TAR(1)\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 GNU TAR Manual\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TAR(1)<\/p>\n
NAME\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
tar – an archiving utility<\/p>\n
SYNOPSIS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
Traditional usage
\ntar {A|c|d|r|t|u|x}[GnSkUWOmpsMBiajJzZhPlRvwo] [ARG…]<\/p>\n
UNIX-style usage
\ntar -A [OPTIONS] ARCHIVE ARCHIVE<\/p>\n
tar -c [-f ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar -d [-f ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar -t [-f ARCHIVE] [OPTIONS] [MEMBER…]<\/p>\n
tar -r [-f ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar -u [-f ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar -x [-f ARCHIVE] [OPTIONS] [MEMBER…]<\/p>\n
GNU-style usage
\ntar {–catenate|–concatenate} [OPTIONS] ARCHIVE ARCHIVE<\/p>\n
tar –create [–file ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar {–diff|–compare} [–file ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar –delete [–file ARCHIVE] [OPTIONS] [MEMBER…]<\/p>\n
tar –append [-f ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar –list [-f ARCHIVE] [OPTIONS] [MEMBER…]<\/p>\n
tar –test-label [–file ARCHIVE] [OPTIONS] [LABEL…]<\/p>\n
tar –update [–file ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar –update [-f ARCHIVE] [OPTIONS] [FILE…]<\/p>\n
tar {–extract|–get} [-f ARCHIVE] [OPTIONS] [MEMBER…]<\/p>\n
NOTE\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
This manpage is a short description of GNU tar.\u00a0 For a detailed
\ndiscussion, including examples and usage recommendations, refer to
\nthe GNU Tar Manual available in texinfo format.\u00a0 If the info reader
\nand the tar documentation are properly installed on your system, the
\ncommand<\/p>\n
info tar<\/p>\n
should give you access to the complete manual.<\/p>\n
You can also view the manual using the info mode in emacs(1), or find
\nit in various formats online at<\/p>\n
http:\/\/www.gnu.org\/software\/tar\/manual<\/p>\n
If any discrepancies occur between this manpage and the GNU Tar
\nManual, the later shall be considered the authoritative source.<\/p>\n
DESCRIPTION\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
GNU tar is an archiving program designed to store multiple files in a
\nsingle file (an archive), and to manipulate such archives.\u00a0 The
\narchive can be either a regular file or a device (e.g. a tape drive,
\nhence the name of the program, which stands for tape archiver), which
\ncan be located either on the local or on a remote machine.<\/p>\n
Option styles
\nOptions to GNU tar can be given in three different styles.\u00a0 In
\ntraditional style, the first argument is a cluster of option letters
\nand all subsequent arguments supply arguments to those options that
\nrequire them.\u00a0 The arguments are read in the same order as the option
\nletters.\u00a0 Any command line words that remain after all options has
\nbeen processed are treated as non-optional arguments: file or archive
\nmember names.<\/p>\n
For example, the c option requires creating the archive, the v option
\nrequests the verbose operation, and the f option takes an argument
\nthat sets the name of the archive to operate upon.\u00a0 The following
\ncommand, written in the traditional style, instructs tar to store all
\nfiles from the directory \/etc into the archive file etc.tar verbosely
\nlisting the files being archived:<\/p>\n
tar cfv a.tar \/etc<\/p>\n
In UNIX or short-option style, each option letter is prefixed with a
\nsingle dash, as in other command line utilities.\u00a0 If an option takes
\nargument, the argument follows it, either as a separate command line
\nword, or immediately following the option.\u00a0 However, if the option
\ntakes an optional argument, the argument must follow the option let\u2010
\nter without any intervening whitespace, as in -g\/tmp\/snar.db.<\/p>\n
Any number of options not taking arguments can be clustered together
\nafter a single dash, e.g. -vkp.\u00a0 Options that take arguments (whether
\nmandatory or optional), can appear at the end of such a cluster, e.g.
\n-vkpf a.tar.<\/p>\n
The example command above written in the short-option style could
\nlook like:<\/p>\n
tar -cvf a.tar \/etc
\nor
\ntar -c -v -f a.tar \/etc<\/p>\n
In GNU or long-option style, each option begins with two dashes and
\nhas a meaningful name, consisting of lower-case letters and dashes.
