Skip to main content

Process Monitoring: Real-Time Watching

Process monitoring watches processes in real-time. See what's happening. Find problems.

Here's the thing: Monitoring processes shows problems. Use it to find what's wrong.

Real-Time Monitoring​

top: Basic Monitor​

top                               # Interactive monitor

Navigation:

  • q = Quit
  • M = Sort by memory
  • P = Sort by CPU
  • 1 = Show all CPUs
  • k = Kill process

My take: top is basic. But it works. Use it.

htop: Better Monitor​

htop                              # Better version

My take: Install htop. It's better. More colorful. Easier to use.

watch: Monitor Commands​

watch -n 2 'ps aux | grep nginx'  # Every 2 seconds
watch -n 5 'systemctl status nginx'

My take: watch monitors commands. Useful for tracking.

Process Trees​

pstree: Process Tree​

pstree                            # Process tree
pstree -p                         # With PIDs
pstree -u                         # With users

My take: pstree shows process relationships. Useful for understanding.

Process Hierarchy​

# Find parent process
ps -o ppid= -p PID

# Find child processes
ps --ppid PID

My take: Understanding hierarchy helps. Find parents. Find children.

Monitoring Specific Processes​

Monitor by Name​

top -p $(pgrep -d, nginx)

My take: Monitor specific processes. Useful for services.

Monitor by User​

top -u username

My take: Monitor user's processes. See what they're doing.

Common Monitoring Patterns​

Watch Service​

watch -n 2 'systemctl status nginx'

Monitor Resource Usage​

htop
# Sort by CPU or memory

Find Resource Hogs​

ps aux --sort=-%cpu | head -10    # Top CPU
ps aux --sort=-%mem | head -10    # Top memory

Common Mistakes (I've Made These)​

  1. Not monitoring: Monitor processes. Find problems early.

  2. Only checking CPU: Check CPU, memory, I/O. All matter.

  3. Not using htop: Install htop. It's better than top.

  4. Ignoring process trees: Trees show relationships. Understand them.

  5. Not setting up alerts: Manual monitoring is tedious. Automate it.

Real-World Examples​

Monitor Web Server​

htop
# Filter for nginx
# Watch CPU and memory

Find Slow Process​

ps aux --sort=-%cpu | head -10

Monitor Process Tree​

pstree -p nginx

What's Next?​

Now that you can monitor processes, you can find problems. Or review Process Management to understand processes better.

Personal note: I used to ignore monitoring. Then processes broke. Now I monitor constantly. It prevents problems. Monitor your processes.