Bartosz Sypytkowski

Yes, this example should cooperate with StatsD, but your issue seems to have a different reason. I'll check it out later.

Your web app may be running using different permissions than your own, i.e. when your running your app on IIS.

The essence of CQRS is to separate reads and writes. For this reason I think it's not necessary to return an updated read model as result of the request. It...

User index is used solely for fast finding user ids given user emails. It should be probably implemented more like separate persistent actor listening on event bus for registered users....

The sole purpose of the default AggregateCoordinator.Receive method is to react on the `Terminated` actor messages, nothing else. Business rules concerned with forwarding messages are handled in classes inheriting from...

Each actor has it's internal mailbox to buffer incoming messages. Problem is that after sending a `Stop` message to actor's mailbox, there may be still some messages posted to it,...

AggregateCoordinator serves few purposes: - It works like a router in the sense that it resolves an aggregates that should receive messages and forwards requests to them. Any messages that...

In akka, each actor is prepared to be potential supervisor. Also each one of them has a default supervisor strategy, which in this case is sufficient.

`IEvent` is custom marker interface defined for purpose of the project [here](https://github.com/Horusiath/AkkaCQRS/blob/master/src/AkkaCQRS.Core/Messages.cs#L14). Each domain event used in this sample implements it. This sets a barrier on programmer preventing from accidentally...

@Lavinski : 1. With current design we already have side effects. Things like setting up receive timeout, even sending a message to another actor, are actually side effects. While it...