1 \input texinfo @c -*-texinfo-*-
2 @comment %**start of header
5 @settitle GNU tar @value{VERSION}
14 @include rendition.texi
17 @c Put everything in one index (arbitrarily chosen to be the concept index).
27 This manual is for @acronym{GNU} @command{tar} (version
28 @value{VERSION}, @value{UPDATED}), which creates and extracts files
31 Copyright @copyright{} 1992, 1994, 1995, 1996, 1997, 1999, 2000, 2001,
32 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
35 Permission is granted to copy, distribute and/or modify this document
36 under the terms of the GNU Free Documentation License, Version 1.1 or
37 any later version published by the Free Software Foundation; with no
38 Invariant Sections, with the Front-Cover Texts being ``A GNU Manual,''
39 and with the Back-Cover Texts as in (a) below. A copy of the license
40 is included in the section entitled "GNU Free Documentation License".
42 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
43 this GNU Manual. Buying copies from GNU Press supports the FSF in
44 developing GNU and promoting software freedom.''
48 @dircategory Archiving
50 * Tar: (tar). Making tape (or disk) archives.
53 @dircategory Individual utilities
55 * tar: (tar)tar invocation. Invoking @GNUTAR{}.
58 @shorttitlepage @acronym{GNU} @command{tar}
61 @title @acronym{GNU} tar: an archiver tool
62 @subtitle @value{RENDITION} @value{VERSION}, @value{UPDATED}
63 @author John Gilmore, Jay Fenlason et al.
66 @vskip 0pt plus 1filll
72 @top @acronym{GNU} tar: an archiver tool
77 @cindex archiving files
79 The first part of this master menu lists the major nodes in this Info
80 document. The rest of the menu lists all the lower level nodes.
83 @c The master menu, created with texinfo-master-menu, goes here.
84 @c FIXME: Submenus for getdate.texi and intern.texi are interpolated by hand.
93 * Date input formats::
100 * Configuring Help Summary::
103 * Free Software Needs Free Documentation::
104 * Copying This Manual::
105 * Index of Command Line Options::
109 --- The Detailed Node Listing ---
113 * Book Contents:: What this Book Contains
114 * Definitions:: Some Definitions
115 * What tar Does:: What @command{tar} Does
116 * Naming tar Archives:: How @command{tar} Archives are Named
117 * Authors:: @GNUTAR{} Authors
118 * Reports:: Reporting bugs or suggestions
120 Tutorial Introduction to @command{tar}
123 * stylistic conventions::
124 * basic tar options:: Basic @command{tar} Operations and Options
125 * frequent operations::
126 * Two Frequent Options::
127 * create:: How to Create Archives
128 * list:: How to List Archives
129 * extract:: How to Extract Members from an Archive
132 Two Frequently Used Options
138 How to Create Archives
140 * prepare for examples::
141 * Creating the archive::
150 How to Extract Members from an Archive
152 * extracting archives::
155 * extracting untrusted archives::
161 * using tar options::
169 The Three Option Styles
171 * Mnemonic Options:: Mnemonic Option Style
172 * Short Options:: Short Option Style
173 * Old Options:: Old Option Style
174 * Mixing:: Mixing Option Styles
176 All @command{tar} Options
178 * Operation Summary::
180 * Short Option Summary::
192 Advanced @GNUTAR{} Operations
201 How to Add Files to Existing Archives: @option{--append}
203 * appending files:: Appending Files to an Archive
210 Options Used by @option{--create}
212 * Ignore Failed Read::
214 Options Used by @option{--extract}
216 * Reading:: Options to Help Read Archives
217 * Writing:: Changing How @command{tar} Writes Files
218 * Scarce:: Coping with Scarce Resources
220 Options to Help Read Archives
222 * read full records::
225 Changing How @command{tar} Writes Files
227 * Dealing with Old Files::
228 * Overwrite Old Files::
233 * Data Modification Times::
234 * Setting Access Permissions::
235 * Directory Modification Times and Permissions::
236 * Writing to Standard Output::
237 * Writing to an External Program::
240 Coping with Scarce Resources
245 Performing Backups and Restoring Files
247 * Full Dumps:: Using @command{tar} to Perform Full Dumps
248 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
249 * Backup Levels:: Levels of Backups
250 * Backup Parameters:: Setting Parameters for Backups and Restoration
251 * Scripted Backups:: Using the Backup Scripts
252 * Scripted Restoration:: Using the Restore Script
254 Setting Parameters for Backups and Restoration
256 * General-Purpose Variables::
257 * Magnetic Tape Control::
259 * backup-specs example:: An Example Text of @file{Backup-specs}
261 Choosing Files and Names for @command{tar}
263 * file:: Choosing the Archive's Name
264 * Selecting Archive Members::
265 * files:: Reading Names from a File
266 * exclude:: Excluding Some Files
267 * wildcards:: Wildcards Patterns and Matching
268 * quoting styles:: Ways of Quoting Special Characters in Names
269 * transform:: Modifying File and Member Names
270 * after:: Operating Only on New Files
271 * recurse:: Descending into Directories
272 * one:: Crossing File System Boundaries
274 Reading Names from a File
280 * problems with exclude::
282 Wildcards Patterns and Matching
284 * controlling pattern-matching::
286 Crossing File System Boundaries
288 * directory:: Changing Directory
289 * absolute:: Absolute File Names
291 Controlling the Archive Format
293 * Portability:: Making @command{tar} Archives More Portable
294 * Compression:: Using Less Space through Compression
295 * Attributes:: Handling File Attributes
296 * cpio:: Comparison of @command{tar} and @command{cpio}
300 * General date syntax:: Common rules.
301 * Calendar date items:: 19 Dec 1994.
302 * Time of day items:: 9:20pm.
303 * Time zone items:: @sc{est}, @sc{pdt}, @sc{gmt}, ...
304 * Day of week items:: Monday and others.
305 * Relative items in date strings:: next tuesday, 2 years ago.
306 * Pure numbers in date strings:: 19931219, 1440.
307 * Seconds since the Epoch:: @@1078100502.
308 * Authors of get_date:: Bellovin, Eggert, Salz, Berets, et al.
310 Making @command{tar} Archives More Portable
312 * Portable Names:: Portable Names
313 * dereference:: Symbolic Links
314 * old:: Old V7 Archives
315 * ustar:: Ustar Archives
316 * gnu:: GNU and old GNU format archives.
317 * posix:: @acronym{POSIX} archives
318 * Checksumming:: Checksumming Problems
319 * Large or Negative Values:: Large files, negative time stamps, etc.
321 @GNUTAR{} and @acronym{POSIX} @command{tar}
323 * PAX keywords:: Controlling Extended Header Keywords.
325 Using Less Space through Compression
327 * gzip:: Creating and Reading Compressed Archives
328 * sparse:: Archiving Sparse Files
330 Tapes and Other Archive Media
332 * Device:: Device selection and switching
333 * Remote Tape Server::
334 * Common Problems and Solutions::
335 * Blocking:: Blocking
336 * Many:: Many archives on one tape
337 * Using Multiple Tapes:: Using Multiple Tapes
338 * label:: Including a Label in the Archive
344 * Format Variations:: Format Variations
345 * Blocking Factor:: The Blocking Factor of an Archive
347 Many Archives on One Tape
349 * Tape Positioning:: Tape Positions and Tape Marks
350 * mt:: The @command{mt} Utility
354 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
355 * Tape Files:: Tape Files
356 * Tarcat:: Concatenate Volumes into a Single Archive
358 GNU tar internals and development
369 * GNU Free Documentation License:: License for copying this manual
375 @chapter Introduction
378 and manipulates @dfn{archives} which are actually collections of
379 many other files; the program provides users with an organized and
380 systematic method for controlling a large amount of data.
381 The name ``tar'' originally came from the phrase ``Tape ARchive'', but
382 archives need not (and these days, typically do not) reside on tapes.
385 * Book Contents:: What this Book Contains
386 * Definitions:: Some Definitions
387 * What tar Does:: What @command{tar} Does
388 * Naming tar Archives:: How @command{tar} Archives are Named
389 * Authors:: @GNUTAR{} Authors
390 * Reports:: Reporting bugs or suggestions
394 @section What this Book Contains
396 The first part of this chapter introduces you to various terms that will
397 recur throughout the book. It also tells you who has worked on @GNUTAR{}
398 and its documentation, and where you should send bug reports
401 The second chapter is a tutorial (@pxref{Tutorial}) which provides a
402 gentle introduction for people who are new to using @command{tar}. It is
403 meant to be self contained, not requiring any reading from subsequent
404 chapters to make sense. It moves from topic to topic in a logical,
405 progressive order, building on information already explained.
407 Although the tutorial is paced and structured to allow beginners to
408 learn how to use @command{tar}, it is not intended solely for beginners.
409 The tutorial explains how to use the three most frequently used
410 operations (@samp{create}, @samp{list}, and @samp{extract}) as well as
411 two frequently used options (@samp{file} and @samp{verbose}). The other
412 chapters do not refer to the tutorial frequently; however, if a section
413 discusses something which is a complex variant of a basic concept, there
414 may be a cross reference to that basic concept. (The entire book,
415 including the tutorial, assumes that the reader understands some basic
416 concepts of using a Unix-type operating system; @pxref{Tutorial}.)
418 The third chapter presents the remaining five operations, and
419 information about using @command{tar} options and option syntax.
421 @FIXME{this sounds more like a @acronym{GNU} Project Manuals Concept [tm] more
422 than the reality. should think about whether this makes sense to say
423 here, or not.} The other chapters are meant to be used as a
424 reference. Each chapter presents everything that needs to be said
425 about a specific topic.
427 One of the chapters (@pxref{Date input formats}) exists in its
428 entirety in other @acronym{GNU} manuals, and is mostly self-contained.
429 In addition, one section of this manual (@pxref{Standard}) contains a
430 big quote which is taken directly from @command{tar} sources.
432 In general, we give both long and short (abbreviated) option names
433 at least once in each section where the relevant option is covered, so
434 that novice readers will become familiar with both styles. (A few
435 options have no short versions, and the relevant sections will
439 @section Some Definitions
443 The @command{tar} program is used to create and manipulate @command{tar}
444 archives. An @dfn{archive} is a single file which contains the contents
445 of many files, while still identifying the names of the files, their
446 owner(s), and so forth. (In addition, archives record access
447 permissions, user and group, size in bytes, and data modification time.
448 Some archives also record the file names in each archived directory, as
449 well as other file and directory information.) You can use @command{tar}
450 to @dfn{create} a new archive in a specified directory.
453 @cindex archive member
456 The files inside an archive are called @dfn{members}. Within this
457 manual, we use the term @dfn{file} to refer only to files accessible in
458 the normal ways (by @command{ls}, @command{cat}, and so forth), and the term
459 @dfn{member} to refer only to the members of an archive. Similarly, a
460 @dfn{file name} is the name of a file, as it resides in the file system,
461 and a @dfn{member name} is the name of an archive member within the
466 The term @dfn{extraction} refers to the process of copying an archive
467 member (or multiple members) into a file in the file system. Extracting
468 all the members of an archive is often called @dfn{extracting the
469 archive}. The term @dfn{unpack} can also be used to refer to the
470 extraction of many or all the members of an archive. Extracting an
471 archive does not destroy the archive's structure, just as creating an
472 archive does not destroy the copies of the files that exist outside of
473 the archive. You may also @dfn{list} the members in a given archive
474 (this is often thought of as ``printing'' them to the standard output,
475 or the command line), or @dfn{append} members to a pre-existing archive.
476 All of these operations can be performed using @command{tar}.
479 @section What @command{tar} Does
482 The @command{tar} program provides the ability to create @command{tar}
483 archives, as well as various other kinds of manipulation. For example,
484 you can use @command{tar} on previously created archives to extract files,
485 to store additional files, or to update or list files which were already
488 Initially, @command{tar} archives were used to store files conveniently on
489 magnetic tape. The name @command{tar} comes from this use; it stands for
490 @code{t}ape @code{ar}chiver. Despite the utility's name, @command{tar} can
491 direct its output to available devices, files, or other programs (using
492 pipes). @command{tar} may even access remote devices or files (as archives).
494 You can use @command{tar} archives in many ways. We want to stress a few
495 of them: storage, backup, and transportation.
497 @FIXME{the following table entries need a bit of work..}
500 Often, @command{tar} archives are used to store related files for
501 convenient file transfer over a network. For example, the
502 @acronym{GNU} Project distributes its software bundled into
503 @command{tar} archives, so that all the files relating to a particular
504 program (or set of related programs) can be transferred as a single
507 A magnetic tape can store several files in sequence. However, the tape
508 has no names for these files; it only knows their relative position on
509 the tape. One way to store several files on one tape and retain their
510 names is by creating a @command{tar} archive. Even when the basic transfer
511 mechanism can keep track of names, as FTP can, the nuisance of handling
512 multiple files, directories, and multiple links makes @command{tar}
515 Archive files are also used for long-term storage. You can think of
516 this as transportation from the present into the future. (It is a
517 science-fiction idiom that you can move through time as well as in
518 space; the idea here is that @command{tar} can be used to move archives in
519 all dimensions, even time!)
522 Because the archive created by @command{tar} is capable of preserving
523 file information and directory structure, @command{tar} is commonly
524 used for performing full and incremental backups of disks. A backup
525 puts a collection of files (possibly pertaining to many users and
526 projects) together on a disk or a tape. This guards against
527 accidental destruction of the information in those files.
528 @GNUTAR{} has special features that allow it to be
529 used to make incremental and full dumps of all the files in a
533 You can create an archive on one system, transfer it to another system,
534 and extract the contents there. This allows you to transport a group of
535 files from one system to another.
538 @node Naming tar Archives
539 @section How @command{tar} Archives are Named
541 Conventionally, @command{tar} archives are given names ending with
542 @samp{.tar}. This is not necessary for @command{tar} to operate properly,
543 but this manual follows that convention in order to accustom readers to
544 it and to make examples more clear.
549 Often, people refer to @command{tar} archives as ``@command{tar} files,'' and
550 archive members as ``files'' or ``entries''. For people familiar with
551 the operation of @command{tar}, this causes no difficulty. However, in
552 this manual, we consistently refer to ``archives'' and ``archive
553 members'' to make learning to use @command{tar} easier for novice users.
556 @section @GNUTAR{} Authors
558 @GNUTAR{} was originally written by John Gilmore,
559 and modified by many people. The @acronym{GNU} enhancements were
560 written by Jay Fenlason, then Joy Kendall, and the whole package has
561 been further maintained by Thomas Bushnell, n/BSG, Fran@,{c}ois
562 Pinard, Paul Eggert, and finally Sergey Poznyakoff with the help of
563 numerous and kind users.
565 We wish to stress that @command{tar} is a collective work, and owes much to
566 all those people who reported problems, offered solutions and other
567 insights, or shared their thoughts and suggestions. An impressive, yet
568 partial list of those contributors can be found in the @file{THANKS}
569 file from the @GNUTAR{} distribution.
571 @FIXME{i want all of these names mentioned, Absolutely. BUT, i'm not
572 sure i want to spell out the history in this detail, at least not for
573 the printed book. i'm just not sure it needs to be said this way.
574 i'll think about it.}
576 @FIXME{History is more important, and surely more interesting, than
577 actual names. Quoting names without history would be meaningless. FP}
579 Jay Fenlason put together a draft of a @GNUTAR{}
580 manual, borrowing notes from the original man page from John Gilmore.
581 This was withdrawn in version 1.11. Thomas Bushnell, n/BSG and Amy
582 Gorin worked on a tutorial and manual for @GNUTAR{}.
583 Fran@,{c}ois Pinard put version 1.11.8 of the manual together by
584 taking information from all these sources and merging them. Melissa
585 Weisshaus finally edited and redesigned the book to create version
586 1.12. The book for versions from 1.14 up to @value{VERSION} were edited
587 by the current maintainer, Sergey Poznyakoff.
589 For version 1.12, Daniel Hagerty contributed a great deal of technical
590 consulting. In particular, he is the primary author of @ref{Backups}.
592 In July, 2003 @GNUTAR{} was put on CVS at savannah.gnu.org
593 (see @url{http://savannah.gnu.org/projects/tar}), and
594 active development and maintenance work has started
595 again. Currently @GNUTAR{} is being maintained by Paul Eggert, Sergey
596 Poznyakoff and Jeff Bailey.
598 Support for @acronym{POSIX} archives was added by Sergey Poznyakoff.
601 @section Reporting bugs or suggestions
604 @cindex reporting bugs
605 If you find problems or have suggestions about this program or manual,
606 please report them to @file{bug-tar@@gnu.org}.
608 When reporting a bug, please be sure to include as much detail as
609 possible, in order to reproduce it. @FIXME{Be more specific, I'd
610 like to make this node as detailed as 'Bug reporting' node in Emacs
614 @chapter Tutorial Introduction to @command{tar}
616 This chapter guides you through some basic examples of three @command{tar}
617 operations: @option{--create}, @option{--list}, and @option{--extract}. If
618 you already know how to use some other version of @command{tar}, then you
619 may not need to read this chapter. This chapter omits most complicated
620 details about how @command{tar} works.
624 * stylistic conventions::
625 * basic tar options:: Basic @command{tar} Operations and Options
626 * frequent operations::
627 * Two Frequent Options::
628 * create:: How to Create Archives
629 * list:: How to List Archives
630 * extract:: How to Extract Members from an Archive
635 @section Assumptions this Tutorial Makes
637 This chapter is paced to allow beginners to learn about @command{tar}
638 slowly. At the same time, we will try to cover all the basic aspects of
639 these three operations. In order to accomplish both of these tasks, we
640 have made certain assumptions about your knowledge before reading this
641 manual, and the hardware you will be using:
645 Before you start to work through this tutorial, you should understand
646 what the terms ``archive'' and ``archive member'' mean
647 (@pxref{Definitions}). In addition, you should understand something
648 about how Unix-type operating systems work, and you should know how to
649 use some basic utilities. For example, you should know how to create,
650 list, copy, rename, edit, and delete files and directories; how to
651 change between directories; and how to figure out where you are in the
652 file system. You should have some basic understanding of directory
653 structure and how files are named according to which directory they are
654 in. You should understand concepts such as standard output and standard
655 input, what various definitions of the term ``argument'' mean, and the
656 differences between relative and absolute path names. @FIXME{and what
660 This manual assumes that you are working from your own home directory
661 (unless we state otherwise). In this tutorial, you will create a
662 directory to practice @command{tar} commands in. When we show path names,
663 we will assume that those paths are relative to your home directory.
664 For example, my home directory path is @file{/home/fsf/melissa}. All of
665 my examples are in a subdirectory of the directory named by that path
666 name; the subdirectory is called @file{practice}.
669 In general, we show examples of archives which exist on (or can be
670 written to, or worked with from) a directory on a hard disk. In most
671 cases, you could write those archives to, or work with them on any other
672 device, such as a tape drive. However, some of the later examples in
673 the tutorial and next chapter will not work on tape drives.
674 Additionally, working with tapes is much more complicated than working
675 with hard disks. For these reasons, the tutorial does not cover working
676 with tape drives. @xref{Media}, for complete information on using
677 @command{tar} archives with tape drives.
679 @FIXME{this is a cop out. need to add some simple tape drive info.}
682 @node stylistic conventions
683 @section Stylistic Conventions
685 In the examples, @samp{$} represents a typical shell prompt. It
686 precedes lines you should type; to make this more clear, those lines are
687 shown in @kbd{this font}, as opposed to lines which represent the
688 computer's response; those lines are shown in @code{this font}, or
689 sometimes @samp{like this}.
691 @c When we have lines which are too long to be
692 @c displayed in any other way, we will show them like this:
694 @node basic tar options
695 @section Basic @command{tar} Operations and Options
697 @command{tar} can take a wide variety of arguments which specify and define
698 the actions it will have on the particular set of files or the archive.
699 The main types of arguments to @command{tar} fall into one of two classes:
700 operations, and options.
702 Some arguments fall into a class called @dfn{operations}; exactly one of
703 these is both allowed and required for any instance of using @command{tar};
704 you may @emph{not} specify more than one. People sometimes speak of
705 @dfn{operating modes}. You are in a particular operating mode when you
706 have specified the operation which specifies it; there are eight
707 operations in total, and thus there are eight operating modes.
709 The other arguments fall into the class known as @dfn{options}. You are
710 not required to specify any options, and you are allowed to specify more
711 than one at a time (depending on the way you are using @command{tar} at
712 that time). Some options are used so frequently, and are so useful for
713 helping you type commands more carefully that they are effectively
714 ``required''. We will discuss them in this chapter.
716 You can write most of the @command{tar} operations and options in any
717 of three forms: long (mnemonic) form, short form, and old style. Some
718 of the operations and options have no short or ``old'' forms; however,
719 the operations and options which we will cover in this tutorial have
720 corresponding abbreviations. @FIXME{make sure this is still the case,
721 at the end}We will indicate those abbreviations appropriately to get
722 you used to seeing them. (Note that the ``old style'' option forms
723 exist in @GNUTAR{} for compatibility with Unix
724 @command{tar}. In this book we present a full discussion of this way
725 of writing options and operations (@pxref{Old Options}), and we discuss
726 the other two styles of writing options (@xref{Mnemonic Options}, and
727 @pxref{Short Options}).
729 In the examples and in the text of this tutorial, we usually use the
730 long forms of operations and options; but the ``short'' forms produce
731 the same result and can make typing long @command{tar} commands easier.
732 For example, instead of typing
735 @kbd{tar --create --verbose --file=afiles.tar apple angst aspic}
741 @kbd{tar -c -v -f afiles.tar apple angst aspic}
747 @kbd{tar -cvf afiles.tar apple angst aspic}
751 For more information on option syntax, see @ref{Advanced tar}. In
752 discussions in the text, when we name an option by its long form, we
753 also give the corresponding short option in parentheses.
755 The term, ``option'', can be confusing at times, since ``operations''
756 are often lumped in with the actual, @emph{optional} ``options'' in certain
757 general class statements. For example, we just talked about ``short and
758 long forms of options and operations''. However, experienced @command{tar}
759 users often refer to these by shorthand terms such as, ``short and long
760 options''. This term assumes that the ``operations'' are included, also.
761 Context will help you determine which definition of ``options'' to use.
763 Similarly, the term ``command'' can be confusing, as it is often used in
764 two different ways. People sometimes refer to @command{tar} ``commands''.
765 A @command{tar} @dfn{command} is the entire command line of user input
766 which tells @command{tar} what to do --- including the operation, options,
767 and any arguments (file names, pipes, other commands, etc). However,
768 you will also sometimes hear the term ``the @command{tar} command''. When
769 the word ``command'' is used specifically like this, a person is usually
770 referring to the @command{tar} @emph{operation}, not the whole line.
771 Again, use context to figure out which of the meanings the speaker
774 @node frequent operations
775 @section The Three Most Frequently Used Operations
777 Here are the three most frequently used operations (both short and long
778 forms), as well as a brief description of their meanings. The rest of
779 this chapter will cover how to use these operations in detail. We will
780 present the rest of the operations in the next chapter.
785 Create a new @command{tar} archive.
788 List the contents of an archive.
791 Extract one or more members from an archive.
794 @node Two Frequent Options
795 @section Two Frequently Used Options
797 To understand how to run @command{tar} in the three operating modes listed
798 previously, you also need to understand how to use two of the options to
799 @command{tar}: @option{--file} (which takes an archive file as an argument)
800 and @option{--verbose}. (You are usually not @emph{required} to specify
801 either of these options when you run @command{tar}, but they can be very
802 useful in making things more clear and helping you avoid errors.)
811 @unnumberedsubsec The @option{--file} Option
814 @opindex file, tutorial
815 @item --file=@var{archive-name}
816 @itemx -f @var{archive-name}
817 Specify the name of an archive file.
820 You can specify an argument for the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option whenever you
821 use @command{tar}; this option determines the name of the archive file
822 that @command{tar} will work on.
825 If you don't specify this argument, then @command{tar} will examine
826 the environment variable @env{TAPE}. If it is set, its value will be
827 used as the archive name. Otherwise, @command{tar} will use the
828 default archive, determined at the compile time. Usually it is
829 standard output or some physical tape drive attached to your machine
830 (you can verify what the default is by running @kbd{tar
831 --show-defaults}, @pxref{defaults}). If there is no tape drive
832 attached, or the default is not meaningful, then @command{tar} will
833 print an error message. The error message might look roughly like one
837 tar: can't open /dev/rmt8 : No such device or address
838 tar: can't open /dev/rsmt0 : I/O error
842 To avoid confusion, we recommend that you always specify an archive file
843 name by using @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) when writing your @command{tar} commands.
844 For more information on using the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option, see
847 @node verbose tutorial
848 @unnumberedsubsec The @option{--verbose} Option
851 @opindex verbose, introduced
854 Show the files being worked on as @command{tar} is running.
857 @option{--verbose} (@option{-v}) shows details about the results of running
858 @command{tar}. This can be especially useful when the results might not be
859 obvious. For example, if you want to see the progress of @command{tar} as
860 it writes files into the archive, you can use the @option{--verbose}
861 option. In the beginning, you may find it useful to use
862 @option{--verbose} at all times; when you are more accustomed to
863 @command{tar}, you will likely want to use it at certain times but not at
864 others. We will use @option{--verbose} at times to help make something
865 clear, and we will give many examples both using and not using
866 @option{--verbose} to show the differences.
868 Each instance of @option{--verbose} on the command line increases the
869 verbosity level by one, so if you need more details on the output,
872 When reading archives (@option{--list}, @option{--extract},
873 @option{--diff}), @command{tar} by default prints only the names of
874 the members being extracted. Using @option{--verbose} will show a full,
875 @command{ls} style member listing.
877 In contrast, when writing archives (@option{--create}, @option{--append},
878 @option{--update}), @command{tar} does not print file names by
879 default. So, a single @option{--verbose} option shows the file names
880 being added to the archive, while two @option{--verbose} options
881 enable the full listing.
883 For example, to create an archive in verbose mode:
886 $ @kbd{tar -cvf afiles.tar apple angst aspic}
893 Creating the same archive with the verbosity level 2 could give:
896 $ @kbd{tar -cvvf afiles.tar apple angst aspic}
897 -rw-r--r-- gray/staff 62373 2006-06-09 12:06 apple
898 -rw-r--r-- gray/staff 11481 2006-06-09 12:06 angst
899 -rw-r--r-- gray/staff 23152 2006-06-09 12:06 aspic
903 This works equally well using short or long forms of options. Using
904 long forms, you would simply write out the mnemonic form of the option
908 $ @kbd{tar --create --verbose --verbose @dots{}}
912 Note that you must double the hyphens properly each time.
914 Later in the tutorial, we will give examples using @w{@option{--verbose
917 The full output consists of six fields:
920 @item File type and permissions in symbolic form.
921 These are displayed in the same format as the first column of
922 @command{ls -l} output (@pxref{What information is listed,
923 format=verbose, Verbose listing, fileutils, GNU file utilities}).
925 @item Owner name and group separated by a slash character.
926 If these data are not available (for example, when listing a @samp{v7} format
927 archive), numeric ID values are printed instead.
929 @item Size of the file, in bytes.
931 @item File modification date in ISO 8601 format.
933 @item File modification time.
936 If the name contains any special characters (white space, newlines,
937 etc.) these are displayed in an unambiguous form using so called
938 @dfn{quoting style}. For the detailed discussion of available styles
939 and on how to use them, see @ref{quoting styles}.
941 Depending on the file type, the name can be followed by some
942 additional information, described in the following table:
945 @item -> @var{link-name}
946 The file or archive member is a @dfn{symbolic link} and
947 @var{link-name} is the name of file it links to.
949 @item link to @var{link-name}
950 The file or archive member is a @dfn{hard link} and @var{link-name} is
951 the name of file it links to.
954 The archive member is an old GNU format long link. You will normally
958 The archive member is an old GNU format long name. You will normally
961 @item --Volume Header--
962 The archive member is a GNU @dfn{volume header} (@pxref{Tape Files}).
964 @item --Continued at byte @var{n}--
965 Encountered only at the beginning of a multy-volume archive
966 (@pxref{Using Multiple Tapes}). This archive member is a continuation
967 from the previous volume. The number @var{n} gives the offset where
968 the original file was split.
970 @item --Mangled file names--
971 This archive member contains @dfn{mangled file names} declarations,
972 a special member type that was used by early versions of @GNUTAR{}.
973 You probably will never encounter this, unless you are reading a very
976 @item unknown file type @var{c}
977 An archive member of unknown type. @var{c} is the type character from
978 the archive header. If you encounter such a message, it means that
979 either your archive contains proprietary member types @GNUTAR{} is not
980 able to handle, or the archive is corrupted.
985 For example, here is an archive listing containing most of the special
986 suffixes explained above:
990 V--------- 0/0 1536 2006-06-09 13:07 MyVolume--Volume Header--
991 -rw-r--r-- gray/staff 456783 2006-06-09 12:06 aspic--Continued at
993 -rw-r--r-- gray/staff 62373 2006-06-09 12:06 apple
994 lrwxrwxrwx gray/staff 0 2006-06-09 13:01 angst -> apple
995 -rw-r--r-- gray/staff 35793 2006-06-09 12:06 blues
996 hrw-r--r-- gray/staff 0 2006-06-09 12:06 music link to blues
1004 @unnumberedsubsec Getting Help: Using the @option{--help} Option
1010 The @option{--help} option to @command{tar} prints out a very brief list of
1011 all operations and option available for the current version of
1012 @command{tar} available on your system.
1016 @section How to Create Archives
1019 @cindex Creation of the archive
1020 @cindex Archive, creation of
1021 One of the basic operations of @command{tar} is @option{--create} (@option{-c}), which
1022 you use to create a @command{tar} archive. We will explain
1023 @option{--create} first because, in order to learn about the other
1024 operations, you will find it useful to have an archive available to
1027 To make this easier, in this section you will first create a directory
1028 containing three files. Then, we will show you how to create an
1029 @emph{archive} (inside the new directory). Both the directory, and
1030 the archive are specifically for you to practice on. The rest of this
1031 chapter and the next chapter will show many examples using this
1032 directory and the files you will create: some of those files may be
1033 other directories and other archives.
1035 The three files you will archive in this example are called
1036 @file{blues}, @file{folk}, and @file{jazz}. The archive is called
1037 @file{collection.tar}.
1039 This section will proceed slowly, detailing how to use @option{--create}
1040 in @code{verbose} mode, and showing examples using both short and long
1041 forms. In the rest of the tutorial, and in the examples in the next
1042 chapter, we will proceed at a slightly quicker pace. This section
1043 moves more slowly to allow beginning users to understand how
1044 @command{tar} works.
1047 * prepare for examples::
1048 * Creating the archive::
1054 @node prepare for examples
1055 @subsection Preparing a Practice Directory for Examples
1057 To follow along with this and future examples, create a new directory
1058 called @file{practice} containing files called @file{blues}, @file{folk}
1059 and @file{jazz}. The files can contain any information you like:
1060 ideally, they should contain information which relates to their names,
1061 and be of different lengths. Our examples assume that @file{practice}
1062 is a subdirectory of your home directory.
1064 Now @command{cd} to the directory named @file{practice}; @file{practice}
1065 is now your @dfn{working directory}. (@emph{Please note}: Although
1066 the full path name of this directory is
1067 @file{/@var{homedir}/practice}, in our examples we will refer to
1068 this directory as @file{practice}; the @var{homedir} is presumed.
1070 In general, you should check that the files to be archived exist where
1071 you think they do (in the working directory) by running @command{ls}.
1072 Because you just created the directory and the files and have changed to
1073 that directory, you probably don't need to do that this time.
1075 It is very important to make sure there isn't already a file in the
1076 working directory with the archive name you intend to use (in this case,
1077 @samp{collection.tar}), or that you don't care about its contents.
1078 Whenever you use @samp{create}, @command{tar} will erase the current
1079 contents of the file named by @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) if it exists. @command{tar}
1080 will not tell you if you are about to overwrite an archive unless you
1081 specify an option which does this (@pxref{backup}, for the
1082 information on how to do so). To add files to an existing archive,
1083 you need to use a different option, such as @option{--append} (@option{-r}); see
1084 @ref{append} for information on how to do this.
1086 @node Creating the archive
1087 @subsection Creating the Archive
1089 @opindex create, introduced
1090 To place the files @file{blues}, @file{folk}, and @file{jazz} into an
1091 archive named @file{collection.tar}, use the following command:
1094 $ @kbd{tar --create --file=collection.tar blues folk jazz}
1097 The order of the arguments is not very important, @emph{when using long
1098 option forms}. You could also say:
1101 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1105 However, you can see that this order is harder to understand; this is
1106 why we will list the arguments in the order that makes the commands
1107 easiest to understand (and we encourage you to do the same when you use
1108 @command{tar}, to avoid errors).
1110 Note that the part of the command which says,
1111 @w{@option{--file=collection.tar}} is considered to be @emph{one} argument.
1112 If you substituted any other string of characters for
1113 @kbd{collection.tar}, then that string would become the name of the
1114 archive file you create.
1116 The order of the options becomes more important when you begin to use
1117 short forms. With short forms, if you type commands in the wrong order
1118 (even if you type them correctly in all other ways), you may end up with
1119 results you don't expect. For this reason, it is a good idea to get
1120 into the habit of typing options in the order that makes inherent sense.
1121 @xref{short create}, for more information on this.
1123 In this example, you type the command as shown above: @option{--create}
1124 is the operation which creates the new archive
1125 (@file{collection.tar}), and @option{--file} is the option which lets
1126 you give it the name you chose. The files, @file{blues}, @file{folk},
1127 and @file{jazz}, are now members of the archive, @file{collection.tar}
1128 (they are @dfn{file name arguments} to the @option{--create} operation.
1129 @xref{Choosing}, for the detailed discussion on these.) Now that they are
1130 in the archive, they are called @emph{archive members}, not files.
1131 (@pxref{Definitions,members}).
1133 When you create an archive, you @emph{must} specify which files you
1134 want placed in the archive. If you do not specify any archive
1135 members, @GNUTAR{} will complain.
1137 If you now list the contents of the working directory (@command{ls}), you will
1138 find the archive file listed as well as the files you saw previously:
1141 blues folk jazz collection.tar
1145 Creating the archive @samp{collection.tar} did not destroy the copies of
1146 the files in the directory.
1148 Keep in mind that if you don't indicate an operation, @command{tar} will not
1149 run and will prompt you for one. If you don't name any files, @command{tar}
1150 will complain. You must have write access to the working directory,
1151 or else you will not be able to create an archive in that directory.
1153 @emph{Caution}: Do not attempt to use @option{--create} (@option{-c}) to add files to
1154 an existing archive; it will delete the archive and write a new one.
1155 Use @option{--append} (@option{-r}) instead. @xref{append}.
1157 @node create verbose
1158 @subsection Running @option{--create} with @option{--verbose}
1160 @opindex create, using with @option{--verbose}
1161 @opindex verbose, using with @option{--create}
1162 If you include the @option{--verbose} (@option{-v}) option on the command line,
1163 @command{tar} will list the files it is acting on as it is working. In
1164 verbose mode, the @code{create} example above would appear as:
1167 $ @kbd{tar --create --verbose --file=collection.tar blues folk jazz}
1173 This example is just like the example we showed which did not use
1174 @option{--verbose}, except that @command{tar} generated the remaining lines
1176 (note the different font styles).
1182 In the rest of the examples in this chapter, we will frequently use
1183 @code{verbose} mode so we can show actions or @command{tar} responses that
1184 you would otherwise not see, and which are important for you to
1188 @subsection Short Forms with @samp{create}
1190 As we said before, the @option{--create} (@option{-c}) operation is one of the most
1191 basic uses of @command{tar}, and you will use it countless times.
1192 Eventually, you will probably want to use abbreviated (or ``short'')
1193 forms of options. A full discussion of the three different forms that
1194 options can take appears in @ref{Styles}; for now, here is what the
1195 previous example (including the @option{--verbose} (@option{-v}) option) looks like
1196 using short option forms:
1199 $ @kbd{tar -cvf collection.tar blues folk jazz}
1206 As you can see, the system responds the same no matter whether you use
1207 long or short option forms.
1209 @FIXME{i don't like how this is worded:} One difference between using
1210 short and long option forms is that, although the exact placement of
1211 arguments following options is no more specific when using short forms,
1212 it is easier to become confused and make a mistake when using short
1213 forms. For example, suppose you attempted the above example in the
1217 $ @kbd{tar -cfv collection.tar blues folk jazz}
1221 In this case, @command{tar} will make an archive file called @file{v},
1222 containing the files @file{blues}, @file{folk}, and @file{jazz}, because
1223 the @samp{v} is the closest ``file name'' to the @option{-f} option, and
1224 is thus taken to be the chosen archive file name. @command{tar} will try
1225 to add a file called @file{collection.tar} to the @file{v} archive file;
1226 if the file @file{collection.tar} did not already exist, @command{tar} will
1227 report an error indicating that this file does not exist. If the file
1228 @file{collection.tar} does already exist (e.g., from a previous command
1229 you may have run), then @command{tar} will add this file to the archive.
1230 Because the @option{-v} option did not get registered, @command{tar} will not
1231 run under @samp{verbose} mode, and will not report its progress.
1233 The end result is that you may be quite confused about what happened,
1234 and possibly overwrite a file. To illustrate this further, we will show
1235 you how an example we showed previously would look using short forms.
1240 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1244 is confusing as it is. When shown using short forms, however, it
1245 becomes much more so:
1248 $ @kbd{tar blues -c folk -f collection.tar jazz}
1252 It would be very easy to put the wrong string of characters
1253 immediately following the @option{-f}, but doing that could sacrifice
1256 For this reason, we recommend that you pay very careful attention to
1257 the order of options and placement of file and archive names,
1258 especially when using short option forms. Not having the option name
1259 written out mnemonically can affect how well you remember which option
1260 does what, and therefore where different names have to be placed.
1263 @subsection Archiving Directories
1265 @cindex Archiving Directories
1266 @cindex Directories, Archiving
1267 You can archive a directory by specifying its directory name as a
1268 file name argument to @command{tar}. The files in the directory will be
1269 archived relative to the working directory, and the directory will be
1270 re-created along with its contents when the archive is extracted.
1272 To archive a directory, first move to its superior directory. If you
1273 have followed the previous instructions in this tutorial, you should
1282 This will put you into the directory which contains @file{practice},
1283 i.e., your home directory. Once in the superior directory, you can
1284 specify the subdirectory, @file{practice}, as a file name argument. To
1285 store @file{practice} in the new archive file @file{music.tar}, type:
1288 $ @kbd{tar --create --verbose --file=music.tar practice}
1292 @command{tar} should output:
1299 practice/collection.tar
1302 Note that the archive thus created is not in the subdirectory
1303 @file{practice}, but rather in the current working directory---the
1304 directory from which @command{tar} was invoked. Before trying to archive a
1305 directory from its superior directory, you should make sure you have
1306 write access to the superior directory itself, not only the directory
1307 you are trying archive with @command{tar}. For example, you will probably
1308 not be able to store your home directory in an archive by invoking
1309 @command{tar} from the root directory; @xref{absolute}. (Note
1310 also that @file{collection.tar}, the original archive file, has itself
1311 been archived. @command{tar} will accept any file as a file to be
1312 archived, regardless of its content. When @file{music.tar} is
1313 extracted, the archive file @file{collection.tar} will be re-written
1314 into the file system).
1316 If you give @command{tar} a command such as
1319 $ @kbd{tar --create --file=foo.tar .}
1323 @command{tar} will report @samp{tar: ./foo.tar is the archive; not
1324 dumped}. This happens because @command{tar} creates the archive
1325 @file{foo.tar} in the current directory before putting any files into
1326 it. Then, when @command{tar} attempts to add all the files in the
1327 directory @file{.} to the archive, it notices that the file
1328 @file{./foo.tar} is the same as the archive @file{foo.tar}, and skips
1329 it. (It makes no sense to put an archive into itself.) @GNUTAR{}
1330 will continue in this case, and create the archive
1331 normally, except for the exclusion of that one file. (@emph{Please
1332 note:} Other implementations of @command{tar} may not be so clever;
1333 they will enter an infinite loop when this happens, so you should not
1334 depend on this behavior unless you are certain you are running
1335 @GNUTAR{}. In general, it is wise to always place the archive outside
1336 of the directory being dumped.
1339 @section How to List Archives
1342 Frequently, you will find yourself wanting to determine exactly what a
1343 particular archive contains. You can use the @option{--list} (@option{-t}) operation
1344 to get the member names as they currently appear in the archive, as well
1345 as various attributes of the files at the time they were archived. For
1346 example, you can examine the archive @file{collection.tar} that you
1347 created in the last section with the command,
1350 $ @kbd{tar --list --file=collection.tar}
1354 The output of @command{tar} would then be:
1363 The archive @file{bfiles.tar} would list as follows:
1372 Be sure to use a @option{--file=@var{archive-name}} (@option{-f
1373 @var{archive-name}}) option just as with @option{--create}
1374 (@option{-c}) to specify the name of the archive.
1376 @opindex list, using with @option{--verbose}
1377 @opindex verbose, using with @option{--list}
1378 If you use the @option{--verbose} (@option{-v}) option with
1379 @option{--list}, then @command{tar} will print out a listing
1380 reminiscent of @w{@samp{ls -l}}, showing owner, file size, and so forth.
1382 If you had used @option{--verbose} (@option{-v}) mode, the example
1383 above would look like:
1386 $ @kbd{tar --list --verbose --file=collection.tar folk}
1387 -rw-r--r-- myself user 62 1990-05-23 10:55 folk
1390 @cindex listing member and file names
1391 @anchor{listing member and file names}
1392 It is important to notice that the output of @kbd{tar --list
1393 --verbose} does not necessarily match that produced by @kbd{tar
1394 --create --verbose} while creating the archive. It is because
1395 @GNUTAR{}, unless told explicitly not to do so, removes some directory
1396 prefixes from file names before storing them in the archive
1397 (@xref{absolute}, for more information). In other
1398 words, in verbose mode @GNUTAR{} shows @dfn{file names} when creating
1399 an archive and @dfn{member names} when listing it. Consider this
1404 $ @kbd{tar cfv archive /etc/mail}
1405 tar: Removing leading `/' from member names
1407 /etc/mail/sendmail.cf
1409 $ @kbd{tar tf archive}
1411 etc/mail/sendmail.cf
1416 @opindex show-stored-names
1417 This default behavior can sometimes be inconvenient. You can force
1418 @GNUTAR{} show member names when creating archive by supplying
1419 @option{--show-stored-names} option.
