Tips

Finding the Biggest Files and Folders in Linux Command Line

Quick tutorial to show you how to find the biggest files on your Linux machine using a few commands that you may already be familiar with du, sort, and head.

Christopher Murray
Christopher Murray

Table of Contents

This is a quick tutorial to show you how to find the biggest files on your Linux machine using a few commands that you may already be familiar with du, sort, and head.

Here's a quick summary:

To find the 10 biggest folders in current directory:

du -h | sort -hr | head -n 10

To find the 10 biggest files and folders in current directory:

du -ah | sort -hr | head -n 10

Read the rest of the article to get a detailed explanation of these commands.

How to find the biggest folders in Linux?

The du command is used for getting the disk usage. Sort command sorts the data as per your requirement. The head command displays the top lines of a text input source.

This is just one combination for getting the biggest files and directories in Linux command line. There can be several other ways to achieve the same result.

Finding Large Files Linux

What happens if you run these three commands together without options? Your output probably won’t be very useful.

When you run these commands, unless specified with du, everything will run automatically using the current working directory as the source file.

Sort without options arranges items in numerical order, but this behavior is a little strange. 100 is considered less than 12 because 2 > 0. That’s definitely not what we want.

Head here defaults to displaying the first 10 items. Depending on the directory you want to analyze, you can tailor this to find large files quickly.

[email protected]:~$ du | sort | head
100    ./.local/share/evolution/addressbook
108    ./.mozilla/firefox/jwqwiz97.default-release/datareporting
112    ./.local/share/gvfs-metadata
12    ./.cache/fontconfig
12    ./.cache/gnome-software/screenshots/112x63
12    ./.cache/thumbnails/fail
12    ./.config/dconf
12    ./.config/evolution
12    ./.config/gnome-control-center/backgrounds
12    ./.config/ibus

Adding Options

So let’s look at what might be more typical options.

Adding -n to sort command means that items will be sorted by numeric value. Adding -r means that the results will be reversed. This is what we want when searching for the largest number.

I’m also going to add -5 to limit our results further than the default for head. This value is something that you should decide based on what you know about the system.

You may want to expand the value to a number greater than 10, or omit it entirely if there are many large files you are trying to filter. Otherwise, you may run it, delete several files, but still have space issues.

Okay, let’s put it all together and see what happens.

[email protected]:~$ du | sort -nr | head -5
1865396    .
1769532    ./Documents
76552    ./.cache
64852    ./.cache/mozilla
64848    ./.cache/mozilla/firefox

That’s better, you can quickly see where the largest files are. You can do better, though. Let’s clean it up with some more options.

How to Count Files in Directory in Linux [5 Examples]
Here are several ways to count the number of files in a directory in Linux command line.

Human-Readable Output

The human options for certain commands help present numbers in a way that is familiar to us. Let’s try adding that to the du command.

[email protected]:~$ du -h | sort -nr | head -5
980K    ./.local/share/app-info
976K    ./.local/share/app-info/xmls
824K    ./.cache/thumbnails
808K    ./.cache/thumbnails/large
804K    ./.local/share/tracker

Corrected Human-Readble Output

Wait a second… Those numbers don’t make any sense. No, they don’t because You have only changed the content to human-readable for the du command. Sort has its own built-in function for human-readable numeric sort with -h. Both must be used to get the desired output. You can run into these kinds of issues often in Linux.

It’s important to experiment and make sure that your results “make sense” before using a command a specific way.

Let’s try it again.

[email protected]:~$ du -h | sort -hr | head -5
1.8G    .
1.7G    ./Documents
75M    ./.cache
64M    ./.cache/mozilla/firefox/jwqwiz97.default-release
64M    ./.cache/mozilla/firefox

That’s more like it.

Where are the largest files?

You can tell from the output that the Documents folder contains some larger files, but if you switch to that folder and run our command again, you don’t get the largest file. You get this:

[email protected]:~/Documents$ du -h | sort -hr | head -5
1.7G    .

This is just telling us what you already know. The current directory, referred to as ., has 1.7G worth of files. That isn’t helpful if you’re trying to find single, unusually large files.

You need to add another flag to du for this task. Using option -a, you can get the output that we’re looking for. Let’s try it.

[email protected]:~/Documents$ du -ah | sort -hr | head -5
1.7G    .
1.1G    ./1gig-file.file
699M    ./doc.tar
2.9M    ./photo-of-woman-wearing-turtleneck-top-2777898.jpg
1.4M    ./semi-opened-laptop-computer-turned-on-on-table-2047905.jpg

Conclusion

Did you enjoy this guide to finding large files in Linux? I hope all of these tips taught you something new.

If you like this guide, please share it on social media. If you have any comments or questions, leave them below.



Join the conversation.