How does the Exchange-to-Exchange binding separate agents from the need to know server queues?

Question

I recently read the topic “Routing Topologies for Performance and Scalability with RabbitMQ,” and one thing I can’t understand why it makes a difference to have a private exchange related to exchanging public topics, and how does it eliminate the need to know about server queues?

Currently I am just posting a public exchange of topics and should not know if there are any lines associated with it. If there is no message in the queue, the message will be dead.

As I understand it, with Ex-to-Ex binding, I should expect the same behavior, if the queue is not there, the message will be dead.

Answer 1

The only time you need to know about the queue is when you use the default exchange, because you need to set the routing key as the name of the queue.

This article does not have a comprehensive treatise on routing topology, and I find this rather obscure. Personally, I use the exchange to exchange bindings with a general exchange, where messages are published, and private exchanges for consumers. An exchange is a fork, and then private exchanges can use any type of exchange they want without affecting the overall configuration of the exchange.

, , . , . .

.