1 \input texinfo @c -*-texinfo-*-
2 @comment %**start of header
5 @settitle GNU tar @value{VERSION}
14 @c 1. Pay attention to @FIXME{}s and @UNREVISED{}s
15 @c 2. Before creating final variant:
16 @c 1.1. Run `make check-options' to make sure all options are properly
18 @c 2.1. Run `make master-menu' (see comment before the master menu).
20 @include rendition.texi
23 @c Put everything in one index (arbitrarily chosen to be the concept index).
33 This manual is for @acronym{GNU} @command{tar} (version
34 @value{VERSION}, @value{UPDATED}), which creates and extracts files
37 Copyright @copyright{} 1992, 1994, 1995, 1996, 1997, 1999, 2000, 2001,
38 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
41 Permission is granted to copy, distribute and/or modify this document
42 under the terms of the GNU Free Documentation License, Version 1.1 or
43 any later version published by the Free Software Foundation; with no
44 Invariant Sections, with the Front-Cover Texts being ``A GNU Manual,''
45 and with the Back-Cover Texts as in (a) below. A copy of the license
46 is included in the section entitled "GNU Free Documentation License".
48 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
49 this GNU Manual. Buying copies from GNU Press supports the FSF in
50 developing GNU and promoting software freedom.''
54 @dircategory Archiving
56 * Tar: (tar). Making tape (or disk) archives.
59 @dircategory Individual utilities
61 * tar: (tar)tar invocation. Invoking @GNUTAR{}.
64 @shorttitlepage @acronym{GNU} @command{tar}
67 @title @acronym{GNU} tar: an archiver tool
68 @subtitle @value{RENDITION} @value{VERSION}, @value{UPDATED}
69 @author John Gilmore, Jay Fenlason et al.
72 @vskip 0pt plus 1filll
78 @top @acronym{GNU} tar: an archiver tool
83 @cindex archiving files
85 The first part of this master menu lists the major nodes in this Info
86 document. The rest of the menu lists all the lower level nodes.
89 @c The master menu goes here.
91 @c NOTE: To update it from within Emacs, make sure mastermenu.el is
92 @c loaded and run texinfo-master-menu.
93 @c To update it from the command line, run
104 * Date input formats::
111 * Configuring Help Summary::
114 * Free Software Needs Free Documentation::
115 * Copying This Manual::
116 * Index of Command Line Options::
120 --- The Detailed Node Listing ---
124 * Book Contents:: What this Book Contains
125 * Definitions:: Some Definitions
126 * What tar Does:: What @command{tar} Does
127 * Naming tar Archives:: How @command{tar} Archives are Named
128 * Authors:: @GNUTAR{} Authors
129 * Reports:: Reporting bugs or suggestions
131 Tutorial Introduction to @command{tar}
134 * stylistic conventions::
135 * basic tar options:: Basic @command{tar} Operations and Options
136 * frequent operations::
137 * Two Frequent Options::
138 * create:: How to Create Archives
139 * list:: How to List Archives
140 * extract:: How to Extract Members from an Archive
143 Two Frequently Used Options
149 How to Create Archives
151 * prepare for examples::
152 * Creating the archive::
161 How to Extract Members from an Archive
163 * extracting archives::
166 * extracting untrusted archives::
172 * using tar options::
180 The Three Option Styles
182 * Mnemonic Options:: Mnemonic Option Style
183 * Short Options:: Short Option Style
184 * Old Options:: Old Option Style
185 * Mixing:: Mixing Option Styles
187 All @command{tar} Options
189 * Operation Summary::
191 * Short Option Summary::
203 Advanced @GNUTAR{} Operations
212 How to Add Files to Existing Archives: @option{--append}
214 * appending files:: Appending Files to an Archive
221 Options Used by @option{--create}
223 * Ignore Failed Read::
225 Options Used by @option{--extract}
227 * Reading:: Options to Help Read Archives
228 * Writing:: Changing How @command{tar} Writes Files
229 * Scarce:: Coping with Scarce Resources
231 Options to Help Read Archives
233 * read full records::
236 Changing How @command{tar} Writes Files
238 * Dealing with Old Files::
239 * Overwrite Old Files::
244 * Data Modification Times::
245 * Setting Access Permissions::
246 * Directory Modification Times and Permissions::
247 * Writing to Standard Output::
248 * Writing to an External Program::
251 Coping with Scarce Resources
256 Performing Backups and Restoring Files
258 * Full Dumps:: Using @command{tar} to Perform Full Dumps
259 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
260 * Backup Levels:: Levels of Backups
261 * Backup Parameters:: Setting Parameters for Backups and Restoration
262 * Scripted Backups:: Using the Backup Scripts
263 * Scripted Restoration:: Using the Restore Script
265 Setting Parameters for Backups and Restoration
267 * General-Purpose Variables::
268 * Magnetic Tape Control::
270 * backup-specs example:: An Example Text of @file{Backup-specs}
272 Choosing Files and Names for @command{tar}
274 * file:: Choosing the Archive's Name
275 * Selecting Archive Members::
276 * files:: Reading Names from a File
277 * exclude:: Excluding Some Files
278 * wildcards:: Wildcards Patterns and Matching
279 * quoting styles:: Ways of Quoting Special Characters in Names
280 * transform:: Modifying File and Member Names
281 * after:: Operating Only on New Files
282 * recurse:: Descending into Directories
283 * one:: Crossing File System Boundaries
285 Reading Names from a File
291 * problems with exclude::
293 Wildcards Patterns and Matching
295 * controlling pattern-matching::
297 Crossing File System Boundaries
299 * directory:: Changing Directory
300 * absolute:: Absolute File Names
304 * General date syntax:: Common rules.
305 * Calendar date items:: 19 Dec 1994.
306 * Time of day items:: 9:20pm.
307 * Time zone items:: @sc{est}, @sc{pdt}, @sc{gmt}.
308 * Day of week items:: Monday and others.
309 * Relative items in date strings:: next tuesday, 2 years ago.
310 * Pure numbers in date strings:: 19931219, 1440.
311 * Seconds since the Epoch:: @@1078100502.
312 * Specifying time zone rules:: TZ="America/New_York", TZ="UTC0".
313 * Authors of get_date:: Bellovin, Eggert, Salz, Berets, et al.
315 Controlling the Archive Format
317 * Portability:: Making @command{tar} Archives More Portable
318 * Compression:: Using Less Space through Compression
319 * Attributes:: Handling File Attributes
320 * cpio:: Comparison of @command{tar} and @command{cpio}
322 Making @command{tar} Archives More Portable
324 * Portable Names:: Portable Names
325 * dereference:: Symbolic Links
326 * old:: Old V7 Archives
327 * ustar:: Ustar Archives
328 * gnu:: GNU and old GNU format archives.
329 * posix:: @acronym{POSIX} archives
330 * Checksumming:: Checksumming Problems
331 * Large or Negative Values:: Large files, negative time stamps, etc.
333 @GNUTAR{} and @acronym{POSIX} @command{tar}
335 * PAX keywords:: Controlling Extended Header Keywords.
337 Using Less Space through Compression
339 * gzip:: Creating and Reading Compressed Archives
340 * sparse:: Archiving Sparse Files
342 Tapes and Other Archive Media
344 * Device:: Device selection and switching
345 * Remote Tape Server::
346 * Common Problems and Solutions::
347 * Blocking:: Blocking
348 * Many:: Many archives on one tape
349 * Using Multiple Tapes:: Using Multiple Tapes
350 * label:: Including a Label in the Archive
356 * Format Variations:: Format Variations
357 * Blocking Factor:: The Blocking Factor of an Archive
359 Many Archives on One Tape
361 * Tape Positioning:: Tape Positions and Tape Marks
362 * mt:: The @command{mt} Utility
366 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
367 * Tape Files:: Tape Files
368 * Tarcat:: Concatenate Volumes into a Single Archive
373 * Generate Mode:: File Generation Mode.
374 * Status Mode:: File Status Mode.
375 * Exec Mode:: Synchronous Execution mode.
379 * Standard:: Basic Tar Format
380 * Extensions:: @acronym{GNU} Extensions to the Archive Format
386 * GNU Free Documentation License:: License for copying this manual
392 @chapter Introduction
395 and manipulates @dfn{archives} which are actually collections of
396 many other files; the program provides users with an organized and
397 systematic method for controlling a large amount of data.
398 The name ``tar'' originally came from the phrase ``Tape ARchive'', but
399 archives need not (and these days, typically do not) reside on tapes.
402 * Book Contents:: What this Book Contains
403 * Definitions:: Some Definitions
404 * What tar Does:: What @command{tar} Does
405 * Naming tar Archives:: How @command{tar} Archives are Named
406 * Authors:: @GNUTAR{} Authors
407 * Reports:: Reporting bugs or suggestions
411 @section What this Book Contains
413 The first part of this chapter introduces you to various terms that will
414 recur throughout the book. It also tells you who has worked on @GNUTAR{}
415 and its documentation, and where you should send bug reports
418 The second chapter is a tutorial (@pxref{Tutorial}) which provides a
419 gentle introduction for people who are new to using @command{tar}. It is
420 meant to be self contained, not requiring any reading from subsequent
421 chapters to make sense. It moves from topic to topic in a logical,
422 progressive order, building on information already explained.
424 Although the tutorial is paced and structured to allow beginners to
425 learn how to use @command{tar}, it is not intended solely for beginners.
426 The tutorial explains how to use the three most frequently used
427 operations (@samp{create}, @samp{list}, and @samp{extract}) as well as
428 two frequently used options (@samp{file} and @samp{verbose}). The other
429 chapters do not refer to the tutorial frequently; however, if a section
430 discusses something which is a complex variant of a basic concept, there
431 may be a cross reference to that basic concept. (The entire book,
432 including the tutorial, assumes that the reader understands some basic
433 concepts of using a Unix-type operating system; @pxref{Tutorial}.)
435 The third chapter presents the remaining five operations, and
436 information about using @command{tar} options and option syntax.
438 @FIXME{this sounds more like a @acronym{GNU} Project Manuals Concept [tm] more
439 than the reality. should think about whether this makes sense to say
440 here, or not.} The other chapters are meant to be used as a
441 reference. Each chapter presents everything that needs to be said
442 about a specific topic.
444 One of the chapters (@pxref{Date input formats}) exists in its
445 entirety in other @acronym{GNU} manuals, and is mostly self-contained.
446 In addition, one section of this manual (@pxref{Standard}) contains a
447 big quote which is taken directly from @command{tar} sources.
449 In general, we give both long and short (abbreviated) option names
450 at least once in each section where the relevant option is covered, so
451 that novice readers will become familiar with both styles. (A few
452 options have no short versions, and the relevant sections will
456 @section Some Definitions
460 The @command{tar} program is used to create and manipulate @command{tar}
461 archives. An @dfn{archive} is a single file which contains the contents
462 of many files, while still identifying the names of the files, their
463 owner(s), and so forth. (In addition, archives record access
464 permissions, user and group, size in bytes, and data modification time.
465 Some archives also record the file names in each archived directory, as
466 well as other file and directory information.) You can use @command{tar}
467 to @dfn{create} a new archive in a specified directory.
470 @cindex archive member
473 The files inside an archive are called @dfn{members}. Within this
474 manual, we use the term @dfn{file} to refer only to files accessible in
475 the normal ways (by @command{ls}, @command{cat}, and so forth), and the term
476 @dfn{member} to refer only to the members of an archive. Similarly, a
477 @dfn{file name} is the name of a file, as it resides in the file system,
478 and a @dfn{member name} is the name of an archive member within the
483 The term @dfn{extraction} refers to the process of copying an archive
484 member (or multiple members) into a file in the file system. Extracting
485 all the members of an archive is often called @dfn{extracting the
486 archive}. The term @dfn{unpack} can also be used to refer to the
487 extraction of many or all the members of an archive. Extracting an
488 archive does not destroy the archive's structure, just as creating an
489 archive does not destroy the copies of the files that exist outside of
490 the archive. You may also @dfn{list} the members in a given archive
491 (this is often thought of as ``printing'' them to the standard output,
492 or the command line), or @dfn{append} members to a pre-existing archive.
493 All of these operations can be performed using @command{tar}.
496 @section What @command{tar} Does
499 The @command{tar} program provides the ability to create @command{tar}
500 archives, as well as various other kinds of manipulation. For example,
501 you can use @command{tar} on previously created archives to extract files,
502 to store additional files, or to update or list files which were already
505 Initially, @command{tar} archives were used to store files conveniently on
506 magnetic tape. The name @command{tar} comes from this use; it stands for
507 @code{t}ape @code{ar}chiver. Despite the utility's name, @command{tar} can
508 direct its output to available devices, files, or other programs (using
509 pipes). @command{tar} may even access remote devices or files (as archives).
511 You can use @command{tar} archives in many ways. We want to stress a few
512 of them: storage, backup, and transportation.
514 @FIXME{the following table entries need a bit of work..}
517 Often, @command{tar} archives are used to store related files for
518 convenient file transfer over a network. For example, the
519 @acronym{GNU} Project distributes its software bundled into
520 @command{tar} archives, so that all the files relating to a particular
521 program (or set of related programs) can be transferred as a single
524 A magnetic tape can store several files in sequence. However, the tape
525 has no names for these files; it only knows their relative position on
526 the tape. One way to store several files on one tape and retain their
527 names is by creating a @command{tar} archive. Even when the basic transfer
528 mechanism can keep track of names, as FTP can, the nuisance of handling
529 multiple files, directories, and multiple links makes @command{tar}
532 Archive files are also used for long-term storage. You can think of
533 this as transportation from the present into the future. (It is a
534 science-fiction idiom that you can move through time as well as in
535 space; the idea here is that @command{tar} can be used to move archives in
536 all dimensions, even time!)
539 Because the archive created by @command{tar} is capable of preserving
540 file information and directory structure, @command{tar} is commonly
541 used for performing full and incremental backups of disks. A backup
542 puts a collection of files (possibly pertaining to many users and
543 projects) together on a disk or a tape. This guards against
544 accidental destruction of the information in those files.
545 @GNUTAR{} has special features that allow it to be
546 used to make incremental and full dumps of all the files in a
550 You can create an archive on one system, transfer it to another system,
551 and extract the contents there. This allows you to transport a group of
552 files from one system to another.
555 @node Naming tar Archives
556 @section How @command{tar} Archives are Named
558 Conventionally, @command{tar} archives are given names ending with
559 @samp{.tar}. This is not necessary for @command{tar} to operate properly,
560 but this manual follows that convention in order to accustom readers to
561 it and to make examples more clear.
566 Often, people refer to @command{tar} archives as ``@command{tar} files,'' and
567 archive members as ``files'' or ``entries''. For people familiar with
568 the operation of @command{tar}, this causes no difficulty. However, in
569 this manual, we consistently refer to ``archives'' and ``archive
570 members'' to make learning to use @command{tar} easier for novice users.
573 @section @GNUTAR{} Authors
575 @GNUTAR{} was originally written by John Gilmore,
576 and modified by many people. The @acronym{GNU} enhancements were
577 written by Jay Fenlason, then Joy Kendall, and the whole package has
578 been further maintained by Thomas Bushnell, n/BSG, Fran@,{c}ois
579 Pinard, Paul Eggert, and finally Sergey Poznyakoff with the help of
580 numerous and kind users.
582 We wish to stress that @command{tar} is a collective work, and owes much to
583 all those people who reported problems, offered solutions and other
584 insights, or shared their thoughts and suggestions. An impressive, yet
585 partial list of those contributors can be found in the @file{THANKS}
586 file from the @GNUTAR{} distribution.
588 @FIXME{i want all of these names mentioned, Absolutely. BUT, i'm not
589 sure i want to spell out the history in this detail, at least not for
590 the printed book. i'm just not sure it needs to be said this way.
591 i'll think about it.}
593 @FIXME{History is more important, and surely more interesting, than
594 actual names. Quoting names without history would be meaningless. FP}
596 Jay Fenlason put together a draft of a @GNUTAR{}
597 manual, borrowing notes from the original man page from John Gilmore.
598 This was withdrawn in version 1.11. Thomas Bushnell, n/BSG and Amy
599 Gorin worked on a tutorial and manual for @GNUTAR{}.
600 Fran@,{c}ois Pinard put version 1.11.8 of the manual together by
601 taking information from all these sources and merging them. Melissa
602 Weisshaus finally edited and redesigned the book to create version
603 1.12. The book for versions from 1.14 up to @value{VERSION} were edited
604 by the current maintainer, Sergey Poznyakoff.
606 For version 1.12, Daniel Hagerty contributed a great deal of technical
607 consulting. In particular, he is the primary author of @ref{Backups}.
609 In July, 2003 @GNUTAR{} was put on CVS at savannah.gnu.org
610 (see @url{http://savannah.gnu.org/projects/tar}), and
611 active development and maintenance work has started
612 again. Currently @GNUTAR{} is being maintained by Paul Eggert, Sergey
613 Poznyakoff and Jeff Bailey.
615 Support for @acronym{POSIX} archives was added by Sergey Poznyakoff.
618 @section Reporting bugs or suggestions
621 @cindex reporting bugs
622 If you find problems or have suggestions about this program or manual,
623 please report them to @file{bug-tar@@gnu.org}.
625 When reporting a bug, please be sure to include as much detail as
626 possible, in order to reproduce it. @FIXME{Be more specific, I'd
627 like to make this node as detailed as 'Bug reporting' node in Emacs
631 @chapter Tutorial Introduction to @command{tar}
633 This chapter guides you through some basic examples of three @command{tar}
634 operations: @option{--create}, @option{--list}, and @option{--extract}. If
635 you already know how to use some other version of @command{tar}, then you
636 may not need to read this chapter. This chapter omits most complicated
637 details about how @command{tar} works.
641 * stylistic conventions::
642 * basic tar options:: Basic @command{tar} Operations and Options
643 * frequent operations::
644 * Two Frequent Options::
645 * create:: How to Create Archives
646 * list:: How to List Archives
647 * extract:: How to Extract Members from an Archive
652 @section Assumptions this Tutorial Makes
654 This chapter is paced to allow beginners to learn about @command{tar}
655 slowly. At the same time, we will try to cover all the basic aspects of
656 these three operations. In order to accomplish both of these tasks, we
657 have made certain assumptions about your knowledge before reading this
658 manual, and the hardware you will be using:
662 Before you start to work through this tutorial, you should understand
663 what the terms ``archive'' and ``archive member'' mean
664 (@pxref{Definitions}). In addition, you should understand something
665 about how Unix-type operating systems work, and you should know how to
666 use some basic utilities. For example, you should know how to create,
667 list, copy, rename, edit, and delete files and directories; how to
668 change between directories; and how to figure out where you are in the
669 file system. You should have some basic understanding of directory
670 structure and how files are named according to which directory they are
671 in. You should understand concepts such as standard output and standard
672 input, what various definitions of the term ``argument'' mean, and the
673 differences between relative and absolute path names. @FIXME{and what
677 This manual assumes that you are working from your own home directory
678 (unless we state otherwise). In this tutorial, you will create a
679 directory to practice @command{tar} commands in. When we show path names,
680 we will assume that those paths are relative to your home directory.
681 For example, my home directory path is @file{/home/fsf/melissa}. All of
682 my examples are in a subdirectory of the directory named by that path
683 name; the subdirectory is called @file{practice}.
686 In general, we show examples of archives which exist on (or can be
687 written to, or worked with from) a directory on a hard disk. In most
688 cases, you could write those archives to, or work with them on any other
689 device, such as a tape drive. However, some of the later examples in
690 the tutorial and next chapter will not work on tape drives.
691 Additionally, working with tapes is much more complicated than working
692 with hard disks. For these reasons, the tutorial does not cover working
693 with tape drives. @xref{Media}, for complete information on using
694 @command{tar} archives with tape drives.
696 @FIXME{this is a cop out. need to add some simple tape drive info.}
699 @node stylistic conventions
700 @section Stylistic Conventions
702 In the examples, @samp{$} represents a typical shell prompt. It
703 precedes lines you should type; to make this more clear, those lines are
704 shown in @kbd{this font}, as opposed to lines which represent the
705 computer's response; those lines are shown in @code{this font}, or
706 sometimes @samp{like this}.
708 @c When we have lines which are too long to be
709 @c displayed in any other way, we will show them like this:
711 @node basic tar options
712 @section Basic @command{tar} Operations and Options
714 @command{tar} can take a wide variety of arguments which specify and define
715 the actions it will have on the particular set of files or the archive.
716 The main types of arguments to @command{tar} fall into one of two classes:
717 operations, and options.
719 Some arguments fall into a class called @dfn{operations}; exactly one of
720 these is both allowed and required for any instance of using @command{tar};
721 you may @emph{not} specify more than one. People sometimes speak of
722 @dfn{operating modes}. You are in a particular operating mode when you
723 have specified the operation which specifies it; there are eight
724 operations in total, and thus there are eight operating modes.
726 The other arguments fall into the class known as @dfn{options}. You are
727 not required to specify any options, and you are allowed to specify more
728 than one at a time (depending on the way you are using @command{tar} at
729 that time). Some options are used so frequently, and are so useful for
730 helping you type commands more carefully that they are effectively
731 ``required''. We will discuss them in this chapter.
733 You can write most of the @command{tar} operations and options in any
734 of three forms: long (mnemonic) form, short form, and old style. Some
735 of the operations and options have no short or ``old'' forms; however,
736 the operations and options which we will cover in this tutorial have
737 corresponding abbreviations. @FIXME{make sure this is still the case,
738 at the end}We will indicate those abbreviations appropriately to get
739 you used to seeing them. (Note that the ``old style'' option forms
740 exist in @GNUTAR{} for compatibility with Unix
741 @command{tar}. In this book we present a full discussion of this way
742 of writing options and operations (@pxref{Old Options}), and we discuss
743 the other two styles of writing options (@xref{Mnemonic Options}, and
744 @pxref{Short Options}).
746 In the examples and in the text of this tutorial, we usually use the
747 long forms of operations and options; but the ``short'' forms produce
748 the same result and can make typing long @command{tar} commands easier.
749 For example, instead of typing
752 @kbd{tar --create --verbose --file=afiles.tar apple angst aspic}
758 @kbd{tar -c -v -f afiles.tar apple angst aspic}
764 @kbd{tar -cvf afiles.tar apple angst aspic}
768 For more information on option syntax, see @ref{Advanced tar}. In
769 discussions in the text, when we name an option by its long form, we
770 also give the corresponding short option in parentheses.
772 The term, ``option'', can be confusing at times, since ``operations''
773 are often lumped in with the actual, @emph{optional} ``options'' in certain
774 general class statements. For example, we just talked about ``short and
775 long forms of options and operations''. However, experienced @command{tar}
776 users often refer to these by shorthand terms such as, ``short and long
777 options''. This term assumes that the ``operations'' are included, also.
778 Context will help you determine which definition of ``options'' to use.
780 Similarly, the term ``command'' can be confusing, as it is often used in
781 two different ways. People sometimes refer to @command{tar} ``commands''.
782 A @command{tar} @dfn{command} is the entire command line of user input
783 which tells @command{tar} what to do --- including the operation, options,
784 and any arguments (file names, pipes, other commands, etc). However,
785 you will also sometimes hear the term ``the @command{tar} command''. When
786 the word ``command'' is used specifically like this, a person is usually
787 referring to the @command{tar} @emph{operation}, not the whole line.
788 Again, use context to figure out which of the meanings the speaker
791 @node frequent operations
792 @section The Three Most Frequently Used Operations
794 Here are the three most frequently used operations (both short and long
795 forms), as well as a brief description of their meanings. The rest of
796 this chapter will cover how to use these operations in detail. We will
797 present the rest of the operations in the next chapter.
802 Create a new @command{tar} archive.
805 List the contents of an archive.
808 Extract one or more members from an archive.
811 @node Two Frequent Options
812 @section Two Frequently Used Options
814 To understand how to run @command{tar} in the three operating modes listed
815 previously, you also need to understand how to use two of the options to
816 @command{tar}: @option{--file} (which takes an archive file as an argument)
817 and @option{--verbose}. (You are usually not @emph{required} to specify
818 either of these options when you run @command{tar}, but they can be very
819 useful in making things more clear and helping you avoid errors.)
828 @unnumberedsubsec The @option{--file} Option
831 @opindex file, tutorial
832 @item --file=@var{archive-name}
833 @itemx -f @var{archive-name}
834 Specify the name of an archive file.
837 You can specify an argument for the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option whenever you
838 use @command{tar}; this option determines the name of the archive file
839 that @command{tar} will work on.
842 If you don't specify this argument, then @command{tar} will examine
843 the environment variable @env{TAPE}. If it is set, its value will be
844 used as the archive name. Otherwise, @command{tar} will use the
845 default archive, determined at the compile time. Usually it is
846 standard output or some physical tape drive attached to your machine
847 (you can verify what the default is by running @kbd{tar
848 --show-defaults}, @pxref{defaults}). If there is no tape drive
849 attached, or the default is not meaningful, then @command{tar} will
850 print an error message. The error message might look roughly like one
854 tar: can't open /dev/rmt8 : No such device or address
855 tar: can't open /dev/rsmt0 : I/O error
859 To avoid confusion, we recommend that you always specify an archive file
860 name by using @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) when writing your @command{tar} commands.
861 For more information on using the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option, see
864 @node verbose tutorial
865 @unnumberedsubsec The @option{--verbose} Option
868 @opindex verbose, introduced
871 Show the files being worked on as @command{tar} is running.
874 @option{--verbose} (@option{-v}) shows details about the results of running
875 @command{tar}. This can be especially useful when the results might not be
876 obvious. For example, if you want to see the progress of @command{tar} as
877 it writes files into the archive, you can use the @option{--verbose}
878 option. In the beginning, you may find it useful to use
879 @option{--verbose} at all times; when you are more accustomed to
880 @command{tar}, you will likely want to use it at certain times but not at
881 others. We will use @option{--verbose} at times to help make something
882 clear, and we will give many examples both using and not using
883 @option{--verbose} to show the differences.
885 Each instance of @option{--verbose} on the command line increases the
886 verbosity level by one, so if you need more details on the output,
889 When reading archives (@option{--list}, @option{--extract},
890 @option{--diff}), @command{tar} by default prints only the names of
891 the members being extracted. Using @option{--verbose} will show a full,
892 @command{ls} style member listing.
894 In contrast, when writing archives (@option{--create}, @option{--append},
895 @option{--update}), @command{tar} does not print file names by
896 default. So, a single @option{--verbose} option shows the file names
897 being added to the archive, while two @option{--verbose} options
898 enable the full listing.
900 For example, to create an archive in verbose mode:
903 $ @kbd{tar -cvf afiles.tar apple angst aspic}
910 Creating the same archive with the verbosity level 2 could give:
913 $ @kbd{tar -cvvf afiles.tar apple angst aspic}
914 -rw-r--r-- gray/staff 62373 2006-06-09 12:06 apple
915 -rw-r--r-- gray/staff 11481 2006-06-09 12:06 angst
916 -rw-r--r-- gray/staff 23152 2006-06-09 12:06 aspic
920 This works equally well using short or long forms of options. Using
921 long forms, you would simply write out the mnemonic form of the option
925 $ @kbd{tar --create --verbose --verbose @dots{}}
929 Note that you must double the hyphens properly each time.
931 Later in the tutorial, we will give examples using @w{@option{--verbose
934 @anchor{verbose member listing}
935 The full output consists of six fields:
938 @item File type and permissions in symbolic form.
939 These are displayed in the same format as the first column of
940 @command{ls -l} output (@pxref{What information is listed,
941 format=verbose, Verbose listing, fileutils, GNU file utilities}).
943 @item Owner name and group separated by a slash character.
944 If these data are not available (for example, when listing a @samp{v7} format
945 archive), numeric ID values are printed instead.
947 @item Size of the file, in bytes.
949 @item File modification date in ISO 8601 format.
951 @item File modification time.
954 If the name contains any special characters (white space, newlines,
955 etc.) these are displayed in an unambiguous form using so called
956 @dfn{quoting style}. For the detailed discussion of available styles
957 and on how to use them, see @ref{quoting styles}.
959 Depending on the file type, the name can be followed by some
960 additional information, described in the following table:
963 @item -> @var{link-name}
964 The file or archive member is a @dfn{symbolic link} and
965 @var{link-name} is the name of file it links to.
967 @item link to @var{link-name}
968 The file or archive member is a @dfn{hard link} and @var{link-name} is
969 the name of file it links to.
972 The archive member is an old GNU format long link. You will normally
976 The archive member is an old GNU format long name. You will normally
979 @item --Volume Header--
980 The archive member is a GNU @dfn{volume header} (@pxref{Tape Files}).
982 @item --Continued at byte @var{n}--
983 Encountered only at the beginning of a multy-volume archive
984 (@pxref{Using Multiple Tapes}). This archive member is a continuation
985 from the previous volume. The number @var{n} gives the offset where
986 the original file was split.
988 @item --Mangled file names--
989 This archive member contains @dfn{mangled file names} declarations,
990 a special member type that was used by early versions of @GNUTAR{}.
991 You probably will never encounter this, unless you are reading a very
994 @item unknown file type @var{c}
995 An archive member of unknown type. @var{c} is the type character from
996 the archive header. If you encounter such a message, it means that
997 either your archive contains proprietary member types @GNUTAR{} is not
998 able to handle, or the archive is corrupted.
1003 For example, here is an archive listing containing most of the special
1004 suffixes explained above:
1008 V--------- 0/0 1536 2006-06-09 13:07 MyVolume--Volume Header--
1009 -rw-r--r-- gray/staff 456783 2006-06-09 12:06 aspic--Continued at
1011 -rw-r--r-- gray/staff 62373 2006-06-09 12:06 apple
1012 lrwxrwxrwx gray/staff 0 2006-06-09 13:01 angst -> apple
1013 -rw-r--r-- gray/staff 35793 2006-06-09 12:06 blues
1014 hrw-r--r-- gray/staff 0 2006-06-09 12:06 music link to blues
1022 @unnumberedsubsec Getting Help: Using the @option{--help} Option
1028 The @option{--help} option to @command{tar} prints out a very brief list of
1029 all operations and option available for the current version of
1030 @command{tar} available on your system.
1034 @section How to Create Archives
1037 @cindex Creation of the archive
1038 @cindex Archive, creation of
1039 One of the basic operations of @command{tar} is @option{--create} (@option{-c}), which
1040 you use to create a @command{tar} archive. We will explain
1041 @option{--create} first because, in order to learn about the other
1042 operations, you will find it useful to have an archive available to
1045 To make this easier, in this section you will first create a directory
1046 containing three files. Then, we will show you how to create an
1047 @emph{archive} (inside the new directory). Both the directory, and
1048 the archive are specifically for you to practice on. The rest of this
1049 chapter and the next chapter will show many examples using this
1050 directory and the files you will create: some of those files may be
1051 other directories and other archives.
1053 The three files you will archive in this example are called
1054 @file{blues}, @file{folk}, and @file{jazz}. The archive is called
1055 @file{collection.tar}.
1057 This section will proceed slowly, detailing how to use @option{--create}
1058 in @code{verbose} mode, and showing examples using both short and long
1059 forms. In the rest of the tutorial, and in the examples in the next
1060 chapter, we will proceed at a slightly quicker pace. This section
1061 moves more slowly to allow beginning users to understand how
1062 @command{tar} works.
1065 * prepare for examples::
1066 * Creating the archive::
1072 @node prepare for examples
1073 @subsection Preparing a Practice Directory for Examples
1075 To follow along with this and future examples, create a new directory
1076 called @file{practice} containing files called @file{blues}, @file{folk}
1077 and @file{jazz}. The files can contain any information you like:
1078 ideally, they should contain information which relates to their names,
1079 and be of different lengths. Our examples assume that @file{practice}
1080 is a subdirectory of your home directory.
1082 Now @command{cd} to the directory named @file{practice}; @file{practice}
1083 is now your @dfn{working directory}. (@emph{Please note}: Although
1084 the full path name of this directory is
1085 @file{/@var{homedir}/practice}, in our examples we will refer to
1086 this directory as @file{practice}; the @var{homedir} is presumed.
1088 In general, you should check that the files to be archived exist where
1089 you think they do (in the working directory) by running @command{ls}.
1090 Because you just created the directory and the files and have changed to
1091 that directory, you probably don't need to do that this time.
1093 It is very important to make sure there isn't already a file in the
1094 working directory with the archive name you intend to use (in this case,
1095 @samp{collection.tar}), or that you don't care about its contents.
1096 Whenever you use @samp{create}, @command{tar} will erase the current
1097 contents of the file named by @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) if it exists. @command{tar}
1098 will not tell you if you are about to overwrite an archive unless you
1099 specify an option which does this (@pxref{backup}, for the
1100 information on how to do so). To add files to an existing archive,
1101 you need to use a different option, such as @option{--append} (@option{-r}); see
1102 @ref{append} for information on how to do this.
1104 @node Creating the archive
1105 @subsection Creating the Archive
1107 @opindex create, introduced
1108 To place the files @file{blues}, @file{folk}, and @file{jazz} into an
1109 archive named @file{collection.tar}, use the following command:
1112 $ @kbd{tar --create --file=collection.tar blues folk jazz}
1115 The order of the arguments is not very important, @emph{when using long
1116 option forms}. You could also say:
1119 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1123 However, you can see that this order is harder to understand; this is
1124 why we will list the arguments in the order that makes the commands
1125 easiest to understand (and we encourage you to do the same when you use
1126 @command{tar}, to avoid errors).
1128 Note that the part of the command which says,
1129 @w{@option{--file=collection.tar}} is considered to be @emph{one} argument.
1130 If you substituted any other string of characters for
1131 @kbd{collection.tar}, then that string would become the name of the
1132 archive file you create.
1134 The order of the options becomes more important when you begin to use
1135 short forms. With short forms, if you type commands in the wrong order
1136 (even if you type them correctly in all other ways), you may end up with
1137 results you don't expect. For this reason, it is a good idea to get
1138 into the habit of typing options in the order that makes inherent sense.
1139 @xref{short create}, for more information on this.
1141 In this example, you type the command as shown above: @option{--create}
1142 is the operation which creates the new archive
1143 (@file{collection.tar}), and @option{--file} is the option which lets
1144 you give it the name you chose. The files, @file{blues}, @file{folk},
1145 and @file{jazz}, are now members of the archive, @file{collection.tar}
1146 (they are @dfn{file name arguments} to the @option{--create} operation.
1147 @xref{Choosing}, for the detailed discussion on these.) Now that they are
1148 in the archive, they are called @emph{archive members}, not files.
1149 (@pxref{Definitions,members}).
1151 When you create an archive, you @emph{must} specify which files you
1152 want placed in the archive. If you do not specify any archive
1153 members, @GNUTAR{} will complain.
1155 If you now list the contents of the working directory (@command{ls}), you will
1156 find the archive file listed as well as the files you saw previously:
1159 blues folk jazz collection.tar
1163 Creating the archive @samp{collection.tar} did not destroy the copies of
1164 the files in the directory.
1166 Keep in mind that if you don't indicate an operation, @command{tar} will not
1167 run and will prompt you for one. If you don't name any files, @command{tar}
1168 will complain. You must have write access to the working directory,
1169 or else you will not be able to create an archive in that directory.
1171 @emph{Caution}: Do not attempt to use @option{--create} (@option{-c}) to add files to
1172 an existing archive; it will delete the archive and write a new one.
1173 Use @option{--append} (@option{-r}) instead. @xref{append}.
1175 @node create verbose
1176 @subsection Running @option{--create} with @option{--verbose}
1178 @opindex create, using with @option{--verbose}
1179 @opindex verbose, using with @option{--create}
1180 If you include the @option{--verbose} (@option{-v}) option on the command line,
1181 @command{tar} will list the files it is acting on as it is working. In
1182 verbose mode, the @code{create} example above would appear as:
1185 $ @kbd{tar --create --verbose --file=collection.tar blues folk jazz}
1191 This example is just like the example we showed which did not use
1192 @option{--verbose}, except that @command{tar} generated the remaining lines
1194 (note the different font styles).
1200 In the rest of the examples in this chapter, we will frequently use
1201 @code{verbose} mode so we can show actions or @command{tar} responses that
1202 you would otherwise not see, and which are important for you to
1206 @subsection Short Forms with @samp{create}
1208 As we said before, the @option{--create} (@option{-c}) operation is one of the most
1209 basic uses of @command{tar}, and you will use it countless times.
1210 Eventually, you will probably want to use abbreviated (or ``short'')
1211 forms of options. A full discussion of the three different forms that
1212 options can take appears in @ref{Styles}; for now, here is what the
1213 previous example (including the @option{--verbose} (@option{-v}) option) looks like
1214 using short option forms:
1217 $ @kbd{tar -cvf collection.tar blues folk jazz}
1224 As you can see, the system responds the same no matter whether you use
1225 long or short option forms.
1227 @FIXME{i don't like how this is worded:} One difference between using
1228 short and long option forms is that, although the exact placement of
1229 arguments following options is no more specific when using short forms,
1230 it is easier to become confused and make a mistake when using short
1231 forms. For example, suppose you attempted the above example in the
1235 $ @kbd{tar -cfv collection.tar blues folk jazz}
1239 In this case, @command{tar} will make an archive file called @file{v},
1240 containing the files @file{blues}, @file{folk}, and @file{jazz}, because
1241 the @samp{v} is the closest ``file name'' to the @option{-f} option, and
1242 is thus taken to be the chosen archive file name. @command{tar} will try
1243 to add a file called @file{collection.tar} to the @file{v} archive file;
1244 if the file @file{collection.tar} did not already exist, @command{tar} will
1245 report an error indicating that this file does not exist. If the file
1246 @file{collection.tar} does already exist (e.g., from a previous command
1247 you may have run), then @command{tar} will add this file to the archive.
1248 Because the @option{-v} option did not get registered, @command{tar} will not
1249 run under @samp{verbose} mode, and will not report its progress.
1251 The end result is that you may be quite confused about what happened,
1252 and possibly overwrite a file. To illustrate this further, we will show
1253 you how an example we showed previously would look using short forms.
1258 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1262 is confusing as it is. When shown using short forms, however, it
1263 becomes much more so:
1266 $ @kbd{tar blues -c folk -f collection.tar jazz}
1270 It would be very easy to put the wrong string of characters
1271 immediately following the @option{-f}, but doing that could sacrifice
1274 For this reason, we recommend that you pay very careful attention to
1275 the order of options and placement of file and archive names,
1276 especially when using short option forms. Not having the option name
1277 written out mnemonically can affect how well you remember which option
1278 does what, and therefore where different names have to be placed.
1281 @subsection Archiving Directories
1283 @cindex Archiving Directories
1284 @cindex Directories, Archiving
1285 You can archive a directory by specifying its directory name as a
1286 file name argument to @command{tar}. The files in the directory will be
1287 archived relative to the working directory, and the directory will be
1288 re-created along with its contents when the archive is extracted.
1290 To archive a directory, first move to its superior directory. If you
1291 have followed the previous instructions in this tutorial, you should
1300 This will put you into the directory which contains @file{practice},
1301 i.e., your home directory. Once in the superior directory, you can
1302 specify the subdirectory, @file{practice}, as a file name argument. To
1303 store @file{practice} in the new archive file @file{music.tar}, type:
1306 $ @kbd{tar --create --verbose --file=music.tar practice}
1310 @command{tar} should output:
1317 practice/collection.tar
1320 Note that the archive thus created is not in the subdirectory
1321 @file{practice}, but rather in the current working directory---the
1322 directory from which @command{tar} was invoked. Before trying to archive a
1323 directory from its superior directory, you should make sure you have
1324 write access to the superior directory itself, not only the directory
1325 you are trying archive with @command{tar}. For example, you will probably
1326 not be able to store your home directory in an archive by invoking
1327 @command{tar} from the root directory; @xref{absolute}. (Note
1328 also that @file{collection.tar}, the original archive file, has itself
1329 been archived. @command{tar} will accept any file as a file to be
1330 archived, regardless of its content. When @file{music.tar} is
1331 extracted, the archive file @file{collection.tar} will be re-written
1332 into the file system).
1334 If you give @command{tar} a command such as
1337 $ @kbd{tar --create --file=foo.tar .}
1341 @command{tar} will report @samp{tar: ./foo.tar is the archive; not
1342 dumped}. This happens because @command{tar} creates the archive
1343 @file{foo.tar} in the current directory before putting any files into
1344 it. Then, when @command{tar} attempts to add all the files in the
1345 directory @file{.} to the archive, it notices that the file
1346 @file{./foo.tar} is the same as the archive @file{foo.tar}, and skips
1347 it. (It makes no sense to put an archive into itself.) @GNUTAR{}
1348 will continue in this case, and create the archive
1349 normally, except for the exclusion of that one file. (@emph{Please
1350 note:} Other implementations of @command{tar} may not be so clever;
1351 they will enter an infinite loop when this happens, so you should not
1352 depend on this behavior unless you are certain you are running
1353 @GNUTAR{}. In general, it is wise to always place the archive outside
1354 of the directory being dumped.
1357 @section How to List Archives
1360 Frequently, you will find yourself wanting to determine exactly what a
1361 particular archive contains. You can use the @option{--list} (@option{-t}) operation
1362 to get the member names as they currently appear in the archive, as well
1363 as various attributes of the files at the time they were archived. For
1364 example, you can examine the archive @file{collection.tar} that you
1365 created in the last section with the command,
1368 $ @kbd{tar --list --file=collection.tar}
1372 The output of @command{tar} would then be:
1381 The archive @file{bfiles.tar} would list as follows:
1390 Be sure to use a @option{--file=@var{archive-name}} (@option{-f
1391 @var{archive-name}}) option just as with @option{--create}
1392 (@option{-c}) to specify the name of the archive.
1394 @opindex list, using with @option{--verbose}
1395 @opindex verbose, using with @option{--list}
1396 If you use the @option{--verbose} (@option{-v}) option with
1397 @option{--list}, then @command{tar} will print out a listing
1398 reminiscent of @w{@samp{ls -l}}, showing owner, file size, and so
1399 forth. This output is described in detail in @ref{verbose member listing}.
1401 If you had used @option{--verbose} (@option{-v}) mode, the example
1402 above would look like:
1405 $ @kbd{tar --list --verbose --file=collection.tar folk}
1406 -rw-r--r-- myself user 62 1990-05-23 10:55 folk
1409 @cindex listing member and file names
1410 @anchor{listing member and file names}
1411 It is important to notice that the output of @kbd{tar --list
1412 --verbose} does not necessarily match that produced by @kbd{tar
1413 --create --verbose} while creating the archive. It is because
1414 @GNUTAR{}, unless told explicitly not to do so, removes some directory
1415 prefixes from file names before storing them in the archive
1416 (@xref{absolute}, for more information). In other
1417 words, in verbose mode @GNUTAR{} shows @dfn{file names} when creating
1418 an archive and @dfn{member names} when listing it. Consider this
1423 $ @kbd{tar cfv archive /etc/mail}
1424 tar: Removing leading `/' from member names
1426 /etc/mail/sendmail.cf
1428 $ @kbd{tar tf archive}
1430 etc/mail/sendmail.cf
1435 @opindex show-stored-names
1436 This default behavior can sometimes be inconvenient. You can force
1437 @GNUTAR{} show member names when creating archive by supplying
1438 @option{--show-stored-names} option.
