1 \input texinfo @c -*-texinfo-*-
2 @comment %**start of header
5 @settitle GNU tar @value{VERSION}
14 @c 1. Pay attention to @FIXME{}s and @UNREVISED{}s
15 @c 2. Before creating final variant:
16 @c 2.1. Run `make check-options' to make sure all options are properly
18 @c 2.2. Run `make master-menu' (see comment before the master menu).
20 @include rendition.texi
26 @c Put everything in one index (arbitrarily chosen to be the concept index).
35 This manual is for @acronym{GNU} @command{tar} (version
36 @value{VERSION}, @value{UPDATED}), which creates and extracts files
39 Copyright @copyright{} 1992, 1994, 1995, 1996, 1997, 1999, 2000, 2001,
40 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
43 Permission is granted to copy, distribute and/or modify this document
44 under the terms of the GNU Free Documentation License, Version 1.3 or
45 any later version published by the Free Software Foundation; with no
46 Invariant Sections, with the Front-Cover Texts being ``A GNU Manual'',
47 and with the Back-Cover Texts as in (a) below. A copy of the license
48 is included in the section entitled ``GNU Free Documentation
51 (a) The FSF's Back-Cover Text is: ``You have the freedom to
52 copy and modify this GNU manual. Buying copies from the FSF
53 supports it in developing GNU and promoting software freedom.''
57 @dircategory Archiving
59 * Tar: (tar). Making tape (or disk) archives.
62 @dircategory Individual utilities
64 * tar: (tar)tar invocation. Invoking @GNUTAR{}.
67 @shorttitlepage @acronym{GNU} @command{tar}
70 @title @acronym{GNU} tar: an archiver tool
71 @subtitle @value{RENDITION} @value{VERSION}, @value{UPDATED}
72 @author John Gilmore, Jay Fenlason et al.
75 @vskip 0pt plus 1filll
81 @top @acronym{GNU} tar: an archiver tool
86 @cindex archiving files
88 The first part of this master menu lists the major nodes in this Info
89 document. The rest of the menu lists all the lower level nodes.
92 @c The master menu goes here.
94 @c NOTE: To update it from within Emacs, make sure mastermenu.el is
95 @c loaded and run texinfo-master-menu.
96 @c To update it from the command line, run
107 * Date input formats::
114 * Configuring Help Summary::
115 * Fixing Snapshot Files::
118 * Free Software Needs Free Documentation::
119 * GNU Free Documentation License::
120 * Index of Command Line Options::
124 --- The Detailed Node Listing ---
128 * Book Contents:: What this Book Contains
129 * Definitions:: Some Definitions
130 * What tar Does:: What @command{tar} Does
131 * Naming tar Archives:: How @command{tar} Archives are Named
132 * Authors:: @GNUTAR{} Authors
133 * Reports:: Reporting bugs or suggestions
135 Tutorial Introduction to @command{tar}
138 * stylistic conventions::
139 * basic tar options:: Basic @command{tar} Operations and Options
140 * frequent operations::
141 * Two Frequent Options::
142 * create:: How to Create Archives
143 * list:: How to List Archives
144 * extract:: How to Extract Members from an Archive
147 Two Frequently Used Options
153 How to Create Archives
155 * prepare for examples::
156 * Creating the archive::
165 How to Extract Members from an Archive
167 * extracting archives::
170 * extracting untrusted archives::
176 * using tar options::
186 The Three Option Styles
188 * Long Options:: Long Option Style
189 * Short Options:: Short Option Style
190 * Old Options:: Old Option Style
191 * Mixing:: Mixing Option Styles
193 All @command{tar} Options
195 * Operation Summary::
197 * Short Option Summary::
209 Advanced @GNUTAR{} Operations
218 How to Add Files to Existing Archives: @option{--append}
220 * appending files:: Appending Files to an Archive
227 Options Used by @option{--create}
229 * override:: Overriding File Metadata.
230 * Ignore Failed Read::
232 Options Used by @option{--extract}
234 * Reading:: Options to Help Read Archives
235 * Writing:: Changing How @command{tar} Writes Files
236 * Scarce:: Coping with Scarce Resources
238 Options to Help Read Archives
240 * read full records::
243 Changing How @command{tar} Writes Files
245 * Dealing with Old Files::
246 * Overwrite Old Files::
251 * Data Modification Times::
252 * Setting Access Permissions::
253 * Directory Modification Times and Permissions::
254 * Writing to Standard Output::
255 * Writing to an External Program::
258 Coping with Scarce Resources
263 Performing Backups and Restoring Files
265 * Full Dumps:: Using @command{tar} to Perform Full Dumps
266 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
267 * Backup Levels:: Levels of Backups
268 * Backup Parameters:: Setting Parameters for Backups and Restoration
269 * Scripted Backups:: Using the Backup Scripts
270 * Scripted Restoration:: Using the Restore Script
272 Setting Parameters for Backups and Restoration
274 * General-Purpose Variables::
275 * Magnetic Tape Control::
277 * backup-specs example:: An Example Text of @file{Backup-specs}
279 Choosing Files and Names for @command{tar}
281 * file:: Choosing the Archive's Name
282 * Selecting Archive Members::
283 * files:: Reading Names from a File
284 * exclude:: Excluding Some Files
285 * wildcards:: Wildcards Patterns and Matching
286 * quoting styles:: Ways of Quoting Special Characters in Names
287 * transform:: Modifying File and Member Names
288 * after:: Operating Only on New Files
289 * recurse:: Descending into Directories
290 * one:: Crossing File System Boundaries
292 Reading Names from a File
298 * problems with exclude::
300 Wildcards Patterns and Matching
302 * controlling pattern-matching::
304 Crossing File System Boundaries
306 * directory:: Changing Directory
307 * absolute:: Absolute File Names
311 * General date syntax:: Common rules.
312 * Calendar date items:: 19 Dec 1994.
313 * Time of day items:: 9:20pm.
314 * Time zone items:: @sc{est}, @sc{pdt}, @sc{gmt}.
315 * Day of week items:: Monday and others.
316 * Relative items in date strings:: next tuesday, 2 years ago.
317 * Pure numbers in date strings:: 19931219, 1440.
318 * Seconds since the Epoch:: @@1078100502.
319 * Specifying time zone rules:: TZ="America/New_York", TZ="UTC0".
320 * Authors of get_date:: Bellovin, Eggert, Salz, Berets, et al.
322 Controlling the Archive Format
324 * Compression:: Using Less Space through Compression
325 * Attributes:: Handling File Attributes
326 * Portability:: Making @command{tar} Archives More Portable
327 * cpio:: Comparison of @command{tar} and @command{cpio}
329 Using Less Space through Compression
331 * gzip:: Creating and Reading Compressed Archives
332 * sparse:: Archiving Sparse Files
334 Creating and Reading Compressed Archives
336 * lbzip2:: Using lbzip2 with @GNUTAR{}.
338 Making @command{tar} Archives More Portable
340 * Portable Names:: Portable Names
341 * dereference:: Symbolic Links
342 * hard links:: Hard Links
343 * old:: Old V7 Archives
344 * ustar:: Ustar Archives
345 * gnu:: GNU and old GNU format archives.
346 * posix:: @acronym{POSIX} archives
347 * Checksumming:: Checksumming Problems
348 * Large or Negative Values:: Large files, negative time stamps, etc.
349 * Other Tars:: How to Extract GNU-Specific Data Using
350 Other @command{tar} Implementations
352 @GNUTAR{} and @acronym{POSIX} @command{tar}
354 * PAX keywords:: Controlling Extended Header Keywords.
356 How to Extract GNU-Specific Data Using Other @command{tar} Implementations
358 * Split Recovery:: Members Split Between Volumes
359 * Sparse Recovery:: Sparse Members
361 Tapes and Other Archive Media
363 * Device:: Device selection and switching
364 * Remote Tape Server::
365 * Common Problems and Solutions::
366 * Blocking:: Blocking
367 * Many:: Many archives on one tape
368 * Using Multiple Tapes:: Using Multiple Tapes
369 * label:: Including a Label in the Archive
375 * Format Variations:: Format Variations
376 * Blocking Factor:: The Blocking Factor of an Archive
378 Many Archives on One Tape
380 * Tape Positioning:: Tape Positions and Tape Marks
381 * mt:: The @command{mt} Utility
385 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
386 * Tape Files:: Tape Files
387 * Tarcat:: Concatenate Volumes into a Single Archive
392 * Standard:: Basic Tar Format
393 * Extensions:: @acronym{GNU} Extensions to the Archive Format
394 * Sparse Formats:: Storing Sparse Files
401 * PAX 0:: PAX Format, Versions 0.0 and 0.1
402 * PAX 1:: PAX Format, Version 1.0
406 * Generate Mode:: File Generation Mode.
407 * Status Mode:: File Status Mode.
408 * Exec Mode:: Synchronous Execution mode.
412 * GNU Free Documentation License:: License for copying this manual
418 @chapter Introduction
421 and manipulates @dfn{archives} which are actually collections of
422 many other files; the program provides users with an organized and
423 systematic method for controlling a large amount of data.
424 The name ``tar'' originally came from the phrase ``Tape ARchive'', but
425 archives need not (and these days, typically do not) reside on tapes.
428 * Book Contents:: What this Book Contains
429 * Definitions:: Some Definitions
430 * What tar Does:: What @command{tar} Does
431 * Naming tar Archives:: How @command{tar} Archives are Named
432 * Authors:: @GNUTAR{} Authors
433 * Reports:: Reporting bugs or suggestions
437 @section What this Book Contains
439 The first part of this chapter introduces you to various terms that will
440 recur throughout the book. It also tells you who has worked on @GNUTAR{}
441 and its documentation, and where you should send bug reports
444 The second chapter is a tutorial (@pxref{Tutorial}) which provides a
445 gentle introduction for people who are new to using @command{tar}. It is
446 meant to be self-contained, not requiring any reading from subsequent
447 chapters to make sense. It moves from topic to topic in a logical,
448 progressive order, building on information already explained.
450 Although the tutorial is paced and structured to allow beginners to
451 learn how to use @command{tar}, it is not intended solely for beginners.
