Thanks @alexey.zimarev for your patience while I try to describe my questions.
We have a very similar implementation based on your book. The problem I’m trying to resolve now is how to handle events that the subscription wasn’t previously handling. Since the subscription stores the checkpoint, once the subscription has run at least once, it will only pay attention to future events.
Our system will have hundreds of different event types across 20+ different bounded contexts, each potentially running in a different microservice. I’d like to come up with a solution that allows us to catch up with event types that a bounded context wasn’t previously handling while still minimizing the number of subscriptions to EventStore that we have.