1441 @item --show-stored-names
1442 Print member (as opposed to @emph{file}) names when creating the archive.
1445 @cindex File name arguments, using @option{--list} with
1446 @opindex list, using with file name arguments
1447 You can specify one or more individual member names as arguments when
1448 using @samp{list}. In this case, @command{tar} will only list the
1449 names of members you identify. For example, @w{@kbd{tar --list
1450 --file=afiles.tar apple}} would only print @file{apple}.
1452 Because @command{tar} preserves paths, file names must be specified as
1453 they appear in the archive (ie., relative to the directory from which
1454 the archive was created). Therefore, it is essential when specifying
1455 member names to @command{tar} that you give the exact member names.
1456 For example, @w{@kbd{tar --list --file=bfiles.tar birds}} would produce an
1457 error message something like @samp{tar: birds: Not found in archive},
1458 because there is no member named @file{birds}, only one named
1459 @file{./birds}. While the names @file{birds} and @file{./birds} name
1460 the same file, @emph{member} names by default are compared verbatim.
1462 However, @w{@kbd{tar --list --file=bfiles.tar baboon}} would respond
1463 with @file{baboon}, because this exact member name is in the archive file
1464 @file{bfiles.tar}. If you are not sure of the exact file name,
1465 use @dfn{globbing patterns}, for example:
1468 $ @kbd{tar --list --file=bfiles.tar --wildcards '*b*'}
1472 will list all members whose name contains @samp{b}. @xref{wildcards},
1473 for a detailed discussion of globbing patterns and related
1474 @command{tar} command line options.
1481 @unnumberedsubsec Listing the Contents of a Stored Directory
1483 To get information about the contents of an archived directory,
1484 use the directory name as a file name argument in conjunction with
1485 @option{--list} (@option{-t}). To find out file attributes, include the
1486 @option{--verbose} (@option{-v}) option.
1488 For example, to find out about files in the directory @file{practice}, in
1489 the archive file @file{music.tar}, type:
1492 $ @kbd{tar --list --verbose --file=music.tar practice}
1495 @command{tar} responds:
1498 drwxrwxrwx myself user 0 1990-05-31 21:49 practice/
1499 -rw-r--r-- myself user 42 1990-05-21 13:29 practice/blues
1500 -rw-r--r-- myself user 62 1990-05-23 10:55 practice/folk
1501 -rw-r--r-- myself user 40 1990-05-21 13:30 practice/jazz
1502 -rw-r--r-- myself user 10240 1990-05-31 21:49 practice/collection.tar
1505 When you use a directory name as a file name argument, @command{tar} acts on
1506 all the files (including sub-directories) in that directory.
1509 @section How to Extract Members from an Archive
1512 @cindex Retrieving files from an archive
1513 @cindex Resurrecting files from an archive
1516 Creating an archive is only half the job---there is no point in storing
1517 files in an archive if you can't retrieve them. The act of retrieving
1518 members from an archive so they can be used and manipulated as
1519 unarchived files again is called @dfn{extraction}. To extract files
1520 from an archive, use the @option{--extract} (@option{--get} or
1521 @option{-x}) operation. As with @option{--create}, specify the name
1522 of the archive with @option{--file} (@option{-f}) option. Extracting
1523 an archive does not modify the archive in any way; you can extract it
1524 multiple times if you want or need to.
1526 Using @option{--extract}, you can extract an entire archive, or specific
1527 files. The files can be directories containing other files, or not. As
1528 with @option{--create} (@option{-c}) and @option{--list} (@option{-t}), you may use the short or the
1529 long form of the operation without affecting the performance.
1532 * extracting archives::
1533 * extracting files::
1535 * extracting untrusted archives::
1536 * failing commands::
1539 @node extracting archives
1540 @subsection Extracting an Entire Archive
1542 To extract an entire archive, specify the archive file name only, with
1543 no individual file names as arguments. For example,
1546 $ @kbd{tar -xvf collection.tar}
1553 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
1554 -rw-r--r-- me user 21 1996-09-23 16:44 blues
1555 -rw-r--r-- me user 20 1996-09-23 16:44 folk
1558 @node extracting files
1559 @subsection Extracting Specific Files
1561 To extract specific archive members, give their exact member names as
1562 arguments, as printed by @option{--list} (@option{-t}). If you had
1563 mistakenly deleted one of the files you had placed in the archive
1564 @file{collection.tar} earlier (say, @file{blues}), you can extract it
1565 from the archive without changing the archive's structure. Its
1566 contents will be identical to the original file @file{blues} that you
1569 First, make sure you are in the @file{practice} directory, and list the
1570 files in the directory. Now, delete the file, @samp{blues}, and list
1571 the files in the directory again.
1573 You can now extract the member @file{blues} from the archive file
1574 @file{collection.tar} like this:
1577 $ @kbd{tar --extract --file=collection.tar blues}
1581 If you list the files in the directory again, you will see that the file
1582 @file{blues} has been restored, with its original permissions, data
1583 modification times, and owner.@footnote{This is only accidentally
1584 true, but not in general. Whereas modification times are always
1585 restored, in most cases, one has to be root for restoring the owner,
1586 and use a special option for restoring permissions. Here, it just
1587 happens that the restoring user is also the owner of the archived
1588 members, and that the current @code{umask} is compatible with original
1589 permissions.} (These parameters will be identical to those which
1590 the file had when you originally placed it in the archive; any changes
1591 you may have made before deleting the file from the file system,
1592 however, will @emph{not} have been made to the archive member.) The
1593 archive file, @samp{collection.tar}, is the same as it was before you
1594 extracted @samp{blues}. You can confirm this by running @command{tar} with
1595 @option{--list} (@option{-t}).
1597 Remember that as with other operations, specifying the exact member
1598 name is important. @w{@kbd{tar --extract --file=bfiles.tar birds}}
1599 will fail, because there is no member named @file{birds}. To extract
1600 the member named @file{./birds}, you must specify @w{@kbd{tar
1601 --extract --file=bfiles.tar ./birds}}. If you don't remember the
1602 exact member names, use @option{--list} (@option{-t}) option
1603 (@pxref{list}). You can also extract those members that match a
1604 specific @dfn{globbing pattern}. For example, to extract from
1605 @file{bfiles.tar} all files that begin with @samp{b}, no matter their
1606 directory prefix, you could type:
1609 $ @kbd{tar -x -f bfiles.tar --wildcards --no-anchored 'b*'}
1613 Here, @option{--wildcards} instructs @command{tar} to treat
1614 command line arguments as globbing patterns and @option{--no-anchored}
1615 informs it that the patterns apply to member names after any @samp{/}
1616 delimiter. The use of globbing patterns is discussed in detail in
1619 You can extract a file to standard output by combining the above options
1620 with the @option{--to-stdout} (@option{-O}) option (@pxref{Writing to Standard
1623 If you give the @option{--verbose} option, then @option{--extract}
1624 will print the names of the archive members as it extracts them.
1627 @subsection Extracting Files that are Directories
1629 Extracting directories which are members of an archive is similar to
1630 extracting other files. The main difference to be aware of is that if
1631 the extracted directory has the same name as any directory already in
1632 the working directory, then files in the extracted directory will be
1633 placed into the directory of the same name. Likewise, if there are
1634 files in the pre-existing directory with the same names as the members
1635 which you extract, the files from the extracted archive will replace
1636 the files already in the working directory (and possible
1637 subdirectories). This will happen regardless of whether or not the
1638 files in the working directory were more recent than those extracted
1639 (there exist, however, special options that alter this behavior
1642 However, if a file was stored with a directory name as part of its file
1643 name, and that directory does not exist under the working directory when
1644 the file is extracted, @command{tar} will create the directory.
1646 We can demonstrate how to use @option{--extract} to extract a directory
1647 file with an example. Change to the @file{practice} directory if you
1648 weren't there, and remove the files @file{folk} and @file{jazz}. Then,
1649 go back to the parent directory and extract the archive
1650 @file{music.tar}. You may either extract the entire archive, or you may
1651 extract only the files you just deleted. To extract the entire archive,
1652 don't give any file names as arguments after the archive name
1653 @file{music.tar}. To extract only the files you deleted, use the
1657 $ @kbd{tar -xvf music.tar practice/folk practice/jazz}
1663 If you were to specify two @option{--verbose} (@option{-v}) options, @command{tar}
1664 would have displayed more detail about the extracted files, as shown
1665 in the example below:
1668 $ @kbd{tar -xvvf music.tar practice/folk practice/jazz}
1669 -rw-r--r-- me user 28 1996-10-18 16:31 practice/jazz
1670 -rw-r--r-- me user 20 1996-09-23 16:44 practice/folk
1674 Because you created the directory with @file{practice} as part of the
1675 file names of each of the files by archiving the @file{practice}
1676 directory as @file{practice}, you must give @file{practice} as part
1677 of the file names when you extract those files from the archive.
1679 @node extracting untrusted archives
1680 @subsection Extracting Archives from Untrusted Sources
1682 Extracting files from archives can overwrite files that already exist.
1683 If you receive an archive from an untrusted source, you should make a
1684 new directory and extract into that directory, so that you don't have
1685 to worry about the extraction overwriting one of your existing files.
1686 For example, if @file{untrusted.tar} came from somewhere else on the
1687 Internet, and you don't necessarily trust its contents, you can
1688 extract it as follows:
1691 $ @kbd{mkdir newdir}
1693 $ @kbd{tar -xvf ../untrusted.tar}
1696 It is also a good practice to examine contents of the archive
1697 before extracting it, using @option{--list} (@option{-t}) option, possibly combined
1698 with @option{--verbose} (@option{-v}).
1700 @node failing commands
1701 @subsection Commands That Will Fail
1703 Here are some sample commands you might try which will not work, and why
1706 If you try to use this command,
1709 $ @kbd{tar -xvf music.tar folk jazz}
1713 you will get the following response:
1716 tar: folk: Not found in archive
1717 tar: jazz: Not found in archive
1722 This is because these files were not originally @emph{in} the parent
1723 directory @file{..}, where the archive is located; they were in the
1724 @file{practice} directory, and their file names reflect this:
1727 $ @kbd{tar -tvf music.tar}
1733 @FIXME{make sure the above works when going through the examples in
1737 Likewise, if you try to use this command,
1740 $ @kbd{tar -tvf music.tar folk jazz}
1744 you would get a similar response. Members with those names are not in the
1745 archive. You must use the correct member names, or wildcards, in order
1746 to extract the files from the archive.
1748 If you have forgotten the correct names of the files in the archive,
1749 use @w{@kbd{tar --list --verbose}} to list them correctly.
1751 @FIXME{more examples, here? hag thinks it's a good idea.}
1754 @section Going Further Ahead in this Manual
1756 @FIXME{need to write up a node here about the things that are going to
1757 be in the rest of the manual.}
1759 @node tar invocation
1760 @chapter Invoking @GNUTAR{}
1763 This chapter is about how one invokes the @GNUTAR{}
1764 command, from the command synopsis (@pxref{Synopsis}). There are
1765 numerous options, and many styles for writing them. One mandatory
1766 option specifies the operation @command{tar} should perform
1767 (@pxref{Operation Summary}), other options are meant to detail how
1768 this operation should be performed (@pxref{Option Summary}).
1769 Non-option arguments are not always interpreted the same way,
1770 depending on what the operation is.
1772 You will find in this chapter everything about option styles and rules for
1773 writing them (@pxref{Styles}). On the other hand, operations and options
1774 are fully described elsewhere, in other chapters. Here, you will find
1775 only synthetic descriptions for operations and options, together with
1776 pointers to other parts of the @command{tar} manual.
1778 Some options are so special they are fully described right in this
1779 chapter. They have the effect of inhibiting the normal operation of
1780 @command{tar} or else, they globally alter the amount of feedback the user
1781 receives about what is going on. These are the @option{--help} and
1782 @option{--version} (@pxref{help}), @option{--verbose} (@pxref{verbose})
1783 and @option{--interactive} options (@pxref{interactive}).
1787 * using tar options::
1797 @section General Synopsis of @command{tar}
1799 The @GNUTAR{} program is invoked as either one of:
1802 @kbd{tar @var{option}@dots{} [@var{name}]@dots{}}
1803 @kbd{tar @var{letter}@dots{} [@var{argument}]@dots{} [@var{option}]@dots{} [@var{name}]@dots{}}
1806 The second form is for when old options are being used.
1808 You can use @command{tar} to store files in an archive, to extract them from
1809 an archive, and to do other types of archive manipulation. The primary
1810 argument to @command{tar}, which is called the @dfn{operation}, specifies
1811 which action to take. The other arguments to @command{tar} are either
1812 @dfn{options}, which change the way @command{tar} performs an operation,
1813 or file names or archive members, which specify the files or members
1814 @command{tar} is to act on.
1816 You can actually type in arguments in any order, even if in this manual
1817 the options always precede the other arguments, to make examples easier
1818 to understand. Further, the option stating the main operation mode
1819 (the @command{tar} main command) is usually given first.
1821 Each @var{name} in the synopsis above is interpreted as an archive member
1822 name when the main command is one of @option{--compare}
1823 (@option{--diff}, @option{-d}), @option{--delete}, @option{--extract}
1824 (@option{--get}, @option{-x}), @option{--list} (@option{-t}) or
1825 @option{--update} (@option{-u}). When naming archive members, you
1826 must give the exact name of the member in the archive, as it is
1827 printed by @option{--list}. For @option{--append} (@option{-r}) and
1828 @option{--create} (@option{-c}), these @var{name} arguments specify
1829 the names of either files or directory hierarchies to place in the archive.
1830 These files or hierarchies should already exist in the file system,
1831 prior to the execution of the @command{tar} command.
1833 @command{tar} interprets relative file names as being relative to the
1834 working directory. @command{tar} will make all file names relative
1835 (by removing leading slashes when archiving or restoring files),
1836 unless you specify otherwise (using the @option{--absolute-names}
1837 option). @xref{absolute}, for more information about
1838 @option{--absolute-names}.
1840 If you give the name of a directory as either a file name or a member
1841 name, then @command{tar} acts recursively on all the files and directories
1842 beneath that directory. For example, the name @file{/} identifies all
1843 the files in the file system to @command{tar}.
1845 The distinction between file names and archive member names is especially
1846 important when shell globbing is used, and sometimes a source of confusion
1847 for newcomers. @xref{wildcards}, for more information about globbing.
1848 The problem is that shells may only glob using existing files in the
1849 file system. Only @command{tar} itself may glob on archive members, so when
1850 needed, you must ensure that wildcard characters reach @command{tar} without
1851 being interpreted by the shell first. Using a backslash before @samp{*}
1852 or @samp{?}, or putting the whole argument between quotes, is usually
1853 sufficient for this.
1855 Even if @var{name}s are often specified on the command line, they
1856 can also be read from a text file in the file system, using the
1857 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}}) option.
1859 If you don't use any file name arguments, @option{--append} (@option{-r}),
1860 @option{--delete} and @option{--concatenate} (@option{--catenate},
1861 @option{-A}) will do nothing, while @option{--create} (@option{-c})
1862 will usually yield a diagnostic and inhibit @command{tar} execution.
1863 The other operations of @command{tar} (@option{--list},
1864 @option{--extract}, @option{--compare}, and @option{--update})
1865 will act on the entire contents of the archive.
1868 @cindex return status
1869 Besides successful exits, @GNUTAR{} may fail for
1870 many reasons. Some reasons correspond to bad usage, that is, when the
1871 @command{tar} command is improperly written. Errors may be
1872 encountered later, while encountering an error processing the archive
1873 or the files. Some errors are recoverable, in which case the failure
1874 is delayed until @command{tar} has completed all its work. Some
1875 errors are such that it would not meaningful, or at least risky, to
1876 continue processing: @command{tar} then aborts processing immediately.
1877 All abnormal exits, whether immediate or delayed, should always be
1878 clearly diagnosed on @code{stderr}, after a line stating the nature of
1881 @GNUTAR{} returns only a few exit statuses. I'm really
1882 aiming simplicity in that area, for now. If you are not using the
1883 @option{--compare} @option{--diff}, @option{-d}) option, zero means
1884 that everything went well, besides maybe innocuous warnings. Nonzero
1885 means that something went wrong. Right now, as of today, ``nonzero''
1886 is almost always 2, except for remote operations, where it may be
1889 @node using tar options
1890 @section Using @command{tar} Options
1892 @GNUTAR{} has a total of eight operating modes which
1893 allow you to perform a variety of tasks. You are required to choose
1894 one operating mode each time you employ the @command{tar} program by
1895 specifying one, and only one operation as an argument to the
1896 @command{tar} command (two lists of four operations each may be found
1897 at @ref{frequent operations} and @ref{Operations}). Depending on
1898 circumstances, you may also wish to customize how the chosen operating
1899 mode behaves. For example, you may wish to change the way the output
1900 looks, or the format of the files that you wish to archive may require
1901 you to do something special in order to make the archive look right.
1903 You can customize and control @command{tar}'s performance by running
1904 @command{tar} with one or more options (such as @option{--verbose}
1905 (@option{-v}), which we used in the tutorial). As we said in the
1906 tutorial, @dfn{options} are arguments to @command{tar} which are (as
1907 their name suggests) optional. Depending on the operating mode, you
1908 may specify one or more options. Different options will have different
1909 effects, but in general they all change details of the operation, such
1910 as archive format, archive name, or level of user interaction. Some
1911 options make sense with all operating modes, while others are
1912 meaningful only with particular modes. You will likely use some
1913 options frequently, while you will only use others infrequently, or
1914 not at all. (A full list of options is available in @pxref{All Options}.)
1916 @vrindex TAR_OPTIONS, environment variable
1917 @anchor{TAR_OPTIONS}
1918 The @env{TAR_OPTIONS} environment variable specifies default options to
1919 be placed in front of any explicit options. For example, if
1920 @code{TAR_OPTIONS} is @samp{-v --unlink-first}, @command{tar} behaves as
1921 if the two options @option{-v} and @option{--unlink-first} had been
1922 specified before any explicit options. Option specifications are
1923 separated by whitespace. A backslash escapes the next character, so it
1924 can be used to specify an option containing whitespace or a backslash.
1926 Note that @command{tar} options are case sensitive. For example, the
1927 options @option{-T} and @option{-t} are different; the first requires an
1928 argument for stating the name of a file providing a list of @var{name}s,
1929 while the second does not require an argument and is another way to
1930 write @option{--list} (@option{-t}).
1932 In addition to the eight operations, there are many options to
1933 @command{tar}, and three different styles for writing both: long (mnemonic)
1934 form, short form, and old style. These styles are discussed below.
1935 Both the options and the operations can be written in any of these three
1938 @FIXME{menu at end of this node. need to think of an actual outline
1939 for this chapter; probably do that after stuff from chapter 4 is
1943 @section The Three Option Styles
1945 There are three styles for writing operations and options to the command
1946 line invoking @command{tar}. The different styles were developed at
1947 different times during the history of @command{tar}. These styles will be
1948 presented below, from the most recent to the oldest.
1950 Some options must take an argument. (For example, @option{--file}
1951 (@option{-f})) takes the name of an archive file as an argument. If
1952 you do not supply an archive file name, @command{tar} will use a
1953 default, but this can be confusing; thus, we recommend that you always
1954 supply a specific archive file name.) Where you @emph{place} the
1955 arguments generally depends on which style of options you choose. We
1956 will detail specific information relevant to each option style in the
1957 sections on the different option styles, below. The differences are
1958 subtle, yet can often be very important; incorrect option placement
1959 can cause you to overwrite a number of important files. We urge you
1960 to note these differences, and only use the option style(s) which
1961 makes the most sense to you until you feel comfortable with the others.
1963 Some options @emph{may} take an argument (currently, there are
1964 two such options: @option{--backup} and @option{--occurrence}). Such
1965 options may have at most long and short forms, they do not have old style
1966 equivalent. The rules for specifying an argument for such options
1967 are stricter than those for specifying mandatory arguments. Please,
1968 pay special attention to them.
1971 * Mnemonic Options:: Mnemonic Option Style
1972 * Short Options:: Short Option Style
1973 * Old Options:: Old Option Style
1974 * Mixing:: Mixing Option Styles
1977 @node Mnemonic Options
1978 @subsection Mnemonic Option Style
1980 @FIXME{have to decide whether or not to replace other occurrences of
1981 "mnemonic" with "long", or *ugh* vice versa.}
1983 Each option has at least one long (or mnemonic) name starting with two
1984 dashes in a row, e.g., @option{--list}. The long names are more clear than
1985 their corresponding short or old names. It sometimes happens that a
1986 single mnemonic option has many different different names which are
1987 synonymous, such as @option{--compare} and @option{--diff}. In addition,
1988 long option names can be given unique abbreviations. For example,
1989 @option{--cre} can be used in place of @option{--create} because there is no
1990 other mnemonic option which begins with @samp{cre}. (One way to find
1991 this out is by trying it and seeing what happens; if a particular
1992 abbreviation could represent more than one option, @command{tar} will tell
1993 you that that abbreviation is ambiguous and you'll know that that
1994 abbreviation won't work. You may also choose to run @samp{tar --help}
1995 to see a list of options. Be aware that if you run @command{tar} with a
1996 unique abbreviation for the long name of an option you didn't want to
1997 use, you are stuck; @command{tar} will perform the command as ordered.)
1999 Mnemonic options are meant to be obvious and easy to remember, and their
2000 meanings are generally easier to discern than those of their
2001 corresponding short options (see below). For example:
2004 $ @kbd{tar --create --verbose --blocking-factor=20 --file=/dev/rmt0}
2008 gives a fairly good set of hints about what the command does, even
2009 for those not fully acquainted with @command{tar}.
2011 Mnemonic options which require arguments take those arguments
2012 immediately following the option name. There are two ways of
2013 specifying a mandatory argument. It can be separated from the
2014 option name either by an equal sign, or by any amount of
2015 white space characters. For example, the @option{--file} option (which
2016 tells the name of the @command{tar} archive) is given a file such as
2017 @file{archive.tar} as argument by using any of the following notations:
2018 @option{--file=archive.tar} or @option{--file archive.tar}.
2020 In contrast, optional arguments must always be introduced using
2021 an equal sign. For example, the @option{--backup} option takes
2022 an optional argument specifying backup type. It must be used
2023 as @option{--backup=@var{backup-type}}.
2026 @subsection Short Option Style
2028 Most options also have a short option name. Short options start with
2029 a single dash, and are followed by a single character, e.g., @option{-t}
2030 (which is equivalent to @option{--list}). The forms are absolutely
2031 identical in function; they are interchangeable.
2033 The short option names are faster to type than long option names.
2035 Short options which require arguments take their arguments immediately
2036 following the option, usually separated by white space. It is also
2037 possible to stick the argument right after the short option name, using
2038 no intervening space. For example, you might write @w{@option{-f
2039 archive.tar}} or @option{-farchive.tar} instead of using
2040 @option{--file=archive.tar}. Both @option{--file=@var{archive-name}} and
2041 @w{@option{-f @var{archive-name}}} denote the option which indicates a
2042 specific archive, here named @file{archive.tar}.
2044 Short options which take optional arguments take their arguments
2045 immediately following the option letter, @emph{without any intervening
2046 white space characters}.
2048 Short options' letters may be clumped together, but you are not
2049 required to do this (as compared to old options; see below). When
2050 short options are clumped as a set, use one (single) dash for them
2051 all, e.g., @w{@samp{@command{tar} -cvf}}. Only the last option in
2052 such a set is allowed to have an argument@footnote{Clustering many
2053 options, the last of which has an argument, is a rather opaque way to
2054 write options. Some wonder if @acronym{GNU} @code{getopt} should not
2055 even be made helpful enough for considering such usages as invalid.}.
2057 When the options are separated, the argument for each option which requires
2058 an argument directly follows that option, as is usual for Unix programs.
2062 $ @kbd{tar -c -v -b 20 -f /dev/rmt0}
2065 If you reorder short options' locations, be sure to move any arguments
2066 that belong to them. If you do not move the arguments properly, you may
2067 end up overwriting files.
2070 @subsection Old Option Style
2073 Like short options, old options are single letters. However, old options
2074 must be written together as a single clumped set, without spaces separating
2075 them or dashes preceding them@footnote{Beware that if you precede options
2076 with a dash, you are announcing the short option style instead of the
2077 old option style; short options are decoded differently.}. This set
2078 of letters must be the first to appear on the command line, after the
2079 @command{tar} program name and some white space; old options cannot appear
2080 anywhere else. The letter of an old option is exactly the same letter as
2081 the corresponding short option. For example, the old option @samp{t} is
2082 the same as the short option @option{-t}, and consequently, the same as the
2083 mnemonic option @option{--list}. So for example, the command @w{@samp{tar
2084 cv}} specifies the option @option{-v} in addition to the operation @option{-c}.
2086 When options that need arguments are given together with the command,
2087 all the associated arguments follow, in the same order as the options.
2088 Thus, the example given previously could also be written in the old
2092 $ @kbd{tar cvbf 20 /dev/rmt0}
2096 Here, @samp{20} is the argument of @option{-b} and @samp{/dev/rmt0} is
2097 the argument of @option{-f}.
2099 On the other hand, this old style syntax makes it difficult to match
2100 option letters with their corresponding arguments, and is often
2101 confusing. In the command @w{@samp{tar cvbf 20 /dev/rmt0}}, for example,
2102 @samp{20} is the argument for @option{-b}, @samp{/dev/rmt0} is the
2103 argument for @option{-f}, and @option{-v} does not have a corresponding
2104 argument. Even using short options like in @w{@samp{tar -c -v -b 20 -f
2105 /dev/rmt0}} is clearer, putting all arguments next to the option they
2108 If you want to reorder the letters in the old option argument, be
2109 sure to reorder any corresponding argument appropriately.
2111 This old way of writing @command{tar} options can surprise even experienced
2112 users. For example, the two commands:
2115 @kbd{tar cfz archive.tar.gz file}
2116 @kbd{tar -cfz archive.tar.gz file}
2120 are quite different. The first example uses @file{archive.tar.gz} as
2121 the value for option @samp{f} and recognizes the option @samp{z}. The
2122 second example, however, uses @file{z} as the value for option
2123 @samp{f} --- probably not what was intended.
2125 Old options are kept for compatibility with old versions of @command{tar}.
2127 This second example could be corrected in many ways, among which the
2128 following are equivalent:
2131 @kbd{tar -czf archive.tar.gz file}
2132 @kbd{tar -cf archive.tar.gz -z file}
2133 @kbd{tar cf archive.tar.gz -z file}
2136 @cindex option syntax, traditional
2137 As far as we know, all @command{tar} programs, @acronym{GNU} and
2138 non-@acronym{GNU}, support old options. @GNUTAR{}
2139 supports them not only for historical reasons, but also because many
2140 people are used to them. For compatibility with Unix @command{tar},
2141 the first argument is always treated as containing command and option
2142 letters even if it doesn't start with @samp{-}. Thus, @samp{tar c} is
2143 equivalent to @w{@samp{tar -c}:} both of them specify the
2144 @option{--create} (@option{-c}) command to create an archive.
2147 @subsection Mixing Option Styles
2149 All three styles may be intermixed in a single @command{tar} command,
2150 so long as the rules for each style are fully
2151 respected@footnote{Before @GNUTAR{} version 1.11.6,
2152 a bug prevented intermixing old style options with mnemonic options in
2153 some cases.}. Old style options and either of the modern styles of
2154 options may be mixed within a single @command{tar} command. However,
2155 old style options must be introduced as the first arguments only,
2156 following the rule for old options (old options must appear directly
2157 after the @command{tar} command and some white space). Modern options
2158 may be given only after all arguments to the old options have been
2159 collected. If this rule is not respected, a modern option might be
2160 falsely interpreted as the value of the argument to one of the old
2163 For example, all the following commands are wholly equivalent, and
2164 illustrate the many combinations and orderings of option styles.
2167 @kbd{tar --create --file=archive.tar}
2168 @kbd{tar --create -f archive.tar}
2169 @kbd{tar --create -farchive.tar}
2170 @kbd{tar --file=archive.tar --create}
2171 @kbd{tar --file=archive.tar -c}
2172 @kbd{tar -c --file=archive.tar}
2173 @kbd{tar -c -f archive.tar}
2174 @kbd{tar -c -farchive.tar}
2175 @kbd{tar -cf archive.tar}
2176 @kbd{tar -cfarchive.tar}
2177 @kbd{tar -f archive.tar --create}
2178 @kbd{tar -f archive.tar -c}
2179 @kbd{tar -farchive.tar --create}
2180 @kbd{tar -farchive.tar -c}
2181 @kbd{tar c --file=archive.tar}
2182 @kbd{tar c -f archive.tar}
2183 @kbd{tar c -farchive.tar}
2184 @kbd{tar cf archive.tar}
2185 @kbd{tar f archive.tar --create}
2186 @kbd{tar f archive.tar -c}
2187 @kbd{tar fc archive.tar}
2190 On the other hand, the following commands are @emph{not} equivalent to
2194 @kbd{tar -f -c archive.tar}
2195 @kbd{tar -fc archive.tar}
2196 @kbd{tar -fcarchive.tar}
2197 @kbd{tar -farchive.tarc}
2198 @kbd{tar cfarchive.tar}
2202 These last examples mean something completely different from what the
2203 user intended (judging based on the example in the previous set which
2204 uses long options, whose intent is therefore very clear). The first
2205 four specify that the @command{tar} archive would be a file named
2206 @option{-c}, @samp{c}, @samp{carchive.tar} or @samp{archive.tarc},
2207 respectively. The first two examples also specify a single non-option,
2208 @var{name} argument having the value @samp{archive.tar}. The last
2209 example contains only old style option letters (repeating option
2210 @samp{c} twice), not all of which are meaningful (eg., @samp{.},
2211 @samp{h}, or @samp{i}), with no argument value. @FIXME{not sure i liked
2212 the first sentence of this paragraph..}
2215 @section All @command{tar} Options
2217 The coming manual sections contain an alphabetical listing of all
2218 @command{tar} operations and options, with brief descriptions and cross
2219 references to more in-depth explanations in the body of the manual.
2220 They also contain an alphabetically arranged table of the short option
2221 forms with their corresponding long option. You can use this table as
2222 a reference for deciphering @command{tar} commands in scripts.
2225 * Operation Summary::
2227 * Short Option Summary::
2230 @node Operation Summary
2231 @subsection Operations
2235 @opindex append, summary
2239 Appends files to the end of the archive. @xref{append}.
2241 @opindex catenate, summary
2245 Same as @option{--concatenate}. @xref{concatenate}.
2247 @opindex compare, summary
2251 Compares archive members with their counterparts in the file
2252 system, and reports differences in file size, mode, owner,
2253 modification date and contents. @xref{compare}.
2255 @opindex concatenate, summary
2259 Appends other @command{tar} archives to the end of the archive.
2262 @opindex create, summary
2266 Creates a new @command{tar} archive. @xref{create}.
2268 @opindex delete, summary
2271 Deletes members from the archive. Don't try this on a archive on a
2272 tape! @xref{delete}.
2274 @opindex diff, summary
2278 Same @option{--compare}. @xref{compare}.
2280 @opindex extract, summary
2284 Extracts members from the archive into the file system. @xref{extract}.
2286 @opindex get, summary
2290 Same as @option{--extract}. @xref{extract}.
2292 @opindex list, summary
2296 Lists the members in an archive. @xref{list}.
2298 @opindex update, summary
2302 Adds files to the end of the archive, but only if they are newer than
2303 their counterparts already in the archive, or if they do not already
2304 exist in the archive. @xref{update}.
2308 @node Option Summary
2309 @subsection @command{tar} Options
2313 @opindex absolute-names, summary
2314 @item --absolute-names
2317 Normally when creating an archive, @command{tar} strips an initial
2318 @samp{/} from member names. This option disables that behavior.
2321 @opindex after-date, summary
2324 (See @option{--newer}, @pxref{after})
2326 @opindex anchored, summary
2328 A pattern must match an initial subsequence of the name's components.
2329 @xref{controlling pattern-matching}.
2331 @opindex atime-preserve, summary
2332 @item --atime-preserve
2333 @itemx --atime-preserve=replace
2334 @itemx --atime-preserve=system
2336 Attempt to preserve the access time of files when reading them. This
2337 option currently is effective only on files that you own, unless you
2338 have superuser privileges.
2340 @option{--atime-preserve=replace} remembers the access time of a file
2341 before reading it, and then restores the access time afterwards. This
2342 may cause problems if other programs are reading the file at the same
2343 time, as the times of their accesses will be lost. On most platforms
2344 restoring the access time also requires @command{tar} to restore the
2345 data modification time too, so this option may also cause problems if
2346 other programs are writing the file at the same time. (Tar attempts
2347 to detect this situation, but cannot do so reliably due to race
2348 conditions.) Worse, on most platforms restoring the access time also
2349 updates the status change time, which means that this option is
2350 incompatible with incremental backups.
2352 @option{--atime-preserve=system} avoids changing time stamps on files,
2353 without interfering with time stamp updates
2354 caused by other programs, so it works better with incremental backups.
2355 However, it requires a special @code{O_NOATIME} option from the
2356 underlying operating and file system implementation, and it also requires
2357 that searching directories does not update their access times. As of
2358 this writing (November 2005) this works only with Linux, and only with
2359 Linux kernels 2.6.8 and later. Worse, there is currently no reliable
2360 way to know whether this feature actually works. Sometimes
2361 @command{tar} knows that it does not work, and if you use
2362 @option{--atime-preserve=system} then @command{tar} complains and
2363 exits right away. But other times @command{tar} might think that the
2364 option works when it actually does not.
2366 Currently @option{--atime-preserve} with no operand defaults to
2367 @option{--atime-preserve=replace}, but this may change in the future
2368 as support for @option{--atime-preserve=system} improves.
2370 If your operating system does not support
2371 @option{--atime-preserve=system}, you might be able to preserve access
2372 times reliably by by using the @command{mount} command. For example,
2373 you can mount the file system read-only, or access the file system via
2374 a read-only loopback mount, or use the @samp{noatime} mount option
2375 available on some systems. However, mounting typically requires
2376 superuser privileges and can be a pain to manage.
2378 @opindex backup, summary
2379 @item --backup=@var{backup-type}
2381 Rather than deleting files from the file system, @command{tar} will
2382 back them up using simple or numbered backups, depending upon
2383 @var{backup-type}. @xref{backup}.
2385 @opindex block-number, summary
2386 @item --block-number
2389 With this option present, @command{tar} prints error messages for read errors
2390 with the block number in the archive file. @xref{block-number}.
2392 @opindex blocking-factor, summary
2393 @item --blocking-factor=@var{blocking}
2394 @itemx -b @var{blocking}
2396 Sets the blocking factor @command{tar} uses to @var{blocking} x 512 bytes per
2397 record. @xref{Blocking Factor}.
2399 @opindex bzip2, summary
2403 This option tells @command{tar} to read or write archives through
2404 @code{bzip2}. @xref{gzip}.
2406 @opindex checkpoint, summary
2407 @item --checkpoint[=@var{number}]
2409 This option directs @command{tar} to print periodic checkpoint
2410 messages as it reads through the archive. It is intended for when you
2411 want a visual indication that @command{tar} is still running, but
2412 don't want to see @option{--verbose} output. For a detailed
2413 description, see @ref{Progress information}.
2415 @opindex check-links, summary
2418 If this option was given, @command{tar} will check the number of links
2419 dumped for each processed file. If this number does not match the
2420 total number of hard links for the file, a warning message will be
2421 output @footnote{Earlier versions of @GNUTAR{} understood @option{-l} as a
2422 synonym for @option{--one-file-system}. The current semantics, which
2423 complies to UNIX98, was introduced with version
2424 1.15.91. @xref{Changes}, for more information.}.
2426 @opindex compress, summary
2427 @opindex uncompress, summary
2432 @command{tar} will use the @command{compress} program when reading or
2433 writing the archive. This allows you to directly act on archives
2434 while saving space. @xref{gzip}.
2436 @opindex confirmation, summary
2437 @item --confirmation
2439 (See @option{--interactive}.) @xref{interactive}.
2441 @opindex delay-directory-restore, summary
2442 @item --delay-directory-restore
2444 Delay setting modification times and permissions of extracted
2445 directories until the end of extraction. @xref{Directory Modification Times and Permissions}.
2447 @opindex dereference, summary
2451 When creating a @command{tar} archive, @command{tar} will archive the
2452 file that a symbolic link points to, rather than archiving the
2453 symlink. @xref{dereference}.
2455 @opindex directory, summary
2456 @item --directory=@var{dir}
2459 When this option is specified, @command{tar} will change its current directory
2460 to @var{dir} before performing any operations. When this option is used
2461 during archive creation, it is order sensitive. @xref{directory}.
2463 @opindex exclude, summary
2464 @item --exclude=@var{pattern}
2466 When performing operations, @command{tar} will skip files that match
2467 @var{pattern}. @xref{exclude}.
2469 @opindex exclude-from, summary
2470 @item --exclude-from=@var{file}
2471 @itemx -X @var{file}
2473 Similar to @option{--exclude}, except @command{tar} will use the list of
2474 patterns in the file @var{file}. @xref{exclude}.
2476 @opindex exclude-caches, summary
2477 @item --exclude-caches
2479 Automatically excludes all directories
2480 containing a cache directory tag. @xref{exclude}.
2482 @opindex file, summary
2483 @item --file=@var{archive}
2484 @itemx -f @var{archive}
2486 @command{tar} will use the file @var{archive} as the @command{tar} archive it
2487 performs operations on, rather than @command{tar}'s compilation dependent
2488 default. @xref{file tutorial}.
2490 @opindex files-from, summary
2491 @item --files-from=@var{file}
2492 @itemx -T @var{file}
2494 @command{tar} will use the contents of @var{file} as a list of archive members
2495 or files to operate on, in addition to those specified on the
2496 command-line. @xref{files}.
2498 @opindex force-local, summary
2501 Forces @command{tar} to interpret the filename given to @option{--file}
2502 as a local file, even if it looks like a remote tape drive name.
2503 @xref{local and remote archives}.
2505 @opindex format, summary
2506 @item --format=@var{format}
2508 Selects output archive format. @var{Format} may be one of the
2513 Creates an archive that is compatible with Unix V7 @command{tar}.
2516 Creates an archive that is compatible with GNU @command{tar} version
2520 Creates archive in GNU tar 1.13 format. Basically it is the same as
2521 @samp{oldgnu} with the only difference in the way it handles long
2525 Creates a @acronym{POSIX.1-1988} compatible archive.
2528 Creates a @acronym{POSIX.1-2001 archive}.
2532 @xref{Formats}, for a detailed discussion of these formats.
2534 @opindex group, summary
2535 @item --group=@var{group}
2537 Files added to the @command{tar} archive will have a group id of @var{group},
2538 rather than the group from the source file. @var{group} is first decoded
2539 as a group symbolic name, but if this interpretation fails, it has to be
2540 a decimal numeric group ID. @FIXME-xref{}
2542 Also see the comments for the @option{--owner=@var{user}} option.
2544 @opindex gzip, summary
2545 @opindex gunzip, summary
2546 @opindex ungzip, summary
2552 This option tells @command{tar} to read or write archives through
2553 @command{gzip}, allowing @command{tar} to directly operate on several
2554 kinds of compressed archives transparently. @xref{gzip}.
2556 @opindex help, summary
2559 @command{tar} will print out a short message summarizing the operations and
2560 options to @command{tar} and exit. @xref{help}.
2562 @opindex ignore-case, summary
2564 Ignore case when matching member or file names with
2565 patterns. @xref{controlling pattern-matching}.
2567 @opindex ignore-command-error, summary
2568 @item --ignore-command-error
2569 Ignore exit codes of subprocesses. @xref{Writing to an External Program}.
2571 @opindex ignore-failed-read, summary
2572 @item --ignore-failed-read
2574 Do not exit unsuccessfully merely because an unreadable file was encountered.
2577 @opindex ignore-zeros, summary
2578 @item --ignore-zeros
2581 With this option, @command{tar} will ignore zeroed blocks in the
2582 archive, which normally signals EOF. @xref{Reading}.
2584 @opindex incremental, summary
2588 Used to inform @command{tar} that it is working with an old
2589 @acronym{GNU}-format incremental backup archive. It is intended
2590 primarily for backwards compatibility only. @xref{Incremental Dumps},
2591 for a detailed discussion of incremental archives.
2593 @opindex index-file, summary
2594 @item --index-file=@var{file}
2596 Send verbose output to @var{file} instead of to standard output.
2598 @opindex info-script, summary
2599 @opindex new-volume-script, summary
2600 @item --info-script=@var{script-file}
2601 @itemx --new-volume-script=@var{script-file}
2602 @itemx -F @var{script-file}
2604 When @command{tar} is performing multi-tape backups, @var{script-file} is run
2605 at the end of each tape. If @var{script-file} exits with nonzero status,
2606 @command{tar} fails immediately. @xref{info-script}, for a detailed
2607 discussion of @var{script-file}.
2609 @opindex interactive, summary
2611 @itemx --confirmation
2614 Specifies that @command{tar} should ask the user for confirmation before
2615 performing potentially destructive options, such as overwriting files.
2618 @opindex keep-newer-files, summary
2619 @item --keep-newer-files
2621 Do not replace existing files that are newer than their archive copies
2622 when extracting files from an archive.
2624 @opindex keep-old-files, summary
2625 @item --keep-old-files
2628 Do not overwrite existing files when extracting files from an archive.
2629 @xref{Keep Old Files}.
2631 @opindex label, summary
2632 @item --label=@var{name}
2633 @itemx -V @var{name}
2635 When creating an archive, instructs @command{tar} to write @var{name}
2636 as a name record in the archive. When extracting or listing archives,
2637 @command{tar} will only operate on archives that have a label matching
2638 the pattern specified in @var{name}. @xref{Tape Files}.
2640 @opindex listed-incremental, summary
2641 @item --listed-incremental=@var{snapshot-file}
2642 @itemx -g @var{snapshot-file}
2644 During a @option{--create} operation, specifies that the archive that
2645 @command{tar} creates is a new @acronym{GNU}-format incremental
2646 backup, using @var{snapshot-file} to determine which files to backup.
2647 With other operations, informs @command{tar} that the archive is in
2648 incremental format. @xref{Incremental Dumps}.
2650 @opindex mode, summary
2651 @item --mode=@var{permissions}
2653 When adding files to an archive, @command{tar} will use
2654 @var{permissions} for the archive members, rather than the permissions
2655 from the files. The program @command{chmod} and this @command{tar}
2656 option share the same syntax for what @var{permissions} might be.
2657 @xref{File permissions, Permissions, File permissions, fileutils,
2658 @acronym{GNU} file utilities}. This reference also has useful
2659 information for those not being overly familiar with the Unix
2662 Of course, @var{permissions} might be plainly specified as an octal number.
2663 However, by using generic symbolic modifications to mode bits, this allows
2664 more flexibility. For example, the value @samp{a+rw} adds read and write
2665 permissions for everybody, while retaining executable bits on directories
2666 or on any other file already marked as executable.