452 The tutorial explains how to use the three most frequently used
453 operations (@samp{create}, @samp{list}, and @samp{extract}) as well as
454 two frequently used options (@samp{file} and @samp{verbose}). The other
455 chapters do not refer to the tutorial frequently; however, if a section
456 discusses something which is a complex variant of a basic concept, there
457 may be a cross-reference to that basic concept. (The entire book,
458 including the tutorial, assumes that the reader understands some basic
459 concepts of using a Unix-type operating system; @pxref{Tutorial}.)
461 The third chapter presents the remaining five operations, and
462 information about using @command{tar} options and option syntax.
464 The other chapters are meant to be used as a reference. Each chapter
465 presents everything that needs to be said about a specific topic.
467 One of the chapters (@pxref{Date input formats}) exists in its
468 entirety in other @acronym{GNU} manuals, and is mostly self-contained.
469 In addition, one section of this manual (@pxref{Standard}) contains a
470 big quote which is taken directly from @command{tar} sources.
472 In general, we give both long and short (abbreviated) option names
473 at least once in each section where the relevant option is covered, so
474 that novice readers will become familiar with both styles. (A few
475 options have no short versions, and the relevant sections will
479 @section Some Definitions
483 The @command{tar} program is used to create and manipulate @command{tar}
484 archives. An @dfn{archive} is a single file which contains the contents
485 of many files, while still identifying the names of the files, their
486 owner(s), and so forth. (In addition, archives record access
487 permissions, user and group, size in bytes, and data modification time.
488 Some archives also record the file names in each archived directory, as
489 well as other file and directory information.) You can use @command{tar}
490 to @dfn{create} a new archive in a specified directory.
493 @cindex archive member
496 The files inside an archive are called @dfn{members}. Within this
497 manual, we use the term @dfn{file} to refer only to files accessible in
498 the normal ways (by @command{ls}, @command{cat}, and so forth), and the term
499 @dfn{member} to refer only to the members of an archive. Similarly, a
500 @dfn{file name} is the name of a file, as it resides in the file system,
501 and a @dfn{member name} is the name of an archive member within the
506 The term @dfn{extraction} refers to the process of copying an archive
507 member (or multiple members) into a file in the file system. Extracting
508 all the members of an archive is often called @dfn{extracting the
509 archive}. The term @dfn{unpack} can also be used to refer to the
510 extraction of many or all the members of an archive. Extracting an
511 archive does not destroy the archive's structure, just as creating an
512 archive does not destroy the copies of the files that exist outside of
513 the archive. You may also @dfn{list} the members in a given archive
514 (this is often thought of as ``printing'' them to the standard output,
515 or the command line), or @dfn{append} members to a pre-existing archive.
516 All of these operations can be performed using @command{tar}.
519 @section What @command{tar} Does
522 The @command{tar} program provides the ability to create @command{tar}
523 archives, as well as various other kinds of manipulation. For example,
524 you can use @command{tar} on previously created archives to extract files,
525 to store additional files, or to update or list files which were already
528 Initially, @command{tar} archives were used to store files conveniently on
529 magnetic tape. The name @command{tar} comes from this use; it stands for
530 @code{t}ape @code{ar}chiver. Despite the utility's name, @command{tar} can
531 direct its output to available devices, files, or other programs (using
532 pipes). @command{tar} may even access remote devices or files (as archives).
534 You can use @command{tar} archives in many ways. We want to stress a few
535 of them: storage, backup, and transportation.
537 @FIXME{the following table entries need a bit of work.}
540 Often, @command{tar} archives are used to store related files for
541 convenient file transfer over a network. For example, the
542 @acronym{GNU} Project distributes its software bundled into
543 @command{tar} archives, so that all the files relating to a particular
544 program (or set of related programs) can be transferred as a single
547 A magnetic tape can store several files in sequence. However, the tape
548 has no names for these files; it only knows their relative position on
549 the tape. One way to store several files on one tape and retain their
550 names is by creating a @command{tar} archive. Even when the basic transfer
551 mechanism can keep track of names, as FTP can, the nuisance of handling
552 multiple files, directories, and multiple links makes @command{tar}
555 Archive files are also used for long-term storage. You can think of
556 this as transportation from the present into the future. (It is a
557 science-fiction idiom that you can move through time as well as in
558 space; the idea here is that @command{tar} can be used to move archives in
559 all dimensions, even time!)
562 Because the archive created by @command{tar} is capable of preserving
563 file information and directory structure, @command{tar} is commonly
564 used for performing full and incremental backups of disks. A backup
565 puts a collection of files (possibly pertaining to many users and
566 projects) together on a disk or a tape. This guards against
567 accidental destruction of the information in those files.
568 @GNUTAR{} has special features that allow it to be
569 used to make incremental and full dumps of all the files in a
573 You can create an archive on one system, transfer it to another system,
574 and extract the contents there. This allows you to transport a group of
575 files from one system to another.
578 @node Naming tar Archives
579 @section How @command{tar} Archives are Named
581 Conventionally, @command{tar} archives are given names ending with
582 @samp{.tar}. This is not necessary for @command{tar} to operate properly,
583 but this manual follows that convention in order to accustom readers to
584 it and to make examples more clear.
589 Often, people refer to @command{tar} archives as ``@command{tar} files,'' and
590 archive members as ``files'' or ``entries''. For people familiar with
591 the operation of @command{tar}, this causes no difficulty. However, in
592 this manual, we consistently refer to ``archives'' and ``archive
593 members'' to make learning to use @command{tar} easier for novice users.
596 @section @GNUTAR{} Authors
598 @GNUTAR{} was originally written by John Gilmore,
599 and modified by many people. The @acronym{GNU} enhancements were
600 written by Jay Fenlason, then Joy Kendall, and the whole package has
601 been further maintained by Thomas Bushnell, n/BSG, Fran@,{c}ois
602 Pinard, Paul Eggert, and finally Sergey Poznyakoff with the help of
603 numerous and kind users.
605 We wish to stress that @command{tar} is a collective work, and owes much to
606 all those people who reported problems, offered solutions and other
607 insights, or shared their thoughts and suggestions. An impressive, yet
608 partial list of those contributors can be found in the @file{THANKS}
609 file from the @GNUTAR{} distribution.
611 @FIXME{i want all of these names mentioned, Absolutely. BUT, i'm not
612 sure i want to spell out the history in this detail, at least not for
613 the printed book. i'm just not sure it needs to be said this way.
614 i'll think about it.}
616 @FIXME{History is more important, and surely more interesting, than
617 actual names. Quoting names without history would be meaningless. FP}
619 Jay Fenlason put together a draft of a @GNUTAR{}
620 manual, borrowing notes from the original man page from John Gilmore.
621 This was withdrawn in version 1.11. Thomas Bushnell, n/BSG and Amy
622 Gorin worked on a tutorial and manual for @GNUTAR{}.
623 Fran@,{c}ois Pinard put version 1.11.8 of the manual together by
624 taking information from all these sources and merging them. Melissa
625 Weisshaus finally edited and redesigned the book to create version
626 1.12. The book for versions from 1.14 up to @value{VERSION} were edited
627 by the current maintainer, Sergey Poznyakoff.
629 For version 1.12, Daniel Hagerty contributed a great deal of technical
630 consulting. In particular, he is the primary author of @ref{Backups}.
632 In July, 2003 @GNUTAR{} was put on CVS at savannah.gnu.org
633 (see @url{http://savannah.gnu.org/projects/tar}), and
634 active development and maintenance work has started
635 again. Currently @GNUTAR{} is being maintained by Paul Eggert, Sergey
636 Poznyakoff and Jeff Bailey.
638 Support for @acronym{POSIX} archives was added by Sergey Poznyakoff.
641 @section Reporting bugs or suggestions
644 @cindex reporting bugs
645 If you find problems or have suggestions about this program or manual,
646 please report them to @file{bug-tar@@gnu.org}.
648 When reporting a bug, please be sure to include as much detail as
649 possible, in order to reproduce it. @FIXME{Be more specific, I'd
650 like to make this node as detailed as 'Bug reporting' node in Emacs
654 @chapter Tutorial Introduction to @command{tar}
656 This chapter guides you through some basic examples of three @command{tar}
657 operations: @option{--create}, @option{--list}, and @option{--extract}. If
658 you already know how to use some other version of @command{tar}, then you
659 may not need to read this chapter. This chapter omits most complicated
660 details about how @command{tar} works.
664 * stylistic conventions::
665 * basic tar options:: Basic @command{tar} Operations and Options
666 * frequent operations::
667 * Two Frequent Options::
668 * create:: How to Create Archives
669 * list:: How to List Archives
670 * extract:: How to Extract Members from an Archive
675 @section Assumptions this Tutorial Makes
677 This chapter is paced to allow beginners to learn about @command{tar}
678 slowly. At the same time, we will try to cover all the basic aspects of
679 these three operations. In order to accomplish both of these tasks, we
680 have made certain assumptions about your knowledge before reading this
681 manual, and the hardware you will be using:
685 Before you start to work through this tutorial, you should understand
686 what the terms ``archive'' and ``archive member'' mean
687 (@pxref{Definitions}). In addition, you should understand something
688 about how Unix-type operating systems work, and you should know how to
689 use some basic utilities. For example, you should know how to create,
690 list, copy, rename, edit, and delete files and directories; how to
691 change between directories; and how to figure out where you are in the
692 file system. You should have some basic understanding of directory
693 structure and how files are named according to which directory they are
694 in. You should understand concepts such as standard output and standard
695 input, what various definitions of the term @samp{argument} mean, and the
696 differences between relative and absolute file names. @FIXME{and what
700 This manual assumes that you are working from your own home directory
701 (unless we state otherwise). In this tutorial, you will create a
702 directory to practice @command{tar} commands in. When we show file names,
703 we will assume that those names are relative to your home directory.
704 For example, my home directory is @file{/home/fsf/melissa}. All of
705 my examples are in a subdirectory of the directory named by that file
706 name; the subdirectory is called @file{practice}.
709 In general, we show examples of archives which exist on (or can be
710 written to, or worked with from) a directory on a hard disk. In most
711 cases, you could write those archives to, or work with them on any other
712 device, such as a tape drive. However, some of the later examples in
713 the tutorial and next chapter will not work on tape drives.
