The guide shows how to install procs, a lightweight yet powerful tool that replaces the cluttered output of ps and top with a clear, hierarchical process tree. It walks through package‑manager commands for Debian/Ubuntu, Fedora/RHEL, and Arch Linux, while pointing out that you can fetch the source from GitHub if your distro lacks it, then demonstrates the basic command procs that prints PID, PPID, CPU%, MEM%, user, and command columns in a neatly indented format. Readers learn to fine‑tune the output with grep filters, --sort options for cpu or mem, JSON piping into jq for automated alerts, and even custom column sets like pid,ppid,user,cpu,threads,command to spotlight the metrics that matter most when diagnosing performance issues. Finally, the article reminds you that while procs is perfect for quick snapshots on a headless server, interactive tools such as htop are still the go‑to choice when you need real‑time graphs or a full TUI, and encourages everyone to try procs to bring order to process chaos.

How to Use the procs Command to Show Active Processes in Linux

You’ll learn how to install procs, a lightweight yet powerful alternative to ps and top, and how to use it to get a clear, hierarchical view of your running processes.

Why Pick procs Over the Classic Tools?

When I started troubleshooting a memory leak on a server, the default top output was a blur of numbers. Switching to procs gave me a tidy tree that let me spot the offending child process in seconds. Unlike ps, it shows real‑time CPU and memory usage, and unlike htop, you don’t need to toggle columns or dig through menus.

Installing procs

• Debian/Ubuntu

  sudo apt-get install procs

• Fedora/RHEL

  sudo dnf install procs

• Arch Linux

  sudo pacman -S procs

If you’re on a distro that doesn’t ship it, just grab the source from < https://github.com/dalance/procs> and build it. The Makefile is a single line: make install.

Basic Usage

Simply type:

procs

You’ll see something like this:

PID   PPID  CPU% MEM% USER     COMMAND
1234    1   0.2  1.3 root     /usr/lib/systemd/systemd
5678  1234 12.5 30.7 user     ./myapp -v

The hierarchy is obvious: child processes are indented under their parents, and the CPU/MEM columns show real‑time usage.

Why this matters: The indentation removes the guesswork of figuring out which process belongs to which parent, a pain in ps -ef where you have to match numbers manually.

Filtering with Grep

If you’re hunting for a particular service:

procs | grep nginx

You’ll get only the lines that contain “nginx”, keeping the output focused. This is handy when your system has hundreds of processes and you need to zero in on one.

Sorting by CPU or Memory

Add --sort followed by a key:

procs --sort=cpu

or

procs --sort=mem

The most resource‑hungry processes will float to the top, making it easy to spot bottlenecks. I’ve seen this happen after a bad driver update that caused an endless loop of high CPU usage; sorting by cpu instantly revealed the culprit.

Using procs in Scripts

Because procs outputs plain text, you can pipe its data into other tools:

procs --json | jq '.[] | select(.mem > 50)'

This line dumps the process list as JSON and then uses jq to filter any that use more than 50 % of RAM. It’s a quick way to alert yourself or trigger an automated cleanup script.

Customizing Columns

Want to see thread count instead of memory? Use:

procs --cols pid,ppid,user,cpu,threads,command

The order of columns is entirely up to you. I usually keep pid, user, and cpu visible, because those are the numbers that matter most when diagnosing a slow machine.

When procs Isn’t Enough

If you need real‑time graphs or interactive sorting, stick with htop. But for a quick snapshot on a headless server where you can’t afford a full TUI, procs is the sweet spot between detail and speed.

That’s all there is to it. Grab procs, give your process list a tidy makeover, and watch the chaos shrink into an easy‑to‑read tree.