See https://issues.apache.org/jira/browse/ARTEMIS-2261

Use case: global-max-size is set to some limit to prevent the broker from falling over when full of user messages.

If this limit is reached and you attempt to control the Broker with Management (over AMQP) using a server assigned temporary response queue, whilst the management requests now processed by the Broker (owing to the work of ARTEMIS-1710), the management responses are lost and don't reach the client.

This occurs because Artemis internally services the request an an AMQP dynamic node [1/2] by creating a temporary queue (AMQPSessionCallback#createTemporaryQueue()). As the name of the temporary queue does not match the management address prefix, the management response messages sent are discarded and a "Address "54111434-5686-4dbb-8a94-dd0182dbe7eb" is full.: ActiveMQAddressFullException[errorType=ADDRESS_FULL message=AMQ229102: Address "54111434-5686-4dbb-8a94-dd0182dbe7eb" is full" results.

[1] http://docs.oasis-open.org/amqp/core/v1.0/os/amqp-core-messaging-v1.0-os.html#type-source
[2] http://docs.oasis-open.org/amqp/core/v1.0/os/amqp-core-messaging-v1.0-os.html#type-target