\nWhen used, the long option can be abbreviated to its initial letters,
\nprovided that this does not create ambiguity.\u00a0 Arguments to long
\noptions are supplied either as a separate command line word, immedi\u2010
\nately following the option, or separated from the option by an equals
\nsign with no intervening whitespace.\u00a0 Optional arguments must always
\nuse the latter method.<\/p>\n
Here are several ways of writing the example command in this style:<\/p>\n
tar –create –file a.tar –verbose \/etc
\nor (abbreviating some options):
\ntar –cre –file=a.tar –verb \/etc<\/p>\n
The options in all three styles can be intermixed, although doing so
\nwith old options is not encouraged.<\/p>\n
Operation mode
\nThe options listed in the table below tell GNU tar what operation it
\nis to perform.\u00a0 Exactly one of them must be given.\u00a0 Meaning of non-
\noptional arguments depends on the operation mode requested.<\/p>\n
-A, –catenate, –concatenate
\nAppend archive to the end of another archive.\u00a0 The arguments
\nare treated as the names of archives to append.\u00a0 All archives
\nmust be of the same format as the archive they are appended
\nto, otherwise the resulting archive might be unusable with
\nnon-GNU implementations of tar.\u00a0 Notice also that when more
\nthan one archive is given, the members from archives other
\nthan the first one will be accessible in the resulting archive
\nonly if using the -i (–ignore-zeros) option.<\/p>\n
Compressed archives cannot be concatenated.<\/p>\n
-c, –create
\nCreate a new archive.\u00a0 Arguments supply the names of the files
\nto be archived.\u00a0 Directories are archived recursively, unless
\nthe –no-recursion option is given.<\/p>\n
-d, –diff, –compare
\nFind differences between archive and file system.\u00a0 The argu\u2010
\nments are optional and specify archive members to compare.\u00a0 If
\nnot given, the current working directory is assumed.<\/p>\n
–delete
\nDelete from the archive.\u00a0 The arguments supply names of the
\narchive members to be removed.\u00a0 At least one argument must be
\ngiven.<\/p>\n
This option does not operate on compressed archives.\u00a0 There is
\nno short option equivalent.<\/p>\n
-r, –append
\nAppend files to the end of an archive.\u00a0 Arguments have the
\nsame meaning as for -c (–create).<\/p>\n
-t, –list
\nList the contents of an archive.\u00a0 Arguments are optional.
\nWhen given, they specify the names of the members to list.<\/p>\n
–test-label
\nTest the archive volume label and exit.\u00a0 When used without
\narguments, it prints the volume label (if any) and exits with
\nstatus 0.\u00a0 When one or more command line arguments are given.
\ntar compares the volume label with each argument.\u00a0 It exits
\nwith code 0 if a match is found, and with code 1 otherwise.
\nNo output is displayed, unless used together with the -v
\n(–verbose) option.<\/p>\n
There is no short option equivalent for this option.<\/p>\n
-u, –update
\nAppend files which are newer than the corresponding copy in
\nthe archive.\u00a0 Arguments have the same meaning as with -c and
\n-r options.<\/p>\n
-x, –extract, –get
\nExtract files from an archive.\u00a0 Arguments are optional.\u00a0 When
\ngiven, they specify names of the archive members to be
\nextracted.<\/p>\n
–show-defaults
\nShow built-in defaults for various tar options and
\nexit.\u00a0 No arguments are allowed.<\/p>\n
-?, –help
\nDisplay a short option summary and exit.\u00a0 No arguments
\nallowed.<\/p>\n
–usage
\nDisplay a list of available options and exit.\u00a0 No argu\u2010
\nments allowed.<\/p>\n
–version
\nPrint program version and copyright information and
\nexit.<\/p>\n
OPTIONS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
Operation modifiers
\n–check-device
\nCheck device numbers when creating incremental archives
\n(default).<\/p>\n
-g, –listed-incremental=FILE
\nHandle new GNU-format incremental backups.\u00a0 FILE is the
\nname of a snapshot file, where tar stores additional
\ninformation which is used to decide which files changed
\nsince the previous incremental dump and, consequently,
\nmust be dumped again.\u00a0 If FILE does not exist when
\ncreating an archive, it will be created and all files
\nwill be added to the resulting archive (the level 0
\ndump).\u00a0 To create incremental archives of non-zero
\nlevel N, create a copy of the snapshot file created
\nduring the level N-1, and use it as FILE.<\/p>\n
When listing or extracting, the actual contents of FILE
\nis not inspected, it is needed only due to syntactical
\nrequirements.\u00a0 It is therefore common practice to use
\n\/dev\/null in its place.<\/p>\n
-G, –incremental
\nHandle old GNU-format incremental backups.<\/p>\n
–ignore-failed-read
\nDo not exit with nonzero on unreadable files.<\/p>\n
–level=NUMBER
\nSet dump level for created listed-incremental archive.
