How to Check Disk Space on Linux (df and du)

A full disk is one of the most common — and most preventable — causes of Linux outages. Failed writes, crashed databases, and broken logging all trace back to a filesystem hitting 100%. The two commands you need are df (how much space is free) and du (what’s using it).

See free space — df

df -h          # all filesystems, human-readable
df -h /var     # just the filesystem holding /var
df -hT         # also show filesystem type

Filesystem      Size  Used Avail Use% Mounted on
/dev/sda1        50G   42G  5.5G  89% /
/dev/sda2       100G   30G   65G  32% /var

The Use% column is what to watch. Anything near 100% is a problem waiting to happen. A full / crashes the server; a full /var usually takes down logging and databases first.

Find what’s using space — du

When a filesystem is filling up, du finds the culprit. Work top-down:

# Biggest items at the top level of /var, sorted smallest to largest
sudo du -sh /var/* | sort -h

# Drill into whatever was biggest — say, logs
sudo du -sh /var/log/* | sort -h

-s summarizes each argument, -h makes sizes human-readable, and sort -h orders by those human-readable sizes. Keep drilling into the largest directory until you find the offender.

Find large files anywhere

To hunt down big individual files regardless of location:

# Files over 100 MB, anywhere
sudo find / -type f -size +100M 2>/dev/null

# Over 1 GB, with sizes, sorted
sudo find / -type f -size +1G -exec du -h {} + 2>/dev/null | sort -h

The 2>/dev/null discards permission-denied noise so you see only the results.

The “disk full but du shows nothing” trap

Sometimes df says the disk is full but du can’t find the space. The usual cause is a deleted file still held open by a running process — the space isn’t freed until the process closes it. Find these:

sudo lsof +L1

This lists open files with a link count of zero (deleted but still open). Restarting the process holding the file releases the space. A common case is a log file that was deleted while the service writing to it kept running.

Check inode usage too

A filesystem can be “full” with space to spare if it’s out of inodes — common when something creates millions of tiny files:

df -i          # inode usage per filesystem

If IUse% is at 100% but Use% isn’t, you’re out of inodes, not space. Find the directory with too many files:

sudo find / -xdev -type f 2>/dev/null | cut -d/ -f2 | sort | uniq -c | sort -rn | head

FAQ

What’s the difference between df and du? df reports free space as the filesystem sees it (fast, whole-filesystem). du adds up actual file sizes in directories (slower, but tells you what’s using the space). They can disagree when deleted-but-open files exist.

Why does df show less total space than my disk size? Filesystems reserve a percentage (often 5%) for root, and formatting overhead takes some. The reserved space is why a “full” disk can still be written to by root.

How do I monitor disk space automatically? A simple cron job that emails when usage crosses a threshold works; for proper alerting, tools like Netdata or Prometheus track it continuously. See the crontab quick reference for the scheduling side.

For the full monitoring toolkit, see Linux monitoring commands, and for memory see how to check memory usage.