1422 @item --show-stored-names
1423 Print member (as opposed to @emph{file}) names when creating the archive.
1426 @cindex File name arguments, using @option{--list} with
1427 @opindex list, using with file name arguments
1428 You can specify one or more individual member names as arguments when
1429 using @samp{list}. In this case, @command{tar} will only list the
1430 names of members you identify. For example, @w{@kbd{tar --list
1431 --file=afiles.tar apple}} would only print @file{apple}.
1433 Because @command{tar} preserves paths, file names must be specified as
1434 they appear in the archive (ie., relative to the directory from which
1435 the archive was created). Therefore, it is essential when specifying
1436 member names to @command{tar} that you give the exact member names.
1437 For example, @w{@kbd{tar --list --file=bfiles.tar birds}} would produce an
1438 error message something like @samp{tar: birds: Not found in archive},
1439 because there is no member named @file{birds}, only one named
1440 @file{./birds}. While the names @file{birds} and @file{./birds} name
1441 the same file, @emph{member} names by default are compared verbatim.
1443 However, @w{@kbd{tar --list --file=bfiles.tar baboon}} would respond
1444 with @file{baboon}, because this exact member name is in the archive file
1445 @file{bfiles.tar}. If you are not sure of the exact file name,
1446 use @dfn{globbing patterns}, for example:
1449 $ @kbd{tar --list --file=bfiles.tar --wildcards '*b*'}
1453 will list all members whose name contains @samp{b}. @xref{wildcards},
1454 for a detailed discussion of globbing patterns and related
1455 @command{tar} command line options.
1462 @unnumberedsubsec Listing the Contents of a Stored Directory
1464 To get information about the contents of an archived directory,
1465 use the directory name as a file name argument in conjunction with
1466 @option{--list} (@option{-t}). To find out file attributes, include the
1467 @option{--verbose} (@option{-v}) option.
1469 For example, to find out about files in the directory @file{practice}, in
1470 the archive file @file{music.tar}, type:
1473 $ @kbd{tar --list --verbose --file=music.tar practice}
1476 @command{tar} responds:
1479 drwxrwxrwx myself user 0 1990-05-31 21:49 practice/
1480 -rw-r--r-- myself user 42 1990-05-21 13:29 practice/blues
1481 -rw-r--r-- myself user 62 1990-05-23 10:55 practice/folk
1482 -rw-r--r-- myself user 40 1990-05-21 13:30 practice/jazz
1483 -rw-r--r-- myself user 10240 1990-05-31 21:49 practice/collection.tar
1486 When you use a directory name as a file name argument, @command{tar} acts on
1487 all the files (including sub-directories) in that directory.
1490 @section How to Extract Members from an Archive
1493 @cindex Retrieving files from an archive
1494 @cindex Resurrecting files from an archive
1497 Creating an archive is only half the job---there is no point in storing
1498 files in an archive if you can't retrieve them. The act of retrieving
1499 members from an archive so they can be used and manipulated as
1500 unarchived files again is called @dfn{extraction}. To extract files
1501 from an archive, use the @option{--extract} (@option{--get} or
1502 @option{-x}) operation. As with @option{--create}, specify the name
1503 of the archive with @option{--file} (@option{-f}) option. Extracting
1504 an archive does not modify the archive in any way; you can extract it
1505 multiple times if you want or need to.
1507 Using @option{--extract}, you can extract an entire archive, or specific
1508 files. The files can be directories containing other files, or not. As
1509 with @option{--create} (@option{-c}) and @option{--list} (@option{-t}), you may use the short or the
1510 long form of the operation without affecting the performance.
1513 * extracting archives::
1514 * extracting files::
1516 * extracting untrusted archives::
1517 * failing commands::
1520 @node extracting archives
1521 @subsection Extracting an Entire Archive
1523 To extract an entire archive, specify the archive file name only, with
1524 no individual file names as arguments. For example,
1527 $ @kbd{tar -xvf collection.tar}
1534 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
1535 -rw-r--r-- me user 21 1996-09-23 16:44 blues
1536 -rw-r--r-- me user 20 1996-09-23 16:44 folk
1539 @node extracting files
1540 @subsection Extracting Specific Files
1542 To extract specific archive members, give their exact member names as
1543 arguments, as printed by @option{--list} (@option{-t}). If you had
1544 mistakenly deleted one of the files you had placed in the archive
1545 @file{collection.tar} earlier (say, @file{blues}), you can extract it
1546 from the archive without changing the archive's structure. Its
1547 contents will be identical to the original file @file{blues} that you
1550 First, make sure you are in the @file{practice} directory, and list the
1551 files in the directory. Now, delete the file, @samp{blues}, and list
1552 the files in the directory again.
1554 You can now extract the member @file{blues} from the archive file
1555 @file{collection.tar} like this:
1558 $ @kbd{tar --extract --file=collection.tar blues}
1562 If you list the files in the directory again, you will see that the file
1563 @file{blues} has been restored, with its original permissions, data
1564 modification times, and owner.@footnote{This is only accidentally
1565 true, but not in general. Whereas modification times are always
1566 restored, in most cases, one has to be root for restoring the owner,
1567 and use a special option for restoring permissions. Here, it just
1568 happens that the restoring user is also the owner of the archived
1569 members, and that the current @code{umask} is compatible with original
1570 permissions.} (These parameters will be identical to those which
1571 the file had when you originally placed it in the archive; any changes
1572 you may have made before deleting the file from the file system,
1573 however, will @emph{not} have been made to the archive member.) The
1574 archive file, @samp{collection.tar}, is the same as it was before you
1575 extracted @samp{blues}. You can confirm this by running @command{tar} with
1576 @option{--list} (@option{-t}).
1578 Remember that as with other operations, specifying the exact member
1579 name is important. @w{@kbd{tar --extract --file=bfiles.tar birds}}
1580 will fail, because there is no member named @file{birds}. To extract
1581 the member named @file{./birds}, you must specify @w{@kbd{tar
1582 --extract --file=bfiles.tar ./birds}}. If you don't remember the
1583 exact member names, use @option{--list} (@option{-t}) option
1584 (@pxref{list}). You can also extract those members that match a
1585 specific @dfn{globbing pattern}. For example, to extract from
1586 @file{bfiles.tar} all files that begin with @samp{b}, no matter their
1587 directory prefix, you could type:
1590 $ @kbd{tar -x -f bfiles.tar --wildcards --no-anchored 'b*'}
1594 Here, @option{--wildcards} instructs @command{tar} to treat
1595 command line arguments as globbing patterns and @option{--no-anchored}
1596 informs it that the patterns apply to member names after any @samp{/}
1597 delimiter. The use of globbing patterns is discussed in detail in
1600 You can extract a file to standard output by combining the above options
1601 with the @option{--to-stdout} (@option{-O}) option (@pxref{Writing to Standard
1604 If you give the @option{--verbose} option, then @option{--extract}
1605 will print the names of the archive members as it extracts them.
1608 @subsection Extracting Files that are Directories
1610 Extracting directories which are members of an archive is similar to
1611 extracting other files. The main difference to be aware of is that if
1612 the extracted directory has the same name as any directory already in
1613 the working directory, then files in the extracted directory will be
1614 placed into the directory of the same name. Likewise, if there are
1615 files in the pre-existing directory with the same names as the members
1616 which you extract, the files from the extracted archive will replace
1617 the files already in the working directory (and possible
1618 subdirectories). This will happen regardless of whether or not the
1619 files in the working directory were more recent than those extracted
1620 (there exist, however, special options that alter this behavior
1623 However, if a file was stored with a directory name as part of its file
1624 name, and that directory does not exist under the working directory when
1625 the file is extracted, @command{tar} will create the directory.
1627 We can demonstrate how to use @option{--extract} to extract a directory
1628 file with an example. Change to the @file{practice} directory if you
1629 weren't there, and remove the files @file{folk} and @file{jazz}. Then,
1630 go back to the parent directory and extract the archive
1631 @file{music.tar}. You may either extract the entire archive, or you may
1632 extract only the files you just deleted. To extract the entire archive,
1633 don't give any file names as arguments after the archive name
1634 @file{music.tar}. To extract only the files you deleted, use the
1638 $ @kbd{tar -xvf music.tar practice/folk practice/jazz}
1644 If you were to specify two @option{--verbose} (@option{-v}) options, @command{tar}
1645 would have displayed more detail about the extracted files, as shown
1646 in the example below:
1649 $ @kbd{tar -xvvf music.tar practice/folk practice/jazz}
1650 -rw-r--r-- me user 28 1996-10-18 16:31 practice/jazz
1651 -rw-r--r-- me user 20 1996-09-23 16:44 practice/folk
1655 Because you created the directory with @file{practice} as part of the
1656 file names of each of the files by archiving the @file{practice}
1657 directory as @file{practice}, you must give @file{practice} as part
1658 of the file names when you extract those files from the archive.
1660 @node extracting untrusted archives
1661 @subsection Extracting Archives from Untrusted Sources
1663 Extracting files from archives can overwrite files that already exist.
1664 If you receive an archive from an untrusted source, you should make a
1665 new directory and extract into that directory, so that you don't have
1666 to worry about the extraction overwriting one of your existing files.
1667 For example, if @file{untrusted.tar} came from somewhere else on the
1668 Internet, and you don't necessarily trust its contents, you can
1669 extract it as follows:
1672 $ @kbd{mkdir newdir}
1674 $ @kbd{tar -xvf ../untrusted.tar}
1677 It is also a good practice to examine contents of the archive
1678 before extracting it, using @option{--list} (@option{-t}) option, possibly combined
1679 with @option{--verbose} (@option{-v}).
1681 @node failing commands
1682 @subsection Commands That Will Fail
1684 Here are some sample commands you might try which will not work, and why
1687 If you try to use this command,
1690 $ @kbd{tar -xvf music.tar folk jazz}
1694 you will get the following response:
1697 tar: folk: Not found in archive
1698 tar: jazz: Not found in archive
1703 This is because these files were not originally @emph{in} the parent
1704 directory @file{..}, where the archive is located; they were in the
1705 @file{practice} directory, and their file names reflect this:
1708 $ @kbd{tar -tvf music.tar}
1714 @FIXME{make sure the above works when going through the examples in
1718 Likewise, if you try to use this command,
1721 $ @kbd{tar -tvf music.tar folk jazz}
1725 you would get a similar response. Members with those names are not in the
1726 archive. You must use the correct member names, or wildcards, in order
1727 to extract the files from the archive.
1729 If you have forgotten the correct names of the files in the archive,
1730 use @w{@kbd{tar --list --verbose}} to list them correctly.
1732 @FIXME{more examples, here? hag thinks it's a good idea.}
1735 @section Going Further Ahead in this Manual
1737 @FIXME{need to write up a node here about the things that are going to
1738 be in the rest of the manual.}
1740 @node tar invocation
1741 @chapter Invoking @GNUTAR{}
1744 This chapter is about how one invokes the @GNUTAR{}
1745 command, from the command synopsis (@pxref{Synopsis}). There are
1746 numerous options, and many styles for writing them. One mandatory
1747 option specifies the operation @command{tar} should perform
1748 (@pxref{Operation Summary}), other options are meant to detail how
1749 this operation should be performed (@pxref{Option Summary}).
1750 Non-option arguments are not always interpreted the same way,
1751 depending on what the operation is.
1753 You will find in this chapter everything about option styles and rules for
1754 writing them (@pxref{Styles}). On the other hand, operations and options
1755 are fully described elsewhere, in other chapters. Here, you will find
1756 only synthetic descriptions for operations and options, together with
1757 pointers to other parts of the @command{tar} manual.
1759 Some options are so special they are fully described right in this
1760 chapter. They have the effect of inhibiting the normal operation of
1761 @command{tar} or else, they globally alter the amount of feedback the user
1762 receives about what is going on. These are the @option{--help} and
1763 @option{--version} (@pxref{help}), @option{--verbose} (@pxref{verbose})
1764 and @option{--interactive} options (@pxref{interactive}).
1768 * using tar options::
1778 @section General Synopsis of @command{tar}
1780 The @GNUTAR{} program is invoked as either one of:
1783 @kbd{tar @var{option}@dots{} [@var{name}]@dots{}}
1784 @kbd{tar @var{letter}@dots{} [@var{argument}]@dots{} [@var{option}]@dots{} [@var{name}]@dots{}}
1787 The second form is for when old options are being used.
1789 You can use @command{tar} to store files in an archive, to extract them from
1790 an archive, and to do other types of archive manipulation. The primary
1791 argument to @command{tar}, which is called the @dfn{operation}, specifies
1792 which action to take. The other arguments to @command{tar} are either
1793 @dfn{options}, which change the way @command{tar} performs an operation,
1794 or file names or archive members, which specify the files or members
1795 @command{tar} is to act on.
1797 You can actually type in arguments in any order, even if in this manual
1798 the options always precede the other arguments, to make examples easier
1799 to understand. Further, the option stating the main operation mode
1800 (the @command{tar} main command) is usually given first.
1802 Each @var{name} in the synopsis above is interpreted as an archive member
1803 name when the main command is one of @option{--compare}
1804 (@option{--diff}, @option{-d}), @option{--delete}, @option{--extract}
1805 (@option{--get}, @option{-x}), @option{--list} (@option{-t}) or
1806 @option{--update} (@option{-u}). When naming archive members, you
1807 must give the exact name of the member in the archive, as it is
1808 printed by @option{--list}. For @option{--append} (@option{-r}) and
1809 @option{--create} (@option{-c}), these @var{name} arguments specify
1810 the names of either files or directory hierarchies to place in the archive.
1811 These files or hierarchies should already exist in the file system,
1812 prior to the execution of the @command{tar} command.
1814 @command{tar} interprets relative file names as being relative to the
1815 working directory. @command{tar} will make all file names relative
1816 (by removing leading slashes when archiving or restoring files),
1817 unless you specify otherwise (using the @option{--absolute-names}
1818 option). @xref{absolute}, for more information about
1819 @option{--absolute-names}.
1821 If you give the name of a directory as either a file name or a member
1822 name, then @command{tar} acts recursively on all the files and directories
1823 beneath that directory. For example, the name @file{/} identifies all
1824 the files in the file system to @command{tar}.
1826 The distinction between file names and archive member names is especially
1827 important when shell globbing is used, and sometimes a source of confusion
1828 for newcomers. @xref{wildcards}, for more information about globbing.
1829 The problem is that shells may only glob using existing files in the
1830 file system. Only @command{tar} itself may glob on archive members, so when
1831 needed, you must ensure that wildcard characters reach @command{tar} without
1832 being interpreted by the shell first. Using a backslash before @samp{*}
1833 or @samp{?}, or putting the whole argument between quotes, is usually
1834 sufficient for this.
1836 Even if @var{name}s are often specified on the command line, they
1837 can also be read from a text file in the file system, using the
1838 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}}) option.
1840 If you don't use any file name arguments, @option{--append} (@option{-r}),
1841 @option{--delete} and @option{--concatenate} (@option{--catenate},
1842 @option{-A}) will do nothing, while @option{--create} (@option{-c})
1843 will usually yield a diagnostic and inhibit @command{tar} execution.
1844 The other operations of @command{tar} (@option{--list},
1845 @option{--extract}, @option{--compare}, and @option{--update})
1846 will act on the entire contents of the archive.
1849 @cindex return status
1850 Besides successful exits, @GNUTAR{} may fail for
1851 many reasons. Some reasons correspond to bad usage, that is, when the
1852 @command{tar} command is improperly written. Errors may be
1853 encountered later, while encountering an error processing the archive
1854 or the files. Some errors are recoverable, in which case the failure
1855 is delayed until @command{tar} has completed all its work. Some
1856 errors are such that it would not meaningful, or at least risky, to
1857 continue processing: @command{tar} then aborts processing immediately.
1858 All abnormal exits, whether immediate or delayed, should always be
1859 clearly diagnosed on @code{stderr}, after a line stating the nature of
1862 @GNUTAR{} returns only a few exit statuses. I'm really
1863 aiming simplicity in that area, for now. If you are not using the
1864 @option{--compare} @option{--diff}, @option{-d}) option, zero means
1865 that everything went well, besides maybe innocuous warnings. Nonzero
1866 means that something went wrong. Right now, as of today, ``nonzero''
1867 is almost always 2, except for remote operations, where it may be
1870 @node using tar options
1871 @section Using @command{tar} Options
1873 @GNUTAR{} has a total of eight operating modes which
1874 allow you to perform a variety of tasks. You are required to choose
1875 one operating mode each time you employ the @command{tar} program by
1876 specifying one, and only one operation as an argument to the
1877 @command{tar} command (two lists of four operations each may be found
1878 at @ref{frequent operations} and @ref{Operations}). Depending on
1879 circumstances, you may also wish to customize how the chosen operating
1880 mode behaves. For example, you may wish to change the way the output
1881 looks, or the format of the files that you wish to archive may require
1882 you to do something special in order to make the archive look right.
1884 You can customize and control @command{tar}'s performance by running
1885 @command{tar} with one or more options (such as @option{--verbose}
1886 (@option{-v}), which we used in the tutorial). As we said in the
1887 tutorial, @dfn{options} are arguments to @command{tar} which are (as
1888 their name suggests) optional. Depending on the operating mode, you
1889 may specify one or more options. Different options will have different
1890 effects, but in general they all change details of the operation, such
1891 as archive format, archive name, or level of user interaction. Some
1892 options make sense with all operating modes, while others are
1893 meaningful only with particular modes. You will likely use some
1894 options frequently, while you will only use others infrequently, or
1895 not at all. (A full list of options is available in @pxref{All Options}.)
1897 @vrindex TAR_OPTIONS, environment variable
1898 @anchor{TAR_OPTIONS}
1899 The @env{TAR_OPTIONS} environment variable specifies default options to
1900 be placed in front of any explicit options. For example, if
1901 @code{TAR_OPTIONS} is @samp{-v --unlink-first}, @command{tar} behaves as
1902 if the two options @option{-v} and @option{--unlink-first} had been
1903 specified before any explicit options. Option specifications are
1904 separated by whitespace. A backslash escapes the next character, so it
1905 can be used to specify an option containing whitespace or a backslash.
1907 Note that @command{tar} options are case sensitive. For example, the
1908 options @option{-T} and @option{-t} are different; the first requires an
1909 argument for stating the name of a file providing a list of @var{name}s,
1910 while the second does not require an argument and is another way to
1911 write @option{--list} (@option{-t}).
1913 In addition to the eight operations, there are many options to
1914 @command{tar}, and three different styles for writing both: long (mnemonic)
1915 form, short form, and old style. These styles are discussed below.
1916 Both the options and the operations can be written in any of these three
1919 @FIXME{menu at end of this node. need to think of an actual outline
1920 for this chapter; probably do that after stuff from chapter 4 is
1924 @section The Three Option Styles
1926 There are three styles for writing operations and options to the command
1927 line invoking @command{tar}. The different styles were developed at
1928 different times during the history of @command{tar}. These styles will be
1929 presented below, from the most recent to the oldest.
1931 Some options must take an argument. (For example, @option{--file}
1932 (@option{-f})) takes the name of an archive file as an argument. If
1933 you do not supply an archive file name, @command{tar} will use a
1934 default, but this can be confusing; thus, we recommend that you always
1935 supply a specific archive file name.) Where you @emph{place} the
1936 arguments generally depends on which style of options you choose. We
1937 will detail specific information relevant to each option style in the
1938 sections on the different option styles, below. The differences are
1939 subtle, yet can often be very important; incorrect option placement
1940 can cause you to overwrite a number of important files. We urge you
1941 to note these differences, and only use the option style(s) which
1942 makes the most sense to you until you feel comfortable with the others.
1944 Some options @emph{may} take an argument (currently, there are
1945 two such options: @option{--backup} and @option{--occurrence}). Such
1946 options may have at most long and short forms, they do not have old style
1947 equivalent. The rules for specifying an argument for such options
1948 are stricter than those for specifying mandatory arguments. Please,
1949 pay special attention to them.
1952 * Mnemonic Options:: Mnemonic Option Style
1953 * Short Options:: Short Option Style
1954 * Old Options:: Old Option Style
1955 * Mixing:: Mixing Option Styles
1958 @node Mnemonic Options
1959 @subsection Mnemonic Option Style
1961 @FIXME{have to decide whether or not to replace other occurrences of
1962 "mnemonic" with "long", or *ugh* vice versa.}
1964 Each option has at least one long (or mnemonic) name starting with two
1965 dashes in a row, e.g., @option{--list}. The long names are more clear than
1966 their corresponding short or old names. It sometimes happens that a
1967 single mnemonic option has many different different names which are
1968 synonymous, such as @option{--compare} and @option{--diff}. In addition,
1969 long option names can be given unique abbreviations. For example,
1970 @option{--cre} can be used in place of @option{--create} because there is no
1971 other mnemonic option which begins with @samp{cre}. (One way to find
1972 this out is by trying it and seeing what happens; if a particular
1973 abbreviation could represent more than one option, @command{tar} will tell
1974 you that that abbreviation is ambiguous and you'll know that that
1975 abbreviation won't work. You may also choose to run @samp{tar --help}
1976 to see a list of options. Be aware that if you run @command{tar} with a
1977 unique abbreviation for the long name of an option you didn't want to
1978 use, you are stuck; @command{tar} will perform the command as ordered.)
1980 Mnemonic options are meant to be obvious and easy to remember, and their
1981 meanings are generally easier to discern than those of their
1982 corresponding short options (see below). For example:
1985 $ @kbd{tar --create --verbose --blocking-factor=20 --file=/dev/rmt0}
1989 gives a fairly good set of hints about what the command does, even
1990 for those not fully acquainted with @command{tar}.
1992 Mnemonic options which require arguments take those arguments
1993 immediately following the option name. There are two ways of
1994 specifying a mandatory argument. It can be separated from the
1995 option name either by an equal sign, or by any amount of
1996 white space characters. For example, the @option{--file} option (which
1997 tells the name of the @command{tar} archive) is given a file such as
1998 @file{archive.tar} as argument by using any of the following notations:
1999 @option{--file=archive.tar} or @option{--file archive.tar}.
2001 In contrast, optional arguments must always be introduced using
2002 an equal sign. For example, the @option{--backup} option takes
2003 an optional argument specifying backup type. It must be used
2004 as @option{--backup=@var{backup-type}}.
2007 @subsection Short Option Style
2009 Most options also have a short option name. Short options start with
2010 a single dash, and are followed by a single character, e.g., @option{-t}
2011 (which is equivalent to @option{--list}). The forms are absolutely
2012 identical in function; they are interchangeable.
2014 The short option names are faster to type than long option names.
2016 Short options which require arguments take their arguments immediately
2017 following the option, usually separated by white space. It is also
2018 possible to stick the argument right after the short option name, using
2019 no intervening space. For example, you might write @w{@option{-f
2020 archive.tar}} or @option{-farchive.tar} instead of using
2021 @option{--file=archive.tar}. Both @option{--file=@var{archive-name}} and
2022 @w{@option{-f @var{archive-name}}} denote the option which indicates a
2023 specific archive, here named @file{archive.tar}.
2025 Short options which take optional arguments take their arguments
2026 immediately following the option letter, @emph{without any intervening
2027 white space characters}.
2029 Short options' letters may be clumped together, but you are not
2030 required to do this (as compared to old options; see below). When
2031 short options are clumped as a set, use one (single) dash for them
2032 all, e.g., @w{@samp{@command{tar} -cvf}}. Only the last option in
2033 such a set is allowed to have an argument@footnote{Clustering many
2034 options, the last of which has an argument, is a rather opaque way to
2035 write options. Some wonder if @acronym{GNU} @code{getopt} should not
2036 even be made helpful enough for considering such usages as invalid.}.
2038 When the options are separated, the argument for each option which requires
2039 an argument directly follows that option, as is usual for Unix programs.
2043 $ @kbd{tar -c -v -b 20 -f /dev/rmt0}
2046 If you reorder short options' locations, be sure to move any arguments
2047 that belong to them. If you do not move the arguments properly, you may
2048 end up overwriting files.
2051 @subsection Old Option Style
2054 Like short options, old options are single letters. However, old options
2055 must be written together as a single clumped set, without spaces separating
2056 them or dashes preceding them@footnote{Beware that if you precede options
2057 with a dash, you are announcing the short option style instead of the
2058 old option style; short options are decoded differently.}. This set
2059 of letters must be the first to appear on the command line, after the
2060 @command{tar} program name and some white space; old options cannot appear
2061 anywhere else. The letter of an old option is exactly the same letter as
2062 the corresponding short option. For example, the old option @samp{t} is
2063 the same as the short option @option{-t}, and consequently, the same as the
2064 mnemonic option @option{--list}. So for example, the command @w{@samp{tar
2065 cv}} specifies the option @option{-v} in addition to the operation @option{-c}.
2067 When options that need arguments are given together with the command,
2068 all the associated arguments follow, in the same order as the options.
2069 Thus, the example given previously could also be written in the old
2073 $ @kbd{tar cvbf 20 /dev/rmt0}
2077 Here, @samp{20} is the argument of @option{-b} and @samp{/dev/rmt0} is
2078 the argument of @option{-f}.
2080 On the other hand, this old style syntax makes it difficult to match
2081 option letters with their corresponding arguments, and is often
2082 confusing. In the command @w{@samp{tar cvbf 20 /dev/rmt0}}, for example,
2083 @samp{20} is the argument for @option{-b}, @samp{/dev/rmt0} is the
2084 argument for @option{-f}, and @option{-v} does not have a corresponding
2085 argument. Even using short options like in @w{@samp{tar -c -v -b 20 -f
2086 /dev/rmt0}} is clearer, putting all arguments next to the option they
2089 If you want to reorder the letters in the old option argument, be
2090 sure to reorder any corresponding argument appropriately.
2092 This old way of writing @command{tar} options can surprise even experienced
2093 users. For example, the two commands:
2096 @kbd{tar cfz archive.tar.gz file}
2097 @kbd{tar -cfz archive.tar.gz file}
2101 are quite different. The first example uses @file{archive.tar.gz} as
2102 the value for option @samp{f} and recognizes the option @samp{z}. The
2103 second example, however, uses @file{z} as the value for option
2104 @samp{f} --- probably not what was intended.
2106 Old options are kept for compatibility with old versions of @command{tar}.
2108 This second example could be corrected in many ways, among which the
2109 following are equivalent:
2112 @kbd{tar -czf archive.tar.gz file}
2113 @kbd{tar -cf archive.tar.gz -z file}
2114 @kbd{tar cf archive.tar.gz -z file}
2117 @cindex option syntax, traditional
2118 As far as we know, all @command{tar} programs, @acronym{GNU} and
2119 non-@acronym{GNU}, support old options. @GNUTAR{}
2120 supports them not only for historical reasons, but also because many
2121 people are used to them. For compatibility with Unix @command{tar},
2122 the first argument is always treated as containing command and option
2123 letters even if it doesn't start with @samp{-}. Thus, @samp{tar c} is
2124 equivalent to @w{@samp{tar -c}:} both of them specify the
2125 @option{--create} (@option{-c}) command to create an archive.
2128 @subsection Mixing Option Styles
2130 All three styles may be intermixed in a single @command{tar} command,
2131 so long as the rules for each style are fully
2132 respected@footnote{Before @GNUTAR{} version 1.11.6,
2133 a bug prevented intermixing old style options with mnemonic options in
2134 some cases.}. Old style options and either of the modern styles of
2135 options may be mixed within a single @command{tar} command. However,
2136 old style options must be introduced as the first arguments only,
2137 following the rule for old options (old options must appear directly
2138 after the @command{tar} command and some white space). Modern options
2139 may be given only after all arguments to the old options have been
2140 collected. If this rule is not respected, a modern option might be
2141 falsely interpreted as the value of the argument to one of the old
2144 For example, all the following commands are wholly equivalent, and
2145 illustrate the many combinations and orderings of option styles.
2148 @kbd{tar --create --file=archive.tar}
2149 @kbd{tar --create -f archive.tar}
2150 @kbd{tar --create -farchive.tar}
2151 @kbd{tar --file=archive.tar --create}
2152 @kbd{tar --file=archive.tar -c}
2153 @kbd{tar -c --file=archive.tar}
2154 @kbd{tar -c -f archive.tar}
2155 @kbd{tar -c -farchive.tar}
2156 @kbd{tar -cf archive.tar}
2157 @kbd{tar -cfarchive.tar}
2158 @kbd{tar -f archive.tar --create}
2159 @kbd{tar -f archive.tar -c}
2160 @kbd{tar -farchive.tar --create}
2161 @kbd{tar -farchive.tar -c}
2162 @kbd{tar c --file=archive.tar}
2163 @kbd{tar c -f archive.tar}
2164 @kbd{tar c -farchive.tar}
2165 @kbd{tar cf archive.tar}
2166 @kbd{tar f archive.tar --create}
2167 @kbd{tar f archive.tar -c}
2168 @kbd{tar fc archive.tar}
2171 On the other hand, the following commands are @emph{not} equivalent to
2175 @kbd{tar -f -c archive.tar}
2176 @kbd{tar -fc archive.tar}
2177 @kbd{tar -fcarchive.tar}
2178 @kbd{tar -farchive.tarc}
2179 @kbd{tar cfarchive.tar}
2183 These last examples mean something completely different from what the
2184 user intended (judging based on the example in the previous set which
2185 uses long options, whose intent is therefore very clear). The first
2186 four specify that the @command{tar} archive would be a file named
2187 @option{-c}, @samp{c}, @samp{carchive.tar} or @samp{archive.tarc},
2188 respectively. The first two examples also specify a single non-option,
2189 @var{name} argument having the value @samp{archive.tar}. The last
2190 example contains only old style option letters (repeating option
2191 @samp{c} twice), not all of which are meaningful (eg., @samp{.},
2192 @samp{h}, or @samp{i}), with no argument value. @FIXME{not sure i liked
2193 the first sentence of this paragraph..}
2196 @section All @command{tar} Options
2198 The coming manual sections contain an alphabetical listing of all
2199 @command{tar} operations and options, with brief descriptions and cross
2200 references to more in-depth explanations in the body of the manual.
2201 They also contain an alphabetically arranged table of the short option
2202 forms with their corresponding long option. You can use this table as
2203 a reference for deciphering @command{tar} commands in scripts.
2206 * Operation Summary::
2208 * Short Option Summary::
2211 @node Operation Summary
2212 @subsection Operations
2216 @opindex append, summary
2220 Appends files to the end of the archive. @xref{append}.
2222 @opindex catenate, summary
2226 Same as @option{--concatenate}. @xref{concatenate}.
2228 @opindex compare, summary
2232 Compares archive members with their counterparts in the file
2233 system, and reports differences in file size, mode, owner,
2234 modification date and contents. @xref{compare}.
2236 @opindex concatenate, summary
2240 Appends other @command{tar} archives to the end of the archive.
2243 @opindex create, summary
2247 Creates a new @command{tar} archive. @xref{create}.
2249 @opindex delete, summary
2252 Deletes members from the archive. Don't try this on a archive on a
2253 tape! @xref{delete}.
2255 @opindex diff, summary
2259 Same @option{--compare}. @xref{compare}.
2261 @opindex extract, summary
2265 Extracts members from the archive into the file system. @xref{extract}.
2267 @opindex get, summary
2271 Same as @option{--extract}. @xref{extract}.
2273 @opindex list, summary
2277 Lists the members in an archive. @xref{list}.
2279 @opindex update, summary
2283 Adds files to the end of the archive, but only if they are newer than
2284 their counterparts already in the archive, or if they do not already
2285 exist in the archive. @xref{update}.
2289 @node Option Summary
2290 @subsection @command{tar} Options
2294 @opindex absolute-names, summary
2295 @item --absolute-names
2298 Normally when creating an archive, @command{tar} strips an initial
2299 @samp{/} from member names. This option disables that behavior.
2302 @opindex after-date, summary
2305 (See @option{--newer}, @pxref{after})
2307 @opindex anchored, summary
2309 A pattern must match an initial subsequence of the name's components.
2310 @xref{controlling pattern-matching}.
2312 @opindex atime-preserve, summary
2313 @item --atime-preserve
2314 @itemx --atime-preserve=replace
2315 @itemx --atime-preserve=system
2317 Attempt to preserve the access time of files when reading them. This
2318 option currently is effective only on files that you own, unless you
2319 have superuser privileges.
2321 @option{--atime-preserve=replace} remembers the access time of a file
2322 before reading it, and then restores the access time afterwards. This
2323 may cause problems if other programs are reading the file at the same
2324 time, as the times of their accesses will be lost. On most platforms
2325 restoring the access time also requires @command{tar} to restore the
2326 data modification time too, so this option may also cause problems if
2327 other programs are writing the file at the same time. (Tar attempts
2328 to detect this situation, but cannot do so reliably due to race
2329 conditions.) Worse, on most platforms restoring the access time also
2330 updates the status change time, which means that this option is
2331 incompatible with incremental backups.
2333 @option{--atime-preserve=system} avoids changing time stamps on files,
2334 without interfering with time stamp updates
2335 caused by other programs, so it works better with incremental backups.
2336 However, it requires a special @code{O_NOATIME} option from the
2337 underlying operating and file system implementation, and it also requires
2338 that searching directories does not update their access times. As of
2339 this writing (November 2005) this works only with Linux, and only with
2340 Linux kernels 2.6.8 and later. Worse, there is currently no reliable
2341 way to know whether this feature actually works. Sometimes
2342 @command{tar} knows that it does not work, and if you use
2343 @option{--atime-preserve=system} then @command{tar} complains and
2344 exits right away. But other times @command{tar} might think that the
2345 option works when it actually does not.
2347 Currently @option{--atime-preserve} with no operand defaults to
2348 @option{--atime-preserve=replace}, but this may change in the future
2349 as support for @option{--atime-preserve=system} improves.
2351 If your operating system does not support
2352 @option{--atime-preserve=system}, you might be able to preserve access
2353 times reliably by by using the @command{mount} command. For example,
2354 you can mount the file system read-only, or access the file system via
2355 a read-only loopback mount, or use the @samp{noatime} mount option
2356 available on some systems. However, mounting typically requires
2357 superuser privileges and can be a pain to manage.
2359 @opindex backup, summary
2360 @item --backup=@var{backup-type}
2362 Rather than deleting files from the file system, @command{tar} will
2363 back them up using simple or numbered backups, depending upon
2364 @var{backup-type}. @xref{backup}.
2366 @opindex block-number, summary
2367 @item --block-number
2370 With this option present, @command{tar} prints error messages for read errors
2371 with the block number in the archive file. @xref{block-number}.
2373 @opindex blocking-factor, summary
2374 @item --blocking-factor=@var{blocking}
2375 @itemx -b @var{blocking}
2377 Sets the blocking factor @command{tar} uses to @var{blocking} x 512 bytes per
2378 record. @xref{Blocking Factor}.
2380 @opindex bzip2, summary
2384 This option tells @command{tar} to read or write archives through
2385 @code{bzip2}. @xref{gzip}.
2387 @opindex checkpoint, summary
2388 @item --checkpoint[=@var{number}]
2390 This option directs @command{tar} to print periodic checkpoint
2391 messages as it reads through the archive. It is intended for when you
2392 want a visual indication that @command{tar} is still running, but
2393 don't want to see @option{--verbose} output. For a detailed
2394 description, see @ref{Progress information}.
2396 @opindex check-links, summary
2399 If this option was given, @command{tar} will check the number of links
2400 dumped for each processed file. If this number does not match the
2401 total number of hard links for the file, a warning message will be
2402 output @footnote{Earlier versions of @GNUTAR{} understood @option{-l} as a
2403 synonym for @option{--one-file-system}. The current semantics, which
2404 complies to UNIX98, was introduced with version
2405 1.15.91. @xref{Changes}, for more information.}.
2407 @opindex compress, summary
2408 @opindex uncompress, summary
2413 @command{tar} will use the @command{compress} program when reading or
2414 writing the archive. This allows you to directly act on archives
2415 while saving space. @xref{gzip}.
2417 @opindex confirmation, summary
2418 @item --confirmation
2420 (See @option{--interactive}.) @xref{interactive}.
2422 @opindex delay-directory-restore, summary
2423 @item --delay-directory-restore
2425 Delay setting modification times and permissions of extracted
2426 directories until the end of extraction. @xref{Directory Modification Times and Permissions}.
2428 @opindex dereference, summary
2432 When creating a @command{tar} archive, @command{tar} will archive the
2433 file that a symbolic link points to, rather than archiving the
2434 symlink. @xref{dereference}.
2436 @opindex directory, summary
2437 @item --directory=@var{dir}
2440 When this option is specified, @command{tar} will change its current directory
2441 to @var{dir} before performing any operations. When this option is used
2442 during archive creation, it is order sensitive. @xref{directory}.
2444 @opindex exclude, summary
2445 @item --exclude=@var{pattern}
2447 When performing operations, @command{tar} will skip files that match
2448 @var{pattern}. @xref{exclude}.
2450 @opindex exclude-from, summary
2451 @item --exclude-from=@var{file}
2452 @itemx -X @var{file}
2454 Similar to @option{--exclude}, except @command{tar} will use the list of
2455 patterns in the file @var{file}. @xref{exclude}.
2457 @opindex exclude-caches, summary
2458 @item --exclude-caches
2460 Automatically excludes all directories
2461 containing a cache directory tag. @xref{exclude}.
2463 @opindex file, summary
2464 @item --file=@var{archive}
2465 @itemx -f @var{archive}
2467 @command{tar} will use the file @var{archive} as the @command{tar} archive it
2468 performs operations on, rather than @command{tar}'s compilation dependent
2469 default. @xref{file tutorial}.
2471 @opindex files-from, summary
2472 @item --files-from=@var{file}
2473 @itemx -T @var{file}
2475 @command{tar} will use the contents of @var{file} as a list of archive members
2476 or files to operate on, in addition to those specified on the
2477 command-line. @xref{files}.
2479 @opindex force-local, summary
2482 Forces @command{tar} to interpret the filename given to @option{--file}
2483 as a local file, even if it looks like a remote tape drive name.
2484 @xref{local and remote archives}.
2486 @opindex format, summary
2487 @item --format=@var{format}
2489 Selects output archive format. @var{Format} may be one of the
2494 Creates an archive that is compatible with Unix V7 @command{tar}.
2497 Creates an archive that is compatible with GNU @command{tar} version
2501 Creates archive in GNU tar 1.13 format. Basically it is the same as
2502 @samp{oldgnu} with the only difference in the way it handles long
2506 Creates a @acronym{POSIX.1-1988} compatible archive.
2509 Creates a @acronym{POSIX.1-2001 archive}.
2513 @xref{Formats}, for a detailed discussion of these formats.
2515 @opindex group, summary
2516 @item --group=@var{group}
2518 Files added to the @command{tar} archive will have a group id of @var{group},
2519 rather than the group from the source file. @var{group} is first decoded
2520 as a group symbolic name, but if this interpretation fails, it has to be
2521 a decimal numeric group ID. @FIXME-xref{}
2523 Also see the comments for the @option{--owner=@var{user}} option.
2525 @opindex gzip, summary
2526 @opindex gunzip, summary
2527 @opindex ungzip, summary
2533 This option tells @command{tar} to read or write archives through
2534 @command{gzip}, allowing @command{tar} to directly operate on several
2535 kinds of compressed archives transparently. @xref{gzip}.
2537 @opindex help, summary
2540 @command{tar} will print out a short message summarizing the operations and
2541 options to @command{tar} and exit. @xref{help}.
2543 @opindex ignore-case, summary
2545 Ignore case when matching member or file names with
2546 patterns. @xref{controlling pattern-matching}.
2548 @opindex ignore-command-error, summary
2549 @item --ignore-command-error
2550 Ignore exit codes of subprocesses. @xref{Writing to an External Program}.
2552 @opindex ignore-failed-read, summary
2553 @item --ignore-failed-read
2555 Do not exit unsuccessfully merely because an unreadable file was encountered.
2558 @opindex ignore-zeros, summary
2559 @item --ignore-zeros
2562 With this option, @command{tar} will ignore zeroed blocks in the
2563 archive, which normally signals EOF. @xref{Reading}.
2565 @opindex incremental, summary
2569 Used to inform @command{tar} that it is working with an old
2570 @acronym{GNU}-format incremental backup archive. It is intended
2571 primarily for backwards compatibility only. @xref{Incremental Dumps},
2572 for a detailed discussion of incremental archives.
2574 @opindex index-file, summary
2575 @item --index-file=@var{file}
2577 Send verbose output to @var{file} instead of to standard output.
2579 @opindex info-script, summary
2580 @opindex new-volume-script, summary
2581 @item --info-script=@var{script-file}
2582 @itemx --new-volume-script=@var{script-file}
2583 @itemx -F @var{script-file}
2585 When @command{tar} is performing multi-tape backups, @var{script-file} is run
2586 at the end of each tape. If @var{script-file} exits with nonzero status,
2587 @command{tar} fails immediately. @xref{info-script}, for a detailed
2588 discussion of @var{script-file}.
2590 @opindex interactive, summary
2592 @itemx --confirmation
2595 Specifies that @command{tar} should ask the user for confirmation before
2596 performing potentially destructive options, such as overwriting files.
2599 @opindex keep-newer-files, summary
2600 @item --keep-newer-files
2602 Do not replace existing files that are newer than their archive copies
2603 when extracting files from an archive.
2605 @opindex keep-old-files, summary
2606 @item --keep-old-files
2609 Do not overwrite existing files when extracting files from an archive.
2610 @xref{Keep Old Files}.
2612 @opindex label, summary
2613 @item --label=@var{name}
2614 @itemx -V @var{name}
2616 When creating an archive, instructs @command{tar} to write @var{name}
2617 as a name record in the archive. When extracting or listing archives,
2618 @command{tar} will only operate on archives that have a label matching
2619 the pattern specified in @var{name}. @xref{Tape Files}.
2621 @opindex listed-incremental, summary
2622 @item --listed-incremental=@var{snapshot-file}
2623 @itemx -g @var{snapshot-file}
2625 During a @option{--create} operation, specifies that the archive that
2626 @command{tar} creates is a new @acronym{GNU}-format incremental
2627 backup, using @var{snapshot-file} to determine which files to backup.
2628 With other operations, informs @command{tar} that the archive is in
2629 incremental format. @xref{Incremental Dumps}.
2631 @opindex mode, summary
2632 @item --mode=@var{permissions}
2634 When adding files to an archive, @command{tar} will use
2635 @var{permissions} for the archive members, rather than the permissions
2636 from the files. The program @command{chmod} and this @command{tar}
2637 option share the same syntax for what @var{permissions} might be.
2638 @xref{File permissions, Permissions, File permissions, fileutils,
2639 @acronym{GNU} file utilities}. This reference also has useful
2640 information for those not being overly familiar with the Unix
2643 Of course, @var{permissions} might be plainly specified as an octal number.
2644 However, by using generic symbolic modifications to mode bits, this allows
2645 more flexibility. For example, the value @samp{a+rw} adds read and write
2646 permissions for everybody, while retaining executable bits on directories
2647 or on any other file already marked as executable.
