Memory Monitoring — Monitoring RAM and Swap Usage in Linux¶

Memory Monitoring is the process of tracking RAM utilization, swap usage, memory allocation, cache usage, and memory-intensive processes to ensure Linux systems operate efficiently. Insufficient available memory can lead to application slowdowns, excessive swapping, Out of Memory (OOM) events, and system instability. Every Linux administrator, DevOps engineer, Cloud Architect, Platform Engineer, and Site Reliability Engineer (SRE) should understand how to monitor memory usage and troubleshoot memory-related issues in production environments.

Learning Path¶

Linux Mastery → Module 12: Monitoring & Logs → Lesson 6

Difficulty: Beginner → Intermediate

Reading Time: 95 Minutes

Course Progress

Course: Linux Mastery

Module: Monitoring & Logs

Lesson: 6 of 10

What You'll Learn¶

After completing this lesson, you'll be able to:

Understand Linux memory management
Monitor RAM and swap usage
Interpret memory statistics
Identify memory-intensive processes
Monitor memory performance
Detect memory bottlenecks
Troubleshoot memory issues
Apply production monitoring best practices

Prerequisites¶

Complete:

Modules 1–11
Module 12 Lessons 1–5

Why Learn Memory Monitoring?¶

Imagine a production application server.

Without monitoring:

Memory Usage Increases

↓

RAM Exhausted

↓

OOM Killer Terminates Process

↓

Application Downtime

With monitoring:

Memory Usage

↓

Alert at 80%

↓

Investigate

↓

Optimize

↓

Stable Application

Monitoring memory helps prevent application failures.

Linux Memory Overview¶

Linux uses memory for:

Running applications
Kernel operations
File system cache
Buffers
Shared memory
Swap

Unused memory is often used for caching to improve performance.

Memory Architecture¶

Applications

↓

RAM

↓

Cache/Buffers

↓

Swap (If Needed)

↓

Disk

RAM is significantly faster than swap storage.

Check Memory Usage¶

The most common command:

free -h

Example:

              total   used   free  shared  buff/cache  available

Mem:          16Gi   6Gi    2Gi     1Gi       8Gi         9Gi

Swap:          4Gi     0B     4Gi

Understanding free Output¶

Column	Description
total	Total RAM
used	Memory currently in use
free	Completely unused RAM
shared	Shared memory
buff/cache	Memory used for buffers and filesystem cache
available	Estimated memory available for new applications

Important

On Linux, a low free value is normal. Focus on the available column, which better represents how much memory can be allocated without swapping.

Check Memory in Megabytes¶

free -m

Check Memory in Gigabytes¶

free -g

Monitor Memory Continuously¶

Refresh every two seconds.

watch free -h

Useful during troubleshooting.

View Memory Statistics¶

Use:

vmstat

Example:

vmstat 2

Refresh every two seconds.

Important columns:

Free memory
Buffers
Cache
Swap activity
CPU usage

Monitor Memory with top¶

top

Memory summary:

MiB Mem

MiB Swap

Press:

Shift + M

Sort processes by memory usage.

Monitor Memory with htop¶

htop

Benefits:

Colorized display
Memory graphs
Interactive interface
Easy process management

Find Memory-Intensive Processes¶

Using ps:

ps aux --sort=-%mem

Top memory consumers:

ps aux --sort=-rss | head

View Process Memory Usage¶

Use:

pmap

Example:

pmap <PID>

Summary only.

pmap -x <PID>

Monitor Swap Usage¶

Display swap.

swapon --show

Or:

free -h

Excessive swap usage often indicates memory pressure.

Check OOM Events¶

The Linux kernel logs Out of Memory events.

View:

dmesg | grep -i oom

Or:

journalctl -k | grep -i oom

Virtual Memory Statistics¶

Detailed information.

cat /proc/meminfo

Example:

MemTotal

MemFree

Cached

Buffers

SwapTotal

Common Commands¶

Memory usage.

free -h

Memory statistics.

vmstat

Process monitoring.

top

Largest memory consumers.

ps aux --sort=-%mem

Kernel OOM messages.

dmesg | grep -i oom

Real Production Examples¶

Check available memory.

free -h

Monitor every second.

watch free -h

Find top memory consumers.

ps aux --sort=-rss | head

Check OOM events.

journalctl -k | grep -i oom

Production Perspective¶

Memory monitoring is essential for:

Kubernetes nodes
Database servers
Java applications
Python services
Web servers
Cloud virtual machines
CI/CD servers
Enterprise applications

Memory exhaustion is one of the most common causes of application instability.

