Text Processing (Grep/Sed/Awk)

Stop chasing column shifts when parsing system logs

🧩 The Challenge

Dealing with logs where the number of spaces between fields changes constantly is a nightmare. I’ve spent way too long trying to count tabs or spaces in an awk statement only to have the script break the moment a filename contains a space.

💡 The Fix

Use awk’s ability to treat multiple whitespace characters as a single field delimiter by default, then just grab the specific columns you actually care about. It saves you from writing complex regex just to pull a PID or a timestamp.

awk '{print $1, $4}' /var/log/my-app.log

⚙️ Why It Works

Since awk treats any run of whitespace as a field separator, you don’t have to worry about whether your log file used one space or a tab. It cleans up the mess for you automatically so you can just index the data by column number.

🚀 Pro-Tip: If you need to output those columns back into a specific format, pipe that into column -t to get everything lined up in a clean table.

Linux Tips & Tricks | © ngelinux.com | 10/26/2026