2649 @opindex multi-volume, summary
2650 @item --multi-volume
2653 Informs @command{tar} that it should create or otherwise operate on a
2654 multi-volume @command{tar} archive. @xref{Using Multiple Tapes}.
2656 @opindex new-volume-script, summary
2657 @item --new-volume-script
2661 @opindex seek, summary
2665 Assume that the archive media supports seeks to arbitrary
2666 locations. Usually @command{tar} determines automatically whether
2667 the archive can be seeked or not. This option is intended for use
2668 in cases when such recognition fails.
2670 @opindex newer, summary
2671 @item --newer=@var{date}
2672 @itemx --after-date=@var{date}
2675 When creating an archive, @command{tar} will only add files that have changed
2676 since @var{date}. If @var{date} begins with @samp{/} or @samp{.}, it
2677 is taken to be the name of a file whose data modification time specifies
2678 the date. @xref{after}.
2680 @opindex newer-mtime, summary
2681 @item --newer-mtime=@var{date}
2683 Like @option{--newer}, but add only files whose
2684 contents have changed (as opposed to just @option{--newer}, which will
2685 also back up files for which any status information has changed).
2687 @opindex no-anchored, summary
2689 An exclude pattern can match any subsequence of the name's components.
2690 @xref{controlling pattern-matching}.
2692 @opindex no-delay-directory-restore, summary
2693 @item --no-delay-directory-restore
2695 Setting modification times and permissions of extracted
2696 directories when all files from this directory has been
2697 extracted. This is the default. @xref{Directory Modification Times and Permissions}.
2699 @opindex no-ignore-case, summary
2700 @item --no-ignore-case
2701 Use case-sensitive matching.
2702 @xref{controlling pattern-matching}.
2704 @opindex no-ignore-command-error, summary
2705 @item --no-ignore-command-error
2706 Print warnings about subprocesses terminated with a non-zero exit
2707 code. @xref{Writing to an External Program}.
2709 @opindex no-quote-chars, summary
2710 @item --no-quote-chars=@var{string}
2711 Remove characters listed in @var{string} from the list of quoted
2712 characters set by the previous @option{--quote-chars} option
2713 (@pxref{quoting styles}).
2715 @opindex no-recursion, summary
2716 @item --no-recursion
2718 With this option, @command{tar} will not recurse into directories.
2721 @opindex no-same-owner, summary
2722 @item --no-same-owner
2725 When extracting an archive, do not attempt to preserve the owner
2726 specified in the @command{tar} archive. This the default behavior
2729 @opindex no-same-permissions, summary
2730 @item --no-same-permissions
2732 When extracting an archive, subtract the user's umask from files from
2733 the permissions specified in the archive. This is the default behavior
2736 @opindex no-wildcards, summary
2737 @item --no-wildcards
2738 Do not use wildcards.
2739 @xref{controlling pattern-matching}.
2741 @opindex no-wildcards-match-slash, summary
2742 @item --no-wildcards-match-slash
2743 Wildcards do not match @samp{/}.
2744 @xref{controlling pattern-matching}.
2746 @opindex null, summary
2749 When @command{tar} is using the @option{--files-from} option, this option
2750 instructs @command{tar} to expect filenames terminated with @option{NUL}, so
2751 @command{tar} can correctly work with file names that contain newlines.
2754 @opindex numeric-owner, summary
2755 @item --numeric-owner
2757 This option will notify @command{tar} that it should use numeric user
2758 and group IDs when creating a @command{tar} file, rather than names.
2762 When extracting files, this option is a synonym for
2763 @option{--no-same-owner}, i.e. it prevents @command{tar} from
2764 restoring ownership of files being extracted.
2766 When creating an archive, @option{-o} is a synonym for
2767 @option{--old-archive}. This behavior is for compatibility
2768 with previous versions of @GNUTAR{}, and will be
2769 removed in the future releases.
2771 @xref{Changes}, for more information.
2773 @opindex occurrence, summary
2774 @item --occurrence[=@var{number}]
2776 This option can be used in conjunction with one of the subcommands
2777 @option{--delete}, @option{--diff}, @option{--extract} or
2778 @option{--list} when a list of files is given either on the command
2779 line or via @option{-T} option.
2781 This option instructs @command{tar} to process only the @var{number}th
2782 occurrence of each named file. @var{Number} defaults to 1, so
2785 tar -x -f archive.tar --occurrence filename
2789 will extract the first occurrence of @file{filename} from @file{archive.tar}
2790 and will terminate without scanning to the end of the archive.
2792 @opindex old-archive, summary
2794 Synonym for @option{--format=v7}.
2796 @opindex one-file-system, summary
2797 @item --one-file-system
2798 Used when creating an archive. Prevents @command{tar} from recursing into
2799 directories that are on different file systems from the current
2800 directory @footnote{Earlier versions of @GNUTAR{} understood @option{-l} as a
2801 synonym for @option{--one-file-system}. This has changed in version
2802 1.15.91. @xref{Changes}, for more information.}.
2804 @opindex overwrite, summary
2807 Overwrite existing files and directory metadata when extracting files
2808 from an archive. @xref{Overwrite Old Files}.
2810 @opindex overwrite-dir, summary
2811 @item --overwrite-dir
2813 Overwrite the metadata of existing directories when extracting files
2814 from an archive. @xref{Overwrite Old Files}.
2816 @opindex owner, summary
2817 @item --owner=@var{user}
2819 Specifies that @command{tar} should use @var{user} as the owner of members
2820 when creating archives, instead of the user associated with the source
2821 file. @var{user} is first decoded as a user symbolic name, but if
2822 this interpretation fails, it has to be a decimal numeric user ID.
2825 There is no value indicating a missing number, and @samp{0} usually means
2826 @code{root}. Some people like to force @samp{0} as the value to offer in
2827 their distributions for the owner of files, because the @code{root} user is
2828 anonymous anyway, so that might as well be the owner of anonymous archives.
2830 This option does not affect extraction from archives.
2832 @opindex transform, summary
2833 @item --transform=@var{sed-expr}
2835 Transform file or member names using @command{sed} replacement expression
2836 @var{sed-expr}. For example,
2839 $ @kbd{tar cf archive.tar --transform 's,^\./,usr/,' .}
2843 will add to @file{archive} files from the current working directory,
2844 replacing initial @samp{./} prefix with @samp{usr/}. For the detailed
2845 discussion, @xref{transform}.
2847 To see transformed member names in verbose listings, use
2848 @option{--show-transformed-names} option
2849 (@pxref{show-transformed-names}).
2851 @opindex quote-chars, summary
2852 @item --quote-chars=@var{string}
2853 Always quote characters from @var{string}, even if the selected
2854 quoting style would not quote them (@pxref{quoting styles}).
2856 @opindex quoting-style, summary
2857 @item --quoting-style=@var{style}
2858 Set quoting style to use when printing member and file names
2859 (@pxref{quoting styles}). Valid @var{style} values are:
2860 @code{literal}, @code{shell}, @code{shell-always}, @code{c},
2861 @code{escape}, @code{locale}, and @code{clocale}. Default quoting
2862 style is @code{escape}, unless overridden while configuring the
2865 @opindex pax-option, summary
2866 @item --pax-option=@var{keyword-list}
2867 This option is meaningful only with @acronym{POSIX.1-2001} archives
2868 (@pxref{posix}). It modifies the way @command{tar} handles the
2869 extended header keywords. @var{Keyword-list} is a comma-separated
2870 list of keyword options. @xref{PAX keywords}, for a detailed
2873 @opindex portability, summary
2875 @itemx --old-archive
2876 Synonym for @option{--format=v7}.
2878 @opindex posix, summary
2880 Same as @option{--format=posix}.
2882 @opindex preserve, summary
2885 Synonymous with specifying both @option{--preserve-permissions} and
2886 @option{--same-order}. @xref{Setting Access Permissions}.
2888 @opindex preserve-order, summary
2889 @item --preserve-order
2891 (See @option{--same-order}; @pxref{Reading}.)
2893 @opindex preserve-permissions, summary
2894 @opindex same-permissions, summary
2895 @item --preserve-permissions
2896 @itemx --same-permissions
2899 When @command{tar} is extracting an archive, it normally subtracts the
2900 users' umask from the permissions specified in the archive and uses
2901 that number as the permissions to create the destination file.
2902 Specifying this option instructs @command{tar} that it should use the
2903 permissions directly from the archive. @xref{Setting Access Permissions}.
2905 @opindex read-full-records, summary
2906 @item --read-full-records
2909 Specifies that @command{tar} should reblock its input, for reading
2910 from pipes on systems with buggy implementations. @xref{Reading}.
2912 @opindex record-size, summary
2913 @item --record-size=@var{size}
2915 Instructs @command{tar} to use @var{size} bytes per record when accessing the
2916 archive. @xref{Blocking Factor}.
2918 @opindex recursion, summary
2921 With this option, @command{tar} recurses into directories.
2924 @opindex recursive-unlink, summary
2925 @item --recursive-unlink
2928 directory hierarchies before extracting directories of the same name
2929 from the archive. @xref{Recursive Unlink}.
2931 @opindex remove-files, summary
2932 @item --remove-files
2934 Directs @command{tar} to remove the source file from the file system after
2935 appending it to an archive. @xref{remove files}.
2937 @opindex restrict, summary
2940 Disable use of some potentially harmful @command{tar} options.
2941 Currently this option disables shell invocaton from multi-volume menu
2942 (@pxref{Using Multiple Tapes}).
2944 @opindex rmt-command, summary
2945 @item --rmt-command=@var{cmd}
2947 Notifies @command{tar} that it should use @var{cmd} instead of
2948 the default @file{/usr/libexec/rmt} (@pxref{Remote Tape Server}).
2950 @opindex rsh-command, summary
2951 @item --rsh-command=@var{cmd}
2953 Notifies @command{tar} that is should use @var{cmd} to communicate with remote
2954 devices. @xref{Device}.
2956 @opindex same-order, summary
2958 @itemx --preserve-order
2961 This option is an optimization for @command{tar} when running on machines with
2962 small amounts of memory. It informs @command{tar} that the list of file
2963 arguments has already been sorted to match the order of files in the
2964 archive. @xref{Reading}.
2966 @opindex same-owner, summary
2969 When extracting an archive, @command{tar} will attempt to preserve the owner
2970 specified in the @command{tar} archive with this option present.
2971 This is the default behavior for the superuser; this option has an
2972 effect only for ordinary users. @xref{Attributes}.
2974 @opindex same-permissions, summary
2975 @item --same-permissions
2977 (See @option{--preserve-permissions}; @pxref{Setting Access Permissions}.)
2979 @opindex show-defaults, summary
2980 @item --show-defaults
2982 Displays the default options used by @command{tar} and exits
2983 successfully. This option is intended for use in shell scripts.
2984 Here is an example of what you can see using this option:
2987 $ tar --show-defaults
2988 --format=gnu -f- -b20 --quoting-style=escape \
2989 --rmt-command=/usr/libexec/rmt --rsh-command=/usr/bin/rsh
2992 @opindex show-omitted-dirs, summary
2993 @item --show-omitted-dirs
2995 Instructs @command{tar} to mention directories its skipping over when
2996 operating on a @command{tar} archive. @xref{show-omitted-dirs}.
2998 @opindex show-transformed-names, summary
2999 @opindex show-stored-names, summary
3000 @item --show-transformed-names
3001 @itemx --show-stored-names
3003 Display file or member names after applying any transformations
3004 (@pxref{transform}). In particular, when used in conjunction with one of
3005 archive creation operations it instructs tar to list the member names
3006 stored in the archive, as opposed to the actual file
3007 names. @xref{listing member and file names}.
3009 @opindex sparse, summary
3013 Invokes a @acronym{GNU} extension when adding files to an archive that handles
3014 sparse files efficiently. @xref{sparse}.
3016 @opindex starting-file, summary
3017 @item --starting-file=@var{name}
3018 @itemx -K @var{name}
3020 This option affects extraction only; @command{tar} will skip extracting
3021 files in the archive until it finds one that matches @var{name}.
3024 @opindex strip-components, summary
3025 @item --strip-components=@var{number}
3026 Strip given @var{number} of leading components from file names before
3027 extraction.@footnote{This option was called @option{--strip-path} in
3028 version 1.14.} For example, if archive @file{archive.tar} contained
3029 @file{/some/file/name}, then running
3032 tar --extract --file archive.tar --strip-components=2
3036 would extract this file to file @file{name}.
3038 @opindex suffix, summary
3039 @item --suffix=@var{suffix}
3041 Alters the suffix @command{tar} uses when backing up files from the default
3042 @samp{~}. @xref{backup}.
3044 @opindex tape-length, summary
3045 @item --tape-length=@var{num}
3048 Specifies the length of tapes that @command{tar} is writing as being
3049 @w{@var{num} x 1024} bytes long. @xref{Using Multiple Tapes}.
3051 @opindex test-label, summary
3054 Reads the volume label. If an argument is specified, test whether it
3055 matches the volume label. @xref{--test-label option}.
3057 @opindex to-command, summary
3058 @item --to-command=@var{command}
3060 During extraction @command{tar} will pipe extracted files to the
3061 standard input of @var{command}. @xref{Writing to an External Program}.
3063 @opindex to-stdout, summary
3067 During extraction, @command{tar} will extract files to stdout rather
3068 than to the file system. @xref{Writing to Standard Output}.
3070 @opindex totals, summary
3073 Displays the total number of bytes written after creating an archive.
3076 @opindex touch, summary
3080 Sets the data modification time of extracted files to the extraction time,
3081 rather than the data modification time stored in the archive.
3082 @xref{Data Modification Times}.
3084 @opindex uncompress, summary
3087 (See @option{--compress}. @pxref{gzip})
3089 @opindex ungzip, summary
3092 (See @option{--gzip}. @pxref{gzip})
3094 @opindex unlink-first, summary
3095 @item --unlink-first
3098 Directs @command{tar} to remove the corresponding file from the file
3099 system before extracting it from the archive. @xref{Unlink First}.
3101 @opindex use-compress-program, summary
3102 @item --use-compress-program=@var{prog}
3104 Instructs @command{tar} to access the archive through @var{prog}, which is
3105 presumed to be a compression program of some sort. @xref{gzip}.
3107 @opindex utc, summary
3110 Display file modification dates in @acronym{UTC}. This option implies
3113 @opindex verbose, summary
3117 Specifies that @command{tar} should be more verbose about the operations its
3118 performing. This option can be specified multiple times for some
3119 operations to increase the amount of information displayed.
3122 @opindex verify, summary
3126 Verifies that the archive was correctly written when creating an
3127 archive. @xref{verify}.
3129 @opindex version, summary
3132 Print information about the program's name, version, origin and legal
3133 status, all on standard output, and then exit successfully.
3136 @opindex volno-file, summary
3137 @item --volno-file=@var{file}
3139 Used in conjunction with @option{--multi-volume}. @command{tar} will keep track
3140 of which volume of a multi-volume archive its working in @var{file}.
3143 @opindex wildcards, summary
3145 Use wildcards when matching member names with patterns.
3146 @xref{controlling pattern-matching}.
3148 @opindex wildcards-match-slash, summary
3149 @item --wildcards-match-slash
3150 Wildcards match @samp{/}.
3151 @xref{controlling pattern-matching}.
3154 @node Short Option Summary
3155 @subsection Short Options Cross Reference
3157 Here is an alphabetized list of all of the short option forms, matching
3158 them with the equivalent long option.
3164 @option{--concatenate}
3168 @option{--read-full-records}
3172 @option{--directory}
3176 @option{--info-script}
3180 @option{--incremental}
3184 @option{--starting-file}
3188 @option{--tape-length}
3192 @option{--multi-volume}
3200 @option{--to-stdout}
3204 @option{--absolute-names}
3208 @option{--block-number}
3216 @option{--files-from}
3220 @option{--unlink-first}
3232 @option{--exclude-from}
3240 @option{--blocking-factor}
3256 @option{--listed-incremental}
3260 @option{--dereference}
3264 @option{--ignore-zeros}
3272 @option{--keep-old-files}
3276 @option{--one-file-system}. Use of this short option is deprecated. It
3277 is retained for compatibility with the earlier versions of GNU
3278 @command{tar}, and will be changed in future releases.
3280 @xref{Changes}, for more information.
3288 When creating --- @option{--no-same-owner}, when extracting ---
3289 @option{--portability}.
3291 The later usage is deprecated. It is retained for compatibility with
3292 the earlier versions of @GNUTAR{}. In the future releases
3293 @option{-o} will be equivalent to @option{--no-same-owner} only.
3297 @option{--preserve-permissions}
3305 @option{--same-order}
3321 @option{--interactive}
3334 @section @GNUTAR{} documentation
3336 @cindex Getting program version number
3338 @cindex Version of the @command{tar} program
3339 Being careful, the first thing is really checking that you are using
3340 @GNUTAR{}, indeed. The @option{--version} option
3341 causes @command{tar} to print information about its name, version,
3342 origin and legal status, all on standard output, and then exit
3343 successfully. For example, @w{@samp{tar --version}} might print:
3346 tar (GNU tar) 1.15.2
3347 Copyright (C) 2006 Free Software Foundation, Inc.
3348 This is free software. You may redistribute copies of it under the terms of
3349 the GNU General Public License <http://www.gnu.org/licenses/gpl.html>.
3350 There is NO WARRANTY, to the extent permitted by law.
3352 Written by John Gilmore and Jay Fenlason.
3356 The first occurrence of @samp{tar} in the result above is the program
3357 name in the package (for example, @command{rmt} is another program),
3358 while the second occurrence of @samp{tar} is the name of the package
3359 itself, containing possibly many programs. The package is currently
3360 named @samp{tar}, after the name of the main program it
3361 contains@footnote{There are plans to merge the @command{cpio} and
3362 @command{tar} packages into a single one which would be called
3363 @code{paxutils}. So, who knows if, one of this days, the
3364 @option{--version} would not output @w{@samp{tar (@acronym{GNU}
3367 @cindex Obtaining help
3368 @cindex Listing all @command{tar} options
3369 @opindex help, introduction
3370 Another thing you might want to do is checking the spelling or meaning
3371 of some particular @command{tar} option, without resorting to this
3372 manual, for once you have carefully read it. @GNUTAR{}
3373 has a short help feature, triggerable through the
3374 @option{--help} option. By using this option, @command{tar} will
3375 print a usage message listing all available options on standard
3376 output, then exit successfully, without doing anything else and
3377 ignoring all other options. Even if this is only a brief summary, it
3378 may be several screens long. So, if you are not using some kind of
3379 scrollable window, you might prefer to use something like:
3382 $ @kbd{tar --help | less}
3386 presuming, here, that you like using @command{less} for a pager. Other
3387 popular pagers are @command{more} and @command{pg}. If you know about some
3388 @var{keyword} which interests you and do not want to read all the
3389 @option{--help} output, another common idiom is doing:
3392 tar --help | grep @var{keyword}
3396 for getting only the pertinent lines. Notice, however, that some
3397 @command{tar} options have long description lines and the above
3398 command will list only the first of them.
3400 The exact look of the option summary displayed by @kbd{tar --help} is
3401 configurable. @xref{Configuring Help Summary}, for a detailed description.
3404 If you only wish to check the spelling of an option, running @kbd{tar
3405 --usage} may be a better choice. This will display a terse list of
3406 @command{tar} option without accompanying explanations.
3408 The short help output is quite succinct, and you might have to get
3409 back to the full documentation for precise points. If you are reading
3410 this paragraph, you already have the @command{tar} manual in some
3411 form. This manual is available in a variety of forms from
3412 @url{http://www.gnu.org/software/tar/manual}. It may printed out of the @GNUTAR{}
3413 distribution, provided you have @TeX{} already installed somewhere,
3414 and a laser printer around. Just configure the distribution, execute
3415 the command @w{@samp{make dvi}}, then print @file{doc/tar.dvi} the
3416 usual way (contact your local guru to know how). If @GNUTAR{}
3417 has been conveniently installed at your place, this
3418 manual is also available in interactive, hypertextual form as an Info
3419 file. Just call @w{@samp{info tar}} or, if you do not have the
3420 @command{info} program handy, use the Info reader provided within
3421 @acronym{GNU} Emacs, calling @samp{tar} from the main Info menu.
3423 There is currently no @code{man} page for @GNUTAR{}.
3424 If you observe such a @code{man} page on the system you are running,
3425 either it does not belong to @GNUTAR{}, or it has not
3426 been produced by @acronym{GNU}. Some package maintainers convert
3427 @kbd{tar --help} output to a man page, using @command{help2man}. In
3428 any case, please bear in mind that the authoritative source of
3429 information about @GNUTAR{} is this Texinfo documentation.
3432 @section Obtaining @GNUTAR{} default values
3434 @opindex show-defaults
3435 @GNUTAR{} has some predefined defaults that are used when you do not
3436 explicitely specify another values. To obtain a list of such
3437 defaults, use @option{--show-defaults} option. This will output the
3438 values in the form of @command{tar} command line options:
3442 @kbd{tar --show-defaults}
3443 --format=gnu -f- -b20 --rmt-command=/etc/rmt --rsh-command=/usr/bin/rsh
3448 The above output shows that this version of @GNUTAR{} defaults to
3449 using @samp{gnu} archive format (@pxref{Formats}), it uses standard
3450 output as the archive, if no @option{--file} option has been given
3451 (@pxref{file tutorial}), the default blocking factor is 20
3452 (@pxref{Blocking Factor}). It also shows the default locations where
3453 @command{tar} will look for @command{rmt} and @command{rsh} binaries.
3456 @section Checking @command{tar} progress
3458 Typically, @command{tar} performs most operations without reporting any
3459 information to the user except error messages. When using @command{tar}
3460 with many options, particularly ones with complicated or
3461 difficult-to-predict behavior, it is possible to make serious mistakes.
3462 @command{tar} provides several options that make observing @command{tar}
3463 easier. These options cause @command{tar} to print information as it
3464 progresses in its job, and you might want to use them just for being
3465 more careful about what is going on, or merely for entertaining
3466 yourself. If you have encountered a problem when operating on an
3467 archive, however, you may need more information than just an error
3468 message in order to solve the problem. The following options can be
3469 helpful diagnostic tools.
3471 @cindex Verbose operation
3473 Normally, the @option{--list} (@option{-t}) command to list an archive
3474 prints just the file names (one per line) and the other commands are
3475 silent. When used with most operations, the @option{--verbose}
3476 (@option{-v}) option causes @command{tar} to print the name of each
3477 file or archive member as it is processed. This and the other options
3478 which make @command{tar} print status information can be useful in
3479 monitoring @command{tar}.
3481 With @option{--create} or @option{--extract}, @option{--verbose} used
3482 once just prints the names of the files or members as they are processed.
3483 Using it twice causes @command{tar} to print a longer listing
3484 (reminiscent of @samp{ls -l}) for each member. Since @option{--list}
3485 already prints the names of the members, @option{--verbose} used once
3486 with @option{--list} causes @command{tar} to print an @samp{ls -l}
3487 type listing of the files in the archive. The following examples both
3488 extract members with long list output:
3491 $ @kbd{tar --extract --file=archive.tar --verbose --verbose}
3492 $ @kbd{tar xvvf archive.tar}
3495 Verbose output appears on the standard output except when an archive is
3496 being written to the standard output, as with @samp{tar --create
3497 --file=- --verbose} (@samp{tar cfv -}, or even @samp{tar cv}---if the
3498 installer let standard output be the default archive). In that case
3499 @command{tar} writes verbose output to the standard error stream.
3501 If @option{--index-file=@var{file}} is specified, @command{tar} sends
3502 verbose output to @var{file} rather than to standard output or standard
3505 @cindex Obtaining total status information
3507 The @option{--totals} option---which is only meaningful when used with
3508 @option{--create} (@option{-c})---causes @command{tar} to print the total
3509 amount written to the archive, after it has been fully created.
3511 @anchor{Progress information}
3512 @cindex Progress information
3514 The @option{--checkpoint} option prints an occasional message
3515 as @command{tar} reads or writes the archive. It is designed for
3516 those who don't need the more detailed (and voluminous) output of
3517 @option{--block-number} (@option{-R}), but do want visual confirmation
3518 that @command{tar} is actually making forward progress. By default it
3519 prints a message each 10 records read or written. This can be changed
3520 by giving it a numeric argument after an equal sign:
3523 $ @kbd{tar -c --checkpoint=1000} /var
3524 tar: Write checkpoint 1000
3525 tar: Write checkpoint 2000
3526 tar: Write checkpoint 3000
3529 This example shows the default checkpoint message used by
3530 @command{tar}. If you place a dot immediately after the equal
3531 sign, it will print a @samp{.} at each checkpoint. For example:
3534 $ @kbd{tar -c --checkpoint=.1000} /var
3538 @opindex show-omitted-dirs
3539 @anchor{show-omitted-dirs}
3540 The @option{--show-omitted-dirs} option, when reading an archive---with
3541 @option{--list} or @option{--extract}, for example---causes a message
3542 to be printed for each directory in the archive which is skipped.
3543 This happens regardless of the reason for skipping: the directory might
3544 not have been named on the command line (implicitly or explicitly),
3545 it might be excluded by the use of the
3546 @option{--exclude=@var{pattern}} option, or some other reason.
3548 @opindex block-number
3549 @cindex Block number where error occurred
3550 @anchor{block-number}
3551 If @option{--block-number} (@option{-R}) is used, @command{tar} prints, along with
3552 every message it would normally produce, the block number within the
3553 archive where the message was triggered. Also, supplementary messages
3554 are triggered when reading blocks full of NULs, or when hitting end of
3555 file on the archive. As of now, if the archive if properly terminated
3556 with a NUL block, the reading of the file may stop before end of file
3557 is met, so the position of end of file will not usually show when
3558 @option{--block-number} (@option{-R}) is used. Note that @GNUTAR{}
3559 drains the archive before exiting when reading the
3560 archive from a pipe.
3562 @cindex Error message, block number of
3563 This option is especially useful when reading damaged archives, since
3564 it helps pinpoint the damaged sections. It can also be used with
3565 @option{--list} (@option{-t}) when listing a file-system backup tape, allowing you to
3566 choose among several backup tapes when retrieving a file later, in
3567 favor of the tape where the file appears earliest (closest to the
3568 front of the tape). @xref{backup}.
3571 @section Asking for Confirmation During Operations
3572 @cindex Interactive operation
3574 Typically, @command{tar} carries out a command without stopping for
3575 further instructions. In some situations however, you may want to
3576 exclude some files and archive members from the operation (for instance
3577 if disk or storage space is tight). You can do this by excluding
3578 certain files automatically (@pxref{Choosing}), or by performing
3579 an operation interactively, using the @option{--interactive} (@option{-w}) option.
3580 @command{tar} also accepts @option{--confirmation} for this option.
3582 @opindex interactive
3583 When the @option{--interactive} (@option{-w}) option is specified, before
3584 reading, writing, or deleting files, @command{tar} first prints a message
3585 for each such file, telling what operation it intends to take, then asks
3586 for confirmation on the terminal. The actions which require
3587 confirmation include adding a file to the archive, extracting a file
3588 from the archive, deleting a file from the archive, and deleting a file
3589 from disk. To confirm the action, you must type a line of input
3590 beginning with @samp{y}. If your input line begins with anything other
3591 than @samp{y}, @command{tar} skips that file.
3593 If @command{tar} is reading the archive from the standard input,
3594 @command{tar} opens the file @file{/dev/tty} to support the interactive
3597 Verbose output is normally sent to standard output, separate from
3598 other error messages. However, if the archive is produced directly
3599 on standard output, then verbose output is mixed with errors on
3600 @code{stderr}. Producing the archive on standard output may be used
3601 as a way to avoid using disk space, when the archive is soon to be
3602 consumed by another process reading it, say. Some people felt the need
3603 of producing an archive on stdout, still willing to segregate between
3604 verbose output and error output. A possible approach would be using a
3605 named pipe to receive the archive, and having the consumer process to
3606 read from that named pipe. This has the advantage of letting standard
3607 output free to receive verbose output, all separate from errors.
3610 @chapter @GNUTAR{} Operations
3623 @section Basic @GNUTAR{} Operations
3625 The basic @command{tar} operations, @option{--create} (@option{-c}),
3626 @option{--list} (@option{-t}) and @option{--extract} (@option{--get},
3627 @option{-x}), are currently presented and described in the tutorial
3628 chapter of this manual. This section provides some complementary notes
3629 for these operations.
3632 @opindex create, complementary notes
3636 Creating an empty archive would have some kind of elegance. One can
3637 initialize an empty archive and later use @option{--append}
3638 (@option{-r}) for adding all members. Some applications would not
3639 welcome making an exception in the way of adding the first archive
3640 member. On the other hand, many people reported that it is
3641 dangerously too easy for @command{tar} to destroy a magnetic tape with
3642 an empty archive@footnote{This is well described in @cite{Unix-haters
3643 Handbook}, by Simson Garfinkel, Daniel Weise & Steven Strassmann, IDG
3644 Books, ISBN 1-56884-203-1.}. The two most common errors are:
3648 Mistakingly using @code{create} instead of @code{extract}, when the
3649 intent was to extract the full contents of an archive. This error
3650 is likely: keys @kbd{c} and @kbd{x} are right next to each other on
3651 the QWERTY keyboard. Instead of being unpacked, the archive then
3652 gets wholly destroyed. When users speak about @dfn{exploding} an
3653 archive, they usually mean something else :-).
3656 Forgetting the argument to @code{file}, when the intent was to create
3657 an archive with a single file in it. This error is likely because a
3658 tired user can easily add the @kbd{f} key to the cluster of option
3659 letters, by the mere force of habit, without realizing the full
3660 consequence of doing so. The usual consequence is that the single
3661 file, which was meant to be saved, is rather destroyed.
3664 So, recognizing the likelihood and the catastrophical nature of these
3665 errors, @GNUTAR{} now takes some distance from elegance, and
3666 cowardly refuses to create an archive when @option{--create} option is
3667 given, there are no arguments besides options, and
3668 @option{--files-from} (@option{-T}) option is @emph{not} used. To get
3669 around the cautiousness of @GNUTAR{} and nevertheless create an
3670 archive with nothing in it, one may still use, as the value for the
3671 @option{--files-from} option, a file with no names in it, as shown in
3672 the following commands:
3675 @kbd{tar --create --file=empty-archive.tar --files-from=/dev/null}
3676 @kbd{tar cfT empty-archive.tar /dev/null}
3679 @opindex extract, complementary notes
3684 A socket is stored, within a @GNUTAR{} archive, as a pipe.
3686 @item @option{--list} (@option{-t})
3688 @GNUTAR{} now shows dates as @samp{1996-08-30},
3689 while it used to show them as @samp{Aug 30 1996}. Preferably,
3690 people should get used to ISO 8601 dates. Local American dates should
3691 be made available again with full date localization support, once
3692 ready. In the meantime, programs not being localizable for dates
3693 should prefer international dates, that's really the way to go.
3695 Look up @url{http://www.ft.uni-erlangen.de/~mskuhn/iso-time.html} if you
3696 are curious, it contains a detailed explanation of the ISO 8601 standard.
3701 @section Advanced @GNUTAR{} Operations
3703 Now that you have learned the basics of using @GNUTAR{}, you may want
3704 to learn about further ways in which @command{tar} can help you.
3706 This chapter presents five, more advanced operations which you probably
3707 won't use on a daily basis, but which serve more specialized functions.
3708 We also explain the different styles of options and why you might want
3709 to use one or another, or a combination of them in your @command{tar}
3710 commands. Additionally, this chapter includes options which allow you to
3711 define the output from @command{tar} more carefully, and provide help and
3712 error correction in special circumstances.
3714 @FIXME{check this after the chapter is actually revised to make sure
3715 it still introduces the info in the chapter correctly : ).}
3727 @subsection The Five Advanced @command{tar} Operations
3730 In the last chapter, you learned about the first three operations to
3731 @command{tar}. This chapter presents the remaining five operations to
3732 @command{tar}: @option{--append}, @option{--update}, @option{--concatenate},
3733 @option{--delete}, and @option{--compare}.
3735 You are not likely to use these operations as frequently as those
3736 covered in the last chapter; however, since they perform specialized
3737 functions, they are quite useful when you do need to use them. We
3738 will give examples using the same directory and files that you created
3739 in the last chapter. As you may recall, the directory is called
3740 @file{practice}, the files are @samp{jazz}, @samp{blues}, @samp{folk},
3741 @samp{rock}, and the two archive files you created are
3742 @samp{collection.tar} and @samp{music.tar}.
3744 We will also use the archive files @samp{afiles.tar} and
3745 @samp{bfiles.tar}. @samp{afiles.tar} contains the members @samp{apple},
3746 @samp{angst}, and @samp{aspic}. @samp{bfiles.tar} contains the members
3747 @samp{./birds}, @samp{baboon}, and @samp{./box}.
3749 Unless we state otherwise, all practicing you do and examples you follow
3750 in this chapter will take place in the @file{practice} directory that
3751 you created in the previous chapter; see @ref{prepare for examples}.
3752 (Below in this section, we will remind you of the state of the examples
3753 where the last chapter left them.)
3755 The five operations that we will cover in this chapter are:
3760 Add new entries to an archive that already exists.
3763 Add more recent copies of archive members to the end of an archive, if
3768 Add one or more pre-existing archives to the end of another archive.
3770 Delete items from an archive (does not work on tapes).
3774 Compare archive members to their counterparts in the file system.
3778 @subsection How to Add Files to Existing Archives: @option{--append}
3782 If you want to add files to an existing archive, you don't need to
3783 create a new archive; you can use @option{--append} (@option{-r}).
3784 The archive must already exist in order to use @option{--append}. (A
3785 related operation is the @option{--update} operation; you can use this
3786 to add newer versions of archive members to an existing archive. To learn how to
3787 do this with @option{--update}, @pxref{update}.)
3789 If you use @option{--append} to add a file that has the same name as an
3790 archive member to an archive containing that archive member, then the
3791 old member is not deleted. What does happen, however, is somewhat
3792 complex. @command{tar} @emph{allows} you to have infinite number of files
3793 with the same name. Some operations treat these same-named members no
3794 differently than any other set of archive members: for example, if you
3795 view an archive with @option{--list} (@option{-t}), you will see all
3796 of those members listed, with their data modification times, owners, etc.
3798 Other operations don't deal with these members as perfectly as you might
3799 prefer; if you were to use @option{--extract} to extract the archive,
3800 only the most recently added copy of a member with the same name as four
3801 other members would end up in the working directory. This is because
3802 @option{--extract} extracts an archive in the order the members appeared
3803 in the archive; the most recently archived members will be extracted
3804 last. Additionally, an extracted member will @emph{replace} a file of
3805 the same name which existed in the directory already, and @command{tar}
3806 will not prompt you about this@footnote{Unless you give it
3807 @option{--keep-old-files} option, or the disk copy is newer than the
3808 the one in the archive and you invoke @command{tar} with
3809 @option{--keep-newer-files} option}. Thus, only the most recently archived
3810 member will end up being extracted, as it will replace the one
3811 extracted before it, and so on.
3813 There exists a special option that allows you to get around this
3814 behavior and extract (or list) only a particular copy of the file.
3815 This is @option{--occurrence} option. If you run @command{tar} with
3816 this option, it will extract only the first copy of the file. You
3817 may also give this option an argument specifying the number of
3818 copy to be extracted. Thus, for example if the archive
3819 @file{archive.tar} contained three copies of file @file{myfile}, then
3823 tar --extract --file archive.tar --occurrence=2 myfile
3827 would extract only the second copy. @xref{Option
3828 Summary,---occurrence}, for the description of @option{--occurrence}
3831 @FIXME{ hag -- you might want to incorporate some of the above into the
3832 MMwtSN node; not sure. i didn't know how to make it simpler...
3834 There are a few ways to get around this. (maybe xref Multiple Members
3835 with the Same Name.}
3837 @cindex Members, replacing with other members
3838 @cindex Replacing members with other members
3839 If you want to replace an archive member, use @option{--delete} to
3840 delete the member you want to remove from the archive, , and then use
3841 @option{--append} to add the member you want to be in the archive. Note
3842 that you can not change the order of the archive; the most recently
3843 added member will still appear last. In this sense, you cannot truly
3844 ``replace'' one member with another. (Replacing one member with another
3845 will not work on certain types of media, such as tapes; see @ref{delete}
3846 and @ref{Media}, for more information.)
3849 * appending files:: Appending Files to an Archive
3853 @node appending files
3854 @subsubsection Appending Files to an Archive
3856 @cindex Adding files to an Archive
3857 @cindex Appending files to an Archive
3858 @cindex Archives, Appending files to
3860 The simplest way to add a file to an already existing archive is the
3861 @option{--append} (@option{-r}) operation, which writes specified
3862 files into the archive whether or not they are already among the
3865 When you use @option{--append}, you @emph{must} specify file name
3866 arguments, as there is no default. If you specify a file that already
3867 exists in the archive, another copy of the file will be added to the
3868 end of the archive. As with other operations, the member names of the
3869 newly added files will be exactly the same as their names given on the
3870 command line. The @option{--verbose} (@option{-v}) option will print
3871 out the names of the files as they are written into the archive.
3873 @option{--append} cannot be performed on some tape drives, unfortunately,
3874 due to deficiencies in the formats those tape drives use. The archive
3875 must be a valid @command{tar} archive, or else the results of using this
3876 operation will be unpredictable. @xref{Media}.
3878 To demonstrate using @option{--append} to add a file to an archive,
3879 create a file called @file{rock} in the @file{practice} directory.
3880 Make sure you are in the @file{practice} directory. Then, run the
3881 following @command{tar} command to add @file{rock} to
3882 @file{collection.tar}:
3885 $ @kbd{tar --append --file=collection.tar rock}
3889 If you now use the @option{--list} (@option{-t}) operation, you will see that
3890 @file{rock} has been added to the archive:
3893 $ @kbd{tar --list --file=collection.tar}
3894 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
3895 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3896 -rw-r--r-- me user 20 1996-09-23 16:44 folk
3897 -rw-r--r-- me user 20 1996-09-23 16:44 rock
3901 @subsubsection Multiple Files with the Same Name
3903 You can use @option{--append} (@option{-r}) to add copies of files
3904 which have been updated since the archive was created. (However, we
3905 do not recommend doing this since there is another @command{tar}
3906 option called @option{--update}; @xref{update}, for more information.
3907 We describe this use of @option{--append} here for the sake of
3908 completeness.) When you extract the archive, the older version will
3909 be effectively lost. This works because files are extracted from an
3910 archive in the order in which they were archived. Thus, when the
3911 archive is extracted, a file archived later in time will replace a
3912 file of the same name which was archived earlier, even though the
3913 older version of the file will remain in the archive unless you delete
3914 all versions of the file.
3916 Supposing you change the file @file{blues} and then append the changed
3917 version to @file{collection.tar}. As you saw above, the original
3918 @file{blues} is in the archive @file{collection.tar}. If you change the
3919 file and append the new version of the file to the archive, there will
3920 be two copies in the archive. When you extract the archive, the older
3921 version of the file will be extracted first, and then replaced by the
3922 newer version when it is extracted.
3924 You can append the new, changed copy of the file @file{blues} to the
3925 archive in this way:
3928 $ @kbd{tar --append --verbose --file=collection.tar blues}
3933 Because you specified the @option{--verbose} option, @command{tar} has
3934 printed the name of the file being appended as it was acted on. Now
3935 list the contents of the archive:
3938 $ @kbd{tar --list --verbose --file=collection.tar}
3939 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
3940 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3941 -rw-r--r-- me user 20 1996-09-23 16:44 folk
3942 -rw-r--r-- me user 20 1996-09-23 16:44 rock
3943 -rw-r--r-- me user 58 1996-10-24 18:30 blues
3947 The newest version of @file{blues} is now at the end of the archive
3948 (note the different creation dates and file sizes). If you extract
3949 the archive, the older version of the file @file{blues} will be
3950 replaced by the newer version. You can confirm this by extracting
3951 the archive and running @samp{ls} on the directory.
3953 If you wish to extract the first occurrence of the file @file{blues}
3954 from the archive, use @option{--occurrence} option, as shown in
3955 the following example:
3958 $ @kbd{tar --extract -vv --occurrence --file=collection.tar blues}
3959 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3962 @xref{Writing}, for more information on @option{--extract} and
3963 @xref{Option Summary, --occurrence}, for the description of
3964 @option{--occurrence} option.
3967 @subsection Updating an Archive
3969 @cindex Updating an archive
3972 In the previous section, you learned how to use @option{--append} to
3973 add a file to an existing archive. A related operation is
3974 @option{--update} (@option{-u}). The @option{--update} operation
3975 updates a @command{tar} archive by comparing the date of the specified
3976 archive members against the date of the file with the same name. If
3977 the file has been modified more recently than the archive member, then
3978 the newer version of the file is added to the archive (as with
3981 Unfortunately, you cannot use @option{--update} with magnetic tape drives.
3982 The operation will fail.
3984 @FIXME{other examples of media on which --update will fail? need to ask
3985 charles and/or mib/thomas/dave shevett..}
3987 Both @option{--update} and @option{--append} work by adding to the end
3988 of the archive. When you extract a file from the archive, only the
3989 version stored last will wind up in the file system, unless you use
3990 the @option{--backup} option. @FIXME-ref{Multiple Members with the
3998 @subsubsection How to Update an Archive Using @option{--update}
4000 You must use file name arguments with the @option{--update}
4001 (@option{-u}) operation. If you don't specify any files,
4002 @command{tar} won't act on any files and won't tell you that it didn't
4003 do anything (which may end up confusing you).
4005 @c note: the above parenthetical added because in fact, this
4006 @c behavior just confused the author. :-)
4008 To see the @option{--update} option at work, create a new file,
4009 @file{classical}, in your practice directory, and some extra text to the
4010 file @file{blues}, using any text editor. Then invoke @command{tar} with
4011 the @samp{update} operation and the @option{--verbose} (@option{-v})
4012 option specified, using the names of all the files in the practice
4013 directory as file name arguments:
4016 $ @kbd{tar --update -v -f collection.tar blues folk rock classical}
4023 Because we have specified verbose mode, @command{tar} prints out the names
4024 of the files it is working on, which in this case are the names of the
4025 files that needed to be updated. If you run @samp{tar --list} and look
4026 at the archive, you will see @file{blues} and @file{classical} at its
4027 end. There will be a total of two versions of the member @samp{blues};
4028 the one at the end will be newer and larger, since you added text before
4031 (The reason @command{tar} does not overwrite the older file when updating
4032 it is because writing to the middle of a section of tape is a difficult
4033 process. Tapes are not designed to go backward. @xref{Media}, for more
4034 information about tapes.
