
Server B, which resides on another network. Server A is a clustered Windows
server utilizing a SAN for a shared drive with its fail-over node.  Server B
is a single server on the network DMZ, without access to the SAN.  The
messages should be consumed only once, by this single consumer on Server B.  

With MQSeries, I would have setup Queue Managers on both nodes, defined an
xmit queue on (between) both nodes, a remote queue definition on Server A,
and a local queue definition on Server B.  This would assure delivery of the
message in the event of a network outtage (after the network recovered).  I
don't know how best to accomplish this with ActiveMQ. Can someone advise me? 

-- 
View this message in context: http://www.nabble.com/IBM-MQSeries-user-seeking-advice-on-ActiveMQ-usage-tp21317810p21317810.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.


Re: IBM MQSeries user seeking advice on ActiveMQ usage

Posted by Joe Fernandez <jo...@ttmsolutions.com>.
Hi Ken,

One option is to deploy your producer on Server A with an 'embedded' broker.
Thus, if for whatever reason your network goes down, the producer can still
use the services of its embedded broker. You would also configure the Server
A broker with a networkConnector to establish a forwarding bridge between
itself and the broker on Server B. The Server B broker would need a
transportConnector to listen for and accept connection requests from the
Server A broker. The Server A broker will forward all messages to the 
Server B broker; provided the network is up, Server B broker has an
appropriate consumer, and there are no consumers on Server A.  

Joe
Free ActiveMQ Reference Guide at http://www.ttmsolutions.com
  


KenCoe wrote:
> 
> Hello, 
> I'm an old user of IBM MQSeries, and, for whatever reason, the ActiveMQ
> terms/architecture is confusing my old brain.  I'm used to communicating
> between Queue Managers using local, remote and xmit queues.  
> 
> My requirement is fairly simple:  
> An application on Server A needs to send message to another application on
> Server B, which resides on another network. Server A is a clustered
> Windows server utilizing a SAN for a shared drive with its fail-over node. 
> Server B is a single server on the network DMZ, without access to the SAN. 
> The messages should be consumed only once, by this single consumer on
> Server B.  
> 
> With MQSeries, I would have setup Queue Managers on both nodes, defined an
> xmit queue on (between) both nodes, a remote queue definition on Server A,
> and a local queue definition on Server B.  This would assure delivery of
> the message in the event of a network outtage (after the network
> recovered).  I don't know how best to accomplish this with ActiveMQ. Can
> someone advise me? 
> 
> 

-- 
View this message in context: http://www.nabble.com/IBM-MQSeries-user-seeking-advice-on-ActiveMQ-usage-tp21317810p21318345.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.