\nCurrently only –level=0 is meaningful: it instructs
\ntar to truncate the snapshot file before dumping,
\nthereby forcing a level 0 dump.<\/p>\n
-n, –seek
\nAssume the archive is seekable.\u00a0 Normally tar
\ndetermines automatically whether the archive can be
\nseeked or not.\u00a0 This option is intended for use in
\ncases when such recognition fails.\u00a0 It takes effect
\nonly if the archive is open for reading (e.g. with
\n–list or –extract options).<\/p>\n
–no-check-device
\nDo not check device numbers when creating incremental
\narchives.<\/p>\n
–no-seek
\nAssume the archive is not seekable.<\/p>\n
–occurrence[=N]
\nProcess only the Nth occurrence of each file in the
\narchive.\u00a0 This option is valid only when used with one
\nof the following subcommands: –delete, –diff,
\n–extract or –list and when a list of files is given
\neither on the command line or via the -T option.\u00a0 The
\ndefault N is 1.<\/p>\n
–restrict
\nDisable the use of some potentially harmful options.<\/p>\n
–sparse-version=MAJOR[.MINOR]
\nSet version of the sparse format to use (implies
\n–sparse).\u00a0 This option implies –sparse.\u00a0 Valid
\nargument values are 0.0, 0.1, and 1.0.\u00a0 For a detailed
\ndiscussion of sparse formats, refer to the GNU Tar
\nManual, appendix D, “Sparse Formats”.\u00a0 Using info
\nreader, it can be accessed running the following
\ncommand: info tar ‘Sparse Formats’.<\/p>\n
-S, –sparse
\nHandle sparse files efficiently.\u00a0 Some files in the
\nfile system may have segments which were actually never
\nwritten (quite often these are database files created
\nby such systems as DBM).\u00a0 When given this option, tar
\nattempts to determine if the file is sparse prior to
\narchiving it, and if so, to reduce the resulting
\narchive size by not dumping empty parts of the file.<\/p>\n
Overwrite control
\nThese options control tar actions when extracting a file over
\nan existing copy on disk.<\/p>\n
-k, –keep-old-files
\nDon’t replace existing files when extracting.<\/p>\n
–keep-newer-files
\nDon’t replace existing files that are newer than their
\narchive copies.<\/p>\n
–no-overwrite-dir
\nPreserve metadata of existing directories.<\/p>\n
–one-top-level[=DIR]
\nExtract all files into DIR, or, if used without
\nargument, into a subdirectory named by the base name of
\nthe archive (minus standard compression suffixes
\nrecognizable by –auto-compress).<\/p>\n
–overwrite
\nOverwrite existing files when extracting.<\/p>\n
–overwrite-dir
\nOverwrite metadata of existing directories when
\nextracting (default).<\/p>\n
–recursive-unlink
\nRecursively remove all files in the directory prior to
\nextracting it.<\/p>\n
–remove-files
\nRemove files from disk after adding them to the
\narchive.<\/p>\n
–skip-old-files
\nDon’t replace existing files when extracting, silently
\nskip over them.<\/p>\n
-U, –unlink-first
\nRemove each file prior to extracting over it.<\/p>\n
-W, –verify
\nVerify the archive after writing it.<\/p>\n
Output stream selection
\n–ignore-command-error<\/p>\n
Ignore subprocess exit codes.<\/p>\n
–no-ignore-command-error
\nTreat non-zero exit codes of children as error
\n(default).<\/p>\n
-O, –to-stdout
\nExtract files to standard output.<\/p>\n
–to-command=COMMAND
\nPipe extracted files to COMMAND.\u00a0 The argument is the
\npathname of an external program, optionally with
\ncommand line arguments.\u00a0 The program will be invoked
\nand the contents of the file being extracted supplied
\nto it on its standard output.\u00a0 Additional data will be
\nsupplied via the following environment variables:<\/p>\n
TAR_FILETYPE
\nType of the file. It is a single letter with the
\nfollowing meaning:<\/p>\n
f\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Regular file
\nd\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Directory
\nl\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Symbolic link
\nh\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Hard link
\nb\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Block device
\nc\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Character device<\/p>\n
Currently only regular files are supported.<\/p>\n
TAR_MODE
\nFile mode, an octal number.<\/p>\n
TAR_FILENAME
\nThe name of the file.<\/p>\n
TAR_REALNAME
\nName of the file as stored in the archive.<\/p>\n
TAR_UNAME
\nName of the file owner.<\/p>\n
TAR_GNAME