714 Additionally, working with tapes is much more complicated than working
715 with hard disks. For these reasons, the tutorial does not cover working
716 with tape drives. @xref{Media}, for complete information on using
717 @command{tar} archives with tape drives.
719 @FIXME{this is a cop out. need to add some simple tape drive info.}
722 @node stylistic conventions
723 @section Stylistic Conventions
725 In the examples, @samp{$} represents a typical shell prompt. It
726 precedes lines you should type; to make this more clear, those lines are
727 shown in @kbd{this font}, as opposed to lines which represent the
728 computer's response; those lines are shown in @code{this font}, or
729 sometimes @samp{like this}.
731 @c When we have lines which are too long to be
732 @c displayed in any other way, we will show them like this:
734 @node basic tar options
735 @section Basic @command{tar} Operations and Options
737 @command{tar} can take a wide variety of arguments which specify and define
738 the actions it will have on the particular set of files or the archive.
739 The main types of arguments to @command{tar} fall into one of two classes:
740 operations, and options.
742 Some arguments fall into a class called @dfn{operations}; exactly one of
743 these is both allowed and required for any instance of using @command{tar};
744 you may @emph{not} specify more than one. People sometimes speak of
745 @dfn{operating modes}. You are in a particular operating mode when you
746 have specified the operation which specifies it; there are eight
747 operations in total, and thus there are eight operating modes.
749 The other arguments fall into the class known as @dfn{options}. You are
750 not required to specify any options, and you are allowed to specify more
751 than one at a time (depending on the way you are using @command{tar} at
752 that time). Some options are used so frequently, and are so useful for
753 helping you type commands more carefully that they are effectively
754 ``required''. We will discuss them in this chapter.
756 You can write most of the @command{tar} operations and options in any
757 of three forms: long (mnemonic) form, short form, and old style. Some
758 of the operations and options have no short or ``old'' forms; however,
759 the operations and options which we will cover in this tutorial have
760 corresponding abbreviations. We will indicate those abbreviations
761 appropriately to get you used to seeing them. Note, that the ``old
762 style'' option forms exist in @GNUTAR{} for compatibility with Unix
763 @command{tar}. In this book we present a full discussion of this way
764 of writing options and operations (@pxref{Old Options}), and we discuss
765 the other two styles of writing options (@xref{Long Options}, and
766 @pxref{Short Options}).
768 In the examples and in the text of this tutorial, we usually use the
769 long forms of operations and options; but the ``short'' forms produce
770 the same result and can make typing long @command{tar} commands easier.
771 For example, instead of typing
774 @kbd{tar --create --verbose --file=afiles.tar apple angst aspic}
780 @kbd{tar -c -v -f afiles.tar apple angst aspic}
786 @kbd{tar -cvf afiles.tar apple angst aspic}
790 For more information on option syntax, see @ref{Advanced tar}. In
791 discussions in the text, when we name an option by its long form, we
792 also give the corresponding short option in parentheses.
794 The term, ``option'', can be confusing at times, since ``operations''
795 are often lumped in with the actual, @emph{optional} ``options'' in certain
796 general class statements. For example, we just talked about ``short and
797 long forms of options and operations''. However, experienced @command{tar}
798 users often refer to these by shorthand terms such as, ``short and long
799 options''. This term assumes that the ``operations'' are included, also.
800 Context will help you determine which definition of ``options'' to use.
802 Similarly, the term ``command'' can be confusing, as it is often used in
803 two different ways. People sometimes refer to @command{tar} ``commands''.
804 A @command{tar} @dfn{command} is the entire command line of user input
805 which tells @command{tar} what to do --- including the operation, options,
806 and any arguments (file names, pipes, other commands, etc.). However,
807 you will also sometimes hear the term ``the @command{tar} command''. When
808 the word ``command'' is used specifically like this, a person is usually
809 referring to the @command{tar} @emph{operation}, not the whole line.
810 Again, use context to figure out which of the meanings the speaker
813 @node frequent operations
814 @section The Three Most Frequently Used Operations
816 Here are the three most frequently used operations (both short and long
817 forms), as well as a brief description of their meanings. The rest of
818 this chapter will cover how to use these operations in detail. We will
819 present the rest of the operations in the next chapter.
824 Create a new @command{tar} archive.
827 List the contents of an archive.
830 Extract one or more members from an archive.
833 @node Two Frequent Options
834 @section Two Frequently Used Options
836 To understand how to run @command{tar} in the three operating modes listed
837 previously, you also need to understand how to use two of the options to
838 @command{tar}: @option{--file} (which takes an archive file as an argument)
839 and @option{--verbose}. (You are usually not @emph{required} to specify
840 either of these options when you run @command{tar}, but they can be very
841 useful in making things more clear and helping you avoid errors.)
850 @unnumberedsubsec The @option{--file} Option
853 @xopindex{file, tutorial}
854 @item --file=@var{archive-name}
855 @itemx -f @var{archive-name}
856 Specify the name of an archive file.
859 You can specify an argument for the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option whenever you
860 use @command{tar}; this option determines the name of the archive file
861 that @command{tar} will work on.
864 If you don't specify this argument, then @command{tar} will examine
865 the environment variable @env{TAPE}. If it is set, its value will be
866 used as the archive name. Otherwise, @command{tar} will use the
867 default archive, determined at compile time. Usually it is
868 standard output or some physical tape drive attached to your machine
869 (you can verify what the default is by running @kbd{tar
870 --show-defaults}, @pxref{defaults}). If there is no tape drive
871 attached, or the default is not meaningful, then @command{tar} will
872 print an error message. The error message might look roughly like one
876 tar: can't open /dev/rmt8 : No such device or address
877 tar: can't open /dev/rsmt0 : I/O error
881 To avoid confusion, we recommend that you always specify an archive file
882 name by using @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) when writing your @command{tar} commands.
883 For more information on using the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option, see
886 @node verbose tutorial
887 @unnumberedsubsec The @option{--verbose} Option
890 @xopindex{verbose, introduced}
893 Show the files being worked on as @command{tar} is running.
896 @option{--verbose} (@option{-v}) shows details about the results of running
897 @command{tar}. This can be especially useful when the results might not be
898 obvious. For example, if you want to see the progress of @command{tar} as
899 it writes files into the archive, you can use the @option{--verbose}
900 option. In the beginning, you may find it useful to use
901 @option{--verbose} at all times; when you are more accustomed to
902 @command{tar}, you will likely want to use it at certain times but not at
903 others. We will use @option{--verbose} at times to help make something
904 clear, and we will give many examples both using and not using
905 @option{--verbose} to show the differences.
907 Each instance of @option{--verbose} on the command line increases the
908 verbosity level by one, so if you need more details on the output,
911 When reading archives (@option{--list}, @option{--extract},
912 @option{--diff}), @command{tar} by default prints only the names of
913 the members being extracted. Using @option{--verbose} will show a full,
914 @command{ls} style member listing.
916 In contrast, when writing archives (@option{--create}, @option{--append},
917 @option{--update}), @command{tar} does not print file names by
918 default. So, a single @option{--verbose} option shows the file names
919 being added to the archive, while two @option{--verbose} options
920 enable the full listing.
922 For example, to create an archive in verbose mode:
925 $ @kbd{tar -cvf afiles.tar apple angst aspic}
932 Creating the same archive with the verbosity level 2 could give:
935 $ @kbd{tar -cvvf afiles.tar apple angst aspic}
936 -rw-r--r-- gray/staff 62373 2006-06-09 12:06 apple
937 -rw-r--r-- gray/staff 11481 2006-06-09 12:06 angst
938 -rw-r--r-- gray/staff 23152 2006-06-09 12:06 aspic
942 This works equally well using short or long forms of options. Using
943 long forms, you would simply write out the mnemonic form of the option
947 $ @kbd{tar --create --verbose --verbose @dots{}}
951 Note that you must double the hyphens properly each time.
953 Later in the tutorial, we will give examples using @w{@option{--verbose
956 @anchor{verbose member listing}
957 The full output consists of six fields:
960 @item File type and permissions in symbolic form.
961 These are displayed in the same format as the first column of
962 @command{ls -l} output (@pxref{What information is listed,
963 format=verbose, Verbose listing, fileutils, GNU file utilities}).
965 @item Owner name and group separated by a slash character.
966 If these data are not available (for example, when listing a @samp{v7} format
967 archive), numeric @acronym{ID} values are printed instead.
969 @item Size of the file, in bytes.
971 @item File modification date in ISO 8601 format.
973 @item File modification time.
976 If the name contains any special characters (white space, newlines,
977 etc.) these are displayed in an unambiguous form using so called
978 @dfn{quoting style}. For the detailed discussion of available styles
979 and on how to use them, see @ref{quoting styles}.
981 Depending on the file type, the name can be followed by some
982 additional information, described in the following table:
985 @item -> @var{link-name}
986 The file or archive member is a @dfn{symbolic link} and
987 @var{link-name} is the name of file it links to.
989 @item link to @var{link-name}
990 The file or archive member is a @dfn{hard link} and @var{link-name} is
991 the name of file it links to.
994 The archive member is an old GNU format long link. You will normally
998 The archive member is an old GNU format long name. You will normally
1001 @item --Volume Header--
1002 The archive member is a GNU @dfn{volume header} (@pxref{Tape Files}).
1004 @item --Continued at byte @var{n}--
1005 Encountered only at the beginning of a multi-volume archive
1006 (@pxref{Using Multiple Tapes}). This archive member is a continuation
1007 from the previous volume. The number @var{n} gives the offset where
1008 the original file was split.
1010 @item unknown file type @var{c}
1011 An archive member of unknown type. @var{c} is the type character from
1012 the archive header. If you encounter such a message, it means that
1013 either your archive contains proprietary member types @GNUTAR{} is not
1014 able to handle, or the archive is corrupted.
