Disk Full

When a filesystem runs out of space, tasks fail with the error No space left on device: deployments, package installs, database writes, logs and file uploads all stop working. The error can also appear while the disk still has free space, if the filesystem has run out of inodes (too many small files).

When the Output of an event contains No space left on device, the Solution section in the event details links to this page.

Quick fix with Cleanup System

Run the Cleanup System utility first. It removes old logs, unused packages and caches without touching your websites or databases, and the Output of the event shows the disk usage before and after. If the disk is still nearly full, follow the steps below to find what is using the space.

1. Check disk space and inodes

Open Terminal on the server's overview page, or connect via SSH as root, then run:

df -h
df -i

• Use% close to 100% in df -h: the filesystem is out of space.
• IUse% at 100% in df -i: the filesystem is out of inodes. Look for folders holding a huge number of small files (cache, sessions, mail queue…).

2. Find what is using the space

List the largest top-level folders, staying on the same filesystem:

sudo du -xhd1 / 2>/dev/null | sort -h

Then drill into the largest one, for example /var:

sudo du -xhd1 /var 2>/dev/null | sort -h

Common culprits: logs in /var/log, Docker data in /var/lib/docker, databases in /var/lib/mysql, local backups, and uploads or caches inside website folders in /home.

3. Browse interactively with ncdu

ncdu lists folders sorted by size and lets you move around and delete files from one screen. Install it if it is missing:

sudo apt install -y ncdu

Then scan the whole filesystem, or a single folder:

sudo ncdu -x /
sudo ncdu /var

Key	Action
↑ ↓	Move through the list
Enter or →	Open a folder
←	Go back to the parent folder
d	Delete the selected file or folder (asks for confirmation)
q	Quit

WARNING

Deleting in ncdu cannot be undone. Check the path before confirming.

4. Free up space safely

Only remove data you have confirmed is no longer needed:

• Package cache and old system journal entries:

  sudo apt clean
  sudo journalctl --vacuum-time=7d

• Docker: check usage first with docker system df. Only prune images, containers or volumes you are sure are unused. Volumes hold application data (for example n8n data), so never remove them blindly.

• Old backups and archives (.zip, .tar.gz, .sql) left in website folders.

• Application logs: rotate or empty them through the application instead of deleting log files that are still being written.

5. Check for deleted files still held open

If df -h still shows the disk as full after you delete large files, a running process may still hold them open:

sudo lsof +L1

Restart the service shown in the output (on the Applications page or with systemctl restart), or reboot the server at a quiet time. Then run df -h again.

After cleanup

1. Run df -h and df -i again to confirm there is free space.
2. Repeat the operation that failed.
3. To keep the disk from filling up again: schedule Cleanup System, review how many backups are kept, or upgrade the server's disk.

DANGER

Never delete system folders, database files (/var/lib/mysql…), configuration files or live website data unless you know what they are and have a backup you can restore.