Skip to main content

e2fsck command man page

E2FSCK(8)     E2FSCK(8)

NAME
       e2fsck - check a Linux ext2/ext3 file system
SYNOPSIS
       e2fsck [ -pacnyrdfkvstDFSV ] [ -b superblock ] [ -B blocksize ] [ -l|-L
       bad_blocks_file ]  [  -C  fd  ] [  -j external-journal   ] [   -E
       extended_options ] device

DESCRIPTION
       e2fsck  is  used to check a Linux second extended file system (ext2fs).
       E2fsck also supports ext2 filesystems containing a journal,  which  are
       also sometimes known as ext3 filesystems, by first applying the journal
       to the filesystem before continuing  with  normal  e2fsck  processing.
       After  the  journal  has been  applied, a filesystem will normally be
       marked as clean. Hence, for ext3 filesystems, e2fsck will normally run
       the  journal  and  exit, unless its superblock indicates that further
       checking is required.

       device is  the  device  file  where  the filesystem  is stored (e.g.
       /dev/hdc1).

       Note  that  in general it is not safe to run e2fsck on mounted filesys-
       tems.  The only exception is if the -n option is specified, and -c, -l,
       or  -L  options are not specified.   However, even if it is safe to do
       so, the results printed by e2fsck are not valid if  the filesystem  is
       mounted.   If e2fsck asks whether or not you should check a filesystem
       which is mounted, the only correct answer is ‘‘no’’.  Only experts  who
       really know what they are doing should consider answering this question
       in any other way.

OPTIONS
       -a     This option does the same thing as the -p option.  It  is  pro-
     vided  for  backwards  compatibility  only; it is suggested that
     people use -p option whenever possible.

       -b superblock
     Instead of using the  normal  superblock,  use  an  alternative
     superblock  specified  by superblock.   This option is normally
     used when the primary superblock has been corrupted.  The loca-
     tion  of the backup superblock is dependent on the filesystem’s
     blocksize.   For filesystems  with  1k  blocksizes,  a backup
     superblock  can  be found at block 8193; for filesystems with 2k
     blocksizes, at block 16384; and  for  4k blocksizes,  at block
     32768.

     Additional  backup  superblocks  can  be determined by using the
     mke2fs program using the -n  option  to print  out  where  the
     superblocks were created.  The -b option to mke2fs, which spec-
     ifies blocksize of the filesystem must be specified in order for
     the superblock locations that are printed out to be accurate.

     If  an alternative superblock is specified and the filesystem is
     not opened read-only, e2fsck will make  sure  that  the  primary
     superblock  is  updated  appropriately  upon  completion of the
     filesystem check.

       -B blocksize
     Normally, e2fsck will search for the superblock at various  dif-
     ferent  block  sizes in an attempt to find the appropriate block
     size.  This search can be fooled in  some cases.  This option
     forces  e2fsck to only try locating the superblock at a particu-
     lar blocksize.  If the superblock is  not  found,  e2fsck  will
     terminate with a fatal error.

       -c     This  option  causes  e2fsck to use badblocks(8) program to do a
     read-only scan of the device in order to find  any  bad  blocks.
     If  any  bad  blocks  are found, they are added to the bad block
     inode to prevent them from being allocated to a file  or direc-
     tory.   If  this option is specified twice, then the bad block
     scan will be done using a non-destructive read-write test.

       -C fd  This option causes e2fsck to write completion information to the
     specified file descriptor so that the progress of the filesystem
     check can be monitored.  This option is typically used  by  pro-
     grams  which  are running e2fsck. If the file descriptor speci-
     fied is 0, e2fsck will print a completion bar as it  goes about
     its  business.   This requires that e2fsck is running on a video
     console or terminal.

       -d     Print  debugging output (useless  unless  you  are   debugging
     e2fsck).

       -D     Optimize directories  in filesystem.  This option causes e2fsck
     to try to optimize all directories, either by reindexing them if
     the  filesystem  supports directory indexing,  or by sorting and
     compressing directories for smaller directories, or for filesys-
     tems using traditional linear directories.

       -E extended_options
     Set  e2fsck  extended options.  Extended options are comma sepa-
     rated, and may take an argument using  the  equals  (’=’) sign.
     The following options are supported:

  ea_ver=extended_attribute_version
 Assume  the  format of the extended attribute blocks
 in the filesystem is the specified  version  number.
 The  version number may  be 1  or 2.  The default
 extended attribute version format is 2.

       -f     Force checking even if the file system seems clean.

       -F     Flush the filesystem device’s buffer  caches  before  beginning.
     Only really useful for doing e2fsck time trials.

       -j external-journal
     Set  the pathname where the external-journal for this filesystem
     can be found.

       -k     When combined with the -c option, any existing bad blocks in the
     bad  blocks  list are preserved, and any new bad blocks found by
     running badblocks(8) will be added to the existing  bad blocks
     list.