1019 For example, here is an archive listing containing most of the special
1020 suffixes explained above:
1024 V--------- 0/0 1536 2006-06-09 13:07 MyVolume--Volume Header--
1025 -rw-r--r-- gray/staff 456783 2006-06-09 12:06 aspic--Continued at
1027 -rw-r--r-- gray/staff 62373 2006-06-09 12:06 apple
1028 lrwxrwxrwx gray/staff 0 2006-06-09 13:01 angst -> apple
1029 -rw-r--r-- gray/staff 35793 2006-06-09 12:06 blues
1030 hrw-r--r-- gray/staff 0 2006-06-09 12:06 music link to blues
1038 @unnumberedsubsec Getting Help: Using the @option{--help} Option
1044 The @option{--help} option to @command{tar} prints out a very brief list of
1045 all operations and option available for the current version of
1046 @command{tar} available on your system.
1050 @section How to Create Archives
1053 @cindex Creation of the archive
1054 @cindex Archive, creation of
1055 One of the basic operations of @command{tar} is @option{--create} (@option{-c}), which
1056 you use to create a @command{tar} archive. We will explain
1057 @option{--create} first because, in order to learn about the other
1058 operations, you will find it useful to have an archive available to
1061 To make this easier, in this section you will first create a directory
1062 containing three files. Then, we will show you how to create an
1063 @emph{archive} (inside the new directory). Both the directory, and
1064 the archive are specifically for you to practice on. The rest of this
1065 chapter and the next chapter will show many examples using this
1066 directory and the files you will create: some of those files may be
1067 other directories and other archives.
1069 The three files you will archive in this example are called
1070 @file{blues}, @file{folk}, and @file{jazz}. The archive is called
1071 @file{collection.tar}.
1073 This section will proceed slowly, detailing how to use @option{--create}
1074 in @code{verbose} mode, and showing examples using both short and long
1075 forms. In the rest of the tutorial, and in the examples in the next
1076 chapter, we will proceed at a slightly quicker pace. This section
1077 moves more slowly to allow beginning users to understand how
1078 @command{tar} works.
1081 * prepare for examples::
1082 * Creating the archive::
1088 @node prepare for examples
1089 @subsection Preparing a Practice Directory for Examples
1091 To follow along with this and future examples, create a new directory
1092 called @file{practice} containing files called @file{blues}, @file{folk}
1093 and @file{jazz}. The files can contain any information you like:
1094 ideally, they should contain information which relates to their names,
1095 and be of different lengths. Our examples assume that @file{practice}
1096 is a subdirectory of your home directory.
1098 Now @command{cd} to the directory named @file{practice}; @file{practice}
1099 is now your @dfn{working directory}. (@emph{Please note}: Although
1100 the full file name of this directory is
1101 @file{/@var{homedir}/practice}, in our examples we will refer to
1102 this directory as @file{practice}; the @var{homedir} is presumed.)
1104 In general, you should check that the files to be archived exist where
1105 you think they do (in the working directory) by running @command{ls}.
1106 Because you just created the directory and the files and have changed to
1107 that directory, you probably don't need to do that this time.
1109 It is very important to make sure there isn't already a file in the
1110 working directory with the archive name you intend to use (in this case,
1111 @samp{collection.tar}), or that you don't care about its contents.
1112 Whenever you use @samp{create}, @command{tar} will erase the current
1113 contents of the file named by @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) if it exists. @command{tar}
1114 will not tell you if you are about to overwrite an archive unless you
1115 specify an option which does this (@pxref{backup}, for the
1116 information on how to do so). To add files to an existing archive,
1117 you need to use a different option, such as @option{--append} (@option{-r}); see
1118 @ref{append} for information on how to do this.
1120 @node Creating the archive
1121 @subsection Creating the Archive
1123 @xopindex{create, introduced}
1124 To place the files @file{blues}, @file{folk}, and @file{jazz} into an
1125 archive named @file{collection.tar}, use the following command:
1128 $ @kbd{tar --create --file=collection.tar blues folk jazz}
1131 The order of the arguments is not very important, @emph{when using long
1132 option forms}. You could also say:
1135 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1139 However, you can see that this order is harder to understand; this is
1140 why we will list the arguments in the order that makes the commands
1141 easiest to understand (and we encourage you to do the same when you use
1142 @command{tar}, to avoid errors).
1144 Note that the sequence
1145 @option{--file=@-collection.tar} is considered to be @emph{one} argument.
1146 If you substituted any other string of characters for
1147 @kbd{collection.tar}, then that string would become the name of the
1148 archive file you create.
1150 The order of the options becomes more important when you begin to use
1151 short forms. With short forms, if you type commands in the wrong order
1152 (even if you type them correctly in all other ways), you may end up with
1153 results you don't expect. For this reason, it is a good idea to get
1154 into the habit of typing options in the order that makes inherent sense.
1155 @xref{short create}, for more information on this.
1157 In this example, you type the command as shown above: @option{--create}
1158 is the operation which creates the new archive
1159 (@file{collection.tar}), and @option{--file} is the option which lets
1160 you give it the name you chose. The files, @file{blues}, @file{folk},
1161 and @file{jazz}, are now members of the archive, @file{collection.tar}
1162 (they are @dfn{file name arguments} to the @option{--create} operation.
1163 @xref{Choosing}, for the detailed discussion on these.) Now that they are
1164 in the archive, they are called @emph{archive members}, not files.
1165 (@pxref{Definitions,members}).
1167 When you create an archive, you @emph{must} specify which files you
1168 want placed in the archive. If you do not specify any archive
1169 members, @GNUTAR{} will complain.
1171 If you now list the contents of the working directory (@command{ls}), you will
1172 find the archive file listed as well as the files you saw previously:
1175 blues folk jazz collection.tar
1179 Creating the archive @samp{collection.tar} did not destroy the copies of
1180 the files in the directory.
1182 Keep in mind that if you don't indicate an operation, @command{tar} will not
1183 run and will prompt you for one. If you don't name any files, @command{tar}
1184 will complain. You must have write access to the working directory,
1185 or else you will not be able to create an archive in that directory.
1187 @emph{Caution}: Do not attempt to use @option{--create} (@option{-c}) to add files to
1188 an existing archive; it will delete the archive and write a new one.
1189 Use @option{--append} (@option{-r}) instead. @xref{append}.
1191 @node create verbose
1192 @subsection Running @option{--create} with @option{--verbose}
1194 @xopindex{create, using with @option{--verbose}}
1195 @xopindex{verbose, using with @option{--create}}
1196 If you include the @option{--verbose} (@option{-v}) option on the command line,
1197 @command{tar} will list the files it is acting on as it is working. In
1198 verbose mode, the @code{create} example above would appear as:
1201 $ @kbd{tar --create --verbose --file=collection.tar blues folk jazz}
1207 This example is just like the example we showed which did not use
1208 @option{--verbose}, except that @command{tar} generated the remaining
1210 lines (note the different font styles).
1216 In the rest of the examples in this chapter, we will frequently use
1217 @code{verbose} mode so we can show actions or @command{tar} responses that
1218 you would otherwise not see, and which are important for you to
1222 @subsection Short Forms with @samp{create}
1224 As we said before, the @option{--create} (@option{-c}) operation is one of the most
1225 basic uses of @command{tar}, and you will use it countless times.
1226 Eventually, you will probably want to use abbreviated (or ``short'')
1227 forms of options. A full discussion of the three different forms that
1228 options can take appears in @ref{Styles}; for now, here is what the
1229 previous example (including the @option{--verbose} (@option{-v}) option) looks like
1230 using short option forms:
1233 $ @kbd{tar -cvf collection.tar blues folk jazz}
1240 As you can see, the system responds the same no matter whether you use
1241 long or short option forms.
1243 @FIXME{i don't like how this is worded:} One difference between using
1244 short and long option forms is that, although the exact placement of
1245 arguments following options is no more specific when using short forms,
1246 it is easier to become confused and make a mistake when using short
1247 forms. For example, suppose you attempted the above example in the
1251 $ @kbd{tar -cfv collection.tar blues folk jazz}
1255 In this case, @command{tar} will make an archive file called @file{v},
1256 containing the files @file{blues}, @file{folk}, and @file{jazz}, because
1257 the @samp{v} is the closest ``file name'' to the @option{-f} option, and
1258 is thus taken to be the chosen archive file name. @command{tar} will try
1259 to add a file called @file{collection.tar} to the @file{v} archive file;
1260 if the file @file{collection.tar} did not already exist, @command{tar} will
1261 report an error indicating that this file does not exist. If the file
1262 @file{collection.tar} does already exist (e.g., from a previous command
1263 you may have run), then @command{tar} will add this file to the archive.
1264 Because the @option{-v} option did not get registered, @command{tar} will not
1265 run under @samp{verbose} mode, and will not report its progress.
1267 The end result is that you may be quite confused about what happened,
1268 and possibly overwrite a file. To illustrate this further, we will show
1269 you how an example we showed previously would look using short forms.
1274 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1278 is confusing as it is. When shown using short forms, however, it
1279 becomes much more so:
1282 $ @kbd{tar blues -c folk -f collection.tar jazz}
1286 It would be very easy to put the wrong string of characters
1287 immediately following the @option{-f}, but doing that could sacrifice
1290 For this reason, we recommend that you pay very careful attention to
1291 the order of options and placement of file and archive names,
1292 especially when using short option forms. Not having the option name
1293 written out mnemonically can affect how well you remember which option
1294 does what, and therefore where different names have to be placed.
1297 @subsection Archiving Directories
1299 @cindex Archiving Directories
1300 @cindex Directories, Archiving
1301 You can archive a directory by specifying its directory name as a
1302 file name argument to @command{tar}. The files in the directory will be
1303 archived relative to the working directory, and the directory will be
1304 re-created along with its contents when the archive is extracted.
