Resolve Disk Full Errors

A full disk causes applications to crash, logs to stop writing, and databases to become corrupted. This guide shows how to quickly find and remove large files.

Check Disk Usage

$df -h

Filesystem      Size  Used Avail Use% Mounted on
/dev/vda1        25G   25G     0  100% /
tmpfs           1.9G     0  1.9G   0% /dev/shm

The root filesystem is 100% full.

Find Large Directories

$du -sh /* 2>/dev/null | sort -rh | head -10

8.5G    /var
6.2G    /home
3.1G    /usr
2.8G    /opt

Drill down into the largest directory:

$du -sh /var/* 2>/dev/null | sort -rh | head -10

7.8G    /var/log
512M    /var/lib
128M    /var/cache

Find Large Files

$find / -type f -size +100M 2>/dev/null | sort -k5 -rn

Or use a more readable approach:

$find / -type f -size +100M -exec ls -lh {} \; 2>/dev/null | sort -k5 -rh | head -20

Common Disk Space Culprits

Large Log Files

$ls -lh /var/log/*.log | sort -k5 -rh | head -10

Truncate a large log file (do not delete — the process may hold the file handle):

$truncate -s 0 /var/log/nginx/access.log

Old Kernel Images

$apt autoremove --purge

Docker Images and Containers

$docker system df
$docker system prune -a

Package Cache

$apt clean
$du -sh /var/cache/apt/

Journald Logs

$journalctl --disk-usage
$journalctl --vacuum-size=500M

Set Up Log Rotation

Prevent logs from growing unbounded:

$nano /etc/logrotate.d/myapp

/var/log/myapp/*.log {
    daily
    rotate 14
    compress
    delaycompress
    missingok
    notifempty
    sharedscripts
    postrotate
        systemctl reload myapp
    endscript
}

[!TIP] If the disk is completely full and you cannot write any files, use truncate -s 0 /path/to/large/file to free space without needing to create a new file.