The container revolution has swept the JVM world, too, and everyone's dockerizing, containerizing, and productionizing their Java applications. Like every other abstraction, containers bring along a unique set of challenges when it comes to monitoring, profiling, and tracing your application. Although Linux containers are just processes with some isolation features and resource quotas sprinkled on top, a lot of the performance tools that work on a standard Linux host are trickier to use with containers. Some traditional performance tools might even lie and report host information when you run them from within a container. In this talk, we will explore a set of modern Linux performance tools for container analysis focused on a set of key scenarios: top-level resource utilization of running containers; high CPU issues and CPU caps; filesystem and disk latency problems; and cross-container database accesses. We will use a variety of open source performance tools, including perf, async-profiler, flame graphs, and BCC; and mention some 3rd party open source and commercial solutions for container monitoring, which rely on an extra monitoring agent.