💾 Disk Space Diagnostics

Over time, logs, NixOS generations, and user data will consume disk space. Use these commands to identify where space is being used.

📊 General Disk Usage

To see an overview of all mounted filesystems:

df -h

Look for the usage percentage of / (root), /nix, and /home.

📁 Identifying Large Files and Folders

To find the largest directories where user data or logs typically accumulate:

doas du -sh /home/* /var/* /root/* /tmp/* | sort -hr | head -n 15

Avoid running du -sh /* as root; traversing the entire /nix/store or virtual filesystems like /proc is extremely slow and may hang your session.

If you prefer an interactive tool, you can temporarily install and use ncdu or dust:

nix shell nixpkgs#ncdu -c ncdu /

📦 Nix Store Usage

The /nix/store grows every time you install new software or rebuild your system. It holds all current and past versions of software.

To see how much space the Nix store is using:

du -sh /nix/store

To clean up old, unused NixOS generations and free up space, see the Maintenance & Cleanup guide.

🌲 Btrfs Specific Diagnostics

Because this platform uses Btrfs, standard df commands might not show the full picture regarding compression and subvolumes.

To get an accurate view of Btrfs space usage across your subvolumes:

doas btrfs filesystem df /

To see raw device usage vs allocated space:

doas btrfs filesystem show /

📜 Journal (Logs) Size

systemd's journal can grow large if left unchecked. The platform includes tuning to bound the log size automatically, but you can check its current usage with:

journalctl --disk-usage

If you need to manually clear space taken by logs:

doas journalctl --vacuum-time=7d  # Keep only the last 7 days
doas journalctl --vacuum-size=1G  # Keep only 1GB of logs