Where does a BasicReject with requeue actually go?

Question

This seems like a simple question, but I'm having a hard time finding a definitive answer. If in RabbitMQ 3.6.1 I have a queue that looks like this:

5  4  3  2  1  <= head

And I consume message 1, then do:

channel.BasicReject(ea.DeliveryTag, true);

Will the 1 end up on the end of the queue or at the head of the queue (assuming for the sake of simplicity that nobody else is consuming the queue at the time)? So will I end up with:

1  5  4  3  2  <= head

or:

5  4  3  2  1  <= head

And is there anyway to control it (one way would be to ack the message and repost it entirely I suppose)? I actually want the first situation because I'm rejecting 1 because a particular resource needed to process that message is currently unavailable. So I'd like to throw it back on the queue to be processed later (when the resource is available) or get picked up by somebody else (who has resources available). But I don't want to throw it back just to keep picking it up again.

Answer 1

I'd say the answer is here, I'll quote a part:

Messages can be returned to the queue using AMQP methods that feature a requeue parameter (basic.recover, basic.reject and basic.nack), or due to a channel closing while holding unacknowledged messages. Any of these scenarios caused messages to be requeued at the back of the queue for RabbitMQ releases earlier than 2.7.0. From RabbitMQ release 2.7.0, messages are always held in the queue in publication order, even in the presence of requeueing or channel closure.

So we could assume that RMQ is implemented in that way the messages are not deleted form the queue (physically deleted) until they are ACKed, they may have a ACKed flag or whatever.