Disk & Storage Commands

Disk & Filesystem Usage

df -h                    # disk usage per mounted filesystem, human-readable
df -hT                     # + filesystem type (ext4, xfs, tmpfs...)
df -i                         # inode usage — a full inode table can block writes even with free space

du -sh /var/log               # total size of a directory
du -sh /var/log/*               # size of each item inside, one level deep
du -sh * | sort -rh | head -10    # find the biggest space consumers in current directory

df reports what the filesystem thinks is used; du sums actual file sizes. A mismatch (df shows full, du doesn’t add up) usually means a deleted file is still held open by a running process.

Block Devices & Partitions

lsblk                      # tree view: disks, partitions, and their mount points
lsblk -f                     # + filesystem type and UUID
sudo fdisk -l                  # list disks and partition tables (legacy-friendly)
sudo parted /dev/sdb print       # partition table for one disk (GPT-friendly)
blkid                              # UUIDs and filesystem types for all block devices

Mounting

mount                       # show all currently mounted filesystems
sudo mount /dev/sdb1 /mnt/data    # mount manually
sudo umount /mnt/data               # unmount
sudo umount -l /mnt/data              # lazy unmount (detach now, clean up when no longer busy)

cat /etc/fstab                          # persistent mount definitions, read at boot
sudo mount -a                             # mount everything in fstab (also validates it without rebooting)
findmnt /var                                # show mount details for a specific path

Formatting

sudo mkfs.ext4 /dev/sdb1     # format a partition as ext4
sudo mkfs.xfs /dev/sdb1        # format as xfs
sudo wipefs -a /dev/sdb1          # wipe existing filesystem signatures before reformatting

Finding What’s Eating Disk Space

sudo lsof +L1                    # files that are deleted but still held open (space not actually freed)
sudo du -x -d1 / | sort -rh | head -20    # largest top-level directories, staying on one filesystem
ncdu /var                             # interactive, navigable disk usage explorer (if installed)

Swap

free -h              # RAM + swap totals and usage
swapon --show           # active swap devices/files
sudo swapoff -a            # disable all swap
sudo swapon -a               # re-enable swap defined in fstab

LVM Quick Reference

sudo pvs                 # physical volumes
sudo vgs                   # volume groups
sudo lvs                     # logical volumes
sudo lvextend -L +10G /dev/vg_data/lv_app   # grow a logical volume
sudo resize2fs /dev/vg_data/lv_app             # grow the ext4 filesystem to match
sudo xfs_growfs /app                             # grow an xfs filesystem (on the MOUNT POINT, not the device)

Production Considerations

Monitor df -i (inode usage) alongside df -h — a filesystem can report free space but still fail writes with “No space left on device” if inodes are exhausted (common with millions of tiny files, like a poorly-rotated log/cache directory).
A full root filesystem can prevent even SSH login or logging from working — alert well before 90% usage, not at 100%.
xfs_growfs targets the mounted path; resize2fs targets the block device — mixing them up is a common operational mistake.

Quick Interview Answer

“df -h shows filesystem-level usage, du -sh sums actual file sizes in a directory — a gap between them usually means a deleted file is still open (lsof +L1 finds it). lsblk and blkid show the block-device/partition layout, mount/umount manage active mounts, and /etc/fstab defines what mounts automatically at boot.”

Common Mistakes

Ignoring inode exhaustion (df -i) and only watching df -h block usage.
Deleting a large log file directly instead of truncating it (> file.log) — the disk space isn’t freed until the writing process closes or restarts.
Running xfs_growfs against a device path instead of the mount point.