\nName of the file owner group.<\/p>\n
TAR_ATIME
\nTime of last access. It is a decimal number,
\nrepresenting seconds since the Epoch.\u00a0 If the
\narchive provides times with nanosecond
\nprecision, the nanoseconds are appended to the
\ntimestamp after a decimal point.<\/p>\n
TAR_MTIME
\nTime of last modification.<\/p>\n
TAR_CTIME
\nTime of last status change.<\/p>\n
TAR_SIZE
\nSize of the file.<\/p>\n
TAR_UID
\nUID of the file owner.<\/p>\n
TAR_GID
\nGID of the file owner.<\/p>\n
Additionally, the following variables contain
\ninformation about tar operation mode and the archive
\nbeing processed:<\/p>\n
TAR_VERSION
\nGNU tar version number.<\/p>\n
TAR_ARCHIVE
\nThe name of the archive tar is processing.<\/p>\n
TAR_BLOCKING_FACTOR
\nCurrent blocking factor, i.e. number of 512-byte
\nblocks in a record.<\/p>\n
TAR_VOLUME
\nOrdinal number of the volume tar is processing
\n(set if reading a multi-volume archive).<\/p>\n
TAR_FORMAT
\nFormat of the archive being processed.\u00a0 One of:
\ngnu, oldgnu, posix, ustar, v7.\u00a0 TAR_SUBCOMMAND A
\nshort option (with a leading dash) describing
\nthe operation tar is executing.<\/p>\n
Handling of file attributes
\n–atime-preserve[=METHOD]
\nPreserve access times on dumped files, either by
\nrestoring the times after reading (METHOD=replace, this
\nis the default) or by not setting the times in the
\nfirst place (METHOD=system)<\/p>\n
–delay-directory-restore
\nDelay setting modification times and permissions of
\nextracted directories until the end of extraction.\u00a0 Use
\nthis option when extracting from an archive which has
\nunusual member ordering.<\/p>\n
–group=NAME
\nForce NAME as group for added files.<\/p>\n
–mode=CHANGES
\nForce symbolic mode CHANGES for added files.<\/p>\n
–mtime=DATE-OR-FILE
\nSet mtime for added files.\u00a0 DATE-OR-FILE is either a
\ndate\/time in almost arbitrary formate, or the name of
\nan existing file.\u00a0 In the latter case the mtime of that
\nfile will be used.<\/p>\n
-m, –touch
\nDon’t extract file modified time.<\/p>\n
–no-delay-directory-restore
\nCancel the effect of the prior
\n–delay-directory-restore option.<\/p>\n
–no-same-owner
\nExtract files as yourself (default for ordinary users).<\/p>\n
–no-same-permissions
\nApply the user’s umask when extracting permissions from
\nthe archive (default for ordinary users).<\/p>\n
–numeric-owner
\nAlways use numbers for user\/group names.<\/p>\n
–owner=NAME
\nForce NAME as owner for added files.<\/p>\n
-p, –preserve-permissions, –same-permissions
\nextract information about file permissions (default for
\nsuperuser)<\/p>\n
–preserve
\nSame as both -p and -s.<\/p>\n
–same-owner
\nTry extracting files with the same ownership as exists
\nin the archive (default for superuser).<\/p>\n
-s, –preserve-order, –same-order
\nSort names to extract to match archive<\/p>\n
–sort=ORDER
\nWhen creating an archive, sort directory entries
\naccording to ORDER, which is one of none, name, or
\ninode.<\/p>\n
The default is –sort=none, which stores archive
\nmembers in the same order as returned by the operating
\nsystem.<\/p>\n
Using –sort=name ensures the member ordering in the
\ncreated archive is uniform and reproducible.<\/p>\n
Using –sort=inode reduces the number of disk seeks
\nmade when creating the archive and thus can
\nconsiderably speed up archivation.\u00a0 This sorting order
\nis supported only if the underlying system provides the
\nnecessary information.<\/p>\n
Device selection and switching
\n-f, –file=ARCHIVE
\nUse archive file or device ARCHIVE.\u00a0 If this option is
\nnot given, tar will first examine the environment
\nvariable `TAPE’.\u00a0 If it is set, its value will be used
\nas the archive name.\u00a0 Otherwise, tar will assume the
\ncompiled-in default.\u00a0 The default value can be
\ninspected either using the –show-defaults option, or
\nat the end of the tar –help output.<\/p>\n
An archive name that has a colon in it specifies a file
\nor device on a remote machine.\u00a0 The part before the
\ncolon is taken as the machine name or IP address, and
\nthe part after it as the file or device pathname, e.g.:<\/p>\n
–file=remotehost:\/dev\/sr0<\/p>\n
An optional username can be prefixed to the hostname,
\nplacing a @ sign between them.<\/p>\n
By default, the remote host is accessed via the rsh(1)
\ncommand.\u00a0 Nowadays it is common to use ssh(1) instead.