4036 @option{--update} (@option{-u}) is not suitable for performing backups for two
4037 reasons: it does not change directory content entries, and it
4038 lengthens the archive every time it is used. The @GNUTAR{}
4039 options intended specifically for backups are more
4040 efficient. If you need to run backups, please consult @ref{Backups}.
4043 @subsection Combining Archives with @option{--concatenate}
4045 @cindex Adding archives to an archive
4046 @cindex Concatenating Archives
4047 @opindex concatenate
4049 @c @cindex @option{-A} described
4050 Sometimes it may be convenient to add a second archive onto the end of
4051 an archive rather than adding individual files to the archive. To add
4052 one or more archives to the end of another archive, you should use the
4053 @option{--concatenate} (@option{--catenate}, @option{-A}) operation.
4055 To use @option{--concatenate}, give the first archive with
4056 @option{--file} option and name the rest of archives to be
4057 concatenated on the command line. The members, and their member
4058 names, will be copied verbatim from those archives to the first one.
4059 @FIXME-ref{This can cause multiple members to have the same name, for
4060 information on how this affects reading the archive, Multiple
4061 Members with the Same Name.}
4062 The new, concatenated archive will be called by the same name as the
4063 one given with the @option{--file} option. As usual, if you omit
4064 @option{--file}, @command{tar} will use the value of the environment
4065 variable @env{TAPE}, or, if this has not been set, the default archive name.
4067 @FIXME{There is no way to specify a new name...}
4069 To demonstrate how @option{--concatenate} works, create two small archives
4070 called @file{bluesrock.tar} and @file{folkjazz.tar}, using the relevant
4071 files from @file{practice}:
4074 $ @kbd{tar -cvf bluesrock.tar blues rock}
4077 $ @kbd{tar -cvf folkjazz.tar folk jazz}
4083 If you like, You can run @samp{tar --list} to make sure the archives
4084 contain what they are supposed to:
4087 $ @kbd{tar -tvf bluesrock.tar}
4088 -rw-r--r-- melissa user 105 1997-01-21 19:42 blues
4089 -rw-r--r-- melissa user 33 1997-01-20 15:34 rock
4090 $ @kbd{tar -tvf jazzfolk.tar}
4091 -rw-r--r-- melissa user 20 1996-09-23 16:44 folk
4092 -rw-r--r-- melissa user 65 1997-01-30 14:15 jazz
4095 We can concatenate these two archives with @command{tar}:
4099 $ @kbd{tar --concatenate --file=bluesrock.tar jazzfolk.tar}
4102 If you now list the contents of the @file{bluesrock.tar}, you will see
4103 that now it also contains the archive members of @file{jazzfolk.tar}:
4106 $ @kbd{tar --list --file=bluesrock.tar}
4113 When you use @option{--concatenate}, the source and target archives must
4114 already exist and must have been created using compatible format
4115 parameters. Notice, that @command{tar} does not check whether the
4116 archives it concatenates have compatible formats, it does not
4117 even check if the files are really tar archives.
4119 Like @option{--append} (@option{-r}), this operation cannot be performed on some
4120 tape drives, due to deficiencies in the formats those tape drives use.
4122 @cindex @code{concatenate} vs @command{cat}
4123 @cindex @command{cat} vs @code{concatenate}
4124 It may seem more intuitive to you to want or try to use @command{cat} to
4125 concatenate two archives instead of using the @option{--concatenate}
4126 operation; after all, @command{cat} is the utility for combining files.
4128 However, @command{tar} archives incorporate an end-of-file marker which
4129 must be removed if the concatenated archives are to be read properly as
4130 one archive. @option{--concatenate} removes the end-of-archive marker
4131 from the target archive before each new archive is appended. If you use
4132 @command{cat} to combine the archives, the result will not be a valid
4133 @command{tar} format archive. If you need to retrieve files from an
4134 archive that was added to using the @command{cat} utility, use the
4135 @option{--ignore-zeros} (@option{-i}) option. @xref{Ignore Zeros}, for further
4136 information on dealing with archives improperly combined using the
4137 @command{cat} shell utility.
4140 @subsection Removing Archive Members Using @option{--delete}
4142 @cindex Deleting files from an archive
4143 @cindex Removing files from an archive
4146 You can remove members from an archive by using the @option{--delete}
4147 option. Specify the name of the archive with @option{--file}
4148 (@option{-f}) and then specify the names of the members to be deleted;
4149 if you list no member names, nothing will be deleted. The
4150 @option{--verbose} option will cause @command{tar} to print the names
4151 of the members as they are deleted. As with @option{--extract}, you
4152 must give the exact member names when using @samp{tar --delete}.
4153 @option{--delete} will remove all versions of the named file from the
4154 archive. The @option{--delete} operation can run very slowly.
4156 Unlike other operations, @option{--delete} has no short form.
4158 @cindex Tapes, using @option{--delete} and
4159 @cindex Deleting from tape archives
4160 This operation will rewrite the archive. You can only use
4161 @option{--delete} on an archive if the archive device allows you to
4162 write to any point on the media, such as a disk; because of this, it
4163 does not work on magnetic tapes. Do not try to delete an archive member
4164 from a magnetic tape; the action will not succeed, and you will be
4165 likely to scramble the archive and damage your tape. There is no safe
4166 way (except by completely re-writing the archive) to delete files from
4167 most kinds of magnetic tape. @xref{Media}.
4169 To delete all versions of the file @file{blues} from the archive
4170 @file{collection.tar} in the @file{practice} directory, make sure you
4171 are in that directory, and then,
4174 $ @kbd{tar --list --file=collection.tar}
4179 $ @kbd{tar --delete --file=collection.tar blues}
4180 $ @kbd{tar --list --file=collection.tar}
4187 @FIXME{Check if the above listing is actually produced after running
4188 all the examples on collection.tar.}
4190 The @option{--delete} option has been reported to work properly when
4191 @command{tar} acts as a filter from @code{stdin} to @code{stdout}.
4194 @subsection Comparing Archive Members with the File System
4195 @cindex Verifying the currency of an archive
4199 The @option{--compare} (@option{-d}), or @option{--diff} operation compares
4200 specified archive members against files with the same names, and then
4201 reports differences in file size, mode, owner, modification date and
4202 contents. You should @emph{only} specify archive member names, not file
4203 names. If you do not name any members, then @command{tar} will compare the
4204 entire archive. If a file is represented in the archive but does not
4205 exist in the file system, @command{tar} reports a difference.
4207 You have to specify the record size of the archive when modifying an
4208 archive with a non-default record size.
4210 @command{tar} ignores files in the file system that do not have
4211 corresponding members in the archive.
4213 The following example compares the archive members @file{rock},
4214 @file{blues} and @file{funk} in the archive @file{bluesrock.tar} with
4215 files of the same name in the file system. (Note that there is no file,
4216 @file{funk}; @command{tar} will report an error message.)
4219 $ @kbd{tar --compare --file=bluesrock.tar rock blues funk}
4222 tar: funk not found in archive
4225 The spirit behind the @option{--compare} (@option{--diff},
4226 @option{-d}) option is to check whether the archive represents the
4227 current state of files on disk, more than validating the integrity of
4228 the archive media. For this later goal, @xref{verify}.
4230 @node create options
4231 @section Options Used by @option{--create}
4233 @opindex create, additional options
4234 The previous chapter described the basics of how to use
4235 @option{--create} (@option{-c}) to create an archive from a set of files.
4236 @xref{create}. This section described advanced options to be used with
4240 * Ignore Failed Read::
4243 @node Ignore Failed Read
4244 @subsection Ignore Fail Read
4247 @item --ignore-failed-read
4248 Do not exit with nonzero on unreadable files or directories.
4251 @node extract options
4252 @section Options Used by @option{--extract}
4255 @opindex extract, additional options
4256 The previous chapter showed how to use @option{--extract} to extract
4257 an archive into the file system. Various options cause @command{tar} to
4258 extract more information than just file contents, such as the owner,
4259 the permissions, the modification date, and so forth. This section
4260 presents options to be used with @option{--extract} when certain special
4261 considerations arise. You may review the information presented in
4262 @ref{extract} for more basic information about the
4263 @option{--extract} operation.
4266 * Reading:: Options to Help Read Archives
4267 * Writing:: Changing How @command{tar} Writes Files
4268 * Scarce:: Coping with Scarce Resources
4272 @subsection Options to Help Read Archives
4273 @cindex Options when reading archives
4276 @cindex Reading incomplete records
4277 @cindex Records, incomplete
4278 @opindex read-full-records
4279 Normally, @command{tar} will request data in full record increments from
4280 an archive storage device. If the device cannot return a full record,
4281 @command{tar} will report an error. However, some devices do not always
4282 return full records, or do not require the last record of an archive to
4283 be padded out to the next record boundary. To keep reading until you
4284 obtain a full record, or to accept an incomplete record if it contains
4285 an end-of-archive marker, specify the @option{--read-full-records} (@option{-B}) option
4286 in conjunction with the @option{--extract} or @option{--list} operations.
4289 The @option{--read-full-records} (@option{-B}) option is turned on by default when
4290 @command{tar} reads an archive from standard input, or from a remote
4291 machine. This is because on BSD Unix systems, attempting to read a
4292 pipe returns however much happens to be in the pipe, even if it is
4293 less than was requested. If this option were not enabled, @command{tar}
4294 would fail as soon as it read an incomplete record from the pipe.
4296 If you're not sure of the blocking factor of an archive, you can
4297 read the archive by specifying @option{--read-full-records} (@option{-B}) and
4298 @option{--blocking-factor=@var{512-size}} (@option{-b
4299 @var{512-size}}), using a blocking factor larger than what the archive
4300 uses. This lets you avoid having to determine the blocking factor
4301 of an archive. @xref{Blocking Factor}.
4304 * read full records::
4308 @node read full records
4309 @unnumberedsubsubsec Reading Full Records
4311 @FIXME{need sentence or so of intro here}
4314 @opindex read-full-records
4315 @item --read-full-records
4317 Use in conjunction with @option{--extract} (@option{--get},
4318 @option{-x}) to read an archive which contains incomplete records, or
4319 one which has a blocking factor less than the one specified.
4323 @unnumberedsubsubsec Ignoring Blocks of Zeros
4325 @cindex End-of-archive blocks, ignoring
4326 @cindex Ignoring end-of-archive blocks
4327 @opindex ignore-zeros
4328 Normally, @command{tar} stops reading when it encounters a block of zeros
4329 between file entries (which usually indicates the end of the archive).
4330 @option{--ignore-zeros} (@option{-i}) allows @command{tar} to
4331 completely read an archive which contains a block of zeros before the
4332 end (i.e., a damaged archive, or one that was created by concatenating
4333 several archives together).
4335 The @option{--ignore-zeros} (@option{-i}) option is turned off by default because many
4336 versions of @command{tar} write garbage after the end-of-archive entry,
4337 since that part of the media is never supposed to be read. @GNUTAR{}
4338 does not write after the end of an archive, but seeks to
4339 maintain compatiblity among archiving utilities.
4342 @item --ignore-zeros
4344 To ignore blocks of zeros (i.e., end-of-archive entries) which may be
4345 encountered while reading an archive. Use in conjunction with
4346 @option{--extract} or @option{--list}.
4350 @subsection Changing How @command{tar} Writes Files
4353 @FIXME{Introductory paragraph}
4356 * Dealing with Old Files::
4357 * Overwrite Old Files::
4359 * Keep Newer Files::
4361 * Recursive Unlink::
4362 * Data Modification Times::
4363 * Setting Access Permissions::
4364 * Directory Modification Times and Permissions::
4365 * Writing to Standard Output::
4366 * Writing to an External Program::
4370 @node Dealing with Old Files
4371 @unnumberedsubsubsec Options Controlling the Overwriting of Existing Files
4373 @opindex overwrite-dir, introduced
4374 When extracting files, if @command{tar} discovers that the extracted
4375 file already exists, it normally replaces the file by removing it before
4376 extracting it, to prevent confusion in the presence of hard or symbolic
4377 links. (If the existing file is a symbolic link, it is removed, not
4378 followed.) However, if a directory cannot be removed because it is
4379 nonempty, @command{tar} normally overwrites its metadata (ownership,
4380 permission, etc.). The @option{--overwrite-dir} option enables this
4381 default behavior. To be more cautious and preserve the metadata of
4382 such a directory, use the @option{--no-overwrite-dir} option.
4384 @cindex Overwriting old files, prevention
4385 @opindex keep-old-files, introduced
4386 To be even more cautious and prevent existing files from being replaced, use
4387 the @option{--keep-old-files} (@option{-k}) option. It causes @command{tar} to refuse
4388 to replace or update a file that already exists, i.e., a file with the
4389 same name as an archive member prevents extraction of that archive
4390 member. Instead, it reports an error.
4392 @opindex overwrite, introduced
4393 To be more aggressive about altering existing files, use the
4394 @option{--overwrite} option. It causes @command{tar} to overwrite
4395 existing files and to follow existing symbolic links when extracting.
4397 @cindex Protecting old files
4398 Some people argue that @GNUTAR{} should not hesitate
4399 to overwrite files with other files when extracting. When extracting
4400 a @command{tar} archive, they expect to see a faithful copy of the
4401 state of the file system when the archive was created. It is debatable
4402 that this would always be a proper behavior. For example, suppose one
4403 has an archive in which @file{usr/local} is a link to
4404 @file{usr/local2}. Since then, maybe the site removed the link and
4405 renamed the whole hierarchy from @file{/usr/local2} to
4406 @file{/usr/local}. Such things happen all the time. I guess it would
4407 not be welcome at all that @GNUTAR{} removes the
4408 whole hierarchy just to make room for the link to be reinstated
4409 (unless it @emph{also} simultaneously restores the full
4410 @file{/usr/local2}, of course!) @GNUTAR{} is indeed
4411 able to remove a whole hierarchy to reestablish a symbolic link, for
4412 example, but @emph{only if} @option{--recursive-unlink} is specified
4413 to allow this behavior. In any case, single files are silently
4416 @opindex unlink-first, introduced
4417 Finally, the @option{--unlink-first} (@option{-U}) option can improve performance in
4418 some cases by causing @command{tar} to remove files unconditionally
4419 before extracting them.
4421 @node Overwrite Old Files
4422 @unnumberedsubsubsec Overwrite Old Files
4427 Overwrite existing files and directory metadata when extracting files
4430 This causes @command{tar} to write extracted files into the file system without
4431 regard to the files already on the system; i.e., files with the same
4432 names as archive members are overwritten when the archive is extracted.
4433 It also causes @command{tar} to extract the ownership, permissions,
4434 and time stamps onto any preexisting files or directories.
4435 If the name of a corresponding file name is a symbolic link, the file
4436 pointed to by the symbolic link will be overwritten instead of the
4437 symbolic link itself (if this is possible). Moreover, special devices,
4438 empty directories and even symbolic links are automatically removed if
4439 they are in the way of extraction.
4441 Be careful when using the @option{--overwrite} option, particularly when
4442 combined with the @option{--absolute-names} (@option{-P}) option, as this combination
4443 can change the contents, ownership or permissions of any file on your
4444 system. Also, many systems do not take kindly to overwriting files that
4445 are currently being executed.
4447 @opindex overwrite-dir
4448 @item --overwrite-dir
4449 Overwrite the metadata of directories when extracting files from an
4450 archive, but remove other files before extracting.
4453 @node Keep Old Files
4454 @unnumberedsubsubsec Keep Old Files
4457 @opindex keep-old-files
4458 @item --keep-old-files
4460 Do not replace existing files from archive. The
4461 @option{--keep-old-files} (@option{-k}) option prevents @command{tar}
4462 from replacing existing files with files with the same name from the
4463 archive. The @option{--keep-old-files} option is meaningless with
4464 @option{--list} (@option{-t}). Prevents @command{tar} from replacing
4465 files in the file system during extraction.
4468 @node Keep Newer Files
4469 @unnumberedsubsubsec Keep Newer Files
4472 @opindex keep-newer-files
4473 @item --keep-newer-files
4474 Do not replace existing files that are newer than their archive
4475 copies. This option is meaningless with @option{--list} (@option{-t}).
4479 @unnumberedsubsubsec Unlink First
4482 @opindex unlink-first
4483 @item --unlink-first
4485 Remove files before extracting over them.
4486 This can make @command{tar} run a bit faster if you know in advance
4487 that the extracted files all need to be removed. Normally this option
4488 slows @command{tar} down slightly, so it is disabled by default.
4491 @node Recursive Unlink
4492 @unnumberedsubsubsec Recursive Unlink
4495 @opindex recursive-unlink
4496 @item --recursive-unlink
4497 When this option is specified, try removing files and directory hierarchies
4498 before extracting over them. @emph{This is a dangerous option!}
4501 If you specify the @option{--recursive-unlink} option,
4502 @command{tar} removes @emph{anything} that keeps you from extracting a file
4503 as far as current permissions will allow it. This could include removal
4504 of the contents of a full directory hierarchy.
4506 @node Data Modification Times
4507 @unnumberedsubsubsec Setting Data Modification Times
4509 @cindex Data modification times of extracted files
4510 @cindex Modification times of extracted files
4511 Normally, @command{tar} sets the data modification times of extracted
4512 files to the corresponding times recorded for the files in the archive, but
4513 limits the permissions of extracted files by the current @code{umask}
4516 To set the data modification times of extracted files to the time when
4517 the files were extracted, use the @option{--touch} (@option{-m}) option in
4518 conjunction with @option{--extract} (@option{--get}, @option{-x}).
4524 Sets the data modification time of extracted archive members to the time
4525 they were extracted, not the time recorded for them in the archive.
4526 Use in conjunction with @option{--extract} (@option{--get}, @option{-x}).
4529 @node Setting Access Permissions
4530 @unnumberedsubsubsec Setting Access Permissions
4532 @cindex Permissions of extracted files
4533 @cindex Modes of extracted files
4534 To set the modes (access permissions) of extracted files to those
4535 recorded for those files in the archive, use @option{--same-permissions}
4536 in conjunction with the @option{--extract} (@option{--get},
4537 @option{-x}) operation.
4540 @opindex preserve-permission
4541 @opindex same-permission
4542 @item --preserve-permission
4543 @itemx --same-permission
4544 @c @itemx --ignore-umask
4546 Set modes of extracted archive members to those recorded in the
4547 archive, instead of current umask settings. Use in conjunction with
4548 @option{--extract} (@option{--get}, @option{-x}).
4551 @node Directory Modification Times and Permissions
4552 @unnumberedsubsubsec Directory Modification Times and Permissions
4554 After sucessfully extracting a file member, @GNUTAR{} normally
4555 restores its permissions and modification times, as described in the
4556 previous sections. This cannot be done for directories, because
4557 after extracting a directory @command{tar} will almost certainly
4558 extract files into that directory and this will cause the directory
4559 modification time to be updated. Moreover, restoring that directory
4560 permissions may not permit file creation within it. Thus, restoring
4561 directory permissions and modification times must be delayed at least
4562 until all files have been extracted into that directory. @GNUTAR{}
4563 restores directories using the following approach.
4565 The extracted directories are created with the mode specified in the
4566 archive, as modified by the umask of the user, which gives sufficient
4567 permissions to allow file creation. The meta-information about the
4568 directory is recorded in the temporary list of directories. When
4569 preparing to extract next archive member, @GNUTAR{} checks if the
4570 directory prefix of this file contains the remembered directory. If
4571 it does not, the program assumes that all files have been extracted
4572 into that directory, restores its modification time and permissions
4573 and removes its entry from the internal list. This approach allows
4574 to correctly restore directory meta-information in the majority of
4575 cases, while keeping memory requirements sufficiently small. It is
4576 based on the fact, that most @command{tar} archives use the predefined
4577 order of members: first the directory, then all the files and
4578 subdirectories in that directory.
4580 However, this is not always true. The most important exception are
4581 incremental archives (@pxref{Incremental Dumps}). The member order in
4582 an incremental archive is reversed: first all directory members are
4583 stored, followed by other (non-directory) members. So, when extracting
4584 from incremental archives, @GNUTAR{} alters the above procedure. It
4585 remebers all restored directories, and restores their meta-data
4586 only after the entire archive has been processed. Notice, that you do
4587 not need to specity any special options for that, as @GNUTAR{}
4588 automatically detects archives in incremental format.
4590 There may be cases, when such processing is required for normal archives
4591 too. Consider the following example:
4595 $ @kbd{tar --no-recursion -cvf archive \
4596 foo foo/file1 bar bar/file foo/file2}
4605 During the normal operation, after encountering @file{bar}
4606 @GNUTAR{} will assume that all files from the directory @file{foo}
4607 were already extracted and will therefore restore its timestamp and
4608 permission bits. However, after extracting @file{foo/file2} the
4609 directory timestamp will be offset again.
4611 To correctly restore directory meta-information in such cases, use
4612 @option{delay-directory-restore} command line option:
4615 @opindex delay-directory-restore
4616 @item --delay-directory-restore
4617 Delays restoring of the modification times and permissions of extracted
4618 directories until the end of extraction. This way, correct
4619 meta-information is restored even if the archive has unusual member
4622 @opindex no-delay-directory-restore
4623 @item --no-delay-directory-restore
4624 Cancel the effect of the previous @option{--delay-directory-restore}.
4625 Use this option if you have used @option{--delay-directory-restore} in
4626 @env{TAR_OPTIONS} variable (@pxref{TAR_OPTIONS}) and wish to
4627 temporarily disable it.
4630 @node Writing to Standard Output
4631 @unnumberedsubsubsec Writing to Standard Output
4633 @cindex Writing extracted files to standard output
4634 @cindex Standard output, writing extracted files to
4635 To write the extracted files to the standard output, instead of
4636 creating the files on the file system, use @option{--to-stdout} (@option{-O}) in
4637 conjunction with @option{--extract} (@option{--get}, @option{-x}). This option is useful if you are
4638 extracting files to send them through a pipe, and do not need to
4639 preserve them in the file system. If you extract multiple members,
4640 they appear on standard output concatenated, in the order they are
4641 found in the archive.
4647 Writes files to the standard output. Use only in conjunction with
4648 @option{--extract} (@option{--get}, @option{-x}). When this option is
4649 used, instead of creating the files specified, @command{tar} writes
4650 the contents of the files extracted to its standard output. This may
4651 be useful if you are only extracting the files in order to send them
4652 through a pipe. This option is meaningless with @option{--list}
4656 This can be useful, for example, if you have a tar archive containing
4657 a big file and don't want to store the file on disk before processing
4658 it. You can use a command like this:
4661 tar -xOzf foo.tgz bigfile | process
4664 or even like this if you want to process the concatenation of the files:
4667 tar -xOzf foo.tgz bigfile1 bigfile2 | process
4670 Hovewer, @option{--to-command} may be more convenient for use with
4671 multiple files. See the next section.
4673 @node Writing to an External Program
4674 @unnumberedsubsubsec Writing to an External Program
4676 You can instruct @command{tar} to send the contents of each extracted
4677 file to the standard input of an external program:
4681 @item --to-command=@var{command}
4682 Extract files and pipe their contents to the standard input of
4683 @var{command}. When this option is used, instead of creating the
4684 files specified, @command{tar} invokes @var{command} and pipes the
4685 contents of the files to its standard output. @var{Command} may
4686 contain command line arguments. The program is executed via
4687 @code{sh -c}. Notice, that @var{command} is executed once for each regular file
4688 extracted. Non-regular files (directories, etc.) are ignored when this
4692 The command can obtain the information about the file it processes
4693 from the following environment variables:
4696 @vrindex TAR_FILETYPE, to-command environment
4698 Type of the file. It is a single letter with the following meaning:
4700 @multitable @columnfractions 0.10 0.90
4701 @item f @tab Regular file
4702 @item d @tab Directory
4703 @item l @tab Symbolic link
4704 @item h @tab Hard link
4705 @item b @tab Block device
4706 @item c @tab Character device
4709 Currently only regular files are supported.
4711 @vrindex TAR_MODE, to-command environment
4713 File mode, an octal number.
4715 @vrindex TAR_FILENAME, to-command environment
4717 The name of the file.
4719 @vrindex TAR_REALNAME, to-command environment
4721 Name of the file as stored in the archive.
4723 @vrindex TAR_UNAME, to-command environment
4725 Name of the file owner.
4727 @vrindex TAR_GNAME, to-command environment
4729 Name of the file owner group.
4731 @vrindex TAR_ATIME, to-command environment
4733 Time of last access. It is a decimal number, representing seconds
4734 since the epoch. If the archive provides times with nanosecond
4735 precision, the nanoseconds are appended to the timestamp after a
4738 @vrindex TAR_MTIME, to-command environment
4740 Time of last modification.
4742 @vrindex TAR_CTIME, to-command environment
4744 Time of last status change.
4746 @vrindex TAR_SIZE, to-command environment
4750 @vrindex TAR_UID, to-command environment
4752 UID of the file owner.
4754 @vrindex TAR_GID, to-command environment
4756 GID of the file owner.
4759 In addition to these variables, @env{TAR_VERSION} contains the
4760 @GNUTAR{} version number.
4762 If @var{command} exits with a non-0 status, @command{tar} will print
4763 an error message similar to the following:
4766 tar: 2345: Child returned status 1
4769 Here, @samp{2345} is the PID of the finished process.
4771 If this behavior is not wanted, use @option{--ignore-command-error}:
4774 @opindex ignore-command-error
4775 @item --ignore-command-error
4776 Ignore exit codes of subprocesses. Notice that if the program
4777 exits on signal or otherwise terminates abnormally, the error message
4778 will be printed even if this option is used.
4780 @opindex no-ignore-command-error
4781 @item --no-ignore-command-error
4782 Cancel the effect of any previous @option{--ignore-command-error}
4783 option. This option is useful if you have set
4784 @option{--ignore-command-error} in @env{TAR_OPTIONS}
4785 (@pxref{TAR_OPTIONS}) and wish to temporarily cancel it.
4789 @unnumberedsubsubsec Removing Files
4791 @FIXME{The section is too terse. Something more to add? An example,
4795 @opindex remove-files
4796 @item --remove-files
4797 Remove files after adding them to the archive.
4801 @subsection Coping with Scarce Resources
4804 @cindex Small memory
4805 @cindex Running out of space
4813 @unnumberedsubsubsec Starting File
4816 @opindex starting-file
4817 @item --starting-file=@var{name}
4818 @itemx -K @var{name}
4819 Starts an operation in the middle of an archive. Use in conjunction
4820 with @option{--extract} (@option{--get}, @option{-x}) or @option{--list} (@option{-t}).
4823 @cindex Middle of the archive, starting in the
4824 If a previous attempt to extract files failed due to lack of disk
4825 space, you can use @option{--starting-file=@var{name}} (@option{-K
4826 @var{name}}) to start extracting only after member @var{name} of the
4827 archive. This assumes, of course, that there is now free space, or
4828 that you are now extracting into a different file system. (You could
4829 also choose to suspend @command{tar}, remove unnecessary files from
4830 the file system, and then restart the same @command{tar} operation.
4831 In this case, @option{--starting-file} is not necessary.
4832 @xref{Incremental Dumps}, @xref{interactive}, and @ref{exclude}.)
4835 @unnumberedsubsubsec Same Order
4838 @cindex Large lists of file names on small machines
4840 @opindex preserve-order
4842 @itemx --preserve-order
4844 To process large lists of file names on machines with small amounts of
4845 memory. Use in conjunction with @option{--compare} (@option{--diff},
4846 @option{-d}), @option{--list} (@option{-t}) or @option{--extract}
4847 (@option{--get}, @option{-x}).
4850 The @option{--same-order} (@option{--preserve-order}, @option{-s}) option tells @command{tar} that the list of file
4851 names to be listed or extracted is sorted in the same order as the
4852 files in the archive. This allows a large list of names to be used,
4853 even on a small machine that would not otherwise be able to hold all
4854 the names in memory at the same time. Such a sorted list can easily be
4855 created by running @samp{tar -t} on the archive and editing its output.
4857 This option is probably never needed on modern computer systems.
4860 @section Backup options
4862 @cindex backup options
4864 @GNUTAR{} offers options for making backups of files
4865 before writing new versions. These options control the details of
4866 these backups. They may apply to the archive itself before it is
4867 created or rewritten, as well as individual extracted members. Other
4868 @acronym{GNU} programs (@command{cp}, @command{install}, @command{ln},
4869 and @command{mv}, for example) offer similar options.
4871 Backup options may prove unexpectedly useful when extracting archives
4872 containing many members having identical name, or when extracting archives
4873 on systems having file name limitations, making different members appear
4874 has having similar names through the side-effect of name truncation.
4875 (This is true only if we have a good scheme for truncated backup names,
4876 which I'm not sure at all: I suspect work is needed in this area.)
4877 When any existing file is backed up before being overwritten by extraction,
4878 then clashing files are automatically be renamed to be unique, and the
4879 true name is kept for only the last file of a series of clashing files.
4880 By using verbose mode, users may track exactly what happens.
4882 At the detail level, some decisions are still experimental, and may
4883 change in the future, we are waiting comments from our users. So, please
4884 do not learn to depend blindly on the details of the backup features.
4885 For example, currently, directories themselves are never renamed through
4886 using these options, so, extracting a file over a directory still has
4887 good chances to fail. Also, backup options apply to created archives,
4888 not only to extracted members. For created archives, backups will not
4889 be attempted when the archive is a block or character device, or when it
4890 refers to a remote file.
4892 For the sake of simplicity and efficiency, backups are made by renaming old
4893 files prior to creation or extraction, and not by copying. The original
4894 name is restored if the file creation fails. If a failure occurs after a
4895 partial extraction of a file, both the backup and the partially extracted
4899 @item --backup[=@var{method}]
4901 @vindex VERSION_CONTROL
4903 Back up files that are about to be overwritten or removed.
4904 Without this option, the original versions are destroyed.
4906 Use @var{method} to determine the type of backups made.
4907 If @var{method} is not specified, use the value of the @env{VERSION_CONTROL}
4908 environment variable. And if @env{VERSION_CONTROL} is not set,
4909 use the @samp{existing} method.
4911 @vindex version-control @r{Emacs variable}
4912 This option corresponds to the Emacs variable @samp{version-control};
4913 the same values for @var{method} are accepted as in Emacs. This option
4914 also allows more descriptive names. The valid @var{method}s are:
4919 @cindex numbered @r{backup method}
4920 Always make numbered backups.
4924 @cindex existing @r{backup method}
4925 Make numbered backups of files that already have them, simple backups
4930 @cindex simple @r{backup method}
4931 Always make simple backups.
4935 @item --suffix=@var{suffix}
4937 @cindex backup suffix
4938 @vindex SIMPLE_BACKUP_SUFFIX
4939 Append @var{suffix} to each backup file made with @option{--backup}. If this
4940 option is not specified, the value of the @env{SIMPLE_BACKUP_SUFFIX}
4941 environment variable is used. And if @env{SIMPLE_BACKUP_SUFFIX} is not
4942 set, the default is @samp{~}, just as in Emacs.
4946 Some people express the desire to @emph{always} use the @option{--backup}
4947 option, by defining some kind of alias or script. This is not as easy
4948 as one may think, due to the fact that old style options should appear first
4949 and consume arguments a bit unpredictably for an alias or script. But,
4950 if you are ready to give up using old style options, you may resort to
4951 using something like (a Bourne shell function here):
4954 tar () @{ /usr/local/bin/tar --backup $*; @}
4958 @section Notable @command{tar} Usages
4961 @FIXME{Using Unix file linking capability to recreate directory
4962 structures---linking files into one subdirectory and then
4963 @command{tar}ring that directory.}
4965 @FIXME{Nice hairy example using absolute-names, newer, etc.}
4968 You can easily use archive files to transport a group of files from
4969 one system to another: put all relevant files into an archive on one
4970 computer system, transfer the archive to another system, and extract
4971 the contents there. The basic transfer medium might be magnetic tape,
4972 Internet FTP, or even electronic mail (though you must encode the
4973 archive with @command{uuencode} in order to transport it properly by
4974 mail). Both machines do not have to use the same operating system, as
4975 long as they both support the @command{tar} program.
4977 For example, here is how you might copy a directory's contents from
4978 one disk to another, while preserving the dates, modes, owners and
4979 link-structure of all the files therein. In this case, the transfer
4980 medium is a @dfn{pipe}, which is one a Unix redirection mechanism:
4983 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xf -)}
4987 You can avoid subshells by using @option{-C} option:
4990 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xf -}
4994 The command also works using short option forms:
4997 $ @kbd{(cd sourcedir; tar --create --file=- . ) \
4998 | (cd targetdir; tar --extract --file=-)}
5000 $ @kbd{tar --directory sourcedir --create --file=- . ) \
5001 | tar --directory targetdir --extract --file=-}
5005 This is one of the easiest methods to transfer a @command{tar} archive.
5008 @section Looking Ahead: The Rest of this Manual
5010 You have now seen how to use all eight of the operations available to
5011 @command{tar}, and a number of the possible options. The next chapter
5012 explains how to choose and change file and archive names, how to use
5013 files to store names of other files which you can then call as
5014 arguments to @command{tar} (this can help you save time if you expect to
5015 archive the same list of files a number of times), and so forth.
5016 @FIXME{in case it's not obvious, i'm making this up in some sense
5017 based on my limited memory of what the next chapter *really* does. i
5018 just wanted to flesh out this final section a little bit so i'd
5019 remember to stick it in here. :-)}
5021 If there are too many files to conveniently list on the command line,
5022 you can list the names in a file, and @command{tar} will read that file.
5025 There are various ways of causing @command{tar} to skip over some files,
5026 and not archive them. @xref{Choosing}.
5029 @chapter Performing Backups and Restoring Files
5032 @GNUTAR{} is distributed along with the scripts
5033 which the Free Software Foundation uses for performing backups. There
5034 is no corresponding scripts available yet for doing restoration of
5035 files. Even if there is a good chance those scripts may be satisfying
5036 to you, they are not the only scripts or methods available for doing
5037 backups and restore. You may well create your own, or use more
5038 sophisticated packages dedicated to that purpose.
5040 Some users are enthusiastic about @code{Amanda} (The Advanced Maryland
5041 Automatic Network Disk Archiver), a backup system developed by James
5042 da Silva @file{jds@@cs.umd.edu} and available on many Unix systems.
5043 This is free software, and it is available at these places:
5046 http://www.cs.umd.edu/projects/amanda/amanda.html
5047 ftp://ftp.cs.umd.edu/pub/amanda
5052 Here is a possible plan for a future documentation about the backuping
5053 scripts which are provided within the @GNUTAR{}
5059 @item what are dumps
5060 @item different levels of dumps
5062 @item full dump = dump everything
5063 @item level 1, level 2 dumps etc
5064 A level @var{n} dump dumps everything changed since the last level
5067 @item how to use scripts for dumps (ie, the concept)
5069 @item scripts to run after editing backup specs (details)
5071 @item Backup Specs, what is it.
5073 @item how to customize
5074 @item actual text of script [/sp/dump/backup-specs]
5078 @item rsh doesn't work
5079 @item rtape isn't installed
5082 @item the @option{--incremental} option of tar
5085 @item write protection
5086 @item types of media, different sizes and types, useful for different things
5087 @item files and tape marks
5088 one tape mark between files, two at end.
5089 @item positioning the tape
5090 MT writes two at end of write,
5091 backspaces over one when writing again.
5097 This chapter documents both the provided shell scripts and @command{tar}
5098 options which are more specific to usage as a backup tool.
5100 To @dfn{back up} a file system means to create archives that contain
5101 all the files in that file system. Those archives can then be used to
5102 restore any or all of those files (for instance if a disk crashes or a
5103 file is accidentally deleted). File system @dfn{backups} are also
5107 * Full Dumps:: Using @command{tar} to Perform Full Dumps
5108 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
5109 * Backup Levels:: Levels of Backups
5110 * Backup Parameters:: Setting Parameters for Backups and Restoration
5111 * Scripted Backups:: Using the Backup Scripts
5112 * Scripted Restoration:: Using the Restore Script
5116 @section Using @command{tar} to Perform Full Dumps
5122 @cindex corrupted archives
5123 Full dumps should only be made when no other people or programs
5124 are modifying files in the file system. If files are modified while
5125 @command{tar} is making the backup, they may not be stored properly in
5126 the archive, in which case you won't be able to restore them if you
5127 have to. (Files not being modified are written with no trouble, and do
5128 not corrupt the entire archive.)
5130 You will want to use the @option{--label=@var{archive-label}}
5131 (@option{-V @var{archive-label}}) option to give the archive a
5132 volume label, so you can tell what this archive is even if the label
5133 falls off the tape, or anything like that.
5135 Unless the file system you are dumping is guaranteed to fit on
5136 one volume, you will need to use the @option{--multi-volume} (@option{-M}) option.
5137 Make sure you have enough tapes on hand to complete the backup.
5139 If you want to dump each file system separately you will need to use
5140 the @option{--one-file-system} option to prevent
5141 @command{tar} from crossing file system boundaries when storing
5144 The @option{--incremental} (@option{-G}) (@pxref{Incremental Dumps})
5145 option is not needed, since this is a complete copy of everything in
5146 the file system, and a full restore from this backup would only be
5147 done onto a completely
5150 Unless you are in a hurry, and trust the @command{tar} program (and your
5151 tapes), it is a good idea to use the @option{--verify} (@option{-W})
5152 option, to make sure your files really made it onto the dump properly.
5153 This will also detect cases where the file was modified while (or just
5154 after) it was being archived. Not all media (notably cartridge tapes)
5155 are capable of being verified, unfortunately.
5157 @node Incremental Dumps
5158 @section Using @command{tar} to Perform Incremental Dumps
5160 @dfn{Incremental backup} is a special form of @GNUTAR{} archive that
5161 stores additional metadata so that exact state of the file system
5162 can be restored when extracting the archive.
5164 @GNUTAR{} currently offers two options for handling incremental
5165 backups: @option{--listed-incremental=@var{snapshot-file}} (@option{-g
5166 @var{snapshot-file}}) and @option{--incremental} (@option{-G}).
5168 @opindex listed-incremental
5169 The option @option{--listed-incremental} instructs tar to operate on
5170 an incremental archive with additional metadata stored in a standalone
5171 file, called a @dfn{snapshot file}. The purpose of this file is to help
5172 determine which files have been changed, added or deleted since the
5173 last backup, so that the next incremental backup will contain only
5174 modified files. The name of the snapshot file is given as an argument
5178 @item --listed-incremental=@var{file}
5179 @itemx -g @var{file}
5180 Handle incremental backups with snapshot data in @var{file}.
5183 To create an incremental backup, you would use
5184 @option{--listed-incremental} together with @option{--create}
5185 (@pxref{create}). For example:
5188 $ @kbd{tar --create \
5189 --file=archive.1.tar \
5190 --listed-incremental=/var/log/usr.snar \
5194 This will create in @file{archive.1.tar} an incremental backup of
5195 the @file{/usr} file system, storing additional metadata in the file
5196 @file{/var/log/usr.snar}. If this file does not exist, it will be
5197 created. The created archive will then be a @dfn{level 0 backup};
5198 please see the next section for more on backup levels.
5200 Otherwise, if the file @file{/var/log/usr.snar} exists, it
5201 determines which files are modified. In this case only these files will be
5202 stored in the archive. Suppose, for example, that after running the
5203 above command, you delete file @file{/usr/doc/old} and create
5204 directory @file{/usr/local/db} with the following contents:
5207 $ @kbd{ls /usr/local/db}
5212 Some time later you create another incremental backup. You will
5216 $ @kbd{tar --create \
5217 --file=archive.2.tar \
5218 --listed-incremental=/var/log/usr.snar \
5220 tar: usr/local/db: Directory is new
5227 The created archive @file{archive.2.tar} will contain only these
5228 three members. This archive is called a @dfn{level 1 backup}. Notice
5229 that @file{/var/log/usr.snar} will be updated with the new data, so if
5230 you plan to create more @samp{level 1} backups, it is necessary to
5231 create a working copy of the snapshot file before running
5232 @command{tar}. The above example will then be modified as follows:
5235 $ @kbd{cp /var/log/usr.snar /var/log/usr.snar-1}
5236 $ @kbd{tar --create \
5237 --file=archive.2.tar \
5238 --listed-incremental=/var/log/usr.snar-1 \
5242 Incremental dumps depend crucially on time stamps, so the results are
5243 unreliable if you modify a file's time stamps during dumping (e.g.,
5244 with the @option{--atime-preserve=replace} option), or if you set the clock
5247 Metadata stored in snapshot files include device numbers, which,
5248 obviously is supposed to be a non-volatile value. However, it turns
5249 out that NFS devices have undependable values when an automounter
5250 gets in the picture. This can lead to a great deal of spurious
5251 redumping in incremental dumps, so it is somewhat useless to compare
5252 two NFS devices numbers over time. The solution implemented currently
5253 is to considers all NFS devices as being equal when it comes to
5254 comparing directories; this is fairly gross, but there does not seem
5255 to be a better way to go.
5257 Note that incremental archives use @command{tar} extensions and may
5258 not be readable by non-@acronym{GNU} versions of the @command{tar} program.
5260 @opindex listed-incremental, using with @option{--extract}
5261 @opindex extract, using with @option{--listed-incremental}
5262 To extract from the incremental dumps, use
5263 @option{--listed-incremental} together with @option{--extract}
5264 option (@pxref{extracting files}). In this case, @command{tar} does
5265 not need to access snapshot file, since all the data necessary for
5266 extraction are stored in the archive itself. So, when extracting, you
5267 can give whatever argument to @option{--listed-incremental}, the usual
5268 practice is to use @option{--listed-incremental=/dev/null}.
5269 Alternatively, you can use @option{--incremental}, which needs no
5270 arguments. In general, @option{--incremental} (@option{-G}) can be
5271 used as a shortcut for @option{--listed-incremental} when listing or
5272 extracting incremental backups (for more information, regarding this
5273 option, @pxref{incremental-op}).