2668 @opindex multi-volume, summary
2669 @item --multi-volume
2672 Informs @command{tar} that it should create or otherwise operate on a
2673 multi-volume @command{tar} archive. @xref{Using Multiple Tapes}.
2675 @opindex new-volume-script, summary
2676 @item --new-volume-script
2680 @opindex seek, summary
2684 Assume that the archive media supports seeks to arbitrary
2685 locations. Usually @command{tar} determines automatically whether
2686 the archive can be seeked or not. This option is intended for use
2687 in cases when such recognition fails.
2689 @opindex newer, summary
2690 @item --newer=@var{date}
2691 @itemx --after-date=@var{date}
2694 When creating an archive, @command{tar} will only add files that have changed
2695 since @var{date}. If @var{date} begins with @samp{/} or @samp{.}, it
2696 is taken to be the name of a file whose data modification time specifies
2697 the date. @xref{after}.
2699 @opindex newer-mtime, summary
2700 @item --newer-mtime=@var{date}
2702 Like @option{--newer}, but add only files whose
2703 contents have changed (as opposed to just @option{--newer}, which will
2704 also back up files for which any status information has changed).
2706 @opindex no-anchored, summary
2708 An exclude pattern can match any subsequence of the name's components.
2709 @xref{controlling pattern-matching}.
2711 @opindex no-delay-directory-restore, summary
2712 @item --no-delay-directory-restore
2714 Setting modification times and permissions of extracted
2715 directories when all files from this directory has been
2716 extracted. This is the default. @xref{Directory Modification Times and Permissions}.
2718 @opindex no-ignore-case, summary
2719 @item --no-ignore-case
2720 Use case-sensitive matching.
2721 @xref{controlling pattern-matching}.
2723 @opindex no-ignore-command-error, summary
2724 @item --no-ignore-command-error
2725 Print warnings about subprocesses terminated with a non-zero exit
2726 code. @xref{Writing to an External Program}.
2728 @opindex no-overwrite-dir, summary
2729 @item --no-overwrite-dir
2731 Preserve metadata of existing directories when extracting files
2732 from an archive. @xref{Overwrite Old Files}.
2734 @opindex no-quote-chars, summary
2735 @item --no-quote-chars=@var{string}
2736 Remove characters listed in @var{string} from the list of quoted
2737 characters set by the previous @option{--quote-chars} option
2738 (@pxref{quoting styles}).
2740 @opindex no-recursion, summary
2741 @item --no-recursion
2743 With this option, @command{tar} will not recurse into directories.
2746 @opindex no-same-owner, summary
2747 @item --no-same-owner
2750 When extracting an archive, do not attempt to preserve the owner
2751 specified in the @command{tar} archive. This the default behavior
2754 @opindex no-same-permissions, summary
2755 @item --no-same-permissions
2757 When extracting an archive, subtract the user's umask from files from
2758 the permissions specified in the archive. This is the default behavior
2761 @opindex no-wildcards, summary
2762 @item --no-wildcards
2763 Do not use wildcards.
2764 @xref{controlling pattern-matching}.
2766 @opindex no-wildcards-match-slash, summary
2767 @item --no-wildcards-match-slash
2768 Wildcards do not match @samp{/}.
2769 @xref{controlling pattern-matching}.
2771 @opindex null, summary
2774 When @command{tar} is using the @option{--files-from} option, this option
2775 instructs @command{tar} to expect filenames terminated with @option{NUL}, so
2776 @command{tar} can correctly work with file names that contain newlines.
2779 @opindex numeric-owner, summary
2780 @item --numeric-owner
2782 This option will notify @command{tar} that it should use numeric user
2783 and group IDs when creating a @command{tar} file, rather than names.
2787 When extracting files, this option is a synonym for
2788 @option{--no-same-owner}, i.e. it prevents @command{tar} from
2789 restoring ownership of files being extracted.
2791 When creating an archive, @option{-o} is a synonym for
2792 @option{--old-archive}. This behavior is for compatibility
2793 with previous versions of @GNUTAR{}, and will be
2794 removed in the future releases.
2796 @xref{Changes}, for more information.
2798 @opindex occurrence, summary
2799 @item --occurrence[=@var{number}]
2801 This option can be used in conjunction with one of the subcommands
2802 @option{--delete}, @option{--diff}, @option{--extract} or
2803 @option{--list} when a list of files is given either on the command
2804 line or via @option{-T} option.
2806 This option instructs @command{tar} to process only the @var{number}th
2807 occurrence of each named file. @var{Number} defaults to 1, so
2810 tar -x -f archive.tar --occurrence filename
2814 will extract the first occurrence of @file{filename} from @file{archive.tar}
2815 and will terminate without scanning to the end of the archive.
2817 @opindex old-archive, summary
2819 Synonym for @option{--format=v7}.
2821 @opindex one-file-system, summary
2822 @item --one-file-system
2823 Used when creating an archive. Prevents @command{tar} from recursing into
2824 directories that are on different file systems from the current
2825 directory @footnote{Earlier versions of @GNUTAR{} understood @option{-l} as a
2826 synonym for @option{--one-file-system}. This has changed in version
2827 1.15.91. @xref{Changes}, for more information.}.
2829 @opindex overwrite, summary
2832 Overwrite existing files and directory metadata when extracting files
2833 from an archive. @xref{Overwrite Old Files}.
2835 @opindex overwrite-dir, summary
2836 @item --overwrite-dir
2838 Overwrite the metadata of existing directories when extracting files
2839 from an archive. @xref{Overwrite Old Files}.
2841 @opindex owner, summary
2842 @item --owner=@var{user}
2844 Specifies that @command{tar} should use @var{user} as the owner of members
2845 when creating archives, instead of the user associated with the source
2846 file. @var{user} is first decoded as a user symbolic name, but if
2847 this interpretation fails, it has to be a decimal numeric user ID.
2850 There is no value indicating a missing number, and @samp{0} usually means
2851 @code{root}. Some people like to force @samp{0} as the value to offer in
2852 their distributions for the owner of files, because the @code{root} user is
2853 anonymous anyway, so that might as well be the owner of anonymous archives.
2855 This option does not affect extraction from archives.
2857 @opindex transform, summary
2858 @item --transform=@var{sed-expr}
2860 Transform file or member names using @command{sed} replacement expression
2861 @var{sed-expr}. For example,
2864 $ @kbd{tar cf archive.tar --transform 's,^\./,usr/,' .}
2868 will add to @file{archive} files from the current working directory,
2869 replacing initial @samp{./} prefix with @samp{usr/}. For the detailed
2870 discussion, @xref{transform}.
2872 To see transformed member names in verbose listings, use
2873 @option{--show-transformed-names} option
2874 (@pxref{show-transformed-names}).
2876 @opindex quote-chars, summary
2877 @item --quote-chars=@var{string}
2878 Always quote characters from @var{string}, even if the selected
2879 quoting style would not quote them (@pxref{quoting styles}).
2881 @opindex quoting-style, summary
2882 @item --quoting-style=@var{style}
2883 Set quoting style to use when printing member and file names
2884 (@pxref{quoting styles}). Valid @var{style} values are:
2885 @code{literal}, @code{shell}, @code{shell-always}, @code{c},
2886 @code{escape}, @code{locale}, and @code{clocale}. Default quoting
2887 style is @code{escape}, unless overridden while configuring the
2890 @opindex pax-option, summary
2891 @item --pax-option=@var{keyword-list}
2892 This option is meaningful only with @acronym{POSIX.1-2001} archives
2893 (@pxref{posix}). It modifies the way @command{tar} handles the
2894 extended header keywords. @var{Keyword-list} is a comma-separated
2895 list of keyword options. @xref{PAX keywords}, for a detailed
2898 @opindex portability, summary
2900 @itemx --old-archive
2901 Synonym for @option{--format=v7}.
2903 @opindex posix, summary
2905 Same as @option{--format=posix}.
2907 @opindex preserve, summary
2910 Synonymous with specifying both @option{--preserve-permissions} and
2911 @option{--same-order}. @xref{Setting Access Permissions}.
2913 @opindex preserve-order, summary
2914 @item --preserve-order
2916 (See @option{--same-order}; @pxref{Reading}.)
2918 @opindex preserve-permissions, summary
2919 @opindex same-permissions, summary
2920 @item --preserve-permissions
2921 @itemx --same-permissions
2924 When @command{tar} is extracting an archive, it normally subtracts the
2925 users' umask from the permissions specified in the archive and uses
2926 that number as the permissions to create the destination file.
2927 Specifying this option instructs @command{tar} that it should use the
2928 permissions directly from the archive. @xref{Setting Access Permissions}.
2930 @opindex read-full-records, summary
2931 @item --read-full-records
2934 Specifies that @command{tar} should reblock its input, for reading
2935 from pipes on systems with buggy implementations. @xref{Reading}.
2937 @opindex record-size, summary
2938 @item --record-size=@var{size}
2940 Instructs @command{tar} to use @var{size} bytes per record when accessing the
2941 archive. @xref{Blocking Factor}.
2943 @opindex recursion, summary
2946 With this option, @command{tar} recurses into directories.
2949 @opindex recursive-unlink, summary
2950 @item --recursive-unlink
2953 directory hierarchies before extracting directories of the same name
2954 from the archive. @xref{Recursive Unlink}.
2956 @opindex remove-files, summary
2957 @item --remove-files
2959 Directs @command{tar} to remove the source file from the file system after
2960 appending it to an archive. @xref{remove files}.
2962 @opindex restrict, summary
2965 Disable use of some potentially harmful @command{tar} options.
2966 Currently this option disables shell invocaton from multi-volume menu
2967 (@pxref{Using Multiple Tapes}).
2969 @opindex rmt-command, summary
2970 @item --rmt-command=@var{cmd}
2972 Notifies @command{tar} that it should use @var{cmd} instead of
2973 the default @file{/usr/libexec/rmt} (@pxref{Remote Tape Server}).
2975 @opindex rsh-command, summary
2976 @item --rsh-command=@var{cmd}
2978 Notifies @command{tar} that is should use @var{cmd} to communicate with remote
2979 devices. @xref{Device}.
2981 @opindex same-order, summary
2983 @itemx --preserve-order
2986 This option is an optimization for @command{tar} when running on machines with
2987 small amounts of memory. It informs @command{tar} that the list of file
2988 arguments has already been sorted to match the order of files in the
2989 archive. @xref{Reading}.
2991 @opindex same-owner, summary
2994 When extracting an archive, @command{tar} will attempt to preserve the owner
2995 specified in the @command{tar} archive with this option present.
2996 This is the default behavior for the superuser; this option has an
2997 effect only for ordinary users. @xref{Attributes}.
2999 @opindex same-permissions, summary
3000 @item --same-permissions
3002 (See @option{--preserve-permissions}; @pxref{Setting Access Permissions}.)
3004 @opindex show-defaults, summary
3005 @item --show-defaults
3007 Displays the default options used by @command{tar} and exits
3008 successfully. This option is intended for use in shell scripts.
3009 Here is an example of what you can see using this option:
3012 $ tar --show-defaults
3013 --format=gnu -f- -b20 --quoting-style=escape \
3014 --rmt-command=/usr/libexec/rmt --rsh-command=/usr/bin/rsh
3017 @opindex show-omitted-dirs, summary
3018 @item --show-omitted-dirs
3020 Instructs @command{tar} to mention directories its skipping over when
3021 operating on a @command{tar} archive. @xref{show-omitted-dirs}.
3023 @opindex show-transformed-names, summary
3024 @opindex show-stored-names, summary
3025 @item --show-transformed-names
3026 @itemx --show-stored-names
3028 Display file or member names after applying any transformations
3029 (@pxref{transform}). In particular, when used in conjunction with one of
3030 archive creation operations it instructs tar to list the member names
3031 stored in the archive, as opposed to the actual file
3032 names. @xref{listing member and file names}.
3034 @opindex sparse, summary
3038 Invokes a @acronym{GNU} extension when adding files to an archive that handles
3039 sparse files efficiently. @xref{sparse}.
3041 @opindex starting-file, summary
3042 @item --starting-file=@var{name}
3043 @itemx -K @var{name}
3045 This option affects extraction only; @command{tar} will skip extracting
3046 files in the archive until it finds one that matches @var{name}.
3049 @opindex strip-components, summary
3050 @item --strip-components=@var{number}
3051 Strip given @var{number} of leading components from file names before
3052 extraction.@footnote{This option was called @option{--strip-path} in
3053 version 1.14.} For example, if archive @file{archive.tar} contained
3054 @file{/some/file/name}, then running
3057 tar --extract --file archive.tar --strip-components=2
3061 would extract this file to file @file{name}.
3063 @opindex suffix, summary
3064 @item --suffix=@var{suffix}
3066 Alters the suffix @command{tar} uses when backing up files from the default
3067 @samp{~}. @xref{backup}.
3069 @opindex tape-length, summary
3070 @item --tape-length=@var{num}
3073 Specifies the length of tapes that @command{tar} is writing as being
3074 @w{@var{num} x 1024} bytes long. @xref{Using Multiple Tapes}.
3076 @opindex test-label, summary
3079 Reads the volume label. If an argument is specified, test whether it
3080 matches the volume label. @xref{--test-label option}.
3082 @opindex to-command, summary
3083 @item --to-command=@var{command}
3085 During extraction @command{tar} will pipe extracted files to the
3086 standard input of @var{command}. @xref{Writing to an External Program}.
3088 @opindex to-stdout, summary
3092 During extraction, @command{tar} will extract files to stdout rather
3093 than to the file system. @xref{Writing to Standard Output}.
3095 @opindex totals, summary
3096 @item --totals[=@var{signo}]
3098 Displays the total number of bytes transferred when processing an
3099 archive. If an argument is given, these data are displayed on
3100 request, when signal @var{signo} is delivered to @command{tar}.
3103 @opindex touch, summary
3107 Sets the data modification time of extracted files to the extraction time,
3108 rather than the data modification time stored in the archive.
3109 @xref{Data Modification Times}.
3111 @opindex uncompress, summary
3114 (See @option{--compress}. @pxref{gzip})
3116 @opindex ungzip, summary
3119 (See @option{--gzip}. @pxref{gzip})
3121 @opindex unlink-first, summary
3122 @item --unlink-first
3125 Directs @command{tar} to remove the corresponding file from the file
3126 system before extracting it from the archive. @xref{Unlink First}.
3128 @opindex use-compress-program, summary
3129 @item --use-compress-program=@var{prog}
3131 Instructs @command{tar} to access the archive through @var{prog}, which is
3132 presumed to be a compression program of some sort. @xref{gzip}.
3134 @opindex utc, summary
3137 Display file modification dates in @acronym{UTC}. This option implies
3140 @opindex verbose, summary
3144 Specifies that @command{tar} should be more verbose about the operations its
3145 performing. This option can be specified multiple times for some
3146 operations to increase the amount of information displayed.
3149 @opindex verify, summary
3153 Verifies that the archive was correctly written when creating an
3154 archive. @xref{verify}.
3156 @opindex version, summary
3159 Print information about the program's name, version, origin and legal
3160 status, all on standard output, and then exit successfully.
3163 @opindex volno-file, summary
3164 @item --volno-file=@var{file}
3166 Used in conjunction with @option{--multi-volume}. @command{tar} will keep track
3167 of which volume of a multi-volume archive its working in @var{file}.
3170 @opindex wildcards, summary
3172 Use wildcards when matching member names with patterns.
3173 @xref{controlling pattern-matching}.
3175 @opindex wildcards-match-slash, summary
3176 @item --wildcards-match-slash
3177 Wildcards match @samp{/}.
3178 @xref{controlling pattern-matching}.
3181 @node Short Option Summary
3182 @subsection Short Options Cross Reference
3184 Here is an alphabetized list of all of the short option forms, matching
3185 them with the equivalent long option.
3191 @option{--concatenate}
3195 @option{--read-full-records}
3199 @option{--directory}
3203 @option{--info-script}
3207 @option{--incremental}
3211 @option{--starting-file}
3215 @option{--tape-length}
3219 @option{--multi-volume}
3227 @option{--to-stdout}
3231 @option{--absolute-names}
3235 @option{--block-number}
3243 @option{--files-from}
3247 @option{--unlink-first}
3259 @option{--exclude-from}
3267 @option{--blocking-factor}
3283 @option{--listed-incremental}
3287 @option{--dereference}
3291 @option{--ignore-zeros}
3299 @option{--keep-old-files}
3303 @option{--one-file-system}. Use of this short option is deprecated. It
3304 is retained for compatibility with the earlier versions of GNU
3305 @command{tar}, and will be changed in future releases.
3307 @xref{Changes}, for more information.
3315 When creating --- @option{--no-same-owner}, when extracting ---
3316 @option{--portability}.
3318 The later usage is deprecated. It is retained for compatibility with
3319 the earlier versions of @GNUTAR{}. In the future releases
3320 @option{-o} will be equivalent to @option{--no-same-owner} only.
3324 @option{--preserve-permissions}
3332 @option{--same-order}
3348 @option{--interactive}
3361 @section @GNUTAR{} documentation
3363 @cindex Getting program version number
3365 @cindex Version of the @command{tar} program
3366 Being careful, the first thing is really checking that you are using
3367 @GNUTAR{}, indeed. The @option{--version} option
3368 causes @command{tar} to print information about its name, version,
3369 origin and legal status, all on standard output, and then exit
3370 successfully. For example, @w{@samp{tar --version}} might print:
3373 tar (GNU tar) 1.15.2
3374 Copyright (C) 2006 Free Software Foundation, Inc.
3375 This is free software. You may redistribute copies of it under the terms of
3376 the GNU General Public License <http://www.gnu.org/licenses/gpl.html>.
3377 There is NO WARRANTY, to the extent permitted by law.
3379 Written by John Gilmore and Jay Fenlason.
3383 The first occurrence of @samp{tar} in the result above is the program
3384 name in the package (for example, @command{rmt} is another program),
3385 while the second occurrence of @samp{tar} is the name of the package
3386 itself, containing possibly many programs. The package is currently
3387 named @samp{tar}, after the name of the main program it
3388 contains@footnote{There are plans to merge the @command{cpio} and
3389 @command{tar} packages into a single one which would be called
3390 @code{paxutils}. So, who knows if, one of this days, the
3391 @option{--version} would not output @w{@samp{tar (@acronym{GNU}
3394 @cindex Obtaining help
3395 @cindex Listing all @command{tar} options
3396 @opindex help, introduction
3397 Another thing you might want to do is checking the spelling or meaning
3398 of some particular @command{tar} option, without resorting to this
3399 manual, for once you have carefully read it. @GNUTAR{}
3400 has a short help feature, triggerable through the
3401 @option{--help} option. By using this option, @command{tar} will
3402 print a usage message listing all available options on standard
3403 output, then exit successfully, without doing anything else and
3404 ignoring all other options. Even if this is only a brief summary, it
3405 may be several screens long. So, if you are not using some kind of
3406 scrollable window, you might prefer to use something like:
3409 $ @kbd{tar --help | less}
3413 presuming, here, that you like using @command{less} for a pager. Other
3414 popular pagers are @command{more} and @command{pg}. If you know about some
3415 @var{keyword} which interests you and do not want to read all the
3416 @option{--help} output, another common idiom is doing:
3419 tar --help | grep @var{keyword}
3423 for getting only the pertinent lines. Notice, however, that some
3424 @command{tar} options have long description lines and the above
3425 command will list only the first of them.
3427 The exact look of the option summary displayed by @kbd{tar --help} is
3428 configurable. @xref{Configuring Help Summary}, for a detailed description.
3431 If you only wish to check the spelling of an option, running @kbd{tar
3432 --usage} may be a better choice. This will display a terse list of
3433 @command{tar} option without accompanying explanations.
3435 The short help output is quite succinct, and you might have to get
3436 back to the full documentation for precise points. If you are reading
3437 this paragraph, you already have the @command{tar} manual in some
3438 form. This manual is available in a variety of forms from
3439 @url{http://www.gnu.org/software/tar/manual}. It may printed out of the @GNUTAR{}
3440 distribution, provided you have @TeX{} already installed somewhere,
3441 and a laser printer around. Just configure the distribution, execute
3442 the command @w{@samp{make dvi}}, then print @file{doc/tar.dvi} the
3443 usual way (contact your local guru to know how). If @GNUTAR{}
3444 has been conveniently installed at your place, this
3445 manual is also available in interactive, hypertextual form as an Info
3446 file. Just call @w{@samp{info tar}} or, if you do not have the
3447 @command{info} program handy, use the Info reader provided within
3448 @acronym{GNU} Emacs, calling @samp{tar} from the main Info menu.
3450 There is currently no @code{man} page for @GNUTAR{}.
3451 If you observe such a @code{man} page on the system you are running,
3452 either it does not belong to @GNUTAR{}, or it has not
3453 been produced by @acronym{GNU}. Some package maintainers convert
3454 @kbd{tar --help} output to a man page, using @command{help2man}. In
3455 any case, please bear in mind that the authoritative source of
3456 information about @GNUTAR{} is this Texinfo documentation.
3459 @section Obtaining @GNUTAR{} default values
3461 @opindex show-defaults
3462 @GNUTAR{} has some predefined defaults that are used when you do not
3463 explicitely specify another values. To obtain a list of such
3464 defaults, use @option{--show-defaults} option. This will output the
3465 values in the form of @command{tar} command line options:
3469 @kbd{tar --show-defaults}
3470 --format=gnu -f- -b20 --rmt-command=/etc/rmt --rsh-command=/usr/bin/rsh
3475 The above output shows that this version of @GNUTAR{} defaults to
3476 using @samp{gnu} archive format (@pxref{Formats}), it uses standard
3477 output as the archive, if no @option{--file} option has been given
3478 (@pxref{file tutorial}), the default blocking factor is 20
3479 (@pxref{Blocking Factor}). It also shows the default locations where
3480 @command{tar} will look for @command{rmt} and @command{rsh} binaries.
3483 @section Checking @command{tar} progress
3485 Typically, @command{tar} performs most operations without reporting any
3486 information to the user except error messages. When using @command{tar}
3487 with many options, particularly ones with complicated or
3488 difficult-to-predict behavior, it is possible to make serious mistakes.
3489 @command{tar} provides several options that make observing @command{tar}
3490 easier. These options cause @command{tar} to print information as it
3491 progresses in its job, and you might want to use them just for being
3492 more careful about what is going on, or merely for entertaining
3493 yourself. If you have encountered a problem when operating on an
3494 archive, however, you may need more information than just an error
3495 message in order to solve the problem. The following options can be
3496 helpful diagnostic tools.
3498 @cindex Verbose operation
3500 Normally, the @option{--list} (@option{-t}) command to list an archive
3501 prints just the file names (one per line) and the other commands are
3502 silent. When used with most operations, the @option{--verbose}
3503 (@option{-v}) option causes @command{tar} to print the name of each
3504 file or archive member as it is processed. This and the other options
3505 which make @command{tar} print status information can be useful in
3506 monitoring @command{tar}.
3508 With @option{--create} or @option{--extract}, @option{--verbose} used
3509 once just prints the names of the files or members as they are processed.
3510 Using it twice causes @command{tar} to print a longer listing
3511 (@xref{verbose member listing}, for the description) for each member.
3512 Since @option{--list} already prints the names of the members,
3513 @option{--verbose} used once with @option{--list} causes @command{tar}
3514 to print an @samp{ls -l} type listing of the files in the archive.
3515 The following examples both extract members with long list output:
3518 $ @kbd{tar --extract --file=archive.tar --verbose --verbose}
3519 $ @kbd{tar xvvf archive.tar}
3522 Verbose output appears on the standard output except when an archive is
3523 being written to the standard output, as with @samp{tar --create
3524 --file=- --verbose} (@samp{tar cfv -}, or even @samp{tar cv}---if the
3525 installer let standard output be the default archive). In that case
3526 @command{tar} writes verbose output to the standard error stream.
3528 If @option{--index-file=@var{file}} is specified, @command{tar} sends
3529 verbose output to @var{file} rather than to standard output or standard
3533 @cindex Obtaining total status information
3535 The @option{--totals} option causes @command{tar} to print on the
3536 standard error the total amount of bytes transferred when processing
3537 an archive. When creating or appending to an archive, this option
3538 prints the number of bytes written to the archive and the average
3539 speed at which they have been written, e.g.:
3543 $ @kbd{tar -c -f archive.tar --totals /home}
3544 Total bytes written: 7924664320 (7.4GiB, 85MiB/s)
3548 When reading an archive, this option displays the number of bytes
3553 $ @kbd{tar -x -f archive.tar --totals}
3554 Total bytes read: 7924664320 (7.4GiB, 95MiB/s)
3558 Finally, when deleting from an archive, the @option{--totals} option
3559 displays both numbers plus number of bytes removed from the archive:
3563 $ @kbd{tar --delete -f foo.tar --totals --wildcards '*~'}
3564 Total bytes read: 9543680 (9.2MiB, 201MiB/s)
3565 Total bytes written: 3829760 (3.7MiB, 81MiB/s)
3566 Total bytes deleted: 1474048
3570 You can also obtain this information on request. When
3571 @option{--totals} is used with an argument, this argument is
3572 interpreted as a symbolic name of a signal, upon delivery of which the
3573 statistics is to be printed:
3576 @item --totals=@var{signo}
3577 Print statistics upon delivery of signal @var{signo}. Valid arguments
3578 are: @code{SIGHUP}, @code{SIGQUIT}, @code{SIGINT}, @code{SIGUSR1} and
3579 @code{SIGUSR2}. Shortened names without @samp{SIG} prefix are also
3583 Both forms of @option{--totals} option can be used simultaneously.
3584 Thus, @kbd{tar -x --totals --totals=USR1} instructs @command{tar} to
3585 extract all members from its default archive and print statistics
3586 after finishing the extraction, as well as when receiving signal
3589 @anchor{Progress information}
3590 @cindex Progress information
3592 The @option{--checkpoint} option prints an occasional message
3593 as @command{tar} reads or writes the archive. It is designed for
3594 those who don't need the more detailed (and voluminous) output of
3595 @option{--block-number} (@option{-R}), but do want visual confirmation
3596 that @command{tar} is actually making forward progress. By default it
3597 prints a message each 10 records read or written. This can be changed
3598 by giving it a numeric argument after an equal sign:
3601 $ @kbd{tar -c --checkpoint=1000} /var
3602 tar: Write checkpoint 1000
3603 tar: Write checkpoint 2000
3604 tar: Write checkpoint 3000
3607 This example shows the default checkpoint message used by
3608 @command{tar}. If you place a dot immediately after the equal
3609 sign, it will print a @samp{.} at each checkpoint. For example:
3612 $ @kbd{tar -c --checkpoint=.1000} /var
3616 @opindex show-omitted-dirs
3617 @anchor{show-omitted-dirs}
3618 The @option{--show-omitted-dirs} option, when reading an archive---with
3619 @option{--list} or @option{--extract}, for example---causes a message
3620 to be printed for each directory in the archive which is skipped.
3621 This happens regardless of the reason for skipping: the directory might
3622 not have been named on the command line (implicitly or explicitly),
3623 it might be excluded by the use of the
3624 @option{--exclude=@var{pattern}} option, or some other reason.
3626 @opindex block-number
3627 @cindex Block number where error occurred
3628 @anchor{block-number}
3629 If @option{--block-number} (@option{-R}) is used, @command{tar} prints, along with
3630 every message it would normally produce, the block number within the
3631 archive where the message was triggered. Also, supplementary messages
3632 are triggered when reading blocks full of NULs, or when hitting end of
3633 file on the archive. As of now, if the archive if properly terminated
3634 with a NUL block, the reading of the file may stop before end of file
3635 is met, so the position of end of file will not usually show when
3636 @option{--block-number} (@option{-R}) is used. Note that @GNUTAR{}
3637 drains the archive before exiting when reading the
3638 archive from a pipe.
3640 @cindex Error message, block number of
3641 This option is especially useful when reading damaged archives, since
3642 it helps pinpoint the damaged sections. It can also be used with
3643 @option{--list} (@option{-t}) when listing a file-system backup tape, allowing you to
3644 choose among several backup tapes when retrieving a file later, in
3645 favor of the tape where the file appears earliest (closest to the
3646 front of the tape). @xref{backup}.
3649 @section Asking for Confirmation During Operations
3650 @cindex Interactive operation
3652 Typically, @command{tar} carries out a command without stopping for
3653 further instructions. In some situations however, you may want to
3654 exclude some files and archive members from the operation (for instance
3655 if disk or storage space is tight). You can do this by excluding
3656 certain files automatically (@pxref{Choosing}), or by performing
3657 an operation interactively, using the @option{--interactive} (@option{-w}) option.
3658 @command{tar} also accepts @option{--confirmation} for this option.
3660 @opindex interactive
3661 When the @option{--interactive} (@option{-w}) option is specified, before
3662 reading, writing, or deleting files, @command{tar} first prints a message
3663 for each such file, telling what operation it intends to take, then asks
3664 for confirmation on the terminal. The actions which require
3665 confirmation include adding a file to the archive, extracting a file
3666 from the archive, deleting a file from the archive, and deleting a file
3667 from disk. To confirm the action, you must type a line of input
3668 beginning with @samp{y}. If your input line begins with anything other
3669 than @samp{y}, @command{tar} skips that file.
3671 If @command{tar} is reading the archive from the standard input,
3672 @command{tar} opens the file @file{/dev/tty} to support the interactive
3675 Verbose output is normally sent to standard output, separate from
3676 other error messages. However, if the archive is produced directly
3677 on standard output, then verbose output is mixed with errors on
3678 @code{stderr}. Producing the archive on standard output may be used
3679 as a way to avoid using disk space, when the archive is soon to be
3680 consumed by another process reading it, say. Some people felt the need
3681 of producing an archive on stdout, still willing to segregate between
3682 verbose output and error output. A possible approach would be using a
3683 named pipe to receive the archive, and having the consumer process to
3684 read from that named pipe. This has the advantage of letting standard
3685 output free to receive verbose output, all separate from errors.
3688 @chapter @GNUTAR{} Operations
3701 @section Basic @GNUTAR{} Operations
3703 The basic @command{tar} operations, @option{--create} (@option{-c}),
3704 @option{--list} (@option{-t}) and @option{--extract} (@option{--get},
3705 @option{-x}), are currently presented and described in the tutorial
3706 chapter of this manual. This section provides some complementary notes
3707 for these operations.
3710 @opindex create, complementary notes
3714 Creating an empty archive would have some kind of elegance. One can
3715 initialize an empty archive and later use @option{--append}
3716 (@option{-r}) for adding all members. Some applications would not
3717 welcome making an exception in the way of adding the first archive
3718 member. On the other hand, many people reported that it is
3719 dangerously too easy for @command{tar} to destroy a magnetic tape with
3720 an empty archive@footnote{This is well described in @cite{Unix-haters
3721 Handbook}, by Simson Garfinkel, Daniel Weise & Steven Strassmann, IDG
3722 Books, ISBN 1-56884-203-1.}. The two most common errors are:
3726 Mistakingly using @code{create} instead of @code{extract}, when the
3727 intent was to extract the full contents of an archive. This error
3728 is likely: keys @kbd{c} and @kbd{x} are right next to each other on
3729 the QWERTY keyboard. Instead of being unpacked, the archive then
3730 gets wholly destroyed. When users speak about @dfn{exploding} an
3731 archive, they usually mean something else :-).
3734 Forgetting the argument to @code{file}, when the intent was to create
3735 an archive with a single file in it. This error is likely because a
3736 tired user can easily add the @kbd{f} key to the cluster of option
3737 letters, by the mere force of habit, without realizing the full
3738 consequence of doing so. The usual consequence is that the single
3739 file, which was meant to be saved, is rather destroyed.
3742 So, recognizing the likelihood and the catastrophical nature of these
3743 errors, @GNUTAR{} now takes some distance from elegance, and
3744 cowardly refuses to create an archive when @option{--create} option is
3745 given, there are no arguments besides options, and
3746 @option{--files-from} (@option{-T}) option is @emph{not} used. To get
3747 around the cautiousness of @GNUTAR{} and nevertheless create an
3748 archive with nothing in it, one may still use, as the value for the
3749 @option{--files-from} option, a file with no names in it, as shown in
3750 the following commands:
3753 @kbd{tar --create --file=empty-archive.tar --files-from=/dev/null}
3754 @kbd{tar cfT empty-archive.tar /dev/null}
3757 @opindex extract, complementary notes
3762 A socket is stored, within a @GNUTAR{} archive, as a pipe.
3764 @item @option{--list} (@option{-t})
3766 @GNUTAR{} now shows dates as @samp{1996-08-30},
3767 while it used to show them as @samp{Aug 30 1996}. Preferably,
3768 people should get used to ISO 8601 dates. Local American dates should
3769 be made available again with full date localization support, once
3770 ready. In the meantime, programs not being localizable for dates
3771 should prefer international dates, that's really the way to go.
3773 Look up @url{http://www.ft.uni-erlangen.de/~mskuhn/iso-time.html} if you
3774 are curious, it contains a detailed explanation of the ISO 8601 standard.
3779 @section Advanced @GNUTAR{} Operations
3781 Now that you have learned the basics of using @GNUTAR{}, you may want
3782 to learn about further ways in which @command{tar} can help you.
3784 This chapter presents five, more advanced operations which you probably
3785 won't use on a daily basis, but which serve more specialized functions.
3786 We also explain the different styles of options and why you might want
3787 to use one or another, or a combination of them in your @command{tar}
3788 commands. Additionally, this chapter includes options which allow you to
3789 define the output from @command{tar} more carefully, and provide help and
3790 error correction in special circumstances.
3792 @FIXME{check this after the chapter is actually revised to make sure
3793 it still introduces the info in the chapter correctly : ).}
3805 @subsection The Five Advanced @command{tar} Operations
3808 In the last chapter, you learned about the first three operations to
3809 @command{tar}. This chapter presents the remaining five operations to
3810 @command{tar}: @option{--append}, @option{--update}, @option{--concatenate},
3811 @option{--delete}, and @option{--compare}.
3813 You are not likely to use these operations as frequently as those
3814 covered in the last chapter; however, since they perform specialized
3815 functions, they are quite useful when you do need to use them. We
3816 will give examples using the same directory and files that you created
3817 in the last chapter. As you may recall, the directory is called
3818 @file{practice}, the files are @samp{jazz}, @samp{blues}, @samp{folk},
3819 @samp{rock}, and the two archive files you created are
3820 @samp{collection.tar} and @samp{music.tar}.
3822 We will also use the archive files @samp{afiles.tar} and
3823 @samp{bfiles.tar}. @samp{afiles.tar} contains the members @samp{apple},
3824 @samp{angst}, and @samp{aspic}. @samp{bfiles.tar} contains the members
3825 @samp{./birds}, @samp{baboon}, and @samp{./box}.
3827 Unless we state otherwise, all practicing you do and examples you follow
3828 in this chapter will take place in the @file{practice} directory that
3829 you created in the previous chapter; see @ref{prepare for examples}.
3830 (Below in this section, we will remind you of the state of the examples
3831 where the last chapter left them.)
3833 The five operations that we will cover in this chapter are:
3838 Add new entries to an archive that already exists.
3841 Add more recent copies of archive members to the end of an archive, if
3846 Add one or more pre-existing archives to the end of another archive.
3848 Delete items from an archive (does not work on tapes).
3852 Compare archive members to their counterparts in the file system.
3856 @subsection How to Add Files to Existing Archives: @option{--append}
3860 If you want to add files to an existing archive, you don't need to
3861 create a new archive; you can use @option{--append} (@option{-r}).
3862 The archive must already exist in order to use @option{--append}. (A
3863 related operation is the @option{--update} operation; you can use this
3864 to add newer versions of archive members to an existing archive. To learn how to
3865 do this with @option{--update}, @pxref{update}.)
3867 If you use @option{--append} to add a file that has the same name as an
3868 archive member to an archive containing that archive member, then the
3869 old member is not deleted. What does happen, however, is somewhat
3870 complex. @command{tar} @emph{allows} you to have infinite number of files
3871 with the same name. Some operations treat these same-named members no
3872 differently than any other set of archive members: for example, if you
3873 view an archive with @option{--list} (@option{-t}), you will see all
3874 of those members listed, with their data modification times, owners, etc.
3876 Other operations don't deal with these members as perfectly as you might
3877 prefer; if you were to use @option{--extract} to extract the archive,
3878 only the most recently added copy of a member with the same name as four
3879 other members would end up in the working directory. This is because
3880 @option{--extract} extracts an archive in the order the members appeared
3881 in the archive; the most recently archived members will be extracted
3882 last. Additionally, an extracted member will @emph{replace} a file of
3883 the same name which existed in the directory already, and @command{tar}
3884 will not prompt you about this@footnote{Unless you give it
3885 @option{--keep-old-files} option, or the disk copy is newer than the
3886 the one in the archive and you invoke @command{tar} with
3887 @option{--keep-newer-files} option}. Thus, only the most recently archived
3888 member will end up being extracted, as it will replace the one
3889 extracted before it, and so on.
3891 There exists a special option that allows you to get around this
3892 behavior and extract (or list) only a particular copy of the file.
3893 This is @option{--occurrence} option. If you run @command{tar} with
3894 this option, it will extract only the first copy of the file. You
3895 may also give this option an argument specifying the number of
3896 copy to be extracted. Thus, for example if the archive
3897 @file{archive.tar} contained three copies of file @file{myfile}, then
3901 tar --extract --file archive.tar --occurrence=2 myfile
3905 would extract only the second copy. @xref{Option
3906 Summary,---occurrence}, for the description of @option{--occurrence}
3909 @FIXME{ hag -- you might want to incorporate some of the above into the
3910 MMwtSN node; not sure. i didn't know how to make it simpler...
3912 There are a few ways to get around this. (maybe xref Multiple Members
3913 with the Same Name.}
3915 @cindex Members, replacing with other members
3916 @cindex Replacing members with other members
3917 If you want to replace an archive member, use @option{--delete} to
3918 delete the member you want to remove from the archive, , and then use
3919 @option{--append} to add the member you want to be in the archive. Note
3920 that you can not change the order of the archive; the most recently
3921 added member will still appear last. In this sense, you cannot truly
3922 ``replace'' one member with another. (Replacing one member with another
3923 will not work on certain types of media, such as tapes; see @ref{delete}
3924 and @ref{Media}, for more information.)
3927 * appending files:: Appending Files to an Archive
3931 @node appending files
3932 @subsubsection Appending Files to an Archive
3934 @cindex Adding files to an Archive
3935 @cindex Appending files to an Archive
3936 @cindex Archives, Appending files to
3938 The simplest way to add a file to an already existing archive is the
3939 @option{--append} (@option{-r}) operation, which writes specified
3940 files into the archive whether or not they are already among the
3943 When you use @option{--append}, you @emph{must} specify file name
3944 arguments, as there is no default. If you specify a file that already
3945 exists in the archive, another copy of the file will be added to the
3946 end of the archive. As with other operations, the member names of the
3947 newly added files will be exactly the same as their names given on the
3948 command line. The @option{--verbose} (@option{-v}) option will print
3949 out the names of the files as they are written into the archive.
3951 @option{--append} cannot be performed on some tape drives, unfortunately,
3952 due to deficiencies in the formats those tape drives use. The archive
3953 must be a valid @command{tar} archive, or else the results of using this
3954 operation will be unpredictable. @xref{Media}.
3956 To demonstrate using @option{--append} to add a file to an archive,
3957 create a file called @file{rock} in the @file{practice} directory.
3958 Make sure you are in the @file{practice} directory. Then, run the
3959 following @command{tar} command to add @file{rock} to
3960 @file{collection.tar}:
3963 $ @kbd{tar --append --file=collection.tar rock}
3967 If you now use the @option{--list} (@option{-t}) operation, you will see that
3968 @file{rock} has been added to the archive:
3971 $ @kbd{tar --list --file=collection.tar}
3972 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
3973 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3974 -rw-r--r-- me user 20 1996-09-23 16:44 folk
3975 -rw-r--r-- me user 20 1996-09-23 16:44 rock
3979 @subsubsection Multiple Members with the Same Name
3981 You can use @option{--append} (@option{-r}) to add copies of files
3982 which have been updated since the archive was created. (However, we
3983 do not recommend doing this since there is another @command{tar}
3984 option called @option{--update}; @xref{update}, for more information.
3985 We describe this use of @option{--append} here for the sake of
3986 completeness.) When you extract the archive, the older version will
3987 be effectively lost. This works because files are extracted from an
3988 archive in the order in which they were archived. Thus, when the
3989 archive is extracted, a file archived later in time will replace a
3990 file of the same name which was archived earlier, even though the
3991 older version of the file will remain in the archive unless you delete
3992 all versions of the file.
3994 Supposing you change the file @file{blues} and then append the changed
3995 version to @file{collection.tar}. As you saw above, the original
3996 @file{blues} is in the archive @file{collection.tar}. If you change the
3997 file and append the new version of the file to the archive, there will
3998 be two copies in the archive. When you extract the archive, the older
3999 version of the file will be extracted first, and then replaced by the
4000 newer version when it is extracted.
4002 You can append the new, changed copy of the file @file{blues} to the
4003 archive in this way:
4006 $ @kbd{tar --append --verbose --file=collection.tar blues}
4011 Because you specified the @option{--verbose} option, @command{tar} has
4012 printed the name of the file being appended as it was acted on. Now
4013 list the contents of the archive:
4016 $ @kbd{tar --list --verbose --file=collection.tar}
4017 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
4018 -rw-r--r-- me user 21 1996-09-23 16:44 blues
4019 -rw-r--r-- me user 20 1996-09-23 16:44 folk
4020 -rw-r--r-- me user 20 1996-09-23 16:44 rock
4021 -rw-r--r-- me user 58 1996-10-24 18:30 blues
4025 The newest version of @file{blues} is now at the end of the archive
4026 (note the different creation dates and file sizes). If you extract
4027 the archive, the older version of the file @file{blues} will be
4028 replaced by the newer version. You can confirm this by extracting
4029 the archive and running @samp{ls} on the directory.
4031 If you wish to extract the first occurrence of the file @file{blues}
4032 from the archive, use @option{--occurrence} option, as shown in
4033 the following example:
4036 $ @kbd{tar --extract -vv --occurrence --file=collection.tar blues}
4037 -rw-r--r-- me user 21 1996-09-23 16:44 blues
4040 @xref{Writing}, for more information on @option{--extract} and
4041 @xref{Option Summary, --occurrence}, for the description of
4042 @option{--occurrence} option.