1306 To archive a directory, first move to its superior directory. If you
1307 have followed the previous instructions in this tutorial, you should
1316 This will put you into the directory which contains @file{practice},
1317 i.e., your home directory. Once in the superior directory, you can
1318 specify the subdirectory, @file{practice}, as a file name argument. To
1319 store @file{practice} in the new archive file @file{music.tar}, type:
1322 $ @kbd{tar --create --verbose --file=music.tar practice}
1326 @command{tar} should output:
1333 practice/collection.tar
1336 Note that the archive thus created is not in the subdirectory
1337 @file{practice}, but rather in the current working directory---the
1338 directory from which @command{tar} was invoked. Before trying to archive a
1339 directory from its superior directory, you should make sure you have
1340 write access to the superior directory itself, not only the directory
1341 you are trying archive with @command{tar}. For example, you will probably
1342 not be able to store your home directory in an archive by invoking
1343 @command{tar} from the root directory; @xref{absolute}. (Note
1344 also that @file{collection.tar}, the original archive file, has itself
1345 been archived. @command{tar} will accept any file as a file to be
1346 archived, regardless of its content. When @file{music.tar} is
1347 extracted, the archive file @file{collection.tar} will be re-written
1348 into the file system).
1350 If you give @command{tar} a command such as
1353 $ @kbd{tar --create --file=foo.tar .}
1357 @command{tar} will report @samp{tar: ./foo.tar is the archive; not
1358 dumped}. This happens because @command{tar} creates the archive
1359 @file{foo.tar} in the current directory before putting any files into
1360 it. Then, when @command{tar} attempts to add all the files in the
1361 directory @file{.} to the archive, it notices that the file
1362 @file{./foo.tar} is the same as the archive @file{foo.tar}, and skips
1363 it. (It makes no sense to put an archive into itself.) @GNUTAR{}
1364 will continue in this case, and create the archive
1365 normally, except for the exclusion of that one file. (@emph{Please
1366 note:} Other implementations of @command{tar} may not be so clever;
1367 they will enter an infinite loop when this happens, so you should not
1368 depend on this behavior unless you are certain you are running
1369 @GNUTAR{}. In general, it is wise to always place the archive outside
1370 of the directory being dumped.)
1373 @section How to List Archives
1376 Frequently, you will find yourself wanting to determine exactly what a
1377 particular archive contains. You can use the @option{--list}
1378 (@option{-t}) operation to get the member names as they currently
1379 appear in the archive, as well as various attributes of the files at
1380 the time they were archived. For example, you can examine the archive
1381 @file{collection.tar} that you created in the last section with the
1385 $ @kbd{tar --list --file=collection.tar}
1389 The output of @command{tar} would then be:
1398 The archive @file{bfiles.tar} would list as follows:
1407 Be sure to use a @option{--file=@var{archive-name}} (@option{-f
1408 @var{archive-name}}) option just as with @option{--create}
1409 (@option{-c}) to specify the name of the archive.
1411 @xopindex{list, using with @option{--verbose}}
1412 @xopindex{verbose, using with @option{--list}}
1413 If you use the @option{--verbose} (@option{-v}) option with
1414 @option{--list}, then @command{tar} will print out a listing
1415 reminiscent of @w{@samp{ls -l}}, showing owner, file size, and so
1416 forth. This output is described in detail in @ref{verbose member listing}.
1418 If you had used @option{--verbose} (@option{-v}) mode, the example
1419 above would look like:
1422 $ @kbd{tar --list --verbose --file=collection.tar folk}
1423 -rw-r--r-- myself user 62 1990-05-23 10:55 folk
1426 @cindex listing member and file names
1427 @anchor{listing member and file names}
1428 It is important to notice that the output of @kbd{tar --list
1429 --verbose} does not necessarily match that produced by @kbd{tar
1430 --create --verbose} while creating the archive. It is because
1431 @GNUTAR{}, unless told explicitly not to do so, removes some directory
1432 prefixes from file names before storing them in the archive
1433 (@xref{absolute}, for more information). In other
1434 words, in verbose mode @GNUTAR{} shows @dfn{file names} when creating
1435 an archive and @dfn{member names} when listing it. Consider this
1440 $ @kbd{tar --create --verbose --file archive /etc/mail}
1441 tar: Removing leading `/' from member names
1443 /etc/mail/sendmail.cf
1445 $ @kbd{tar --test --file archive}
1447 etc/mail/sendmail.cf
1452 @opindex show-stored-names
1453 This default behavior can sometimes be inconvenient. You can force
1454 @GNUTAR{} show member names when creating archive by supplying
1455 @option{--show-stored-names} option.
1458 @item --show-stored-names
1459 Print member (as opposed to @emph{file}) names when creating the archive.
1462 @cindex File name arguments, using @option{--list} with
1463 @xopindex{list, using with file name arguments}
1464 You can specify one or more individual member names as arguments when
1465 using @samp{list}. In this case, @command{tar} will only list the
1466 names of members you identify. For example, @w{@kbd{tar --list
1467 --file=afiles.tar apple}} would only print @file{apple}.
1469 Because @command{tar} preserves file names, these must be specified as
1470 they appear in the archive (i.e., relative to the directory from which
1471 the archive was created). Therefore, it is essential when specifying
1472 member names to @command{tar} that you give the exact member names.
1473 For example, @w{@kbd{tar --list --file=bfiles.tar birds}} would produce an
1474 error message something like @samp{tar: birds: Not found in archive},
1475 because there is no member named @file{birds}, only one named
1476 @file{./birds}. While the names @file{birds} and @file{./birds} name
1477 the same file, @emph{member} names by default are compared verbatim.
1479 However, @w{@kbd{tar --list --file=bfiles.tar baboon}} would respond
1480 with @file{baboon}, because this exact member name is in the archive file
1481 @file{bfiles.tar}. If you are not sure of the exact file name,
1482 use @dfn{globbing patterns}, for example:
1485 $ @kbd{tar --list --file=bfiles.tar --wildcards '*b*'}
1489 will list all members whose name contains @samp{b}. @xref{wildcards},
1490 for a detailed discussion of globbing patterns and related
1491 @command{tar} command line options.
1498 @unnumberedsubsec Listing the Contents of a Stored Directory
1500 To get information about the contents of an archived directory,
1501 use the directory name as a file name argument in conjunction with
1502 @option{--list} (@option{-t}). To find out file attributes, include the
1503 @option{--verbose} (@option{-v}) option.
1505 For example, to find out about files in the directory @file{practice}, in
1506 the archive file @file{music.tar}, type:
1509 $ @kbd{tar --list --verbose --file=music.tar practice}
1512 @command{tar} responds:
1515 drwxrwxrwx myself user 0 1990-05-31 21:49 practice/
1516 -rw-r--r-- myself user 42 1990-05-21 13:29 practice/blues
1517 -rw-r--r-- myself user 62 1990-05-23 10:55 practice/folk
1518 -rw-r--r-- myself user 40 1990-05-21 13:30 practice/jazz
1519 -rw-r--r-- myself user 10240 1990-05-31 21:49 practice/collection.tar
1522 When you use a directory name as a file name argument, @command{tar} acts on
1523 all the files (including sub-directories) in that directory.
1526 @section How to Extract Members from an Archive
1528 @cindex Retrieving files from an archive
1529 @cindex Resurrecting files from an archive
1532 Creating an archive is only half the job---there is no point in storing
1533 files in an archive if you can't retrieve them. The act of retrieving
1534 members from an archive so they can be used and manipulated as
1535 unarchived files again is called @dfn{extraction}. To extract files
1536 from an archive, use the @option{--extract} (@option{--get} or
1537 @option{-x}) operation. As with @option{--create}, specify the name
1538 of the archive with @option{--file} (@option{-f}) option. Extracting
1539 an archive does not modify the archive in any way; you can extract it
1540 multiple times if you want or need to.
1542 Using @option{--extract}, you can extract an entire archive, or specific
1543 files. The files can be directories containing other files, or not. As
1544 with @option{--create} (@option{-c}) and @option{--list} (@option{-t}), you may use the short or the
1545 long form of the operation without affecting the performance.
1548 * extracting archives::
1549 * extracting files::
1551 * extracting untrusted archives::
1552 * failing commands::
1555 @node extracting archives
1556 @subsection Extracting an Entire Archive
1558 To extract an entire archive, specify the archive file name only, with
1559 no individual file names as arguments. For example,
1562 $ @kbd{tar -xvf collection.tar}
1569 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
1570 -rw-r--r-- me user 21 1996-09-23 16:44 blues
1571 -rw-r--r-- me user 20 1996-09-23 16:44 folk
1574 @node extracting files
1575 @subsection Extracting Specific Files
1577 To extract specific archive members, give their exact member names as
1578 arguments, as printed by @option{--list} (@option{-t}). If you had
1579 mistakenly deleted one of the files you had placed in the archive
1580 @file{collection.tar} earlier (say, @file{blues}), you can extract it
1581 from the archive without changing the archive's structure. Its
1582 contents will be identical to the original file @file{blues} that you
1585 First, make sure you are in the @file{practice} directory, and list the
1586 files in the directory. Now, delete the file, @samp{blues}, and list
1587 the files in the directory again.
1589 You can now extract the member @file{blues} from the archive file
1590 @file{collection.tar} like this:
1593 $ @kbd{tar --extract --file=collection.tar blues}
1597 If you list the files in the directory again, you will see that the file
1598 @file{blues} has been restored, with its original permissions, data
1599 modification times, and owner.@footnote{This is only accidentally
1600 true, but not in general. Whereas modification times are always
1601 restored, in most cases, one has to be root for restoring the owner,
1602 and use a special option for restoring permissions. Here, it just
1603 happens that the restoring user is also the owner of the archived
1604 members, and that the current @code{umask} is compatible with original
1605 permissions.} (These parameters will be identical to those which
1606 the file had when you originally placed it in the archive; any changes
1607 you may have made before deleting the file from the file system,
1608 however, will @emph{not} have been made to the archive member.) The
1609 archive file, @samp{collection.tar}, is the same as it was before you
1610 extracted @samp{blues}. You can confirm this by running @command{tar} with
1611 @option{--list} (@option{-t}).
