CRITICAL

How to Fix Linux Error ENOSPC: No Space Left on Device

Quick Fix Summary

TL;DR

Run `df -h` to check disk space and `df -i` to check inodes, then delete files or increase storage.

ENOSPC indicates the filesystem has no space for new data or metadata. This can be due to exhausted disk blocks or inodes, halting writes and potentially crashing applications.

Diagnosis & Causes

  • Physical disk or partition is 100% full.
  • Inode table is exhausted (too many small files).
  • Docker or container logs consuming all space.
  • Unrotated application or system logs.
  • Temporary files not being cleaned up.

    • Recovery Steps

    1

    Step 1: Diagnose the Root Cause

    First, identify if the issue is block space or inode exhaustion. Use `df` with different flags.

    bash 
    df -h
df -i
    2

    Step 2: Free Up Block Space (If Disk is Full)

    If `df -h` shows 100% use, find and remove large, unnecessary files. Start with common culprits like logs, caches, and temporary directories.

    bash 
    du -sh /* 2>/dev/null | sort -rh | head -10
sudo find /var/log -type f -name "*.log" -size +100M
sudo journalctl --vacuum-size=200M
    3

    Step 3: Free Up Inodes (If Inodes are Exhausted)

    If `df -i` shows 100% IUse, you have too many small files. Find directories with high inode counts and clean them.

    bash 
    sudo find / -xdev -type f | cut -d "/" -f 2 | sort | uniq -c | sort -rn
sudo find /path/to/suspect/dir -type f -delete
    4

    Step 4: Prevent Recurrence & Proactive Monitoring

    Implement log rotation, set up monitoring alerts for disk and inode usage, and schedule cleanup tasks.

    bash 
    sudo logrotate -f /etc/logrotate.conf
echo "*/30 * * * * root /usr/bin/find /tmp -type f -mtime +7 -delete" | sudo tee -a /etc/crontab

    Architect's Pro Tip

    "For Docker hosts, `docker system prune -a -f` and setting log driver with `--log-opt max-size` in `/etc/docker/daemon.json` prevents ENOSPC from runaway container logs."

    Frequently Asked Questions

    My `df -h` shows free space, but I still get ENOSPC. Why?

    This is classic inode exhaustion. Run `df -i`. You have free blocks but no free inodes (metadata entries) to create new files, directories, or symlinks.

    What's the fastest way to clear space in a production emergency?

    Target `/var/log`, `/tmp`, and container log directories (`/var/lib/docker/containers/*/*-json.log`). Use `truncate -s 0 <logfile>` to zero out large logs without deleting them, preserving file handles.

    Can I increase inodes on a live filesystem?

    No. Inode count is set at filesystem creation (`mkfs` with `-N`). You must back up data, recreate the filesystem with more inodes, and restore.

    Related Linux Guides

    502 Bad Gateway

    How to Fix 502 Bad Gateway in Nginx/Apache on Ubuntu 24.04
    502 Bad Gateway

    How to Fix 502 Bad Gateway in NGINX on Ubuntu 24.04 LTS
    ImagePullBackOff

    Kubernetes Troubleshooting Guide: Diagnosing ImagePullBackOff on Linux Nodes