4045 @subsection Updating an Archive
4047 @cindex Updating an archive
4050 In the previous section, you learned how to use @option{--append} to
4051 add a file to an existing archive. A related operation is
4052 @option{--update} (@option{-u}). The @option{--update} operation
4053 updates a @command{tar} archive by comparing the date of the specified
4054 archive members against the date of the file with the same name. If
4055 the file has been modified more recently than the archive member, then
4056 the newer version of the file is added to the archive (as with
4059 Unfortunately, you cannot use @option{--update} with magnetic tape drives.
4060 The operation will fail.
4062 @FIXME{other examples of media on which --update will fail? need to ask
4063 charles and/or mib/thomas/dave shevett..}
4065 Both @option{--update} and @option{--append} work by adding to the end
4066 of the archive. When you extract a file from the archive, only the
4067 version stored last will wind up in the file system, unless you use
4068 the @option{--backup} option. @xref{multiple}, for a detailed discussion.
4075 @subsubsection How to Update an Archive Using @option{--update}
4077 You must use file name arguments with the @option{--update}
4078 (@option{-u}) operation. If you don't specify any files,
4079 @command{tar} won't act on any files and won't tell you that it didn't
4080 do anything (which may end up confusing you).
4082 @c note: the above parenthetical added because in fact, this
4083 @c behavior just confused the author. :-)
4085 To see the @option{--update} option at work, create a new file,
4086 @file{classical}, in your practice directory, and some extra text to the
4087 file @file{blues}, using any text editor. Then invoke @command{tar} with
4088 the @samp{update} operation and the @option{--verbose} (@option{-v})
4089 option specified, using the names of all the files in the practice
4090 directory as file name arguments:
4093 $ @kbd{tar --update -v -f collection.tar blues folk rock classical}
4100 Because we have specified verbose mode, @command{tar} prints out the names
4101 of the files it is working on, which in this case are the names of the
4102 files that needed to be updated. If you run @samp{tar --list} and look
4103 at the archive, you will see @file{blues} and @file{classical} at its
4104 end. There will be a total of two versions of the member @samp{blues};
4105 the one at the end will be newer and larger, since you added text before
4108 (The reason @command{tar} does not overwrite the older file when updating
4109 it is because writing to the middle of a section of tape is a difficult
4110 process. Tapes are not designed to go backward. @xref{Media}, for more
4111 information about tapes.
4113 @option{--update} (@option{-u}) is not suitable for performing backups for two
4114 reasons: it does not change directory content entries, and it
4115 lengthens the archive every time it is used. The @GNUTAR{}
4116 options intended specifically for backups are more
4117 efficient. If you need to run backups, please consult @ref{Backups}.
4120 @subsection Combining Archives with @option{--concatenate}
4122 @cindex Adding archives to an archive
4123 @cindex Concatenating Archives
4124 @opindex concatenate
4126 @c @cindex @option{-A} described
4127 Sometimes it may be convenient to add a second archive onto the end of
4128 an archive rather than adding individual files to the archive. To add
4129 one or more archives to the end of another archive, you should use the
4130 @option{--concatenate} (@option{--catenate}, @option{-A}) operation.
4132 To use @option{--concatenate}, give the first archive with
4133 @option{--file} option and name the rest of archives to be
4134 concatenated on the command line. The members, and their member
4135 names, will be copied verbatim from those archives to the first one.
4136 @footnote{This can cause multiple members to have the same name, for
4137 information on how this affects reading the archive, @ref{multiple}.}
4138 The new, concatenated archive will be called by the same name as the
4139 one given with the @option{--file} option. As usual, if you omit
4140 @option{--file}, @command{tar} will use the value of the environment
4141 variable @env{TAPE}, or, if this has not been set, the default archive name.
4143 @FIXME{There is no way to specify a new name...}
4145 To demonstrate how @option{--concatenate} works, create two small archives
4146 called @file{bluesrock.tar} and @file{folkjazz.tar}, using the relevant
4147 files from @file{practice}:
4150 $ @kbd{tar -cvf bluesrock.tar blues rock}
4153 $ @kbd{tar -cvf folkjazz.tar folk jazz}
4159 If you like, You can run @samp{tar --list} to make sure the archives
4160 contain what they are supposed to:
4163 $ @kbd{tar -tvf bluesrock.tar}
4164 -rw-r--r-- melissa user 105 1997-01-21 19:42 blues
4165 -rw-r--r-- melissa user 33 1997-01-20 15:34 rock
4166 $ @kbd{tar -tvf jazzfolk.tar}
4167 -rw-r--r-- melissa user 20 1996-09-23 16:44 folk
4168 -rw-r--r-- melissa user 65 1997-01-30 14:15 jazz
4171 We can concatenate these two archives with @command{tar}:
4175 $ @kbd{tar --concatenate --file=bluesrock.tar jazzfolk.tar}
4178 If you now list the contents of the @file{bluesrock.tar}, you will see
4179 that now it also contains the archive members of @file{jazzfolk.tar}:
4182 $ @kbd{tar --list --file=bluesrock.tar}
4189 When you use @option{--concatenate}, the source and target archives must
4190 already exist and must have been created using compatible format
4191 parameters. Notice, that @command{tar} does not check whether the
4192 archives it concatenates have compatible formats, it does not
4193 even check if the files are really tar archives.
4195 Like @option{--append} (@option{-r}), this operation cannot be performed on some
4196 tape drives, due to deficiencies in the formats those tape drives use.
4198 @cindex @code{concatenate} vs @command{cat}
4199 @cindex @command{cat} vs @code{concatenate}
4200 It may seem more intuitive to you to want or try to use @command{cat} to
4201 concatenate two archives instead of using the @option{--concatenate}
4202 operation; after all, @command{cat} is the utility for combining files.
4204 However, @command{tar} archives incorporate an end-of-file marker which
4205 must be removed if the concatenated archives are to be read properly as
4206 one archive. @option{--concatenate} removes the end-of-archive marker
4207 from the target archive before each new archive is appended. If you use
4208 @command{cat} to combine the archives, the result will not be a valid
4209 @command{tar} format archive. If you need to retrieve files from an
4210 archive that was added to using the @command{cat} utility, use the
4211 @option{--ignore-zeros} (@option{-i}) option. @xref{Ignore Zeros}, for further
4212 information on dealing with archives improperly combined using the
4213 @command{cat} shell utility.
4216 @subsection Removing Archive Members Using @option{--delete}
4218 @cindex Deleting files from an archive
4219 @cindex Removing files from an archive
4222 You can remove members from an archive by using the @option{--delete}
4223 option. Specify the name of the archive with @option{--file}
4224 (@option{-f}) and then specify the names of the members to be deleted;
4225 if you list no member names, nothing will be deleted. The
4226 @option{--verbose} option will cause @command{tar} to print the names
4227 of the members as they are deleted. As with @option{--extract}, you
4228 must give the exact member names when using @samp{tar --delete}.
4229 @option{--delete} will remove all versions of the named file from the
4230 archive. The @option{--delete} operation can run very slowly.
4232 Unlike other operations, @option{--delete} has no short form.
4234 @cindex Tapes, using @option{--delete} and
4235 @cindex Deleting from tape archives
4236 This operation will rewrite the archive. You can only use
4237 @option{--delete} on an archive if the archive device allows you to
4238 write to any point on the media, such as a disk; because of this, it
4239 does not work on magnetic tapes. Do not try to delete an archive member
4240 from a magnetic tape; the action will not succeed, and you will be
4241 likely to scramble the archive and damage your tape. There is no safe
4242 way (except by completely re-writing the archive) to delete files from
4243 most kinds of magnetic tape. @xref{Media}.
4245 To delete all versions of the file @file{blues} from the archive
4246 @file{collection.tar} in the @file{practice} directory, make sure you
4247 are in that directory, and then,
4250 $ @kbd{tar --list --file=collection.tar}
4255 $ @kbd{tar --delete --file=collection.tar blues}
4256 $ @kbd{tar --list --file=collection.tar}
4263 @FIXME{Check if the above listing is actually produced after running
4264 all the examples on collection.tar.}
4266 The @option{--delete} option has been reported to work properly when
4267 @command{tar} acts as a filter from @code{stdin} to @code{stdout}.
4270 @subsection Comparing Archive Members with the File System
4271 @cindex Verifying the currency of an archive
4275 The @option{--compare} (@option{-d}), or @option{--diff} operation compares
4276 specified archive members against files with the same names, and then
4277 reports differences in file size, mode, owner, modification date and
4278 contents. You should @emph{only} specify archive member names, not file
4279 names. If you do not name any members, then @command{tar} will compare the
4280 entire archive. If a file is represented in the archive but does not
4281 exist in the file system, @command{tar} reports a difference.
4283 You have to specify the record size of the archive when modifying an
4284 archive with a non-default record size.
4286 @command{tar} ignores files in the file system that do not have
4287 corresponding members in the archive.
4289 The following example compares the archive members @file{rock},
4290 @file{blues} and @file{funk} in the archive @file{bluesrock.tar} with
4291 files of the same name in the file system. (Note that there is no file,
4292 @file{funk}; @command{tar} will report an error message.)
4295 $ @kbd{tar --compare --file=bluesrock.tar rock blues funk}
4298 tar: funk not found in archive
4301 The spirit behind the @option{--compare} (@option{--diff},
4302 @option{-d}) option is to check whether the archive represents the
4303 current state of files on disk, more than validating the integrity of
4304 the archive media. For this later goal, @xref{verify}.
4306 @node create options
4307 @section Options Used by @option{--create}
4309 @opindex create, additional options
4310 The previous chapter described the basics of how to use
4311 @option{--create} (@option{-c}) to create an archive from a set of files.
4312 @xref{create}. This section described advanced options to be used with
4316 * Ignore Failed Read::
4319 @node Ignore Failed Read
4320 @subsection Ignore Fail Read
4323 @item --ignore-failed-read
4324 Do not exit with nonzero on unreadable files or directories.
4327 @node extract options
4328 @section Options Used by @option{--extract}
4331 @opindex extract, additional options
4332 The previous chapter showed how to use @option{--extract} to extract
4333 an archive into the file system. Various options cause @command{tar} to
4334 extract more information than just file contents, such as the owner,
4335 the permissions, the modification date, and so forth. This section
4336 presents options to be used with @option{--extract} when certain special
4337 considerations arise. You may review the information presented in
4338 @ref{extract} for more basic information about the
4339 @option{--extract} operation.
4342 * Reading:: Options to Help Read Archives
4343 * Writing:: Changing How @command{tar} Writes Files
4344 * Scarce:: Coping with Scarce Resources
4348 @subsection Options to Help Read Archives
4349 @cindex Options when reading archives
4352 @cindex Reading incomplete records
4353 @cindex Records, incomplete
4354 @opindex read-full-records
4355 Normally, @command{tar} will request data in full record increments from
4356 an archive storage device. If the device cannot return a full record,
4357 @command{tar} will report an error. However, some devices do not always
4358 return full records, or do not require the last record of an archive to
4359 be padded out to the next record boundary. To keep reading until you
4360 obtain a full record, or to accept an incomplete record if it contains
4361 an end-of-archive marker, specify the @option{--read-full-records} (@option{-B}) option
4362 in conjunction with the @option{--extract} or @option{--list} operations.
4365 The @option{--read-full-records} (@option{-B}) option is turned on by default when
4366 @command{tar} reads an archive from standard input, or from a remote
4367 machine. This is because on BSD Unix systems, attempting to read a
4368 pipe returns however much happens to be in the pipe, even if it is
4369 less than was requested. If this option were not enabled, @command{tar}
4370 would fail as soon as it read an incomplete record from the pipe.
4372 If you're not sure of the blocking factor of an archive, you can
4373 read the archive by specifying @option{--read-full-records} (@option{-B}) and
4374 @option{--blocking-factor=@var{512-size}} (@option{-b
4375 @var{512-size}}), using a blocking factor larger than what the archive
4376 uses. This lets you avoid having to determine the blocking factor
4377 of an archive. @xref{Blocking Factor}.
4380 * read full records::
4384 @node read full records
4385 @unnumberedsubsubsec Reading Full Records
4387 @FIXME{need sentence or so of intro here}
4390 @opindex read-full-records
4391 @item --read-full-records
4393 Use in conjunction with @option{--extract} (@option{--get},
4394 @option{-x}) to read an archive which contains incomplete records, or
4395 one which has a blocking factor less than the one specified.
4399 @unnumberedsubsubsec Ignoring Blocks of Zeros
4401 @cindex End-of-archive blocks, ignoring
4402 @cindex Ignoring end-of-archive blocks
4403 @opindex ignore-zeros
4404 Normally, @command{tar} stops reading when it encounters a block of zeros
4405 between file entries (which usually indicates the end of the archive).
4406 @option{--ignore-zeros} (@option{-i}) allows @command{tar} to
4407 completely read an archive which contains a block of zeros before the
4408 end (i.e., a damaged archive, or one that was created by concatenating
4409 several archives together).
4411 The @option{--ignore-zeros} (@option{-i}) option is turned off by default because many
4412 versions of @command{tar} write garbage after the end-of-archive entry,
4413 since that part of the media is never supposed to be read. @GNUTAR{}
4414 does not write after the end of an archive, but seeks to
4415 maintain compatiblity among archiving utilities.
4418 @item --ignore-zeros
4420 To ignore blocks of zeros (i.e., end-of-archive entries) which may be
4421 encountered while reading an archive. Use in conjunction with
4422 @option{--extract} or @option{--list}.
4426 @subsection Changing How @command{tar} Writes Files
4429 @FIXME{Introductory paragraph}
4432 * Dealing with Old Files::
4433 * Overwrite Old Files::
4435 * Keep Newer Files::
4437 * Recursive Unlink::
4438 * Data Modification Times::
4439 * Setting Access Permissions::
4440 * Directory Modification Times and Permissions::
4441 * Writing to Standard Output::
4442 * Writing to an External Program::
4446 @node Dealing with Old Files
4447 @unnumberedsubsubsec Options Controlling the Overwriting of Existing Files
4449 @opindex overwrite-dir, introduced
4450 When extracting files, if @command{tar} discovers that the extracted
4451 file already exists, it normally replaces the file by removing it before
4452 extracting it, to prevent confusion in the presence of hard or symbolic
4453 links. (If the existing file is a symbolic link, it is removed, not
4454 followed.) However, if a directory cannot be removed because it is
4455 nonempty, @command{tar} normally overwrites its metadata (ownership,
4456 permission, etc.). The @option{--overwrite-dir} option enables this
4457 default behavior. To be more cautious and preserve the metadata of
4458 such a directory, use the @option{--no-overwrite-dir} option.
4460 @cindex Overwriting old files, prevention
4461 @opindex keep-old-files, introduced
4462 To be even more cautious and prevent existing files from being replaced, use
4463 the @option{--keep-old-files} (@option{-k}) option. It causes @command{tar} to refuse
4464 to replace or update a file that already exists, i.e., a file with the
4465 same name as an archive member prevents extraction of that archive
4466 member. Instead, it reports an error.
4468 @opindex overwrite, introduced
4469 To be more aggressive about altering existing files, use the
4470 @option{--overwrite} option. It causes @command{tar} to overwrite
4471 existing files and to follow existing symbolic links when extracting.
4473 @cindex Protecting old files
4474 Some people argue that @GNUTAR{} should not hesitate
4475 to overwrite files with other files when extracting. When extracting
4476 a @command{tar} archive, they expect to see a faithful copy of the
4477 state of the file system when the archive was created. It is debatable
4478 that this would always be a proper behavior. For example, suppose one
4479 has an archive in which @file{usr/local} is a link to
4480 @file{usr/local2}. Since then, maybe the site removed the link and
4481 renamed the whole hierarchy from @file{/usr/local2} to
4482 @file{/usr/local}. Such things happen all the time. I guess it would
4483 not be welcome at all that @GNUTAR{} removes the
4484 whole hierarchy just to make room for the link to be reinstated
4485 (unless it @emph{also} simultaneously restores the full
4486 @file{/usr/local2}, of course!) @GNUTAR{} is indeed
4487 able to remove a whole hierarchy to reestablish a symbolic link, for
4488 example, but @emph{only if} @option{--recursive-unlink} is specified
4489 to allow this behavior. In any case, single files are silently
4492 @opindex unlink-first, introduced
4493 Finally, the @option{--unlink-first} (@option{-U}) option can improve performance in
4494 some cases by causing @command{tar} to remove files unconditionally
4495 before extracting them.
4497 @node Overwrite Old Files
4498 @unnumberedsubsubsec Overwrite Old Files
4503 Overwrite existing files and directory metadata when extracting files
4506 This causes @command{tar} to write extracted files into the file system without
4507 regard to the files already on the system; i.e., files with the same
4508 names as archive members are overwritten when the archive is extracted.
4509 It also causes @command{tar} to extract the ownership, permissions,
4510 and time stamps onto any preexisting files or directories.
4511 If the name of a corresponding file name is a symbolic link, the file
4512 pointed to by the symbolic link will be overwritten instead of the
4513 symbolic link itself (if this is possible). Moreover, special devices,
4514 empty directories and even symbolic links are automatically removed if
4515 they are in the way of extraction.
4517 Be careful when using the @option{--overwrite} option, particularly when
4518 combined with the @option{--absolute-names} (@option{-P}) option, as this combination
4519 can change the contents, ownership or permissions of any file on your
4520 system. Also, many systems do not take kindly to overwriting files that
4521 are currently being executed.
4523 @opindex overwrite-dir
4524 @item --overwrite-dir
4525 Overwrite the metadata of directories when extracting files from an
4526 archive, but remove other files before extracting.
4529 @node Keep Old Files
4530 @unnumberedsubsubsec Keep Old Files
4533 @opindex keep-old-files
4534 @item --keep-old-files
4536 Do not replace existing files from archive. The
4537 @option{--keep-old-files} (@option{-k}) option prevents @command{tar}
4538 from replacing existing files with files with the same name from the
4539 archive. The @option{--keep-old-files} option is meaningless with
4540 @option{--list} (@option{-t}). Prevents @command{tar} from replacing
4541 files in the file system during extraction.
4544 @node Keep Newer Files
4545 @unnumberedsubsubsec Keep Newer Files
4548 @opindex keep-newer-files
4549 @item --keep-newer-files
4550 Do not replace existing files that are newer than their archive
4551 copies. This option is meaningless with @option{--list} (@option{-t}).
4555 @unnumberedsubsubsec Unlink First
4558 @opindex unlink-first
4559 @item --unlink-first
4561 Remove files before extracting over them.
4562 This can make @command{tar} run a bit faster if you know in advance
4563 that the extracted files all need to be removed. Normally this option
4564 slows @command{tar} down slightly, so it is disabled by default.
4567 @node Recursive Unlink
4568 @unnumberedsubsubsec Recursive Unlink
4571 @opindex recursive-unlink
4572 @item --recursive-unlink
4573 When this option is specified, try removing files and directory hierarchies
4574 before extracting over them. @emph{This is a dangerous option!}
4577 If you specify the @option{--recursive-unlink} option,
4578 @command{tar} removes @emph{anything} that keeps you from extracting a file
4579 as far as current permissions will allow it. This could include removal
4580 of the contents of a full directory hierarchy.
4582 @node Data Modification Times
4583 @unnumberedsubsubsec Setting Data Modification Times
4585 @cindex Data modification times of extracted files
4586 @cindex Modification times of extracted files
4587 Normally, @command{tar} sets the data modification times of extracted
4588 files to the corresponding times recorded for the files in the archive, but
4589 limits the permissions of extracted files by the current @code{umask}
4592 To set the data modification times of extracted files to the time when
4593 the files were extracted, use the @option{--touch} (@option{-m}) option in
4594 conjunction with @option{--extract} (@option{--get}, @option{-x}).
4600 Sets the data modification time of extracted archive members to the time
4601 they were extracted, not the time recorded for them in the archive.
4602 Use in conjunction with @option{--extract} (@option{--get}, @option{-x}).
4605 @node Setting Access Permissions
4606 @unnumberedsubsubsec Setting Access Permissions
4608 @cindex Permissions of extracted files
4609 @cindex Modes of extracted files
4610 To set the modes (access permissions) of extracted files to those
4611 recorded for those files in the archive, use @option{--same-permissions}
4612 in conjunction with the @option{--extract} (@option{--get},
4613 @option{-x}) operation.
4616 @opindex preserve-permissions
4617 @opindex same-permissions
4618 @item --preserve-permissions
4619 @itemx --same-permissions
4620 @c @itemx --ignore-umask
4622 Set modes of extracted archive members to those recorded in the
4623 archive, instead of current umask settings. Use in conjunction with
4624 @option{--extract} (@option{--get}, @option{-x}).
4627 @node Directory Modification Times and Permissions
4628 @unnumberedsubsubsec Directory Modification Times and Permissions
4630 After sucessfully extracting a file member, @GNUTAR{} normally
4631 restores its permissions and modification times, as described in the
4632 previous sections. This cannot be done for directories, because
4633 after extracting a directory @command{tar} will almost certainly
4634 extract files into that directory and this will cause the directory
4635 modification time to be updated. Moreover, restoring that directory
4636 permissions may not permit file creation within it. Thus, restoring
4637 directory permissions and modification times must be delayed at least
4638 until all files have been extracted into that directory. @GNUTAR{}
4639 restores directories using the following approach.
4641 The extracted directories are created with the mode specified in the
4642 archive, as modified by the umask of the user, which gives sufficient
4643 permissions to allow file creation. The meta-information about the
4644 directory is recorded in the temporary list of directories. When
4645 preparing to extract next archive member, @GNUTAR{} checks if the
4646 directory prefix of this file contains the remembered directory. If
4647 it does not, the program assumes that all files have been extracted
4648 into that directory, restores its modification time and permissions
4649 and removes its entry from the internal list. This approach allows
4650 to correctly restore directory meta-information in the majority of
4651 cases, while keeping memory requirements sufficiently small. It is
4652 based on the fact, that most @command{tar} archives use the predefined
4653 order of members: first the directory, then all the files and
4654 subdirectories in that directory.
4656 However, this is not always true. The most important exception are
4657 incremental archives (@pxref{Incremental Dumps}). The member order in
4658 an incremental archive is reversed: first all directory members are
4659 stored, followed by other (non-directory) members. So, when extracting
4660 from incremental archives, @GNUTAR{} alters the above procedure. It
4661 remebers all restored directories, and restores their meta-data
4662 only after the entire archive has been processed. Notice, that you do
4663 not need to specity any special options for that, as @GNUTAR{}
4664 automatically detects archives in incremental format.
4666 There may be cases, when such processing is required for normal archives
4667 too. Consider the following example:
4671 $ @kbd{tar --no-recursion -cvf archive \
4672 foo foo/file1 bar bar/file foo/file2}
4681 During the normal operation, after encountering @file{bar}
4682 @GNUTAR{} will assume that all files from the directory @file{foo}
4683 were already extracted and will therefore restore its timestamp and
4684 permission bits. However, after extracting @file{foo/file2} the
4685 directory timestamp will be offset again.
4687 To correctly restore directory meta-information in such cases, use
4688 @option{delay-directory-restore} command line option:
4691 @opindex delay-directory-restore
4692 @item --delay-directory-restore
4693 Delays restoring of the modification times and permissions of extracted
4694 directories until the end of extraction. This way, correct
4695 meta-information is restored even if the archive has unusual member
4698 @opindex no-delay-directory-restore
4699 @item --no-delay-directory-restore
4700 Cancel the effect of the previous @option{--delay-directory-restore}.
4701 Use this option if you have used @option{--delay-directory-restore} in
4702 @env{TAR_OPTIONS} variable (@pxref{TAR_OPTIONS}) and wish to
4703 temporarily disable it.
4706 @node Writing to Standard Output
4707 @unnumberedsubsubsec Writing to Standard Output
4709 @cindex Writing extracted files to standard output
4710 @cindex Standard output, writing extracted files to
4711 To write the extracted files to the standard output, instead of
4712 creating the files on the file system, use @option{--to-stdout} (@option{-O}) in
4713 conjunction with @option{--extract} (@option{--get}, @option{-x}). This option is useful if you are
4714 extracting files to send them through a pipe, and do not need to
4715 preserve them in the file system. If you extract multiple members,
4716 they appear on standard output concatenated, in the order they are
4717 found in the archive.
4723 Writes files to the standard output. Use only in conjunction with
4724 @option{--extract} (@option{--get}, @option{-x}). When this option is
4725 used, instead of creating the files specified, @command{tar} writes
4726 the contents of the files extracted to its standard output. This may
4727 be useful if you are only extracting the files in order to send them
4728 through a pipe. This option is meaningless with @option{--list}
4732 This can be useful, for example, if you have a tar archive containing
4733 a big file and don't want to store the file on disk before processing
4734 it. You can use a command like this:
4737 tar -xOzf foo.tgz bigfile | process
4740 or even like this if you want to process the concatenation of the files:
4743 tar -xOzf foo.tgz bigfile1 bigfile2 | process
4746 Hovewer, @option{--to-command} may be more convenient for use with
4747 multiple files. See the next section.
4749 @node Writing to an External Program
4750 @unnumberedsubsubsec Writing to an External Program
4752 You can instruct @command{tar} to send the contents of each extracted
4753 file to the standard input of an external program:
4757 @item --to-command=@var{command}
4758 Extract files and pipe their contents to the standard input of
4759 @var{command}. When this option is used, instead of creating the
4760 files specified, @command{tar} invokes @var{command} and pipes the
4761 contents of the files to its standard output. @var{Command} may
4762 contain command line arguments. The program is executed via
4763 @code{sh -c}. Notice, that @var{command} is executed once for each regular file
4764 extracted. Non-regular files (directories, etc.) are ignored when this
4768 The command can obtain the information about the file it processes
4769 from the following environment variables:
4772 @vrindex TAR_FILETYPE, to-command environment
4774 Type of the file. It is a single letter with the following meaning:
4776 @multitable @columnfractions 0.10 0.90
4777 @item f @tab Regular file
4778 @item d @tab Directory
4779 @item l @tab Symbolic link
4780 @item h @tab Hard link
4781 @item b @tab Block device
4782 @item c @tab Character device
4785 Currently only regular files are supported.
4787 @vrindex TAR_MODE, to-command environment
4789 File mode, an octal number.
4791 @vrindex TAR_FILENAME, to-command environment
4793 The name of the file.
4795 @vrindex TAR_REALNAME, to-command environment
4797 Name of the file as stored in the archive.
4799 @vrindex TAR_UNAME, to-command environment
4801 Name of the file owner.
4803 @vrindex TAR_GNAME, to-command environment
4805 Name of the file owner group.
4807 @vrindex TAR_ATIME, to-command environment
4809 Time of last access. It is a decimal number, representing seconds
4810 since the epoch. If the archive provides times with nanosecond
4811 precision, the nanoseconds are appended to the timestamp after a
4814 @vrindex TAR_MTIME, to-command environment
4816 Time of last modification.
4818 @vrindex TAR_CTIME, to-command environment
4820 Time of last status change.
4822 @vrindex TAR_SIZE, to-command environment
4826 @vrindex TAR_UID, to-command environment
4828 UID of the file owner.
4830 @vrindex TAR_GID, to-command environment
4832 GID of the file owner.
4835 In addition to these variables, @env{TAR_VERSION} contains the
4836 @GNUTAR{} version number.
4838 If @var{command} exits with a non-0 status, @command{tar} will print
4839 an error message similar to the following:
4842 tar: 2345: Child returned status 1
4845 Here, @samp{2345} is the PID of the finished process.
4847 If this behavior is not wanted, use @option{--ignore-command-error}:
4850 @opindex ignore-command-error
4851 @item --ignore-command-error
4852 Ignore exit codes of subprocesses. Notice that if the program
4853 exits on signal or otherwise terminates abnormally, the error message
4854 will be printed even if this option is used.
4856 @opindex no-ignore-command-error
4857 @item --no-ignore-command-error
4858 Cancel the effect of any previous @option{--ignore-command-error}
4859 option. This option is useful if you have set
4860 @option{--ignore-command-error} in @env{TAR_OPTIONS}
4861 (@pxref{TAR_OPTIONS}) and wish to temporarily cancel it.
4865 @unnumberedsubsubsec Removing Files
4867 @FIXME{The section is too terse. Something more to add? An example,
4871 @opindex remove-files
4872 @item --remove-files
4873 Remove files after adding them to the archive.
4877 @subsection Coping with Scarce Resources
4880 @cindex Small memory
4881 @cindex Running out of space
4889 @unnumberedsubsubsec Starting File
4892 @opindex starting-file
4893 @item --starting-file=@var{name}
4894 @itemx -K @var{name}
4895 Starts an operation in the middle of an archive. Use in conjunction
4896 with @option{--extract} (@option{--get}, @option{-x}) or @option{--list} (@option{-t}).
4899 @cindex Middle of the archive, starting in the
4900 If a previous attempt to extract files failed due to lack of disk
4901 space, you can use @option{--starting-file=@var{name}} (@option{-K
4902 @var{name}}) to start extracting only after member @var{name} of the
4903 archive. This assumes, of course, that there is now free space, or
4904 that you are now extracting into a different file system. (You could
4905 also choose to suspend @command{tar}, remove unnecessary files from
4906 the file system, and then restart the same @command{tar} operation.
4907 In this case, @option{--starting-file} is not necessary.
4908 @xref{Incremental Dumps}, @xref{interactive}, and @ref{exclude}.)
4911 @unnumberedsubsubsec Same Order
4914 @cindex Large lists of file names on small machines
4916 @opindex preserve-order
4918 @itemx --preserve-order
4920 To process large lists of file names on machines with small amounts of
4921 memory. Use in conjunction with @option{--compare} (@option{--diff},
4922 @option{-d}), @option{--list} (@option{-t}) or @option{--extract}
4923 (@option{--get}, @option{-x}).
4926 The @option{--same-order} (@option{--preserve-order}, @option{-s}) option tells @command{tar} that the list of file
4927 names to be listed or extracted is sorted in the same order as the
4928 files in the archive. This allows a large list of names to be used,
4929 even on a small machine that would not otherwise be able to hold all
4930 the names in memory at the same time. Such a sorted list can easily be
4931 created by running @samp{tar -t} on the archive and editing its output.
4933 This option is probably never needed on modern computer systems.
4936 @section Backup options
4938 @cindex backup options
4940 @GNUTAR{} offers options for making backups of files
4941 before writing new versions. These options control the details of
4942 these backups. They may apply to the archive itself before it is
4943 created or rewritten, as well as individual extracted members. Other
4944 @acronym{GNU} programs (@command{cp}, @command{install}, @command{ln},
4945 and @command{mv}, for example) offer similar options.
4947 Backup options may prove unexpectedly useful when extracting archives
4948 containing many members having identical name, or when extracting archives
4949 on systems having file name limitations, making different members appear
4950 has having similar names through the side-effect of name truncation.
4951 (This is true only if we have a good scheme for truncated backup names,
4952 which I'm not sure at all: I suspect work is needed in this area.)
4953 When any existing file is backed up before being overwritten by extraction,
4954 then clashing files are automatically be renamed to be unique, and the
4955 true name is kept for only the last file of a series of clashing files.
4956 By using verbose mode, users may track exactly what happens.
4958 At the detail level, some decisions are still experimental, and may
4959 change in the future, we are waiting comments from our users. So, please
4960 do not learn to depend blindly on the details of the backup features.
4961 For example, currently, directories themselves are never renamed through
4962 using these options, so, extracting a file over a directory still has
4963 good chances to fail. Also, backup options apply to created archives,
4964 not only to extracted members. For created archives, backups will not
4965 be attempted when the archive is a block or character device, or when it
4966 refers to a remote file.
4968 For the sake of simplicity and efficiency, backups are made by renaming old
4969 files prior to creation or extraction, and not by copying. The original
4970 name is restored if the file creation fails. If a failure occurs after a
4971 partial extraction of a file, both the backup and the partially extracted
4975 @item --backup[=@var{method}]
4977 @vindex VERSION_CONTROL
4979 Back up files that are about to be overwritten or removed.
4980 Without this option, the original versions are destroyed.
4982 Use @var{method} to determine the type of backups made.
4983 If @var{method} is not specified, use the value of the @env{VERSION_CONTROL}
4984 environment variable. And if @env{VERSION_CONTROL} is not set,
4985 use the @samp{existing} method.
4987 @vindex version-control @r{Emacs variable}
4988 This option corresponds to the Emacs variable @samp{version-control};
4989 the same values for @var{method} are accepted as in Emacs. This option
4990 also allows more descriptive names. The valid @var{method}s are:
4995 @cindex numbered @r{backup method}
4996 Always make numbered backups.
5000 @cindex existing @r{backup method}
5001 Make numbered backups of files that already have them, simple backups
5006 @cindex simple @r{backup method}
5007 Always make simple backups.
5011 @item --suffix=@var{suffix}
5013 @cindex backup suffix
5014 @vindex SIMPLE_BACKUP_SUFFIX
5015 Append @var{suffix} to each backup file made with @option{--backup}. If this
5016 option is not specified, the value of the @env{SIMPLE_BACKUP_SUFFIX}
5017 environment variable is used. And if @env{SIMPLE_BACKUP_SUFFIX} is not
5018 set, the default is @samp{~}, just as in Emacs.
5023 @section Notable @command{tar} Usages
5026 @FIXME{Using Unix file linking capability to recreate directory
5027 structures---linking files into one subdirectory and then
5028 @command{tar}ring that directory.}
5030 @FIXME{Nice hairy example using absolute-names, newer, etc.}
5033 You can easily use archive files to transport a group of files from
5034 one system to another: put all relevant files into an archive on one
5035 computer system, transfer the archive to another system, and extract
5036 the contents there. The basic transfer medium might be magnetic tape,
5037 Internet FTP, or even electronic mail (though you must encode the
5038 archive with @command{uuencode} in order to transport it properly by
5039 mail). Both machines do not have to use the same operating system, as
5040 long as they both support the @command{tar} program.
5042 For example, here is how you might copy a directory's contents from
5043 one disk to another, while preserving the dates, modes, owners and
5044 link-structure of all the files therein. In this case, the transfer
5045 medium is a @dfn{pipe}, which is one a Unix redirection mechanism:
5048 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xf -)}
5052 You can avoid subshells by using @option{-C} option:
5055 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xf -}
5059 The command also works using short option forms:
5062 $ @kbd{(cd sourcedir; tar --create --file=- . ) \
5063 | (cd targetdir; tar --extract --file=-)}
5065 $ @kbd{tar --directory sourcedir --create --file=- . ) \
5066 | tar --directory targetdir --extract --file=-}
5070 This is one of the easiest methods to transfer a @command{tar} archive.
5073 @section Looking Ahead: The Rest of this Manual
5075 You have now seen how to use all eight of the operations available to
5076 @command{tar}, and a number of the possible options. The next chapter
5077 explains how to choose and change file and archive names, how to use
5078 files to store names of other files which you can then call as
5079 arguments to @command{tar} (this can help you save time if you expect to
5080 archive the same list of files a number of times), and so forth.
5081 @FIXME{in case it's not obvious, i'm making this up in some sense
5082 based on my limited memory of what the next chapter *really* does. i
5083 just wanted to flesh out this final section a little bit so i'd
5084 remember to stick it in here. :-)}
5086 If there are too many files to conveniently list on the command line,
5087 you can list the names in a file, and @command{tar} will read that file.
5090 There are various ways of causing @command{tar} to skip over some files,
5091 and not archive them. @xref{Choosing}.
5094 @chapter Performing Backups and Restoring Files
5097 @GNUTAR{} is distributed along with the scripts
5098 which the Free Software Foundation uses for performing backups. There
5099 is no corresponding scripts available yet for doing restoration of
5100 files. Even if there is a good chance those scripts may be satisfying
5101 to you, they are not the only scripts or methods available for doing
5102 backups and restore. You may well create your own, or use more
5103 sophisticated packages dedicated to that purpose.
5105 Some users are enthusiastic about @code{Amanda} (The Advanced Maryland
5106 Automatic Network Disk Archiver), a backup system developed by James
5107 da Silva @file{jds@@cs.umd.edu} and available on many Unix systems.
5108 This is free software, and it is available at these places:
5111 http://www.cs.umd.edu/projects/amanda/amanda.html
5112 ftp://ftp.cs.umd.edu/pub/amanda
5117 Here is a possible plan for a future documentation about the backuping
5118 scripts which are provided within the @GNUTAR{}
5124 @item what are dumps
5125 @item different levels of dumps
5127 @item full dump = dump everything
5128 @item level 1, level 2 dumps etc
5129 A level @var{n} dump dumps everything changed since the last level
5132 @item how to use scripts for dumps (ie, the concept)
5134 @item scripts to run after editing backup specs (details)
5136 @item Backup Specs, what is it.
5138 @item how to customize
5139 @item actual text of script [/sp/dump/backup-specs]
5143 @item rsh doesn't work
5144 @item rtape isn't installed
5147 @item the @option{--incremental} option of tar
5150 @item write protection
5151 @item types of media, different sizes and types, useful for different things
5152 @item files and tape marks
5153 one tape mark between files, two at end.
5154 @item positioning the tape
5155 MT writes two at end of write,
5156 backspaces over one when writing again.
5162 This chapter documents both the provided shell scripts and @command{tar}
5163 options which are more specific to usage as a backup tool.
5165 To @dfn{back up} a file system means to create archives that contain
5166 all the files in that file system. Those archives can then be used to
5167 restore any or all of those files (for instance if a disk crashes or a
5168 file is accidentally deleted). File system @dfn{backups} are also
5172 * Full Dumps:: Using @command{tar} to Perform Full Dumps
5173 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
5174 * Backup Levels:: Levels of Backups
5175 * Backup Parameters:: Setting Parameters for Backups and Restoration
5176 * Scripted Backups:: Using the Backup Scripts
5177 * Scripted Restoration:: Using the Restore Script
5181 @section Using @command{tar} to Perform Full Dumps
5187 @cindex corrupted archives
5188 Full dumps should only be made when no other people or programs
5189 are modifying files in the file system. If files are modified while
5190 @command{tar} is making the backup, they may not be stored properly in
5191 the archive, in which case you won't be able to restore them if you
5192 have to. (Files not being modified are written with no trouble, and do
5193 not corrupt the entire archive.)
5195 You will want to use the @option{--label=@var{archive-label}}
5196 (@option{-V @var{archive-label}}) option to give the archive a
5197 volume label, so you can tell what this archive is even if the label
5198 falls off the tape, or anything like that.
5200 Unless the file system you are dumping is guaranteed to fit on
5201 one volume, you will need to use the @option{--multi-volume} (@option{-M}) option.
5202 Make sure you have enough tapes on hand to complete the backup.
5204 If you want to dump each file system separately you will need to use
5205 the @option{--one-file-system} option to prevent
5206 @command{tar} from crossing file system boundaries when storing
5209 The @option{--incremental} (@option{-G}) (@pxref{Incremental Dumps})
5210 option is not needed, since this is a complete copy of everything in
5211 the file system, and a full restore from this backup would only be
5212 done onto a completely
5215 Unless you are in a hurry, and trust the @command{tar} program (and your
5216 tapes), it is a good idea to use the @option{--verify} (@option{-W})
5217 option, to make sure your files really made it onto the dump properly.
5218 This will also detect cases where the file was modified while (or just
5219 after) it was being archived. Not all media (notably cartridge tapes)
5220 are capable of being verified, unfortunately.
5222 @node Incremental Dumps
5223 @section Using @command{tar} to Perform Incremental Dumps
5225 @dfn{Incremental backup} is a special form of @GNUTAR{} archive that
5226 stores additional metadata so that exact state of the file system
5227 can be restored when extracting the archive.
5229 @GNUTAR{} currently offers two options for handling incremental
5230 backups: @option{--listed-incremental=@var{snapshot-file}} (@option{-g
5231 @var{snapshot-file}}) and @option{--incremental} (@option{-G}).
5233 @opindex listed-incremental
5234 The option @option{--listed-incremental} instructs tar to operate on
5235 an incremental archive with additional metadata stored in a standalone
5236 file, called a @dfn{snapshot file}. The purpose of this file is to help
5237 determine which files have been changed, added or deleted since the
5238 last backup, so that the next incremental backup will contain only
5239 modified files. The name of the snapshot file is given as an argument
5243 @item --listed-incremental=@var{file}
5244 @itemx -g @var{file}
5245 Handle incremental backups with snapshot data in @var{file}.
5248 To create an incremental backup, you would use
5249 @option{--listed-incremental} together with @option{--create}
5250 (@pxref{create}). For example:
5253 $ @kbd{tar --create \
5254 --file=archive.1.tar \
5255 --listed-incremental=/var/log/usr.snar \
5259 This will create in @file{archive.1.tar} an incremental backup of
5260 the @file{/usr} file system, storing additional metadata in the file
5261 @file{/var/log/usr.snar}. If this file does not exist, it will be
5262 created. The created archive will then be a @dfn{level 0 backup};
5263 please see the next section for more on backup levels.
5265 Otherwise, if the file @file{/var/log/usr.snar} exists, it
5266 determines which files are modified. In this case only these files will be
5267 stored in the archive. Suppose, for example, that after running the
5268 above command, you delete file @file{/usr/doc/old} and create
5269 directory @file{/usr/local/db} with the following contents:
5272 $ @kbd{ls /usr/local/db}
5277 Some time later you create another incremental backup. You will
5281 $ @kbd{tar --create \
5282 --file=archive.2.tar \
5283 --listed-incremental=/var/log/usr.snar \
5285 tar: usr/local/db: Directory is new
5292 The created archive @file{archive.2.tar} will contain only these
5293 three members. This archive is called a @dfn{level 1 backup}. Notice
5294 that @file{/var/log/usr.snar} will be updated with the new data, so if
5295 you plan to create more @samp{level 1} backups, it is necessary to
5296 create a working copy of the snapshot file before running
5297 @command{tar}. The above example will then be modified as follows:
5300 $ @kbd{cp /var/log/usr.snar /var/log/usr.snar-1}
5301 $ @kbd{tar --create \
5302 --file=archive.2.tar \
5303 --listed-incremental=/var/log/usr.snar-1 \
5307 Incremental dumps depend crucially on time stamps, so the results are
5308 unreliable if you modify a file's time stamps during dumping (e.g.,
5309 with the @option{--atime-preserve=replace} option), or if you set the clock
5312 Metadata stored in snapshot files include device numbers, which,
5313 obviously is supposed to be a non-volatile value. However, it turns
5314 out that NFS devices have undependable values when an automounter
5315 gets in the picture. This can lead to a great deal of spurious
5316 redumping in incremental dumps, so it is somewhat useless to compare
5317 two NFS devices numbers over time. The solution implemented currently
5318 is to considers all NFS devices as being equal when it comes to
5319 comparing directories; this is fairly gross, but there does not seem
5320 to be a better way to go.
5322 Note that incremental archives use @command{tar} extensions and may
5323 not be readable by non-@acronym{GNU} versions of the @command{tar} program.
5325 @opindex listed-incremental, using with @option{--extract}
5326 @opindex extract, using with @option{--listed-incremental}
5327 To extract from the incremental dumps, use
5328 @option{--listed-incremental} together with @option{--extract}
5329 option (@pxref{extracting files}). In this case, @command{tar} does
5330 not need to access snapshot file, since all the data necessary for
5331 extraction are stored in the archive itself. So, when extracting, you
5332 can give whatever argument to @option{--listed-incremental}, the usual
5333 practice is to use @option{--listed-incremental=/dev/null}.
