How to handle maintenance to your queuing server (e.g. rabbitMQ)

Question

I'm new to queuing and looking into options that could be done operationally. I'm sorry if I have any conceptual misunderstandings related to queuing.

I watched this video where the presenter made a point that one benefit of SQS over RabbitMQ was that if you ever had to do maintenance on your RabbitMQ server that you could potentially have a bad time.

How are you supposed to handle maintenance on a rabbitMQ server? Are we supposed to be sending every message to a load balancer which then redirects those messages to the appropriate queue?

• Ewan: short answer : yes. stackoverflow.com/questions/28207327/… but you can apply the same logic to anything. Using the cloud version costs more but someone has handled the hard bits
• Ben Cottrell: RabbitMQ has some out-of-the-box capabilities for High-Availability and clustering in case nodes need to be taken down, although whether that's enough for you really depends on your availability requirements, you may still have additional work in front of you - blog.rabbitmq.com/posts/2011/10/…
• nanotek: If either of you would like to put your comments into an answer I will upvote / accept it