Hi Everybody, I am new to ActiveMQ. For the past 3 months I have been using HornetQ with JBoss 5.1 and am pretty much feed up with HornetQ, JBoss and the whole Resource Adapter XML tag soup. The lessons that I have learnt by using JBoss+HornetQ and by reading through the Geronimo documentation are
The Resource Adapter tag soup is ok to work with if all you want is a single server. But the moment you try to set up a cluster (master/slave or load balanced) of brokers things either break down or are so hard to do that you give up halfway. Setting up a cluster of JMS brokers on any App Server including Geronimo is poorly documented. Which is why I have decided that it may be a good idea to go with a pure JMS setup. I intend to use ActiveMQ is this manner. We have multiple Web Servers and Job Servers each running on their own individual JVMs. These JVMs totaling 8 at present need to submit messages onto a queue. There will be a set of consumers which will read these message, create SOLR documents and post them to a SOLR server for Indexing. The whole idea of using messaging, is that creating a SOLR document is time consuming. The application cannot afford to wait for this to complete. So the application (web servers and job servers) just puts a message onto a Queue and a Consumer would then create the SOLR doc and post it to SOLR. >From what I seen of the code in PooledConnectionFactory, it seems to be good for use inside the JVMs hosting the Web Servers or the Job Servers. But my question is, How do I use PooledConnnectionFactory? Is this intended to be instantiated only once inside a particular JVM? That is should I create Singleton (Manager kind of) to work with PooledConnection Factory? With Regards Swapnonil Mukherjee -- View this message in context: http://activemq.2283324.n4.nabble.com/How-do-I-use-the-PooledConnectionFactory-tp2536577p2536577.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
