How to check disk space

Quick answer: df -h shows how full each filesystem is; du -sh /var/* 2>/dev/null | sort -h shows what's eating space; ncdu / gives an interactive browser. Keep at least 10–15% free — a 100% full disk breaks databases and even SSH logins.

Overview

Disk usage grows silently: logs, caches, old backups, docker images. Checking takes seconds and prevents the classic "everything stopped working" incident caused by a full root filesystem.

Before you start

• SSH access.
• Optionally install ncdu:

  apt install -y ncdu     # Ubuntu / Debian
  dnf install -y ncdu     # AlmaLinux / Rocky

Step-by-step guide

1. Overall usage:

   df -h

   Look at the / line (and /var, /home if separate).
2. Find the heavy directories:

   du -xh / --max-depth=2 2>/dev/null | sort -h | tail -20

3. Interactive exploration:

   ncdu -x /

   Navigate with arrows, delete carefully with d.
4. Usual suspects:

   journalctl --disk-usage
   du -sh /var/log /var/cache /tmp
   docker system df        # if you use Docker

   Trim the systemd journal: journalctl --vacuum-size=200M; clean apt cache: apt clean; prune docker: docker system prune.
5. Watch inodes too — millions of tiny files can exhaust them while df shows free space:

   df -i

Common issues

• Deleted a huge file but space didn't return? A process still holds it open — find it with lsof +L1 and restart that service.
• df and du disagree? Same cause as above, or data hidden under a mount point.
• Disk genuinely too small? Upgrade the disk — Storage VPS plans offer large volumes at low cost.

When to contact support

If the disk is 100% full and the server misbehaves badly (can't log in, database down), see the out-of-disk emergency guide or open a support ticket.

Frequently asked questions

How much free space should I keep?

Keep at least 10-15% of the disk free at all times. A filesystem that hits 100% breaks databases, stops logging and can even prevent SSH logins, so act well before you reach that point.

I deleted files but space did not come back — why?

A running process still holds the deleted files open, so the kernel cannot release their blocks yet. Find the holder with lsof +L1 and restart that service to reclaim the space.

What fills the disk most often?

Logs (/var/log, journald), package caches, old backups and Docker images. journalctl --vacuum-size and docker system prune reclaim a lot.

Related articles

• How to expand disk space after upgrade
• How to upgrade VPS resources
• How to check VPS CPU, RAM and disk usage
• Server is out of disk space

Need a hand? Contact Cloud2Y support →