Process & Resource Monitoring (Top/Htop/Ps/Systemd)

Stop staring at flickering htop output to find short-lived processes

🧩 The Challenge

You know that feeling when a script or a cron job is acting weird, but it finishes before you can even get your eyes on it in htop? I’ve spent way too many Friday afternoons trying to catch a flickering process that’s only alive for a fraction of a second.

💡 The Fix

Use systemd-cgtop to look at the resource consumption across your control groups instead of focusing on individual PIDs. It’s a total game changer when you’re trying to figure out which service is actually chewing through your CPU while the process itself is playing hide-and-seek.

systemd-cgtop --depth=2

⚙️ Why It Works

Since most of your background noise is wrapped in service units anyway, this aggregates the data at the cgroup level. You get a much steadier view of resource usage that isn’t dependent on the process staying alive for multiple refresh intervals.

🚀 Pro-Tip: Keep an eye on the IO columns if you suspect a hidden log-writer is thrashing your disk.

Linux Tips & Tricks | © ngelinux.com | 7/15/2026