5275 When extracting from the incremental backup @GNUTAR{} attempts to
5276 restore the exact state the file system had when the archive was
5277 created. In particular, it will @emph{delete} those files in the file
5278 system that did not exist in their directories when the archive was
5279 created. If you have created several levels of incremental files,
5280 then in order to restore the exact contents the file system had when
5281 the last level was created, you will need to restore from all backups
5282 in turn. Continuing our example, to restore the state of @file{/usr}
5283 file system, one would do@footnote{Notice, that since both archives
5284 were created withouth @option{-P} option (@pxref{absolute}), these
5285 commands should be run from the root file system.}:
5288 $ @kbd{tar --extract \
5289 --listed-incremental=/dev/null \
5290 --file archive.1.tar}
5291 $ @kbd{tar --extract \
5292 --listed-incremental=/dev/null \
5293 --file archive.2.tar}
5296 To list the contents of an incremental archive, use @option{--list}
5297 (@pxref{list}), as usual. To obtain more information about the
5298 archive, use @option{--listed-incremental} or @option{--incremental}
5299 combined with two @option{--verbose} options@footnote{Two
5300 @option{--verbose} options were selected to avoid breaking usual
5301 verbose listing output (@option{--list --verbose}) when using in
5304 @opindex incremental, using with @option{--list}
5305 @opindex listed-incremental, using with @option{--list}
5306 @opindex list, using with @option{--incremental}
5307 @opindex list, using with @option{--listed-incremental}
5308 Versions of @GNUTAR{} up to 1.15.1 used to dump verbatim binary
5309 contents of the DUMPDIR header (with terminating nulls) when
5310 @option{--incremental} or @option{--listed-incremental} option was
5311 given, no matter what the verbosity level. This behavior, and,
5312 especially, the binary output it produced were considered incovenient
5313 and were changed in version 1.16}:
5316 @kbd{tar --list --incremental --verbose --verbose archive.tar}
5319 This command will print, for each directory in the archive, the list
5320 of files in that directory at the time the archive was created. This
5321 information is put out in a format which is both human-readable and
5322 unambiguous for a program: each file name is printed as
5329 where @var{x} is a letter describing the status of the file: @samp{Y}
5330 if the file is present in the archive, @samp{N} if the file is not
5331 included in the archive, or a @samp{D} if the file is a directory (and
5332 is included in the archive). @xref{Dumpdir}, for the detailed
5333 description of dumpdirs and status codes. Each such
5334 line is terminated by a newline character. The last line is followed
5335 by an additional newline to indicate the end of the data.
5337 @anchor{incremental-op}The option @option{--incremental} (@option{-G})
5338 gives the same behavior as @option{--listed-incremental} when used
5339 with @option{--list} and @option{--extract} options. When used with
5340 @option{--create} option, it creates an incremental archive without
5341 creating snapshot file. Thus, it is impossible to create several
5342 levels of incremental backups with @option{--incremental} option.
5345 @section Levels of Backups
5347 An archive containing all the files in the file system is called a
5348 @dfn{full backup} or @dfn{full dump}. You could insure your data by
5349 creating a full dump every day. This strategy, however, would waste a
5350 substantial amount of archive media and user time, as unchanged files
5351 are daily re-archived.
5353 It is more efficient to do a full dump only occasionally. To back up
5354 files between full dumps, you can use @dfn{incremental dumps}. A @dfn{level
5355 one} dump archives all the files that have changed since the last full
5358 A typical dump strategy would be to perform a full dump once a week,
5359 and a level one dump once a day. This means some versions of files
5360 will in fact be archived more than once, but this dump strategy makes
5361 it possible to restore a file system to within one day of accuracy by
5362 only extracting two archives---the last weekly (full) dump and the
5363 last daily (level one) dump. The only information lost would be in
5364 files changed or created since the last daily backup. (Doing dumps
5365 more than once a day is usually not worth the trouble).
5367 @GNUTAR{} comes with scripts you can use to do full
5368 and level-one (actually, even level-two and so on) dumps. Using
5369 scripts (shell programs) to perform backups and restoration is a
5370 convenient and reliable alternative to typing out file name lists
5371 and @command{tar} commands by hand.
5373 Before you use these scripts, you need to edit the file
5374 @file{backup-specs}, which specifies parameters used by the backup
5375 scripts and by the restore script. This file is usually located
5376 in @file{/etc/backup} directory. @xref{Backup Parameters}, for its
5377 detailed description. Once the backup parameters are set, you can
5378 perform backups or restoration by running the appropriate script.
5380 The name of the backup script is @code{backup}. The name of the
5381 restore script is @code{restore}. The following sections describe
5382 their use in detail.
5384 @emph{Please Note:} The backup and restoration scripts are
5385 designed to be used together. While it is possible to restore files by
5386 hand from an archive which was created using a backup script, and to create
5387 an archive by hand which could then be extracted using the restore script,
5388 it is easier to use the scripts. @xref{Incremental Dumps}, before
5389 making such an attempt.
5391 @node Backup Parameters
5392 @section Setting Parameters for Backups and Restoration
5394 The file @file{backup-specs} specifies backup parameters for the
5395 backup and restoration scripts provided with @command{tar}. You must
5396 edit @file{backup-specs} to fit your system configuration and schedule
5397 before using these scripts.
5399 Syntactically, @file{backup-specs} is a shell script, containing
5400 mainly variable assignments. However, any valid shell construct
5401 is allowed in this file. Particularly, you may wish to define
5402 functions within that script (e.g., see @code{RESTORE_BEGIN} below).
5403 For more information about shell script syntax, please refer to
5404 @url{http://www.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html#ta
5405 g_02, the definition of the Shell Command Language}. See also
5406 @ref{Top,,Bash Features,bashref,Bash Reference Manual}.
5408 The shell variables controlling behavior of @code{backup} and
5409 @code{restore} are described in the following subsections.
5412 * General-Purpose Variables::
5413 * Magnetic Tape Control::
5415 * backup-specs example:: An Example Text of @file{Backup-specs}
5418 @node General-Purpose Variables
5419 @subsection General-Purpose Variables
5421 @defvr {Backup variable} ADMINISTRATOR
5422 The user name of the backup administrator. @code{Backup} scripts
5423 sends a backup report to this address.
5426 @defvr {Backup variable} BACKUP_HOUR
5427 The hour at which the backups are done. This can be a number from 0
5428 to 23, or the time specification in form @var{hours}:@var{minutes},
5429 or the string @samp{now}.
5431 This variable is used by @code{backup}. Its value may be overridden
5432 using @option{--time} option (@pxref{Scripted Backups}).
5435 @defvr {Backup variable} TAPE_FILE
5437 The device @command{tar} writes the archive to. If @var{TAPE_FILE}
5438 is a remote archive (@pxref{remote-dev}), backup script will suppose
5439 that your @command{mt} is able to access remote devices. If @var{RSH}
5440 (@pxref{RSH}) is set, @option{--rsh-command} option will be added to
5441 invocations of @command{mt}.
5444 @defvr {Backup variable} BLOCKING
5446 The blocking factor @command{tar} will use when writing the dump archive.
5447 @xref{Blocking Factor}.
5450 @defvr {Backup variable} BACKUP_DIRS
5452 A list of file systems to be dumped (for @code{backup}), or restored
5453 (for @code{restore}). You can include any directory
5454 name in the list --- subdirectories on that file system will be
5455 included, regardless of how they may look to other networked machines.
5456 Subdirectories on other file systems will be ignored.
5458 The host name specifies which host to run @command{tar} on, and should
5459 normally be the host that actually contains the file system. However,
5460 the host machine must have @GNUTAR{} installed, and
5461 must be able to access the directory containing the backup scripts and
5462 their support files using the same file name that is used on the
5463 machine where the scripts are run (ie. what @command{pwd} will print
5464 when in that directory on that machine). If the host that contains
5465 the file system does not have this capability, you can specify another
5466 host as long as it can access the file system through NFS.
5468 If the list of file systems is very long you may wish to put it
5469 in a separate file. This file is usually named
5470 @file{/etc/backup/dirs}, but this name may be overridden in
5471 @file{backup-specs} using @code{DIRLIST} variable.
5474 @defvr {Backup variable} DIRLIST
5476 A path to the file containing the list of the file systems to backup
5477 or restore. By default it is @file{/etc/backup/dirs}.
5480 @defvr {Backup variable} BACKUP_FILES
5482 A list of individual files to be dumped (for @code{backup}), or restored
5483 (for @code{restore}). These should be accessible from the machine on
5484 which the backup script is run.
5486 If the list of file systems is very long you may wish to store it
5487 in a separate file. This file is usually named
5488 @file{/etc/backup/files}, but this name may be overridden in
5489 @file{backup-specs} using @code{FILELIST} variable.
5492 @defvr {Backup variable} FILELIST
5494 A path to the file containing the list of the individual files to backup
5495 or restore. By default it is @file{/etc/backup/files}.
5498 @defvr {Backup variable} MT
5500 Full file name of @command{mt} binary.
5503 @defvr {Backup variable} RSH
5505 Full file name of @command{rsh} binary or its equivalent. You may wish to
5506 set it to @code{ssh}, to improve security. In this case you will have
5507 to use public key authentication.
5510 @defvr {Backup variable} RSH_COMMAND
5512 Full file name of @command{rsh} binary on remote mashines. This will
5513 be passed via @option{--rsh-command} option to the remote invocation
5517 @defvr {Backup variable} VOLNO_FILE
5519 Name of temporary file to hold volume numbers. This needs to be accessible
5520 by all the machines which have file systems to be dumped.
5523 @defvr {Backup variable} XLIST
5525 Name of @dfn{exclude file list}. An @dfn{exclude file list} is a file
5526 located on the remote machine and containing the list of files to
5527 be excluded from the backup. Exclude file lists are searched in
5528 /etc/tar-backup directory. A common use for exclude file lists
5529 is to exclude files containing security-sensitive information
5530 (e.g., @file{/etc/shadow} from backups).
5532 This variable affects only @code{backup}.
5535 @defvr {Backup variable} SLEEP_TIME
5537 Time to sleep between dumps of any two successive file systems
5539 This variable affects only @code{backup}.
5542 @defvr {Backup variable} DUMP_REMIND_SCRIPT
5544 Script to be run when it's time to insert a new tape in for the next
5545 volume. Administrators may want to tailor this script for their site.
5546 If this variable isn't set, @GNUTAR{} will display its built-in
5547 prompt, and will expect confirmation from the console.
5549 The built-in prompt for POSIX locale is:
5552 Prepare volume #@var{n} for `@var{archive}' and hit return:
5556 where @var{n} is the ordinal number of the volume to be created and
5557 @var{archive} is archive file or device name.
5559 If you run @GNUTAR{} under a different locale, the translation of
5560 the above prompt to the locale's language will be used.
5564 @defvr {Backup variable} SLEEP_MESSAGE
5566 Message to display on the terminal while waiting for dump time. Usually
5567 this will just be some literal text.
5570 @defvr {Backup variable} TAR
5572 Full file name of the @GNUTAR{} executable. If this is not set, backup
5573 scripts will search @command{tar} in the current shell path.
5576 @node Magnetic Tape Control
5577 @subsection Magnetic Tape Control
5579 Backup scripts access tape device using special @dfn{hook functions}.
5580 These functions take a single argument -- the name of the tape
5581 device. Their names are kept in the following variables:
5583 @defvr {Backup variable} MT_BEGIN
5584 The name of @dfn{begin} function. This function is called before
5585 accessing the drive. By default it retensions the tape:
5591 mt -f "$1" retension
5596 @defvr {Backup variable} MT_REWIND
5597 The name of @dfn{rewind} function. The default definition is as
5610 @defvr {Backup variable} MT_OFFLINE
5611 The name of the function switching the tape off line. By default
5612 it is defined as follows:
5615 MT_OFFLINE=mt_offline
5623 @defvr {Backup variable} MT_STATUS
5624 The name of the function used to obtain the status of the archive device,
5625 including error count. Default definition:
5637 @subsection User Hooks
5639 @dfn{User hooks} are shell functions executed before and after
5640 each @command{tar} invocation. Thus, there are @dfn{backup
5641 hooks}, which are executed before and after dumping each file
5642 system, and @dfn{restore hooks}, executed before and
5643 after restoring a file system. Each user hook is a shell function
5644 taking four arguments:
5646 @deffn {User Hook Function} hook @var{level} @var{host} @var{fs} @var{fsname}
5651 Current backup or restore level.
5654 Name or IP address of the host machine being dumped or restored.
5657 Full path name to the file system being dumped or restored.
5660 File system name with directory separators replaced with colons. This
5661 is useful, e.g., for creating unique files.
5665 Following variables keep the names of user hook functions
5667 @defvr {Backup variable} DUMP_BEGIN
5668 Dump begin function. It is executed before dumping the file system.
5671 @defvr {Backup variable} DUMP_END
5672 Executed after dumping the file system.
5675 @defvr {Backup variable} RESTORE_BEGIN
5676 Executed before restoring the file system.
5679 @defvr {Backup variable} RESTORE_END
5680 Executed after restoring the file system.
5683 @node backup-specs example
5684 @subsection An Example Text of @file{Backup-specs}
5686 The following is an example of @file{backup-specs}:
5689 # site-specific parameters for file system backup.
5691 ADMINISTRATOR=friedman
5693 TAPE_FILE=/dev/nrsmt0
5695 # Use @code{ssh} instead of the less secure @code{rsh}
5697 RSH_COMMAND=/usr/bin/ssh
5699 # Override MT_STATUS function:
5705 # Disable MT_OFFLINE function
5722 apple-gunkies:/com/mailer/gnu
5723 apple-gunkies:/com/archive/gnu"
5725 BACKUP_FILES="/com/mailer/aliases /com/mailer/league*[a-z]"
5729 @node Scripted Backups
5730 @section Using the Backup Scripts
5732 The syntax for running a backup script is:
5735 backup --level=@var{level} --time=@var{time}
5738 The @option{level} option requests the dump level. Thus, to produce
5739 a full dump, specify @code{--level=0} (this is the default, so
5740 @option{--level} may be omitted if its value is @code{0}).
5741 @footnote{For backward compatibility, the @code{backup} will also
5742 try to deduce the requested dump level from the name of the
5743 script itself. If the name consists of a string @samp{level-}
5744 followed by a single decimal digit, that digit is taken as
5745 the dump level number. Thus, you may create a link from @code{backup}
5746 to @code{level-1} and then run @code{level-1} whenever you need to
5747 create a level one dump.}
5749 The @option{--time} option determines when should the backup be
5750 run. @var{Time} may take three forms:
5753 @item @var{hh}:@var{mm}
5755 The dump must be run at @var{hh} hours @var{mm} minutes.
5759 The dump must be run at @var{hh} hours
5763 The dump must be run immediately.
5766 You should start a script with a tape or disk mounted. Once you
5767 start a script, it prompts you for new tapes or disks as it
5768 needs them. Media volumes don't have to correspond to archive
5769 files --- a multi-volume archive can be started in the middle of a
5770 tape that already contains the end of another multi-volume archive.
5771 The @code{restore} script prompts for media by its archive volume,
5772 so to avoid an error message you should keep track of which tape
5773 (or disk) contains which volume of the archive (@pxref{Scripted
5776 The backup scripts write two files on the file system. The first is a
5777 record file in @file{/etc/tar-backup/}, which is used by the scripts
5778 to store and retrieve information about which files were dumped. This
5779 file is not meant to be read by humans, and should not be deleted by
5780 them. @xref{Snapshot Files}, for a more detailed explanation of this
5783 The second file is a log file containing the names of the file systems
5784 and files dumped, what time the backup was made, and any error
5785 messages that were generated, as well as how much space was left in
5786 the media volume after the last volume of the archive was written.
5787 You should check this log file after every backup. The file name is
5788 @file{log-@var{mm-dd-yyyy}-level-@var{n}}, where @var{mm-dd-yyyy}
5789 represents current date, and @var{n} represents current dump level number.
5791 The script also prints the name of each system being dumped to the
5794 Following is the full list of options accepted by @code{backup}
5798 @item -l @var{level}
5799 @itemx --level=@var{level}
5800 Do backup level @var{level} (default 0).
5804 Force backup even if today's log file already exists.
5806 @item -v[@var{level}]
5807 @itemx --verbose[=@var{level}]
5808 Set verbosity level. The higher the level is, the more debugging
5809 information will be output during execution. Devault @var{level}
5810 is 100, which means the highest debugging level.
5812 @item -t @var{start-time}
5813 @itemx --time=@var{start-time}
5814 Wait till @var{time}, then do backup.
5818 Display short help message and exit.
5822 Display information about the program's name, version, origin and legal
5823 status, all on standard output, and then exit successfully.
5827 @node Scripted Restoration
5828 @section Using the Restore Script
5830 To restore files that were archived using a scripted backup, use the
5831 @code{restore} script. Its usage is quite straightforward. In the
5832 simplest form, invoke @code{restore --all}, it will
5833 then restore all the file systems and files specified in
5834 @file{backup-specs} (@pxref{General-Purpose Variables,BACKUP_DIRS}).
5836 You may select the file systems (and/or files) to restore by
5837 giving @code{restore} list of @dfn{patterns} in its command
5838 line. For example, running
5845 will restore all file systems on the machine @samp{albert}. A more
5846 complicated example:
5849 restore 'albert:*' '*:/var'
5853 This command will restore all file systems on the machine @samp{albert}
5854 as well as @file{/var} file system on all machines.
5856 By default @code{restore} will start restoring files from the lowest
5857 available dump level (usually zero) and will continue through
5858 all available dump levels. There may be situations where such a
5859 thorough restore is not necessary. For example, you may wish to
5860 restore only files from the recent level one backup. To do so,
5861 use @option{--level} option, as shown in the example below:
5867 The full list of options accepted by @code{restore} follows:
5872 Restore all file systems and files specified in @file{backup-specs}
5874 @item -l @var{level}
5875 @itemx --level=@var{level}
5876 Start restoring from the given backup level, instead of the default 0.
5878 @item -v[@var{level}]
5879 @itemx --verbose[=@var{level}]
5880 Set verbosity level. The higher the level is, the more debugging
5881 information will be output during execution. Devault @var{level}
5882 is 100, which means the highest debugging level.
5886 Display short help message and exit.
5890 Display information about the program's name, version, origin and legal
5891 status, all on standard output, and then exit successfully.
5894 You should start the restore script with the media containing the
5895 first volume of the archive mounted. The script will prompt for other
5896 volumes as they are needed. If the archive is on tape, you don't need
5897 to rewind the tape to to its beginning---if the tape head is
5898 positioned past the beginning of the archive, the script will rewind
5899 the tape as needed. @FIXME-xref{Media, for a discussion of tape
5903 @strong{Warning:} The script will delete files from the active file
5904 system if they were not in the file system when the archive was made.
5907 @xref{Incremental Dumps}, for an explanation of how the script makes
5911 @chapter Choosing Files and Names for @command{tar}
5914 Certain options to @command{tar} enable you to specify a name for your
5915 archive. Other options let you decide which files to include or exclude
5916 from the archive, based on when or whether files were modified, whether
5917 the file names do or don't match specified patterns, or whether files
5918 are in specified directories.
5920 This chapter discusses these options in detail.
5923 * file:: Choosing the Archive's Name
5924 * Selecting Archive Members::
5925 * files:: Reading Names from a File
5926 * exclude:: Excluding Some Files
5927 * wildcards:: Wildcards Patterns and Matching
5928 * quoting styles:: Ways of Quoting Special Characters in Names
5929 * transform:: Modifying File and Member Names
5930 * after:: Operating Only on New Files
5931 * recurse:: Descending into Directories
5932 * one:: Crossing File System Boundaries
5936 @section Choosing and Naming Archive Files
5939 @cindex Naming an archive
5940 @cindex Archive Name
5941 @cindex Choosing an archive file
5942 @cindex Where is the archive?
5943 By default, @command{tar} uses an archive file name that was compiled when
5944 it was built on the system; usually this name refers to some physical
5945 tape drive on the machine. However, the person who installed @command{tar}
5946 on the system may not have set the default to a meaningful value as far as
5947 most users are concerned. As a result, you will usually want to tell
5948 @command{tar} where to find (or create) the archive. The
5949 @option{--file=@var{archive-name}} (@option{-f @var{archive-name}})
5950 option allows you to either specify or name a file to use as the archive
5951 instead of the default archive file location.
5954 @opindex file, short description
5955 @item --file=@var{archive-name}
5956 @itemx -f @var{archive-name}
5957 Name the archive to create or operate on. Use in conjunction with
5961 For example, in this @command{tar} command,
5964 $ @kbd{tar -cvf collection.tar blues folk jazz}
5968 @file{collection.tar} is the name of the archive. It must directly
5969 follow the @option{-f} option, since whatever directly follows @option{-f}
5970 @emph{will} end up naming the archive. If you neglect to specify an
5971 archive name, you may end up overwriting a file in the working directory
5972 with the archive you create since @command{tar} will use this file's name
5973 for the archive name.
5975 An archive can be saved as a file in the file system, sent through a
5976 pipe or over a network, or written to an I/O device such as a tape,
5977 floppy disk, or CD write drive.
5979 @cindex Writing new archives
5980 @cindex Archive creation
5981 If you do not name the archive, @command{tar} uses the value of the
5982 environment variable @env{TAPE} as the file name for the archive. If
5983 that is not available, @command{tar} uses a default, compiled-in archive
5984 name, usually that for tape unit zero (ie. @file{/dev/tu00}).
5986 @cindex Standard input and output
5987 @cindex tar to standard input and output
5988 If you use @file{-} as an @var{archive-name}, @command{tar} reads the
5989 archive from standard input (when listing or extracting files), or
5990 writes it to standard output (when creating an archive). If you use
5991 @file{-} as an @var{archive-name} when modifying an archive,
5992 @command{tar} reads the original archive from its standard input and
5993 writes the entire new archive to its standard output.
5995 The following example is a convenient way of copying directory
5996 hierarchy from @file{sourcedir} to @file{targetdir}.
5999 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xpf -)}
6002 The @option{-C} option allows to avoid using subshells:
6005 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xpf -}
6008 In both examples above, the leftmost @command{tar} invocation archives
6009 the contents of @file{sourcedir} to the standard output, while the
6010 rightmost one reads this archive from its standard input and
6011 extracts it. The @option{-p} option tells it to restore permissions
6012 of the extracted files.
6014 @cindex Remote devices
6015 @cindex tar to a remote device
6017 To specify an archive file on a device attached to a remote machine,
6021 @kbd{--file=@var{hostname}:/@var{dev}/@var{file-name}}
6025 @command{tar} will complete the remote connection, if possible, and
6026 prompt you for a username and password. If you use
6027 @option{--file=@@@var{hostname}:/@var{dev}/@var{file-name}}, @command{tar}
6028 will complete the remote connection, if possible, using your username
6029 as the username on the remote machine.
6031 @cindex Local and remote archives
6032 @anchor{local and remote archives}
6033 If the archive file name includes a colon (@samp{:}), then it is assumed
6034 to be a file on another machine. If the archive file is
6035 @samp{@var{user}@@@var{host}:@var{file}}, then @var{file} is used on the
6036 host @var{host}. The remote host is accessed using the @command{rsh}
6037 program, with a username of @var{user}. If the username is omitted
6038 (along with the @samp{@@} sign), then your user name will be used.
6039 (This is the normal @command{rsh} behavior.) It is necessary for the
6040 remote machine, in addition to permitting your @command{rsh} access, to
6041 have the @file{rmt} program installed (This command is included in
6042 the @GNUTAR{} distribution and by default is installed under
6043 @file{@var{prefix}/libexec/rmt}, were @var{prefix} means your
6044 installation prefix). If you need to use a file whose name includes a
6045 colon, then the remote tape drive behavior
6046 can be inhibited by using the @option{--force-local} option.
6048 When the archive is being created to @file{/dev/null}, @GNUTAR{}
6049 tries to minimize input and output operations. The Amanda backup
6050 system, when used with @GNUTAR{}, has an initial sizing pass which
6053 @node Selecting Archive Members
6054 @section Selecting Archive Members
6055 @cindex Specifying files to act on
6056 @cindex Specifying archive members
6058 @dfn{File Name arguments} specify which files in the file system
6059 @command{tar} operates on, when creating or adding to an archive, or which
6060 archive members @command{tar} operates on, when reading or deleting from
6061 an archive. @xref{Operations}.
6063 To specify file names, you can include them as the last arguments on
6064 the command line, as follows:
6066 @kbd{tar} @var{operation} [@var{option1} @var{option2} @dots{}] [@var{file name-1} @var{file name-2} @dots{}]
6069 If a file name begins with dash (@samp{-}), precede it with
6070 @option{--add-file} option to prevent it from being treated as an
6073 If you specify a directory name as a file name argument, all the files
6074 in that directory are operated on by @command{tar}.
6076 If you do not specify files, @command{tar} behavior differs depending
6077 on the operation mode as described below:
6079 When @command{tar} is invoked with @option{--create} (@option{-c}),
6080 @command{tar} will stop immediately, reporting the following:
6084 $ @kbd{tar cf a.tar}
6085 tar: Cowardly refusing to create an empty archive
6086 Try `tar --help' or `tar --usage' for more information.
6090 If you specify either @option{--list} (@option{-t}) or
6091 @option{--extract} (@option{--get}, @option{-x}), @command{tar}
6092 operates on all the archive members in the archive.
6094 If run with @option{--diff} option, tar will compare the archive with
6095 the contents of the current working directory.
6097 If you specify any other operation, @command{tar} does nothing.
6099 By default, @command{tar} takes file names from the command line. However,
6100 there are other ways to specify file or member names, or to modify the
6101 manner in which @command{tar} selects the files or members upon which to
6102 operate. In general, these methods work both for specifying the names
6103 of files and archive members.
6106 @section Reading Names from a File
6108 @cindex Reading file names from a file
6109 @cindex Lists of file names
6110 @cindex File Name arguments, alternatives
6111 Instead of giving the names of files or archive members on the command
6112 line, you can put the names into a file, and then use the
6113 @option{--files-from=@var{file-of-names}} (@option{-T
6114 @var{file-of-names}}) option to @command{tar}. Give the name of the
6115 file which contains the list of files to include as the argument to
6116 @option{--files-from}. In the list, the file names should be separated by
6117 newlines. You will frequently use this option when you have generated
6118 the list of files to archive with the @command{find} utility.
6122 @item --files-from=@var{file-name}
6123 @itemx -T @var{file-name}
6124 Get names to extract or create from file @var{file-name}.
6127 If you give a single dash as a file name for @option{--files-from}, (i.e.,
6128 you specify either @code{--files-from=-} or @code{-T -}), then the file
6129 names are read from standard input.
6131 Unless you are running @command{tar} with @option{--create}, you can not use
6132 both @code{--files-from=-} and @code{--file=-} (@code{-f -}) in the same
6135 Any number of @option{-T} options can be given in the command line.
6137 The following example shows how to use @command{find} to generate a list of
6138 files smaller than 400K in length and put that list into a file
6139 called @file{small-files}. You can then use the @option{-T} option to
6140 @command{tar} to specify the files from that file, @file{small-files}, to
6141 create the archive @file{little.tgz}. (The @option{-z} option to
6142 @command{tar} compresses the archive with @command{gzip}; @pxref{gzip} for
6146 $ @kbd{find . -size -400 -print > small-files}
6147 $ @kbd{tar -c -v -z -T small-files -f little.tgz}
6151 In the file list given by @option{-T} option, any file name beginning
6152 with @samp{-} character is considered a @command{tar} option and is
6153 processed accordingly.@footnote{Versions of @GNUTAR{} up to 1.15.1
6154 recognized only @option{-C} option in file lists, and only if the
6155 option and its argument occupied two consecutive lines.} For example,
6156 the common use of this feature is to change to another directory by
6157 specifying @option{-C} option:
6167 $ @kbd{tar -c -f foo.tar --files-from list}
6172 In this example, @command{tar} will first switch to @file{/etc}
6173 directory and add files @file{passwd} and @file{hosts} to the
6174 archive. Then it will change to @file{/lib} directory and will archive
6175 the file @file{libc.a}. Thus, the resulting archive @file{foo.tar} will
6180 $ @kbd{tar tf foo.tar}
6188 @opindex directory, using in @option{--files-from} argument
6189 Notice that the option parsing algorithm used with @option{-T} is
6190 stricter than the one used by shell. Namely, when specifying option
6191 arguments, you should observe the following rules:
6195 When using short (single-letter) option form, its argument must
6196 immediately follow the option letter, without any intervening
6197 whitespace. For example: @code{-Cdir}.
6200 When using long option form, the option argument must be separated
6201 from the option by a single equal sign. No whitespace is allowed on
6202 any side of the equal sign. For example: @code{--directory=dir}.
6205 For both short and long option forms, the option argument can be given
6206 on the next line after the option name, e.g.:
6227 If you happen to have a file whose name starts with @samp{-},
6228 precede it with @option{--add-file} option to prevent it from
6229 being recognized as an option. For example: @code{--add-file=--my-file}.
6236 @subsection @code{NUL} Terminated File Names
6238 @cindex File names, terminated by @code{NUL}
6239 @cindex @code{NUL} terminated file names
6240 The @option{--null} option causes
6241 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}})
6242 to read file names terminated by a @code{NUL} instead of a newline, so
6243 files whose names contain newlines can be archived using
6244 @option{--files-from}.
6249 Only consider @code{NUL} terminated file names, instead of files that
6250 terminate in a newline.
6253 The @option{--null} option is just like the one in @acronym{GNU}
6254 @command{xargs} and @command{cpio}, and is useful with the
6255 @option{-print0} predicate of @acronym{GNU} @command{find}. In
6256 @command{tar}, @option{--null} also disables special handling for
6257 file names that begin with dash.
6259 This example shows how to use @command{find} to generate a list of files
6260 larger than 800K in length and put that list into a file called
6261 @file{long-files}. The @option{-print0} option to @command{find} is just
6262 like @option{-print}, except that it separates files with a @code{NUL}
6263 rather than with a newline. You can then run @command{tar} with both the
6264 @option{--null} and @option{-T} options to specify that @command{tar} get the
6265 files from that file, @file{long-files}, to create the archive
6266 @file{big.tgz}. The @option{--null} option to @command{tar} will cause
6267 @command{tar} to recognize the @code{NUL} separator between files.
6270 $ @kbd{find . -size +800 -print0 > long-files}
6271 $ @kbd{tar -c -v --null --files-from=long-files --file=big.tar}
6274 @FIXME{say anything else here to conclude the section?}
6277 @section Excluding Some Files
6280 @cindex File names, excluding files by
6281 @cindex Excluding files by name and pattern
6282 @cindex Excluding files by file system
6283 To avoid operating on files whose names match a particular pattern,
6284 use the @option{--exclude} or @option{--exclude-from} options.
6288 @item --exclude=@var{pattern}
6289 Causes @command{tar} to ignore files that match the @var{pattern}.
6293 The @option{--exclude=@var{pattern}} option prevents any file or
6294 member whose name matches the shell wildcard (@var{pattern}) from
6296 For example, to create an archive with all the contents of the directory
6297 @file{src} except for files whose names end in @file{.o}, use the
6298 command @samp{tar -cf src.tar --exclude='*.o' src}.
6300 You may give multiple @option{--exclude} options.
6303 @opindex exclude-from
6304 @item --exclude-from=@var{file}
6305 @itemx -X @var{file}
6306 Causes @command{tar} to ignore files that match the patterns listed in
6310 @findex exclude-from
6311 Use the @option{--exclude-from} option to read a
6312 list of patterns, one per line, from @var{file}; @command{tar} will
6313 ignore files matching those patterns. Thus if @command{tar} is
6314 called as @w{@samp{tar -c -X foo .}} and the file @file{foo} contains a
6315 single line @file{*.o}, no files whose names end in @file{.o} will be
6316 added to the archive.
6319 @opindex exclude-caches
6320 @item --exclude-caches
6321 Causes @command{tar} to ignore directories containing a cache directory tag.
6324 @findex exclude-caches
6325 When creating an archive, the @option{--exclude-caches} option causes
6326 @command{tar} to exclude all directories that contain a @dfn{cache
6327 directory tag}. A cache directory tag is a short file with the
6328 well-known name @file{CACHEDIR.TAG} and having a standard header
6329 specified in @url{http://www.brynosaurus.com/cachedir/spec.html}.
6330 Various applications write cache directory tags into directories they
6331 use to hold regenerable, non-precious data, so that such data can be
6332 more easily excluded from backups.
6335 * problems with exclude::
6338 @node problems with exclude
6339 @unnumberedsubsec Problems with Using the @code{exclude} Options
6341 @opindex exclude, potential problems with
6342 Some users find @samp{exclude} options confusing. Here are some common
6347 The main operating mode of @command{tar} does not act on a path name
6348 explicitly listed on the command line if one of its file name
6349 components is excluded. In the example above, if
6350 you create an archive and exclude files that end with @samp{*.o}, but
6351 explicitly name the file @samp{dir.o/foo} after all the options have been
6352 listed, @samp{dir.o/foo} will be excluded from the archive.
6355 You can sometimes confuse the meanings of @option{--exclude} and
6356 @option{--exclude-from}. Be careful: use @option{--exclude} when files
6357 to be excluded are given as a pattern on the command line. Use
6358 @option{--exclude-from} to introduce the name of a file which contains
6359 a list of patterns, one per line; each of these patterns can exclude
6360 zero, one, or many files.
6363 When you use @option{--exclude=@var{pattern}}, be sure to quote the
6364 @var{pattern} parameter, so @GNUTAR{} sees wildcard characters
6365 like @samp{*}. If you do not do this, the shell might expand the
6366 @samp{*} itself using files at hand, so @command{tar} might receive a
6367 list of files instead of one pattern, or none at all, making the
6368 command somewhat illegal. This might not correspond to what you want.
6373 $ @kbd{tar -c -f @var{archive.tar} --exclude '*.o' @var{directory}}
6381 $ @kbd{tar -c -f @var{archive.tar} --exclude *.o @var{directory}}
6385 You must use use shell syntax, or globbing, rather than @code{regexp}
6386 syntax, when using exclude options in @command{tar}. If you try to use
6387 @code{regexp} syntax to describe files to be excluded, your command
6391 @FIXME{The change in semantics must have occurred before 1.11,
6392 so I doubt if it is worth mentioning at all. Anyway, should at
6393 least specify in which version the semantics changed.}
6394 In earlier versions of @command{tar}, what is now the
6395 @option{--exclude-from} option was called @option{--exclude} instead.
6396 Now, @option{--exclude} applies to patterns listed on the command
6397 line and @option{--exclude-from} applies to patterns listed in a
6403 @section Wildcards Patterns and Matching
6405 @dfn{Globbing} is the operation by which @dfn{wildcard} characters,
6406 @samp{*} or @samp{?} for example, are replaced and expanded into all
6407 existing files matching the given pattern. @GNUTAR{} can use wildcard
6408 patterns for matching (or globbing) archive members when extracting
6409 from or listing an archive. Wildcard patterns are also used for
6410 verifying volume labels of @command{tar} archives. This section has the
6411 purpose of explaining wildcard syntax for @command{tar}.
6413 @FIXME{the next few paragraphs need work.}
6415 A @var{pattern} should be written according to shell syntax, using wildcard
6416 characters to effect globbing. Most characters in the pattern stand
6417 for themselves in the matched string, and case is significant: @samp{a}
6418 will match only @samp{a}, and not @samp{A}. The character @samp{?} in the
6419 pattern matches any single character in the matched string. The character
6420 @samp{*} in the pattern matches zero, one, or more single characters in
6421 the matched string. The character @samp{\} says to take the following
6422 character of the pattern @emph{literally}; it is useful when one needs to
6423 match the @samp{?}, @samp{*}, @samp{[} or @samp{\} characters, themselves.
6425 The character @samp{[}, up to the matching @samp{]}, introduces a character
6426 class. A @dfn{character class} is a list of acceptable characters
6427 for the next single character of the matched string. For example,
6428 @samp{[abcde]} would match any of the first five letters of the alphabet.
6429 Note that within a character class, all of the ``special characters''
6430 listed above other than @samp{\} lose their special meaning; for example,
6431 @samp{[-\\[*?]]} would match any of the characters, @samp{-}, @samp{\},
6432 @samp{[}, @samp{*}, @samp{?}, or @samp{]}. (Due to parsing constraints,
6433 the characters @samp{-} and @samp{]} must either come @emph{first} or
6434 @emph{last} in a character class.)
6436 @cindex Excluding characters from a character class
6437 @cindex Character class, excluding characters from
6438 If the first character of the class after the opening @samp{[}
6439 is @samp{!} or @samp{^}, then the meaning of the class is reversed.
6440 Rather than listing character to match, it lists those characters which
6441 are @emph{forbidden} as the next single character of the matched string.
6443 Other characters of the class stand for themselves. The special
6444 construction @samp{[@var{a}-@var{e}]}, using an hyphen between two
6445 letters, is meant to represent all characters between @var{a} and
6448 @FIXME{need to add a sentence or so here to make this clear for those
6449 who don't have dan around.}
6451 Periods (@samp{.}) or forward slashes (@samp{/}) are not considered
6452 special for wildcard matches. However, if a pattern completely matches
6453 a directory prefix of a matched string, then it matches the full matched
6454 string: thus, excluding a directory also excludes all the files beneath it.
6457 * controlling pattern-matching::
6460 @node controlling pattern-matching
6461 @unnumberedsubsec Controlling Pattern-Matching
6463 For the purposes of this section, we call @dfn{exclusion members} all
6464 member names obtained while processing @option{--exclude} and
6465 @option{--exclude-from} options, and @dfn{inclusion members} those
6466 member names that were given in the command line or read from the file
6467 specified with @option{--files-from} option.
6469 These two pairs of member lists are used in the following operations:
6470 @option{--diff}, @option{--extract}, @option{--list},
6473 There are no inclusion members in create mode (@option{--create} and
6474 @option{--append}), since in this mode the names obtained from the
6475 command line refer to @emph{files}, not archive members.
6477 By default, inclusion members are compared with archive members
6478 literally @footnote{Notice that earlier @GNUTAR{} versions used
6479 globbing for inclusion members, which contradicted to UNIX98
6480 specification and was not documented. @xref{Changes}, for more
6481 information on this and other changes.} and exclusion members are
6482 treated as globbing patterns. For example:
6486 $ @kbd{tar tf foo.tar}
6491 # @i{Member names are used verbatim:}
6492 $ @kbd{tar -xf foo.tar -v '[remarks]'}
6494 # @i{Exclude member names are globbed:}
6495 $ @kbd{tar -xf foo.tar -v --exclude '*.c'}
6501 This behavior can be altered by using the following options:
6506 Treat all member names as wildcards.
6508 @opindex no-wildcards
6509 @item --no-wildcards
6510 Treat all member names as literal strings.
6513 Thus, to extract files whose names end in @samp{.c}, you can use:
6516 $ @kbd{tar -xf foo.tar -v --wildcards '*.c'}
6522 Notice quoting of the pattern to prevent the shell from interpreting
6525 The effect of @option{--wildcards} option is cancelled by
6526 @option{--no-wildcards}. This can be used to pass part of
6527 the command line arguments verbatim and other part as globbing
6528 patterns. For example, the following invocation:
6531 $ @kbd{tar -xf foo.tar --wildcards '*.txt' --no-wildcards '[remarks]'}
6535 instructs @command{tar} to extract from @file{foo.tar} all files whose
6536 names end in @samp{.txt} and the file named @file{[remarks]}.
6538 Normally, a pattern matches a name if an initial subsequence of the
6539 name's components matches the pattern, where @samp{*}, @samp{?}, and
6540 @samp{[...]} are the usual shell wildcards, @samp{\} escapes wildcards,
6541 and wildcards can match @samp{/}.
6543 Other than optionally stripping leading @samp{/} from names
6544 (@pxref{absolute}), patterns and names are used as-is. For
6545 example, trailing @samp{/} is not trimmed from a user-specified name
6546 before deciding whether to exclude it.
6548 However, this matching procedure can be altered by the options listed
6549 below. These options accumulate. For example:
6552 --ignore-case --exclude='makefile' --no-ignore-case ---exclude='readme'
6556 ignores case when excluding @samp{makefile}, but not when excluding
6561 @opindex no-anchored
6563 @itemx --no-anchored
6564 If anchored, a pattern must match an initial subsequence
6565 of the name's components. Otherwise, the pattern can match any
6566 subsequence. Default is @option{--no-anchored} for exclusion members
6567 and @option{--anchored} inclusion members.
6569 @opindex ignore-case
6570 @opindex no-ignore-case
6572 @itemx --no-ignore-case
6573 When ignoring case, upper-case patterns match lower-case names and vice versa.
6574 When not ignoring case (the default), matching is case-sensitive.
6576 @opindex wildcards-match-slash
6577 @opindex no-wildcards-match-slash
6578 @item --wildcards-match-slash
6579 @itemx --no-wildcards-match-slash
6580 When wildcards match slash (the default for exclusion members), a
6581 wildcard like @samp{*} in the pattern can match a @samp{/} in the
6582 name. Otherwise, @samp{/} is matched only by @samp{/}.
6586 The @option{--recursion} and @option{--no-recursion} options
6587 (@pxref{recurse}) also affect how member patterns are interpreted. If
6588 recursion is in effect, a pattern matches a name if it matches any of
6589 the name's parent directories.
6591 The following table summarizes pattern-matching default values:
6593 @multitable @columnfractions .3 .7
6594 @headitem Members @tab Default settings
6595 @item Inclusion @tab @option{--no-wildcards --anchored --no-wildcards-match-slash}
6596 @item Exclusion @tab @option{--wildcards --no-anchored --wildcards-match-slash}
6599 @node quoting styles
6600 @section Quoting Member Names
6602 When displaying member names, @command{tar} takes care to avoid
6603 ambiguities caused by certain characters. This is called @dfn{name
6604 quoting}. The characters in question are:
6607 @item Non-printable control characters:
6609 @multitable @columnfractions 0.20 0.10 0.60
6610 @headitem Character @tab ASCII @tab Character name
6611 @item \a @tab 7 @tab Audible bell
6612 @item \b @tab 8 @tab Backspace
6613 @item \f @tab 12 @tab Form feed
6614 @item \n @tab 10 @tab New line
6615 @item \r @tab 13 @tab Carriage return
6616 @item \t @tab 9 @tab Horizontal tabulation
6617 @item \v @tab 11 @tab Vertical tabulation
6620 @item Space (ASCII 32)
6622 @item Single and double quotes (@samp{'} and @samp{"})
6624 @item Backslash (@samp{\})
6627 The exact way @command{tar} uses to quote these characters depends on
6628 the @dfn{quoting style}. The default quoting style, called
6629 @dfn{escape} (see below), uses backslash notation to represent control
6630 characters, space and backslash. Using this quoting style, control
6631 characters are represented as listed in column @samp{Character} in the
6632 above table, a space is printed as @samp{\ } and a backslash as @samp{\\}.
6634 @GNUTAR{} offers seven distinct quoting styles, which can be selected
6635 using @option{--quoting-style} option:
6638 @item --quoting-style=@var{style}
6639 @opindex quoting-style
6641 Sets quoting style. Valid values for @var{style} argument are:
6642 literal, shell, shell-always, c, escape, locale, clocale.
6645 These styles are described in detail below. To illustrate their
6646 effect, we will use an imaginary tar archive @file{arch.tar}
6647 containing the following members:
6651 # 1. Contains horizontal tabulation character.
