Cluster rabbitmq node synchronization #904
Description
I use a cluster RabbitMQ with two nodes (Node A and Node B). I use classic queues, exchanges and bindings. The queues are created on Node A and are visible and reachable from Node B. These queues, exchanges and bindings are created by my java client currently connected to Node A.
Everything works correctly when we send message to exchange using routing key.
The problem occurs when the Node A crashes. My java client lost connection, connect to Node B and recreate queues, exchanges and bindings on the Node B.
After that, some bindings correctly works but some others cause a routing error from exchange.
Is it normal ? Am I doing something wrong?
Could you please tell me how to fix this problem ?