🚀 Day 3 of My DevOps Journey – Linux Process Management & System Monitoring

👋 Recap & Focus

Today’s goal was to understand how Linux handles processes and how to monitor system resources. I learned how to start, stop, observe, and manage processes using tools like ps, top, htop, and more.

I also practiced running CPU-intensive processes in the background and saw how to manage them using jobs, fg, bg, and kill. This is core to DevOps work when dealing with real-world servers and services.

---

✅ Step 1: Viewing Running Processes

I started by exploring how to view all active processes:

ps aux          # See all processes
top             # Live process monitor
htop            # (installed with sudo apt install htop)

• ps aux gave me a static snapshot.

• top and htop helped me monitor CPU and memory usage in real time.

---

✅ Step 2: Killing Processes

I tested how to terminate processes using:

kill <PID>        # Kill by Process ID
pkill <name>      # Kill by process name
killall <name>    # Kill all instances of a process

❌ Where I Got Stuck

At one point, I ran:

ps aux | grep firefox
kill 8535

But it returned:

bash: kill: (8535) - No such process

🔍 I realized I was trying to kill the grep command itself, not Firefox! The correct way was to look for the real Firefox PID (not the grep line) or use pgrep firefox.

---

✅ Step 3: Background & Foreground Jobs

I used sleep and yes to test background processes:

sleep 300 &
jobs
fg %1
bg %1
kill %1

❌ What Tripped Me Up

After running multiple sleep 300 &, I had jobs [1] and [2]. When I ran fg %1, it said:

bash: fg: job has terminated

Then kill %1 gave:

no such job

✔️ The reason? That sleep process already ended before I could bring it to foreground or kill it.

---

✅ Step 4: System Info Commands

I used the following to get memory, CPU, and disk stats:

df -h          # Disk space
free -m        # RAM usage
uptime         # Load average and uptime
hostname       # Hostname
uname -a       # Kernel info

📊 These commands will be very useful later when monitoring cloud VMs or CI/CD agents.

---

✅ Step 5: Run & Monitor a CPU-Heavy Process

I ran:

yes > /dev/null &

This command runs a loop printing "y" forever, using almost 100% CPU.

🔍 Observation:

Using top, I saw it quickly jumped to the top of the process list.

pkill yes     # To kill it quickly

✅ I also learned to kill individual PIDs using kill <PID> and confirm with pgrep yes.

---

✅ Where I Got Stuck Again

When I tried this:

kill yes

It failed with:

bash: kill: yes: arguments must be process or job IDs

✔️ I fixed it by using either pkill yes or kill <PID> after finding the process with ps aux | grep yes.

---

✅ Sleep Duration Confusion

I tested with different sleep times :

sleep 300 &     # 5 minutes (300 sec)
sleep 9000 &    # 2.5 hours (9000 sec)
sleep 9999 &    # 2 hours 46 min (9999 sec)

✔️ All work the same — just different durations. Great way to simulate long-running background processes.

---

📘 Final Commands Reference

---

📦 GitHub Repo

📍 DevOps Journal – Day3.md

---

🧠 Reflection

“Today I learned how to observe, control, and kill processes — and how real Linux systems behave under pressure. It felt like looking under the hood of the operating system.”

---

🙌 Let's Connect

• 💼 LinkedIn

• 💻 GitHub

---

#devops #linux #processmanagement #100DaysOfDevOps #learninginpublic #top #kill #htop