On Page 224, you talk about messages arriving out of order. In that section, you talk about a message PaymentAccepted was received before another message was arrived (with the users address).

How do I know a Message B should be processed only after Message A is processed? In the example, should I access the address of a customer (for example John Doe) from the database and if it is not there, should I conclude that another message has not arrived?

Yeah, you got it. If the customer's address is not in the database, you know that a message should put it there at some point. So you put the message back on the queue and it gets tried again later.

At some point you have stop retrying messages because there is a problem. NServiceBus (and any message bus) will put the message on the error queue after a certain number of re-tries, so you will be notified of the problem.

Trying to fix my brain to think "messaging" way

It does require a bit of brain re-wiring but you seem to be getting there quite quickly... which is good because you will have to teach all of your colleagues soon

Many of my colleagues and our director are DDD fans. The director has been to Eric vans training. Jimmy' Bogards team has developed some projects for us. In one of the project, we use his MediatR.

We have occasional study on DDD and micro services. The challenge is always bringing the interest to practice in the midst of business priorities. The migration to a distributed environment is a big challenge too.

Your DDD book sits on my desk and has recommended to colleges.

Thanks. I'm really happy that you liked it and recommend it. I enjoyed it answering your questions, and I can tell you have a good understanding of the material