5334 Alternatively, you can use @option{--incremental}, which needs no
5335 arguments. In general, @option{--incremental} (@option{-G}) can be
5336 used as a shortcut for @option{--listed-incremental} when listing or
5337 extracting incremental backups (for more information, regarding this
5338 option, @pxref{incremental-op}).
5340 When extracting from the incremental backup @GNUTAR{} attempts to
5341 restore the exact state the file system had when the archive was
5342 created. In particular, it will @emph{delete} those files in the file
5343 system that did not exist in their directories when the archive was
5344 created. If you have created several levels of incremental files,
5345 then in order to restore the exact contents the file system had when
5346 the last level was created, you will need to restore from all backups
5347 in turn. Continuing our example, to restore the state of @file{/usr}
5348 file system, one would do@footnote{Notice, that since both archives
5349 were created withouth @option{-P} option (@pxref{absolute}), these
5350 commands should be run from the root file system.}:
5353 $ @kbd{tar --extract \
5354 --listed-incremental=/dev/null \
5355 --file archive.1.tar}
5356 $ @kbd{tar --extract \
5357 --listed-incremental=/dev/null \
5358 --file archive.2.tar}
5361 To list the contents of an incremental archive, use @option{--list}
5362 (@pxref{list}), as usual. To obtain more information about the
5363 archive, use @option{--listed-incremental} or @option{--incremental}
5364 combined with two @option{--verbose} options@footnote{Two
5365 @option{--verbose} options were selected to avoid breaking usual
5366 verbose listing output (@option{--list --verbose}) when using in
5369 @opindex incremental, using with @option{--list}
5370 @opindex listed-incremental, using with @option{--list}
5371 @opindex list, using with @option{--incremental}
5372 @opindex list, using with @option{--listed-incremental}
5373 Versions of @GNUTAR{} up to 1.15.1 used to dump verbatim binary
5374 contents of the DUMPDIR header (with terminating nulls) when
5375 @option{--incremental} or @option{--listed-incremental} option was
5376 given, no matter what the verbosity level. This behavior, and,
5377 especially, the binary output it produced were considered incovenient
5378 and were changed in version 1.16}:
5381 @kbd{tar --list --incremental --verbose --verbose archive.tar}
5384 This command will print, for each directory in the archive, the list
5385 of files in that directory at the time the archive was created. This
5386 information is put out in a format which is both human-readable and
5387 unambiguous for a program: each file name is printed as
5394 where @var{x} is a letter describing the status of the file: @samp{Y}
5395 if the file is present in the archive, @samp{N} if the file is not
5396 included in the archive, or a @samp{D} if the file is a directory (and
5397 is included in the archive). @xref{Dumpdir}, for the detailed
5398 description of dumpdirs and status codes. Each such
5399 line is terminated by a newline character. The last line is followed
5400 by an additional newline to indicate the end of the data.
5402 @anchor{incremental-op}The option @option{--incremental} (@option{-G})
5403 gives the same behavior as @option{--listed-incremental} when used
5404 with @option{--list} and @option{--extract} options. When used with
5405 @option{--create} option, it creates an incremental archive without
5406 creating snapshot file. Thus, it is impossible to create several
5407 levels of incremental backups with @option{--incremental} option.
5410 @section Levels of Backups
5412 An archive containing all the files in the file system is called a
5413 @dfn{full backup} or @dfn{full dump}. You could insure your data by
5414 creating a full dump every day. This strategy, however, would waste a
5415 substantial amount of archive media and user time, as unchanged files
5416 are daily re-archived.
5418 It is more efficient to do a full dump only occasionally. To back up
5419 files between full dumps, you can use @dfn{incremental dumps}. A @dfn{level
5420 one} dump archives all the files that have changed since the last full
5423 A typical dump strategy would be to perform a full dump once a week,
5424 and a level one dump once a day. This means some versions of files
5425 will in fact be archived more than once, but this dump strategy makes
5426 it possible to restore a file system to within one day of accuracy by
5427 only extracting two archives---the last weekly (full) dump and the
5428 last daily (level one) dump. The only information lost would be in
5429 files changed or created since the last daily backup. (Doing dumps
5430 more than once a day is usually not worth the trouble).
5432 @GNUTAR{} comes with scripts you can use to do full
5433 and level-one (actually, even level-two and so on) dumps. Using
5434 scripts (shell programs) to perform backups and restoration is a
5435 convenient and reliable alternative to typing out file name lists
5436 and @command{tar} commands by hand.
5438 Before you use these scripts, you need to edit the file
5439 @file{backup-specs}, which specifies parameters used by the backup
5440 scripts and by the restore script. This file is usually located
5441 in @file{/etc/backup} directory. @xref{Backup Parameters}, for its
5442 detailed description. Once the backup parameters are set, you can
5443 perform backups or restoration by running the appropriate script.
5445 The name of the backup script is @code{backup}. The name of the
5446 restore script is @code{restore}. The following sections describe
5447 their use in detail.
5449 @emph{Please Note:} The backup and restoration scripts are
5450 designed to be used together. While it is possible to restore files by
5451 hand from an archive which was created using a backup script, and to create
5452 an archive by hand which could then be extracted using the restore script,
5453 it is easier to use the scripts. @xref{Incremental Dumps}, before
5454 making such an attempt.
5456 @node Backup Parameters
5457 @section Setting Parameters for Backups and Restoration
5459 The file @file{backup-specs} specifies backup parameters for the
5460 backup and restoration scripts provided with @command{tar}. You must
5461 edit @file{backup-specs} to fit your system configuration and schedule
5462 before using these scripts.
5464 Syntactically, @file{backup-specs} is a shell script, containing
5465 mainly variable assignments. However, any valid shell construct
5466 is allowed in this file. Particularly, you may wish to define
5467 functions within that script (e.g., see @code{RESTORE_BEGIN} below).
5468 For more information about shell script syntax, please refer to
5469 @url{http://www.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html#ta
5470 g_02, the definition of the Shell Command Language}. See also
5471 @ref{Top,,Bash Features,bashref,Bash Reference Manual}.
5473 The shell variables controlling behavior of @code{backup} and
5474 @code{restore} are described in the following subsections.
5477 * General-Purpose Variables::
5478 * Magnetic Tape Control::
5480 * backup-specs example:: An Example Text of @file{Backup-specs}
5483 @node General-Purpose Variables
5484 @subsection General-Purpose Variables
5486 @defvr {Backup variable} ADMINISTRATOR
5487 The user name of the backup administrator. @code{Backup} scripts
5488 sends a backup report to this address.
5491 @defvr {Backup variable} BACKUP_HOUR
5492 The hour at which the backups are done. This can be a number from 0
5493 to 23, or the time specification in form @var{hours}:@var{minutes},
5494 or the string @samp{now}.
5496 This variable is used by @code{backup}. Its value may be overridden
5497 using @option{--time} option (@pxref{Scripted Backups}).
5500 @defvr {Backup variable} TAPE_FILE
5502 The device @command{tar} writes the archive to. If @var{TAPE_FILE}
5503 is a remote archive (@pxref{remote-dev}), backup script will suppose
5504 that your @command{mt} is able to access remote devices. If @var{RSH}
5505 (@pxref{RSH}) is set, @option{--rsh-command} option will be added to
5506 invocations of @command{mt}.
5509 @defvr {Backup variable} BLOCKING
5511 The blocking factor @command{tar} will use when writing the dump archive.
5512 @xref{Blocking Factor}.
5515 @defvr {Backup variable} BACKUP_DIRS
5517 A list of file systems to be dumped (for @code{backup}), or restored
5518 (for @code{restore}). You can include any directory
5519 name in the list --- subdirectories on that file system will be
5520 included, regardless of how they may look to other networked machines.
5521 Subdirectories on other file systems will be ignored.
5523 The host name specifies which host to run @command{tar} on, and should
5524 normally be the host that actually contains the file system. However,
5525 the host machine must have @GNUTAR{} installed, and
5526 must be able to access the directory containing the backup scripts and
5527 their support files using the same file name that is used on the
5528 machine where the scripts are run (ie. what @command{pwd} will print
5529 when in that directory on that machine). If the host that contains
5530 the file system does not have this capability, you can specify another
5531 host as long as it can access the file system through NFS.
5533 If the list of file systems is very long you may wish to put it
5534 in a separate file. This file is usually named
5535 @file{/etc/backup/dirs}, but this name may be overridden in
5536 @file{backup-specs} using @code{DIRLIST} variable.
5539 @defvr {Backup variable} DIRLIST
5541 A path to the file containing the list of the file systems to backup
5542 or restore. By default it is @file{/etc/backup/dirs}.
5545 @defvr {Backup variable} BACKUP_FILES
5547 A list of individual files to be dumped (for @code{backup}), or restored
5548 (for @code{restore}). These should be accessible from the machine on
5549 which the backup script is run.
5551 If the list of file systems is very long you may wish to store it
5552 in a separate file. This file is usually named
5553 @file{/etc/backup/files}, but this name may be overridden in
5554 @file{backup-specs} using @code{FILELIST} variable.
5557 @defvr {Backup variable} FILELIST
5559 A path to the file containing the list of the individual files to backup
5560 or restore. By default it is @file{/etc/backup/files}.
5563 @defvr {Backup variable} MT
5565 Full file name of @command{mt} binary.
5568 @defvr {Backup variable} RSH
5570 Full file name of @command{rsh} binary or its equivalent. You may wish to
5571 set it to @code{ssh}, to improve security. In this case you will have
5572 to use public key authentication.
5575 @defvr {Backup variable} RSH_COMMAND
5577 Full file name of @command{rsh} binary on remote mashines. This will
5578 be passed via @option{--rsh-command} option to the remote invocation
5582 @defvr {Backup variable} VOLNO_FILE
5584 Name of temporary file to hold volume numbers. This needs to be accessible
5585 by all the machines which have file systems to be dumped.
5588 @defvr {Backup variable} XLIST
5590 Name of @dfn{exclude file list}. An @dfn{exclude file list} is a file
5591 located on the remote machine and containing the list of files to
5592 be excluded from the backup. Exclude file lists are searched in
5593 /etc/tar-backup directory. A common use for exclude file lists
5594 is to exclude files containing security-sensitive information
5595 (e.g., @file{/etc/shadow} from backups).
5597 This variable affects only @code{backup}.
5600 @defvr {Backup variable} SLEEP_TIME
5602 Time to sleep between dumps of any two successive file systems
5604 This variable affects only @code{backup}.
5607 @defvr {Backup variable} DUMP_REMIND_SCRIPT
5609 Script to be run when it's time to insert a new tape in for the next
5610 volume. Administrators may want to tailor this script for their site.
5611 If this variable isn't set, @GNUTAR{} will display its built-in
5612 prompt, and will expect confirmation from the console.
5614 The built-in prompt for POSIX locale is:
5617 Prepare volume #@var{n} for `@var{archive}' and hit return:
5621 where @var{n} is the ordinal number of the volume to be created and
5622 @var{archive} is archive file or device name.
5624 If you run @GNUTAR{} under a different locale, the translation of
5625 the above prompt to the locale's language will be used.
5629 @defvr {Backup variable} SLEEP_MESSAGE
5631 Message to display on the terminal while waiting for dump time. Usually
5632 this will just be some literal text.
5635 @defvr {Backup variable} TAR
5637 Full file name of the @GNUTAR{} executable. If this is not set, backup
5638 scripts will search @command{tar} in the current shell path.
5641 @node Magnetic Tape Control
5642 @subsection Magnetic Tape Control
5644 Backup scripts access tape device using special @dfn{hook functions}.
5645 These functions take a single argument -- the name of the tape
5646 device. Their names are kept in the following variables:
5648 @defvr {Backup variable} MT_BEGIN
5649 The name of @dfn{begin} function. This function is called before
5650 accessing the drive. By default it retensions the tape:
5656 mt -f "$1" retension
5661 @defvr {Backup variable} MT_REWIND
5662 The name of @dfn{rewind} function. The default definition is as
5675 @defvr {Backup variable} MT_OFFLINE
5676 The name of the function switching the tape off line. By default
5677 it is defined as follows:
5680 MT_OFFLINE=mt_offline
5688 @defvr {Backup variable} MT_STATUS
5689 The name of the function used to obtain the status of the archive device,
5690 including error count. Default definition:
5702 @subsection User Hooks
5704 @dfn{User hooks} are shell functions executed before and after
5705 each @command{tar} invocation. Thus, there are @dfn{backup
5706 hooks}, which are executed before and after dumping each file
5707 system, and @dfn{restore hooks}, executed before and
5708 after restoring a file system. Each user hook is a shell function
5709 taking four arguments:
5711 @deffn {User Hook Function} hook @var{level} @var{host} @var{fs} @var{fsname}
5716 Current backup or restore level.
5719 Name or IP address of the host machine being dumped or restored.
5722 Full path name to the file system being dumped or restored.
5725 File system name with directory separators replaced with colons. This
5726 is useful, e.g., for creating unique files.
5730 Following variables keep the names of user hook functions
5732 @defvr {Backup variable} DUMP_BEGIN
5733 Dump begin function. It is executed before dumping the file system.
5736 @defvr {Backup variable} DUMP_END
5737 Executed after dumping the file system.
5740 @defvr {Backup variable} RESTORE_BEGIN
5741 Executed before restoring the file system.
5744 @defvr {Backup variable} RESTORE_END
5745 Executed after restoring the file system.
5748 @node backup-specs example
5749 @subsection An Example Text of @file{Backup-specs}
5751 The following is an example of @file{backup-specs}:
5754 # site-specific parameters for file system backup.
5756 ADMINISTRATOR=friedman
5758 TAPE_FILE=/dev/nrsmt0
5760 # Use @code{ssh} instead of the less secure @code{rsh}
5762 RSH_COMMAND=/usr/bin/ssh
5764 # Override MT_STATUS function:
5770 # Disable MT_OFFLINE function
5787 apple-gunkies:/com/mailer/gnu
5788 apple-gunkies:/com/archive/gnu"
5790 BACKUP_FILES="/com/mailer/aliases /com/mailer/league*[a-z]"
5794 @node Scripted Backups
5795 @section Using the Backup Scripts
5797 The syntax for running a backup script is:
5800 backup --level=@var{level} --time=@var{time}
5803 The @option{level} option requests the dump level. Thus, to produce
5804 a full dump, specify @code{--level=0} (this is the default, so
5805 @option{--level} may be omitted if its value is @code{0}).
5806 @footnote{For backward compatibility, the @code{backup} will also
5807 try to deduce the requested dump level from the name of the
5808 script itself. If the name consists of a string @samp{level-}
5809 followed by a single decimal digit, that digit is taken as
5810 the dump level number. Thus, you may create a link from @code{backup}
5811 to @code{level-1} and then run @code{level-1} whenever you need to
5812 create a level one dump.}
5814 The @option{--time} option determines when should the backup be
5815 run. @var{Time} may take three forms:
5818 @item @var{hh}:@var{mm}
5820 The dump must be run at @var{hh} hours @var{mm} minutes.
5824 The dump must be run at @var{hh} hours
5828 The dump must be run immediately.
5831 You should start a script with a tape or disk mounted. Once you
5832 start a script, it prompts you for new tapes or disks as it
5833 needs them. Media volumes don't have to correspond to archive
5834 files --- a multi-volume archive can be started in the middle of a
5835 tape that already contains the end of another multi-volume archive.
5836 The @code{restore} script prompts for media by its archive volume,
5837 so to avoid an error message you should keep track of which tape
5838 (or disk) contains which volume of the archive (@pxref{Scripted
5841 The backup scripts write two files on the file system. The first is a
5842 record file in @file{/etc/tar-backup/}, which is used by the scripts
5843 to store and retrieve information about which files were dumped. This
5844 file is not meant to be read by humans, and should not be deleted by
5845 them. @xref{Snapshot Files}, for a more detailed explanation of this
5848 The second file is a log file containing the names of the file systems
5849 and files dumped, what time the backup was made, and any error
5850 messages that were generated, as well as how much space was left in
5851 the media volume after the last volume of the archive was written.
5852 You should check this log file after every backup. The file name is
5853 @file{log-@var{mm-dd-yyyy}-level-@var{n}}, where @var{mm-dd-yyyy}
5854 represents current date, and @var{n} represents current dump level number.
5856 The script also prints the name of each system being dumped to the
5859 Following is the full list of options accepted by @code{backup}
5863 @item -l @var{level}
5864 @itemx --level=@var{level}
5865 Do backup level @var{level} (default 0).
5869 Force backup even if today's log file already exists.
5871 @item -v[@var{level}]
5872 @itemx --verbose[=@var{level}]
5873 Set verbosity level. The higher the level is, the more debugging
5874 information will be output during execution. Devault @var{level}
5875 is 100, which means the highest debugging level.
5877 @item -t @var{start-time}
5878 @itemx --time=@var{start-time}
5879 Wait till @var{time}, then do backup.
5883 Display short help message and exit.
5887 Display information about the program's name, version, origin and legal
5888 status, all on standard output, and then exit successfully.
5892 @node Scripted Restoration
5893 @section Using the Restore Script
5895 To restore files that were archived using a scripted backup, use the
5896 @code{restore} script. Its usage is quite straightforward. In the
5897 simplest form, invoke @code{restore --all}, it will
5898 then restore all the file systems and files specified in
5899 @file{backup-specs} (@pxref{General-Purpose Variables,BACKUP_DIRS}).
5901 You may select the file systems (and/or files) to restore by
5902 giving @code{restore} list of @dfn{patterns} in its command
5903 line. For example, running
5910 will restore all file systems on the machine @samp{albert}. A more
5911 complicated example:
5914 restore 'albert:*' '*:/var'
5918 This command will restore all file systems on the machine @samp{albert}
5919 as well as @file{/var} file system on all machines.
5921 By default @code{restore} will start restoring files from the lowest
5922 available dump level (usually zero) and will continue through
5923 all available dump levels. There may be situations where such a
5924 thorough restore is not necessary. For example, you may wish to
5925 restore only files from the recent level one backup. To do so,
5926 use @option{--level} option, as shown in the example below:
5932 The full list of options accepted by @code{restore} follows:
5937 Restore all file systems and files specified in @file{backup-specs}
5939 @item -l @var{level}
5940 @itemx --level=@var{level}
5941 Start restoring from the given backup level, instead of the default 0.
5943 @item -v[@var{level}]
5944 @itemx --verbose[=@var{level}]
5945 Set verbosity level. The higher the level is, the more debugging
5946 information will be output during execution. Devault @var{level}
5947 is 100, which means the highest debugging level.
5951 Display short help message and exit.
5955 Display information about the program's name, version, origin and legal
5956 status, all on standard output, and then exit successfully.
5959 You should start the restore script with the media containing the
5960 first volume of the archive mounted. The script will prompt for other
5961 volumes as they are needed. If the archive is on tape, you don't need
5962 to rewind the tape to to its beginning---if the tape head is
5963 positioned past the beginning of the archive, the script will rewind
5964 the tape as needed. @FIXME-xref{Media, for a discussion of tape
5968 @strong{Warning:} The script will delete files from the active file
5969 system if they were not in the file system when the archive was made.
5972 @xref{Incremental Dumps}, for an explanation of how the script makes
5976 @chapter Choosing Files and Names for @command{tar}
5979 Certain options to @command{tar} enable you to specify a name for your
5980 archive. Other options let you decide which files to include or exclude
5981 from the archive, based on when or whether files were modified, whether
5982 the file names do or don't match specified patterns, or whether files
5983 are in specified directories.
5985 This chapter discusses these options in detail.
5988 * file:: Choosing the Archive's Name
5989 * Selecting Archive Members::
5990 * files:: Reading Names from a File
5991 * exclude:: Excluding Some Files
5992 * wildcards:: Wildcards Patterns and Matching
5993 * quoting styles:: Ways of Quoting Special Characters in Names
5994 * transform:: Modifying File and Member Names
5995 * after:: Operating Only on New Files
5996 * recurse:: Descending into Directories
5997 * one:: Crossing File System Boundaries
6001 @section Choosing and Naming Archive Files
6004 @cindex Naming an archive
6005 @cindex Archive Name
6006 @cindex Choosing an archive file
6007 @cindex Where is the archive?
6008 By default, @command{tar} uses an archive file name that was compiled when
6009 it was built on the system; usually this name refers to some physical
6010 tape drive on the machine. However, the person who installed @command{tar}
6011 on the system may not have set the default to a meaningful value as far as
6012 most users are concerned. As a result, you will usually want to tell
6013 @command{tar} where to find (or create) the archive. The
6014 @option{--file=@var{archive-name}} (@option{-f @var{archive-name}})
6015 option allows you to either specify or name a file to use as the archive
6016 instead of the default archive file location.
6019 @opindex file, short description
6020 @item --file=@var{archive-name}
6021 @itemx -f @var{archive-name}
6022 Name the archive to create or operate on. Use in conjunction with
6026 For example, in this @command{tar} command,
6029 $ @kbd{tar -cvf collection.tar blues folk jazz}
6033 @file{collection.tar} is the name of the archive. It must directly
6034 follow the @option{-f} option, since whatever directly follows @option{-f}
6035 @emph{will} end up naming the archive. If you neglect to specify an
6036 archive name, you may end up overwriting a file in the working directory
6037 with the archive you create since @command{tar} will use this file's name
6038 for the archive name.
6040 An archive can be saved as a file in the file system, sent through a
6041 pipe or over a network, or written to an I/O device such as a tape,
6042 floppy disk, or CD write drive.
6044 @cindex Writing new archives
6045 @cindex Archive creation
6046 If you do not name the archive, @command{tar} uses the value of the
6047 environment variable @env{TAPE} as the file name for the archive. If
6048 that is not available, @command{tar} uses a default, compiled-in archive
6049 name, usually that for tape unit zero (ie. @file{/dev/tu00}).
6051 @cindex Standard input and output
6052 @cindex tar to standard input and output
6053 If you use @file{-} as an @var{archive-name}, @command{tar} reads the
6054 archive from standard input (when listing or extracting files), or
6055 writes it to standard output (when creating an archive). If you use
6056 @file{-} as an @var{archive-name} when modifying an archive,
6057 @command{tar} reads the original archive from its standard input and
6058 writes the entire new archive to its standard output.
6060 The following example is a convenient way of copying directory
6061 hierarchy from @file{sourcedir} to @file{targetdir}.
6064 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xpf -)}
6067 The @option{-C} option allows to avoid using subshells:
6070 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xpf -}
6073 In both examples above, the leftmost @command{tar} invocation archives
6074 the contents of @file{sourcedir} to the standard output, while the
6075 rightmost one reads this archive from its standard input and
6076 extracts it. The @option{-p} option tells it to restore permissions
6077 of the extracted files.
6079 @cindex Remote devices
6080 @cindex tar to a remote device
6082 To specify an archive file on a device attached to a remote machine,
6086 @kbd{--file=@var{hostname}:/@var{dev}/@var{file-name}}
6090 @command{tar} will complete the remote connection, if possible, and
6091 prompt you for a username and password. If you use
6092 @option{--file=@@@var{hostname}:/@var{dev}/@var{file-name}}, @command{tar}
6093 will complete the remote connection, if possible, using your username
6094 as the username on the remote machine.
6096 @cindex Local and remote archives
6097 @anchor{local and remote archives}
6098 If the archive file name includes a colon (@samp{:}), then it is assumed
6099 to be a file on another machine. If the archive file is
6100 @samp{@var{user}@@@var{host}:@var{file}}, then @var{file} is used on the
6101 host @var{host}. The remote host is accessed using the @command{rsh}
6102 program, with a username of @var{user}. If the username is omitted
6103 (along with the @samp{@@} sign), then your user name will be used.
6104 (This is the normal @command{rsh} behavior.) It is necessary for the
6105 remote machine, in addition to permitting your @command{rsh} access, to
6106 have the @file{rmt} program installed (This command is included in
6107 the @GNUTAR{} distribution and by default is installed under
6108 @file{@var{prefix}/libexec/rmt}, were @var{prefix} means your
6109 installation prefix). If you need to use a file whose name includes a
6110 colon, then the remote tape drive behavior
6111 can be inhibited by using the @option{--force-local} option.
6113 When the archive is being created to @file{/dev/null}, @GNUTAR{}
6114 tries to minimize input and output operations. The Amanda backup
6115 system, when used with @GNUTAR{}, has an initial sizing pass which
6118 @node Selecting Archive Members
6119 @section Selecting Archive Members
6120 @cindex Specifying files to act on
6121 @cindex Specifying archive members
6123 @dfn{File Name arguments} specify which files in the file system
6124 @command{tar} operates on, when creating or adding to an archive, or which
6125 archive members @command{tar} operates on, when reading or deleting from
6126 an archive. @xref{Operations}.
6128 To specify file names, you can include them as the last arguments on
6129 the command line, as follows:
6131 @kbd{tar} @var{operation} [@var{option1} @var{option2} @dots{}] [@var{file name-1} @var{file name-2} @dots{}]
6134 If a file name begins with dash (@samp{-}), precede it with
6135 @option{--add-file} option to prevent it from being treated as an
6138 If you specify a directory name as a file name argument, all the files
6139 in that directory are operated on by @command{tar}.
6141 If you do not specify files, @command{tar} behavior differs depending
6142 on the operation mode as described below:
6144 When @command{tar} is invoked with @option{--create} (@option{-c}),
6145 @command{tar} will stop immediately, reporting the following:
6149 $ @kbd{tar cf a.tar}
6150 tar: Cowardly refusing to create an empty archive
6151 Try `tar --help' or `tar --usage' for more information.
6155 If you specify either @option{--list} (@option{-t}) or
6156 @option{--extract} (@option{--get}, @option{-x}), @command{tar}
6157 operates on all the archive members in the archive.
6159 If run with @option{--diff} option, tar will compare the archive with
6160 the contents of the current working directory.
6162 If you specify any other operation, @command{tar} does nothing.
6164 By default, @command{tar} takes file names from the command line. However,
6165 there are other ways to specify file or member names, or to modify the
6166 manner in which @command{tar} selects the files or members upon which to
6167 operate. In general, these methods work both for specifying the names
6168 of files and archive members.
6171 @section Reading Names from a File
6173 @cindex Reading file names from a file
6174 @cindex Lists of file names
6175 @cindex File Name arguments, alternatives
6176 Instead of giving the names of files or archive members on the command
6177 line, you can put the names into a file, and then use the
6178 @option{--files-from=@var{file-of-names}} (@option{-T
6179 @var{file-of-names}}) option to @command{tar}. Give the name of the
6180 file which contains the list of files to include as the argument to
6181 @option{--files-from}. In the list, the file names should be separated by
6182 newlines. You will frequently use this option when you have generated
6183 the list of files to archive with the @command{find} utility.
6187 @item --files-from=@var{file-name}
6188 @itemx -T @var{file-name}
6189 Get names to extract or create from file @var{file-name}.
6192 If you give a single dash as a file name for @option{--files-from}, (i.e.,
6193 you specify either @code{--files-from=-} or @code{-T -}), then the file
6194 names are read from standard input.
6196 Unless you are running @command{tar} with @option{--create}, you can not use
6197 both @code{--files-from=-} and @code{--file=-} (@code{-f -}) in the same
6200 Any number of @option{-T} options can be given in the command line.
6202 The following example shows how to use @command{find} to generate a list of
6203 files smaller than 400K in length and put that list into a file
6204 called @file{small-files}. You can then use the @option{-T} option to
6205 @command{tar} to specify the files from that file, @file{small-files}, to
6206 create the archive @file{little.tgz}. (The @option{-z} option to
6207 @command{tar} compresses the archive with @command{gzip}; @pxref{gzip} for
6211 $ @kbd{find . -size -400 -print > small-files}
6212 $ @kbd{tar -c -v -z -T small-files -f little.tgz}
6216 In the file list given by @option{-T} option, any file name beginning
6217 with @samp{-} character is considered a @command{tar} option and is
6218 processed accordingly.@footnote{Versions of @GNUTAR{} up to 1.15.1
6219 recognized only @option{-C} option in file lists, and only if the
6220 option and its argument occupied two consecutive lines.} For example,
6221 the common use of this feature is to change to another directory by
6222 specifying @option{-C} option:
6232 $ @kbd{tar -c -f foo.tar --files-from list}
6237 In this example, @command{tar} will first switch to @file{/etc}
6238 directory and add files @file{passwd} and @file{hosts} to the
6239 archive. Then it will change to @file{/lib} directory and will archive
6240 the file @file{libc.a}. Thus, the resulting archive @file{foo.tar} will
6245 $ @kbd{tar tf foo.tar}
6253 @opindex directory, using in @option{--files-from} argument
6254 Notice that the option parsing algorithm used with @option{-T} is
6255 stricter than the one used by shell. Namely, when specifying option
6256 arguments, you should observe the following rules:
6260 When using short (single-letter) option form, its argument must
6261 immediately follow the option letter, without any intervening
6262 whitespace. For example: @code{-Cdir}.
6265 When using long option form, the option argument must be separated
6266 from the option by a single equal sign. No whitespace is allowed on
6267 any side of the equal sign. For example: @code{--directory=dir}.
6270 For both short and long option forms, the option argument can be given
6271 on the next line after the option name, e.g.:
6292 If you happen to have a file whose name starts with @samp{-},
6293 precede it with @option{--add-file} option to prevent it from
6294 being recognized as an option. For example: @code{--add-file=--my-file}.
6301 @subsection @code{NUL} Terminated File Names
6303 @cindex File names, terminated by @code{NUL}
6304 @cindex @code{NUL} terminated file names
6305 The @option{--null} option causes
6306 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}})
6307 to read file names terminated by a @code{NUL} instead of a newline, so
6308 files whose names contain newlines can be archived using
6309 @option{--files-from}.
6314 Only consider @code{NUL} terminated file names, instead of files that
6315 terminate in a newline.
6318 The @option{--null} option is just like the one in @acronym{GNU}
6319 @command{xargs} and @command{cpio}, and is useful with the
6320 @option{-print0} predicate of @acronym{GNU} @command{find}. In
6321 @command{tar}, @option{--null} also disables special handling for
6322 file names that begin with dash.
6324 This example shows how to use @command{find} to generate a list of files
6325 larger than 800K in length and put that list into a file called
6326 @file{long-files}. The @option{-print0} option to @command{find} is just
6327 like @option{-print}, except that it separates files with a @code{NUL}
6328 rather than with a newline. You can then run @command{tar} with both the
6329 @option{--null} and @option{-T} options to specify that @command{tar} get the
6330 files from that file, @file{long-files}, to create the archive
6331 @file{big.tgz}. The @option{--null} option to @command{tar} will cause
6332 @command{tar} to recognize the @code{NUL} separator between files.
6335 $ @kbd{find . -size +800 -print0 > long-files}
6336 $ @kbd{tar -c -v --null --files-from=long-files --file=big.tar}
6339 @FIXME{say anything else here to conclude the section?}
6342 @section Excluding Some Files
6345 @cindex File names, excluding files by
6346 @cindex Excluding files by name and pattern
6347 @cindex Excluding files by file system
6348 To avoid operating on files whose names match a particular pattern,
6349 use the @option{--exclude} or @option{--exclude-from} options.
6353 @item --exclude=@var{pattern}
6354 Causes @command{tar} to ignore files that match the @var{pattern}.
6358 The @option{--exclude=@var{pattern}} option prevents any file or
6359 member whose name matches the shell wildcard (@var{pattern}) from
6361 For example, to create an archive with all the contents of the directory
6362 @file{src} except for files whose names end in @file{.o}, use the
6363 command @samp{tar -cf src.tar --exclude='*.o' src}.
6365 You may give multiple @option{--exclude} options.
6368 @opindex exclude-from
6369 @item --exclude-from=@var{file}
6370 @itemx -X @var{file}
6371 Causes @command{tar} to ignore files that match the patterns listed in
6375 @findex exclude-from
6376 Use the @option{--exclude-from} option to read a
6377 list of patterns, one per line, from @var{file}; @command{tar} will
6378 ignore files matching those patterns. Thus if @command{tar} is
6379 called as @w{@samp{tar -c -X foo .}} and the file @file{foo} contains a
6380 single line @file{*.o}, no files whose names end in @file{.o} will be
6381 added to the archive.
6384 @opindex exclude-caches
6385 @item --exclude-caches
6386 Causes @command{tar} to ignore directories containing a cache directory tag.
6389 @findex exclude-caches
6390 When creating an archive, the @option{--exclude-caches} option causes
6391 @command{tar} to exclude all directories that contain a @dfn{cache
6392 directory tag}. A cache directory tag is a short file with the
6393 well-known name @file{CACHEDIR.TAG} and having a standard header
6394 specified in @url{http://www.brynosaurus.com/cachedir/spec.html}.
6395 Various applications write cache directory tags into directories they
6396 use to hold regenerable, non-precious data, so that such data can be
6397 more easily excluded from backups.
6400 * problems with exclude::
6403 @node problems with exclude
6404 @unnumberedsubsec Problems with Using the @code{exclude} Options
6406 @opindex exclude, potential problems with
6407 Some users find @samp{exclude} options confusing. Here are some common
6412 The main operating mode of @command{tar} does not act on a path name
6413 explicitly listed on the command line if one of its file name
6414 components is excluded. In the example above, if
6415 you create an archive and exclude files that end with @samp{*.o}, but
6416 explicitly name the file @samp{dir.o/foo} after all the options have been
6417 listed, @samp{dir.o/foo} will be excluded from the archive.
6420 You can sometimes confuse the meanings of @option{--exclude} and
6421 @option{--exclude-from}. Be careful: use @option{--exclude} when files
6422 to be excluded are given as a pattern on the command line. Use
6423 @option{--exclude-from} to introduce the name of a file which contains
6424 a list of patterns, one per line; each of these patterns can exclude
6425 zero, one, or many files.
6428 When you use @option{--exclude=@var{pattern}}, be sure to quote the
6429 @var{pattern} parameter, so @GNUTAR{} sees wildcard characters
6430 like @samp{*}. If you do not do this, the shell might expand the
6431 @samp{*} itself using files at hand, so @command{tar} might receive a
6432 list of files instead of one pattern, or none at all, making the
6433 command somewhat illegal. This might not correspond to what you want.
6438 $ @kbd{tar -c -f @var{archive.tar} --exclude '*.o' @var{directory}}
6446 $ @kbd{tar -c -f @var{archive.tar} --exclude *.o @var{directory}}
6450 You must use use shell syntax, or globbing, rather than @code{regexp}
6451 syntax, when using exclude options in @command{tar}. If you try to use
6452 @code{regexp} syntax to describe files to be excluded, your command
6456 @FIXME{The change in semantics must have occurred before 1.11,
6457 so I doubt if it is worth mentioning at all. Anyway, should at
6458 least specify in which version the semantics changed.}
6459 In earlier versions of @command{tar}, what is now the
6460 @option{--exclude-from} option was called @option{--exclude} instead.
6461 Now, @option{--exclude} applies to patterns listed on the command
6462 line and @option{--exclude-from} applies to patterns listed in a
6468 @section Wildcards Patterns and Matching
6470 @dfn{Globbing} is the operation by which @dfn{wildcard} characters,
6471 @samp{*} or @samp{?} for example, are replaced and expanded into all
6472 existing files matching the given pattern. @GNUTAR{} can use wildcard
6473 patterns for matching (or globbing) archive members when extracting
6474 from or listing an archive. Wildcard patterns are also used for
6475 verifying volume labels of @command{tar} archives. This section has the
6476 purpose of explaining wildcard syntax for @command{tar}.
6478 @FIXME{the next few paragraphs need work.}
6480 A @var{pattern} should be written according to shell syntax, using wildcard
6481 characters to effect globbing. Most characters in the pattern stand
6482 for themselves in the matched string, and case is significant: @samp{a}
6483 will match only @samp{a}, and not @samp{A}. The character @samp{?} in the
6484 pattern matches any single character in the matched string. The character
6485 @samp{*} in the pattern matches zero, one, or more single characters in
6486 the matched string. The character @samp{\} says to take the following
6487 character of the pattern @emph{literally}; it is useful when one needs to
6488 match the @samp{?}, @samp{*}, @samp{[} or @samp{\} characters, themselves.
6490 The character @samp{[}, up to the matching @samp{]}, introduces a character
6491 class. A @dfn{character class} is a list of acceptable characters
6492 for the next single character of the matched string. For example,
6493 @samp{[abcde]} would match any of the first five letters of the alphabet.
6494 Note that within a character class, all of the ``special characters''
6495 listed above other than @samp{\} lose their special meaning; for example,
6496 @samp{[-\\[*?]]} would match any of the characters, @samp{-}, @samp{\},
6497 @samp{[}, @samp{*}, @samp{?}, or @samp{]}. (Due to parsing constraints,
6498 the characters @samp{-} and @samp{]} must either come @emph{first} or
6499 @emph{last} in a character class.)
6501 @cindex Excluding characters from a character class
6502 @cindex Character class, excluding characters from
6503 If the first character of the class after the opening @samp{[}
6504 is @samp{!} or @samp{^}, then the meaning of the class is reversed.
6505 Rather than listing character to match, it lists those characters which
6506 are @emph{forbidden} as the next single character of the matched string.
6508 Other characters of the class stand for themselves. The special
6509 construction @samp{[@var{a}-@var{e}]}, using an hyphen between two
6510 letters, is meant to represent all characters between @var{a} and
6513 @FIXME{need to add a sentence or so here to make this clear for those
6514 who don't have dan around.}
6516 Periods (@samp{.}) or forward slashes (@samp{/}) are not considered
6517 special for wildcard matches. However, if a pattern completely matches
6518 a directory prefix of a matched string, then it matches the full matched
6519 string: thus, excluding a directory also excludes all the files beneath it.
6522 * controlling pattern-matching::
6525 @node controlling pattern-matching
6526 @unnumberedsubsec Controlling Pattern-Matching
6528 For the purposes of this section, we call @dfn{exclusion members} all
6529 member names obtained while processing @option{--exclude} and
6530 @option{--exclude-from} options, and @dfn{inclusion members} those
6531 member names that were given in the command line or read from the file
6532 specified with @option{--files-from} option.
6534 These two pairs of member lists are used in the following operations:
6535 @option{--diff}, @option{--extract}, @option{--list},
6538 There are no inclusion members in create mode (@option{--create} and
6539 @option{--append}), since in this mode the names obtained from the
6540 command line refer to @emph{files}, not archive members.
6542 By default, inclusion members are compared with archive members
6543 literally @footnote{Notice that earlier @GNUTAR{} versions used
6544 globbing for inclusion members, which contradicted to UNIX98
6545 specification and was not documented. @xref{Changes}, for more
6546 information on this and other changes.} and exclusion members are
6547 treated as globbing patterns. For example:
6551 $ @kbd{tar tf foo.tar}
6556 # @i{Member names are used verbatim:}
6557 $ @kbd{tar -xf foo.tar -v '[remarks]'}
6559 # @i{Exclude member names are globbed:}
6560 $ @kbd{tar -xf foo.tar -v --exclude '*.c'}
6566 This behavior can be altered by using the following options:
6571 Treat all member names as wildcards.
6573 @opindex no-wildcards
6574 @item --no-wildcards
6575 Treat all member names as literal strings.
6578 Thus, to extract files whose names end in @samp{.c}, you can use:
6581 $ @kbd{tar -xf foo.tar -v --wildcards '*.c'}
6587 Notice quoting of the pattern to prevent the shell from interpreting
6590 The effect of @option{--wildcards} option is cancelled by
6591 @option{--no-wildcards}. This can be used to pass part of
6592 the command line arguments verbatim and other part as globbing
6593 patterns. For example, the following invocation:
6596 $ @kbd{tar -xf foo.tar --wildcards '*.txt' --no-wildcards '[remarks]'}
6600 instructs @command{tar} to extract from @file{foo.tar} all files whose
6601 names end in @samp{.txt} and the file named @file{[remarks]}.
6603 Normally, a pattern matches a name if an initial subsequence of the
6604 name's components matches the pattern, where @samp{*}, @samp{?}, and
6605 @samp{[...]} are the usual shell wildcards, @samp{\} escapes wildcards,
6606 and wildcards can match @samp{/}.
6608 Other than optionally stripping leading @samp{/} from names
6609 (@pxref{absolute}), patterns and names are used as-is. For
6610 example, trailing @samp{/} is not trimmed from a user-specified name
6611 before deciding whether to exclude it.
6613 However, this matching procedure can be altered by the options listed
6614 below. These options accumulate. For example:
6617 --ignore-case --exclude='makefile' --no-ignore-case ---exclude='readme'
6621 ignores case when excluding @samp{makefile}, but not when excluding
6626 @opindex no-anchored
6628 @itemx --no-anchored
6629 If anchored, a pattern must match an initial subsequence
6630 of the name's components. Otherwise, the pattern can match any
6631 subsequence. Default is @option{--no-anchored} for exclusion members
6632 and @option{--anchored} inclusion members.
6634 @opindex ignore-case
6635 @opindex no-ignore-case
6637 @itemx --no-ignore-case
6638 When ignoring case, upper-case patterns match lower-case names and vice versa.
6639 When not ignoring case (the default), matching is case-sensitive.
6641 @opindex wildcards-match-slash
6642 @opindex no-wildcards-match-slash
6643 @item --wildcards-match-slash
6644 @itemx --no-wildcards-match-slash
6645 When wildcards match slash (the default for exclusion members), a
6646 wildcard like @samp{*} in the pattern can match a @samp{/} in the
6647 name. Otherwise, @samp{/} is matched only by @samp{/}.
6651 The @option{--recursion} and @option{--no-recursion} options
6652 (@pxref{recurse}) also affect how member patterns are interpreted. If
6653 recursion is in effect, a pattern matches a name if it matches any of
6654 the name's parent directories.
6656 The following table summarizes pattern-matching default values:
6658 @multitable @columnfractions .3 .7
6659 @headitem Members @tab Default settings
6660 @item Inclusion @tab @option{--no-wildcards --anchored --no-wildcards-match-slash}
6661 @item Exclusion @tab @option{--wildcards --no-anchored --wildcards-match-slash}
6664 @node quoting styles
6665 @section Quoting Member Names
6667 When displaying member names, @command{tar} takes care to avoid
6668 ambiguities caused by certain characters. This is called @dfn{name
6669 quoting}. The characters in question are:
6672 @item Non-printable control characters:
6674 @multitable @columnfractions 0.20 0.10 0.60
6675 @headitem Character @tab ASCII @tab Character name
6676 @item \a @tab 7 @tab Audible bell
6677 @item \b @tab 8 @tab Backspace
6678 @item \f @tab 12 @tab Form feed
6679 @item \n @tab 10 @tab New line
6680 @item \r @tab 13 @tab Carriage return
6681 @item \t @tab 9 @tab Horizontal tabulation
6682 @item \v @tab 11 @tab Vertical tabulation
6685 @item Space (ASCII 32)
6687 @item Single and double quotes (@samp{'} and @samp{"})
6689 @item Backslash (@samp{\})
6692 The exact way @command{tar} uses to quote these characters depends on
6693 the @dfn{quoting style}. The default quoting style, called
6694 @dfn{escape} (see below), uses backslash notation to represent control
6695 characters, space and backslash. Using this quoting style, control
6696 characters are represented as listed in column @samp{Character} in the
6697 above table, a space is printed as @samp{\ } and a backslash as @samp{\\}.