\nYou can do so by giving the following command line
\noption:<\/p>\n
–rsh-command=\/usr\/bin\/ssh<\/p>\n
The remote machine should have the rmt(8) command
\ninstalled.\u00a0 If its pathname does not match tar’s
\ndefault, you can inform tar about the correct pathname
\nusing the –rmt-command option.<\/p>\n
–force-local
\nArchive file is local even if it has a colon.<\/p>\n
-F, –info-script=COMMAND, –new-volume-script=COMMAND
\nRun COMMAND at the end of each tape (implies -M).\u00a0 The
\ncommand can include arguments.\u00a0 When started, it will
\ninherit tar’s environment plus the following variables:<\/p>\n
TAR_VERSION
\nGNU tar version number.<\/p>\n
TAR_ARCHIVE
\nThe name of the archive tar is processing.<\/p>\n
TAR_BLOCKING_FACTOR
\nCurrent blocking factor, i.e. number of 512-byte
\nblocks in a record.<\/p>\n
TAR_VOLUME
\nOrdinal number of the volume tar is processing
\n(set if reading a multi-volume archive).<\/p>\n
TAR_FORMAT
\nFormat of the archive being processed.\u00a0 One of:
\ngnu, oldgnu, posix, ustar, v7.<\/p>\n
TAR_SUBCOMMAND
\nA short option (with a leading dash) describing
\nthe operation tar is executing.<\/p>\n
TAR_FD File descriptor which can be used to communicate
\nthe new volume name to tar.<\/p>\n
If the info script fails, tar exits; otherwise, it
\nbegins writing the next volume.<\/p>\n
-L, –tape-length=
\nChange tape after writing Nx1024 bytes.\u00a0 If N is fol\u2010
\nlowed by a size suffix (see the subsection Size suf\u2010
\nfixes below), the suffix specifies the multiplicative
\nfactor to be used instead of 1024.<\/p>\n
This option implies -M.<\/p>\n
-M, –multi-volume
\nCreate\/list\/extract multi-volume archive.<\/p>\n
–rmt-command=COMMAND
\nUse COMMAND instead of rmt when accessing remote ar\u2010
\nchives.\u00a0 See the description of the -f option, above.<\/p>\n
–rsh-command=COMMAND
\nUse COMMAND instead of rsh when accessing remote ar\u2010
\nchives.\u00a0 See the description of the -f option, above.<\/p>\n
–volno-file=FILE
\nWhen this option is used in conjunction with
\n–multi-volume, tar will keep track of which volume of
\na multi-volume archive it is working in FILE.<\/p>\n
Device blocking
\n-b, –blocking-factor=BLOCKS
\nSet record size to BLOCKSx512 bytes.<\/p>\n
-B, –read-full-records
\nWhen listing or extracting, accept incomplete input
\nrecords after end-of-file marker.<\/p>\n
-i, –ignore-zeros
\nIgnore zeroed blocks in archive.\u00a0 Normally two consecu\u2010
\ntive 512-blocks filled with zeroes mean EOF and tar
\nstops reading after encountering them.\u00a0 This option
\ninstructs it to read further and is useful when reading
\narchives created with the -A option.<\/p>\n
–record-size=NUMBER
\nSet record size.\u00a0 NUMBER is the number of bytes per
\nrecord.\u00a0 It must be multiple of 512.\u00a0 It can can be
\nsuffixed with a size suffix, e.g. –record-size=10K,
\nfor 10 Kilobytes.\u00a0 See the subsection Size suffixes,
\nfor a list of valid suffixes.<\/p>\n
Archive format selection
\n-H, –format=FORMAT
\nCreate archive of the given format.\u00a0 Valid formats are:<\/p>\n
gnu\u00a0\u00a0\u00a0 GNU tar 1.13.x format<\/p>\n
oldgnu GNU format as per tar <= 1.12.<\/p>\n
pax, posix
\nPOSIX 1003.1-2001 (pax) format.<\/p>\n
ustar\u00a0 POSIX 1003.1-1988 (ustar) format.<\/p>\n
v7\u00a0\u00a0\u00a0\u00a0 Old V7 tar format.<\/p>\n
–old-archive, –portability
\nSame as –format=v7.<\/p>\n
–pax-option=keyword[[:]=value][,keyword[[:]=value]]…
\nControl pax keywords when creating PAX archives (-H
\npax).\u00a0 This option is equivalent to the -o option of
\nthe pax(1)utility.<\/p>\n
–posix
\nSame as –format=posix.<\/p>\n
-V, –label=TEXT
\nCreate archive with volume name TEXT.\u00a0 If listing or
\nextracting, use TEXT as a globbing pattern for volume
\nname.<\/p>\n
Compression options
\n-a, –auto-compress
\nUse archive suffix to determine the compression pro\u2010
\ngram.<\/p>\n
-I, –use-compress-program=COMMAND
\nFilter data through COMMAND.\u00a0 It must accept the -d
\noption, for decompression.\u00a0 The argument can contain
\ncommand line options.<\/p>\n
-j, –bzip2
\nFilter the archive through bzip2(1).<\/p>\n
-J, –xz
\nFilter the archive through xz(1).<\/p>\n
–lzip Filter the archive through lzip(1).<\/p>\n
–lzma Filter the archive through lzma(1).<\/p>\n
–lzop Filter the archive through lzop(1).<\/p>\n
–no-auto-compress
\nDo not use archive suffix to determine the compression
\nprogram.<\/p>\n
-z, –gzip, –gunzip, –ungzip
\nFilter the archive through gzip(1).<\/p>\n
-Z, –compress, –uncompress
\nFilter the archive through compress(1).<\/p>\n
Local file selection
\n–add-file=FILE
\nAdd FILE to the archive (useful if its name starts with
\na dash).<\/p>\n
–backup[=CONTROL]
\nBackup before removal.\u00a0 The CONTROL argument, if sup\u2010
\nplied, controls the backup policy.\u00a0 Its valid values
\nare:<\/p>\n
none, off
\nNever make backups.<\/p>\n
t, numbered
\nMake numbered backups.<\/p>\n
nil, existing
\nMake numbered backups if numbered backups exist,
\nsimple backups otherwise.<\/p>\n
never, simple
\nAlways make simple backups<\/p>\n
If CONTROL is not given, the value is taken from the
\nVERSION_CONTROL environment variable.\u00a0 If it is not
\nset, existing is assumed.<\/p>\n
-C, –directory=DIR
\nChange to directory DIR.<\/p>\n
–exclude=PATTERN
\nExclude files matching PATTERN, a glob(3)-style wild\u2010
\ncard pattern.<\/p>\n
–exclude-backups
\nExclude backup and lock files.<\/p>\n
–exclude-caches
\nExclude contents of directories containing file
\nCACHEDIR.TAG, except for the tag file itself.<\/p>\n
–exclude-caches-all
\nExclude directories containing file CACHEDIR.TAG and
\nthe file itself.<\/p>\n
–exclude-caches-under
\nExclude everything under directories containing
\nCACHEDIR.TAG<\/p>\n
–exclude-ignore=FILE
\nBefore dumping a directory, see if it contains FILE.
