In this deck, I quickly summarize how people have dealt with logging in Docker historically and then describe a comprehensive approach to logging and monitoring in Docker, based on research and customer interviews. While Docker adds a welcome layer of abstraction to the deployment of applications, it also challenges assumptions on how those applications should be managed. I discuss a comprehensive approach for collecting logs and metrics into a centralized platform and dissect the latest additions to Docker itself (log drivers, stats).