2:32 AM. PagerDuty wakes you up. DynamoDB is throttling. Should you wake up the team and fiercely charge to resolve the issue, or can it wait for tomorrow? Understanding the business impact and the affected users are the key points to making this decision. Those data points are usually not easy to obtain, especially in highly distributed asynchronous architectures like serverless. In this session, we will share guidelines on what needs to be part of your serverless application in order to be able to answer those questions in a matter of minutes. The main operational questions, when things go bad: - What is the user functionality being affected? - Which users were affected and how? - What is the root cause of these issues? Getting a good night's sleep is within arm’s reach... Other talks at this conference 🚀🪐 https://www.conf42.com/cloud2022 — 0:00 Intro 1:27 Talk