Ultimate Guide to Freeing Up Disk Space on Linux: Clean Logs, Cache, and More

Summary ⁓ Discover the best tips to free up disk space on Linux. Learn how to safely clean logs, remove temporary files, manage large directories, and optimize storage using tools like du and journalctl. Keep your system efficient with automated maintenance and practical space-saving techniques.

“Disk space is like pizza — you always think you have enough until you run out at the worst possible moment.

Linux servers generally perform really well and we all love it, but software running on it often conks off due to low disk space availability. It is important that you know the method and commands that will help you claim the precious disk space. Let's see how...

Find the culprit: Who's hogging up the space?

Disk Free (df)

$ df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/root        29G   23G  6.7G  78% /
devtmpfs        3.9G     0  3.9G   0% /dev
tmpfs           3.9G     0  3.9G   0% /dev/shm
tmpfs           789M  1.1M  788M   1% /run
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs           3.9G     0  3.9G   0% /sys/fs/cgroup
/dev/loop1       64M   64M     0 100% /snap/core20/2379
/dev/loop2       92M   92M     0 100% /snap/lxd/24061
/dev/loop3       92M   92M     0 100% /snap/lxd/29619
/dev/loop4       39M   39M     0 100% /snap/snapd/21759
/dev/sda15      105M  6.1M   99M   6% /boot/efi
/dev/sdb1        16G   28K   15G   1% /mnt
/dev/loop6       64M   64M     0 100% /snap/core20/2434
/dev/loop0       45M   45M     0 100% /snap/snapd/23258
tmpfs           789M     0  789M   0% /run/user/1000

Understanding the output

/dev/root: This is the main root filesystem where most of your system files and user data reside. It is nearly 78% full.
/dev/shm, /run, /sys/fs/cgroup, tmpfs: These are temporary filesystems (RAM-backed) used for various runtime operations. Their usage is minimal.
/snap/...: Snap packages are mounted as loopback devices. Their sizes are fixed and fully utilized (100% use is expected).
/boot/efi: Partition for EFI boot data. It’s only 6% used.
/mnt: Likely an external or secondary disk with plenty of free space (16G total, only 28K used).

Disk Usage (du)

$ sudo du -hxd1 / | sort -h
4.0K	/media
4.0K	/srv
16K	/lost+found
16K	/opt
24K	/snap
36K	/root
68K	/tmp
5.6M	/etc
476M	/boot
4.5G	/usr
5.9G	/var
12G	/home
23G	/

Understanding the output

/home folder on root is using the most amount of data 12GB . Digging in is easy, simply add the folder path you are interested in. In a few tries, I was able to figure it out, that since it is a build server where multiple agents were deployed (and recently upgraded), there was junk lying around. Trying the good old rm -rf culprit-folders gave me my space back.

$ sudo du -hxd1 /home | sort -h
12G	/home
12G	/home/azureuser

$ sudo du -hxd1 /home/azureuser | sort -h
...
...
...
189M	/home/azureuser/.nvm
2.6G	/home/azureuser/myagent-4
2.8G	/home/azureuser/myagent-3
2.9G	/home/azureuser/myagent-2
3.1G	/home/azureuser/myagent-1
12G	/home/azureuser

$ sudo du -hxd1 /home/azureuser/myagent-1 | sort -h
54M	/home/azureuser/myagent-1/_diag
105M	/home/azureuser/myagent-1/bin.3.241.0
107M	/home/azureuser/myagent-1/bin.4.248.0
317M	/home/azureuser/myagent-1/externals.3.241.0
317M	/home/azureuser/myagent-1/externals.4.248.0
2.2G	/home/azureuser/myagent-1/_work
3.1G	/home/azureuser/myagent-1

The flags in the command imply:

-h: Human-readable sizes.
-x: Restrict the scan to the /dev/root filesystem (avoid crossing into other mounts like /mnt or /snap).
-d1: Show only the sizes of top-level directories.

Translation: Show me all the big directories in /, but please sort them nicely so I can pick on the fattest one.

You’ll likely see the usual suspects: /var, /home, /usr. Spoiler alert: /var is often the worst offender.

Log Files: Do You Really Need ALL That History?

Your system loves keeping a diary, especially in /var/log. But, let’s face it: Do you really need logs from six months ago when you only debugged that server crash once? No.

