Oracle provides a lot of wait events, performance statistics and metrics by default, but from time to time these metrics are still not enough to drill down the root cause of a (performance) issue. This session provides you with insights about specific OS kernel/debug features (primarily *NIX) and Oracle debug tools in general for revealing the Oracle kernel code flow in the safest way and how to interpret the outcome at its best. It includes several demos and a real life case about troubleshooting a performance issue with the mentioned methodology. All the used tools are free to use and not limited to a specific Oracle RDBMS edition.