I am using kafka to implement event sourcing. Based on a threat in http://search-hadoop.com/kafka, I created sepparate topic for my events, e.g. UserCreatedEvent and UserUpdateEmailEvent. Both events are using userId as the message key. However assuming each topic has several partition, how to setup the consumer?
My consumer is listening to both events / topics. But I am not sure if my consumer didn't miss the event for user A from both topics because in topic UserCreated, key userA can end up in partition 1 but on UserUpdateEmail topic, key userA can end up in other partition. Or does it mean I need to make consumer to listen to all partition? But this approach is not scalable. Thank you, Fajar
