e2fsck

Check filesystem, reporting any damaged blocks

$ sudo e2fsck [/dev/sdXN]

$ sudo e2fsck -p [/dev/sdXN]

$ sudo e2fsck -c [/dev/sdXN]

$ sudo e2fsck -f [/dev/sdXN]

$ sudo e2fsck -fccky [/dev/sdXN]

e2fsck [options] device

-a
    Preen: auto-fix minor errors without prompts (boot default)

-b superblock
    Use alternate superblock location

-B blocksize
    Filesystem block size (usually auto)

-c [interval] [blocks]
    Check for bad blocks with badblocks(8)

-C fd
    Progress info to file descriptor fd (0=stdout)

-d
    Debug: print verbose debug info

-D
    Optimize directories (htree rebuild)

-E extended_options
    Ext4 extended opts (e.g., ^extended_options)

-f
    Force check even if marked clean

-F
    Flush FS buffers before checking

-I inode_size
    Set expected inode size

-l badblocks_file
    Add bad blocks from file to list

-L badblocks_file
    Replace bad blocks list from file

-m reserved_blocks_percent
    Set reserved blocks percentage

-n
    Read-only: check without changes

-p
    Auto-repair like -a, but fix more aggressively

-r repair_fd
    Display repair list to fd (deprecated)

-t fsck_time
    Set last fsck time

-v
    Verbose output

-y
    Answer 'yes' to all prompts

-z undo_file
    Create undo file for replay

e2fsck is the primary tool for checking and repairing ext2, ext3, and ext4 filesystems on Linux. It verifies the integrity of critical structures including superblocks, inodes, block and inode bitmaps, directories, and journals (for ext3/4). Errors are classified by severity, with options for read-only verification, automatic fixes, or interactive prompts.

Invoked automatically during boot via fsck (as /sbin/fsck.ext2, etc.) if the filesystem isn't marked clean. Manual runs require an unmounted device to prevent corruption. Key features include bad block detection via badblocks(8), alternate superblock usage for damaged primaries, directory optimization, and progress reporting.

In preen mode (-a or -p), it fixes non-critical issues silently, ideal for boot-time use. Force mode (-f) checks clean filesystems. For disasters, interactive mode lists fixes; -y assumes yes to all. Large filesystems may take hours; monitor with -C 0 for stdout progress.

Essential for maintenance, but improper use risks data loss. Pair with tune2fs for features like lazy_itable_init to speed checks.

Always unmount filesystem first (except boot/root).
Can cause data loss on severe corruption.
Time-intensive on large drives (>1TB).
Online resize2fs may allow mounted checks, but risky.

dumpe2fs -h device | grep superblock
Lists backups; use with -b if primary damaged.

e2fsck -f -y /dev/sda1
Force check, auto-yes repair.
e2fsck -n -C 0 /dev/sda1
Read-only check with progress.

Replay journal first with tune2fs -n or debugfs; e2fsck -E journal_only checks only journal.

Part of e2fsprogs suite, created 1993 by Remy Card for ext2fs. Enhanced by Theodore Ts'o for ext3 (2001 journaling) and ext4 (2008 extents/large fs). Widely used in Linux distros; active development continues for modern features like metadata checksums.

fsck(8), fsck.ext4(8), tune2fs(8), mke2fs(8), dumpe2fs(8), debugfs(8), badblocks(8), resize2fs(8)