[jboss-dev-forums] [Design of Messaging on JBoss (Messaging/JBoss)] - Re: JBMESSAGING-674 - Propagating changes on ClusteredConnec

[clebert.suconic@jboss.com]

"timfox" wrote : If the client view is being updated, as you say, when a node crashes, then the only other possibility AFAICT is that the connection factory deploy/undeploy even is being triggered when a node crashes.
  | 
  | Can you explain how this happens?

The update is happening on ConnectionFactoryJNDIMapper::onReplicationChange..


Look at the method's implementation, you will see a call to updateClusteredClients:


  | line 359:            endpoint.updateClusteredClients(delArr, failoverMap);
  | 


And the LoadBalancingPolicy is being updated at

anonymous wrote : 
  | line 348:            del.setDelegates(delArr);
  | 

This is the very same routine we were using to rebing the ConnectionFactory when the ViewChanged, fired by the ClusteredPostOffice's replicator. We don't undeploy/deploy when a node crashes... we rebing the ConnectionFActory and the place I was telling you.

anonymous wrote : I am just saying that creating a new interface is redundant since you are not calling the methods polymorphically. 

Ah... I see what you mean.... I wanted to do more refactoring on it and have it totally poliymorphically as you were saying but I would need to move some stuff for that.  I will take a look if would be possible to refactor it a little bit more.

View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4004840#4004840

Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4004840