6653 # 2. Contains newline character
6656 # 3. Contains a space
6658 # 4. Contains double quotes
6660 # 5. Contains single quotes
6662 # 6. Contains a backslash character:
6667 Here is how usual @command{ls} command would have listed them, if they
6668 had existed in the current working directory:
6686 No quoting, display each character as is:
6690 $ @kbd{tar tf arch.tar --quoting-style=literal}
6703 Display characters the same way Bourne shell does:
6704 control characters, except @samp{\t} and @samp{\n}, are printed using
6705 backslash escapes, @samp{\t} and @samp{\n} are printed as is, and a
6706 single quote is printed as @samp{\'}. If a name contains any quoted
6707 characters, it is enclosed in single quotes. In particular, if a name
6708 contains single quotes, it is printed as several single-quoted strings:
6712 $ @kbd{tar tf arch.tar --quoting-style=shell}
6715 './a'\''single'\''quote'
6725 Same as @samp{shell}, but the names are always enclosed in single
6730 $ @kbd{tar tf arch.tar --quoting-style=shell-always}
6733 './a'\''single'\''quote'
6743 Use the notation of the C programming language. All names are
6744 enclosed in double quotes. Control characters are quoted using
6745 backslash notations, double quotes are represented as @samp{\"},
6746 backslash characters are represented as @samp{\\}. Single quotes and
6747 spaces are not quoted:
6751 $ @kbd{tar tf arch.tar --quoting-style=c}
6755 "./a\"double\"quote"
6763 Control characters are printed using backslash notation, a space is
6764 printed as @samp{\ } and a backslash as @samp{\\}. This is the
6765 default quoting style, unless it was changed when configured the
6770 $ @kbd{tar tf arch.tar --quoting-style=escape}
6782 Control characters, single quote and backslash are printed using
6783 backslash notation. All names are quoted using left and right
6784 quotation marks, appropriate to the current locale. If it does not
6785 define quotation marks, use @samp{`} as left and @samp{'} as right
6786 quotation marks. Any occurrences of the right quotation mark in a
6787 name are escaped with @samp{\}, for example:
6793 $ @kbd{tar tf arch.tar --quoting-style=locale}
6796 `./a\'single\'quote'
6805 Same as @samp{locale}, but @samp{"} is used for both left and right
6806 quotation marks, if not provided by the currently selected locale:
6810 $ @kbd{tar tf arch.tar --quoting-style=clocale}
6814 "./a\"double\"quote"
6822 You can specify which characters should be quoted in addition to those
6823 implied by the current quoting style:
6826 @item --quote-chars=@var{string}
6827 Always quote characters from @var{string}, even if the selected
6828 quoting style would not quote them.
6831 For example, using @samp{escape} quoting (compare with the usual
6832 escape listing above):
6836 $ @kbd{tar tf arch.tar --quoting-style=escape --quote-chars=' "'}
6847 To disable quoting of such additional characters, use the following
6851 @item --no-quote-chars=@var{string}
6852 Remove characters listed in @var{string} from the list of quoted
6853 characters set by the previous @option{--quote-chars} option.
6856 This option is particularly useful if you have added
6857 @option{--quote-chars} to your @env{TAR_OPTIONS} (@pxref{TAR_OPTIONS})
6858 and wish to disable it for the current invocation.
6860 Note, that @option{--no-quote-chars} does @emph{not} disable those
6861 characters that are quoted by default in the selected quoting style.
6864 @section Modifying File and Member Names
6866 @command{Tar} archives contain detailed information about files stored
6867 in them and full file names are part of that information. When
6868 storing file to an archive, its file name is recorded in the archive
6869 along with the actual file contents. When restoring from an archive,
6870 a file is created on disk with exactly the same name as that stored
6871 in the archive. In the majority of cases this is the desired behavior
6872 of a file archiver. However, there are some cases when it is not.
6874 First of all, it is often unsafe to extract archive members with
6875 absolute file names or those that begin with a @file{../}. @GNUTAR{}
6876 takes special precautions when extracting such names and provides a
6877 special option for handling them, which is described in
6880 Secondly, you may wish to extract file names without some leading
6881 directory components, or with otherwise modified names. In other
6882 cases it is desirable to store files under differing names in the
6885 @GNUTAR{} provides two options for these needs.
6888 @opindex strip-components
6889 @item --strip-components=@var{number}
6890 Strip given @var{number} of leading components from file names before
6894 For example, suppose you have archived whole @file{/usr} hierarchy to
6895 a tar archive named @file{usr.tar}. Among other files, this archive
6896 contains @file{usr/include/stdlib.h}, which you wish to extract to
6897 the current working directory. To do so, you type:
6900 $ @kbd{tar -xf usr.tar --strip=2 usr/include/stdlib.h}
6903 The option @option{--strip=2} instructs @command{tar} to strip the
6904 two leading components (@file{usr/} and @file{include/}) off the file
6907 If you add to the above invocation @option{--verbose} (@option{-v})
6908 option, you will note that the verbose listing still contains the
6909 full file name, with the two removed components still in place. This
6910 can be inconvenient, so @command{tar} provides a special option for
6911 altering this behavior:
6913 @anchor{show-transformed-names}
6915 @opindex --show-transformed-names
6916 @item --show-transformed-names
6917 Display file or member names with all requested transformations
6926 $ @kbd{tar -xf usr.tar -v --strip=2 usr/include/stdlib.h}
6927 usr/include/stdlib.h
6928 $ @kbd{tar -xf usr.tar -v --strip=2 --show-transformed usr/include/stdlib.h}
6933 Notice that in both cases the file is @file{stdlib.h} extracted to the
6934 current working directory, @option{--show-transformed-names} affects
6935 only the way its name is displayed.
6937 This option is especially useful for verifying whether the invocation
6938 will have the desired effect. Thus, before running
6941 $ @kbd{tar -x --strip=@var{n}}
6945 it is often advisable to run
6948 $ @kbd{tar -t -v --show-transformed --strip=@var{n}}
6952 to make sure the command will produce the intended results.
6954 In case you need to apply more complex modifications to the file name,
6955 @GNUTAR{} provides a general-purpose transformation option:
6958 @opindex --transform
6959 @item --transform=@var{expression}
6960 Modify file names using supplied @var{expression}.
6964 The @var{expression} is a @command{sed}-like replace expression of the
6968 s/@var{regexp}/@var{replace}/[@var{flags}]
6972 where @var{regexp} is a @dfn{regular expression}, @var{replace} is a
6973 replacement for each file name part that matches @var{regexp}. Both
6974 @var{regexp} and @var{replace} are described in detail in
6975 @ref{The "s" Command, The "s" Command, The `s' Command, sed, GNU sed}.
6977 Supported @var{flags} are:
6981 Apply the replacement to @emph{all} matches to the @var{regexp}, not
6985 Use case-insensitive matching
6988 @var{regexp} is an @dfn{extended regular expression} (@pxref{Extended
6989 regexps, Extended regular expressions, Extended regular expressions,
6993 Only replace the @var{number}th match of the @var{regexp}.
6995 Note: the @var{posix} standard does not specify what should happen
6996 when you mix the @samp{g} and @var{number} modifiers. @GNUTAR{}
6997 follows the GNU @command{sed} implementation in this regard, so
6998 the the interaction is defined to be: ignore matches before the
6999 @var{number}th, and then match and replace all matches from the
7004 Any delimiter can be used in lieue of @samp{/}, the only requirement being
7005 that it be used consistently throughout the expression. For example,
7006 the following two expressions are equivalent:
7015 Changing delimiters is often useful when the @var{regex} contains
7016 slashes. For example, it is more convenient to write @code{s,/,-,} than
7019 Here are several examples of @option{--transform} usage:
7022 @item Extract @file{usr/} hierarchy into @file{usr/local/}:
7025 $ @kbd{tar --transform='s,usr/,usr/local/,' -x -f arch.tar}
7028 @item Strip two leading directory components (equivalent to
7029 @option{--strip-components=2}):
7032 $ @kbd{tar --transform='s,/*[^/]*/[^/]*/,,' -x -f arch.tar}
7035 @item Prepend @file{/prefix/} to each file name:
7038 $ @kbd{tar --transform 's,^,/prefix/,' -x -f arch.tar}
7041 @item Convert each file name to lower case:
7044 $ @kbd{tar --transform 's/.*/\L&/' -x -f arch.tar}
7049 Unlike @option{--strip-components}, @option{--transform} can be used
7050 in any @GNUTAR{} operation mode. For example, the following command
7051 adds files to the archive while replacing the leading @file{usr/}
7052 component with @file{var/}:
7055 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' /}
7058 To test @option{--transform} effect we suggest using
7059 @option{--show-transformed-names} option:
7062 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' \
7063 --verbose --show-transformed-names /}
7066 If both @option{--strip-components} and @option{--transform} are used
7067 together, then @option{--transform} is applied first, and the required
7068 number of components is then stripped from its result.
7071 @section Operating Only on New Files
7074 @cindex Excluding file by age
7075 @cindex Data Modification time, excluding files by
7076 @cindex Modification time, excluding files by
7077 @cindex Age, excluding files by
7078 The @option{--after-date=@var{date}} (@option{--newer=@var{date}},
7079 @option{-N @var{date}}) option causes @command{tar} to only work on
7080 files whose data modification or status change times are newer than
7081 the @var{date} given. If @var{date} starts with @samp{/} or @samp{.},
7082 it is taken to be a file name; the data modification time of that file
7083 is used as the date. If you use this option when creating or appending
7084 to an archive, the archive will only include new files. If you use
7085 @option{--after-date} when extracting an archive, @command{tar} will
7086 only extract files newer than the @var{date} you specify.
7088 If you only want @command{tar} to make the date comparison based on
7089 modification of the file's data (rather than status
7090 changes), then use the @option{--newer-mtime=@var{date}} option.
7092 You may use these options with any operation. Note that these options
7093 differ from the @option{--update} (@option{-u}) operation in that they
7094 allow you to specify a particular date against which @command{tar} can
7095 compare when deciding whether or not to archive the files.
7100 @item --after-date=@var{date}
7101 @itemx --newer=@var{date}
7102 @itemx -N @var{date}
7103 Only store files newer than @var{date}.
7105 Acts on files only if their data modification or status change times are
7106 later than @var{date}. Use in conjunction with any operation.
7108 If @var{date} starts with @samp{/} or @samp{.}, it is taken to be a file
7109 name; the data modification time of that file is used as the date.
7111 @opindex newer-mtime
7112 @item --newer-mtime=@var{date}
7113 Acts like @option{--after-date}, but only looks at data modification times.
7116 These options limit @command{tar} to operate only on files which have
7117 been modified after the date specified. A file's status is considered to have
7118 changed if its contents have been modified, or if its owner,
7119 permissions, and so forth, have been changed. (For more information on
7120 how to specify a date, see @ref{Date input formats}; remember that the
7121 entire date argument must be quoted if it contains any spaces.)
7123 Gurus would say that @option{--after-date} tests both the data
7124 modification time (@code{mtime}, the time the contents of the file
7125 were last modified) and the status change time (@code{ctime}, the time
7126 the file's status was last changed: owner, permissions, etc.@:)
7127 fields, while @option{--newer-mtime} tests only the @code{mtime}
7130 To be precise, @option{--after-date} checks @emph{both} @code{mtime} and
7131 @code{ctime} and processes the file if either one is more recent than
7132 @var{date}, while @option{--newer-mtime} only checks @code{mtime} and
7133 disregards @code{ctime}. Neither does it use @code{atime} (the last time the
7134 contents of the file were looked at).
7136 Date specifiers can have embedded spaces. Because of this, you may need
7137 to quote date arguments to keep the shell from parsing them as separate
7138 arguments. For example, the following command will add to the archive
7139 all the files modified less than two days ago:
7142 $ @kbd{tar -cf foo.tar --newer-mtime '2 days ago'}
7146 @strong{Please Note:} @option{--after-date} and @option{--newer-mtime}
7147 should not be used for incremental backups. @xref{Incremental Dumps},
7148 for proper way of creating incremental backups.
7152 @section Descending into Directories
7154 @cindex Avoiding recursion in directories
7155 @cindex Descending directories, avoiding
7156 @cindex Directories, avoiding recursion
7157 @cindex Recursion in directories, avoiding
7159 @FIXME{arrggh! this is still somewhat confusing to me. :-< }
7161 Usually, @command{tar} will recursively explore all directories (either
7162 those given on the command line or through the @option{--files-from}
7163 option) for the various files they contain. However, you may not always
7164 want @command{tar} to act this way.
7166 @opindex no-recursion
7167 The @option{--no-recursion} option inhibits @command{tar}'s recursive descent
7168 into specified directories. If you specify @option{--no-recursion}, you can
7169 use the @command{find} utility for hunting through levels of directories to
7170 construct a list of file names which you could then pass to @command{tar}.
7171 @command{find} allows you to be more selective when choosing which files to
7172 archive; see @ref{files}, for more information on using @command{find} with
7173 @command{tar}, or look.
7176 @item --no-recursion
7177 Prevents @command{tar} from recursively descending directories.
7181 Requires @command{tar} to recursively descend directories.
7182 This is the default.
7185 When you use @option{--no-recursion}, @GNUTAR{} grabs
7186 directory entries themselves, but does not descend on them
7187 recursively. Many people use @command{find} for locating files they
7188 want to back up, and since @command{tar} @emph{usually} recursively
7189 descends on directories, they have to use the @samp{@w{-not -type d}}
7190 test in their @command{find} invocation (@pxref{Type, Type, Type test,
7191 find, Finding Files}), as they usually do not want all the files in a
7192 directory. They then use the @option{--files-from} option to archive
7193 the files located via @command{find}.
7195 The problem when restoring files archived in this manner is that the
7196 directories themselves are not in the archive; so the
7197 @option{--same-permissions} (@option{--preserve-permissions},
7198 @option{-p}) option does not affect them---while users might really
7199 like it to. Specifying @option{--no-recursion} is a way to tell
7200 @command{tar} to grab only the directory entries given to it, adding
7201 no new files on its own. To summarize, if you use @command{find} to
7202 create a list of files to be stored in an archive, use it as follows:
7206 $ @kbd{find @var{dir} @var{tests} | \
7207 tar -cf @var{archive} -T - --no-recursion}
7211 The @option{--no-recursion} option also applies when extracting: it
7212 causes @command{tar} to extract only the matched directory entries, not
7213 the files under those directories.
7215 The @option{--no-recursion} option also affects how globbing patterns
7216 are interpreted (@pxref{controlling pattern-matching}).
7218 The @option{--no-recursion} and @option{--recursion} options apply to
7219 later options and operands, and can be overridden by later occurrences
7220 of @option{--no-recursion} and @option{--recursion}. For example:
7223 $ @kbd{tar -cf jams.tar --no-recursion grape --recursion grape/concord}
7227 creates an archive with one entry for @file{grape}, and the recursive
7228 contents of @file{grape/concord}, but no entries under @file{grape}
7229 other than @file{grape/concord}.
7232 @section Crossing File System Boundaries
7233 @cindex File system boundaries, not crossing
7236 @command{tar} will normally automatically cross file system boundaries in
7237 order to archive files which are part of a directory tree. You can
7238 change this behavior by running @command{tar} and specifying
7239 @option{--one-file-system}. This option only affects files that are
7240 archived because they are in a directory that is being archived;
7241 @command{tar} will still archive files explicitly named on the command line
7242 or through @option{--files-from}, regardless of where they reside.
7245 @opindex one-file-system
7246 @item --one-file-system
7247 Prevents @command{tar} from crossing file system boundaries when
7248 archiving. Use in conjunction with any write operation.
7251 The @option{--one-file-system} option causes @command{tar} to modify its
7252 normal behavior in archiving the contents of directories. If a file in
7253 a directory is not on the same file system as the directory itself, then
7254 @command{tar} will not archive that file. If the file is a directory
7255 itself, @command{tar} will not archive anything beneath it; in other words,
7256 @command{tar} will not cross mount points.
7258 This option is useful for making full or incremental archival backups of
7259 a file system. If this option is used in conjunction with
7260 @option{--verbose} (@option{-v}), files that are excluded are
7261 mentioned by name on the standard error.
7264 * directory:: Changing Directory
7265 * absolute:: Absolute File Names
7269 @subsection Changing the Working Directory
7272 @FIXME{need to read over this node now for continuity; i've switched
7273 things around some.}
7275 @cindex Changing directory mid-stream
7276 @cindex Directory, changing mid-stream
7277 @cindex Working directory, specifying
7278 To change the working directory in the middle of a list of file names,
7279 either on the command line or in a file specified using
7280 @option{--files-from} (@option{-T}), use @option{--directory} (@option{-C}).
7281 This will change the working directory to the specified directory
7282 after that point in the list.
7286 @item --directory=@var{directory}
7287 @itemx -C @var{directory}
7288 Changes the working directory in the middle of a command line.
7294 $ @kbd{tar -c -f jams.tar grape prune -C food cherry}
7298 will place the files @file{grape} and @file{prune} from the current
7299 directory into the archive @file{jams.tar}, followed by the file
7300 @file{cherry} from the directory @file{food}. This option is especially
7301 useful when you have several widely separated files that you want to
7302 store in the same archive.
7304 Note that the file @file{cherry} is recorded in the archive under the
7305 precise name @file{cherry}, @emph{not} @file{food/cherry}. Thus, the
7306 archive will contain three files that all appear to have come from the
7307 same directory; if the archive is extracted with plain @samp{tar
7308 --extract}, all three files will be written in the current directory.
7310 Contrast this with the command,
7313 $ @kbd{tar -c -f jams.tar grape prune -C food red/cherry}
7317 which records the third file in the archive under the name
7318 @file{red/cherry} so that, if the archive is extracted using
7319 @samp{tar --extract}, the third file will be written in a subdirectory
7320 named @file{orange-colored}.
7322 You can use the @option{--directory} option to make the archive
7323 independent of the original name of the directory holding the files.
7324 The following command places the files @file{/etc/passwd},
7325 @file{/etc/hosts}, and @file{/lib/libc.a} into the archive
7329 $ @kbd{tar -c -f foo.tar -C /etc passwd hosts -C /lib libc.a}
7333 However, the names of the archive members will be exactly what they were
7334 on the command line: @file{passwd}, @file{hosts}, and @file{libc.a}.
7335 They will not appear to be related by file name to the original
7336 directories where those files were located.
7338 Note that @option{--directory} options are interpreted consecutively. If
7339 @option{--directory} specifies a relative file name, it is interpreted
7340 relative to the then current directory, which might not be the same as
7341 the original current working directory of @command{tar}, due to a previous
7342 @option{--directory} option.
7344 When using @option{--files-from} (@pxref{files}), you can put various
7345 @command{tar} options (including @option{-C}) in the file list. Notice,
7346 however, that in this case the option and its argument may not be
7347 separated by whitespace. If you use short option, its argument must
7348 either follow the option letter immediately, without any intervening
7349 whitespace, or occupy the next line. Otherwise, if you use long
7350 option, separate its argument by an equal sign.
7352 For instance, the file list for the above example will be:
7367 To use it, you would invoke @command{tar} as follows:
7370 $ @kbd{tar -c -f foo.tar --files-from list}
7373 Notice also that you can only use the short option variant in the file
7374 list, i.e., always use @option{-C}, not @option{--directory}.
7376 The interpretation of @option{--directory} is disabled by
7377 @option{--null} option.
7380 @subsection Absolute File Names
7384 @opindex absolute-names
7385 @item --absolute-names
7387 Do not strip leading slashes from file names, and permit file names
7388 containing a @file{..} file name component.
7391 By default, @GNUTAR{} drops a leading @samp{/} on
7392 input or output, and complains about file names containing a @file{..}
7393 component. This option turns off this behavior.
7395 When @command{tar} extracts archive members from an archive, it strips any
7396 leading slashes (@samp{/}) from the member name. This causes absolute
7397 member names in the archive to be treated as relative file names. This
7398 allows you to have such members extracted wherever you want, instead of
7399 being restricted to extracting the member in the exact directory named
7400 in the archive. For example, if the archive member has the name
7401 @file{/etc/passwd}, @command{tar} will extract it as if the name were
7402 really @file{etc/passwd}.
7404 File names containing @file{..} can cause problems when extracting, so
7405 @command{tar} normally warns you about such files when creating an
7406 archive, and rejects attempts to extracts such files.
7408 Other @command{tar} programs do not do this. As a result, if you
7409 create an archive whose member names start with a slash, they will be
7410 difficult for other people with a non-@GNUTAR{}
7411 program to use. Therefore, @GNUTAR{} also strips
7412 leading slashes from member names when putting members into the
7413 archive. For example, if you ask @command{tar} to add the file
7414 @file{/bin/ls} to an archive, it will do so, but the member name will
7415 be @file{bin/ls}.@footnote{A side effect of this is that when
7416 @option{--create} is used with @option{--verbose} the resulting output
7417 is not, generally speaking, the same as the one you'd get running
7418 @kbd{tar --list} command. This may be important if you use some
7419 scripts for comparing both outputs. @xref{listing member and file names},
7420 for the information on how to handle this case.}
7422 If you use the @option{--absolute-names} (@option{-P}) option,
7423 @command{tar} will do none of these transformations.
7425 To archive or extract files relative to the root directory, specify
7426 the @option{--absolute-names} (@option{-P}) option.
7428 Normally, @command{tar} acts on files relative to the working
7429 directory---ignoring superior directory names when archiving, and
7430 ignoring leading slashes when extracting.
7432 When you specify @option{--absolute-names} (@option{-P}),
7433 @command{tar} stores file names including all superior directory
7434 names, and preserves leading slashes. If you only invoked
7435 @command{tar} from the root directory you would never need the
7436 @option{--absolute-names} option, but using this option
7437 may be more convenient than switching to root.
7439 @FIXME{Should be an example in the tutorial/wizardry section using this
7440 to transfer files between systems.}
7442 @FIXME{Is write access an issue?}
7445 @item --absolute-names
7446 Preserves full file names (including superior directory names) when
7447 archiving files. Preserves leading slash when extracting files.
7451 @FIXME{this is still horrible; need to talk with dan on monday.}
7453 @command{tar} prints out a message about removing the @samp{/} from
7454 file names. This message appears once per @GNUTAR{}
7455 invocation. It represents something which ought to be told; ignoring
7456 what it means can cause very serious surprises, later.
7458 Some people, nevertheless, do not want to see this message. Wanting to
7459 play really dangerously, one may of course redirect @command{tar} standard
7460 error to the sink. For example, under @command{sh}:
7463 $ @kbd{tar -c -f archive.tar /home 2> /dev/null}
7467 Another solution, both nicer and simpler, would be to change to
7468 the @file{/} directory first, and then avoid absolute notation.
7472 $ @kbd{(cd / && tar -c -f archive.tar home)}
7474 $ @kbd{tar -c -f archive.tar -C / home}
7477 @include getdate.texi
7480 @chapter Controlling the Archive Format
7482 @cindex Tar archive formats
7483 Due to historical reasons, there are several formats of tar archives.
7484 All of them are based on the same principles, but have some subtle
7485 differences that often make them incompatible with each other.
7487 GNU tar is able to create and handle archives in a variety of formats.
7488 The most frequently used formats are (in alphabetical order):
7492 Format used by @GNUTAR{} versions up to 1.13.25. This format derived
7493 from an early @acronym{POSIX} standard, adding some improvements such as
7494 sparse file handling and incremental archives. Unfortunately these
7495 features were implemented in a way incompatible with other archive
7498 Archives in @samp{gnu} format are able to hold pathnames of unlimited
7502 Format used by @GNUTAR{} of versions prior to 1.12.
7505 Archive format, compatible with the V7 implementation of tar. This
7506 format imposes a number of limitations. The most important of them
7510 @item The maximum length of a file name is limited to 99 characters.
7511 @item The maximum length of a symbolic link is limited to 99 characters.
7512 @item It is impossible to store special files (block and character
7513 devices, fifos etc.)
7514 @item Maximum value of user or group ID is limited to 2097151 (7777777
7516 @item V7 archives do not contain symbolic ownership information (user
7517 and group name of the file owner).
7520 This format has traditionally been used by Automake when producing
7521 Makefiles. This practice will change in the future, in the meantime,
7522 however this means that projects containing filenames more than 99
7523 characters long will not be able to use @GNUTAR{} @value{VERSION} and
7524 Automake prior to 1.9.
7527 Archive format defined by @acronym{POSIX.1-1988} specification. It stores
7528 symbolic ownership information. It is also able to store
7529 special files. However, it imposes several restrictions as well:
7532 @item The maximum length of a file name is limited to 256 characters,
7533 provided that the filename can be split at directory separator in
7534 two parts, first of them being at most 155 bytes long. So, in most
7535 cases the maximum file name length will be shorter than 256
7537 @item The maximum length of a symbolic link name is limited to
7539 @item Maximum size of a file the archive is able to accomodate
7541 @item Maximum value of UID/GID is 2097151.
7542 @item Maximum number of bits in device major and minor numbers is 21.
7546 Format used by J@"org Schilling @command{star}
7547 implementation. @GNUTAR{} is able to read @samp{star} archives but
7548 currently does not produce them.
7551 Archive format defined by @acronym{POSIX.1-2001} specification. This is the
7552 most flexible and feature-rich format. It does not impose any
7553 restrictions on file sizes or filename lengths. This format is quite
7554 recent, so not all tar implementations are able to handle it properly.
7555 However, this format is designed in such a way that any tar
7556 implementation able to read @samp{ustar} archives will be able to read
7557 most @samp{posix} archives as well, with the only exception that any
7558 additional information (such as long file names etc.) will in such
7559 case be extracted as plain text files along with the files it refers to.
7561 This archive format will be the default format for future versions
7566 The following table summarizes the limitations of each of these
7569 @multitable @columnfractions .10 .20 .20 .20 .20
7570 @headitem Format @tab UID @tab File Size @tab Path Name @tab Devn
7571 @item gnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
7572 @item oldgnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
7573 @item v7 @tab 2097151 @tab 8GB @tab 99 @tab n/a
7574 @item ustar @tab 2097151 @tab 8GB @tab 256 @tab 21
7575 @item posix @tab Unlimited @tab Unlimited @tab Unlimited @tab Unlimited
7578 The default format for @GNUTAR{} is defined at compilation
7579 time. You may check it by running @command{tar --help}, and examining
7580 the last lines of its output. Usually, @GNUTAR{} is configured
7581 to create archives in @samp{gnu} format, however, future version will
7582 switch to @samp{posix}.
7585 * Portability:: Making @command{tar} Archives More Portable
7586 * Compression:: Using Less Space through Compression
7587 * Attributes:: Handling File Attributes
7588 * cpio:: Comparison of @command{tar} and @command{cpio}
7592 @section Making @command{tar} Archives More Portable
7594 Creating a @command{tar} archive on a particular system that is meant to be
7595 useful later on many other machines and with other versions of @command{tar}
7596 is more challenging than you might think. @command{tar} archive formats
7597 have been evolving since the first versions of Unix. Many such formats
7598 are around, and are not always compatible with each other. This section
7599 discusses a few problems, and gives some advice about making @command{tar}
7600 archives more portable.
7602 One golden rule is simplicity. For example, limit your @command{tar}
7603 archives to contain only regular files and directories, avoiding
7604 other kind of special files. Do not attempt to save sparse files or
7605 contiguous files as such. Let's discuss a few more problems, in turn.
7607 @FIXME{Discuss GNU extensions (incremental backups, multi-volume
7608 archives and archive labels) in GNU and PAX formats.}
7611 * Portable Names:: Portable Names
7612 * dereference:: Symbolic Links
7613 * old:: Old V7 Archives
7614 * ustar:: Ustar Archives
7615 * gnu:: GNU and old GNU format archives.
7616 * posix:: @acronym{POSIX} archives
7617 * Checksumming:: Checksumming Problems
7618 * Large or Negative Values:: Large files, negative time stamps, etc.
7621 @node Portable Names
7622 @subsection Portable Names
7624 Use portable file and member names. A name is portable if it contains
7625 only ASCII letters and digits, @samp{/}, @samp{.}, @samp{_}, and
7626 @samp{-}; it cannot be empty, start with @samp{-} or @samp{//}, or
7627 contain @samp{/-}. Avoid deep directory nesting. For portability to
7628 old Unix hosts, limit your file name components to 14 characters or
7631 If you intend to have your @command{tar} archives to be read under
7632 MSDOS, you should not rely on case distinction for file names, and you
7633 might use the @acronym{GNU} @command{doschk} program for helping you
7634 further diagnosing illegal MSDOS names, which are even more limited
7638 @subsection Symbolic Links
7639 @cindex File names, using symbolic links
7640 @cindex Symbolic link as file name
7642 @opindex dereference
7643 Normally, when @command{tar} archives a symbolic link, it writes a
7644 block to the archive naming the target of the link. In that way, the
7645 @command{tar} archive is a faithful record of the file system contents.
7646 @option{--dereference} (@option{-h}) is used with @option{--create} (@option{-c}), and causes
7647 @command{tar} to archive the files symbolic links point to, instead of
7648 the links themselves. When this option is used, when @command{tar}
7649 encounters a symbolic link, it will archive the linked-to file,
7650 instead of simply recording the presence of a symbolic link.
7652 The name under which the file is stored in the file system is not
7653 recorded in the archive. To record both the symbolic link name and
7654 the file name in the system, archive the file under both names. If
7655 all links were recorded automatically by @command{tar}, an extracted file
7656 might be linked to a file name that no longer exists in the file
7659 If a linked-to file is encountered again by @command{tar} while creating
7660 the same archive, an entire second copy of it will be stored. (This
7661 @emph{might} be considered a bug.)
7663 So, for portable archives, do not archive symbolic links as such,
7664 and use @option{--dereference} (@option{-h}): many systems do not support
7665 symbolic links, and moreover, your distribution might be unusable if
7666 it contains unresolved symbolic links.
7669 @subsection Old V7 Archives
7670 @cindex Format, old style
7671 @cindex Old style format
7672 @cindex Old style archives
7673 @cindex v7 archive format
7675 Certain old versions of @command{tar} cannot handle additional
7676 information recorded by newer @command{tar} programs. To create an
7677 archive in V7 format (not ANSI), which can be read by these old
7678 versions, specify the @option{--format=v7} option in
7679 conjunction with the @option{--create} (@option{-c}) (@command{tar} also
7680 accepts @option{--portability} or @samp{op-old-archive} for this
7681 option). When you specify it,
7682 @command{tar} leaves out information about directories, pipes, fifos,
7683 contiguous files, and device files, and specifies file ownership by
7684 group and user IDs instead of group and user names.
7686 When updating an archive, do not use @option{--format=v7}
7687 unless the archive was created using this option.
7689 In most cases, a @emph{new} format archive can be read by an @emph{old}
7690 @command{tar} program without serious trouble, so this option should
7691 seldom be needed. On the other hand, most modern @command{tar}s are
7692 able to read old format archives, so it might be safer for you to
7693 always use @option{--format=v7} for your distributions.
7696 @subsection Ustar Archive Format
7698 @cindex ustar archive format
7699 Archive format defined by @acronym{POSIX}.1-1988 specification is called
7700 @code{ustar}. Although it is more flexible than the V7 format, it
7701 still has many restrictions (@xref{Formats,ustar}, for the detailed
7702 description of @code{ustar} format). Along with V7 format,
7703 @code{ustar} format is a good choice for archives intended to be read
7704 with other implementations of @command{tar}.
7706 To create archive in @code{ustar} format, use @option{--format=ustar}
7707 option in conjunction with the @option{--create} (@option{-c}).
7710 @subsection @acronym{GNU} and old @GNUTAR{} format
7712 @cindex GNU archive format
7713 @cindex Old GNU archive format
7714 @GNUTAR{} was based on an early draft of the
7715 @acronym{POSIX} 1003.1 @code{ustar} standard. @acronym{GNU} extensions to
7716 @command{tar}, such as the support for file names longer than 100
7717 characters, use portions of the @command{tar} header record which were
7718 specified in that @acronym{POSIX} draft as unused. Subsequent changes in
7719 @acronym{POSIX} have allocated the same parts of the header record for
7720 other purposes. As a result, @GNUTAR{} format is
7721 incompatible with the current @acronym{POSIX} specification, and with
7722 @command{tar} programs that follow it.
7724 In the majority of cases, @command{tar} will be configured to create
7725 this format by default. This will change in the future releases, since
7726 we plan to make @samp{posix} format the default.
7728 To force creation a @GNUTAR{} archive, use option
7729 @option{--format=gnu}.
7732 @subsection @GNUTAR{} and @acronym{POSIX} @command{tar}
7734 @cindex POSIX archive format
7735 @cindex PAX archive format
7736 Starting from version 1.14 @GNUTAR{} features full support for
7737 @acronym{POSIX.1-2001} archives.
7739 A @acronym{POSIX} conformant archive will be created if @command{tar}
7740 was given @option{--format=posix} (@option{--format=pax}) option. No
7741 special option is required to read and extract from a @acronym{POSIX}
7745 * PAX keywords:: Controlling Extended Header Keywords.
7749 @subsubsection Controlling Extended Header Keywords
7753 @item --pax-option=@var{keyword-list}
7754 Handle keywords in @acronym{PAX} extended headers. This option is
7755 equivalent to @option{-o} option of the @command{pax} utility.
7758 @var{Keyword-list} is a comma-separated
7759 list of keyword options, each keyword option taking one of
7760 the following forms:
7763 @item delete=@var{pattern}
7764 When used with one of archive-creation commands,
7765 this option instructs @command{tar} to omit from extended header records
7766 that it produces any keywords matching the string @var{pattern}.
7768 When used in extract or list mode, this option instructs tar
7769 to ignore any keywords matching the given @var{pattern} in the extended
7770 header records. In both cases, matching is performed using the pattern
7771 matching notation described in @acronym{POSIX 1003.2}, 3.13
7772 (@pxref{wildcards}). For example:
7775 --pax-option delete=security.*
7778 would suppress security-related information.
7780 @item exthdr.name=@var{string}
7782 This keyword allows user control over the name that is written into the
7783 ustar header blocks for the extended headers. The name is obtained
7784 from @var{string} after making the following substitutions:
7786 @multitable @columnfractions .25 .55
7787 @headitem Meta-character @tab Replaced By
7788 @item %d @tab The directory name of the file, equivalent to the
7789 result of the @command{dirname} utility on the translated pathname.
7790 @item %f @tab The filename of the file, equivalent to the result
7791 of the @command{basename} utility on the translated pathname.
7792 @item %p @tab The process ID of the @command{tar} process.
7793 @item %% @tab A @samp{%} character.
7796 Any other @samp{%} characters in @var{string} produce undefined
7799 If no option @samp{exthdr.name=string} is specified, @command{tar}
7800 will use the following default value:
7806 @item globexthdr.name=@var{string}
7807 This keyword allows user control over the name that is written into
7808 the ustar header blocks for global extended header records. The name
7809 is obtained from the contents of @var{string}, after making
7810 the following substitutions:
7812 @multitable @columnfractions .25 .55
7813 @headitem Meta-character @tab Replaced By
7814 @item %n @tab An integer that represents the
7815 sequence number of the global extended header record in the archive,
7817 @item %p @tab The process ID of the @command{tar} process.
7818 @item %% @tab A @samp{%} character.
7821 Any other @samp{%} characters in @var{string} produce undefined results.
7823 If no option @samp{globexthdr.name=string} is specified, @command{tar}
7824 will use the following default value:
7827 $TMPDIR/GlobalHead.%p.%n
7831 where @samp{$TMPDIR} represents the value of the @var{TMPDIR}
7832 environment variable. If @var{TMPDIR} is not set, @command{tar}
7835 @item @var{keyword}=@var{value}
7836 When used with one of archive-creation commands, these keyword/value pairs
7837 will be included at the beginning of the archive in a global extended
7838 header record. When used with one of archive-reading commands,
7839 @command{tar} will behave as if it has encountered these keyword/value
7840 pairs at the beginning of the archive in a global extended header
7843 @item @var{keyword}:=@var{value}
7844 When used with one of archive-creation commands, these keyword/value pairs
7845 will be included as records at the beginning of an extended header for
7846 each file. This is effectively equivalent to @var{keyword}=@var{value}
7847 form except that it creates no global extended header records.
7849 When used with one of archive-reading commands, @command{tar} will
7850 behave as if these keyword/value pairs were included as records at the
7851 end of each extended header; thus, they will override any global or
7852 file-specific extended header record keywords of the same names.
7853 For example, in the command:
7856 tar --format=posix --create \
7857 --file archive --pax-option gname:=user .
7860 the group name will be forced to a new value for all files
7861 stored in the archive.
7865 @subsection Checksumming Problems
7867 SunOS and HP-UX @command{tar} fail to accept archives created using
7868 @GNUTAR{} and containing non-ASCII file names, that
7869 is, file names having characters with the eight bit set, because they
7870 use signed checksums, while @GNUTAR{} uses unsigned
7871 checksums while creating archives, as per @acronym{POSIX} standards. On
7872 reading, @GNUTAR{} computes both checksums and
7873 accept any. It is somewhat worrying that a lot of people may go
7874 around doing backup of their files using faulty (or at least
7875 non-standard) software, not learning about it until it's time to
7876 restore their missing files with an incompatible file extractor, or
7879 @GNUTAR{} compute checksums both ways, and accept
7880 any on read, so @acronym{GNU} tar can read Sun tapes even with their
7881 wrong checksums. @GNUTAR{} produces the standard
7882 checksum, however, raising incompatibilities with Sun. That is to
7883 say, @GNUTAR{} has not been modified to
7884 @emph{produce} incorrect archives to be read by buggy @command{tar}'s.
7885 I've been told that more recent Sun @command{tar} now read standard
7886 archives, so maybe Sun did a similar patch, after all?
7888 The story seems to be that when Sun first imported @command{tar}
7889 sources on their system, they recompiled it without realizing that
7890 the checksums were computed differently, because of a change in
7891 the default signing of @code{char}'s in their compiler. So they
7892 started computing checksums wrongly. When they later realized their
7893 mistake, they merely decided to stay compatible with it, and with
7894 themselves afterwards. Presumably, but I do not really know, HP-UX
7895 has chosen that their @command{tar} archives to be compatible with Sun's.
7896 The current standards do not favor Sun @command{tar} format. In any
7897 case, it now falls on the shoulders of SunOS and HP-UX users to get
7898 a @command{tar} able to read the good archives they receive.
7900 @node Large or Negative Values
7901 @subsection Large or Negative Values
7902 @cindex large values
7903 @cindex future time stamps
7904 @cindex negative time stamps
7907 The above sections suggest to use @samp{oldest possible} archive
7908 format if in doubt. However, sometimes it is not possible. If you
7909 attempt to archive a file whose metadata cannot be represented using
7910 required format, @GNUTAR{} will print error message and ignore such a
7911 file. You will than have to switch to a format that is able to
7912 handle such values. The format summary table (@pxref{Formats}) will
7915 In particular, when trying to archive files larger than 8GB or with
7916 timestamps not in the range 1970-01-01 00:00:00 through 2242-03-16
7917 12:56:31 @sc{utc}, you will have to chose between @acronym{GNU} and
7918 @acronym{POSIX} archive formats. When considering which format to
7919 choose, bear in mind that the @acronym{GNU} format uses
7920 two's-complement base-256 notation to store values that do not fit
7921 into standard @acronym{ustar} range. Such archives can generally be
7922 read only by a @GNUTAR{} implementation. Moreover, they sometimes
7923 cannot be correctly restored on another hosts even by @GNUTAR{}. For
7924 example, using two's complement representation for negative time
7925 stamps that assumes a signed 32-bit @code{time_t} generates archives
7926 that are not portable to hosts with differing @code{time_t}
7929 On the other hand, @acronym{POSIX} archives, generally speaking, can
7930 be extracted by any tar implementation that understands older
7931 @acronym{ustar} format. The only exception are files larger than 8GB.
7933 @FIXME{Describe how @acronym{POSIX} archives are extracted by non
7937 @section Using Less Space through Compression
7940 * gzip:: Creating and Reading Compressed Archives
7941 * sparse:: Archiving Sparse Files
7945 @subsection Creating and Reading Compressed Archives
7946 @cindex Compressed archives
7947 @cindex Storing archives in compressed format
7949 @GNUTAR{} is able to create and read compressed archives. It supports
7950 @command{gzip} and @command{bzip2} compression programs. For backward
7951 compatibilty, it also supports @command{compress} command, although
7952 we strongly recommend against using it, since there is a patent
7953 covering the algorithm it uses and you could be sued for patent
7954 infringement merely by running @command{compress}! Besides, it is less
7955 effective than @command{gzip} and @command{bzip2}.
7957 Creating a compressed archive is simple: you just specify a
7958 @dfn{compression option} along with the usual archive creation
7959 commands. The compression option is @option{-z} (@option{--gzip}) to
7960 create a @command{gzip} compressed archive, @option{-j}
7961 (@option{--bzip2}) to create a @command{bzip2} compressed archive, and
7962 @option{-Z} (@option{--compress}) to use @command{compress} program.
7966 $ @kbd{tar cfz archive.tar.gz .}
7969 Reading compressed archive is even simpler: you don't need to specify
7970 any additional options as @GNUTAR{} recognizes its format
7971 automatically. Thus, the following commands will list and extract the
7972 archive created in previous example:
7975 # List the compressed archive
7976 $ @kbd{tar tf archive.tar.gz}
7977 # Extract the compressed archive
7978 $ @kbd{tar xf archive.tar.gz}
7981 The only case when you have to specify a decompression option while
7982 reading the archive is when reading from a pipe or from a tape drive
7983 that does not support random access. However, in this case @GNUTAR{}
7984 will indicate which option you should use. For example:
7987 $ @kbd{cat archive.tar.gz | tar tf -}
7988 tar: Archive is compressed. Use -z option
7989 tar: Error is not recoverable: exiting now
7992 If you see such diagnostics, just add the suggested option to the
7993 invocation of @GNUTAR{}:
7996 $ @kbd{cat archive.tar.gz | tar tfz -}
7999 Notice also, that there are several restrictions on operations on
8000 compressed archives. First of all, compressed archives cannot be
8001 modified, i.e., you cannot update (@option{--update} (@option{-u})) them or delete
8002 (@option{--delete}) members from them. Likewise, you cannot append
8003 another @command{tar} archive to a compressed archive using
8004 @option{--append} (@option{-r})). Secondly, multi-volume archives cannot be
8007 The following table summarizes compression options used by @GNUTAR{}.
8015 Filter the archive through @command{gzip}.
8017 You can use @option{--gzip} and @option{--gunzip} on physical devices
8018 (tape drives, etc.) and remote files as well as on normal files; data
8019 to or from such devices or remote files is reblocked by another copy
8020 of the @command{tar} program to enforce the specified (or default) record
8021 size. The default compression parameters are used; if you need to
8022 override them, set @env{GZIP} environment variable, e.g.:
8025 $ @kbd{GZIP=--best tar cfz archive.tar.gz subdir}
8029 Another way would be to avoid the @option{--gzip} (@option{--gunzip}, @option{--ungzip}, @option{-z}) option and run
8030 @command{gzip} explicitly:
8033 $ @kbd{tar cf - subdir | gzip --best -c - > archive.tar.gz}
8036 @cindex corrupted archives
8037 About corrupted compressed archives: @command{gzip}'ed files have no
8038 redundancy, for maximum compression. The adaptive nature of the
8039 compression scheme means that the compression tables are implicitly
8040 spread all over the archive. If you lose a few blocks, the dynamic
8041 construction of the compression tables becomes unsynchronized, and there
8042 is little chance that you could recover later in the archive.
