How can you use Bash to monitor system performance?

1 Install dependencies

Before you start, you need to make sure that you have some essential tools installed on your system. These include the sysstat package, which contains commands like sar, iostat, mpstat, and pidstat, and the bc package, which is a calculator utility. You can install them using your package manager, such as apt, yum, or pacman. For example, on Debian-based systems, you can run: sudo apt update && sudo apt install sysstat bc

2 Monitor CPU usage

One of the most important indicators of system performance is the CPU usage, which shows how much of the processor's capacity is being utilized by different processes. You can use the top command to get a real-time overview of the CPU usage and other system information, such as memory, load average, and process list. You can also use the mpstat command to get more detailed statistics about the CPU usage per core and per mode, such as user, system, idle, or interrupt. For example, to get the CPU usage per core every second, you can run: mpstat -P ALL 1

3 Monitor memory usage

Another key indicator of system performance is the memory usage, which shows how much of the RAM and swap space is being used by the system and the processes. You can use the free command to get a summary of the memory usage and availability, including the buffers and cache. You can also use the vmstat command to get more detailed information about the memory usage and swapping activity, as well as the CPU and disk IO. For example, to get the memory usage every second, you can run: vmstat 1

4 Monitor disk usage

The disk usage shows how much of the disk space is being used and how much is free. You can use the df command to get a report of the disk usage per filesystem, including the size, used, available, and percentage. You can also use the du command to get the disk usage per directory or file, which can help you identify the largest or most frequently accessed files. For example, to get the disk usage of the home directory in human-readable format, you can run: du -sh /home

5 Monitor disk IO

The disk IO shows how much data is being transferred to and from the disk devices. You can use the iostat command to get a report of the disk IO per device, including the transfers per second, kilobytes per second, average queue size, and average service time. You can also use the pidstat command to get the disk IO per process, which can help you identify the most IO-intensive processes. For example, to get the disk IO per device every second, you can run: iostat -x 1

6 Monitor network traffic

The network traffic shows how much data is being sent and received over the network interfaces. You can use the ifconfig command to get a summary of the network traffic per interface, including the bytes, packets, errors, and collisions. You can also use the sar command to get more detailed statistics about the network traffic per interface and per protocol, such as TCP, UDP, or ICMP. For example, to get the network traffic per interface every second, you can run: sar -n DEV 1

7 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?