\nIf so, read exclusion patterns from this file.\u00a0 The
\npatterns affect only the directory itself.<\/p>\n
–exclude-ignore-recursive=FILE
\nSame as –exclude-ignore, except that patterns from
\nFILE affect both the directory and all its subdirecto\u2010
\nries.<\/p>\n
–exclude-tag=FILE
\nExclude contents of directories containing FILE, except
\nfor FILE itself.<\/p>\n
–exclude-tag-all=FILE
\nExclude directories containing FILE.<\/p>\n
–exclude-tag-under=FILE
\nExclude everything under directories containing FILE.<\/p>\n
–exclude-vcs
\nExclude version control system directories.<\/p>\n
–exclude-vcs-ignores
\nExclude files that match patterns read from VCS-spe\u2010
\ncific ignore files.\u00a0 Supported files are: .cvsignore,
\n.gitignore, .bzrignore, and .hgignore.<\/p>\n
-h, –dereference
\nFollow symlinks; archive and dump the files they point
\nto.<\/p>\n
–hard-dereference
\nFollow hard links; archive and dump the files they
\nrefer to.<\/p>\n
-K, –starting-file=MEMBER
\nBegin at the given member in the archive.<\/p>\n
–newer-mtime=DATE
\nWork on files whose data changed after the DATE.\u00a0 If
\nDATE starts with \/ or . it is taken to be a file name;
\nthe mtime of that file is used as the date.<\/p>\n
–no-null
\nDisable the effect of the previous –null option.<\/p>\n
–no-recursion
\nAvoid descending automatically in directories.<\/p>\n
–no-unquote
\nDo not unquote input file or member names.<\/p>\n
–null Instruct subsequent -T options to read null-terminated
\nnames, disable handling of the -C option read from the
\nfile.<\/p>\n
-N, –newer=DATE, –after-date=DATE
\nOnly store files newer than DATE.\u00a0 If DATE starts with
\n\/ or . it is taken to be a file name; the ctime of that
\nfile is used as the date.<\/p>\n
–one-file-system
\nStay in local file system when creating archive.<\/p>\n
-P, –absolute-names
\nDon’t strip leading slashes from file names when creat\u2010
\ning archives.<\/p>\n
–recursion
\nRecurse into directories (default).<\/p>\n
–suffix=STRING
\nBackup before removal, override usual suffix.\u00a0 Default
\nsuffix is ~, unless overridden by environment variable
\nSIMPLE_BACKUP_SUFFIX.<\/p>\n
-T, –files-from=FILE
\nGet names to extract or create from FILE.<\/p>\n
–unquote
\nUnquote file or member names (default).<\/p>\n
-X, –exclude-from=FILE
\nExclude files matching patterns listed in FILE.<\/p>\n
File name transformations
\n–strip-components=NUMBER
\nStrip NUMBER leading components from file names on
\nextraction.<\/p>\n
–transform=EXPRESSION, –xform=EXPRESSION
\nUse sed replace EXPRESSION to transform file names.<\/p>\n
File name matching options
\nThese options affect both exclude and include patterns.<\/p>\n
–anchored
\nPatterns match file name start.<\/p>\n
–ignore-case
\nIgnore case.<\/p>\n
–no-anchored
\nPatterns match after any \/ (default for exclusion).<\/p>\n
–no-ignore-case
\nCase sensitive matching (default).<\/p>\n
–no-wildcards
\nVerbatim string matching.<\/p>\n
–no-wildcards-match-slash
\nWildcards do not match \/.<\/p>\n
–wildcards
\nUse wildcards (default for exclusion).<\/p>\n
–wildcards-match-slash
\nWildcards match \/ (default for exclusion).<\/p>\n
Informative output
\n–checkpoint[=N]
\nDisplay progress messages every Nth record (default
\n10).<\/p>\n
–checkpoint-action=ACTION
\nRun ACTION on each checkpoint.<\/p>\n
–full-time
\nPrint file time to its full resolution.<\/p>\n
–index-file=FILE
\nSend verbose output to FILE.<\/p>\n
-l, –check-links
\nPrint a message if not all links are dumped.<\/p>\n
–no-quote-chars=STRING
\nDisable quoting for characters from STRING.<\/p>\n
–quote-chars=STRING
\nAdditionally quote characters from STRING.<\/p>\n
–quoting-style=STYLE
\nSet quoting style for file and member names.\u00a0 Valid
\nvalues for STYLE are literal, shell, shell-always, c,
\nc-maybe, escape, locale, clocale.<\/p>\n
-R, –block-number
\nShow block number within archive with each message.<\/p>\n
–show-omitted-dirs
\nWhen listing or extracting, list each directory that
\ndoes not match search criteria.<\/p>\n
–show-transformed-names, –show-stored-names
\nShow file or archive names after transformation by
\n–strip and –transform options.<\/p>\n
–totals[=SIGNAL]
\nPrint total bytes after processing the archive.\u00a0 If
\nSIGNAL is given, print total bytes when this signal is
\ndelivered.\u00a0 Allowed signals are: SIGHUP, SIGQUIT, SIG\u2010