Hands-on Lab¶

Task 1¶

Display memory usage.

free -h

Task 2¶

Refresh memory usage continuously.

watch free -h

Task 3¶

Display virtual memory statistics.

vmstat 2

Task 4¶

Open top.

top

Press:

Shift + M

Sort by memory usage.

Task 5¶

Identify the largest memory-consuming processes.

ps aux --sort=-%mem | head

Task 6¶

Display detailed memory information.

cat /proc/meminfo

Task 7¶

Display swap usage.

swapon --show

Task 8¶

Search for Out of Memory events.

journalctl -k | grep -i oom

Command Deep Dive¶

Command	Purpose	Production Example
`free -h`	Display memory usage	Capacity monitoring
`vmstat`	Virtual memory statistics	Performance analysis
`top`	Real-time process monitoring	Troubleshooting
`htop`	Interactive monitoring	System administration
`ps aux --sort=-%mem`	Largest memory consumers	Memory leak detection
`cat /proc/meminfo`	Detailed memory information	Low-level diagnostics

Common Memory Monitoring Mistakes¶

Mistake	Solution
Focusing only on the `free` column	Monitor the `available` column as well
Ignoring swap usage	Investigate sustained swap activity
Never checking OOM events	Review kernel logs regularly
Looking only at total memory	Identify individual memory-consuming processes
Assuming high cache usage is always a problem	Linux uses available RAM efficiently for caching

Production Troubleshooting Scenario¶

Scenario

A production API server becomes slow.

Investigation:

free -h

Result:

Available Memory

↓

Very Low

ps aux --sort=-rss | head

A Java process is consuming most of the RAM.

Further investigation confirms a memory leak.

After tuning the application and restarting the service:

Memory Usage

↓

Normal

↓

Performance Restored

Best Practices¶

Monitor available memory instead of only free memory.
Configure alerts before memory usage becomes critical.
Monitor swap usage regularly.
Investigate repeated OOM events immediately.
Identify long-running memory-intensive processes.
Monitor application memory trends over time.
Combine memory monitoring with CPU and disk metrics.
Use centralized monitoring tools such as Prometheus and Grafana for historical analysis.

Common Mistakes¶

❌ Assuming low free memory always indicates a problem.

✅ Verify low free memory always indicates a problem instead of assuming it.

❌ Ignoring swap activity.

✅ Always review swap activity.

❌ Never checking kernel OOM events.

✅ Always checking kernel OOM events.

❌ Monitoring only total memory usage.

✅ Avoid this mistake: monitoring only total memory usage.

❌ Ignoring gradual memory growth that may indicate a memory leak.

✅ Always review gradual memory growth that may indicate a memory leak.

Interview Questions¶

Beginner¶

What does free -h display?
What is swap memory?
Which command displays virtual memory statistics?
What is the purpose of vmstat?

Intermediate¶

What is the difference between used, free, and available memory?
How do you identify memory-intensive processes?
What causes Linux to use swap?
How do you investigate an OOM event?

Architect Level¶

How would you monitor memory across hundreds of Linux servers?
How would you troubleshoot a Kubernetes node experiencing memory pressure?
How would you distinguish between normal cache usage and an application memory leak?

Summary¶

In this lesson, you learned:

Linux memory management
RAM and swap monitoring
Memory statistics
Process memory analysis
OOM event investigation
Memory troubleshooting
Performance monitoring
Production best practices

Memory monitoring is a critical part of maintaining healthy Linux systems. By tracking RAM usage, swap activity, memory-intensive processes, and kernel OOM events, administrators can identify bottlenecks early, optimize application performance, and prevent production outages caused by memory exhaustion.

Key Takeaways¶

Use free -h to monitor memory utilization.
Focus on the available memory column rather than only free memory.
Monitor swap usage and investigate sustained swapping.
Use vmstat, top, and htop for deeper analysis.
Identify high-memory processes using ps.
Review OOM events to diagnose memory-related failures.

What's Next?¶

CPU Monitoring — Monitoring Processor Performance in Linux

You'll explore:

Understanding CPU utilization
Monitoring CPU performance
Load average
Using top, uptime, vmstat, and related tools
Identifying CPU-intensive processes
Troubleshooting CPU bottlenecks
Production CPU monitoring best practices

By the end of the lesson, you'll be able to monitor CPU performance, analyze system load, identify resource-intensive processes, and troubleshoot CPU-related performance issues in production Linux environments.