8044 There are pending suggestions for having a per-volume or per-file
8045 compression in @GNUTAR{}. This would allow for viewing the
8046 contents without decompression, and for resynchronizing decompression at
8047 every volume or file, in case of corrupted archives. Doing so, we might
8048 lose some compressibility. But this would have make recovering easier.
8049 So, there are pros and cons. We'll see!
8054 Filter the archive through @code{bzip2}. Otherwise like @option{--gzip}.
8061 Filter the archive through @command{compress}. Otherwise like @option{--gzip}.
8063 The @acronym{GNU} Project recommends you not use
8064 @command{compress}, because there is a patent covering the algorithm it
8065 uses. You could be sued for patent infringement merely by running
8068 @opindex use-compress-program
8069 @item --use-compress-program=@var{prog}
8070 Use external compression program @var{prog}. Use this option if you
8071 have a compression program that @GNUTAR{} does not support. There
8072 are two requirements to which @var{prog} should comply:
8074 First, when called without options, it should read data from standard
8075 input, compress it and output it on standard output.
8077 Secondly, if called with @option{-d} argument, it should do exactly
8078 the opposite, i.e., read the compressed data from the standard input
8079 and produce uncompressed data on the standard output.
8082 @cindex gpg, using with tar
8083 @cindex gnupg, using with tar
8084 @cindex Using encrypted archives
8085 The @option{--use-compress-program} option, in particular, lets you
8086 implement your own filters, not necessarily dealing with
8087 compression/decomression. For example, suppose you wish to implement
8088 PGP encryption on top of compression, using @command{gpg} (@pxref{Top,
8089 gpg, gpg ---- encryption and signing tool, gpg, GNU Privacy Guard
8090 Manual}). The following script does that:
8096 -d) gpg --decrypt - | gzip -d -c;;
8097 '') gzip -c | gpg -s ;;
8098 *) echo "Unknown option $1">&2; exit 1;;
8103 Suppose you name it @file{gpgz} and save it somewhere in your
8104 @env{PATH}. Then the following command will create a commpressed
8105 archive signed with your private key:
8108 $ @kbd{tar -cf foo.tar.gpgz --use-compress=gpgz .}
8112 Likewise, the following command will list its contents:
8115 $ @kbd{tar -tf foo.tar.gpgz --use-compress=gpgz .}
8119 The above is based on the following discussion:
8121 I have one question, or maybe it's a suggestion if there isn't a way
8122 to do it now. I would like to use @option{--gzip}, but I'd also like
8123 the output to be fed through a program like @acronym{GNU}
8124 @command{ecc} (actually, right now that's @samp{exactly} what I'd like
8125 to use :-)), basically adding ECC protection on top of compression.
8126 It seems as if this should be quite easy to do, but I can't work out
8127 exactly how to go about it. Of course, I can pipe the standard output
8128 of @command{tar} through @command{ecc}, but then I lose (though I
8129 haven't started using it yet, I confess) the ability to have
8130 @command{tar} use @command{rmt} for it's I/O (I think).
8132 I think the most straightforward thing would be to let me specify a
8133 general set of filters outboard of compression (preferably ordered,
8134 so the order can be automatically reversed on input operations, and
8135 with the options they require specifiable), but beggars shouldn't be
8136 choosers and anything you decide on would be fine with me.
8138 By the way, I like @command{ecc} but if (as the comments say) it can't
8139 deal with loss of block sync, I'm tempted to throw some time at adding
8140 that capability. Supposing I were to actually do such a thing and
8141 get it (apparently) working, do you accept contributed changes to
8142 utilities like that? (Leigh Clayton @file{loc@@soliton.com}, May 1995).
8144 Isn't that exactly the role of the
8145 @option{--use-compress-prog=@var{program}} option?
8146 I never tried it myself, but I suspect you may want to write a
8147 @var{prog} script or program able to filter stdin to stdout to
8148 way you want. It should recognize the @option{-d} option, for when
8149 extraction is needed rather than creation.
8151 It has been reported that if one writes compressed data (through the
8152 @option{--gzip} or @option{--compress} options) to a DLT and tries to use
8153 the DLT compression mode, the data will actually get bigger and one will
8154 end up with less space on the tape.
8158 @subsection Archiving Sparse Files
8159 @cindex Sparse Files
8166 Handle sparse files efficiently.
8169 This option causes all files to be put in the archive to be tested for
8170 sparseness, and handled specially if they are. The @option{--sparse}
8171 (@option{-S}) option is useful when many @code{dbm} files, for example, are being
8172 backed up. Using this option dramatically decreases the amount of
8173 space needed to store such a file.
8175 In later versions, this option may be removed, and the testing and
8176 treatment of sparse files may be done automatically with any special
8177 @acronym{GNU} options. For now, it is an option needing to be specified on
8178 the command line with the creation or updating of an archive.
8180 Files in the file system occasionally have ``holes.'' A hole in a file
8181 is a section of the file's contents which was never written. The
8182 contents of a hole read as all zeros. On many operating systems,
8183 actual disk storage is not allocated for holes, but they are counted
8184 in the length of the file. If you archive such a file, @command{tar}
8185 could create an archive longer than the original. To have @command{tar}
8186 attempt to recognize the holes in a file, use @option{--sparse} (@option{-S}). When
8187 you use this option, then, for any file using less disk space than
8188 would be expected from its length, @command{tar} searches the file for
8189 consecutive stretches of zeros. It then records in the archive for
8190 the file where the consecutive stretches of zeros are, and only
8191 archives the ``real contents'' of the file. On extraction (using
8192 @option{--sparse} is not needed on extraction) any such
8193 files have holes created wherever the continuous stretches of zeros
8194 were found. Thus, if you use @option{--sparse}, @command{tar} archives
8195 won't take more space than the original.
8197 A file is sparse if it contains blocks of zeros whose existence is
8198 recorded, but that have no space allocated on disk. When you specify
8199 the @option{--sparse} option in conjunction with the @option{--create}
8200 (@option{-c}) operation, @command{tar} tests all files for sparseness
8201 while archiving. If @command{tar} finds a file to be sparse, it uses a
8202 sparse representation of the file in the archive. @xref{create}, for
8203 more information about creating archives.
8205 @option{--sparse} is useful when archiving files, such as dbm files,
8206 likely to contain many nulls. This option dramatically
8207 decreases the amount of space needed to store such an archive.
8210 @strong{Please Note:} Always use @option{--sparse} when performing file
8211 system backups, to avoid archiving the expanded forms of files stored
8212 sparsely in the system.
8214 Even if your system has no sparse files currently, some may be
8215 created in the future. If you use @option{--sparse} while making file
8216 system backups as a matter of course, you can be assured the archive
8217 will never take more space on the media than the files take on disk
8218 (otherwise, archiving a disk filled with sparse files might take
8219 hundreds of tapes). @xref{Incremental Dumps}.
8222 @command{tar} ignores the @option{--sparse} option when reading an archive.
8227 Files stored sparsely in the file system are represented sparsely in
8228 the archive. Use in conjunction with write operations.
8231 However, users should be well aware that at archive creation time,
8232 @GNUTAR{} still has to read whole disk file to
8233 locate the @dfn{holes}, and so, even if sparse files use little space
8234 on disk and in the archive, they may sometimes require inordinate
8235 amount of time for reading and examining all-zero blocks of a file.
8236 Although it works, it's painfully slow for a large (sparse) file, even
8237 though the resulting tar archive may be small. (One user reports that
8238 dumping a @file{core} file of over 400 megabytes, but with only about
8239 3 megabytes of actual data, took about 9 minutes on a Sun Sparcstation
8240 ELC, with full CPU utilization.)
8242 This reading is required in all cases and is not related to the fact
8243 the @option{--sparse} option is used or not, so by merely @emph{not}
8244 using the option, you are not saving time@footnote{Well! We should say
8245 the whole truth, here. When @option{--sparse} is selected while creating
8246 an archive, the current @command{tar} algorithm requires sparse files to be
8247 read twice, not once. We hope to develop a new archive format for saving
8248 sparse files in which one pass will be sufficient.}.
8250 Programs like @command{dump} do not have to read the entire file; by
8251 examining the file system directly, they can determine in advance
8252 exactly where the holes are and thus avoid reading through them. The
8253 only data it need read are the actual allocated data blocks.
8254 @GNUTAR{} uses a more portable and straightforward
8255 archiving approach, it would be fairly difficult that it does
8256 otherwise. Elizabeth Zwicky writes to @file{comp.unix.internals}, on
8260 What I did say is that you cannot tell the difference between a hole and an
8261 equivalent number of nulls without reading raw blocks. @code{st_blocks} at
8262 best tells you how many holes there are; it doesn't tell you @emph{where}.
8263 Just as programs may, conceivably, care what @code{st_blocks} is (care
8264 to name one that does?), they may also care where the holes are (I have
8265 no examples of this one either, but it's equally imaginable).
8267 I conclude from this that good archivers are not portable. One can
8268 arguably conclude that if you want a portable program, you can in good
8269 conscience restore files with as many holes as possible, since you can't
8274 @section Handling File Attributes
8277 When @command{tar} reads files, it updates their access times. To
8278 avoid this, use the @option{--atime-preserve[=METHOD]} option, which can either
8279 reset the access time retroactively or avoid changing it in the first
8282 Handling of file attributes
8285 @opindex atime-preserve
8286 @item --atime-preserve
8287 @itemx --atime-preserve=replace
8288 @itemx --atime-preserve=system
8289 Preserve the access times of files that are read. This works only for
8290 files that you own, unless you have superuser privileges.
8292 @option{--atime-preserve=replace} works on most systems, but it also
8293 restores the data modification time and updates the status change
8294 time. Hence it doesn't interact with incremental dumps nicely
8295 (@pxref{Backups}), and it can set access or data modification times
8296 incorrectly if other programs access the file while @command{tar} is
8299 @option{--atime-preserve=system} avoids changing the access time in
8300 the first place, if the operating system supports this.
8301 Unfortunately, this may or may not work on any given operating system
8302 or file system. If @command{tar} knows for sure it won't work, it
8303 complains right away.
8305 Currently @option{--atime-preserve} with no operand defaults to
8306 @option{--atime-preserve=replace}, but this is intended to change to
8307 @option{--atime-preserve=system} when the latter is better-supported.
8312 Do not extract data modification time.
8314 When this option is used, @command{tar} leaves the data modification times
8315 of the files it extracts as the times when the files were extracted,
8316 instead of setting it to the times recorded in the archive.
8318 This option is meaningless with @option{--list} (@option{-t}).
8322 Create extracted files with the same ownership they have in the
8325 This is the default behavior for the superuser,
8326 so this option is meaningful only for non-root users, when @command{tar}
8327 is executed on those systems able to give files away. This is
8328 considered as a security flaw by many people, at least because it
8329 makes quite difficult to correctly account users for the disk space
8330 they occupy. Also, the @code{suid} or @code{sgid} attributes of
8331 files are easily and silently lost when files are given away.
8333 When writing an archive, @command{tar} writes the user id and user name
8334 separately. If it can't find a user name (because the user id is not
8335 in @file{/etc/passwd}), then it does not write one. When restoring,
8336 it tries to look the name (if one was written) up in
8337 @file{/etc/passwd}. If it fails, then it uses the user id stored in
8338 the archive instead.
8340 @opindex no-same-owner
8341 @item --no-same-owner
8343 Do not attempt to restore ownership when extracting. This is the
8344 default behavior for ordinary users, so this option has an effect
8345 only for the superuser.
8347 @opindex numeric-owner
8348 @item --numeric-owner
8349 The @option{--numeric-owner} option allows (ANSI) archives to be written
8350 without user/group name information or such information to be ignored
8351 when extracting. It effectively disables the generation and/or use
8352 of user/group name information. This option forces extraction using
8353 the numeric ids from the archive, ignoring the names.
8355 This is useful in certain circumstances, when restoring a backup from
8356 an emergency floppy with different passwd/group files for example.
8357 It is otherwise impossible to extract files with the right ownerships
8358 if the password file in use during the extraction does not match the
8359 one belonging to the file system(s) being extracted. This occurs,
8360 for example, if you are restoring your files after a major crash and
8361 had booted from an emergency floppy with no password file or put your
8362 disk into another machine to do the restore.
8364 The numeric ids are @emph{always} saved into @command{tar} archives.
8365 The identifying names are added at create time when provided by the
8366 system, unless @option{--old-archive} (@option{-o}) is used. Numeric ids could be
8367 used when moving archives between a collection of machines using
8368 a centralized management for attribution of numeric ids to users
8369 and groups. This is often made through using the NIS capabilities.
8371 When making a @command{tar} file for distribution to other sites, it
8372 is sometimes cleaner to use a single owner for all files in the
8373 distribution, and nicer to specify the write permission bits of the
8374 files as stored in the archive independently of their actual value on
8375 the file system. The way to prepare a clean distribution is usually
8376 to have some Makefile rule creating a directory, copying all needed
8377 files in that directory, then setting ownership and permissions as
8378 wanted (there are a lot of possible schemes), and only then making a
8379 @command{tar} archive out of this directory, before cleaning
8380 everything out. Of course, we could add a lot of options to
8381 @GNUTAR{} for fine tuning permissions and ownership.
8382 This is not the good way, I think. @GNUTAR{} is
8383 already crowded with options and moreover, the approach just explained
8384 gives you a great deal of control already.
8386 @opindex same-permissions, short description
8387 @opindex preserve-permissions, short description
8389 @itemx --same-permissions
8390 @itemx --preserve-permissions
8391 Extract all protection information.
8393 This option causes @command{tar} to set the modes (access permissions) of
8394 extracted files exactly as recorded in the archive. If this option
8395 is not used, the current @code{umask} setting limits the permissions
8396 on extracted files. This option is by default enabled when
8397 @command{tar} is executed by a superuser.
8400 This option is meaningless with @option{--list} (@option{-t}).
8404 Same as both @option{--same-permissions} and @option{--same-order}.
8406 The @option{--preserve} option has no equivalent short option name.
8407 It is equivalent to @option{--same-permissions} plus @option{--same-order}.
8409 @FIXME{I do not see the purpose of such an option. (Neither I. FP.)
8410 Neither do I. --Sergey}
8415 @section Comparison of @command{tar} and @command{cpio}
8418 @FIXME{Reorganize the following material}
8420 The @command{cpio} archive formats, like @command{tar}, do have maximum
8421 pathname lengths. The binary and old ASCII formats have a max path
8422 length of 256, and the new ASCII and CRC ASCII formats have a max
8423 path length of 1024. @acronym{GNU} @command{cpio} can read and write archives
8424 with arbitrary pathname lengths, but other @command{cpio} implementations
8425 may crash unexplainedly trying to read them.
8427 @command{tar} handles symbolic links in the form in which it comes in BSD;
8428 @command{cpio} doesn't handle symbolic links in the form in which it comes
8429 in System V prior to SVR4, and some vendors may have added symlinks
8430 to their system without enhancing @command{cpio} to know about them.
8431 Others may have enhanced it in a way other than the way I did it
8432 at Sun, and which was adopted by AT&T (and which is, I think, also
8433 present in the @command{cpio} that Berkeley picked up from AT&T and put
8434 into a later BSD release---I think I gave them my changes).
8436 (SVR4 does some funny stuff with @command{tar}; basically, its @command{cpio}
8437 can handle @command{tar} format input, and write it on output, and it
8438 probably handles symbolic links. They may not have bothered doing
8439 anything to enhance @command{tar} as a result.)
8441 @command{cpio} handles special files; traditional @command{tar} doesn't.
8443 @command{tar} comes with V7, System III, System V, and BSD source;
8444 @command{cpio} comes only with System III, System V, and later BSD
8445 (4.3-tahoe and later).
8447 @command{tar}'s way of handling multiple hard links to a file can handle
8448 file systems that support 32-bit inumbers (e.g., the BSD file system);
8449 @command{cpio}s way requires you to play some games (in its "binary"
8450 format, i-numbers are only 16 bits, and in its "portable ASCII" format,
8451 they're 18 bits---it would have to play games with the "file system ID"
8452 field of the header to make sure that the file system ID/i-number pairs
8453 of different files were always different), and I don't know which
8454 @command{cpio}s, if any, play those games. Those that don't might get
8455 confused and think two files are the same file when they're not, and
8456 make hard links between them.
8458 @command{tar}s way of handling multiple hard links to a file places only
8459 one copy of the link on the tape, but the name attached to that copy
8460 is the @emph{only} one you can use to retrieve the file; @command{cpio}s
8461 way puts one copy for every link, but you can retrieve it using any
8465 What type of check sum (if any) is used, and how is this calculated.
8468 See the attached manual pages for @command{tar} and @command{cpio} format.
8469 @command{tar} uses a checksum which is the sum of all the bytes in the
8470 @command{tar} header for a file; @command{cpio} uses no checksum.
8473 If anyone knows why @command{cpio} was made when @command{tar} was present
8477 It wasn't. @command{cpio} first showed up in PWB/UNIX 1.0; no
8478 generally-available version of UNIX had @command{tar} at the time. I don't
8479 know whether any version that was generally available @emph{within AT&T}
8480 had @command{tar}, or, if so, whether the people within AT&T who did
8481 @command{cpio} knew about it.
8483 On restore, if there is a corruption on a tape @command{tar} will stop at
8484 that point, while @command{cpio} will skip over it and try to restore the
8487 The main difference is just in the command syntax and header format.
8489 @command{tar} is a little more tape-oriented in that everything is blocked
8490 to start on a record boundary.
8493 Is there any differences between the ability to recover crashed
8494 archives between the two of them. (Is there any chance of recovering
8495 crashed archives at all.)
8498 Theoretically it should be easier under @command{tar} since the blocking
8499 lets you find a header with some variation of @samp{dd skip=@var{nn}}.
8500 However, modern @command{cpio}'s and variations have an option to just
8501 search for the next file header after an error with a reasonable chance
8502 of resyncing. However, lots of tape driver software won't allow you to
8503 continue past a media error which should be the only reason for getting
8504 out of sync unless a file changed sizes while you were writing the
8508 If anyone knows why @command{cpio} was made when @command{tar} was present
8509 at the unix scene, please tell me about this too.
8512 Probably because it is more media efficient (by not blocking everything
8513 and using only the space needed for the headers where @command{tar}
8514 always uses 512 bytes per file header) and it knows how to archive
8517 You might want to look at the freely available alternatives. The
8518 major ones are @command{afio}, @GNUTAR{}, and
8519 @command{pax}, each of which have their own extensions with some
8520 backwards compatibility.
8522 Sparse files were @command{tar}red as sparse files (which you can
8523 easily test, because the resulting archive gets smaller, and
8524 @acronym{GNU} @command{cpio} can no longer read it).
8527 @chapter Tapes and Other Archive Media
8530 A few special cases about tape handling warrant more detailed
8531 description. These special cases are discussed below.
8533 Many complexities surround the use of @command{tar} on tape drives. Since
8534 the creation and manipulation of archives located on magnetic tape was
8535 the original purpose of @command{tar}, it contains many features making
8536 such manipulation easier.
8538 Archives are usually written on dismountable media---tape cartridges,
8539 mag tapes, or floppy disks.
8541 The amount of data a tape or disk holds depends not only on its size,
8542 but also on how it is formatted. A 2400 foot long reel of mag tape
8543 holds 40 megabytes of data when formatted at 1600 bits per inch. The
8544 physically smaller EXABYTE tape cartridge holds 2.3 gigabytes.
8546 Magnetic media are re-usable---once the archive on a tape is no longer
8547 needed, the archive can be erased and the tape or disk used over.
8548 Media quality does deteriorate with use, however. Most tapes or disks
8549 should be discarded when they begin to produce data errors. EXABYTE
8550 tape cartridges should be discarded when they generate an @dfn{error
8551 count} (number of non-usable bits) of more than 10k.
8553 Magnetic media are written and erased using magnetic fields, and
8554 should be protected from such fields to avoid damage to stored data.
8555 Sticking a floppy disk to a filing cabinet using a magnet is probably
8559 * Device:: Device selection and switching
8560 * Remote Tape Server::
8561 * Common Problems and Solutions::
8562 * Blocking:: Blocking
8563 * Many:: Many archives on one tape
8564 * Using Multiple Tapes:: Using Multiple Tapes
8565 * label:: Including a Label in the Archive
8567 * Write Protection::
8571 @section Device Selection and Switching
8575 @item -f [@var{hostname}:]@var{file}
8576 @itemx --file=[@var{hostname}:]@var{file}
8577 Use archive file or device @var{file} on @var{hostname}.
8580 This option is used to specify the file name of the archive @command{tar}
8583 If the file name is @samp{-}, @command{tar} reads the archive from standard
8584 input (when listing or extracting), or writes it to standard output
8585 (when creating). If the @samp{-} file name is given when updating an
8586 archive, @command{tar} will read the original archive from its standard
8587 input, and will write the entire new archive to its standard output.
8589 If the file name contains a @samp{:}, it is interpreted as
8590 @samp{hostname:file name}. If the @var{hostname} contains an @dfn{at}
8591 sign (@samp{@@}), it is treated as @samp{user@@hostname:file name}. In
8592 either case, @command{tar} will invoke the command @command{rsh} (or
8593 @command{remsh}) to start up an @command{/usr/libexec/rmt} on the remote
8594 machine. If you give an alternate login name, it will be given to the
8596 Naturally, the remote machine must have an executable
8597 @command{/usr/libexec/rmt}. This program is free software from the
8598 University of California, and a copy of the source code can be found
8599 with the sources for @command{tar}; it's compiled and installed by default.
8600 The exact path to this utility is determined when configuring the package.
8601 It is @file{@var{prefix}/libexec/rmt}, where @var{prefix} stands for
8602 your installation prefix. This location may also be overridden at
8603 runtime by using @option{rmt-command=@var{command}} option (@xref{Option Summary,
8604 ---rmt-command}, for detailed description of this option. @xref{Remote
8605 Tape Server}, for the description of @command{rmt} command).
8607 If this option is not given, but the environment variable @env{TAPE}
8608 is set, its value is used; otherwise, old versions of @command{tar}
8609 used a default archive name (which was picked when @command{tar} was
8610 compiled). The default is normally set up to be the @dfn{first} tape
8611 drive or other transportable I/O medium on the system.
8613 Starting with version 1.11.5, @GNUTAR{} uses
8614 standard input and standard output as the default device, and I will
8615 not try anymore supporting automatic device detection at installation
8616 time. This was failing really in too many cases, it was hopeless.
8617 This is now completely left to the installer to override standard
8618 input and standard output for default device, if this seems
8619 preferable. Further, I think @emph{most} actual usages of
8620 @command{tar} are done with pipes or disks, not really tapes,
8621 cartridges or diskettes.
8623 Some users think that using standard input and output is running
8624 after trouble. This could lead to a nasty surprise on your screen if
8625 you forget to specify an output file name---especially if you are going
8626 through a network or terminal server capable of buffering large amounts
8627 of output. We had so many bug reports in that area of configuring
8628 default tapes automatically, and so many contradicting requests, that
8629 we finally consider the problem to be portably intractable. We could
8630 of course use something like @samp{/dev/tape} as a default, but this
8631 is @emph{also} running after various kind of trouble, going from hung
8632 processes to accidental destruction of real tapes. After having seen
8633 all this mess, using standard input and output as a default really
8634 sounds like the only clean choice left, and a very useful one too.
8636 @GNUTAR{} reads and writes archive in records, I
8637 suspect this is the main reason why block devices are preferred over
8638 character devices. Most probably, block devices are more efficient
8639 too. The installer could also check for @samp{DEFTAPE} in
8640 @file{<sys/mtio.h>}.
8643 @opindex force-local, short description
8645 Archive file is local even if it contains a colon.
8647 @opindex rsh-command
8648 @item --rsh-command=@var{command}
8649 Use remote @var{command} instead of @command{rsh}. This option exists
8650 so that people who use something other than the standard @command{rsh}
8651 (e.g., a Kerberized @command{rsh}) can access a remote device.
8653 When this command is not used, the shell command found when
8654 the @command{tar} program was installed is used instead. This is
8655 the first found of @file{/usr/ucb/rsh}, @file{/usr/bin/remsh},
8656 @file{/usr/bin/rsh}, @file{/usr/bsd/rsh} or @file{/usr/bin/nsh}.
8657 The installer may have overridden this by defining the environment
8658 variable @env{RSH} @emph{at installation time}.
8661 Specify drive and density.
8663 @opindex multi-volume, short description
8665 @itemx --multi-volume
8666 Create/list/extract multi-volume archive.
8668 This option causes @command{tar} to write a @dfn{multi-volume} archive---one
8669 that may be larger than will fit on the medium used to hold it.
8670 @xref{Multi-Volume Archives}.
8672 @opindex tape-length, short description
8674 @itemx --tape-length=@var{num}
8675 Change tape after writing @var{num} x 1024 bytes.
8677 This option might be useful when your tape drivers do not properly
8678 detect end of physical tapes. By being slightly conservative on the
8679 maximum tape length, you might avoid the problem entirely.
8681 @opindex info-script, short description
8682 @opindex new-volume-script, short description
8684 @itemx --info-script=@var{file}
8685 @itemx --new-volume-script=@var{file}
8686 Execute @file{file} at end of each tape. This implies
8687 @option{--multi-volume} (@option{-M}). @xref{info-script}, for a detailed
8688 description of this option.
8691 @node Remote Tape Server
8692 @section The Remote Tape Server
8694 @cindex remote tape drive
8696 In order to access the tape drive on a remote machine, @command{tar}
8697 uses the remote tape server written at the University of California at
8698 Berkeley. The remote tape server must be installed as
8699 @file{@var{prefix}/libexec/rmt} on any machine whose tape drive you
8700 want to use. @command{tar} calls @command{rmt} by running an
8701 @command{rsh} or @command{remsh} to the remote machine, optionally
8702 using a different login name if one is supplied.
8704 A copy of the source for the remote tape server is provided. It is
8705 Copyright @copyright{} 1983 by the Regents of the University of
8706 California, but can be freely distributed. It is compiled and
8707 installed by default.
8709 @cindex absolute file names
8710 Unless you use the @option{--absolute-names} (@option{-P}) option,
8711 @GNUTAR{} will not allow you to create an archive that contains
8712 absolute file names (a file name beginning with @samp{/}.) If you try,
8713 @command{tar} will automatically remove the leading @samp{/} from the
8714 file names it stores in the archive. It will also type a warning
8715 message telling you what it is doing.
8717 When reading an archive that was created with a different
8718 @command{tar} program, @GNUTAR{} automatically
8719 extracts entries in the archive which have absolute file names as if
8720 the file names were not absolute. This is an important feature. A
8721 visitor here once gave a @command{tar} tape to an operator to restore;
8722 the operator used Sun @command{tar} instead of @GNUTAR{},
8723 and the result was that it replaced large portions of
8724 our @file{/bin} and friends with versions from the tape; needless to
8725 say, we were unhappy about having to recover the file system from
8728 For example, if the archive contained a file @file{/usr/bin/computoy},
8729 @GNUTAR{} would extract the file to @file{usr/bin/computoy},
8730 relative to the current directory. If you want to extract the files in
8731 an archive to the same absolute names that they had when the archive
8732 was created, you should do a @samp{cd /} before extracting the files
8733 from the archive, or you should either use the @option{--absolute-names}
8734 option, or use the command @samp{tar -C / @dots{}}.
8736 @cindex Ultrix 3.1 and write failure
8737 Some versions of Unix (Ultrix 3.1 is known to have this problem),
8738 can claim that a short write near the end of a tape succeeded,
8739 when it actually failed. This will result in the -M option not
8740 working correctly. The best workaround at the moment is to use a
8741 significantly larger blocking factor than the default 20.
8743 In order to update an archive, @command{tar} must be able to backspace the
8744 archive in order to reread or rewrite a record that was just read (or
8745 written). This is currently possible only on two kinds of files: normal
8746 disk files (or any other file that can be backspaced with @samp{lseek}),
8747 and industry-standard 9-track magnetic tape (or any other kind of tape
8748 that can be backspaced with the @code{MTIOCTOP} @code{ioctl}.
8750 This means that the @option{--append}, @option{--concatenate}, and
8751 @option{--delete} commands will not work on any other kind of file.
8752 Some media simply cannot be backspaced, which means these commands and
8753 options will never be able to work on them. These non-backspacing
8754 media include pipes and cartridge tape drives.
8756 Some other media can be backspaced, and @command{tar} will work on them
8757 once @command{tar} is modified to do so.
8759 Archives created with the @option{--multi-volume}, @option{--label}, and
8760 @option{--incremental} (@option{-G}) options may not be readable by other version
8761 of @command{tar}. In particular, restoring a file that was split over
8762 a volume boundary will require some careful work with @command{dd}, if
8763 it can be done at all. Other versions of @command{tar} may also create
8764 an empty file whose name is that of the volume header. Some versions
8765 of @command{tar} may create normal files instead of directories archived
8766 with the @option{--incremental} (@option{-G}) option.
8768 @node Common Problems and Solutions
8769 @section Some Common Problems and their Solutions
8776 no such file or directory
8779 errors from @command{tar}:
8780 directory checksum error
8783 errors from media/system:
8794 @dfn{Block} and @dfn{record} terminology is rather confused, and it
8795 is also confusing to the expert reader. On the other hand, readers
8796 who are new to the field have a fresh mind, and they may safely skip
8797 the next two paragraphs, as the remainder of this manual uses those
8798 two terms in a quite consistent way.
8800 John Gilmore, the writer of the public domain @command{tar} from which
8801 @GNUTAR{} was originally derived, wrote (June 1995):
8804 The nomenclature of tape drives comes from IBM, where I believe
8805 they were invented for the IBM 650 or so. On IBM mainframes, what
8806 is recorded on tape are tape blocks. The logical organization of
8807 data is into records. There are various ways of putting records into
8808 blocks, including @code{F} (fixed sized records), @code{V} (variable
8809 sized records), @code{FB} (fixed blocked: fixed size records, @var{n}
8810 to a block), @code{VB} (variable size records, @var{n} to a block),
8811 @code{VSB} (variable spanned blocked: variable sized records that can
8812 occupy more than one block), etc. The @code{JCL} @samp{DD RECFORM=}
8813 parameter specified this to the operating system.
8815 The Unix man page on @command{tar} was totally confused about this.
8816 When I wrote @code{PD TAR}, I used the historically correct terminology
8817 (@command{tar} writes data records, which are grouped into blocks).
8818 It appears that the bogus terminology made it into @acronym{POSIX} (no surprise
8819 here), and now Fran@,{c}ois has migrated that terminology back
8820 into the source code too.
8823 The term @dfn{physical block} means the basic transfer chunk from or
8824 to a device, after which reading or writing may stop without anything
8825 being lost. In this manual, the term @dfn{block} usually refers to
8826 a disk physical block, @emph{assuming} that each disk block is 512
8827 bytes in length. It is true that some disk devices have different
8828 physical blocks, but @command{tar} ignore these differences in its own
8829 format, which is meant to be portable, so a @command{tar} block is always
8830 512 bytes in length, and @dfn{block} always mean a @command{tar} block.
8831 The term @dfn{logical block} often represents the basic chunk of
8832 allocation of many disk blocks as a single entity, which the operating
8833 system treats somewhat atomically; this concept is only barely used
8836 The term @dfn{physical record} is another way to speak of a physical
8837 block, those two terms are somewhat interchangeable. In this manual,
8838 the term @dfn{record} usually refers to a tape physical block,
8839 @emph{assuming} that the @command{tar} archive is kept on magnetic tape.
8840 It is true that archives may be put on disk or used with pipes,
8841 but nevertheless, @command{tar} tries to read and write the archive one
8842 @dfn{record} at a time, whatever the medium in use. One record is made
8843 up of an integral number of blocks, and this operation of putting many
8844 disk blocks into a single tape block is called @dfn{reblocking}, or
8845 more simply, @dfn{blocking}. The term @dfn{logical record} refers to
8846 the logical organization of many characters into something meaningful
8847 to the application. The term @dfn{unit record} describes a small set
8848 of characters which are transmitted whole to or by the application,
8849 and often refers to a line of text. Those two last terms are unrelated
8850 to what we call a @dfn{record} in @GNUTAR{}.
8852 When writing to tapes, @command{tar} writes the contents of the archive
8853 in chunks known as @dfn{records}. To change the default blocking
8854 factor, use the @option{--blocking-factor=@var{512-size}} (@option{-b
8855 @var{512-size}}) option. Each record will then be composed of
8856 @var{512-size} blocks. (Each @command{tar} block is 512 bytes.
8857 @xref{Standard}.) Each file written to the archive uses at least one
8858 full record. As a result, using a larger record size can result in
8859 more wasted space for small files. On the other hand, a larger record
8860 size can often be read and written much more efficiently.
8862 Further complicating the problem is that some tape drives ignore the
8863 blocking entirely. For these, a larger record size can still improve
8864 performance (because the software layers above the tape drive still
8865 honor the blocking), but not as dramatically as on tape drives that
8868 When reading an archive, @command{tar} can usually figure out the
8869 record size on itself. When this is the case, and a non-standard
8870 record size was used when the archive was created, @command{tar} will
8871 print a message about a non-standard blocking factor, and then operate
8872 normally. On some tape devices, however, @command{tar} cannot figure
8873 out the record size itself. On most of those, you can specify a
8874 blocking factor (with @option{--blocking-factor}) larger than the
8875 actual blocking factor, and then use the @option{--read-full-records}
8876 (@option{-B}) option. (If you specify a blocking factor with
8877 @option{--blocking-factor} and don't use the
8878 @option{--read-full-records} option, then @command{tar} will not
8879 attempt to figure out the recording size itself.) On some devices,
8880 you must always specify the record size exactly with
8881 @option{--blocking-factor} when reading, because @command{tar} cannot
8882 figure it out. In any case, use @option{--list} (@option{-t}) before
8883 doing any extractions to see whether @command{tar} is reading the archive
8886 @command{tar} blocks are all fixed size (512 bytes), and its scheme for
8887 putting them into records is to put a whole number of them (one or
8888 more) into each record. @command{tar} records are all the same size;
8889 at the end of the file there's a block containing all zeros, which
8890 is how you tell that the remainder of the last record(s) are garbage.
8892 In a standard @command{tar} file (no options), the block size is 512
8893 and the record size is 10240, for a blocking factor of 20. What the
8894 @option{--blocking-factor} option does is sets the blocking factor,
8895 changing the record size while leaving the block size at 512 bytes.
8896 20 was fine for ancient 800 or 1600 bpi reel-to-reel tape drives;
8897 most tape drives these days prefer much bigger records in order to
8898 stream and not waste tape. When writing tapes for myself, some tend
8899 to use a factor of the order of 2048, say, giving a record size of
8900 around one megabyte.
8902 If you use a blocking factor larger than 20, older @command{tar}
8903 programs might not be able to read the archive, so we recommend this
8904 as a limit to use in practice. @GNUTAR{}, however,
8905 will support arbitrarily large record sizes, limited only by the
8906 amount of virtual memory or the physical characteristics of the tape
8910 * Format Variations:: Format Variations
8911 * Blocking Factor:: The Blocking Factor of an Archive
8914 @node Format Variations
8915 @subsection Format Variations
8916 @cindex Format Parameters
8917 @cindex Format Options
8918 @cindex Options, archive format specifying
8919 @cindex Options, format specifying
8922 Format parameters specify how an archive is written on the archive
8923 media. The best choice of format parameters will vary depending on
8924 the type and number of files being archived, and on the media used to
8927 To specify format parameters when accessing or creating an archive,
8928 you can use the options described in the following sections.
8929 If you do not specify any format parameters, @command{tar} uses
8930 default parameters. You cannot modify a compressed archive.
8931 If you create an archive with the @option{--blocking-factor} option
8932 specified (@pxref{Blocking Factor}), you must specify that
8933 blocking-factor when operating on the archive. @xref{Formats}, for other
8934 examples of format parameter considerations.
8936 @node Blocking Factor
8937 @subsection The Blocking Factor of an Archive
8938 @cindex Blocking Factor
8940 @cindex Number of blocks per record
8941 @cindex Number of bytes per record
8942 @cindex Bytes per record
8943 @cindex Blocks per record
8946 @opindex blocking-factor
8947 The data in an archive is grouped into blocks, which are 512 bytes.
8948 Blocks are read and written in whole number multiples called
8949 @dfn{records}. The number of blocks in a record (ie. the size of a
8950 record in units of 512 bytes) is called the @dfn{blocking factor}.
8951 The @option{--blocking-factor=@var{512-size}} (@option{-b
8952 @var{512-size}}) option specifies the blocking factor of an archive.
8953 The default blocking factor is typically 20 (i.e., 10240 bytes), but
8954 can be specified at installation. To find out the blocking factor of
8955 an existing archive, use @samp{tar --list --file=@var{archive-name}}.
8956 This may not work on some devices.
8958 Records are separated by gaps, which waste space on the archive media.
8959 If you are archiving on magnetic tape, using a larger blocking factor
8960 (and therefore larger records) provides faster throughput and allows you
8961 to fit more data on a tape (because there are fewer gaps). If you are
8962 archiving on cartridge, a very large blocking factor (say 126 or more)
8963 greatly increases performance. A smaller blocking factor, on the other
8964 hand, may be useful when archiving small files, to avoid archiving lots
8965 of nulls as @command{tar} fills out the archive to the end of the record.
8966 In general, the ideal record size depends on the size of the
8967 inter-record gaps on the tape you are using, and the average size of the
8968 files you are archiving. @xref{create}, for information on
8971 @FIXME{Need example of using a cartridge with blocking factor=126 or more.}
8973 Archives with blocking factors larger than 20 cannot be read
8974 by very old versions of @command{tar}, or by some newer versions
8975 of @command{tar} running on old machines with small address spaces.
8976 With @GNUTAR{}, the blocking factor of an archive is limited
8977 only by the maximum record size of the device containing the archive,
8978 or by the amount of available virtual memory.
8980 Also, on some systems, not using adequate blocking factors, as sometimes
8981 imposed by the device drivers, may yield unexpected diagnostics. For
8982 example, this has been reported:
8985 Cannot write to /dev/dlt: Invalid argument
8989 In such cases, it sometimes happen that the @command{tar} bundled by
8990 the system is aware of block size idiosyncrasies, while @GNUTAR{}
8991 requires an explicit specification for the block size,
8992 which it cannot guess. This yields some people to consider
8993 @GNUTAR{} is misbehaving, because by comparison,
8994 @cite{the bundle @command{tar} works OK}. Adding @w{@kbd{-b 256}},
8995 for example, might resolve the problem.
8997 If you use a non-default blocking factor when you create an archive, you
8998 must specify the same blocking factor when you modify that archive. Some
8999 archive devices will also require you to specify the blocking factor when
9000 reading that archive, however this is not typically the case. Usually, you
9001 can use @option{--list} (@option{-t}) without specifying a blocking factor---@command{tar}
9002 reports a non-default record size and then lists the archive members as
9003 it would normally. To extract files from an archive with a non-standard
9004 blocking factor (particularly if you're not sure what the blocking factor
9005 is), you can usually use the @option{--read-full-records} (@option{-B}) option while
9006 specifying a blocking factor larger then the blocking factor of the archive
9007 (ie. @samp{tar --extract --read-full-records --blocking-factor=300}.
9008 @xref{list}, for more information on the @option{--list} (@option{-t})
9009 operation. @xref{Reading}, for a more detailed explanation of that option.
9012 @item --blocking-factor=@var{number}
9013 @itemx -b @var{number}
9014 Specifies the blocking factor of an archive. Can be used with any
9015 operation, but is usually not necessary with @option{--list} (@option{-t}).
9021 @item -b @var{blocks}
9022 @itemx --blocking-factor=@var{blocks}
9023 Set record size to @math{@var{blocks} * 512} bytes.
9025 This option is used to specify a @dfn{blocking factor} for the archive.
9026 When reading or writing the archive, @command{tar}, will do reads and writes
9027 of the archive in records of @math{@var{block}*512} bytes. This is true
9028 even when the archive is compressed. Some devices requires that all
9029 write operations be a multiple of a certain size, and so, @command{tar}
9030 pads the archive out to the next record boundary.
9032 The default blocking factor is set when @command{tar} is compiled, and is
9033 typically 20. Blocking factors larger than 20 cannot be read by very
9034 old versions of @command{tar}, or by some newer versions of @command{tar}
9035 running on old machines with small address spaces.
9037 With a magnetic tape, larger records give faster throughput and fit
9038 more data on a tape (because there are fewer inter-record gaps).
9039 If the archive is in a disk file or a pipe, you may want to specify
9040 a smaller blocking factor, since a large one will result in a large
9041 number of null bytes at the end of the archive.
9043 When writing cartridge or other streaming tapes, a much larger
9044 blocking factor (say 126 or more) will greatly increase performance.
9045 However, you must specify the same blocking factor when reading or
9046 updating the archive.
9048 Apparently, Exabyte drives have a physical block size of 8K bytes.
9049 If we choose our blocksize as a multiple of 8k bytes, then the problem
9050 seems to disappear. Id est, we are using block size of 112 right
9051 now, and we haven't had the problem since we switched@dots{}
9053 With @GNUTAR{} the blocking factor is limited only
9054 by the maximum record size of the device containing the archive, or by
9055 the amount of available virtual memory.
9057 However, deblocking or reblocking is virtually avoided in a special
9058 case which often occurs in practice, but which requires all the
9059 following conditions to be simultaneously true:
9062 the archive is subject to a compression option,
9064 the archive is not handled through standard input or output, nor
9065 redirected nor piped,
9067 the archive is directly handled to a local disk, instead of any special
9070 @option{--blocking-factor} is not explicitly specified on the @command{tar}
9074 If the output goes directly to a local disk, and not through
9075 stdout, then the last write is not extended to a full record size.
