Container Basics On Linux (Namespaces/Cgroups)

Stop the PID namespace from hiding your container process trees

🧩 The Challenge

You ever run top or ps inside a container and see a clean list, only to realize you can’t actually see the child processes that spawned from your main app? It’s a massive headache when you’re trying to debug why a sub-process is stuck in a zombie state and you have no visibility into the parent-child relationship.

💡 The Fix

Use the nsenter tool to jump into the specific namespace of the container process so you can view the process tree exactly how the kernel sees it from inside the jail. It keeps you from having to install ps or top inside your lightweight distroless images.

nsenter -t <pid> -p -m ps auxf

⚙️ Why It Works

By targeting the PID and mount namespaces of the host process, you bypass the container’s isolation entirely to look at the process hierarchy directly from the kernel’s perspective. It effectively strips away the wall the container built around your process tree.

🚀 Pro-Tip: Alias this to your shell config so you aren’t hunting for that PID every single time.

Linux Tips & Tricks | © ngelinux.com | 10/5/2026