\nINT, SIGUSR1, and SIGUSR2.\u00a0 The SIG prefix can be omit\u2010
\nted.<\/p>\n
–utc\u00a0 Print file modification times in UTC.<\/p>\n
-v, –verbose
\nVerbosely list files processed.<\/p>\n
–warning=KEYWORD
\nEnable or disable warning messages identified by KEY\u2010
\nWORD.\u00a0 The messages are suppressed if KEYWORD is pre\u2010
\nfixed with no- and enabled otherwise.<\/p>\n
Multiple –warning messages accumulate.<\/p>\n
Keywords controlling general tar operation:<\/p>\n
all\u00a0\u00a0\u00a0 Enable all warning messages.\u00a0 This is the
\ndefault.<\/p>\n
none\u00a0\u00a0 Disable all warning messages.<\/p>\n
filename-with-nuls
\n“%s: file name read contains nul character”<\/p>\n
alone-zero-block
\n“A lone zero block at %s”<\/p>\n
Keywords applicable for tar –create:<\/p>\n
cachedir
\n“%s: contains a cache directory tag %s; %s”<\/p>\n
file-shrank
\n“%s: File shrank by %s bytes; padding with
\nzeros”<\/p>\n
xdev\u00a0\u00a0 “%s: file is on a different filesystem; not
\ndumped”<\/p>\n
file-ignored
\n“%s: Unknown file type; file ignored”
\n“%s: socket ignored”
\n“%s: door ignored”<\/p>\n
file-unchanged
\n“%s: file is unchanged; not dumped”<\/p>\n
ignore-archive
\n“%s: file is the archive; not dumped”<\/p>\n
file-removed
\n“%s: File removed before we read it”<\/p>\n
file-changed
\n“%s: file changed as we read it”<\/p>\n
Keywords applicable for tar –extract:<\/p>\n
existing-file
\n“%s: skipping existing file”<\/p>\n
timestamp
\n“%s: implausibly old time stamp %s”
\n“%s: time stamp %s is %s s in the future”<\/p>\n
contiguous-cast
\n“Extracting contiguous files as regular files”<\/p>\n
symlink-cast
\n“Attempting extraction of symbolic links as hard
\nlinks”<\/p>\n
unknown-cast
\n“%s: Unknown file type ‘%c’, extracted as normal
\nfile”<\/p>\n
ignore-newer
\n“Current %s is newer or same age”<\/p>\n
unknown-keyword
\n“Ignoring unknown extended header keyword ‘%s'”<\/p>\n
decompress-program
\nControls verbose description of failures occur\u2010
\nring when trying to run alternative decompressor
\nprograms.\u00a0 This warning is disabled by default
\n(unless –verbose is used).\u00a0 A common example of
\nwhat you can get when using this warning is:<\/p>\n
$ tar –warning=decompress-program -x -f archive.Z
\ntar (child): cannot run compress: No such file or directory
\ntar (child): trying gzip<\/p>\n
This means that tar first tried to decompress
\narchive.Z using compress, and, when that failed,
\nswitched to gzip.<\/p>\n
record-size
\n“Record size = %lu blocks”<\/p>\n
Keywords controlling incremental extraction:<\/p>\n
rename-directory
\n“%s: Directory has been renamed from %s”
\n“%s: Directory has been renamed”<\/p>\n
new-directory
\n“%s: Directory is new”<\/p>\n
xdev\u00a0\u00a0 “%s: directory is on a different device: not
\npurging”<\/p>\n
bad-dumpdir
\n“Malformed dumpdir: ‘X’ never used”<\/p>\n
-w, –interactive, –confirmation
\nAsk for confirmation for every action.<\/p>\n
Compatibility options
\n-o\u00a0\u00a0\u00a0\u00a0 When creating, same as –old-archive.\u00a0 When extracting,
\nsame as –no-same-owner.<\/p>\n
Size suffixes
\nSuffix\u00a0\u00a0\u00a0 Units\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Byte Equivalent
\nb\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Blocks\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 512
\nB\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Kilobytes\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 1024
\nc\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Bytes\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE
\nG\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Gigabytes\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 1024^3
\nK\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Kilobytes\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 1024
\nk\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Kilobytes\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 1024
\nM\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Megabytes\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 1024^2
\nP\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Petabytes\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 1024^5
\nT\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Terabytes\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 1024^4
\nw\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Words\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE x 2<\/p>\n