6699 @GNUTAR{} offers seven distinct quoting styles, which can be selected
6700 using @option{--quoting-style} option:
6703 @item --quoting-style=@var{style}
6704 @opindex quoting-style
6706 Sets quoting style. Valid values for @var{style} argument are:
6707 literal, shell, shell-always, c, escape, locale, clocale.
6710 These styles are described in detail below. To illustrate their
6711 effect, we will use an imaginary tar archive @file{arch.tar}
6712 containing the following members:
6716 # 1. Contains horizontal tabulation character.
6718 # 2. Contains newline character
6721 # 3. Contains a space
6723 # 4. Contains double quotes
6725 # 5. Contains single quotes
6727 # 6. Contains a backslash character:
6732 Here is how usual @command{ls} command would have listed them, if they
6733 had existed in the current working directory:
6751 No quoting, display each character as is:
6755 $ @kbd{tar tf arch.tar --quoting-style=literal}
6768 Display characters the same way Bourne shell does:
6769 control characters, except @samp{\t} and @samp{\n}, are printed using
6770 backslash escapes, @samp{\t} and @samp{\n} are printed as is, and a
6771 single quote is printed as @samp{\'}. If a name contains any quoted
6772 characters, it is enclosed in single quotes. In particular, if a name
6773 contains single quotes, it is printed as several single-quoted strings:
6777 $ @kbd{tar tf arch.tar --quoting-style=shell}
6780 './a'\''single'\''quote'
6790 Same as @samp{shell}, but the names are always enclosed in single
6795 $ @kbd{tar tf arch.tar --quoting-style=shell-always}
6798 './a'\''single'\''quote'
6808 Use the notation of the C programming language. All names are
6809 enclosed in double quotes. Control characters are quoted using
6810 backslash notations, double quotes are represented as @samp{\"},
6811 backslash characters are represented as @samp{\\}. Single quotes and
6812 spaces are not quoted:
6816 $ @kbd{tar tf arch.tar --quoting-style=c}
6820 "./a\"double\"quote"
6828 Control characters are printed using backslash notation, a space is
6829 printed as @samp{\ } and a backslash as @samp{\\}. This is the
6830 default quoting style, unless it was changed when configured the
6835 $ @kbd{tar tf arch.tar --quoting-style=escape}
6847 Control characters, single quote and backslash are printed using
6848 backslash notation. All names are quoted using left and right
6849 quotation marks, appropriate to the current locale. If it does not
6850 define quotation marks, use @samp{`} as left and @samp{'} as right
6851 quotation marks. Any occurrences of the right quotation mark in a
6852 name are escaped with @samp{\}, for example:
6858 $ @kbd{tar tf arch.tar --quoting-style=locale}
6861 `./a\'single\'quote'
6870 Same as @samp{locale}, but @samp{"} is used for both left and right
6871 quotation marks, if not provided by the currently selected locale:
6875 $ @kbd{tar tf arch.tar --quoting-style=clocale}
6879 "./a\"double\"quote"
6887 You can specify which characters should be quoted in addition to those
6888 implied by the current quoting style:
6891 @item --quote-chars=@var{string}
6892 Always quote characters from @var{string}, even if the selected
6893 quoting style would not quote them.
6896 For example, using @samp{escape} quoting (compare with the usual
6897 escape listing above):
6901 $ @kbd{tar tf arch.tar --quoting-style=escape --quote-chars=' "'}
6912 To disable quoting of such additional characters, use the following
6916 @item --no-quote-chars=@var{string}
6917 Remove characters listed in @var{string} from the list of quoted
6918 characters set by the previous @option{--quote-chars} option.
6921 This option is particularly useful if you have added
6922 @option{--quote-chars} to your @env{TAR_OPTIONS} (@pxref{TAR_OPTIONS})
6923 and wish to disable it for the current invocation.
6925 Note, that @option{--no-quote-chars} does @emph{not} disable those
6926 characters that are quoted by default in the selected quoting style.
6929 @section Modifying File and Member Names
6931 @command{Tar} archives contain detailed information about files stored
6932 in them and full file names are part of that information. When
6933 storing file to an archive, its file name is recorded in the archive
6934 along with the actual file contents. When restoring from an archive,
6935 a file is created on disk with exactly the same name as that stored
6936 in the archive. In the majority of cases this is the desired behavior
6937 of a file archiver. However, there are some cases when it is not.
6939 First of all, it is often unsafe to extract archive members with
6940 absolute file names or those that begin with a @file{../}. @GNUTAR{}
6941 takes special precautions when extracting such names and provides a
6942 special option for handling them, which is described in
6945 Secondly, you may wish to extract file names without some leading
6946 directory components, or with otherwise modified names. In other
6947 cases it is desirable to store files under differing names in the
6950 @GNUTAR{} provides two options for these needs.
6953 @opindex strip-components
6954 @item --strip-components=@var{number}
6955 Strip given @var{number} of leading components from file names before
6959 For example, suppose you have archived whole @file{/usr} hierarchy to
6960 a tar archive named @file{usr.tar}. Among other files, this archive
6961 contains @file{usr/include/stdlib.h}, which you wish to extract to
6962 the current working directory. To do so, you type:
6965 $ @kbd{tar -xf usr.tar --strip=2 usr/include/stdlib.h}
6968 The option @option{--strip=2} instructs @command{tar} to strip the
6969 two leading components (@file{usr/} and @file{include/}) off the file
6972 If you add to the above invocation @option{--verbose} (@option{-v})
6973 option, you will note that the verbose listing still contains the
6974 full file name, with the two removed components still in place. This
6975 can be inconvenient, so @command{tar} provides a special option for
6976 altering this behavior:
6978 @anchor{show-transformed-names}
6980 @opindex show-transformed-names
6981 @item --show-transformed-names
6982 Display file or member names with all requested transformations
6991 $ @kbd{tar -xf usr.tar -v --strip=2 usr/include/stdlib.h}
6992 usr/include/stdlib.h
6993 $ @kbd{tar -xf usr.tar -v --strip=2 --show-transformed usr/include/stdlib.h}
6998 Notice that in both cases the file is @file{stdlib.h} extracted to the
6999 current working directory, @option{--show-transformed-names} affects
7000 only the way its name is displayed.
7002 This option is especially useful for verifying whether the invocation
7003 will have the desired effect. Thus, before running
7006 $ @kbd{tar -x --strip=@var{n}}
7010 it is often advisable to run
7013 $ @kbd{tar -t -v --show-transformed --strip=@var{n}}
7017 to make sure the command will produce the intended results.
7019 In case you need to apply more complex modifications to the file name,
7020 @GNUTAR{} provides a general-purpose transformation option:
7024 @item --transform=@var{expression}
7025 Modify file names using supplied @var{expression}.
7029 The @var{expression} is a @command{sed}-like replace expression of the
7033 s/@var{regexp}/@var{replace}/[@var{flags}]
7037 where @var{regexp} is a @dfn{regular expression}, @var{replace} is a
7038 replacement for each file name part that matches @var{regexp}. Both
7039 @var{regexp} and @var{replace} are described in detail in
7040 @ref{The "s" Command, The "s" Command, The `s' Command, sed, GNU sed}.
7042 Supported @var{flags} are:
7046 Apply the replacement to @emph{all} matches to the @var{regexp}, not
7050 Use case-insensitive matching
7053 @var{regexp} is an @dfn{extended regular expression} (@pxref{Extended
7054 regexps, Extended regular expressions, Extended regular expressions,
7058 Only replace the @var{number}th match of the @var{regexp}.
7060 Note: the @var{posix} standard does not specify what should happen
7061 when you mix the @samp{g} and @var{number} modifiers. @GNUTAR{}
7062 follows the GNU @command{sed} implementation in this regard, so
7063 the the interaction is defined to be: ignore matches before the
7064 @var{number}th, and then match and replace all matches from the
7069 Any delimiter can be used in lieue of @samp{/}, the only requirement being
7070 that it be used consistently throughout the expression. For example,
7071 the following two expressions are equivalent:
7080 Changing delimiters is often useful when the @var{regex} contains
7081 slashes. For example, it is more convenient to write @code{s,/,-,} than
7084 Here are several examples of @option{--transform} usage:
7087 @item Extract @file{usr/} hierarchy into @file{usr/local/}:
7090 $ @kbd{tar --transform='s,usr/,usr/local/,' -x -f arch.tar}
7093 @item Strip two leading directory components (equivalent to
7094 @option{--strip-components=2}):
7097 $ @kbd{tar --transform='s,/*[^/]*/[^/]*/,,' -x -f arch.tar}
7100 @item Prepend @file{/prefix/} to each file name:
7103 $ @kbd{tar --transform 's,^,/prefix/,' -x -f arch.tar}
7106 @item Convert each file name to lower case:
7109 $ @kbd{tar --transform 's/.*/\L&/' -x -f arch.tar}
7114 Unlike @option{--strip-components}, @option{--transform} can be used
7115 in any @GNUTAR{} operation mode. For example, the following command
7116 adds files to the archive while replacing the leading @file{usr/}
7117 component with @file{var/}:
7120 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' /}
7123 To test @option{--transform} effect we suggest using
7124 @option{--show-transformed-names} option:
7127 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' \
7128 --verbose --show-transformed-names /}
7131 If both @option{--strip-components} and @option{--transform} are used
7132 together, then @option{--transform} is applied first, and the required
7133 number of components is then stripped from its result.
7136 @section Operating Only on New Files
7139 @cindex Excluding file by age
7140 @cindex Data Modification time, excluding files by
7141 @cindex Modification time, excluding files by
7142 @cindex Age, excluding files by
7143 The @option{--after-date=@var{date}} (@option{--newer=@var{date}},
7144 @option{-N @var{date}}) option causes @command{tar} to only work on
7145 files whose data modification or status change times are newer than
7146 the @var{date} given. If @var{date} starts with @samp{/} or @samp{.},
7147 it is taken to be a file name; the data modification time of that file
7148 is used as the date. If you use this option when creating or appending
7149 to an archive, the archive will only include new files. If you use
7150 @option{--after-date} when extracting an archive, @command{tar} will
7151 only extract files newer than the @var{date} you specify.
7153 If you only want @command{tar} to make the date comparison based on
7154 modification of the file's data (rather than status
7155 changes), then use the @option{--newer-mtime=@var{date}} option.
7157 You may use these options with any operation. Note that these options
7158 differ from the @option{--update} (@option{-u}) operation in that they
7159 allow you to specify a particular date against which @command{tar} can
7160 compare when deciding whether or not to archive the files.
7165 @item --after-date=@var{date}
7166 @itemx --newer=@var{date}
7167 @itemx -N @var{date}
7168 Only store files newer than @var{date}.
7170 Acts on files only if their data modification or status change times are
7171 later than @var{date}. Use in conjunction with any operation.
7173 If @var{date} starts with @samp{/} or @samp{.}, it is taken to be a file
7174 name; the data modification time of that file is used as the date.
7176 @opindex newer-mtime
7177 @item --newer-mtime=@var{date}
7178 Acts like @option{--after-date}, but only looks at data modification times.
7181 These options limit @command{tar} to operate only on files which have
7182 been modified after the date specified. A file's status is considered to have
7183 changed if its contents have been modified, or if its owner,
7184 permissions, and so forth, have been changed. (For more information on
7185 how to specify a date, see @ref{Date input formats}; remember that the
7186 entire date argument must be quoted if it contains any spaces.)
7188 Gurus would say that @option{--after-date} tests both the data
7189 modification time (@code{mtime}, the time the contents of the file
7190 were last modified) and the status change time (@code{ctime}, the time
7191 the file's status was last changed: owner, permissions, etc.@:)
7192 fields, while @option{--newer-mtime} tests only the @code{mtime}
7195 To be precise, @option{--after-date} checks @emph{both} @code{mtime} and
7196 @code{ctime} and processes the file if either one is more recent than
7197 @var{date}, while @option{--newer-mtime} only checks @code{mtime} and
7198 disregards @code{ctime}. Neither does it use @code{atime} (the last time the
7199 contents of the file were looked at).
7201 Date specifiers can have embedded spaces. Because of this, you may need
7202 to quote date arguments to keep the shell from parsing them as separate
7203 arguments. For example, the following command will add to the archive
7204 all the files modified less than two days ago:
7207 $ @kbd{tar -cf foo.tar --newer-mtime '2 days ago'}
7211 @strong{Please Note:} @option{--after-date} and @option{--newer-mtime}
7212 should not be used for incremental backups. @xref{Incremental Dumps},
7213 for proper way of creating incremental backups.
7217 @section Descending into Directories
7219 @cindex Avoiding recursion in directories
7220 @cindex Descending directories, avoiding
7221 @cindex Directories, avoiding recursion
7222 @cindex Recursion in directories, avoiding
7224 @FIXME{arrggh! this is still somewhat confusing to me. :-< }
7226 Usually, @command{tar} will recursively explore all directories (either
7227 those given on the command line or through the @option{--files-from}
7228 option) for the various files they contain. However, you may not always
7229 want @command{tar} to act this way.
7231 @opindex no-recursion
7232 The @option{--no-recursion} option inhibits @command{tar}'s recursive descent
7233 into specified directories. If you specify @option{--no-recursion}, you can
7234 use the @command{find} utility for hunting through levels of directories to
7235 construct a list of file names which you could then pass to @command{tar}.
7236 @command{find} allows you to be more selective when choosing which files to
7237 archive; see @ref{files}, for more information on using @command{find} with
7238 @command{tar}, or look.
7241 @item --no-recursion
7242 Prevents @command{tar} from recursively descending directories.
7246 Requires @command{tar} to recursively descend directories.
7247 This is the default.
7250 When you use @option{--no-recursion}, @GNUTAR{} grabs
7251 directory entries themselves, but does not descend on them
7252 recursively. Many people use @command{find} for locating files they
7253 want to back up, and since @command{tar} @emph{usually} recursively
7254 descends on directories, they have to use the @samp{@w{-not -type d}}
7255 test in their @command{find} invocation (@pxref{Type, Type, Type test,
7256 find, Finding Files}), as they usually do not want all the files in a
7257 directory. They then use the @option{--files-from} option to archive
7258 the files located via @command{find}.
7260 The problem when restoring files archived in this manner is that the
7261 directories themselves are not in the archive; so the
7262 @option{--same-permissions} (@option{--preserve-permissions},
7263 @option{-p}) option does not affect them---while users might really
7264 like it to. Specifying @option{--no-recursion} is a way to tell
7265 @command{tar} to grab only the directory entries given to it, adding
7266 no new files on its own. To summarize, if you use @command{find} to
7267 create a list of files to be stored in an archive, use it as follows:
7271 $ @kbd{find @var{dir} @var{tests} | \
7272 tar -cf @var{archive} -T - --no-recursion}
7276 The @option{--no-recursion} option also applies when extracting: it
7277 causes @command{tar} to extract only the matched directory entries, not
7278 the files under those directories.
7280 The @option{--no-recursion} option also affects how globbing patterns
7281 are interpreted (@pxref{controlling pattern-matching}).
7283 The @option{--no-recursion} and @option{--recursion} options apply to
7284 later options and operands, and can be overridden by later occurrences
7285 of @option{--no-recursion} and @option{--recursion}. For example:
7288 $ @kbd{tar -cf jams.tar --no-recursion grape --recursion grape/concord}
7292 creates an archive with one entry for @file{grape}, and the recursive
7293 contents of @file{grape/concord}, but no entries under @file{grape}
7294 other than @file{grape/concord}.
7297 @section Crossing File System Boundaries
7298 @cindex File system boundaries, not crossing
7301 @command{tar} will normally automatically cross file system boundaries in
7302 order to archive files which are part of a directory tree. You can
7303 change this behavior by running @command{tar} and specifying
7304 @option{--one-file-system}. This option only affects files that are
7305 archived because they are in a directory that is being archived;
7306 @command{tar} will still archive files explicitly named on the command line
7307 or through @option{--files-from}, regardless of where they reside.
7310 @opindex one-file-system
7311 @item --one-file-system
7312 Prevents @command{tar} from crossing file system boundaries when
7313 archiving. Use in conjunction with any write operation.
7316 The @option{--one-file-system} option causes @command{tar} to modify its
7317 normal behavior in archiving the contents of directories. If a file in
7318 a directory is not on the same file system as the directory itself, then
7319 @command{tar} will not archive that file. If the file is a directory
7320 itself, @command{tar} will not archive anything beneath it; in other words,
7321 @command{tar} will not cross mount points.
7323 This option is useful for making full or incremental archival backups of
7324 a file system. If this option is used in conjunction with
7325 @option{--verbose} (@option{-v}), files that are excluded are
7326 mentioned by name on the standard error.
7329 * directory:: Changing Directory
7330 * absolute:: Absolute File Names
7334 @subsection Changing the Working Directory
7337 @FIXME{need to read over this node now for continuity; i've switched
7338 things around some.}
7340 @cindex Changing directory mid-stream
7341 @cindex Directory, changing mid-stream
7342 @cindex Working directory, specifying
7343 To change the working directory in the middle of a list of file names,
7344 either on the command line or in a file specified using
7345 @option{--files-from} (@option{-T}), use @option{--directory} (@option{-C}).
7346 This will change the working directory to the specified directory
7347 after that point in the list.
7351 @item --directory=@var{directory}
7352 @itemx -C @var{directory}
7353 Changes the working directory in the middle of a command line.
7359 $ @kbd{tar -c -f jams.tar grape prune -C food cherry}
7363 will place the files @file{grape} and @file{prune} from the current
7364 directory into the archive @file{jams.tar}, followed by the file
7365 @file{cherry} from the directory @file{food}. This option is especially
7366 useful when you have several widely separated files that you want to
7367 store in the same archive.
7369 Note that the file @file{cherry} is recorded in the archive under the
7370 precise name @file{cherry}, @emph{not} @file{food/cherry}. Thus, the
7371 archive will contain three files that all appear to have come from the
7372 same directory; if the archive is extracted with plain @samp{tar
7373 --extract}, all three files will be written in the current directory.
7375 Contrast this with the command,
7378 $ @kbd{tar -c -f jams.tar grape prune -C food red/cherry}
7382 which records the third file in the archive under the name
7383 @file{red/cherry} so that, if the archive is extracted using
7384 @samp{tar --extract}, the third file will be written in a subdirectory
7385 named @file{orange-colored}.
7387 You can use the @option{--directory} option to make the archive
7388 independent of the original name of the directory holding the files.
7389 The following command places the files @file{/etc/passwd},
7390 @file{/etc/hosts}, and @file{/lib/libc.a} into the archive
7394 $ @kbd{tar -c -f foo.tar -C /etc passwd hosts -C /lib libc.a}
7398 However, the names of the archive members will be exactly what they were
7399 on the command line: @file{passwd}, @file{hosts}, and @file{libc.a}.
7400 They will not appear to be related by file name to the original
7401 directories where those files were located.
7403 Note that @option{--directory} options are interpreted consecutively. If
7404 @option{--directory} specifies a relative file name, it is interpreted
7405 relative to the then current directory, which might not be the same as
7406 the original current working directory of @command{tar}, due to a previous
7407 @option{--directory} option.
7409 When using @option{--files-from} (@pxref{files}), you can put various
7410 @command{tar} options (including @option{-C}) in the file list. Notice,
7411 however, that in this case the option and its argument may not be
7412 separated by whitespace. If you use short option, its argument must
7413 either follow the option letter immediately, without any intervening
7414 whitespace, or occupy the next line. Otherwise, if you use long
7415 option, separate its argument by an equal sign.
7417 For instance, the file list for the above example will be:
7432 To use it, you would invoke @command{tar} as follows:
7435 $ @kbd{tar -c -f foo.tar --files-from list}
7438 Notice also that you can only use the short option variant in the file
7439 list, i.e., always use @option{-C}, not @option{--directory}.
7441 The interpretation of @option{--directory} is disabled by
7442 @option{--null} option.
7445 @subsection Absolute File Names
7449 @opindex absolute-names
7450 @item --absolute-names
7452 Do not strip leading slashes from file names, and permit file names
7453 containing a @file{..} file name component.
7456 By default, @GNUTAR{} drops a leading @samp{/} on
7457 input or output, and complains about file names containing a @file{..}
7458 component. This option turns off this behavior.
7460 When @command{tar} extracts archive members from an archive, it strips any
7461 leading slashes (@samp{/}) from the member name. This causes absolute
7462 member names in the archive to be treated as relative file names. This
7463 allows you to have such members extracted wherever you want, instead of
7464 being restricted to extracting the member in the exact directory named
7465 in the archive. For example, if the archive member has the name
7466 @file{/etc/passwd}, @command{tar} will extract it as if the name were
7467 really @file{etc/passwd}.
7469 File names containing @file{..} can cause problems when extracting, so
7470 @command{tar} normally warns you about such files when creating an
7471 archive, and rejects attempts to extracts such files.
7473 Other @command{tar} programs do not do this. As a result, if you
7474 create an archive whose member names start with a slash, they will be
7475 difficult for other people with a non-@GNUTAR{}
7476 program to use. Therefore, @GNUTAR{} also strips
7477 leading slashes from member names when putting members into the
7478 archive. For example, if you ask @command{tar} to add the file
7479 @file{/bin/ls} to an archive, it will do so, but the member name will
7480 be @file{bin/ls}.@footnote{A side effect of this is that when
7481 @option{--create} is used with @option{--verbose} the resulting output
7482 is not, generally speaking, the same as the one you'd get running
7483 @kbd{tar --list} command. This may be important if you use some
7484 scripts for comparing both outputs. @xref{listing member and file names},
7485 for the information on how to handle this case.}
7487 If you use the @option{--absolute-names} (@option{-P}) option,
7488 @command{tar} will do none of these transformations.
7490 To archive or extract files relative to the root directory, specify
7491 the @option{--absolute-names} (@option{-P}) option.
7493 Normally, @command{tar} acts on files relative to the working
7494 directory---ignoring superior directory names when archiving, and
7495 ignoring leading slashes when extracting.
7497 When you specify @option{--absolute-names} (@option{-P}),
7498 @command{tar} stores file names including all superior directory
7499 names, and preserves leading slashes. If you only invoked
7500 @command{tar} from the root directory you would never need the
7501 @option{--absolute-names} option, but using this option
7502 may be more convenient than switching to root.
7504 @FIXME{Should be an example in the tutorial/wizardry section using this
7505 to transfer files between systems.}
7507 @FIXME{Is write access an issue?}
7510 @item --absolute-names
7511 Preserves full file names (including superior directory names) when
7512 archiving files. Preserves leading slash when extracting files.
7516 @FIXME{this is still horrible; need to talk with dan on monday.}
7518 @command{tar} prints out a message about removing the @samp{/} from
7519 file names. This message appears once per @GNUTAR{}
7520 invocation. It represents something which ought to be told; ignoring
7521 what it means can cause very serious surprises, later.
7523 Some people, nevertheless, do not want to see this message. Wanting to
7524 play really dangerously, one may of course redirect @command{tar} standard
7525 error to the sink. For example, under @command{sh}:
7528 $ @kbd{tar -c -f archive.tar /home 2> /dev/null}
7532 Another solution, both nicer and simpler, would be to change to
7533 the @file{/} directory first, and then avoid absolute notation.
7537 $ @kbd{(cd / && tar -c -f archive.tar home)}
7539 $ @kbd{tar -c -f archive.tar -C / home}
7542 @include getdate.texi
7545 @chapter Controlling the Archive Format
7547 @cindex Tar archive formats
7548 Due to historical reasons, there are several formats of tar archives.
7549 All of them are based on the same principles, but have some subtle
7550 differences that often make them incompatible with each other.
7552 GNU tar is able to create and handle archives in a variety of formats.
7553 The most frequently used formats are (in alphabetical order):
7557 Format used by @GNUTAR{} versions up to 1.13.25. This format derived
7558 from an early @acronym{POSIX} standard, adding some improvements such as
7559 sparse file handling and incremental archives. Unfortunately these
7560 features were implemented in a way incompatible with other archive
7563 Archives in @samp{gnu} format are able to hold pathnames of unlimited
7567 Format used by @GNUTAR{} of versions prior to 1.12.
7570 Archive format, compatible with the V7 implementation of tar. This
7571 format imposes a number of limitations. The most important of them
7575 @item The maximum length of a file name is limited to 99 characters.
7576 @item The maximum length of a symbolic link is limited to 99 characters.
7577 @item It is impossible to store special files (block and character
7578 devices, fifos etc.)
7579 @item Maximum value of user or group ID is limited to 2097151 (7777777
7581 @item V7 archives do not contain symbolic ownership information (user
7582 and group name of the file owner).
7585 This format has traditionally been used by Automake when producing
7586 Makefiles. This practice will change in the future, in the meantime,
7587 however this means that projects containing filenames more than 99
7588 characters long will not be able to use @GNUTAR{} @value{VERSION} and
7589 Automake prior to 1.9.
7592 Archive format defined by @acronym{POSIX.1-1988} specification. It stores
7593 symbolic ownership information. It is also able to store
7594 special files. However, it imposes several restrictions as well:
7597 @item The maximum length of a file name is limited to 256 characters,
7598 provided that the filename can be split at directory separator in
7599 two parts, first of them being at most 155 bytes long. So, in most
7600 cases the maximum file name length will be shorter than 256
7602 @item The maximum length of a symbolic link name is limited to
7604 @item Maximum size of a file the archive is able to accomodate
7606 @item Maximum value of UID/GID is 2097151.
7607 @item Maximum number of bits in device major and minor numbers is 21.
7611 Format used by J@"org Schilling @command{star}
7612 implementation. @GNUTAR{} is able to read @samp{star} archives but
7613 currently does not produce them.
7616 Archive format defined by @acronym{POSIX.1-2001} specification. This is the
7617 most flexible and feature-rich format. It does not impose any
7618 restrictions on file sizes or filename lengths. This format is quite
7619 recent, so not all tar implementations are able to handle it properly.
7620 However, this format is designed in such a way that any tar
7621 implementation able to read @samp{ustar} archives will be able to read
7622 most @samp{posix} archives as well, with the only exception that any
7623 additional information (such as long file names etc.) will in such
7624 case be extracted as plain text files along with the files it refers to.
7626 This archive format will be the default format for future versions
7631 The following table summarizes the limitations of each of these
7634 @multitable @columnfractions .10 .20 .20 .20 .20
7635 @headitem Format @tab UID @tab File Size @tab Path Name @tab Devn
7636 @item gnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
7637 @item oldgnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
7638 @item v7 @tab 2097151 @tab 8GB @tab 99 @tab n/a
7639 @item ustar @tab 2097151 @tab 8GB @tab 256 @tab 21
7640 @item posix @tab Unlimited @tab Unlimited @tab Unlimited @tab Unlimited
7643 The default format for @GNUTAR{} is defined at compilation
7644 time. You may check it by running @command{tar --help}, and examining
7645 the last lines of its output. Usually, @GNUTAR{} is configured
7646 to create archives in @samp{gnu} format, however, future version will
7647 switch to @samp{posix}.
7650 * Portability:: Making @command{tar} Archives More Portable
7651 * Compression:: Using Less Space through Compression
7652 * Attributes:: Handling File Attributes
7653 * cpio:: Comparison of @command{tar} and @command{cpio}
7657 @section Making @command{tar} Archives More Portable
7659 Creating a @command{tar} archive on a particular system that is meant to be
7660 useful later on many other machines and with other versions of @command{tar}
7661 is more challenging than you might think. @command{tar} archive formats
7662 have been evolving since the first versions of Unix. Many such formats
7663 are around, and are not always compatible with each other. This section
7664 discusses a few problems, and gives some advice about making @command{tar}
7665 archives more portable.
7667 One golden rule is simplicity. For example, limit your @command{tar}
7668 archives to contain only regular files and directories, avoiding
7669 other kind of special files. Do not attempt to save sparse files or
7670 contiguous files as such. Let's discuss a few more problems, in turn.
7672 @FIXME{Discuss GNU extensions (incremental backups, multi-volume
7673 archives and archive labels) in GNU and PAX formats.}
7676 * Portable Names:: Portable Names
7677 * dereference:: Symbolic Links
7678 * old:: Old V7 Archives
7679 * ustar:: Ustar Archives
7680 * gnu:: GNU and old GNU format archives.
7681 * posix:: @acronym{POSIX} archives
7682 * Checksumming:: Checksumming Problems
7683 * Large or Negative Values:: Large files, negative time stamps, etc.
7686 @node Portable Names
7687 @subsection Portable Names
7689 Use portable file and member names. A name is portable if it contains
7690 only ASCII letters and digits, @samp{/}, @samp{.}, @samp{_}, and
7691 @samp{-}; it cannot be empty, start with @samp{-} or @samp{//}, or
7692 contain @samp{/-}. Avoid deep directory nesting. For portability to
7693 old Unix hosts, limit your file name components to 14 characters or
7696 If you intend to have your @command{tar} archives to be read under
7697 MSDOS, you should not rely on case distinction for file names, and you
7698 might use the @acronym{GNU} @command{doschk} program for helping you
7699 further diagnosing illegal MSDOS names, which are even more limited
7703 @subsection Symbolic Links
7704 @cindex File names, using symbolic links
7705 @cindex Symbolic link as file name
7707 @opindex dereference
7708 Normally, when @command{tar} archives a symbolic link, it writes a
7709 block to the archive naming the target of the link. In that way, the
7710 @command{tar} archive is a faithful record of the file system contents.
7711 @option{--dereference} (@option{-h}) is used with @option{--create} (@option{-c}), and causes
7712 @command{tar} to archive the files symbolic links point to, instead of
7713 the links themselves. When this option is used, when @command{tar}
7714 encounters a symbolic link, it will archive the linked-to file,
7715 instead of simply recording the presence of a symbolic link.
7717 The name under which the file is stored in the file system is not
7718 recorded in the archive. To record both the symbolic link name and
7719 the file name in the system, archive the file under both names. If
7720 all links were recorded automatically by @command{tar}, an extracted file
7721 might be linked to a file name that no longer exists in the file
7724 If a linked-to file is encountered again by @command{tar} while creating
7725 the same archive, an entire second copy of it will be stored. (This
7726 @emph{might} be considered a bug.)
7728 So, for portable archives, do not archive symbolic links as such,
7729 and use @option{--dereference} (@option{-h}): many systems do not support
7730 symbolic links, and moreover, your distribution might be unusable if
7731 it contains unresolved symbolic links.
7734 @subsection Old V7 Archives
7735 @cindex Format, old style
7736 @cindex Old style format
7737 @cindex Old style archives
7738 @cindex v7 archive format
7740 Certain old versions of @command{tar} cannot handle additional
7741 information recorded by newer @command{tar} programs. To create an
7742 archive in V7 format (not ANSI), which can be read by these old
7743 versions, specify the @option{--format=v7} option in
7744 conjunction with the @option{--create} (@option{-c}) (@command{tar} also
7745 accepts @option{--portability} or @samp{op-old-archive} for this
7746 option). When you specify it,
7747 @command{tar} leaves out information about directories, pipes, fifos,
7748 contiguous files, and device files, and specifies file ownership by
7749 group and user IDs instead of group and user names.
7751 When updating an archive, do not use @option{--format=v7}
7752 unless the archive was created using this option.
7754 In most cases, a @emph{new} format archive can be read by an @emph{old}
7755 @command{tar} program without serious trouble, so this option should
7756 seldom be needed. On the other hand, most modern @command{tar}s are
7757 able to read old format archives, so it might be safer for you to
7758 always use @option{--format=v7} for your distributions.
7761 @subsection Ustar Archive Format
7763 @cindex ustar archive format
7764 Archive format defined by @acronym{POSIX}.1-1988 specification is called
7765 @code{ustar}. Although it is more flexible than the V7 format, it
7766 still has many restrictions (@xref{Formats,ustar}, for the detailed
7767 description of @code{ustar} format). Along with V7 format,
7768 @code{ustar} format is a good choice for archives intended to be read
7769 with other implementations of @command{tar}.
7771 To create archive in @code{ustar} format, use @option{--format=ustar}
7772 option in conjunction with the @option{--create} (@option{-c}).
7775 @subsection @acronym{GNU} and old @GNUTAR{} format
7777 @cindex GNU archive format
7778 @cindex Old GNU archive format
7779 @GNUTAR{} was based on an early draft of the
7780 @acronym{POSIX} 1003.1 @code{ustar} standard. @acronym{GNU} extensions to
7781 @command{tar}, such as the support for file names longer than 100
7782 characters, use portions of the @command{tar} header record which were
7783 specified in that @acronym{POSIX} draft as unused. Subsequent changes in
7784 @acronym{POSIX} have allocated the same parts of the header record for
7785 other purposes. As a result, @GNUTAR{} format is
7786 incompatible with the current @acronym{POSIX} specification, and with
7787 @command{tar} programs that follow it.
7789 In the majority of cases, @command{tar} will be configured to create
7790 this format by default. This will change in the future releases, since
7791 we plan to make @samp{posix} format the default.
7793 To force creation a @GNUTAR{} archive, use option
7794 @option{--format=gnu}.
7797 @subsection @GNUTAR{} and @acronym{POSIX} @command{tar}
7799 @cindex POSIX archive format
7800 @cindex PAX archive format
7801 Starting from version 1.14 @GNUTAR{} features full support for
7802 @acronym{POSIX.1-2001} archives.
7804 A @acronym{POSIX} conformant archive will be created if @command{tar}
7805 was given @option{--format=posix} (@option{--format=pax}) option. No
7806 special option is required to read and extract from a @acronym{POSIX}
7810 * PAX keywords:: Controlling Extended Header Keywords.
7814 @subsubsection Controlling Extended Header Keywords
7818 @item --pax-option=@var{keyword-list}
7819 Handle keywords in @acronym{PAX} extended headers. This option is
7820 equivalent to @option{-o} option of the @command{pax} utility.
7823 @var{Keyword-list} is a comma-separated
7824 list of keyword options, each keyword option taking one of
7825 the following forms:
7828 @item delete=@var{pattern}
7829 When used with one of archive-creation commands,
7830 this option instructs @command{tar} to omit from extended header records
7831 that it produces any keywords matching the string @var{pattern}.
7833 When used in extract or list mode, this option instructs tar
7834 to ignore any keywords matching the given @var{pattern} in the extended
7835 header records. In both cases, matching is performed using the pattern
7836 matching notation described in @acronym{POSIX 1003.2}, 3.13
7837 (@pxref{wildcards}). For example:
7840 --pax-option delete=security.*
7843 would suppress security-related information.
7845 @item exthdr.name=@var{string}
7847 This keyword allows user control over the name that is written into the
7848 ustar header blocks for the extended headers. The name is obtained
7849 from @var{string} after making the following substitutions:
7851 @multitable @columnfractions .25 .55
7852 @headitem Meta-character @tab Replaced By
7853 @item %d @tab The directory name of the file, equivalent to the
7854 result of the @command{dirname} utility on the translated pathname.
7855 @item %f @tab The filename of the file, equivalent to the result
7856 of the @command{basename} utility on the translated pathname.
7857 @item %p @tab The process ID of the @command{tar} process.
7858 @item %% @tab A @samp{%} character.
7861 Any other @samp{%} characters in @var{string} produce undefined
7864 If no option @samp{exthdr.name=string} is specified, @command{tar}
7865 will use the following default value:
7871 @item globexthdr.name=@var{string}
7872 This keyword allows user control over the name that is written into
7873 the ustar header blocks for global extended header records. The name
7874 is obtained from the contents of @var{string}, after making
7875 the following substitutions:
7877 @multitable @columnfractions .25 .55
7878 @headitem Meta-character @tab Replaced By
7879 @item %n @tab An integer that represents the
7880 sequence number of the global extended header record in the archive,
7882 @item %p @tab The process ID of the @command{tar} process.
7883 @item %% @tab A @samp{%} character.
7886 Any other @samp{%} characters in @var{string} produce undefined results.
7888 If no option @samp{globexthdr.name=string} is specified, @command{tar}
7889 will use the following default value:
7892 $TMPDIR/GlobalHead.%p.%n
7896 where @samp{$TMPDIR} represents the value of the @var{TMPDIR}
7897 environment variable. If @var{TMPDIR} is not set, @command{tar}
7900 @item @var{keyword}=@var{value}
7901 When used with one of archive-creation commands, these keyword/value pairs
7902 will be included at the beginning of the archive in a global extended
7903 header record. When used with one of archive-reading commands,
7904 @command{tar} will behave as if it has encountered these keyword/value
7905 pairs at the beginning of the archive in a global extended header
7908 @item @var{keyword}:=@var{value}
7909 When used with one of archive-creation commands, these keyword/value pairs
7910 will be included as records at the beginning of an extended header for
7911 each file. This is effectively equivalent to @var{keyword}=@var{value}
7912 form except that it creates no global extended header records.
7914 When used with one of archive-reading commands, @command{tar} will
7915 behave as if these keyword/value pairs were included as records at the
7916 end of each extended header; thus, they will override any global or
7917 file-specific extended header record keywords of the same names.
7918 For example, in the command:
7921 tar --format=posix --create \
7922 --file archive --pax-option gname:=user .
7925 the group name will be forced to a new value for all files
7926 stored in the archive.
7930 @subsection Checksumming Problems
7932 SunOS and HP-UX @command{tar} fail to accept archives created using
7933 @GNUTAR{} and containing non-ASCII file names, that
7934 is, file names having characters with the eight bit set, because they
7935 use signed checksums, while @GNUTAR{} uses unsigned
7936 checksums while creating archives, as per @acronym{POSIX} standards. On
7937 reading, @GNUTAR{} computes both checksums and
7938 accept any. It is somewhat worrying that a lot of people may go
7939 around doing backup of their files using faulty (or at least
7940 non-standard) software, not learning about it until it's time to
7941 restore their missing files with an incompatible file extractor, or
7944 @GNUTAR{} compute checksums both ways, and accept
7945 any on read, so @acronym{GNU} tar can read Sun tapes even with their
7946 wrong checksums. @GNUTAR{} produces the standard
7947 checksum, however, raising incompatibilities with Sun. That is to
7948 say, @GNUTAR{} has not been modified to
7949 @emph{produce} incorrect archives to be read by buggy @command{tar}'s.
7950 I've been told that more recent Sun @command{tar} now read standard
7951 archives, so maybe Sun did a similar patch, after all?
7953 The story seems to be that when Sun first imported @command{tar}
7954 sources on their system, they recompiled it without realizing that
7955 the checksums were computed differently, because of a change in
7956 the default signing of @code{char}'s in their compiler. So they
7957 started computing checksums wrongly. When they later realized their
7958 mistake, they merely decided to stay compatible with it, and with
7959 themselves afterwards. Presumably, but I do not really know, HP-UX
7960 has chosen that their @command{tar} archives to be compatible with Sun's.
7961 The current standards do not favor Sun @command{tar} format. In any
7962 case, it now falls on the shoulders of SunOS and HP-UX users to get
7963 a @command{tar} able to read the good archives they receive.
7965 @node Large or Negative Values
7966 @subsection Large or Negative Values
7967 @cindex large values
7968 @cindex future time stamps
7969 @cindex negative time stamps
7972 The above sections suggest to use @samp{oldest possible} archive
7973 format if in doubt. However, sometimes it is not possible. If you
7974 attempt to archive a file whose metadata cannot be represented using
7975 required format, @GNUTAR{} will print error message and ignore such a
7976 file. You will than have to switch to a format that is able to
7977 handle such values. The format summary table (@pxref{Formats}) will
7980 In particular, when trying to archive files larger than 8GB or with
7981 timestamps not in the range 1970-01-01 00:00:00 through 2242-03-16
7982 12:56:31 @sc{utc}, you will have to chose between @acronym{GNU} and
7983 @acronym{POSIX} archive formats. When considering which format to
7984 choose, bear in mind that the @acronym{GNU} format uses
7985 two's-complement base-256 notation to store values that do not fit
7986 into standard @acronym{ustar} range. Such archives can generally be
7987 read only by a @GNUTAR{} implementation. Moreover, they sometimes
7988 cannot be correctly restored on another hosts even by @GNUTAR{}. For
7989 example, using two's complement representation for negative time
7990 stamps that assumes a signed 32-bit @code{time_t} generates archives
7991 that are not portable to hosts with differing @code{time_t}
7994 On the other hand, @acronym{POSIX} archives, generally speaking, can
7995 be extracted by any tar implementation that understands older
7996 @acronym{ustar} format. The only exception are files larger than 8GB.
7998 @FIXME{Describe how @acronym{POSIX} archives are extracted by non
8002 @section Using Less Space through Compression
8005 * gzip:: Creating and Reading Compressed Archives
8006 * sparse:: Archiving Sparse Files
8010 @subsection Creating and Reading Compressed Archives
8011 @cindex Compressed archives
8012 @cindex Storing archives in compressed format
8014 @GNUTAR{} is able to create and read compressed archives. It supports
8015 @command{gzip} and @command{bzip2} compression programs. For backward
8016 compatibilty, it also supports @command{compress} command, although
8017 we strongly recommend against using it, since there is a patent
8018 covering the algorithm it uses and you could be sued for patent
8019 infringement merely by running @command{compress}! Besides, it is less
8020 effective than @command{gzip} and @command{bzip2}.
8022 Creating a compressed archive is simple: you just specify a
8023 @dfn{compression option} along with the usual archive creation
8024 commands. The compression option is @option{-z} (@option{--gzip}) to
8025 create a @command{gzip} compressed archive, @option{-j}
8026 (@option{--bzip2}) to create a @command{bzip2} compressed archive, and
8027 @option{-Z} (@option{--compress}) to use @command{compress} program.
8031 $ @kbd{tar cfz archive.tar.gz .}
8034 Reading compressed archive is even simpler: you don't need to specify
8035 any additional options as @GNUTAR{} recognizes its format
8036 automatically. Thus, the following commands will list and extract the
8037 archive created in previous example:
8040 # List the compressed archive
8041 $ @kbd{tar tf archive.tar.gz}
8042 # Extract the compressed archive
8043 $ @kbd{tar xf archive.tar.gz}
8046 The only case when you have to specify a decompression option while
8047 reading the archive is when reading from a pipe or from a tape drive
8048 that does not support random access. However, in this case @GNUTAR{}
8049 will indicate which option you should use. For example:
8052 $ @kbd{cat archive.tar.gz | tar tf -}
8053 tar: Archive is compressed. Use -z option
8054 tar: Error is not recoverable: exiting now
8057 If you see such diagnostics, just add the suggested option to the
8058 invocation of @GNUTAR{}:
8061 $ @kbd{cat archive.tar.gz | tar tfz -}
8064 Notice also, that there are several restrictions on operations on
8065 compressed archives. First of all, compressed archives cannot be
8066 modified, i.e., you cannot update (@option{--update} (@option{-u})) them or delete
8067 (@option{--delete}) members from them. Likewise, you cannot append
8068 another @command{tar} archive to a compressed archive using
8069 @option{--append} (@option{-r})). Secondly, multi-volume archives cannot be
8072 The following table summarizes compression options used by @GNUTAR{}.
8080 Filter the archive through @command{gzip}.
8082 You can use @option{--gzip} and @option{--gunzip} on physical devices
8083 (tape drives, etc.) and remote files as well as on normal files; data
8084 to or from such devices or remote files is reblocked by another copy
8085 of the @command{tar} program to enforce the specified (or default) record
8086 size. The default compression parameters are used; if you need to
8087 override them, set @env{GZIP} environment variable, e.g.:
8090 $ @kbd{GZIP=--best tar cfz archive.tar.gz subdir}
8094 Another way would be to avoid the @option{--gzip} (@option{--gunzip}, @option{--ungzip}, @option{-z}) option and run
8095 @command{gzip} explicitly:
8098 $ @kbd{tar cf - subdir | gzip --best -c - > archive.tar.gz}
8101 @cindex corrupted archives
8102 About corrupted compressed archives: @command{gzip}'ed files have no
8103 redundancy, for maximum compression. The adaptive nature of the
8104 compression scheme means that the compression tables are implicitly
8105 spread all over the archive. If you lose a few blocks, the dynamic
8106 construction of the compression tables becomes unsynchronized, and there
8107 is little chance that you could recover later in the archive.
