Shell Script for Collecting Linux CPU, Memory, and Disk I/O Metrics

Today we share a Bash script that gathers Linux system information such as CPU core count, utilization percentages, context switches, interrupts, load averages, memory usage, swap statistics, and disk I/O metrics.

The script first determines the server's IP address using

ifconfig | grep inet | grep -vE 'inet6|127.0.0.1' | awk '{print $2}'

, then extracts the total number of CPU cores from /proc/cpuinfo.

CPU usage is obtained via top -b -n 1 combined with grep Cpu, awk, and cut to isolate user, system, idle, and iowait percentages.

Context switches and interrupt counts are read from vmstat -n 1 1 using sed and awk, while the task queue length is also derived from the same output.

Load averages for the past 1, 5, and 15 minutes are parsed from uptime with awk and cut.

Memory statistics, including total, used, free, and swap values, are collected with free and filtered via grep, awk, and sed.

Disk I/O statistics for a specific device (e.g., /dev/sda) are obtained using iostat -kx, then filtered with grep and awk to report read/write request rates, queue length, average request time, service time, and CPU utilization.

Finally, the script echoes each metric with descriptive labels, providing a concise snapshot of the server's performance.