Monitoring System Resources with top and htop

1. The Classic Monitor: top

The top command is a standard utility found on virtually every Linux and Unix-like system, including your Fedora installation. Because it's always available, knowing how to use it is a fundamental skill for any system administrator or developer.

To get started, simply open your terminal and run the command:

top

You'll see a full-screen display that updates every few seconds. It can look intimidating at first, so let's break it down. The screen is split into two main parts:

1. The Summary Area (Top): The first five lines provide a high-level overview of your system's health.
2. The Process List (Bottom): A detailed list of running processes, sorted by CPU usage by default.

To understand what all this information means, let's watch a video that provides a clear and detailed explanation.

After watching the video, you should be familiar with the layout. Here are the most common interactive commands you'll use while top is running:

• q: Quit and return to the terminal prompt.
• P (Shift + p): Sort the process list by Processor (%CPU) usage (this is often the default).
• M (Shift + m): Sort the process list by Memory (%MEM) usage.
• u: Filter the list to show processes owned by a specific user. top will prompt you for the username.
• c: Toggle the COMMAND column between showing just the process name and the full command path with arguments.
• k: Kill a process. top will prompt you for the PID of the process you want to terminate.
• 1 (the number one): Toggle the CPU usage display between a single, aggregated summary and a separate line for each CPU core.

For a more comprehensive list of commands, the "Using top" section in the OneUptime article is a great reference to bookmark.

2. The Modern Successor: htop

While top is powerful and ubiquitous, htop is an enhanced, more user-friendly alternative that many developers and sysadmins prefer. It presents the same information but in a more visually intuitive way, with color-coding, graphical meters, and full mouse support.

Installing htop

htop is not always installed by default, but it's readily available in Fedora's repositories. To install it, use the dnf package manager you learned about previously:

sudo dnf install htop

The installation process is also covered in the following video.

Exploring the htop Interface

Once installed, run htop from your terminal:

htop

You'll immediately notice the differences:

• Visual Meters: Graphical bars at the top show usage for each CPU core, memory, and swap space.
• Color Coding: Makes it easy to distinguish between different types of processes and resource usage levels.
• Mouse Support: You can click on column headers to sort, scroll through the process list with your mouse wheel, and click on processes to select them.
• Function Key Menu: A helpful menu at the bottom of the screen shows you all the available actions, mapped to the F1-F10 keys.

Let's explore some of the most useful features of htop.

Here's a summary of the most important htop functions:

• F10, q: Quit htop.
• F6 (or mouse click): Sort by a specific column. A menu will appear letting you choose.
• F3: Search for a process by name. It will highlight the first match.
• F4: Filter the list, showing only processes whose command contains the text you enter.
• F5: Toggle Tree view, which shows parent-child relationships between processes. This is very useful for web servers like Nginx or Apache, which spawn worker processes.
• F9: Kill a process. This brings up a menu of signals you can send to the selected process.
• F2: Setup menu, where you can customize meters, columns, and colors.

The OneUptime article, "How to Handle System Monitoring with top and htop," has a great table under the section Comparing top vs htop that summarizes the key advantages of htop.

3. Do-It-Yourself: Hunting a Runaway Process

Now it's time to put this knowledge into practice. A common scenario you'll face as a developer is a "runaway script"—a piece of code stuck in an infinite loop that consumes 100% of a CPU core. Let's simulate this and use top and htop to find and fix the problem.

Step 1: Create the Runaway Script
Open a text editor and create a file named busy_script.py with the following content. This simple script creates an infinite loop that will max out one of your CPU cores.

# busy_script.py
print("Starting a busy loop...")
while True:



    # This loop does nothing but consume CPU time
    pass

Save the file.

Step 2: Run the Script
In your terminal, run the script in the background using the & symbol. This will start the process and immediately give you your terminal prompt back.

python3 busy_script.py &

Step 3: Hunt the Process with top

1. Launch top.
2. You should see a python3 process at or near the top of the list, with a %CPU value close to 100.0.
3. Press 1 to see the per-core CPU usage. You'll notice that one core is maxed out while others are relatively idle.
4. Note the PID of the python3 process from the first column.
5. Press k to activate the kill prompt. top will suggest a PID to kill (usually the top one).
6. Type the PID of your python3 script and press Enter. At the next prompt, just press Enter again to send the default SIGTERM signal.
7. The python3 process will disappear from the list.
8. Press q to exit top.

Step 4: Hunt the Process with htop
Let's do it again, this time with the more user-friendly tool.

1. Start the script again in the background: python3 busy_script.py &
2. Launch htop.
3. The python3 process should be at the top of the list, highlighted and showing high CPU usage on one of the meters.
4. Use your arrow keys or mouse to ensure the python3 process is selected (highlighted).
5. Press F9 to bring up the "Kill" menu.
6. The default signal, 15 SIGTERM, is already selected. Press Enter to send it.
7. The process will be terminated and vanish from the list.
8. Press F10 or q to quit.

This exercise demonstrates the core workflow for identifying and dealing with resource-intensive processes—a vital skill for maintaining a healthy server.

Conclusion

Congratulations! You've moved beyond static snapshots and can now perform live monitoring of your system's health. You've learned to use both the universally available top and the feature-rich htop to keep an eye on your server's performance.

Key Takeaways:

• top and htop provide a real-time, dynamic view of running processes and system resource usage.
• top is the classic, pre-installed tool. Key commands include P for CPU sort, M for memory sort, and k to kill.
• htop is a modern, user-friendly alternative with colors, graphs, and mouse support. Key functions include F6 (Sort), F4 (Filter), and F9 (Kill).
• Both tools are essential for diagnosing performance bottlenecks by identifying processes with high CPU or memory consumption.

In this lesson, you used the interactive "kill" functions in top and htop. These are convenient shortcuts for a more fundamental Linux concept: sending signals to processes. In our next lesson, we will dive deeper into this, learning how to use the kill and killall commands directly to terminate, pause, and control processes with precision.