8109 There are pending suggestions for having a per-volume or per-file
8110 compression in @GNUTAR{}. This would allow for viewing the
8111 contents without decompression, and for resynchronizing decompression at
8112 every volume or file, in case of corrupted archives. Doing so, we might
8113 lose some compressibility. But this would have make recovering easier.
8114 So, there are pros and cons. We'll see!
8119 Filter the archive through @code{bzip2}. Otherwise like @option{--gzip}.
8126 Filter the archive through @command{compress}. Otherwise like @option{--gzip}.
8128 The @acronym{GNU} Project recommends you not use
8129 @command{compress}, because there is a patent covering the algorithm it
8130 uses. You could be sued for patent infringement merely by running
8133 @opindex use-compress-program
8134 @item --use-compress-program=@var{prog}
8135 Use external compression program @var{prog}. Use this option if you
8136 have a compression program that @GNUTAR{} does not support. There
8137 are two requirements to which @var{prog} should comply:
8139 First, when called without options, it should read data from standard
8140 input, compress it and output it on standard output.
8142 Secondly, if called with @option{-d} argument, it should do exactly
8143 the opposite, i.e., read the compressed data from the standard input
8144 and produce uncompressed data on the standard output.
8147 @cindex gpg, using with tar
8148 @cindex gnupg, using with tar
8149 @cindex Using encrypted archives
8150 The @option{--use-compress-program} option, in particular, lets you
8151 implement your own filters, not necessarily dealing with
8152 compression/decomression. For example, suppose you wish to implement
8153 PGP encryption on top of compression, using @command{gpg} (@pxref{Top,
8154 gpg, gpg ---- encryption and signing tool, gpg, GNU Privacy Guard
8155 Manual}). The following script does that:
8161 -d) gpg --decrypt - | gzip -d -c;;
8162 '') gzip -c | gpg -s ;;
8163 *) echo "Unknown option $1">&2; exit 1;;
8168 Suppose you name it @file{gpgz} and save it somewhere in your
8169 @env{PATH}. Then the following command will create a commpressed
8170 archive signed with your private key:
8173 $ @kbd{tar -cf foo.tar.gpgz --use-compress=gpgz .}
8177 Likewise, the following command will list its contents:
8180 $ @kbd{tar -tf foo.tar.gpgz --use-compress=gpgz .}
8184 The above is based on the following discussion:
8186 I have one question, or maybe it's a suggestion if there isn't a way
8187 to do it now. I would like to use @option{--gzip}, but I'd also like
8188 the output to be fed through a program like @acronym{GNU}
8189 @command{ecc} (actually, right now that's @samp{exactly} what I'd like
8190 to use :-)), basically adding ECC protection on top of compression.
8191 It seems as if this should be quite easy to do, but I can't work out
8192 exactly how to go about it. Of course, I can pipe the standard output
8193 of @command{tar} through @command{ecc}, but then I lose (though I
8194 haven't started using it yet, I confess) the ability to have
8195 @command{tar} use @command{rmt} for it's I/O (I think).
8197 I think the most straightforward thing would be to let me specify a
8198 general set of filters outboard of compression (preferably ordered,
8199 so the order can be automatically reversed on input operations, and
8200 with the options they require specifiable), but beggars shouldn't be
8201 choosers and anything you decide on would be fine with me.
8203 By the way, I like @command{ecc} but if (as the comments say) it can't
8204 deal with loss of block sync, I'm tempted to throw some time at adding
8205 that capability. Supposing I were to actually do such a thing and
8206 get it (apparently) working, do you accept contributed changes to
8207 utilities like that? (Leigh Clayton @file{loc@@soliton.com}, May 1995).
8209 Isn't that exactly the role of the
8210 @option{--use-compress-prog=@var{program}} option?
8211 I never tried it myself, but I suspect you may want to write a
8212 @var{prog} script or program able to filter stdin to stdout to
8213 way you want. It should recognize the @option{-d} option, for when
8214 extraction is needed rather than creation.
8216 It has been reported that if one writes compressed data (through the
8217 @option{--gzip} or @option{--compress} options) to a DLT and tries to use
8218 the DLT compression mode, the data will actually get bigger and one will
8219 end up with less space on the tape.
8223 @subsection Archiving Sparse Files
8224 @cindex Sparse Files
8231 Handle sparse files efficiently.
8234 This option causes all files to be put in the archive to be tested for
8235 sparseness, and handled specially if they are. The @option{--sparse}
8236 (@option{-S}) option is useful when many @code{dbm} files, for example, are being
8237 backed up. Using this option dramatically decreases the amount of
8238 space needed to store such a file.
8240 In later versions, this option may be removed, and the testing and
8241 treatment of sparse files may be done automatically with any special
8242 @acronym{GNU} options. For now, it is an option needing to be specified on
8243 the command line with the creation or updating of an archive.
8245 Files in the file system occasionally have ``holes.'' A hole in a file
8246 is a section of the file's contents which was never written. The
8247 contents of a hole read as all zeros. On many operating systems,
8248 actual disk storage is not allocated for holes, but they are counted
8249 in the length of the file. If you archive such a file, @command{tar}
8250 could create an archive longer than the original. To have @command{tar}
8251 attempt to recognize the holes in a file, use @option{--sparse} (@option{-S}). When
8252 you use this option, then, for any file using less disk space than
8253 would be expected from its length, @command{tar} searches the file for
8254 consecutive stretches of zeros. It then records in the archive for
8255 the file where the consecutive stretches of zeros are, and only
8256 archives the ``real contents'' of the file. On extraction (using
8257 @option{--sparse} is not needed on extraction) any such
8258 files have holes created wherever the continuous stretches of zeros
8259 were found. Thus, if you use @option{--sparse}, @command{tar} archives
8260 won't take more space than the original.
8262 A file is sparse if it contains blocks of zeros whose existence is
8263 recorded, but that have no space allocated on disk. When you specify
8264 the @option{--sparse} option in conjunction with the @option{--create}
8265 (@option{-c}) operation, @command{tar} tests all files for sparseness
8266 while archiving. If @command{tar} finds a file to be sparse, it uses a
8267 sparse representation of the file in the archive. @xref{create}, for
8268 more information about creating archives.
8270 @option{--sparse} is useful when archiving files, such as dbm files,
8271 likely to contain many nulls. This option dramatically
8272 decreases the amount of space needed to store such an archive.
8275 @strong{Please Note:} Always use @option{--sparse} when performing file
8276 system backups, to avoid archiving the expanded forms of files stored
8277 sparsely in the system.
8279 Even if your system has no sparse files currently, some may be
8280 created in the future. If you use @option{--sparse} while making file
8281 system backups as a matter of course, you can be assured the archive
8282 will never take more space on the media than the files take on disk
8283 (otherwise, archiving a disk filled with sparse files might take
8284 hundreds of tapes). @xref{Incremental Dumps}.
8287 @command{tar} ignores the @option{--sparse} option when reading an archive.
8292 Files stored sparsely in the file system are represented sparsely in
8293 the archive. Use in conjunction with write operations.
8296 However, users should be well aware that at archive creation time,
8297 @GNUTAR{} still has to read whole disk file to
8298 locate the @dfn{holes}, and so, even if sparse files use little space
8299 on disk and in the archive, they may sometimes require inordinate
8300 amount of time for reading and examining all-zero blocks of a file.
8301 Although it works, it's painfully slow for a large (sparse) file, even
8302 though the resulting tar archive may be small. (One user reports that
8303 dumping a @file{core} file of over 400 megabytes, but with only about
8304 3 megabytes of actual data, took about 9 minutes on a Sun Sparcstation
8305 ELC, with full CPU utilization.)
8307 This reading is required in all cases and is not related to the fact
8308 the @option{--sparse} option is used or not, so by merely @emph{not}
8309 using the option, you are not saving time@footnote{Well! We should say
8310 the whole truth, here. When @option{--sparse} is selected while creating
8311 an archive, the current @command{tar} algorithm requires sparse files to be
8312 read twice, not once. We hope to develop a new archive format for saving
8313 sparse files in which one pass will be sufficient.}.
8315 Programs like @command{dump} do not have to read the entire file; by
8316 examining the file system directly, they can determine in advance
8317 exactly where the holes are and thus avoid reading through them. The
8318 only data it need read are the actual allocated data blocks.
8319 @GNUTAR{} uses a more portable and straightforward
8320 archiving approach, it would be fairly difficult that it does
8321 otherwise. Elizabeth Zwicky writes to @file{comp.unix.internals}, on
8325 What I did say is that you cannot tell the difference between a hole and an
8326 equivalent number of nulls without reading raw blocks. @code{st_blocks} at
8327 best tells you how many holes there are; it doesn't tell you @emph{where}.
8328 Just as programs may, conceivably, care what @code{st_blocks} is (care
8329 to name one that does?), they may also care where the holes are (I have
8330 no examples of this one either, but it's equally imaginable).
8332 I conclude from this that good archivers are not portable. One can
8333 arguably conclude that if you want a portable program, you can in good
8334 conscience restore files with as many holes as possible, since you can't
8339 @section Handling File Attributes
8342 When @command{tar} reads files, it updates their access times. To
8343 avoid this, use the @option{--atime-preserve[=METHOD]} option, which can either
8344 reset the access time retroactively or avoid changing it in the first
8347 Handling of file attributes
8350 @opindex atime-preserve
8351 @item --atime-preserve
8352 @itemx --atime-preserve=replace
8353 @itemx --atime-preserve=system
8354 Preserve the access times of files that are read. This works only for
8355 files that you own, unless you have superuser privileges.
8357 @option{--atime-preserve=replace} works on most systems, but it also
8358 restores the data modification time and updates the status change
8359 time. Hence it doesn't interact with incremental dumps nicely
8360 (@pxref{Backups}), and it can set access or data modification times
8361 incorrectly if other programs access the file while @command{tar} is
8364 @option{--atime-preserve=system} avoids changing the access time in
8365 the first place, if the operating system supports this.
8366 Unfortunately, this may or may not work on any given operating system
8367 or file system. If @command{tar} knows for sure it won't work, it
8368 complains right away.
8370 Currently @option{--atime-preserve} with no operand defaults to
8371 @option{--atime-preserve=replace}, but this is intended to change to
8372 @option{--atime-preserve=system} when the latter is better-supported.
8377 Do not extract data modification time.
8379 When this option is used, @command{tar} leaves the data modification times
8380 of the files it extracts as the times when the files were extracted,
8381 instead of setting it to the times recorded in the archive.
8383 This option is meaningless with @option{--list} (@option{-t}).
8387 Create extracted files with the same ownership they have in the
8390 This is the default behavior for the superuser,
8391 so this option is meaningful only for non-root users, when @command{tar}
8392 is executed on those systems able to give files away. This is
8393 considered as a security flaw by many people, at least because it
8394 makes quite difficult to correctly account users for the disk space
8395 they occupy. Also, the @code{suid} or @code{sgid} attributes of
8396 files are easily and silently lost when files are given away.
8398 When writing an archive, @command{tar} writes the user id and user name
8399 separately. If it can't find a user name (because the user id is not
8400 in @file{/etc/passwd}), then it does not write one. When restoring,
8401 it tries to look the name (if one was written) up in
8402 @file{/etc/passwd}. If it fails, then it uses the user id stored in
8403 the archive instead.
8405 @opindex no-same-owner
8406 @item --no-same-owner
8408 Do not attempt to restore ownership when extracting. This is the
8409 default behavior for ordinary users, so this option has an effect
8410 only for the superuser.
8412 @opindex numeric-owner
8413 @item --numeric-owner
8414 The @option{--numeric-owner} option allows (ANSI) archives to be written
8415 without user/group name information or such information to be ignored
8416 when extracting. It effectively disables the generation and/or use
8417 of user/group name information. This option forces extraction using
8418 the numeric ids from the archive, ignoring the names.
8420 This is useful in certain circumstances, when restoring a backup from
8421 an emergency floppy with different passwd/group files for example.
8422 It is otherwise impossible to extract files with the right ownerships
8423 if the password file in use during the extraction does not match the
8424 one belonging to the file system(s) being extracted. This occurs,
8425 for example, if you are restoring your files after a major crash and
8426 had booted from an emergency floppy with no password file or put your
8427 disk into another machine to do the restore.
8429 The numeric ids are @emph{always} saved into @command{tar} archives.
8430 The identifying names are added at create time when provided by the
8431 system, unless @option{--old-archive} (@option{-o}) is used. Numeric ids could be
8432 used when moving archives between a collection of machines using
8433 a centralized management for attribution of numeric ids to users
8434 and groups. This is often made through using the NIS capabilities.
8436 When making a @command{tar} file for distribution to other sites, it
8437 is sometimes cleaner to use a single owner for all files in the
8438 distribution, and nicer to specify the write permission bits of the
8439 files as stored in the archive independently of their actual value on
8440 the file system. The way to prepare a clean distribution is usually
8441 to have some Makefile rule creating a directory, copying all needed
8442 files in that directory, then setting ownership and permissions as
8443 wanted (there are a lot of possible schemes), and only then making a
8444 @command{tar} archive out of this directory, before cleaning
8445 everything out. Of course, we could add a lot of options to
8446 @GNUTAR{} for fine tuning permissions and ownership.
8447 This is not the good way, I think. @GNUTAR{} is
8448 already crowded with options and moreover, the approach just explained
8449 gives you a great deal of control already.
8451 @opindex same-permissions, short description
8452 @opindex preserve-permissions, short description
8454 @itemx --same-permissions
8455 @itemx --preserve-permissions
8456 Extract all protection information.
8458 This option causes @command{tar} to set the modes (access permissions) of
8459 extracted files exactly as recorded in the archive. If this option
8460 is not used, the current @code{umask} setting limits the permissions
8461 on extracted files. This option is by default enabled when
8462 @command{tar} is executed by a superuser.
8465 This option is meaningless with @option{--list} (@option{-t}).
8469 Same as both @option{--same-permissions} and @option{--same-order}.
8471 The @option{--preserve} option has no equivalent short option name.
8472 It is equivalent to @option{--same-permissions} plus @option{--same-order}.
8474 @FIXME{I do not see the purpose of such an option. (Neither I. FP.)
8475 Neither do I. --Sergey}
8480 @section Comparison of @command{tar} and @command{cpio}
8483 @FIXME{Reorganize the following material}
8485 The @command{cpio} archive formats, like @command{tar}, do have maximum
8486 pathname lengths. The binary and old ASCII formats have a max path
8487 length of 256, and the new ASCII and CRC ASCII formats have a max
8488 path length of 1024. @acronym{GNU} @command{cpio} can read and write archives
8489 with arbitrary pathname lengths, but other @command{cpio} implementations
8490 may crash unexplainedly trying to read them.
8492 @command{tar} handles symbolic links in the form in which it comes in BSD;
8493 @command{cpio} doesn't handle symbolic links in the form in which it comes
8494 in System V prior to SVR4, and some vendors may have added symlinks
8495 to their system without enhancing @command{cpio} to know about them.
8496 Others may have enhanced it in a way other than the way I did it
8497 at Sun, and which was adopted by AT&T (and which is, I think, also
8498 present in the @command{cpio} that Berkeley picked up from AT&T and put
8499 into a later BSD release---I think I gave them my changes).
8501 (SVR4 does some funny stuff with @command{tar}; basically, its @command{cpio}
8502 can handle @command{tar} format input, and write it on output, and it
8503 probably handles symbolic links. They may not have bothered doing
8504 anything to enhance @command{tar} as a result.)
8506 @command{cpio} handles special files; traditional @command{tar} doesn't.
8508 @command{tar} comes with V7, System III, System V, and BSD source;
8509 @command{cpio} comes only with System III, System V, and later BSD
8510 (4.3-tahoe and later).
8512 @command{tar}'s way of handling multiple hard links to a file can handle
8513 file systems that support 32-bit inumbers (e.g., the BSD file system);
8514 @command{cpio}s way requires you to play some games (in its "binary"
8515 format, i-numbers are only 16 bits, and in its "portable ASCII" format,
8516 they're 18 bits---it would have to play games with the "file system ID"
8517 field of the header to make sure that the file system ID/i-number pairs
8518 of different files were always different), and I don't know which
8519 @command{cpio}s, if any, play those games. Those that don't might get
8520 confused and think two files are the same file when they're not, and
8521 make hard links between them.
8523 @command{tar}s way of handling multiple hard links to a file places only
8524 one copy of the link on the tape, but the name attached to that copy
8525 is the @emph{only} one you can use to retrieve the file; @command{cpio}s
8526 way puts one copy for every link, but you can retrieve it using any
8530 What type of check sum (if any) is used, and how is this calculated.
8533 See the attached manual pages for @command{tar} and @command{cpio} format.
8534 @command{tar} uses a checksum which is the sum of all the bytes in the
8535 @command{tar} header for a file; @command{cpio} uses no checksum.
8538 If anyone knows why @command{cpio} was made when @command{tar} was present
8542 It wasn't. @command{cpio} first showed up in PWB/UNIX 1.0; no
8543 generally-available version of UNIX had @command{tar} at the time. I don't
8544 know whether any version that was generally available @emph{within AT&T}
8545 had @command{tar}, or, if so, whether the people within AT&T who did
8546 @command{cpio} knew about it.
8548 On restore, if there is a corruption on a tape @command{tar} will stop at
8549 that point, while @command{cpio} will skip over it and try to restore the
8552 The main difference is just in the command syntax and header format.
8554 @command{tar} is a little more tape-oriented in that everything is blocked
8555 to start on a record boundary.
8558 Is there any differences between the ability to recover crashed
8559 archives between the two of them. (Is there any chance of recovering
8560 crashed archives at all.)
8563 Theoretically it should be easier under @command{tar} since the blocking
8564 lets you find a header with some variation of @samp{dd skip=@var{nn}}.
8565 However, modern @command{cpio}'s and variations have an option to just
8566 search for the next file header after an error with a reasonable chance
8567 of resyncing. However, lots of tape driver software won't allow you to
8568 continue past a media error which should be the only reason for getting
8569 out of sync unless a file changed sizes while you were writing the
8573 If anyone knows why @command{cpio} was made when @command{tar} was present
8574 at the unix scene, please tell me about this too.
8577 Probably because it is more media efficient (by not blocking everything
8578 and using only the space needed for the headers where @command{tar}
8579 always uses 512 bytes per file header) and it knows how to archive
8582 You might want to look at the freely available alternatives. The
8583 major ones are @command{afio}, @GNUTAR{}, and
8584 @command{pax}, each of which have their own extensions with some
8585 backwards compatibility.
8587 Sparse files were @command{tar}red as sparse files (which you can
8588 easily test, because the resulting archive gets smaller, and
8589 @acronym{GNU} @command{cpio} can no longer read it).
8592 @chapter Tapes and Other Archive Media
8595 A few special cases about tape handling warrant more detailed
8596 description. These special cases are discussed below.
8598 Many complexities surround the use of @command{tar} on tape drives. Since
8599 the creation and manipulation of archives located on magnetic tape was
8600 the original purpose of @command{tar}, it contains many features making
8601 such manipulation easier.
8603 Archives are usually written on dismountable media---tape cartridges,
8604 mag tapes, or floppy disks.
8606 The amount of data a tape or disk holds depends not only on its size,
8607 but also on how it is formatted. A 2400 foot long reel of mag tape
8608 holds 40 megabytes of data when formatted at 1600 bits per inch. The
8609 physically smaller EXABYTE tape cartridge holds 2.3 gigabytes.
8611 Magnetic media are re-usable---once the archive on a tape is no longer
8612 needed, the archive can be erased and the tape or disk used over.
8613 Media quality does deteriorate with use, however. Most tapes or disks
8614 should be discarded when they begin to produce data errors. EXABYTE
8615 tape cartridges should be discarded when they generate an @dfn{error
8616 count} (number of non-usable bits) of more than 10k.
8618 Magnetic media are written and erased using magnetic fields, and
8619 should be protected from such fields to avoid damage to stored data.
8620 Sticking a floppy disk to a filing cabinet using a magnet is probably
8624 * Device:: Device selection and switching
8625 * Remote Tape Server::
8626 * Common Problems and Solutions::
8627 * Blocking:: Blocking
8628 * Many:: Many archives on one tape
8629 * Using Multiple Tapes:: Using Multiple Tapes
8630 * label:: Including a Label in the Archive
8632 * Write Protection::
8636 @section Device Selection and Switching
8640 @item -f [@var{hostname}:]@var{file}
8641 @itemx --file=[@var{hostname}:]@var{file}
8642 Use archive file or device @var{file} on @var{hostname}.
8645 This option is used to specify the file name of the archive @command{tar}
8648 If the file name is @samp{-}, @command{tar} reads the archive from standard
8649 input (when listing or extracting), or writes it to standard output
8650 (when creating). If the @samp{-} file name is given when updating an
8651 archive, @command{tar} will read the original archive from its standard
8652 input, and will write the entire new archive to its standard output.
8654 If the file name contains a @samp{:}, it is interpreted as
8655 @samp{hostname:file name}. If the @var{hostname} contains an @dfn{at}
8656 sign (@samp{@@}), it is treated as @samp{user@@hostname:file name}. In
8657 either case, @command{tar} will invoke the command @command{rsh} (or
8658 @command{remsh}) to start up an @command{/usr/libexec/rmt} on the remote
8659 machine. If you give an alternate login name, it will be given to the
8661 Naturally, the remote machine must have an executable
8662 @command{/usr/libexec/rmt}. This program is free software from the
8663 University of California, and a copy of the source code can be found
8664 with the sources for @command{tar}; it's compiled and installed by default.
8665 The exact path to this utility is determined when configuring the package.
8666 It is @file{@var{prefix}/libexec/rmt}, where @var{prefix} stands for
8667 your installation prefix. This location may also be overridden at
8668 runtime by using @option{rmt-command=@var{command}} option (@xref{Option Summary,
8669 ---rmt-command}, for detailed description of this option. @xref{Remote
8670 Tape Server}, for the description of @command{rmt} command).
8672 If this option is not given, but the environment variable @env{TAPE}
8673 is set, its value is used; otherwise, old versions of @command{tar}
8674 used a default archive name (which was picked when @command{tar} was
8675 compiled). The default is normally set up to be the @dfn{first} tape
8676 drive or other transportable I/O medium on the system.
8678 Starting with version 1.11.5, @GNUTAR{} uses
8679 standard input and standard output as the default device, and I will
8680 not try anymore supporting automatic device detection at installation
8681 time. This was failing really in too many cases, it was hopeless.
8682 This is now completely left to the installer to override standard
8683 input and standard output for default device, if this seems
8684 preferable. Further, I think @emph{most} actual usages of
8685 @command{tar} are done with pipes or disks, not really tapes,
8686 cartridges or diskettes.
8688 Some users think that using standard input and output is running
8689 after trouble. This could lead to a nasty surprise on your screen if
8690 you forget to specify an output file name---especially if you are going
8691 through a network or terminal server capable of buffering large amounts
8692 of output. We had so many bug reports in that area of configuring
8693 default tapes automatically, and so many contradicting requests, that
8694 we finally consider the problem to be portably intractable. We could
8695 of course use something like @samp{/dev/tape} as a default, but this
8696 is @emph{also} running after various kind of trouble, going from hung
8697 processes to accidental destruction of real tapes. After having seen
8698 all this mess, using standard input and output as a default really
8699 sounds like the only clean choice left, and a very useful one too.
8701 @GNUTAR{} reads and writes archive in records, I
8702 suspect this is the main reason why block devices are preferred over
8703 character devices. Most probably, block devices are more efficient
8704 too. The installer could also check for @samp{DEFTAPE} in
8705 @file{<sys/mtio.h>}.
8708 @opindex force-local, short description
8710 Archive file is local even if it contains a colon.
8712 @opindex rsh-command
8713 @item --rsh-command=@var{command}
8714 Use remote @var{command} instead of @command{rsh}. This option exists
8715 so that people who use something other than the standard @command{rsh}
8716 (e.g., a Kerberized @command{rsh}) can access a remote device.
8718 When this command is not used, the shell command found when
8719 the @command{tar} program was installed is used instead. This is
8720 the first found of @file{/usr/ucb/rsh}, @file{/usr/bin/remsh},
8721 @file{/usr/bin/rsh}, @file{/usr/bsd/rsh} or @file{/usr/bin/nsh}.
8722 The installer may have overridden this by defining the environment
8723 variable @env{RSH} @emph{at installation time}.
8726 Specify drive and density.
8728 @opindex multi-volume, short description
8730 @itemx --multi-volume
8731 Create/list/extract multi-volume archive.
8733 This option causes @command{tar} to write a @dfn{multi-volume} archive---one
8734 that may be larger than will fit on the medium used to hold it.
8735 @xref{Multi-Volume Archives}.
8737 @opindex tape-length, short description
8739 @itemx --tape-length=@var{num}
8740 Change tape after writing @var{num} x 1024 bytes.
8742 This option might be useful when your tape drivers do not properly
8743 detect end of physical tapes. By being slightly conservative on the
8744 maximum tape length, you might avoid the problem entirely.
8746 @opindex info-script, short description
8747 @opindex new-volume-script, short description
8749 @itemx --info-script=@var{file}
8750 @itemx --new-volume-script=@var{file}
8751 Execute @file{file} at end of each tape. This implies
8752 @option{--multi-volume} (@option{-M}). @xref{info-script}, for a detailed
8753 description of this option.
8756 @node Remote Tape Server
8757 @section The Remote Tape Server
8759 @cindex remote tape drive
8761 In order to access the tape drive on a remote machine, @command{tar}
8762 uses the remote tape server written at the University of California at
8763 Berkeley. The remote tape server must be installed as
8764 @file{@var{prefix}/libexec/rmt} on any machine whose tape drive you
8765 want to use. @command{tar} calls @command{rmt} by running an
8766 @command{rsh} or @command{remsh} to the remote machine, optionally
8767 using a different login name if one is supplied.
8769 A copy of the source for the remote tape server is provided. It is
8770 Copyright @copyright{} 1983 by the Regents of the University of
8771 California, but can be freely distributed. It is compiled and
8772 installed by default.
8774 @cindex absolute file names
8775 Unless you use the @option{--absolute-names} (@option{-P}) option,
8776 @GNUTAR{} will not allow you to create an archive that contains
8777 absolute file names (a file name beginning with @samp{/}.) If you try,
8778 @command{tar} will automatically remove the leading @samp{/} from the
8779 file names it stores in the archive. It will also type a warning
8780 message telling you what it is doing.
8782 When reading an archive that was created with a different
8783 @command{tar} program, @GNUTAR{} automatically
8784 extracts entries in the archive which have absolute file names as if
8785 the file names were not absolute. This is an important feature. A
8786 visitor here once gave a @command{tar} tape to an operator to restore;
8787 the operator used Sun @command{tar} instead of @GNUTAR{},
8788 and the result was that it replaced large portions of
8789 our @file{/bin} and friends with versions from the tape; needless to
8790 say, we were unhappy about having to recover the file system from
8793 For example, if the archive contained a file @file{/usr/bin/computoy},
8794 @GNUTAR{} would extract the file to @file{usr/bin/computoy},
8795 relative to the current directory. If you want to extract the files in
8796 an archive to the same absolute names that they had when the archive
8797 was created, you should do a @samp{cd /} before extracting the files
8798 from the archive, or you should either use the @option{--absolute-names}
8799 option, or use the command @samp{tar -C / @dots{}}.
8801 @cindex Ultrix 3.1 and write failure
8802 Some versions of Unix (Ultrix 3.1 is known to have this problem),
8803 can claim that a short write near the end of a tape succeeded,
8804 when it actually failed. This will result in the -M option not
8805 working correctly. The best workaround at the moment is to use a
8806 significantly larger blocking factor than the default 20.
8808 In order to update an archive, @command{tar} must be able to backspace the
8809 archive in order to reread or rewrite a record that was just read (or
8810 written). This is currently possible only on two kinds of files: normal
8811 disk files (or any other file that can be backspaced with @samp{lseek}),
8812 and industry-standard 9-track magnetic tape (or any other kind of tape
8813 that can be backspaced with the @code{MTIOCTOP} @code{ioctl}.
8815 This means that the @option{--append}, @option{--concatenate}, and
8816 @option{--delete} commands will not work on any other kind of file.
8817 Some media simply cannot be backspaced, which means these commands and
8818 options will never be able to work on them. These non-backspacing
8819 media include pipes and cartridge tape drives.
8821 Some other media can be backspaced, and @command{tar} will work on them
8822 once @command{tar} is modified to do so.
8824 Archives created with the @option{--multi-volume}, @option{--label}, and
8825 @option{--incremental} (@option{-G}) options may not be readable by other version
8826 of @command{tar}. In particular, restoring a file that was split over
8827 a volume boundary will require some careful work with @command{dd}, if
8828 it can be done at all. Other versions of @command{tar} may also create
8829 an empty file whose name is that of the volume header. Some versions
8830 of @command{tar} may create normal files instead of directories archived
8831 with the @option{--incremental} (@option{-G}) option.
8833 @node Common Problems and Solutions
8834 @section Some Common Problems and their Solutions
8841 no such file or directory
8844 errors from @command{tar}:
8845 directory checksum error
8848 errors from media/system:
8859 @dfn{Block} and @dfn{record} terminology is rather confused, and it
8860 is also confusing to the expert reader. On the other hand, readers
8861 who are new to the field have a fresh mind, and they may safely skip
8862 the next two paragraphs, as the remainder of this manual uses those
8863 two terms in a quite consistent way.
8865 John Gilmore, the writer of the public domain @command{tar} from which
8866 @GNUTAR{} was originally derived, wrote (June 1995):
8869 The nomenclature of tape drives comes from IBM, where I believe
8870 they were invented for the IBM 650 or so. On IBM mainframes, what
8871 is recorded on tape are tape blocks. The logical organization of
8872 data is into records. There are various ways of putting records into
8873 blocks, including @code{F} (fixed sized records), @code{V} (variable
8874 sized records), @code{FB} (fixed blocked: fixed size records, @var{n}
8875 to a block), @code{VB} (variable size records, @var{n} to a block),
8876 @code{VSB} (variable spanned blocked: variable sized records that can
8877 occupy more than one block), etc. The @code{JCL} @samp{DD RECFORM=}
8878 parameter specified this to the operating system.
8880 The Unix man page on @command{tar} was totally confused about this.
8881 When I wrote @code{PD TAR}, I used the historically correct terminology
8882 (@command{tar} writes data records, which are grouped into blocks).
8883 It appears that the bogus terminology made it into @acronym{POSIX} (no surprise
8884 here), and now Fran@,{c}ois has migrated that terminology back
8885 into the source code too.
8888 The term @dfn{physical block} means the basic transfer chunk from or
8889 to a device, after which reading or writing may stop without anything
8890 being lost. In this manual, the term @dfn{block} usually refers to
8891 a disk physical block, @emph{assuming} that each disk block is 512
8892 bytes in length. It is true that some disk devices have different
8893 physical blocks, but @command{tar} ignore these differences in its own
8894 format, which is meant to be portable, so a @command{tar} block is always
8895 512 bytes in length, and @dfn{block} always mean a @command{tar} block.
8896 The term @dfn{logical block} often represents the basic chunk of
8897 allocation of many disk blocks as a single entity, which the operating
8898 system treats somewhat atomically; this concept is only barely used
8901 The term @dfn{physical record} is another way to speak of a physical
8902 block, those two terms are somewhat interchangeable. In this manual,
8903 the term @dfn{record} usually refers to a tape physical block,
8904 @emph{assuming} that the @command{tar} archive is kept on magnetic tape.
8905 It is true that archives may be put on disk or used with pipes,
8906 but nevertheless, @command{tar} tries to read and write the archive one
8907 @dfn{record} at a time, whatever the medium in use. One record is made
8908 up of an integral number of blocks, and this operation of putting many
8909 disk blocks into a single tape block is called @dfn{reblocking}, or
8910 more simply, @dfn{blocking}. The term @dfn{logical record} refers to
8911 the logical organization of many characters into something meaningful
8912 to the application. The term @dfn{unit record} describes a small set
8913 of characters which are transmitted whole to or by the application,
8914 and often refers to a line of text. Those two last terms are unrelated
8915 to what we call a @dfn{record} in @GNUTAR{}.
8917 When writing to tapes, @command{tar} writes the contents of the archive
8918 in chunks known as @dfn{records}. To change the default blocking
8919 factor, use the @option{--blocking-factor=@var{512-size}} (@option{-b
8920 @var{512-size}}) option. Each record will then be composed of
8921 @var{512-size} blocks. (Each @command{tar} block is 512 bytes.
8922 @xref{Standard}.) Each file written to the archive uses at least one
8923 full record. As a result, using a larger record size can result in
8924 more wasted space for small files. On the other hand, a larger record
8925 size can often be read and written much more efficiently.
8927 Further complicating the problem is that some tape drives ignore the
8928 blocking entirely. For these, a larger record size can still improve
8929 performance (because the software layers above the tape drive still
8930 honor the blocking), but not as dramatically as on tape drives that
8933 When reading an archive, @command{tar} can usually figure out the
8934 record size on itself. When this is the case, and a non-standard
8935 record size was used when the archive was created, @command{tar} will
8936 print a message about a non-standard blocking factor, and then operate
8937 normally. On some tape devices, however, @command{tar} cannot figure
8938 out the record size itself. On most of those, you can specify a
8939 blocking factor (with @option{--blocking-factor}) larger than the
8940 actual blocking factor, and then use the @option{--read-full-records}
8941 (@option{-B}) option. (If you specify a blocking factor with
8942 @option{--blocking-factor} and don't use the
8943 @option{--read-full-records} option, then @command{tar} will not
8944 attempt to figure out the recording size itself.) On some devices,
8945 you must always specify the record size exactly with
8946 @option{--blocking-factor} when reading, because @command{tar} cannot
8947 figure it out. In any case, use @option{--list} (@option{-t}) before
8948 doing any extractions to see whether @command{tar} is reading the archive
8951 @command{tar} blocks are all fixed size (512 bytes), and its scheme for
8952 putting them into records is to put a whole number of them (one or
8953 more) into each record. @command{tar} records are all the same size;
8954 at the end of the file there's a block containing all zeros, which
8955 is how you tell that the remainder of the last record(s) are garbage.
8957 In a standard @command{tar} file (no options), the block size is 512
8958 and the record size is 10240, for a blocking factor of 20. What the
8959 @option{--blocking-factor} option does is sets the blocking factor,
8960 changing the record size while leaving the block size at 512 bytes.
8961 20 was fine for ancient 800 or 1600 bpi reel-to-reel tape drives;
8962 most tape drives these days prefer much bigger records in order to
8963 stream and not waste tape. When writing tapes for myself, some tend
8964 to use a factor of the order of 2048, say, giving a record size of
8965 around one megabyte.
8967 If you use a blocking factor larger than 20, older @command{tar}
8968 programs might not be able to read the archive, so we recommend this
8969 as a limit to use in practice. @GNUTAR{}, however,
8970 will support arbitrarily large record sizes, limited only by the
8971 amount of virtual memory or the physical characteristics of the tape
8975 * Format Variations:: Format Variations
8976 * Blocking Factor:: The Blocking Factor of an Archive
8979 @node Format Variations
8980 @subsection Format Variations
8981 @cindex Format Parameters
8982 @cindex Format Options
8983 @cindex Options, archive format specifying
8984 @cindex Options, format specifying
8987 Format parameters specify how an archive is written on the archive
8988 media. The best choice of format parameters will vary depending on
8989 the type and number of files being archived, and on the media used to
8992 To specify format parameters when accessing or creating an archive,
8993 you can use the options described in the following sections.
8994 If you do not specify any format parameters, @command{tar} uses
8995 default parameters. You cannot modify a compressed archive.
8996 If you create an archive with the @option{--blocking-factor} option
8997 specified (@pxref{Blocking Factor}), you must specify that
8998 blocking-factor when operating on the archive. @xref{Formats}, for other
8999 examples of format parameter considerations.
9001 @node Blocking Factor
9002 @subsection The Blocking Factor of an Archive
9003 @cindex Blocking Factor
9005 @cindex Number of blocks per record
9006 @cindex Number of bytes per record
9007 @cindex Bytes per record
9008 @cindex Blocks per record
9011 @opindex blocking-factor
9012 The data in an archive is grouped into blocks, which are 512 bytes.
9013 Blocks are read and written in whole number multiples called
9014 @dfn{records}. The number of blocks in a record (ie. the size of a
9015 record in units of 512 bytes) is called the @dfn{blocking factor}.
9016 The @option{--blocking-factor=@var{512-size}} (@option{-b
9017 @var{512-size}}) option specifies the blocking factor of an archive.
9018 The default blocking factor is typically 20 (i.e., 10240 bytes), but
9019 can be specified at installation. To find out the blocking factor of
9020 an existing archive, use @samp{tar --list --file=@var{archive-name}}.
9021 This may not work on some devices.
9023 Records are separated by gaps, which waste space on the archive media.
9024 If you are archiving on magnetic tape, using a larger blocking factor
9025 (and therefore larger records) provides faster throughput and allows you
9026 to fit more data on a tape (because there are fewer gaps). If you are
9027 archiving on cartridge, a very large blocking factor (say 126 or more)
9028 greatly increases performance. A smaller blocking factor, on the other
9029 hand, may be useful when archiving small files, to avoid archiving lots
9030 of nulls as @command{tar} fills out the archive to the end of the record.
9031 In general, the ideal record size depends on the size of the
9032 inter-record gaps on the tape you are using, and the average size of the
9033 files you are archiving. @xref{create}, for information on
9036 @FIXME{Need example of using a cartridge with blocking factor=126 or more.}
9038 Archives with blocking factors larger than 20 cannot be read
9039 by very old versions of @command{tar}, or by some newer versions
9040 of @command{tar} running on old machines with small address spaces.
9041 With @GNUTAR{}, the blocking factor of an archive is limited
9042 only by the maximum record size of the device containing the archive,
9043 or by the amount of available virtual memory.
9045 Also, on some systems, not using adequate blocking factors, as sometimes
9046 imposed by the device drivers, may yield unexpected diagnostics. For
9047 example, this has been reported:
9050 Cannot write to /dev/dlt: Invalid argument
9054 In such cases, it sometimes happen that the @command{tar} bundled by
9055 the system is aware of block size idiosyncrasies, while @GNUTAR{}
9056 requires an explicit specification for the block size,
9057 which it cannot guess. This yields some people to consider
9058 @GNUTAR{} is misbehaving, because by comparison,
9059 @cite{the bundle @command{tar} works OK}. Adding @w{@kbd{-b 256}},
9060 for example, might resolve the problem.
9062 If you use a non-default blocking factor when you create an archive, you
9063 must specify the same blocking factor when you modify that archive. Some
9064 archive devices will also require you to specify the blocking factor when
9065 reading that archive, however this is not typically the case. Usually, you
9066 can use @option{--list} (@option{-t}) without specifying a blocking factor---@command{tar}
9067 reports a non-default record size and then lists the archive members as
9068 it would normally. To extract files from an archive with a non-standard
9069 blocking factor (particularly if you're not sure what the blocking factor
9070 is), you can usually use the @option{--read-full-records} (@option{-B}) option while
9071 specifying a blocking factor larger then the blocking factor of the archive
9072 (ie. @samp{tar --extract --read-full-records --blocking-factor=300}.
9073 @xref{list}, for more information on the @option{--list} (@option{-t})
9074 operation. @xref{Reading}, for a more detailed explanation of that option.
9077 @item --blocking-factor=@var{number}
9078 @itemx -b @var{number}
9079 Specifies the blocking factor of an archive. Can be used with any
9080 operation, but is usually not necessary with @option{--list} (@option{-t}).
9086 @item -b @var{blocks}
9087 @itemx --blocking-factor=@var{blocks}
9088 Set record size to @math{@var{blocks} * 512} bytes.
9090 This option is used to specify a @dfn{blocking factor} for the archive.
9091 When reading or writing the archive, @command{tar}, will do reads and writes
9092 of the archive in records of @math{@var{block}*512} bytes. This is true
9093 even when the archive is compressed. Some devices requires that all
9094 write operations be a multiple of a certain size, and so, @command{tar}
9095 pads the archive out to the next record boundary.
9097 The default blocking factor is set when @command{tar} is compiled, and is
9098 typically 20. Blocking factors larger than 20 cannot be read by very
9099 old versions of @command{tar}, or by some newer versions of @command{tar}
9100 running on old machines with small address spaces.
9102 With a magnetic tape, larger records give faster throughput and fit
9103 more data on a tape (because there are fewer inter-record gaps).
9104 If the archive is in a disk file or a pipe, you may want to specify
9105 a smaller blocking factor, since a large one will result in a large
9106 number of null bytes at the end of the archive.
9108 When writing cartridge or other streaming tapes, a much larger
9109 blocking factor (say 126 or more) will greatly increase performance.
9110 However, you must specify the same blocking factor when reading or
9111 updating the archive.
9113 Apparently, Exabyte drives have a physical block size of 8K bytes.
9114 If we choose our blocksize as a multiple of 8k bytes, then the problem
9115 seems to disappear. Id est, we are using block size of 112 right
9116 now, and we haven't had the problem since we switched@dots{}
9118 With @GNUTAR{} the blocking factor is limited only
9119 by the maximum record size of the device containing the archive, or by
9120 the amount of available virtual memory.
9122 However, deblocking or reblocking is virtually avoided in a special
9123 case which often occurs in practice, but which requires all the
9124 following conditions to be simultaneously true:
9127 the archive is subject to a compression option,
9129 the archive is not handled through standard input or output, nor
9130 redirected nor piped,
9132 the archive is directly handled to a local disk, instead of any special
9135 @option{--blocking-factor} is not explicitly specified on the @command{tar}
9139 If the output goes directly to a local disk, and not through
9140 stdout, then the last write is not extended to a full record size.