1613 Remember that as with other operations, specifying the exact member
1614 name is important. @w{@kbd{tar --extract --file=bfiles.tar birds}}
1615 will fail, because there is no member named @file{birds}. To extract
1616 the member named @file{./birds}, you must specify @w{@kbd{tar
1617 --extract --file=bfiles.tar ./birds}}. If you don't remember the
1618 exact member names, use @option{--list} (@option{-t}) option
1619 (@pxref{list}). You can also extract those members that match a
1620 specific @dfn{globbing pattern}. For example, to extract from
1621 @file{bfiles.tar} all files that begin with @samp{b}, no matter their
1622 directory prefix, you could type:
1625 $ @kbd{tar -x -f bfiles.tar --wildcards --no-anchored 'b*'}
1629 Here, @option{--wildcards} instructs @command{tar} to treat
1630 command line arguments as globbing patterns and @option{--no-anchored}
1631 informs it that the patterns apply to member names after any @samp{/}
1632 delimiter. The use of globbing patterns is discussed in detail in
1635 You can extract a file to standard output by combining the above options
1636 with the @option{--to-stdout} (@option{-O}) option (@pxref{Writing to Standard
1639 If you give the @option{--verbose} option, then @option{--extract}
1640 will print the names of the archive members as it extracts them.
1643 @subsection Extracting Files that are Directories
1645 Extracting directories which are members of an archive is similar to
1646 extracting other files. The main difference to be aware of is that if
1647 the extracted directory has the same name as any directory already in
1648 the working directory, then files in the extracted directory will be
1649 placed into the directory of the same name. Likewise, if there are
1650 files in the pre-existing directory with the same names as the members
1651 which you extract, the files from the extracted archive will replace
1652 the files already in the working directory (and possible
1653 subdirectories). This will happen regardless of whether or not the
1654 files in the working directory were more recent than those extracted
1655 (there exist, however, special options that alter this behavior
1658 However, if a file was stored with a directory name as part of its file
1659 name, and that directory does not exist under the working directory when
1660 the file is extracted, @command{tar} will create the directory.
1662 We can demonstrate how to use @option{--extract} to extract a directory
1663 file with an example. Change to the @file{practice} directory if you
1664 weren't there, and remove the files @file{folk} and @file{jazz}. Then,
1665 go back to the parent directory and extract the archive
1666 @file{music.tar}. You may either extract the entire archive, or you may
1667 extract only the files you just deleted. To extract the entire archive,
1668 don't give any file names as arguments after the archive name
1669 @file{music.tar}. To extract only the files you deleted, use the
1673 $ @kbd{tar -xvf music.tar practice/folk practice/jazz}
1679 If you were to specify two @option{--verbose} (@option{-v}) options, @command{tar}
1680 would have displayed more detail about the extracted files, as shown
1681 in the example below:
1684 $ @kbd{tar -xvvf music.tar practice/folk practice/jazz}
1685 -rw-r--r-- me user 28 1996-10-18 16:31 practice/jazz
1686 -rw-r--r-- me user 20 1996-09-23 16:44 practice/folk
1690 Because you created the directory with @file{practice} as part of the
1691 file names of each of the files by archiving the @file{practice}
1692 directory as @file{practice}, you must give @file{practice} as part
1693 of the file names when you extract those files from the archive.
1695 @node extracting untrusted archives
1696 @subsection Extracting Archives from Untrusted Sources
1698 Extracting files from archives can overwrite files that already exist.
1699 If you receive an archive from an untrusted source, you should make a
1700 new directory and extract into that directory, so that you don't have
1701 to worry about the extraction overwriting one of your existing files.
1702 For example, if @file{untrusted.tar} came from somewhere else on the
1703 Internet, and you don't necessarily trust its contents, you can
1704 extract it as follows:
1707 $ @kbd{mkdir newdir}
1709 $ @kbd{tar -xvf ../untrusted.tar}
1712 It is also a good practice to examine contents of the archive
1713 before extracting it, using @option{--list} (@option{-t}) option, possibly combined
1714 with @option{--verbose} (@option{-v}).
1716 @node failing commands
1717 @subsection Commands That Will Fail
1719 Here are some sample commands you might try which will not work, and why
1722 If you try to use this command,
1725 $ @kbd{tar -xvf music.tar folk jazz}
1729 you will get the following response:
1732 tar: folk: Not found in archive
1733 tar: jazz: Not found in archive
1737 This is because these files were not originally @emph{in} the parent
1738 directory @file{..}, where the archive is located; they were in the
1739 @file{practice} directory, and their file names reflect this:
1742 $ @kbd{tar -tvf music.tar}
1748 @FIXME{make sure the above works when going through the examples in
1752 Likewise, if you try to use this command,
1755 $ @kbd{tar -tvf music.tar folk jazz}
1759 you would get a similar response. Members with those names are not in the
1760 archive. You must use the correct member names, or wildcards, in order
1761 to extract the files from the archive.
1763 If you have forgotten the correct names of the files in the archive,
1764 use @w{@kbd{tar --list --verbose}} to list them correctly.
1766 @FIXME{more examples, here? hag thinks it's a good idea.}
1769 @section Going Further Ahead in this Manual
1772 @FIXME{need to write up a node here about the things that are going to
1773 be in the rest of the manual.}
1775 @node tar invocation
1776 @chapter Invoking @GNUTAR{}
1778 This chapter is about how one invokes the @GNUTAR{}
1779 command, from the command synopsis (@pxref{Synopsis}). There are
1780 numerous options, and many styles for writing them. One mandatory
1781 option specifies the operation @command{tar} should perform
1782 (@pxref{Operation Summary}), other options are meant to detail how
1783 this operation should be performed (@pxref{Option Summary}).
1784 Non-option arguments are not always interpreted the same way,
1785 depending on what the operation is.
1787 You will find in this chapter everything about option styles and rules for
1788 writing them (@pxref{Styles}). On the other hand, operations and options
1789 are fully described elsewhere, in other chapters. Here, you will find
1790 only synthetic descriptions for operations and options, together with
1791 pointers to other parts of the @command{tar} manual.
1793 Some options are so special they are fully described right in this
1794 chapter. They have the effect of inhibiting the normal operation of
1795 @command{tar} or else, they globally alter the amount of feedback the user
1796 receives about what is going on. These are the @option{--help} and
1797 @option{--version} (@pxref{help}), @option{--verbose} (@pxref{verbose})
1798 and @option{--interactive} options (@pxref{interactive}).
1802 * using tar options::
1814 @section General Synopsis of @command{tar}
1816 The @GNUTAR{} program is invoked as either one of:
1819 @kbd{tar @var{option}@dots{} [@var{name}]@dots{}}
1820 @kbd{tar @var{letter}@dots{} [@var{argument}]@dots{} [@var{option}]@dots{} [@var{name}]@dots{}}
1823 The second form is for when old options are being used.
1825 You can use @command{tar} to store files in an archive, to extract them from
1826 an archive, and to do other types of archive manipulation. The primary
1827 argument to @command{tar}, which is called the @dfn{operation}, specifies
1828 which action to take. The other arguments to @command{tar} are either
1829 @dfn{options}, which change the way @command{tar} performs an operation,
1830 or file names or archive members, which specify the files or members
1831 @command{tar} is to act on.
1833 You can actually type in arguments in any order, even if in this manual
1834 the options always precede the other arguments, to make examples easier
1835 to understand. Further, the option stating the main operation mode
1836 (the @command{tar} main command) is usually given first.
1838 Each @var{name} in the synopsis above is interpreted as an archive member
1839 name when the main command is one of @option{--compare}
1840 (@option{--diff}, @option{-d}), @option{--delete}, @option{--extract}
1841 (@option{--get}, @option{-x}), @option{--list} (@option{-t}) or
1842 @option{--update} (@option{-u}). When naming archive members, you
1843 must give the exact name of the member in the archive, as it is
1844 printed by @option{--list}. For @option{--append} (@option{-r}) and
1845 @option{--create} (@option{-c}), these @var{name} arguments specify
1846 the names of either files or directory hierarchies to place in the archive.
1847 These files or hierarchies should already exist in the file system,
1848 prior to the execution of the @command{tar} command.
1850 @command{tar} interprets relative file names as being relative to the
1851 working directory. @command{tar} will make all file names relative
1852 (by removing leading slashes when archiving or restoring files),
1853 unless you specify otherwise (using the @option{--absolute-names}
1854 option). @xref{absolute}, for more information about
1855 @option{--absolute-names}.
1857 If you give the name of a directory as either a file name or a member
1858 name, then @command{tar} acts recursively on all the files and directories
1859 beneath that directory. For example, the name @file{/} identifies all
1860 the files in the file system to @command{tar}.
1862 The distinction between file names and archive member names is especially
1863 important when shell globbing is used, and sometimes a source of confusion
1864 for newcomers. @xref{wildcards}, for more information about globbing.
1865 The problem is that shells may only glob using existing files in the
1866 file system. Only @command{tar} itself may glob on archive members, so when
1867 needed, you must ensure that wildcard characters reach @command{tar} without
1868 being interpreted by the shell first. Using a backslash before @samp{*}
1869 or @samp{?}, or putting the whole argument between quotes, is usually
1870 sufficient for this.
1872 Even if @var{name}s are often specified on the command line, they
1873 can also be read from a text file in the file system, using the
1874 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}}) option.
1876 If you don't use any file name arguments, @option{--append} (@option{-r}),
1877 @option{--delete} and @option{--concatenate} (@option{--catenate},
1878 @option{-A}) will do nothing, while @option{--create} (@option{-c})
1879 will usually yield a diagnostic and inhibit @command{tar} execution.
1880 The other operations of @command{tar} (@option{--list},
1881 @option{--extract}, @option{--compare}, and @option{--update})
1882 will act on the entire contents of the archive.
1885 @cindex return status
1886 Besides successful exits, @GNUTAR{} may fail for
1887 many reasons. Some reasons correspond to bad usage, that is, when the
1888 @command{tar} command line is improperly written. Errors may be
1889 encountered later, while processing the archive or the files. Some
1890 errors are recoverable, in which case the failure is delayed until
1891 @command{tar} has completed all its work. Some errors are such that
1892 it would be not meaningful, or at least risky, to continue processing:
1893 @command{tar} then aborts processing immediately. All abnormal exits,
1894 whether immediate or delayed, should always be clearly diagnosed on
1895 @code{stderr}, after a line stating the nature of the error.
1897 Possible exit codes of @GNUTAR{} are summarized in the following
1902 @samp{Successful termination}.
