Distributed system are like a game of Minesweeper, where unexpected failures lurk beneath the surface. How do we make sure you published this event, didn’t miss that message, maybe we’ve already seen one, or actually do we even have to care? So called delivery guarantees are here to stay and we’d better know how to uphold these assurances when needed. Through real-world, practical examples, we will explore a range of patterns, tools and techniques I learned and used over the years, evaluating their advantages and limitations and if they’re actually applicable to the task at hand.