9141 Otherwise, reblocking occurs. Here are a few other remarks on this
9147 @command{gzip} will complain about trailing garbage if asked to
9148 uncompress a compressed archive on tape, there is an option to turn
9149 the message off, but it breaks the regularity of simply having to use
9150 @samp{@var{prog} -d} for decompression. It would be nice if gzip was
9151 silently ignoring any number of trailing zeros. I'll ask Jean-loup
9152 Gailly, by sending a copy of this message to him.
9155 @command{compress} does not show this problem, but as Jean-loup pointed
9156 out to Michael, @samp{compress -d} silently adds garbage after
9157 the result of decompression, which tar ignores because it already
9158 recognized its end-of-file indicator. So this bug may be safely
9162 @samp{gzip -d -q} will be silent about the trailing zeros indeed,
9163 but will still return an exit status of 2 which tar reports in turn.
9164 @command{tar} might ignore the exit status returned, but I hate doing
9165 that, as it weakens the protection @command{tar} offers users against
9166 other possible problems at decompression time. If @command{gzip} was
9167 silently skipping trailing zeros @emph{and} also avoiding setting the
9168 exit status in this innocuous case, that would solve this situation.
9171 @command{tar} should become more solid at not stopping to read a pipe at
9172 the first null block encountered. This inelegantly breaks the pipe.
9173 @command{tar} should rather drain the pipe out before exiting itself.
9176 @opindex ignore-zeros, short description
9178 @itemx --ignore-zeros
9179 Ignore blocks of zeros in archive (means EOF).
9181 The @option{--ignore-zeros} (@option{-i}) option causes @command{tar} to ignore blocks
9182 of zeros in the archive. Normally a block of zeros indicates the
9183 end of the archive, but when reading a damaged archive, or one which
9184 was created by concatenating several archives together, this option
9185 allows @command{tar} to read the entire archive. This option is not on
9186 by default because many versions of @command{tar} write garbage after
9189 Note that this option causes @command{tar} to read to the end of the
9190 archive file, which may sometimes avoid problems when multiple files
9191 are stored on a single physical tape.
9193 @opindex read-full-records, short description
9195 @itemx --read-full-records
9196 Reblock as we read (for reading 4.2BSD pipes).
9198 If @option{--read-full-records} is used, @command{tar}
9199 will not panic if an attempt to read a record from the archive does
9200 not return a full record. Instead, @command{tar} will keep reading
9201 until it has obtained a full
9204 This option is turned on by default when @command{tar} is reading
9205 an archive from standard input, or from a remote machine. This is
9206 because on BSD Unix systems, a read of a pipe will return however
9207 much happens to be in the pipe, even if it is less than @command{tar}
9208 requested. If this option was not used, @command{tar} would fail as
9209 soon as it read an incomplete record from the pipe.
9211 This option is also useful with the commands for updating an archive.
9217 @FIXME{Appropriate options should be moved here from elsewhere.}
9219 @cindex blocking factor
9220 @cindex tape blocking
9222 When handling various tapes or cartridges, you have to take care of
9223 selecting a proper blocking, that is, the number of disk blocks you
9224 put together as a single tape block on the tape, without intervening
9225 tape gaps. A @dfn{tape gap} is a small landing area on the tape
9226 with no information on it, used for decelerating the tape to a
9227 full stop, and for later regaining the reading or writing speed.
9228 When the tape driver starts reading a record, the record has to
9229 be read whole without stopping, as a tape gap is needed to stop the
9230 tape motion without loosing information.
9232 @cindex Exabyte blocking
9233 @cindex DAT blocking
9234 Using higher blocking (putting more disk blocks per tape block) will use
9235 the tape more efficiently as there will be less tape gaps. But reading
9236 such tapes may be more difficult for the system, as more memory will be
9237 required to receive at once the whole record. Further, if there is a
9238 reading error on a huge record, this is less likely that the system will
9239 succeed in recovering the information. So, blocking should not be too
9240 low, nor it should be too high. @command{tar} uses by default a blocking of
9241 20 for historical reasons, and it does not really matter when reading or
9242 writing to disk. Current tape technology would easily accommodate higher
9243 blockings. Sun recommends a blocking of 126 for Exabytes and 96 for DATs.
9244 We were told that for some DLT drives, the blocking should be a multiple
9245 of 4Kb, preferably 64Kb (@w{@kbd{-b 128}}) or 256 for decent performance.
9246 Other manufacturers may use different recommendations for the same tapes.
9247 This might also depends of the buffering techniques used inside modern
9248 tape controllers. Some imposes a minimum blocking, or a maximum blocking.
9249 Others request blocking to be some exponent of two.
9251 So, there is no fixed rule for blocking. But blocking at read time
9252 should ideally be the same as blocking used at write time. At one place
9253 I know, with a wide variety of equipment, they found it best to use a
9254 blocking of 32 to guarantee that their tapes are fully interchangeable.
9256 I was also told that, for recycled tapes, prior erasure (by the same
9257 drive unit that will be used to create the archives) sometimes lowers
9258 the error rates observed at rewriting time.
9260 I might also use @option{--number-blocks} instead of
9261 @option{--block-number}, so @option{--block} will then expand to
9262 @option{--blocking-factor} unambiguously.
9265 @section Many Archives on One Tape
9267 @FIXME{Appropriate options should be moved here from elsewhere.}
9269 @findex ntape @r{device}
9270 Most tape devices have two entries in the @file{/dev} directory, or
9271 entries that come in pairs, which differ only in the minor number for
9272 this device. Let's take for example @file{/dev/tape}, which often
9273 points to the only or usual tape device of a given system. There might
9274 be a corresponding @file{/dev/nrtape} or @file{/dev/ntape}. The simpler
9275 name is the @emph{rewinding} version of the device, while the name
9276 having @samp{nr} in it is the @emph{no rewinding} version of the same
9279 A rewinding tape device will bring back the tape to its beginning point
9280 automatically when this device is opened or closed. Since @command{tar}
9281 opens the archive file before using it and closes it afterwards, this
9282 means that a simple:
9285 $ @kbd{tar cf /dev/tape @var{directory}}
9289 will reposition the tape to its beginning both prior and after saving
9290 @var{directory} contents to it, thus erasing prior tape contents and
9291 making it so that any subsequent write operation will destroy what has
9294 @cindex tape positioning
9295 So, a rewinding device is normally meant to hold one and only one file.
9296 If you want to put more than one @command{tar} archive on a given tape, you
9297 will need to avoid using the rewinding version of the tape device. You
9298 will also have to pay special attention to tape positioning. Errors in
9299 positioning may overwrite the valuable data already on your tape. Many
9300 people, burnt by past experiences, will only use rewinding devices and
9301 limit themselves to one file per tape, precisely to avoid the risk of
9302 such errors. Be fully aware that writing at the wrong position on a
9303 tape loses all information past this point and most probably until the
9304 end of the tape, and this destroyed information @emph{cannot} be
9307 To save @var{directory-1} as a first archive at the beginning of a
9308 tape, and leave that tape ready for a second archive, you should use:
9311 $ @kbd{mt -f /dev/nrtape rewind}
9312 $ @kbd{tar cf /dev/nrtape @var{directory-1}}
9316 @dfn{Tape marks} are special magnetic patterns written on the tape
9317 media, which are later recognizable by the reading hardware. These
9318 marks are used after each file, when there are many on a single tape.
9319 An empty file (that is to say, two tape marks in a row) signal the
9320 logical end of the tape, after which no file exist. Usually,
9321 non-rewinding tape device drivers will react to the close request issued
9322 by @command{tar} by first writing two tape marks after your archive, and by
9323 backspacing over one of these. So, if you remove the tape at that time
9324 from the tape drive, it is properly terminated. But if you write
9325 another file at the current position, the second tape mark will be
9326 erased by the new information, leaving only one tape mark between files.
9328 So, you may now save @var{directory-2} as a second archive after the
9329 first on the same tape by issuing the command:
9332 $ @kbd{tar cf /dev/nrtape @var{directory-2}}
9336 and so on for all the archives you want to put on the same tape.
9338 Another usual case is that you do not write all the archives the same
9339 day, and you need to remove and store the tape between two archive
9340 sessions. In general, you must remember how many files are already
9341 saved on your tape. Suppose your tape already has 16 files on it, and
9342 that you are ready to write the 17th. You have to take care of skipping
9343 the first 16 tape marks before saving @var{directory-17}, say, by using
9347 $ @kbd{mt -f /dev/nrtape rewind}
9348 $ @kbd{mt -f /dev/nrtape fsf 16}
9349 $ @kbd{tar cf /dev/nrtape @var{directory-17}}
9352 In all the previous examples, we put aside blocking considerations, but
9353 you should do the proper things for that as well. @xref{Blocking}.
9356 * Tape Positioning:: Tape Positions and Tape Marks
9357 * mt:: The @command{mt} Utility
9360 @node Tape Positioning
9361 @subsection Tape Positions and Tape Marks
9364 Just as archives can store more than one file from the file system,
9365 tapes can store more than one archive file. To keep track of where
9366 archive files (or any other type of file stored on tape) begin and
9367 end, tape archive devices write magnetic @dfn{tape marks} on the
9368 archive media. Tape drives write one tape mark between files,
9369 two at the end of all the file entries.
9371 If you think of data as a series of records "rrrr"'s, and tape marks as
9372 "*"'s, a tape might look like the following:
9375 rrrr*rrrrrr*rrrrr*rr*rrrrr**-------------------------
9378 Tape devices read and write tapes using a read/write @dfn{tape
9379 head}---a physical part of the device which can only access one
9380 point on the tape at a time. When you use @command{tar} to read or
9381 write archive data from a tape device, the device will begin reading
9382 or writing from wherever on the tape the tape head happens to be,
9383 regardless of which archive or what part of the archive the tape
9384 head is on. Before writing an archive, you should make sure that no
9385 data on the tape will be overwritten (unless it is no longer needed).
9386 Before reading an archive, you should make sure the tape head is at
9387 the beginning of the archive you want to read. You can do it manually
9388 via @code{mt} utility (@pxref{mt}). The @code{restore} script does
9389 that automatically (@pxref{Scripted Restoration}).
9391 If you want to add new archive file entries to a tape, you should
9392 advance the tape to the end of the existing file entries, backspace
9393 over the last tape mark, and write the new archive file. If you were
9394 to add two archives to the example above, the tape might look like the
9398 rrrr*rrrrrr*rrrrr*rr*rrrrr*rrr*rrrr**----------------
9402 @subsection The @command{mt} Utility
9405 @FIXME{Is it true that this only works on non-block devices?
9406 should explain the difference, (fixed or variable).}
9407 @xref{Blocking Factor}.
9409 You can use the @command{mt} utility to advance or rewind a tape past a
9410 specified number of archive files on the tape. This will allow you
9411 to move to the beginning of an archive before extracting or reading
9412 it, or to the end of all the archives before writing a new one.
9413 @FIXME{Why isn't there an "advance 'til you find two tape marks
9416 The syntax of the @command{mt} command is:
9419 @kbd{mt [-f @var{tapename}] @var{operation} [@var{number}]}
9422 where @var{tapename} is the name of the tape device, @var{number} is
9423 the number of times an operation is performed (with a default of one),
9424 and @var{operation} is one of the following:
9426 @FIXME{is there any use for record operations?}
9431 Writes @var{number} tape marks at the current position on the tape.
9434 Moves tape position forward @var{number} files.
9437 Moves tape position back @var{number} files.
9440 Rewinds the tape. (Ignores @var{number}).
9444 Rewinds the tape and takes the tape device off-line. (Ignores @var{number}).
9447 Prints status information about the tape unit.
9451 @FIXME{Is there a better way to frob the spacing on the list?}
9453 If you don't specify a @var{tapename}, @command{mt} uses the environment
9454 variable @env{TAPE}; if @env{TAPE} is not set, @command{mt} will use
9455 the default device specified in your @file{sys/mtio.h} file
9456 (@code{DEFTAPE} variable). If this is not defined, the program will
9457 display a descriptive error message and exit with code 1.
9459 @command{mt} returns a 0 exit status when the operation(s) were
9460 successful, 1 if the command was unrecognized, and 2 if an operation
9463 @node Using Multiple Tapes
9464 @section Using Multiple Tapes
9467 Often you might want to write a large archive, one larger than will fit
9468 on the actual tape you are using. In such a case, you can run multiple
9469 @command{tar} commands, but this can be inconvenient, particularly if you
9470 are using options like @option{--exclude=@var{pattern}} or dumping entire file systems.
9471 Therefore, @command{tar} supports multiple tapes automatically.
9473 Use @option{--multi-volume} (@option{-M}) on the command line, and
9474 then @command{tar} will, when it reaches the end of the tape, prompt
9475 for another tape, and continue the archive. Each tape will have an
9476 independent archive, and can be read without needing the other. (As
9477 an exception to this, the file that @command{tar} was archiving when
9478 it ran out of tape will usually be split between the two archives; in
9479 this case you need to extract from the first archive, using
9480 @option{--multi-volume}, and then put in the second tape when
9481 prompted, so @command{tar} can restore both halves of the file.)
9483 @GNUTAR{} multi-volume archives do not use a truly portable format.
9484 You need @GNUTAR{} at both ends to process them properly.
9486 When prompting for a new tape, @command{tar} accepts any of the following
9491 Request @command{tar} to explain possible responses
9493 Request @command{tar} to exit immediately.
9494 @item n @var{file-name}
9495 Request @command{tar} to write the next volume on the file @var{file-name}.
9497 Request @command{tar} to run a subshell. This option can be disabled
9498 by giving @option{--restrict} command line option to @command{tar}.
9500 Request @command{tar} to begin writing the next volume.
9503 (You should only type @samp{y} after you have changed the tape;
9504 otherwise @command{tar} will write over the volume it just finished.)
9506 @cindex End-of-archive info script
9508 @anchor{info-script}
9509 @opindex info-script
9510 @opindex new-volume-script
9511 If you want more elaborate behavior than this, give @command{tar} the
9512 @option{--info-script=@var{script-name}}
9513 (@option{--new-volume-script=@var{script-name}}, @option{-F
9514 @var{script-name}}) option. The file @var{script-name} is expected to
9515 be a program (or shell script) to be run instead of the normal
9516 prompting procedure. It is executed without any command line
9517 arguments. Additional data is passed to it via the following
9518 environment variables:
9521 @vrindex TAR_VERSION, info script environment variable
9523 @GNUTAR{} version number.
9525 @vrindex TAR_ARCHIVE, info script environment variable
9527 The name of the archive @command{tar} is processing.
9529 @vrindex TAR_VOLUME, info script environment variable
9531 Ordinal number of the volume @command{tar} is about to start.
9533 @vrindex TAR_SUBCOMMAND, info script environment variable
9534 @item TAR_SUBCOMMAND
9535 Short option describing the operation @command{tar} is executing
9536 @xref{Operations}, for a complete list of subcommand options.
9538 @vrindex TAR_FORMAT, info script environment variable
9540 Format of the archive being processed. @xref{Formats}, for a complete
9541 list of archive format names.
9544 The info script can instruct @command{tar} to use new archive name,
9545 by writing in to file descriptor 3 (see below for an
9548 If the info script fails, @command{tar} exits; otherwise, it begins
9549 writing the next volume.
9551 The method @command{tar} uses to detect end of tape is not perfect, and
9552 fails on some operating systems or on some devices. You can use the
9553 @option{--tape-length=@var{size}} (@option{-L @var{size}}) option if
9554 @command{tar} can't detect the end of the tape itself. This option
9555 selects @option{--multi-volume} (@option{-M}) automatically. The
9556 @var{size} argument should then be the usable size of the tape in
9557 units of 1024 bytes. But for many devices, and floppy disks in
9558 particular, this option is never required for real, as far as we know.
9560 @cindex Volume number file
9564 The volume number used by @command{tar} in its tape-change prompt
9565 can be changed; if you give the
9566 @option{--volno-file=@var{file-of-number}} option, then
9567 @var{file-of-number} should be an unexisting file to be created, or
9568 else, a file already containing a decimal number. That number will be
9569 used as the volume number of the first volume written. When
9570 @command{tar} is finished, it will rewrite the file with the
9571 now-current volume number. (This does not change the volume number
9572 written on a tape label, as per @ref{label}, it @emph{only} affects
9573 the number used in the prompt.)
9575 If you want @command{tar} to cycle through a series of files or tape
9576 drives, there are three approaches to choose from. First of all, you
9577 can give @command{tar} multiple @option{--file} options. In this case
9578 the specified files will be used, in sequence, as the successive
9579 volumes of the archive. Only when the first one in the sequence needs
9580 to be used again will @command{tar} prompt for a tape change (or run
9581 the info script). Secondly, you can use the @samp{n} response to the
9582 tape-change prompt, and, finally, you can use an info script, that
9583 writes new archive name to file descriptor. The following example
9584 illustrates this approach:
9589 echo Preparing volume $TAR_VOLUME of $TAR_ARCHIVE.
9591 name=`expr $TAR_ARCHIVE : '\(.*\)-.*'`
9592 case $TAR_SUBCOMMAND in
9594 -d|-x|-t) test -r $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME || exit 1
9599 echo $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME >&3
9603 Each volume of a multi-volume archive is an independent @command{tar}
9604 archive, complete in itself. For example, you can list or extract any
9605 volume alone; just don't specify @option{--multi-volume}
9606 (@option{-M}). However, if one file in the archive is split across
9607 volumes, the only way to extract it successfully is with a
9608 multi-volume extract command @option{--extract --multi-volume}
9609 (@option{-xM}) starting on or before the volume where the file begins.
9611 For example, let's presume someone has two tape drives on a system
9612 named @file{/dev/tape0} and @file{/dev/tape1}. For having @GNUTAR{}
9613 to switch to the second drive when it needs to write the
9614 second tape, and then back to the first tape, etc., just do either of:
9617 $ @kbd{tar --create --multi-volume --file=/dev/tape0 --file=/dev/tape1 @var{files}}
9618 $ @kbd{tar cMff /dev/tape0 /dev/tape1 @var{files}}
9622 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
9623 * Tape Files:: Tape Files
9624 * Tarcat:: Concatenate Volumes into a Single Archive
9628 @node Multi-Volume Archives
9629 @subsection Archives Longer than One Tape or Disk
9630 @cindex Multi-volume archives
9633 @opindex multi-volume
9634 To create an archive that is larger than will fit on a single unit of
9635 the media, use the @option{--multi-volume} (@option{-M}) option in conjunction with
9636 the @option{--create} option (@pxref{create}). A @dfn{multi-volume}
9637 archive can be manipulated like any other archive (provided the
9638 @option{--multi-volume} option is specified), but is stored on more
9639 than one tape or disk.
9641 When you specify @option{--multi-volume}, @command{tar} does not report an
9642 error when it comes to the end of an archive volume (when reading), or
9643 the end of the media (when writing). Instead, it prompts you to load
9644 a new storage volume. If the archive is on a magnetic tape, you
9645 should change tapes when you see the prompt; if the archive is on a
9646 floppy disk, you should change disks; etc.
9648 You can read each individual volume of a multi-volume archive as if it
9649 were an archive by itself. For example, to list the contents of one
9650 volume, use @option{--list}, without @option{--multi-volume} specified.
9651 To extract an archive member from one volume (assuming it is described
9652 that volume), use @option{--extract}, again without
9653 @option{--multi-volume}.
9655 If an archive member is split across volumes (ie. its entry begins on
9656 one volume of the media and ends on another), you need to specify
9657 @option{--multi-volume} to extract it successfully. In this case, you
9658 should load the volume where the archive member starts, and use
9659 @samp{tar --extract --multi-volume}---@command{tar} will prompt for later
9660 volumes as it needs them. @xref{extracting archives}, for more
9661 information about extracting archives.
9663 @option{--info-script=@var{script-name}}
9664 (@option{--new-volume-script=@var{script-name}}, @option{-F
9665 @var{script-name}}) (@pxref{info-script}) is like
9666 @option{--multi-volume} (@option{-M}), except that @command{tar} does
9667 not prompt you directly to change media volumes when a volume is
9668 full---instead, @command{tar} runs commands you have stored in
9669 @var{script-name}. For example, this option can be used to eject
9670 cassettes, or to broadcast messages such as @samp{Someone please come
9671 change my tape} when performing unattended backups. When
9672 @var{script-name} is done, @command{tar} will assume that the media
9675 Multi-volume archives can be modified like any other archive. To add
9676 files to a multi-volume archive, you need to only mount the last
9677 volume of the archive media (and new volumes, if needed). For all
9678 other operations, you need to use the entire archive.
9680 If a multi-volume archive was labeled using
9681 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9682 (@pxref{label}) when it was created, @command{tar} will not
9683 automatically label volumes which are added later. To label
9684 subsequent volumes, specify @option{--label=@var{archive-label}} again
9685 in conjunction with the @option{--append}, @option{--update} or
9686 @option{--concatenate} operation.
9688 @cindex Labeling multi-volume archives
9691 @FIXME{There should be a sample program here, including an exit
9692 before end. Is the exit status even checked in tar? :-(}
9695 @item --multi-volume
9697 Creates a multi-volume archive, when used in conjunction with
9698 @option{--create} (@option{-c}). To perform any other operation on a multi-volume
9699 archive, specify @option{--multi-volume} in conjunction with that
9702 @item --info-script=@var{program-file}
9703 @itemx --new-volume-script=@var{program-file}
9704 @itemx -F @var{program-file}
9705 Creates a multi-volume archive via a script. Used in conjunction with
9706 @option{--create} (@option{-c}). @xref{info-script}, dor a detailed discussion.
9709 Beware that there is @emph{no} real standard about the proper way, for
9710 a @command{tar} archive, to span volume boundaries. If you have a
9711 multi-volume created by some vendor's @command{tar}, there is almost
9712 no chance you could read all the volumes with @GNUTAR{}.
9713 The converse is also true: you may not expect
9714 multi-volume archives created by @GNUTAR{} to be
9715 fully recovered by vendor's @command{tar}. Since there is little
9716 chance that, in mixed system configurations, some vendor's
9717 @command{tar} will work on another vendor's machine, and there is a
9718 great chance that @GNUTAR{} will work on most of
9719 them, your best bet is to install @GNUTAR{} on all
9720 machines between which you know exchange of files is possible.
9723 @subsection Tape Files
9726 To give the archive a name which will be recorded in it, use the
9727 @option{--label=@var{volume-label}} (@option{-V @var{volume-label}})
9728 option. This will write a special block identifying
9729 @var{volume-label} as the name of the archive to the front of the
9730 archive which will be displayed when the archive is listed with
9731 @option{--list}. If you are creating a multi-volume archive with
9732 @option{--multi-volume} (@pxref{Using Multiple Tapes}), then the
9733 volume label will have @samp{Volume @var{nnn}} appended to the name
9734 you give, where @var{nnn} is the number of the volume of the archive.
9735 (If you use the @option{--label=@var{volume-label}}) option when
9736 reading an archive, it checks to make sure the label on the tape
9737 matches the one you give. @xref{label}.
9739 When @command{tar} writes an archive to tape, it creates a single
9740 tape file. If multiple archives are written to the same tape, one
9741 after the other, they each get written as separate tape files. When
9742 extracting, it is necessary to position the tape at the right place
9743 before running @command{tar}. To do this, use the @command{mt} command.
9744 For more information on the @command{mt} command and on the organization
9745 of tapes into a sequence of tape files, see @ref{mt}.
9747 People seem to often do:
9750 @kbd{--label="@var{some-prefix} `date +@var{some-format}`"}
9753 or such, for pushing a common date in all volumes or an archive set.
9756 @subsection Concatenate Volumes into a Single Archive
9759 Sometimes it is necessary to convert existing @GNUTAR{} multi-volume
9760 archive to a single @command{tar} archive. Simply concatenating all
9761 volumes into one will not work, since each volume carries an additional
9762 information at the beginning. @GNUTAR{} is shipped with the shell
9763 script @command{tarcat} designed for this purpose.
9765 The script takes a list of files comprising a multi-volume archive
9766 and creates the resulting archive at the standard output. For example:
9769 @kbd{tarcat vol.1 vol.2 vol.3 | tar tf -}
9772 The script implements a simple heuristics to determine the format of
9773 the first volume file and to decide how to process the rest of the
9774 files. However, it makes no attempt to verify whether the files are
9775 given in order or even if they are valid @command{tar} archives.
9776 It uses @command{dd} and does not filter its standard error, so you
9777 will usually see lots of spurious messages.
9779 @FIXME{The script is not installed. Should we install it?}
9782 @section Including a Label in the Archive
9783 @cindex Labeling an archive
9784 @cindex Labels on the archive media
9788 To avoid problems caused by misplaced paper labels on the archive
9789 media, you can include a @dfn{label} entry---an archive member which
9790 contains the name of the archive---in the archive itself. Use the
9791 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9792 option in conjunction with the @option{--create} operation to include
9793 a label entry in the archive as it is being created.
9796 @item --label=@var{archive-label}
9797 @itemx -V @var{archive-label}
9798 Includes an @dfn{archive-label} at the beginning of the archive when
9799 the archive is being created, when used in conjunction with the
9800 @option{--create} operation. Checks to make sure the archive label
9801 matches the one specified (when used in conjunction with any other
9805 If you create an archive using both
9806 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9807 and @option{--multi-volume} (@option{-M}), each volume of the archive
9808 will have an archive label of the form @samp{@var{archive-label}
9809 Volume @var{n}}, where @var{n} is 1 for the first volume, 2 for the
9810 next, and so on. @xref{Using Multiple Tapes}, for information on
9811 creating multiple volume archives.
9813 @cindex Volume label, listing
9814 @cindex Listing volume label
9815 The volume label will be displayed by @option{--list} along with
9816 the file contents. If verbose display is requested, it will also be
9817 explicitely marked as in the example below:
9821 $ @kbd{tar --verbose --list --file=iamanarchive}
9822 V--------- 0 0 0 1992-03-07 12:01 iamalabel--Volume Header--
9823 -rw-r--r-- ringo user 40 1990-05-21 13:30 iamafilename
9828 @anchor{--test-label option}
9829 However, @option{--list} option will cause listing entire
9830 contents of the archive, which may be undesirable (for example, if the
9831 archive is stored on a tape). You can request checking only the volume
9832 by specifying @option{--test-label} option. This option reads only the
9833 first block of an archive, so it can be used with slow storage
9834 devices. For example:
9838 $ @kbd{tar --test-label --file=iamanarchive}
9843 If @option{--test-label} is used with a single command line
9844 argument, @command{tar} compares the volume label with the
9845 argument. It exits with code 0 if the two strings match, and with code
9846 2 otherwise. In this case no output is displayed. For example:
9850 $ @kbd{tar --test-label --file=iamanarchive 'iamalable'}
9852 $ @kbd{tar --test-label --file=iamanarchive 'iamalable' alabel}
9857 If you request any operation, other than @option{--create}, along
9858 with using @option{--label} option, @command{tar} will first check if
9859 the archive label matches the one specified and will refuse to proceed
9860 if it does not. Use this as a safety precaution to avoid accidentally
9861 overwriting existing archives. For example, if you wish to add files
9862 to @file{archive}, presumably labelled with string @samp{My volume},
9867 $ @kbd{tar -rf archive --label 'My volume' .}
9868 tar: Archive not labeled to match `My volume'
9873 in case its label does not match. This will work even if
9874 @file{archive} is not labelled at all.
9876 Similarly, @command{tar} will refuse to list or extract the
9877 archive if its label doesn't match the @var{archive-label}
9878 specified. In those cases, @var{archive-label} argument is interpreted
9879 as a globbing-style pattern which must match the actual magnetic
9880 volume label. @xref{exclude}, for a precise description of how match
9881 is attempted@footnote{Previous versions of @command{tar} used full
9882 regular expression matching, or before that, only exact string
9883 matching, instead of wildcard matchers. We decided for the sake of
9884 simplicity to use a uniform matching device through
9885 @command{tar}.}. If the switch @option{--multi-volume} (@option{-M}) is being used,
9886 the volume label matcher will also suffix @var{archive-label} by
9887 @w{@samp{ Volume [1-9]*}} if the initial match fails, before giving
9888 up. Since the volume numbering is automatically added in labels at
9889 creation time, it sounded logical to equally help the user taking care
9890 of it when the archive is being read.
9892 The @option{--label} was once called @option{--volume}, but is not
9893 available under that name anymore.
9895 You can also use @option{--label} to get a common information on
9896 all tapes of a series. For having this information different in each
9897 series created through a single script used on a regular basis, just
9898 manage to get some date string as part of the label. For example:
9902 $ @kbd{tar cfMV /dev/tape "Daily backup for `date +%Y-%m-%d`"}
9903 $ @kbd{tar --create --file=/dev/tape --multi-volume \
9904 --volume="Daily backup for `date +%Y-%m-%d`"}
9908 Also note that each label has its own date and time, which corresponds
9909 to when @GNUTAR{} initially attempted to write it,
9910 often soon after the operator launches @command{tar} or types the
9911 carriage return telling that the next tape is ready. Comparing date
9912 labels does give an idea of tape throughput only if the delays for
9913 rewinding tapes and the operator switching them were negligible, which
9914 is usually not the case.
9917 @section Verifying Data as It is Stored
9918 @cindex Verifying a write operation
9919 @cindex Double-checking a write operation
9924 @opindex verify, short description
9925 Attempt to verify the archive after writing.
9928 This option causes @command{tar} to verify the archive after writing it.
9929 Each volume is checked after it is written, and any discrepancies
9930 are recorded on the standard error output.
9932 Verification requires that the archive be on a back-space-able medium.
9933 This means pipes, some cartridge tape drives, and some other devices
9936 You can insure the accuracy of an archive by comparing files in the
9937 system with archive members. @command{tar} can compare an archive to the
9938 file system as the archive is being written, to verify a write
9939 operation, or can compare a previously written archive, to insure that
9942 @opindex verify, using with @option{--create}
9943 @opindex create, using with @option{--verify}
9944 To check for discrepancies in an archive immediately after it is
9945 written, use the @option{--verify} (@option{-W}) option in conjunction with
9946 the @option{--create} operation. When this option is
9947 specified, @command{tar} checks archive members against their counterparts
9948 in the file system, and reports discrepancies on the standard error.
9950 To verify an archive, you must be able to read it from before the end
9951 of the last written entry. This option is useful for detecting data
9952 errors on some tapes. Archives written to pipes, some cartridge tape
9953 drives, and some other devices cannot be verified.
9955 One can explicitly compare an already made archive with the file
9956 system by using the @option{--compare} (@option{--diff}, @option{-d})
9957 option, instead of using the more automatic @option{--verify} option.
9960 Note that these two options have a slightly different intent. The
9961 @option{--compare} option checks how identical are the logical contents of some
9962 archive with what is on your disks, while the @option{--verify} option is
9963 really for checking if the physical contents agree and if the recording
9964 media itself is of dependable quality. So, for the @option{--verify}
9965 operation, @command{tar} tries to defeat all in-memory cache pertaining to
9966 the archive, while it lets the speed optimization undisturbed for the
9967 @option{--compare} option. If you nevertheless use @option{--compare} for
9968 media verification, you may have to defeat the in-memory cache yourself,
9969 maybe by opening and reclosing the door latch of your recording unit,
9970 forcing some doubt in your operating system about the fact this is really
9971 the same volume as the one just written or read.
9973 The @option{--verify} option would not be necessary if drivers were indeed
9974 able to detect dependably all write failures. This sometimes require many
9975 magnetic heads, some able to read after the writes occurred. One would
9976 not say that drivers unable to detect all cases are necessarily flawed,
9977 as long as programming is concerned.
9979 The @option{--verify} (@option{-W}) option will not work in
9980 conjunction with the @option{--multi-volume} (@option{-M}) option or
9981 the @option{--append} (@option{-r}), @option{--update} (@option{-u})
9982 and @option{--delete} operations. @xref{Operations}, for more
9983 information on these operations.
9985 Also, since @command{tar} normally strips leading @samp{/} from file
9986 names (@pxref{absolute}), a command like @samp{tar --verify -cf
9987 /tmp/foo.tar /etc} will work as desired only if the working directory is
9988 @file{/}, as @command{tar} uses the archive's relative member names
9989 (e.g., @file{etc/motd}) when verifying the archive.
9991 @node Write Protection
9992 @section Write Protection
9994 Almost all tapes and diskettes, and in a few rare cases, even disks can
9995 be @dfn{write protected}, to protect data on them from being changed.
9996 Once an archive is written, you should write protect the media to prevent
9997 the archive from being accidentally overwritten or deleted. (This will
9998 protect the archive from being changed with a tape or floppy drive---it
9999 will not protect it from magnet fields or other physical hazards).
10001 The write protection device itself is usually an integral part of the
10002 physical media, and can be a two position (write enabled/write
10003 disabled) switch, a notch which can be popped out or covered, a ring
10004 which can be removed from the center of a tape reel, or some other
10005 changeable feature.
10010 This appendix lists some important user-visible changes between
10011 version @GNUTAR{} @value{VERSION} and previous versions. An up-to-date
10012 version of this document is available at
10013 @uref{http://www.gnu.org/@/software/@/tar/manual/changes.html,the
10014 @GNUTAR{} documentation page}.
10017 @item Use of globbing patterns when listing and extracting.
10019 Previous versions of GNU tar assumed shell-style globbing when
10020 extracting from or listing an archive. For example:
10023 $ @kbd{tar xf foo.tar '*.c'}
10026 would extract all files whose names end in @samp{.c}. This behavior
10027 was not documented and was incompatible with traditional tar
10028 implementations. Therefore, starting from version 1.15.91, GNU tar
10029 no longer uses globbing by default. For example, the above invocation
10030 is now interpreted as a request to extract from the archive the file
10033 To facilitate transition to the new behavior for those users who got
10034 used to the previous incorrect one, @command{tar} will print a warning
10035 if it finds out that a requested member was not found in the archive
10036 and its name looks like a globbing pattern. For example:
10039 $ @kbd{tar xf foo.tar '*.c'}
10040 tar: Pattern matching characters used in file names. Please,
10041 tar: use --wildcards to enable pattern matching, or --no-wildcards to
10042 tar: suppress this warning.
10043 tar: *.c: Not found in archive
10044 tar: Error exit delayed from previous errors
10047 To treat member names as globbing patterns, use --wildcards option.
10048 If you want to tar to mimic the behavior of versions prior to 1.15.91,
10049 add this option to your @env{TAR_OPTIONS} variable.
10051 @xref{wildcards}, for the detailed discussion of the use of globbing
10052 patterns by @GNUTAR{}.
10054 @item Use of short option @option{-o}.
10056 Earlier versions of @GNUTAR{} understood @option{-o} command line
10057 option as a synonym for @option{--old-archive}.
10059 @GNUTAR{} starting from version 1.13.90 understands this option as
10060 a synonym for @option{--no-same-owner}. This is compatible with
10061 UNIX98 @command{tar} implementations.
10063 However, to facilitate transition, @option{-o} option retains its
10064 old semantics when it is used with one of archive-creation commands.
10065 Users are encouraged to use @option{--format=oldgnu} instead.
10067 It is especially important, since versions of @acronym{GNU} Automake
10068 up to and including 1.8.4 invoke tar with this option to produce
10069 distribution tarballs. @xref{Formats,v7}, for the detailed discussion
10070 of this issue and its implications.
10072 @FIXME{Change the first argument to tar-formats when the new Automake is
10073 out. The proposition to add @anchor{} to the appropriate place of its
10074 docs was accepted by Automake people --Sergey 2006-05-25}.
10075 @xref{Options, tar-v7, Changing Automake's Behavior,
10076 automake, GNU Automake}, for a description on how to use various
10077 archive formats with @command{automake}.
10079 Future versions of @GNUTAR{} will understand @option{-o} only as a
10080 synonym for @option{--no-same-owner}.
10082 @item Use of short option @option{-l}
10084 Earlier versions of @GNUTAR{} understood @option{-l} option as a
10085 synonym for @option{--one-file-system}. Since such usage contradicted
10086 to UNIX98 specification and harmed compatibility with other
10087 implementation, it was declared deprecated in version 1.14. However,
10088 to facilitate transition to its new semantics, it was supported by
10089 versions 1.15 and 1.15.90. The present use of @option{-l} as a short
10090 variant of @option{--check-links} was introduced in version 1.15.91.
10092 @item Use of options @option{--portability} and @option{--old-archive}
10094 These options are deprecated. Please use @option{--format=v7} instead.
10096 @item Use of option @option{--posix}
10098 This option is deprecated. Please use @option{--format=posix} instead.
10101 @node Configuring Help Summary
10102 @appendix Configuring Help Summary
10104 Running @kbd{tar --help} displays the short @command{tar} option
10105 summary (@pxref{help}). This summary is organised by @dfn{groups} of
10106 semantically close options. The options within each group are printed
10107 in the following order: a short option, eventually followed by a list
10108 of corresponding long option names, followed by a short description of
10109 the option. For example, here is an excerpt from the actual @kbd{tar
10113 Main operation mode:
10115 -A, --catenate, --concatenate append tar files to an archive
10116 -c, --create create a new archive
10117 -d, --diff, --compare find differences between archive and
10119 --delete delete from the archive
10122 @vrindex ARGP_HELP_FMT, environment variable
10123 The exact visual representation of the help output is configurable via
10124 @env{ARGP_HELP_FMT} environment variable. The value of this variable
10125 is a comma-separated list of @dfn{format variable} assignments. There
10126 are two kinds of format variables. An @dfn{offset variable} keeps the
10127 offset of some part of help output text from the leftmost column on
10128 the screen. A @dfn{boolean} variable is a flag that toggles some
10129 output feature on or off. Depending on the type of the corresponding
10130 variable, there are two kinds of assignments:
10133 @item Offset assignment
10135 The assignment to an offset variable has the following syntax:
10138 @var{variable}=@var{value}
10142 where @var{variable} is the variable name, and @var{value} is a
10143 numeric value to be assigned to the variable.
10145 @item Boolean assignment
10147 To assign @code{true} value to a variable, simply put this variable name. To
10148 assign @code{false} value, prefix the variable name with @samp{no-}. For
10153 # Assign @code{true} value:
10155 # Assign @code{false} value:
10161 Following variables are declared:
10163 @deftypevr {Help Output} boolean dup-args
10164 If true, arguments for an option are shown with both short and long
10165 options, even when a given option has both forms, for example:
10168 -f ARCHIVE, --file=ARCHIVE use archive file or device ARCHIVE
10171 If false, then if an option has both short and long forms, the
10172 argument is only shown with the long one, for example:
10175 -f, --file=ARCHIVE use archive file or device ARCHIVE
10179 and a message indicating that the argument is applicable to both
10180 forms is printed below the options. This message can be disabled
10181 using @code{dup-args-note} (see below).
10183 The default is false.
10186 @deftypevr {Help Output} boolean dup-args-note
10187 If this variable is true, which is the default, the following notice
10188 is displayed at the end of the help output:
10191 Mandatory or optional arguments to long options are also mandatory or
10192 optional for any corresponding short options.
10195 Setting @code{no-dup-args-note} inhibits this message. Normally, only one of
10196 variables @code{dup-args} or @code{dup-args-note} should be set.
10199 @deftypevr {Help Output} offset short-opt-col
10200 Column in which short options start. Default is 2.
10204 $ @kbd{tar --help|grep ARCHIVE}
10205 -f, --file=ARCHIVE use archive file or device ARCHIVE
10206 $ @kbd{ARGP_HELP_FMT=short-opt-col=6 tar --help|grep ARCHIVE}
10207 -f, --file=ARCHIVE use archive file or device ARCHIVE
10212 @deftypevr {Help Output} offset long-opt-col
10213 Column in which long options start. Default is 6. For example:
10217 $ @kbd{tar --help|grep ARCHIVE}
10218 -f, --file=ARCHIVE use archive file or device ARCHIVE
10219 $ @kbd{ARGP_HELP_FMT=long-opt-col=16 tar --help|grep ARCHIVE}
10220 -f, --file=ARCHIVE use archive file or device ARCHIVE
10225 @deftypevr {Help Output} offset doc-opt-col
10226 Column in which @dfn{doc options} start. A doc option isn't actually
10227 an option, but rather an arbitrary piece of documentation that is
10228 displayed in much the same manner as the options. For example, in
10229 the description of @option{--format} option:
10233 -H, --format=FORMAT create archive of the given format.
10235 FORMAT is one of the following:
10237 gnu GNU tar 1.13.x format
10238 oldgnu GNU format as per tar <= 1.12
10239 pax POSIX 1003.1-2001 (pax) format
10241 ustar POSIX 1003.1-1988 (ustar) format
10242 v7 old V7 tar format
10247 the format names are doc options. Thus, if you set
10248 @kbd{ARGP_HELP_FMT=doc-opt-col=6} the above part of the help output
10249 will look as follows:
10253 -H, --format=FORMAT create archive of the given format.
10255 FORMAT is one of the following:
10257 gnu GNU tar 1.13.x format
10258 oldgnu GNU format as per tar <= 1.12
10259 pax POSIX 1003.1-2001 (pax) format
10261 ustar POSIX 1003.1-1988 (ustar) format
10262 v7 old V7 tar format
10267 @deftypevr {Help Output} offset opt-doc-col
10268 Column in which option description starts. Default is 29.
10272 $ @kbd{tar --help|grep ARCHIVE}
10273 -f, --file=ARCHIVE use archive file or device ARCHIVE
10274 $ @kbd{ARGP_HELP_FMT=opt-doc-col=19 tar --help|grep ARCHIVE}
10275 -f, --file=ARCHIVE use archive file or device ARCHIVE
10276 $ @kbd{ARGP_HELP_FMT=opt-doc-col=9 tar --help|grep ARCHIVE}
10278 use archive file or device ARCHIVE
10283 Notice, that the description starts on a separate line if
10284 @code{opt-doc-col} value is too small.
10287 @deftypevr {Help Output} offset header-col
10288 Column in which @dfn{group headers} are printed. A group header is a
10289 descriptive text preceding an option group. For example, in the
10293 Main operation mode:
10295 -A, --catenate, --concatenate append tar files to
10297 -c, --create create a new archive
10300 @samp{Main operation mode:} is the group header.
10302 The default value is 1.
10305 @deftypevr {Help Output} offset usage-indent
10306 Indentation of wrapped usage lines. Affects @option{--usage}
10307 output. Default is 12.
10310 @deftypevr {Help Output} offset rmargin
10311 Right margin of the text output. Used for wrapping.
10316 @include genfile.texi
10318 @node Tar Internals
10319 @appendix Tar Internals
10320 @include intern.texi
10322 @node Free Software Needs Free Documentation
10323 @appendix Free Software Needs Free Documentation
10324 @include freemanuals.texi
10326 @node Copying This Manual
10327 @appendix Copying This Manual
10330 * GNU Free Documentation License:: License for copying this manual
10335 @node Index of Command Line Options
10336 @appendix Index of Command Line Options
10338 This appendix contains an index of all @GNUTAR{} long command line
10339 options. The options are listed without the preceeding double-dash.
10341 @FIXME{Provide an index of short options}
10342 @c Do not forget to check if all options are indexed (see maintenance
10343 @c notes at the beginning of this document.
10355 @c Local variables:
10356 @c texinfo-column-for-description: 32