1905 @samp{Some files differ}. If tar was invoked with @option{--compare}
1906 (@option{--diff}, @option{-d}) command line option, this means that
1907 some files in the archive differ from their disk counterparts
1908 (@pxref{compare}). If tar was given @option{--create},
1909 @option{--append} or @option{--update} option, this exit code means
1910 that some files were changed while being archived and so the resulting
1911 archive does not contain the exact copy of the file set.
1914 @samp{Fatal error}. This means that some fatal, unrecoverable error
1918 If @command{tar} has invoked a subprocess and that subprocess exited with a
1919 nonzero exit code, @command{tar} exits with that code as well.
1920 This can happen, for example, if @command{tar} was given some
1921 compression option (@pxref{gzip}) and the external compressor program
1922 failed. Another example is @command{rmt} failure during backup to the
1923 remote device (@pxref{Remote Tape Server}).
1925 @node using tar options
1926 @section Using @command{tar} Options
1928 @GNUTAR{} has a total of eight operating modes which
1929 allow you to perform a variety of tasks. You are required to choose
1930 one operating mode each time you employ the @command{tar} program by
1931 specifying one, and only one operation as an argument to the
1932 @command{tar} command (the corresponding options may be found
1933 at @ref{frequent operations} and @ref{Operations}). Depending on
1934 circumstances, you may also wish to customize how the chosen operating
1935 mode behaves. For example, you may wish to change the way the output
1936 looks, or the format of the files that you wish to archive may require
1937 you to do something special in order to make the archive look right.
1939 You can customize and control @command{tar}'s performance by running
1940 @command{tar} with one or more options (such as @option{--verbose}
1941 (@option{-v}), which we used in the tutorial). As we said in the
1942 tutorial, @dfn{options} are arguments to @command{tar} which are (as
1943 their name suggests) optional. Depending on the operating mode, you
1944 may specify one or more options. Different options will have different
1945 effects, but in general they all change details of the operation, such
1946 as archive format, archive name, or level of user interaction. Some
1947 options make sense with all operating modes, while others are
1948 meaningful only with particular modes. You will likely use some
1949 options frequently, while you will only use others infrequently, or
1950 not at all. (A full list of options is available in @pxref{All Options}.)
1952 @vrindex TAR_OPTIONS, environment variable
1953 @anchor{TAR_OPTIONS}
1954 The @env{TAR_OPTIONS} environment variable specifies default options to
1955 be placed in front of any explicit options. For example, if
1956 @code{TAR_OPTIONS} is @samp{-v --unlink-first}, @command{tar} behaves as
1957 if the two options @option{-v} and @option{--unlink-first} had been
1958 specified before any explicit options. Option specifications are
1959 separated by whitespace. A backslash escapes the next character, so it
1960 can be used to specify an option containing whitespace or a backslash.
1962 Note that @command{tar} options are case sensitive. For example, the
1963 options @option{-T} and @option{-t} are different; the first requires an
1964 argument for stating the name of a file providing a list of @var{name}s,
1965 while the second does not require an argument and is another way to
1966 write @option{--list} (@option{-t}).
1968 In addition to the eight operations, there are many options to
1969 @command{tar}, and three different styles for writing both: long (mnemonic)
1970 form, short form, and old style. These styles are discussed below.
1971 Both the options and the operations can be written in any of these three
1974 @FIXME{menu at end of this node. need to think of an actual outline
1975 for this chapter; probably do that after stuff from chapter 4 is
1979 @section The Three Option Styles
1981 There are three styles for writing operations and options to the command
1982 line invoking @command{tar}. The different styles were developed at
1983 different times during the history of @command{tar}. These styles will be
1984 presented below, from the most recent to the oldest.
1986 Some options must take an argument@footnote{For example, @option{--file}
1987 (@option{-f}) takes the name of an archive file as an argument. If
1988 you do not supply an archive file name, @command{tar} will use a
1989 default, but this can be confusing; thus, we recommend that you always
1990 supply a specific archive file name.}. Where you @emph{place} the
1991 arguments generally depends on which style of options you choose. We
1992 will detail specific information relevant to each option style in the
1993 sections on the different option styles, below. The differences are
1994 subtle, yet can often be very important; incorrect option placement
1995 can cause you to overwrite a number of important files. We urge you
1996 to note these differences, and only use the option style(s) which
1997 makes the most sense to you until you feel comfortable with the others.
1999 Some options @emph{may} take an argument. Such options may have at
2000 most long and short forms, they do not have old style equivalent. The
2001 rules for specifying an argument for such options are stricter than
2002 those for specifying mandatory arguments. Please, pay special
2006 * Long Options:: Long Option Style
2007 * Short Options:: Short Option Style
2008 * Old Options:: Old Option Style
2009 * Mixing:: Mixing Option Styles
2013 @subsection Long Option Style
2015 @cindex long options
2016 @cindex options, long style
2017 @cindex options, GNU style
2018 @cindex options, mnemonic names
2019 Each option has at least one @dfn{long} (or @dfn{mnemonic}) name starting with two
2020 dashes in a row, e.g., @option{--list}. The long names are more clear than
2021 their corresponding short or old names. It sometimes happens that a
2022 single long option has many different names which are
2023 synonymous, such as @option{--compare} and @option{--diff}. In addition,
2024 long option names can be given unique abbreviations. For example,
2025 @option{--cre} can be used in place of @option{--create} because there is no
2026 other long option which begins with @samp{cre}. (One way to find
2027 this out is by trying it and seeing what happens; if a particular
2028 abbreviation could represent more than one option, @command{tar} will tell
2029 you that that abbreviation is ambiguous and you'll know that that
2030 abbreviation won't work. You may also choose to run @samp{tar --help}
2031 to see a list of options. Be aware that if you run @command{tar} with a
2032 unique abbreviation for the long name of an option you didn't want to
2033 use, you are stuck; @command{tar} will perform the command as ordered.)
2035 Long options are meant to be obvious and easy to remember, and their
2036 meanings are generally easier to discern than those of their
2037 corresponding short options (see below). For example:
2040 $ @kbd{tar --create --verbose --blocking-factor=20 --file=/dev/rmt0}
2044 gives a fairly good set of hints about what the command does, even
2045 for those not fully acquainted with @command{tar}.
2047 @cindex arguments to long options
2048 @cindex long options with mandatory arguments
2049 Long options which require arguments take those arguments
2050 immediately following the option name. There are two ways of
2051 specifying a mandatory argument. It can be separated from the
2052 option name either by an equal sign, or by any amount of
2053 white space characters. For example, the @option{--file} option (which
2054 tells the name of the @command{tar} archive) is given a file such as
2055 @file{archive.tar} as argument by using any of the following notations:
2056 @option{--file=archive.tar} or @option{--file archive.tar}.
2058 @cindex optional arguments to long options
2059 @cindex long options with optional arguments
2060 In contrast, optional arguments must always be introduced using
2061 an equal sign. For example, the @option{--backup} option takes
2062 an optional argument specifying backup type. It must be used
2063 as @option{--backup=@var{backup-type}}.
2066 @subsection Short Option Style
2068 @cindex short options
2069 @cindex options, short style
2070 @cindex options, traditional
2071 Most options also have a @dfn{short option} name. Short options start with
2072 a single dash, and are followed by a single character, e.g., @option{-t}
2073 (which is equivalent to @option{--list}). The forms are absolutely
2074 identical in function; they are interchangeable.
2076 The short option names are faster to type than long option names.
2078 @cindex arguments to short options
2079 @cindex short options with mandatory arguments
2080 Short options which require arguments take their arguments immediately
2081 following the option, usually separated by white space. It is also
2082 possible to stick the argument right after the short option name, using
2083 no intervening space. For example, you might write @w{@option{-f
2084 archive.tar}} or @option{-farchive.tar} instead of using
2085 @option{--file=archive.tar}. Both @option{--file=@var{archive-name}} and
2086 @w{@option{-f @var{archive-name}}} denote the option which indicates a
2087 specific archive, here named @file{archive.tar}.
2089 @cindex optional arguments to short options
2090 @cindex short options with optional arguments
2091 Short options which take optional arguments take their arguments
2092 immediately following the option letter, @emph{without any intervening
2093 white space characters}.
2095 Short options' letters may be clumped together, but you are not
2096 required to do this (as compared to old options; see below). When
2097 short options are clumped as a set, use one (single) dash for them
2098 all, e.g., @w{@samp{@command{tar} -cvf}}. Only the last option in
2099 such a set is allowed to have an argument@footnote{Clustering many
2100 options, the last of which has an argument, is a rather opaque way to
2101 write options. Some wonder if @acronym{GNU} @code{getopt} should not
2102 even be made helpful enough for considering such usages as invalid.}.
2104 When the options are separated, the argument for each option which requires
2105 an argument directly follows that option, as is usual for Unix programs.
2109 $ @kbd{tar -c -v -b 20 -f /dev/rmt0}
2112 If you reorder short options' locations, be sure to move any arguments
2113 that belong to them. If you do not move the arguments properly, you may
2114 end up overwriting files.
2117 @subsection Old Option Style
2118 @cindex options, old style
2119 @cindex old option style
2121 Like short options, @dfn{old options} are single letters. However, old options
2122 must be written together as a single clumped set, without spaces separating
2123 them or dashes preceding them@footnote{Beware that if you precede options
2124 with a dash, you are announcing the short option style instead of the
2125 old option style; short options are decoded differently.}. This set
2126 of letters must be the first to appear on the command line, after the
2127 @command{tar} program name and some white space; old options cannot appear
2128 anywhere else. The letter of an old option is exactly the same letter as
2129 the corresponding short option. For example, the old option @samp{t} is
2130 the same as the short option @option{-t}, and consequently, the same as the
2131 long option @option{--list}. So for example, the command @w{@samp{tar
2132 cv}} specifies the option @option{-v} in addition to the operation @option{-c}.
2134 @cindex arguments to old options
2135 @cindex old options with mandatory arguments
2136 When options that need arguments are given together with the command,
2137 all the associated arguments follow, in the same order as the options.
2138 Thus, the example given previously could also be written in the old
2142 $ @kbd{tar cvbf 20 /dev/rmt0}
2146 Here, @samp{20} is the argument of @option{-b} and @samp{/dev/rmt0} is
2147 the argument of @option{-f}.
