You can use the API output to create copies of messages in a queue, or to verify PUT and GET operations.
You can also post messages to a topic, and then create as many administrative subscriptions of the target queues as possible. Something can then get and write messages from one of these target queues. The problem is that MQ changes the message identifier between publication and consumption, while in the queue it remains static.
There is no built-in MQ function for recording messages. You can use linear logs and clear logs later, but they do not necessarily capture all messages due to optimization. (For example, a PUT message for a waiting getter outside the synchronization point). However, there is at least one commercial product for scanning transaction logs to check message activity.
The philosophy of MQ as a whole is that it is a delivery mechanism and deals with envelope data for routing and delivery, but does not apply to payload data. WAS, IIB, and other broker / transformation mechanisms is where IBM delivered all the features related to the usefulness of messages.
T.Rob source share