Bash Script to Monitor Server Health with CPU and Memory Usage

Use a Bash script with commands like top -bn1 for CPU, free -m for memory, and df -h for disk space, combined in a script to monitor server health.

Examples

InputRun script on a server with 10% CPU, 2048MB free memory, 50% disk usage

OutputCPU Load: 10% Memory Usage: 1024MB used / 2048MB total Disk Usage: 50% used on /

InputRun script on a server with 85% CPU, 512MB free memory, 90% disk usage

OutputCPU Load: 85% Memory Usage: 3584MB used / 4096MB total Disk Usage: 90% used on /

InputRun script on a server with 0% CPU, 0MB free memory, 100% disk usage

OutputCPU Load: 0% Memory Usage: 8192MB used / 8192MB total Disk Usage: 100% used on /

How to Think About It

To monitor server health, gather key metrics like CPU load, memory usage, and disk space using simple system commands. Combine these commands in a script that runs them one after another and prints the results clearly for easy reading.

Algorithm

1

Get current CPU load using a system command.

2

Get memory usage details using a system command.

3

Get disk usage information using a system command.

4

Format and print these values in a readable way.

Code

bash

#!/bin/bash

cpu_load=$(top -bn1 | grep 'Cpu(s)' | awk '{print 100 - $8"%"}')
mem_used=$(free -m | awk '/Mem:/ {print $3}')
mem_total=$(free -m | awk '/Mem:/ {print $2}')
disk_usage=$(df -h / | awk 'NR==2 {print $5}')

echo "CPU Load: $cpu_load"
echo "Memory Usage: $mem_used MB used / $mem_total MB total"
echo "Disk Usage: $disk_usage used on /"

Dry Run

Let's trace the script running on a server with 15.3% CPU load, 2048MB used memory out of 4096MB, and 45% disk usage.

1

Get CPU load

Command output: 'Cpu(s): 15.3%us, ... idle 84.7%' → cpu_load=100 - 84.7 = 15.3%

2

Get memory usage

free -m output line: 'Mem: 4096 2048 2048 ...' → mem_used=2048, mem_total=4096

3

Get disk usage

df -h / output line: '/dev/sda1 100G 45G 55G 45% /' → disk_usage=45%

Metric	Value
CPU Load	15.3%
Memory Used	2048 MB
Memory Total	4096 MB
Disk Usage	45%

Why This Works

Step 1: CPU Load Calculation

The script uses top -bn1 to get CPU stats and calculates CPU usage by subtracting idle percentage from 100.

Step 2: Memory Usage Extraction

It uses free -m to get memory in megabytes and extracts used and total memory values.

Step 3: Disk Usage Check

It runs df -h / to get disk usage of the root partition and extracts the used percentage.

Alternative Approaches

Using vmstat and awk

bash

#!/bin/bash
cpu_load=$(vmstat 1 2 | tail -1 | awk '{print 100 - $15"%"}')
mem_used=$(free -m | awk '/Mem:/ {print $3}')
mem_total=$(free -m | awk '/Mem:/ {print $2}')
disk_usage=$(df -h / | awk 'NR==2 {print $5}')
echo "CPU Load: $cpu_load"
echo "Memory Usage: $mem_used MB used / $mem_total MB total"
echo "Disk Usage: $disk_usage used on /"

vmstat gives a snapshot of CPU usage but requires waiting for 2 samples; slightly slower but more accurate.

Using sar command

bash

#!/bin/bash
cpu_load=$(sar 1 1 | grep Average | awk '{print 100 - $8"%"}')
mem_used=$(free -m | awk '/Mem:/ {print $3}')
mem_total=$(free -m | awk '/Mem:/ {print $2}')
disk_usage=$(df -h / | awk 'NR==2 {print $5}')
echo "CPU Load: $cpu_load"
echo "Memory Usage: $mem_used MB used / $mem_total MB total"
echo "Disk Usage: $disk_usage used on /"

sar provides detailed system activity reports but may not be installed by default.

Complexity: O(1) time, O(1) space

Time Complexity

The script runs a fixed number of system commands once, so time complexity is constant O(1).

Space Complexity

It uses a few variables to store command outputs, so space complexity is constant O(1).

Which Approach is Fastest?

Using top -bn1 is fast and available by default; vmstat and sar may be slower or require installation.

Approach	Time	Space	Best For
top command	O(1)	O(1)	Quick, default availability
vmstat command	O(1) but slower	O(1)	More accurate CPU snapshot
sar command	O(1) but slower	O(1)	Detailed system reports, requires install