RETURN VALUE\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
Tar exit code indicates whether it was able to successfully
\nperform the requested operation, and if not, what kind of
\nerror occurred.<\/p>\n
0\u00a0\u00a0\u00a0\u00a0\u00a0 Successful termination.<\/p>\n
1\u00a0\u00a0\u00a0\u00a0\u00a0 Some files differ.\u00a0 If tar was invoked with the
\n–compare (–diff, -d) command line option, this means
\nthat some files in the archive differ from their disk
\ncounterparts.\u00a0 If tar was given one of the –create,
\n–append or –update options, this exit code means that
\nsome files were changed while being archived and so the
\nresulting archive does not contain the exact copy of
\nthe file set.<\/p>\n
2\u00a0\u00a0\u00a0\u00a0\u00a0 Fatal error.\u00a0 This means that some fatal, unrecoverable
\nerror occurred.<\/p>\n
If a subprocess that had been invoked by tar exited with a
\nnonzero exit code, tar itself exits with that code as well.
\nThis can happen, for example, if a compression option (e.g.
\n-z) was used and the external compressor program failed.
\nAnother example is rmt failure during backup to a remote
\ndevice.<\/p>\n
SEE ALSO\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
bzip2(1), compress(1), gzip(1), lzma(1), lzop(1), rmt(8),
\nsymlink(7), tar(5), xz(1).<\/p>\n
Complete tar manual: run info tar or use emacs(1) info mode to
\nread it.<\/p>\n
Online copies of GNU tar documentation in various formats can
\nbe found at:<\/p>\n
http:\/\/www.gnu.org\/software\/tar\/manual<\/p>\n
BUG REPORTS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
Report bugs to <bug-tar@gnu.org>.<\/p>\n
COPYRIGHT\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
Copyright \u00a9 2013 Free Software Foundation, Inc.
\nLicense GPLv3+: GNU GPL version 3 or later
\n<http:\/\/gnu.org\/licenses\/gpl.html>
\nThis is free software: you are free to change and redistribute
\nit.\u00a0 There is NO WARRANTY, to the extent permitted by law.<\/p>\n
COLOPHON\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top<\/p>\n
This page is part of the tar (an archiver program) project.
\nInformation about the project can be found at
\n\u27e8http:\/\/savannah.gnu.org\/projects\/tar\/\u27e9.\u00a0 If you have a bug
\nreport for this manual page, see
\n\u27e8http:\/\/savannah.gnu.org\/bugs\/?group=tar\u27e9.\u00a0 This page was
\nobtained from the project’s upstream Git repository
\n(git:\/\/git.savannah.gnu.org\/tar.git) on 2014-12-30.\u00a0 If you
\ndiscover any rendering problems in this HTML version of the
\npage, or you believe there is a better or more up-to-date
\nsource for the page, or you have corrections or improvements
\nto the information in this COLOPHON (which is not part of the
\noriginal manual page), send a mail to\u00a0man-pages@man7.org<\/p>\n
TAR\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 August 16, 2014\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TAR(1)<\/p>\n","protected":false},"excerpt":{"rendered":"
TAR(1)\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 GNU TAR Manual\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TAR(1) NAME\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top tar – an archiving utility SYNOPSIS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 top Traditional usage tar {A|c|d|r|t|u|x}[GnSkUWOmpsMBiajJzZhPlRvwo] [ARG…] UNIX-style usage tar -A [OPTIONS] ARCHIVE ARCHIVE tar -c [-f ARCHIVE] [OPTIONS] [FILE…] tar -d [-f ARCHIVE] [OPTIONS] [FILE…] tar -t [-f ARCHIVE] [OPTIONS] [MEMBER…] tar -r [-f ARCHIVE] [OPTIONS] [FILE…] tar -u [-f ARCHIVE] [OPTIONS] […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_sitemap_exclude":false,"_sitemap_priority":"","_sitemap_frequency":"","footnotes":""},"categories":[8],"tags":[],"class_list":["post-365","post","type-post","status-publish","format-standard","hentry","category-shell"],"a3_pvc":{"activated":false,"total_views":0,"today_views":0},"_links":{"self":[{"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/posts\/365","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=365"}],"version-history":[{"count":0,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/posts\/365\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=365"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=365"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=365"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}