Downtime sucks. Especially if you're running a high-scale environment and every minute you're down, you're losing lots of money. Our experience at Jelly Button with 4M daily active players taught us a lot about how to prevent downtime and how to effectively deal with crises. In this session, I will present our guidelines for downtime prevention and management, talk about best practices and effective communication. Relevant for both cloud and on-premise deployments.