9076 Otherwise, reblocking occurs. Here are a few other remarks on this
9082 @command{gzip} will complain about trailing garbage if asked to
9083 uncompress a compressed archive on tape, there is an option to turn
9084 the message off, but it breaks the regularity of simply having to use
9085 @samp{@var{prog} -d} for decompression. It would be nice if gzip was
9086 silently ignoring any number of trailing zeros. I'll ask Jean-loup
9087 Gailly, by sending a copy of this message to him.
9090 @command{compress} does not show this problem, but as Jean-loup pointed
9091 out to Michael, @samp{compress -d} silently adds garbage after
9092 the result of decompression, which tar ignores because it already
9093 recognized its end-of-file indicator. So this bug may be safely
9097 @samp{gzip -d -q} will be silent about the trailing zeros indeed,
9098 but will still return an exit status of 2 which tar reports in turn.
9099 @command{tar} might ignore the exit status returned, but I hate doing
9100 that, as it weakens the protection @command{tar} offers users against
9101 other possible problems at decompression time. If @command{gzip} was
9102 silently skipping trailing zeros @emph{and} also avoiding setting the
9103 exit status in this innocuous case, that would solve this situation.
9106 @command{tar} should become more solid at not stopping to read a pipe at
9107 the first null block encountered. This inelegantly breaks the pipe.
9108 @command{tar} should rather drain the pipe out before exiting itself.
9111 @opindex ignore-zeros, short description
9113 @itemx --ignore-zeros
9114 Ignore blocks of zeros in archive (means EOF).
9116 The @option{--ignore-zeros} (@option{-i}) option causes @command{tar} to ignore blocks
9117 of zeros in the archive. Normally a block of zeros indicates the
9118 end of the archive, but when reading a damaged archive, or one which
9119 was created by concatenating several archives together, this option
9120 allows @command{tar} to read the entire archive. This option is not on
9121 by default because many versions of @command{tar} write garbage after
9124 Note that this option causes @command{tar} to read to the end of the
9125 archive file, which may sometimes avoid problems when multiple files
9126 are stored on a single physical tape.
9128 @opindex read-full-records, short description
9130 @itemx --read-full-records
9131 Reblock as we read (for reading 4.2BSD pipes).
9133 If @option{--read-full-records} is used, @command{tar}
9134 will not panic if an attempt to read a record from the archive does
9135 not return a full record. Instead, @command{tar} will keep reading
9136 until it has obtained a full
9139 This option is turned on by default when @command{tar} is reading
9140 an archive from standard input, or from a remote machine. This is
9141 because on BSD Unix systems, a read of a pipe will return however
9142 much happens to be in the pipe, even if it is less than @command{tar}
9143 requested. If this option was not used, @command{tar} would fail as
9144 soon as it read an incomplete record from the pipe.
9146 This option is also useful with the commands for updating an archive.
9152 @FIXME{Appropriate options should be moved here from elsewhere.}
9154 @cindex blocking factor
9155 @cindex tape blocking
9157 When handling various tapes or cartridges, you have to take care of
9158 selecting a proper blocking, that is, the number of disk blocks you
9159 put together as a single tape block on the tape, without intervening
9160 tape gaps. A @dfn{tape gap} is a small landing area on the tape
9161 with no information on it, used for decelerating the tape to a
9162 full stop, and for later regaining the reading or writing speed.
9163 When the tape driver starts reading a record, the record has to
9164 be read whole without stopping, as a tape gap is needed to stop the
9165 tape motion without loosing information.
9167 @cindex Exabyte blocking
9168 @cindex DAT blocking
9169 Using higher blocking (putting more disk blocks per tape block) will use
9170 the tape more efficiently as there will be less tape gaps. But reading
9171 such tapes may be more difficult for the system, as more memory will be
9172 required to receive at once the whole record. Further, if there is a
9173 reading error on a huge record, this is less likely that the system will
9174 succeed in recovering the information. So, blocking should not be too
9175 low, nor it should be too high. @command{tar} uses by default a blocking of
9176 20 for historical reasons, and it does not really matter when reading or
9177 writing to disk. Current tape technology would easily accommodate higher
9178 blockings. Sun recommends a blocking of 126 for Exabytes and 96 for DATs.
9179 We were told that for some DLT drives, the blocking should be a multiple
9180 of 4Kb, preferably 64Kb (@w{@kbd{-b 128}}) or 256 for decent performance.
9181 Other manufacturers may use different recommendations for the same tapes.
9182 This might also depends of the buffering techniques used inside modern
9183 tape controllers. Some imposes a minimum blocking, or a maximum blocking.
9184 Others request blocking to be some exponent of two.
9186 So, there is no fixed rule for blocking. But blocking at read time
9187 should ideally be the same as blocking used at write time. At one place
9188 I know, with a wide variety of equipment, they found it best to use a
9189 blocking of 32 to guarantee that their tapes are fully interchangeable.
9191 I was also told that, for recycled tapes, prior erasure (by the same
9192 drive unit that will be used to create the archives) sometimes lowers
9193 the error rates observed at rewriting time.
9195 I might also use @option{--number-blocks} instead of
9196 @option{--block-number}, so @option{--block} will then expand to
9197 @option{--blocking-factor} unambiguously.
9200 @section Many Archives on One Tape
9202 @FIXME{Appropriate options should be moved here from elsewhere.}
9204 @findex ntape @r{device}
9205 Most tape devices have two entries in the @file{/dev} directory, or
9206 entries that come in pairs, which differ only in the minor number for
9207 this device. Let's take for example @file{/dev/tape}, which often
9208 points to the only or usual tape device of a given system. There might
9209 be a corresponding @file{/dev/nrtape} or @file{/dev/ntape}. The simpler
9210 name is the @emph{rewinding} version of the device, while the name
9211 having @samp{nr} in it is the @emph{no rewinding} version of the same
9214 A rewinding tape device will bring back the tape to its beginning point
9215 automatically when this device is opened or closed. Since @command{tar}
9216 opens the archive file before using it and closes it afterwards, this
9217 means that a simple:
9220 $ @kbd{tar cf /dev/tape @var{directory}}
9224 will reposition the tape to its beginning both prior and after saving
9225 @var{directory} contents to it, thus erasing prior tape contents and
9226 making it so that any subsequent write operation will destroy what has
9229 @cindex tape positioning
9230 So, a rewinding device is normally meant to hold one and only one file.
9231 If you want to put more than one @command{tar} archive on a given tape, you
9232 will need to avoid using the rewinding version of the tape device. You
9233 will also have to pay special attention to tape positioning. Errors in
9234 positioning may overwrite the valuable data already on your tape. Many
9235 people, burnt by past experiences, will only use rewinding devices and
9236 limit themselves to one file per tape, precisely to avoid the risk of
9237 such errors. Be fully aware that writing at the wrong position on a
9238 tape loses all information past this point and most probably until the
9239 end of the tape, and this destroyed information @emph{cannot} be
9242 To save @var{directory-1} as a first archive at the beginning of a
9243 tape, and leave that tape ready for a second archive, you should use:
9246 $ @kbd{mt -f /dev/nrtape rewind}
9247 $ @kbd{tar cf /dev/nrtape @var{directory-1}}
9251 @dfn{Tape marks} are special magnetic patterns written on the tape
9252 media, which are later recognizable by the reading hardware. These
9253 marks are used after each file, when there are many on a single tape.
9254 An empty file (that is to say, two tape marks in a row) signal the
9255 logical end of the tape, after which no file exist. Usually,
9256 non-rewinding tape device drivers will react to the close request issued
9257 by @command{tar} by first writing two tape marks after your archive, and by
9258 backspacing over one of these. So, if you remove the tape at that time
9259 from the tape drive, it is properly terminated. But if you write
9260 another file at the current position, the second tape mark will be
9261 erased by the new information, leaving only one tape mark between files.
9263 So, you may now save @var{directory-2} as a second archive after the
9264 first on the same tape by issuing the command:
9267 $ @kbd{tar cf /dev/nrtape @var{directory-2}}
9271 and so on for all the archives you want to put on the same tape.
9273 Another usual case is that you do not write all the archives the same
9274 day, and you need to remove and store the tape between two archive
9275 sessions. In general, you must remember how many files are already
9276 saved on your tape. Suppose your tape already has 16 files on it, and
9277 that you are ready to write the 17th. You have to take care of skipping
9278 the first 16 tape marks before saving @var{directory-17}, say, by using
9282 $ @kbd{mt -f /dev/nrtape rewind}
9283 $ @kbd{mt -f /dev/nrtape fsf 16}
9284 $ @kbd{tar cf /dev/nrtape @var{directory-17}}
9287 In all the previous examples, we put aside blocking considerations, but
9288 you should do the proper things for that as well. @xref{Blocking}.
9291 * Tape Positioning:: Tape Positions and Tape Marks
9292 * mt:: The @command{mt} Utility
9295 @node Tape Positioning
9296 @subsection Tape Positions and Tape Marks
9299 Just as archives can store more than one file from the file system,
9300 tapes can store more than one archive file. To keep track of where
9301 archive files (or any other type of file stored on tape) begin and
9302 end, tape archive devices write magnetic @dfn{tape marks} on the
9303 archive media. Tape drives write one tape mark between files,
9304 two at the end of all the file entries.
9306 If you think of data as a series of records "rrrr"'s, and tape marks as
9307 "*"'s, a tape might look like the following:
9310 rrrr*rrrrrr*rrrrr*rr*rrrrr**-------------------------
9313 Tape devices read and write tapes using a read/write @dfn{tape
9314 head}---a physical part of the device which can only access one
9315 point on the tape at a time. When you use @command{tar} to read or
9316 write archive data from a tape device, the device will begin reading
9317 or writing from wherever on the tape the tape head happens to be,
9318 regardless of which archive or what part of the archive the tape
9319 head is on. Before writing an archive, you should make sure that no
9320 data on the tape will be overwritten (unless it is no longer needed).
9321 Before reading an archive, you should make sure the tape head is at
9322 the beginning of the archive you want to read. You can do it manually
9323 via @code{mt} utility (@pxref{mt}). The @code{restore} script does
9324 that automatically (@pxref{Scripted Restoration}).
9326 If you want to add new archive file entries to a tape, you should
9327 advance the tape to the end of the existing file entries, backspace
9328 over the last tape mark, and write the new archive file. If you were
9329 to add two archives to the example above, the tape might look like the
9333 rrrr*rrrrrr*rrrrr*rr*rrrrr*rrr*rrrr**----------------
9337 @subsection The @command{mt} Utility
9340 @FIXME{Is it true that this only works on non-block devices?
9341 should explain the difference, (fixed or variable).}
9342 @xref{Blocking Factor}.
9344 You can use the @command{mt} utility to advance or rewind a tape past a
9345 specified number of archive files on the tape. This will allow you
9346 to move to the beginning of an archive before extracting or reading
9347 it, or to the end of all the archives before writing a new one.
9348 @FIXME{Why isn't there an "advance 'til you find two tape marks
9351 The syntax of the @command{mt} command is:
9354 @kbd{mt [-f @var{tapename}] @var{operation} [@var{number}]}
9357 where @var{tapename} is the name of the tape device, @var{number} is
9358 the number of times an operation is performed (with a default of one),
9359 and @var{operation} is one of the following:
9361 @FIXME{is there any use for record operations?}
9366 Writes @var{number} tape marks at the current position on the tape.
9369 Moves tape position forward @var{number} files.
9372 Moves tape position back @var{number} files.
9375 Rewinds the tape. (Ignores @var{number}).
9379 Rewinds the tape and takes the tape device off-line. (Ignores @var{number}).
9382 Prints status information about the tape unit.
9386 @FIXME{Is there a better way to frob the spacing on the list?}
9388 If you don't specify a @var{tapename}, @command{mt} uses the environment
9389 variable @env{TAPE}; if @env{TAPE} is not set, @command{mt} will use
9390 the default device specified in your @file{sys/mtio.h} file
9391 (@code{DEFTAPE} variable). If this is not defined, the program will
9392 display a descriptive error message and exit with code 1.
9394 @command{mt} returns a 0 exit status when the operation(s) were
9395 successful, 1 if the command was unrecognized, and 2 if an operation
9398 @node Using Multiple Tapes
9399 @section Using Multiple Tapes
9402 Often you might want to write a large archive, one larger than will fit
9403 on the actual tape you are using. In such a case, you can run multiple
9404 @command{tar} commands, but this can be inconvenient, particularly if you
9405 are using options like @option{--exclude=@var{pattern}} or dumping entire file systems.
9406 Therefore, @command{tar} supports multiple tapes automatically.
9408 Use @option{--multi-volume} (@option{-M}) on the command line, and
9409 then @command{tar} will, when it reaches the end of the tape, prompt
9410 for another tape, and continue the archive. Each tape will have an
9411 independent archive, and can be read without needing the other. (As
9412 an exception to this, the file that @command{tar} was archiving when
9413 it ran out of tape will usually be split between the two archives; in
9414 this case you need to extract from the first archive, using
9415 @option{--multi-volume}, and then put in the second tape when
9416 prompted, so @command{tar} can restore both halves of the file.)
9418 @GNUTAR{} multi-volume archives do not use a truly portable format.
9419 You need @GNUTAR{} at both ends to process them properly.
9421 When prompting for a new tape, @command{tar} accepts any of the following
9426 Request @command{tar} to explain possible responses
9428 Request @command{tar} to exit immediately.
9429 @item n @var{file-name}
9430 Request @command{tar} to write the next volume on the file @var{file-name}.
9432 Request @command{tar} to run a subshell. This option can be disabled
9433 by giving @option{--restrict} command line option to @command{tar}.
9435 Request @command{tar} to begin writing the next volume.
9438 (You should only type @samp{y} after you have changed the tape;
9439 otherwise @command{tar} will write over the volume it just finished.)
9441 @cindex End-of-archive info script
9443 @anchor{info-script}
9444 @opindex info-script
9445 @opindex new-volume-script
9446 If you want more elaborate behavior than this, give @command{tar} the
9447 @option{--info-script=@var{script-name}}
9448 (@option{--new-volume-script=@var{script-name}}, @option{-F
9449 @var{script-name}}) option. The file @var{script-name} is expected to
9450 be a program (or shell script) to be run instead of the normal
9451 prompting procedure. It is executed without any command line
9452 arguments. Additional data is passed to it via the following
9453 environment variables:
9456 @vrindex TAR_VERSION, info script environment variable
9458 @GNUTAR{} version number.
9460 @vrindex TAR_ARCHIVE, info script environment variable
9462 The name of the archive @command{tar} is processing.
9464 @vrindex TAR_VOLUME, info script environment variable
9466 Ordinal number of the volume @command{tar} is about to start.
9468 @vrindex TAR_SUBCOMMAND, info script environment variable
9469 @item TAR_SUBCOMMAND
9470 Short option describing the operation @command{tar} is executing
9471 @xref{Operations}, for a complete list of subcommand options.
9473 @vrindex TAR_FORMAT, info script environment variable
9475 Format of the archive being processed. @xref{Formats}, for a complete
9476 list of archive format names.
9479 The info script can instruct @command{tar} to use new archive name,
9480 by writing in to file descriptor 3 (see below for an
9483 If the info script fails, @command{tar} exits; otherwise, it begins
9484 writing the next volume.
9486 The method @command{tar} uses to detect end of tape is not perfect, and
9487 fails on some operating systems or on some devices. You can use the
9488 @option{--tape-length=@var{size}} (@option{-L @var{size}}) option if
9489 @command{tar} can't detect the end of the tape itself. This option
9490 selects @option{--multi-volume} (@option{-M}) automatically. The
9491 @var{size} argument should then be the usable size of the tape in
9492 units of 1024 bytes. But for many devices, and floppy disks in
9493 particular, this option is never required for real, as far as we know.
9495 @cindex Volume number file
9499 The volume number used by @command{tar} in its tape-change prompt
9500 can be changed; if you give the
9501 @option{--volno-file=@var{file-of-number}} option, then
9502 @var{file-of-number} should be an unexisting file to be created, or
9503 else, a file already containing a decimal number. That number will be
9504 used as the volume number of the first volume written. When
9505 @command{tar} is finished, it will rewrite the file with the
9506 now-current volume number. (This does not change the volume number
9507 written on a tape label, as per @ref{label}, it @emph{only} affects
9508 the number used in the prompt.)
9510 If you want @command{tar} to cycle through a series of files or tape
9511 drives, there are three approaches to choose from. First of all, you
9512 can give @command{tar} multiple @option{--file} options. In this case
9513 the specified files will be used, in sequence, as the successive
9514 volumes of the archive. Only when the first one in the sequence needs
9515 to be used again will @command{tar} prompt for a tape change (or run
9516 the info script). Secondly, you can use the @samp{n} response to the
9517 tape-change prompt, and, finally, you can use an info script, that
9518 writes new archive name to file descriptor. The following example
9519 illustrates this approach:
9524 echo Preparing volume $TAR_VOLUME of $TAR_ARCHIVE.
9526 name=`expr $TAR_ARCHIVE : '\(.*\)-.*'`
9527 case $TAR_SUBCOMMAND in
9529 -d|-x|-t) test -r $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME || exit 1
9534 echo $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME >&3
9538 Each volume of a multi-volume archive is an independent @command{tar}
9539 archive, complete in itself. For example, you can list or extract any
9540 volume alone; just don't specify @option{--multi-volume}
9541 (@option{-M}). However, if one file in the archive is split across
9542 volumes, the only way to extract it successfully is with a
9543 multi-volume extract command @option{--extract --multi-volume}
9544 (@option{-xM}) starting on or before the volume where the file begins.
9546 For example, let's presume someone has two tape drives on a system
9547 named @file{/dev/tape0} and @file{/dev/tape1}. For having @GNUTAR{}
9548 to switch to the second drive when it needs to write the
9549 second tape, and then back to the first tape, etc., just do either of:
9552 $ @kbd{tar --create --multi-volume --file=/dev/tape0 --file=/dev/tape1 @var{files}}
9553 $ @kbd{tar cMff /dev/tape0 /dev/tape1 @var{files}}
9557 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
9558 * Tape Files:: Tape Files
9559 * Tarcat:: Concatenate Volumes into a Single Archive
9563 @node Multi-Volume Archives
9564 @subsection Archives Longer than One Tape or Disk
9565 @cindex Multi-volume archives
9568 @opindex multi-volume
9569 To create an archive that is larger than will fit on a single unit of
9570 the media, use the @option{--multi-volume} (@option{-M}) option in conjunction with
9571 the @option{--create} option (@pxref{create}). A @dfn{multi-volume}
9572 archive can be manipulated like any other archive (provided the
9573 @option{--multi-volume} option is specified), but is stored on more
9574 than one tape or disk.
9576 When you specify @option{--multi-volume}, @command{tar} does not report an
9577 error when it comes to the end of an archive volume (when reading), or
9578 the end of the media (when writing). Instead, it prompts you to load
9579 a new storage volume. If the archive is on a magnetic tape, you
9580 should change tapes when you see the prompt; if the archive is on a
9581 floppy disk, you should change disks; etc.
9583 You can read each individual volume of a multi-volume archive as if it
9584 were an archive by itself. For example, to list the contents of one
9585 volume, use @option{--list}, without @option{--multi-volume} specified.
9586 To extract an archive member from one volume (assuming it is described
9587 that volume), use @option{--extract}, again without
9588 @option{--multi-volume}.
9590 If an archive member is split across volumes (ie. its entry begins on
9591 one volume of the media and ends on another), you need to specify
9592 @option{--multi-volume} to extract it successfully. In this case, you
9593 should load the volume where the archive member starts, and use
9594 @samp{tar --extract --multi-volume}---@command{tar} will prompt for later
9595 volumes as it needs them. @xref{extracting archives}, for more
9596 information about extracting archives.
9598 @option{--info-script=@var{script-name}}
9599 (@option{--new-volume-script=@var{script-name}}, @option{-F
9600 @var{script-name}}) (@pxref{info-script}) is like
9601 @option{--multi-volume} (@option{-M}), except that @command{tar} does
9602 not prompt you directly to change media volumes when a volume is
9603 full---instead, @command{tar} runs commands you have stored in
9604 @var{script-name}. For example, this option can be used to eject
9605 cassettes, or to broadcast messages such as @samp{Someone please come
9606 change my tape} when performing unattended backups. When
9607 @var{script-name} is done, @command{tar} will assume that the media
9610 Multi-volume archives can be modified like any other archive. To add
9611 files to a multi-volume archive, you need to only mount the last
9612 volume of the archive media (and new volumes, if needed). For all
9613 other operations, you need to use the entire archive.
9615 If a multi-volume archive was labeled using
9616 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9617 (@pxref{label}) when it was created, @command{tar} will not
9618 automatically label volumes which are added later. To label
9619 subsequent volumes, specify @option{--label=@var{archive-label}} again
9620 in conjunction with the @option{--append}, @option{--update} or
9621 @option{--concatenate} operation.
9623 @cindex Labeling multi-volume archives
9626 @FIXME{There should be a sample program here, including an exit
9627 before end. Is the exit status even checked in tar? :-(}
9630 @item --multi-volume
9632 Creates a multi-volume archive, when used in conjunction with
9633 @option{--create} (@option{-c}). To perform any other operation on a multi-volume
9634 archive, specify @option{--multi-volume} in conjunction with that
9637 @item --info-script=@var{program-file}
9638 @itemx --new-volume-script=@var{program-file}
9639 @itemx -F @var{program-file}
9640 Creates a multi-volume archive via a script. Used in conjunction with
9641 @option{--create} (@option{-c}). @xref{info-script}, dor a detailed discussion.
9644 Beware that there is @emph{no} real standard about the proper way, for
9645 a @command{tar} archive, to span volume boundaries. If you have a
9646 multi-volume created by some vendor's @command{tar}, there is almost
9647 no chance you could read all the volumes with @GNUTAR{}.
9648 The converse is also true: you may not expect
9649 multi-volume archives created by @GNUTAR{} to be
9650 fully recovered by vendor's @command{tar}. Since there is little
9651 chance that, in mixed system configurations, some vendor's
9652 @command{tar} will work on another vendor's machine, and there is a
9653 great chance that @GNUTAR{} will work on most of
9654 them, your best bet is to install @GNUTAR{} on all
9655 machines between which you know exchange of files is possible.
9658 @subsection Tape Files
9661 To give the archive a name which will be recorded in it, use the
9662 @option{--label=@var{volume-label}} (@option{-V @var{volume-label}})
9663 option. This will write a special block identifying
9664 @var{volume-label} as the name of the archive to the front of the
9665 archive which will be displayed when the archive is listed with
9666 @option{--list}. If you are creating a multi-volume archive with
9667 @option{--multi-volume} (@pxref{Using Multiple Tapes}), then the
9668 volume label will have @samp{Volume @var{nnn}} appended to the name
9669 you give, where @var{nnn} is the number of the volume of the archive.
9670 (If you use the @option{--label=@var{volume-label}}) option when
9671 reading an archive, it checks to make sure the label on the tape
9672 matches the one you give. @xref{label}.
9674 When @command{tar} writes an archive to tape, it creates a single
9675 tape file. If multiple archives are written to the same tape, one
9676 after the other, they each get written as separate tape files. When
9677 extracting, it is necessary to position the tape at the right place
9678 before running @command{tar}. To do this, use the @command{mt} command.
9679 For more information on the @command{mt} command and on the organization
9680 of tapes into a sequence of tape files, see @ref{mt}.
9682 People seem to often do:
9685 @kbd{--label="@var{some-prefix} `date +@var{some-format}`"}
9688 or such, for pushing a common date in all volumes or an archive set.
9691 @subsection Concatenate Volumes into a Single Archive
9694 Sometimes it is necessary to convert existing @GNUTAR{} multi-volume
9695 archive to a single @command{tar} archive. Simply concatenating all
9696 volumes into one will not work, since each volume carries an additional
9697 information at the beginning. @GNUTAR{} is shipped with the shell
9698 script @command{tarcat} designed for this purpose.
9700 The script takes a list of files comprising a multi-volume archive
9701 and creates the resulting archive at the standard output. For example:
9704 @kbd{tarcat vol.1 vol.2 vol.3 | tar tf -}
9707 The script implements a simple heuristics to determine the format of
9708 the first volume file and to decide how to process the rest of the
9709 files. However, it makes no attempt to verify whether the files are
9710 given in order or even if they are valid @command{tar} archives.
9711 It uses @command{dd} and does not filter its standard error, so you
9712 will usually see lots of spurious messages.
9714 @FIXME{The script is not installed. Should we install it?}
9717 @section Including a Label in the Archive
9718 @cindex Labeling an archive
9719 @cindex Labels on the archive media
9723 To avoid problems caused by misplaced paper labels on the archive
9724 media, you can include a @dfn{label} entry---an archive member which
9725 contains the name of the archive---in the archive itself. Use the
9726 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9727 option in conjunction with the @option{--create} operation to include
9728 a label entry in the archive as it is being created.
9731 @item --label=@var{archive-label}
9732 @itemx -V @var{archive-label}
9733 Includes an @dfn{archive-label} at the beginning of the archive when
9734 the archive is being created, when used in conjunction with the
9735 @option{--create} operation. Checks to make sure the archive label
9736 matches the one specified (when used in conjunction with any other
9740 If you create an archive using both
9741 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9742 and @option{--multi-volume} (@option{-M}), each volume of the archive
9743 will have an archive label of the form @samp{@var{archive-label}
9744 Volume @var{n}}, where @var{n} is 1 for the first volume, 2 for the
9745 next, and so on. @xref{Using Multiple Tapes}, for information on
9746 creating multiple volume archives.
9748 @cindex Volume label, listing
9749 @cindex Listing volume label
9750 The volume label will be displayed by @option{--list} along with
9751 the file contents. If verbose display is requested, it will also be
9752 explicitely marked as in the example below:
9756 $ @kbd{tar --verbose --list --file=iamanarchive}
9757 V--------- 0 0 0 1992-03-07 12:01 iamalabel--Volume Header--
9758 -rw-r--r-- ringo user 40 1990-05-21 13:30 iamafilename
9763 @anchor{--test-label option}
9764 However, @option{--list} option will cause listing entire
9765 contents of the archive, which may be undesirable (for example, if the
9766 archive is stored on a tape). You can request checking only the volume
9767 by specifying @option{--test-label} option. This option reads only the
9768 first block of an archive, so it can be used with slow storage
9769 devices. For example:
9773 $ @kbd{tar --test-label --file=iamanarchive}
9778 If @option{--test-label} is used with a single command line
9779 argument, @command{tar} compares the volume label with the
9780 argument. It exits with code 0 if the two strings match, and with code
9781 2 otherwise. In this case no output is displayed. For example:
9785 $ @kbd{tar --test-label --file=iamanarchive 'iamalable'}
9787 $ @kbd{tar --test-label --file=iamanarchive 'iamalable' alabel}
9792 If you request any operation, other than @option{--create}, along
9793 with using @option{--label} option, @command{tar} will first check if
9794 the archive label matches the one specified and will refuse to proceed
9795 if it does not. Use this as a safety precaution to avoid accidentally
9796 overwriting existing archives. For example, if you wish to add files
9797 to @file{archive}, presumably labelled with string @samp{My volume},
9802 $ @kbd{tar -rf archive --label 'My volume' .}
9803 tar: Archive not labeled to match `My volume'
9808 in case its label does not match. This will work even if
9809 @file{archive} is not labelled at all.
9811 Similarly, @command{tar} will refuse to list or extract the
9812 archive if its label doesn't match the @var{archive-label}
9813 specified. In those cases, @var{archive-label} argument is interpreted
9814 as a globbing-style pattern which must match the actual magnetic
9815 volume label. @xref{exclude}, for a precise description of how match
9816 is attempted@footnote{Previous versions of @command{tar} used full
9817 regular expression matching, or before that, only exact string
9818 matching, instead of wildcard matchers. We decided for the sake of
9819 simplicity to use a uniform matching device through
9820 @command{tar}.}. If the switch @option{--multi-volume} (@option{-M}) is being used,
9821 the volume label matcher will also suffix @var{archive-label} by
9822 @w{@samp{ Volume [1-9]*}} if the initial match fails, before giving
9823 up. Since the volume numbering is automatically added in labels at
9824 creation time, it sounded logical to equally help the user taking care
9825 of it when the archive is being read.
9827 The @option{--label} was once called @option{--volume}, but is not
9828 available under that name anymore.
9830 You can also use @option{--label} to get a common information on
9831 all tapes of a series. For having this information different in each
9832 series created through a single script used on a regular basis, just
9833 manage to get some date string as part of the label. For example:
9837 $ @kbd{tar cfMV /dev/tape "Daily backup for `date +%Y-%m-%d`"}
9838 $ @kbd{tar --create --file=/dev/tape --multi-volume \
9839 --volume="Daily backup for `date +%Y-%m-%d`"}
9843 Also note that each label has its own date and time, which corresponds
9844 to when @GNUTAR{} initially attempted to write it,
9845 often soon after the operator launches @command{tar} or types the
9846 carriage return telling that the next tape is ready. Comparing date
9847 labels does give an idea of tape throughput only if the delays for
9848 rewinding tapes and the operator switching them were negligible, which
9849 is usually not the case.
9852 @section Verifying Data as It is Stored
9853 @cindex Verifying a write operation
9854 @cindex Double-checking a write operation
9859 @opindex verify, short description
9860 Attempt to verify the archive after writing.
9863 This option causes @command{tar} to verify the archive after writing it.
9864 Each volume is checked after it is written, and any discrepancies
9865 are recorded on the standard error output.
9867 Verification requires that the archive be on a back-space-able medium.
9868 This means pipes, some cartridge tape drives, and some other devices
9871 You can insure the accuracy of an archive by comparing files in the
9872 system with archive members. @command{tar} can compare an archive to the
9873 file system as the archive is being written, to verify a write
9874 operation, or can compare a previously written archive, to insure that
9877 @opindex verify, using with @option{--create}
9878 @opindex create, using with @option{--verify}
9879 To check for discrepancies in an archive immediately after it is
9880 written, use the @option{--verify} (@option{-W}) option in conjunction with
9881 the @option{--create} operation. When this option is
9882 specified, @command{tar} checks archive members against their counterparts
9883 in the file system, and reports discrepancies on the standard error.
9885 To verify an archive, you must be able to read it from before the end
9886 of the last written entry. This option is useful for detecting data
9887 errors on some tapes. Archives written to pipes, some cartridge tape
9888 drives, and some other devices cannot be verified.
9890 One can explicitly compare an already made archive with the file
9891 system by using the @option{--compare} (@option{--diff}, @option{-d})
9892 option, instead of using the more automatic @option{--verify} option.
9895 Note that these two options have a slightly different intent. The
9896 @option{--compare} option checks how identical are the logical contents of some
9897 archive with what is on your disks, while the @option{--verify} option is
9898 really for checking if the physical contents agree and if the recording
9899 media itself is of dependable quality. So, for the @option{--verify}
9900 operation, @command{tar} tries to defeat all in-memory cache pertaining to
9901 the archive, while it lets the speed optimization undisturbed for the
9902 @option{--compare} option. If you nevertheless use @option{--compare} for
9903 media verification, you may have to defeat the in-memory cache yourself,
9904 maybe by opening and reclosing the door latch of your recording unit,
9905 forcing some doubt in your operating system about the fact this is really
9906 the same volume as the one just written or read.
9908 The @option{--verify} option would not be necessary if drivers were indeed
9909 able to detect dependably all write failures. This sometimes require many
9910 magnetic heads, some able to read after the writes occurred. One would
9911 not say that drivers unable to detect all cases are necessarily flawed,
9912 as long as programming is concerned.
9914 The @option{--verify} (@option{-W}) option will not work in
9915 conjunction with the @option{--multi-volume} (@option{-M}) option or
9916 the @option{--append} (@option{-r}), @option{--update} (@option{-u})
9917 and @option{--delete} operations. @xref{Operations}, for more
9918 information on these operations.
9920 Also, since @command{tar} normally strips leading @samp{/} from file
9921 names (@pxref{absolute}), a command like @samp{tar --verify -cf
9922 /tmp/foo.tar /etc} will work as desired only if the working directory is
9923 @file{/}, as @command{tar} uses the archive's relative member names
9924 (e.g., @file{etc/motd}) when verifying the archive.
9926 @node Write Protection
9927 @section Write Protection
9929 Almost all tapes and diskettes, and in a few rare cases, even disks can
9930 be @dfn{write protected}, to protect data on them from being changed.
9931 Once an archive is written, you should write protect the media to prevent
9932 the archive from being accidentally overwritten or deleted. (This will
9933 protect the archive from being changed with a tape or floppy drive---it
9934 will not protect it from magnet fields or other physical hazards).
9936 The write protection device itself is usually an integral part of the
9937 physical media, and can be a two position (write enabled/write
9938 disabled) switch, a notch which can be popped out or covered, a ring
9939 which can be removed from the center of a tape reel, or some other
9945 This appendix lists some important user-visible changes between
9946 version @GNUTAR{} @value{VERSION} and previous versions. An up-to-date
9947 version of this document is available at
9948 @uref{http://www.gnu.org/@/software/@/tar/manual/changes.html,the
9949 @GNUTAR{} documentation page}.
9952 @item Use of globbing patterns when listing and extracting.
9954 Previous versions of GNU tar assumed shell-style globbing when
9955 extracting from or listing an archive. For example:
9958 $ @kbd{tar xf foo.tar '*.c'}
9961 would extract all files whose names end in @samp{.c}. This behavior
9962 was not documented and was incompatible with traditional tar
9963 implementations. Therefore, starting from version 1.15.91, GNU tar
9964 no longer uses globbing by default. For example, the above invocation
9965 is now interpreted as a request to extract from the archive the file
9968 To facilitate transition to the new behavior for those users who got
9969 used to the previous incorrect one, @command{tar} will print a warning
9970 if it finds out that a requested member was not found in the archive
9971 and its name looks like a globbing pattern. For example:
9974 $ @kbd{tar xf foo.tar '*.c'}
9975 tar: Pattern matching characters used in file names. Please,
9976 tar: use --wildcards to enable pattern matching, or --no-wildcards to
9977 tar: suppress this warning.
9978 tar: *.c: Not found in archive
9979 tar: Error exit delayed from previous errors
9982 To treat member names as globbing patterns, use --wildcards option.
9983 If you want to tar to mimic the behavior of versions prior to 1.15.91,
9984 add this option to your @env{TAR_OPTIONS} variable.
9986 @xref{wildcards}, for the detailed discussion of the use of globbing
9987 patterns by @GNUTAR{}.
9989 @item Use of short option @option{-o}.
9991 Earlier versions of @GNUTAR{} understood @option{-o} command line
9992 option as a synonym for @option{--old-archive}.
9994 @GNUTAR{} starting from version 1.13.90 understands this option as
9995 a synonym for @option{--no-same-owner}. This is compatible with
9996 UNIX98 @command{tar} implementations.
9998 However, to facilitate transition, @option{-o} option retains its
9999 old semantics when it is used with one of archive-creation commands.
10000 Users are encouraged to use @option{--format=oldgnu} instead.
10002 It is especially important, since versions of @acronym{GNU} Automake
10003 up to and including 1.8.4 invoke tar with this option to produce
10004 distribution tarballs. @xref{Formats,v7}, for the detailed discussion
10005 of this issue and its implications.
10007 @FIXME{Change the first argument to tar-formats when the new Automake is
10008 out. The proposition to add @anchor{} to the appropriate place of its
10009 docs was accepted by Automake people --Sergey 2006-05-25}.
10010 @xref{Options, tar-v7, Changing Automake's Behavior,
10011 automake, GNU Automake}, for a description on how to use various
10012 archive formats with @command{automake}.
10014 Future versions of @GNUTAR{} will understand @option{-o} only as a
10015 synonym for @option{--no-same-owner}.
10017 @item Use of short option @option{-l}
10019 Earlier versions of @GNUTAR{} understood @option{-l} option as a
10020 synonym for @option{--one-file-system}. Since such usage contradicted
10021 to UNIX98 specification and harmed compatibility with other
10022 implementation, it was declared deprecated in version 1.14. However,
10023 to facilitate transition to its new semantics, it was supported by
10024 versions 1.15 and 1.15.90. The present use of @option{-l} as a short
10025 variant of @option{--check-links} was introduced in version 1.15.91.
10027 @item Use of options @option{--portability} and @option{--old-archive}
10029 These options are deprecated. Please use @option{--format=v7} instead.
10031 @item Use of option @option{--posix}
10033 This option is deprecated. Please use @option{--format=posix} instead.
10036 @node Configuring Help Summary
10037 @appendix Configuring Help Summary
10039 Running @kbd{tar --help} displays the short @command{tar} option
10040 summary (@pxref{help}). This summary is organised by @dfn{groups} of
10041 semantically close options. The options within each group are printed
10042 in the following order: a short option, eventually followed by a list
10043 of corresponding long option names, followed by a short description of
10044 the option. For example, here is an excerpt from the actual @kbd{tar
10048 Main operation mode:
10050 -A, --catenate, --concatenate append tar files to an archive
10051 -c, --create create a new archive
10052 -d, --diff, --compare find differences between archive and
10054 --delete delete from the archive
10057 @vrindex ARGP_HELP_FMT, environment variable
10058 The exact visual representation of the help output is configurable via
10059 @env{ARGP_HELP_FMT} environment variable. The value of this variable
10060 is a comma-separated list of @dfn{format variable} assignments. There
10061 are two kinds of format variables. An @dfn{offset variable} keeps the
10062 offset of some part of help output text from the leftmost column on
10063 the screen. A @dfn{boolean} variable is a flag that toggles some
10064 output feature on or off. Depending on the type of the corresponding
10065 variable, there are two kinds of assignments:
10068 @item Offset assignment
10070 The assignment to an offset variable has the following syntax:
10073 @var{variable}=@var{value}
10077 where @var{variable} is the variable name, and @var{value} is a
10078 numeric value to be assigned to the variable.
10080 @item Boolean assignment
10082 To assign @code{true} value to a variable, simply put this variable name. To
10083 assign @code{false} value, prefix the variable name with @samp{no-}. For
10088 # Assign @code{true} value:
10090 # Assign @code{false} value:
10096 Following variables are declared:
10098 @deftypevr {Help Output} boolean dup-args
10099 If true, arguments for an option are shown with both short and long
10100 options, even when a given option has both forms, for example:
10103 -f ARCHIVE, --file=ARCHIVE use archive file or device ARCHIVE
10106 If false, then if an option has both short and long forms, the
10107 argument is only shown with the long one, for example:
10110 -f, --file=ARCHIVE use archive file or device ARCHIVE
10114 and a message indicating that the argument is applicable to both
10115 forms is printed below the options. This message can be disabled
10116 using @code{dup-args-note} (see below).
10118 The default is false.
10121 @deftypevr {Help Output} boolean dup-args-note
10122 If this variable is true, which is the default, the following notice
10123 is displayed at the end of the help output:
10126 Mandatory or optional arguments to long options are also mandatory or
10127 optional for any corresponding short options.
10130 Setting @code{no-dup-args-note} inhibits this message. Normally, only one of
10131 variables @code{dup-args} or @code{dup-args-note} should be set.
10134 @deftypevr {Help Output} offset short-opt-col
10135 Column in which short options start. Default is 2.
10139 $ @kbd{tar --help|grep ARCHIVE}
10140 -f, --file=ARCHIVE use archive file or device ARCHIVE
10141 $ @kbd{ARGP_HELP_FMT=short-opt-col=6 tar --help|grep ARCHIVE}
10142 -f, --file=ARCHIVE use archive file or device ARCHIVE
10147 @deftypevr {Help Output} offset long-opt-col
10148 Column in which long options start. Default is 6. For example:
10152 $ @kbd{tar --help|grep ARCHIVE}
10153 -f, --file=ARCHIVE use archive file or device ARCHIVE
10154 $ @kbd{ARGP_HELP_FMT=long-opt-col=16 tar --help|grep ARCHIVE}
10155 -f, --file=ARCHIVE use archive file or device ARCHIVE
10160 @deftypevr {Help Output} offset doc-opt-col
10161 Column in which @dfn{doc options} start. A doc option isn't actually
10162 an option, but rather an arbitrary piece of documentation that is
10163 displayed in much the same manner as the options. For example, in
10164 the description of @option{--format} option:
10168 -H, --format=FORMAT create archive of the given format.
10170 FORMAT is one of the following:
10172 gnu GNU tar 1.13.x format
10173 oldgnu GNU format as per tar <= 1.12
10174 pax POSIX 1003.1-2001 (pax) format
10176 ustar POSIX 1003.1-1988 (ustar) format
10177 v7 old V7 tar format
10182 the format names are doc options. Thus, if you set
10183 @kbd{ARGP_HELP_FMT=doc-opt-col=6} the above part of the help output
10184 will look as follows:
10188 -H, --format=FORMAT create archive of the given format.
10190 FORMAT is one of the following:
10192 gnu GNU tar 1.13.x format
10193 oldgnu GNU format as per tar <= 1.12
10194 pax POSIX 1003.1-2001 (pax) format
10196 ustar POSIX 1003.1-1988 (ustar) format
10197 v7 old V7 tar format
10202 @deftypevr {Help Output} offset opt-doc-col
10203 Column in which option description starts. Default is 29.
10207 $ @kbd{tar --help|grep ARCHIVE}
10208 -f, --file=ARCHIVE use archive file or device ARCHIVE
10209 $ @kbd{ARGP_HELP_FMT=opt-doc-col=19 tar --help|grep ARCHIVE}
10210 -f, --file=ARCHIVE use archive file or device ARCHIVE
10211 $ @kbd{ARGP_HELP_FMT=opt-doc-col=9 tar --help|grep ARCHIVE}
10213 use archive file or device ARCHIVE
10218 Notice, that the description starts on a separate line if
10219 @code{opt-doc-col} value is too small.
10222 @deftypevr {Help Output} offset header-col
10223 Column in which @dfn{group headers} are printed. A group header is a
10224 descriptive text preceding an option group. For example, in the
10228 Main operation mode:
10230 -A, --catenate, --concatenate append tar files to
10232 -c, --create create a new archive
10235 @samp{Main operation mode:} is the group header.
10237 The default value is 1.
10240 @deftypevr {Help Output} offset usage-indent
10241 Indentation of wrapped usage lines. Affects @option{--usage}
10242 output. Default is 12.
10245 @deftypevr {Help Output} offset rmargin
10246 Right margin of the text output. Used for wrapping.
10251 @include genfile.texi
10253 @node Tar Internals
10254 @appendix Tar Internals
10255 @include intern.texi
10257 @node Free Software Needs Free Documentation
10258 @appendix Free Software Needs Free Documentation
10259 @include freemanuals.texi
10261 @node Copying This Manual
10262 @appendix Copying This Manual
10265 * GNU Free Documentation License:: License for copying this manual
10270 @node Index of Command Line Options
10271 @appendix Index of Command Line Options
10273 This appendix contains an index of all @GNUTAR{} long command line
10274 options. The options are listed without the preceeding double-dash.
10277 @item Make sure @emph{all} options are indexed.
10278 @item Provide an index of short options
10292 @c Local variables:
10293 @c texinfo-column-for-description: 32