Here’s how to trim the fat:

To check which logs are hogging space:

$ sudo du -ah /var/log | sort -h | tail -n 10

If you’re paranoid about accidentally deleting something important, compress them instead:

$ sudo gzip /var/log/<filename>

To remove the outdated ones:

$ sudo snap remove <package-name> --revision=<old-revision>

Clean Package Caches: Because You Don’t Need a Backlog

Ever downloaded software and wondered why your disk is mysteriously shrinking? That’s because package managers like apt keep a cache of installation files.

To clean it up:

$ sudo apt clean

And while you are it, kick out the unused dependencies:

$ sudo apt autoremove
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following packages will be REMOVED:
...
  linux-image-5.15.0-1065-azure linux-image-5.15.0-1067-azure linux-image-5.15.0-1068-azure linux-image-5.15.0-1070-azure
  linux-image-5.15.0-1071-azure linux-image-5.15.0-1072-azure linux-modules-5.15.0-1064-azure linux-modules-5.15.0-1065-azure
  linux-modules-5.15.0-1067-azure linux-modules-5.15.0-1068-azure linux-modules-5.15.0-1070-azure linux-modules-5.15.0-1071-azure
  linux-modules-5.15.0-1072-azure linux-tools-5.15.0-1064-azure linux-tools-5.15.0-1065-azure linux-tools-5.15.0-1067-azure
  linux-tools-5.15.0-1068-azure linux-tools-5.15.0-1070-azure linux-tools-5.15.0-1071-azure linux-tools-5.15.0-1072-azure
0 upgraded, 0 newly installed, 56 to remove and 47 not upgraded.
After this operation, 2019 MB disk space will be freed.
Do you want to continue? [Y/n]

Press Y if you are good to remove these.

Look for Hidden Giants: The Largest Files

When in doubt, find the truly greedy files with this command:

$ sudo find / -xdev -type f -exec du -h {} + | sort -rh | head -n 10
175M	/usr/bin/mongod
143M	/usr/bin/mongosh
140M	/home/azureuser/vsts-agent-linux-x64-3.241.0.tar.gz
124M	/usr/bin/mongos
123M	/var/log/btmp.1
112M	/usr/lib/mongosh_crypt_v1.so
109M	/var/lib/docker/buildkit/metadata_v2.db
103M	/usr/bin/dockerd
100M	/var/lib/mongodb/journal/WiredTigerPreplog.0000000002
100M	/var/lib/mongodb/journal/WiredTigerPreplog.0000000001

This will point out the top 10 offenders. Approach with caution, though—don’t delete system files unless you want your Linux system to stage a rebellion.

Handle the Big Boss: /var/log/journal

The systemd-journald logs in /var/log/journal can balloon to ridiculous sizes. If you’re running low on space, it’s okay to trim them down.

To safely clean them:

$ sudo journalctl --vacuum-size=500M
sudo journalctl --vacuum-time=2weeks

Move Big Files to an External Drive

Got some big files you can’t delete but don’t need on your main drive? Move them to an external disk:

$ sudo mv /path/to/large/files /mnt/external_drive/

After a little bit of clean up here is how my drive looks like:

Filesystem      Size  Used Avail Use% Mounted on
/dev/root        29G   14G   16G  46% /
devtmpfs        3.9G     0  3.9G   0% /dev
tmpfs           3.9G     0  3.9G   0% /dev/shm
tmpfs           789M  920K  789M   1% /run
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs           3.9G     0  3.9G   0% /sys/fs/cgroup
/dev/loop1       64M   64M     0 100% /snap/core20/2379
/dev/loop2       92M   92M     0 100% /snap/lxd/24061
/dev/loop3       92M   92M     0 100% /snap/lxd/29619
/dev/loop4       39M   39M     0 100% /snap/snapd/21759
/dev/sda15      105M  6.1M   99M   6% /boot/efi
/dev/sdb1        16G   28K   15G   1% /mnt
/dev/loop6       64M   64M     0 100% /snap/core20/2434
/dev/loop0       45M   45M     0 100% /snap/snapd/23258
tmpfs           789M     0  789M   0% /run/user/1000

So, next time your Linux system gives you the low-disk-space warning, smile and say: “I’ve got this.” Because now you do.

Happy cleaning! 🧹