If you're talking about queues, that's expected behavior. A queue message exists on only one broker in an NoB, so the two on-prem brokers will complete for queue messages from your us-east master.
If you're talking about topics, then that's very unexpected behavior and we should dig deeper. And to address your explicit question, I'm not aware of any known issues with networkConnectors between those two versions, though of course there could be ones that haven't been discovered. Tim On Sep 8, 2017 8:40 AM, "mtod" <m...@thetods.net> wrote: My Setup: Systems: >OS: Windows Servers 2016 >ActiveMQ Version: 5.14.5 >Java: 1.8.141 I have setup 2 - 3 node Master/Slave/Slave clusters on AWS 2 on-prem ActiveMQ servers >1 running 5.14.5 >1 running 5.10.0 Current network configuration Germany (AWS) <-> USEast (AWS) <-> Cambridge (on-prem) I have traffic flowing between all servers except for the 5.10.0 ActiveMQ it seems that everytime I enable the network connector the 2 on-prem boxes compete for messages only one will get them. So the main question is there a issue using network connectors between different ActiveMQ versions such as 5.10.0 and 5.14.5? Thanks Mike -- Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html