I have found issue related to core-bridges and cluster (no backup). I receive duplicated messages.

I have two separate clusters (EAP+HQ) each has two nodes. There are two core-bridges from the first cluster to the second cluster. If I send messages into the first cluster I will be able to see them on the second node, this works fine. But problem occurs when I kill one node from the second cluster and I restart this server. I can see duplicated messages in the second cluster. This issue does not occurs with small amount of messages, e.g. for 800 messages I have not seen this issue but I have seen it with 5 000 messages.

Here is configuration:
cluster A = A1 and A2
cluster B = B1 and B2

Cluster A contains: inputQueue + core-bridge (inputQueue -> outputQueue on cluster B)
A1-core-bridge-B1
A2-core-bridge-B2

Cluster B contains outputQueue.

Test scenario:
1. Start both clusters
2. Stop both bridges on the A cluster via JMX console
3. Start client and push 5000 messages to A1 node
4. Cluster A will do LB, A1=2500 A2=2500
5. Start both bridges on the A cluster via JMX console
6. Wait till all messages will be delivered to cluster B
7. Check messages count on Bx nodes, there are correct values (2500 + 2500)
8. Kill one of B servers with kill -9
9. Restart killed server
10. Check messages count on Bx nodes, there are invalid values

We have tried to consume all messages from cluster B, we have received same count of messages as we had seen in jmx-console.

1. Can I have two core-bridges between two HQ clusters? I have not found any notice in the documentation about it.
2. Is there something missing in configuration?

This scenario is based on LODH #4 test-case and for this reason it is blocker for release and test development.
Andy, please could you take look at this issue and let me know if you are able to reproduce this issue? I am attaching test.