Slightly off topic, but given that there are a number of reasons why you can open a tcp socket but not have a usable daemon behind it, you probably want an application monitor in your load balancer rather than just a tcp monitor.
You could use the same code in your monitoring system to implement a part of a round-trip checker, to automatically test your messaging from end to end. On Fri, Jan 04, 2013 at 06:08:46AM -0800, jliezers wrote: > Yes, the load balancers are performing some kind on 'tcp probe' every 10 > seconds which probably explains the exceptions. > > I now need to determine if I need to fix the load balancer config or if I > can safely ignore the exceptions. > > > > > > -- > View this message in context: > http://activemq.2283324.n4.nabble.com/Transport-Connection-EOFException-tp4661326p4661342.html > Sent from the ActiveMQ - User mailing list archive at Nabble.com.