       -l filename
     Add  the block numbers listed in the file specified by filename
     to the list of bad blocks.  The format of this file is the  same
     as the one generated by the badblocks(8) program. Note that the
     block numbers are based on  the  blocksize  of  the  filesystem.
     Hence,  badblocks(8) must be given the blocksize of the filesys-
     tem in order to obtain correct results.  As a result, it is much
     simpler  and safer to use the -c option to e2fsck, since it will
     assure that the correct parameters are passed to the  badblocks
     program.

       -L filename
     Set  the bad  blocks list to be the list of blocks specified by
     filename. (This option is the same as the -l option, except the
     bad  blocks list is cleared before the blocks listed in the file
     are added to the bad blocks list.)

       -n     Open the filesystem read-only, and assume an answer of  ‘no’  to
     all  questions. Allows e2fsck to  be used non-interactively.
     (Note: if the -c, -l, or -L options are specified in addition to
     the -n option, then the filesystem will be opened read-write, to
     permit the bad-blocks list to be updated.   However,  no other
     changes will be made to the filesystem.) This option may not be
     specified at the same time as the -p or -y options.

       -p     Automatically repair ("preen") the  file system.  This option
     will  case  e2fsck  to automatically fix any filesystem problems
     that can be safely fixed without human intervention.  If e2fsck
     discovers a  problem which may require the system administrator
     to take  additional  corrective  action, e2fsck will  print  a
     description  of the problem and then exit with the value 4 logi-
     cally or’ed into the exit code.  (See the EXIT  CODE  section.)
     This  option  is normally used by the system’s boot scripts.  It
     may not be specified at the same time as the -n or -y options.

       -r     This option does nothing at all; it is provided only  for back-
     wards compatibility.

       -s     This  option  will  byte-swap the filesystem so that it is using
     the normalized, standard byte-order (which  is  i386  or little
     endian). If  the  filesystem  is already in the standard byte-
     order, e2fsck will take no action.

       -S     This option will byte-swap the  filesystem,  regardless  of  its
     current byte-order.

       -t     Print  timing  statistics for  e2fsck. If this option is used
     twice, additional timing statistics are printed  on  a  pass  by
     pass basis.

       -v     Verbose mode.

       -V     Print version information and exit.

       -y     Assume  an answer of ‘yes’ to all questions; allows e2fsck to be
     used non-interactively.  This option may not be specified at the
     same time as the -n or -p options.

EXIT CODE
       The  exit  code returned  by e2fsck is the sum of the following condi-
       tions:
   0 - No errors
   1 - File system errors corrected
   2 - File system errors corrected, system should
  be rebooted
   4 - File system errors left uncorrected
   8 - Operational error
   16 - Usage or syntax error
   32 - E2fsck canceled by user request
   128 - Shared library error

SIGNALS
       The following signals have the following effect when sent to e2fsck.

       SIGUSR1
     This signal causes e2fsck to start displaying a completion  bar.
     (See discussion of the -C option.)

       SIGUSR2
     This signal causes e2fsck to stop displaying a completion bar.

REPORTING BUGS
       Almost  any  piece of software will have bugs.  If you manage to find a
       filesystem which causes e2fsck to crash, or which e2fsck is  unable  to
       repair, please report it to the author.

       Please  include as  much  information  as possible in your bug report.
       Ideally, include a complete transcript of the e2fsck run, so I can  see
       exactly what  error  messages  are displayed.  (Make sure the messages
       printed by e2fsck are in English; if your system has been configured so
       that  e2fsck’s  messages have  been  translated into another language,
       please set the the LC_ALL environment variable to C so that  the tran-
       script  of  e2fsck’s  output  will  be  useful  to  me.) If you have a
       writable filesystem where the transcript can be stored, the  script(1)
       program is a handy way to save the output of e2fsck to a file.

       It  is  also  useful  to send the output of dumpe2fs(8). If a specific
       inode or inodes seems to be giving  e2fsck  trouble,  try  running  the
       debugfs(8)  command  and send the output of the stat(1u) command run on
       the relevant inode(s).  If the inode is a directory, the debugfs  dump
       command will allow you to extract the contents of the directory inode,
       which can sent to me after being first run  through  uuencode(1).   The
       most useful data you can send to help reproduce the bug is a compressed
       raw image dump of the filesystem, generated using e2image(8).  See  the
       e2image(8) man page for more details.

       Always include the full version string which e2fsck displays when it is
       run, so I know which version you are running.

AUTHOR
       This version of e2fsck was written by Theodore Ts’o <tytso@mit.edu>.

SEE ALSO
       badblocks(8),   dumpe2fs(8),   debugfs(8),    e2image(8),    mke2fs(8),
       tune2fs(8)

E2fsprogs version 1.39   May 2006     E2FSCK(8)

Comments

Popular posts from this blog