2149 On the other hand, this old style syntax makes it difficult to match
2150 option letters with their corresponding arguments, and is often
2151 confusing. In the command @w{@samp{tar cvbf 20 /dev/rmt0}}, for example,
2152 @samp{20} is the argument for @option{-b}, @samp{/dev/rmt0} is the
2153 argument for @option{-f}, and @option{-v} does not have a corresponding
2154 argument. Even using short options like in @w{@samp{tar -c -v -b 20 -f
2155 /dev/rmt0}} is clearer, putting all arguments next to the option they
2158 If you want to reorder the letters in the old option argument, be
2159 sure to reorder any corresponding argument appropriately.
2161 This old way of writing @command{tar} options can surprise even experienced
2162 users. For example, the two commands:
2165 @kbd{tar cfz archive.tar.gz file}
2166 @kbd{tar -cfz archive.tar.gz file}
2170 are quite different. The first example uses @file{archive.tar.gz} as
2171 the value for option @samp{f} and recognizes the option @samp{z}. The
2172 second example, however, uses @file{z} as the value for option
2173 @samp{f} --- probably not what was intended.
2175 Old options are kept for compatibility with old versions of @command{tar}.
2177 This second example could be corrected in many ways, among which the
2178 following are equivalent:
2181 @kbd{tar -czf archive.tar.gz file}
2182 @kbd{tar -cf archive.tar.gz -z file}
2183 @kbd{tar cf archive.tar.gz -z file}
2186 @cindex option syntax, traditional
2187 As far as we know, all @command{tar} programs, @acronym{GNU} and
2188 non-@acronym{GNU}, support old options. @GNUTAR{}
2189 supports them not only for historical reasons, but also because many
2190 people are used to them. For compatibility with Unix @command{tar},
2191 the first argument is always treated as containing command and option
2192 letters even if it doesn't start with @samp{-}. Thus, @samp{tar c} is
2193 equivalent to @w{@samp{tar -c}:} both of them specify the
2194 @option{--create} (@option{-c}) command to create an archive.
2197 @subsection Mixing Option Styles
2199 @cindex options, mixing different styles
2200 All three styles may be intermixed in a single @command{tar} command,
2201 so long as the rules for each style are fully
2202 respected@footnote{Before @GNUTAR{} version 1.11.6,
2203 a bug prevented intermixing old style options with long options in
2204 some cases.}. Old style options and either of the modern styles of
2205 options may be mixed within a single @command{tar} command. However,
2206 old style options must be introduced as the first arguments only,
2207 following the rule for old options (old options must appear directly
2208 after the @command{tar} command and some white space). Modern options
2209 may be given only after all arguments to the old options have been
2210 collected. If this rule is not respected, a modern option might be
2211 falsely interpreted as the value of the argument to one of the old
2214 For example, all the following commands are wholly equivalent, and
2215 illustrate the many combinations and orderings of option styles.
2218 @kbd{tar --create --file=archive.tar}
2219 @kbd{tar --create -f archive.tar}
2220 @kbd{tar --create -farchive.tar}
2221 @kbd{tar --file=archive.tar --create}
2222 @kbd{tar --file=archive.tar -c}
2223 @kbd{tar -c --file=archive.tar}
2224 @kbd{tar -c -f archive.tar}
2225 @kbd{tar -c -farchive.tar}
2226 @kbd{tar -cf archive.tar}
2227 @kbd{tar -cfarchive.tar}
2228 @kbd{tar -f archive.tar --create}
2229 @kbd{tar -f archive.tar -c}
2230 @kbd{tar -farchive.tar --create}
2231 @kbd{tar -farchive.tar -c}
2232 @kbd{tar c --file=archive.tar}
2233 @kbd{tar c -f archive.tar}
2234 @kbd{tar c -farchive.tar}
2235 @kbd{tar cf archive.tar}
2236 @kbd{tar f archive.tar --create}
2237 @kbd{tar f archive.tar -c}
2238 @kbd{tar fc archive.tar}
2241 On the other hand, the following commands are @emph{not} equivalent to
2245 @kbd{tar -f -c archive.tar}
2246 @kbd{tar -fc archive.tar}
2247 @kbd{tar -fcarchive.tar}
2248 @kbd{tar -farchive.tarc}
2249 @kbd{tar cfarchive.tar}
2253 These last examples mean something completely different from what the
2254 user intended (judging based on the example in the previous set which
2255 uses long options, whose intent is therefore very clear). The first
2256 four specify that the @command{tar} archive would be a file named
2257 @option{-c}, @samp{c}, @samp{carchive.tar} or @samp{archive.tarc},
2258 respectively. The first two examples also specify a single non-option,
2259 @var{name} argument having the value @samp{archive.tar}. The last
2260 example contains only old style option letters (repeating option
2261 @samp{c} twice), not all of which are meaningful (eg., @samp{.},
2262 @samp{h}, or @samp{i}), with no argument value. @FIXME{not sure i liked
2263 the first sentence of this paragraph..}
2266 @section All @command{tar} Options
2268 The coming manual sections contain an alphabetical listing of all
2269 @command{tar} operations and options, with brief descriptions and
2270 cross-references to more in-depth explanations in the body of the manual.
2271 They also contain an alphabetically arranged table of the short option
2272 forms with their corresponding long option. You can use this table as
2273 a reference for deciphering @command{tar} commands in scripts.
2276 * Operation Summary::
2278 * Short Option Summary::
2281 @node Operation Summary
2282 @subsection Operations
2290 Appends files to the end of the archive. @xref{append}.
2292 @opsummary{catenate}
2296 Same as @option{--concatenate}. @xref{concatenate}.
2302 Compares archive members with their counterparts in the file
2303 system, and reports differences in file size, mode, owner,
2304 modification date and contents. @xref{compare}.
2306 @opsummary{concatenate}
2310 Appends other @command{tar} archives to the end of the archive.
2317 Creates a new @command{tar} archive. @xref{create}.
2322 Deletes members from the archive. Don't try this on an archive on a
2323 tape! @xref{delete}.
2329 Same @option{--compare}. @xref{compare}.
2335 Extracts members from the archive into the file system. @xref{extract}.
2341 Same as @option{--extract}. @xref{extract}.
2347 Lists the members in an archive. @xref{list}.
2353 Adds files to the end of the archive, but only if they are newer than
2354 their counterparts already in the archive, or if they do not already
2355 exist in the archive. @xref{update}.
2359 @node Option Summary
2360 @subsection @command{tar} Options
2364 @opsummary{absolute-names}
2365 @item --absolute-names
2368 Normally when creating an archive, @command{tar} strips an initial
2369 @samp{/} from member names. This option disables that behavior.
2372 @opsummary{after-date}
2375 (See @option{--newer}, @pxref{after})
2377 @opsummary{anchored}
2379 A pattern must match an initial subsequence of the name's components.
2380 @xref{controlling pattern-matching}.
2382 @opsummary{atime-preserve}
2383 @item --atime-preserve
2384 @itemx --atime-preserve=replace
2385 @itemx --atime-preserve=system
2387 Attempt to preserve the access time of files when reading them. This
2388 option currently is effective only on files that you own, unless you
2389 have superuser privileges.
2391 @option{--atime-preserve=replace} remembers the access time of a file
2392 before reading it, and then restores the access time afterwards. This
2393 may cause problems if other programs are reading the file at the same
2394 time, as the times of their accesses will be lost. On most platforms
2395 restoring the access time also requires @command{tar} to restore the
2396 data modification time too, so this option may also cause problems if
2397 other programs are writing the file at the same time (@command{tar} attempts
2398 to detect this situation, but cannot do so reliably due to race
2399 conditions). Worse, on most platforms restoring the access time also
2400 updates the status change time, which means that this option is
2401 incompatible with incremental backups.
2403 @option{--atime-preserve=system} avoids changing time stamps on files,
2404 without interfering with time stamp updates
2405 caused by other programs, so it works better with incremental backups.
2406 However, it requires a special @code{O_NOATIME} option from the
2407 underlying operating and file system implementation, and it also requires
2408 that searching directories does not update their access times. As of
2409 this writing (November 2005) this works only with Linux, and only with
2410 Linux kernels 2.6.8 and later. Worse, there is currently no reliable
2411 way to know whether this feature actually works. Sometimes
2412 @command{tar} knows that it does not work, and if you use
2413 @option{--atime-preserve=system} then @command{tar} complains and
2414 exits right away. But other times @command{tar} might think that the
2415 option works when it actually does not.
2417 Currently @option{--atime-preserve} with no operand defaults to
2418 @option{--atime-preserve=replace}, but this may change in the future
2419 as support for @option{--atime-preserve=system} improves.
2421 If your operating or file system does not support
2422 @option{--atime-preserve=@-system}, you might be able to preserve access
2423 times reliably by using the @command{mount} command. For example,
2424 you can mount the file system read-only, or access the file system via
2425 a read-only loopback mount, or use the @samp{noatime} mount option
2426 available on some systems. However, mounting typically requires
2427 superuser privileges and can be a pain to manage.
2429 @opsummary{auto-compress}
2430 @item --auto-compress
2433 During a @option{--create} operation, enables automatic compressed
2434 format recognition based on the archive suffix. The effect of this
2435 option is cancelled by @option{--no-auto-compress}. @xref{gzip}.
2438 @item --backup=@var{backup-type}
2440 Rather than deleting files from the file system, @command{tar} will
2441 back them up using simple or numbered backups, depending upon
2442 @var{backup-type}. @xref{backup}.
2444 @opsummary{block-number}
2445 @item --block-number
2448 With this option present, @command{tar} prints error messages for read errors
2449 with the block number in the archive file. @xref{block-number}.
2451 @opsummary{blocking-factor}
2452 @item --blocking-factor=@var{blocking}
2453 @itemx -b @var{blocking}
2455 Sets the blocking factor @command{tar} uses to @var{blocking} x 512 bytes per
2456 record. @xref{Blocking Factor}.
2462 This option tells @command{tar} to read or write archives through
2463 @code{bzip2}. @xref{gzip}.
2465 @opsummary{check-device}
2466 @item --check-device
2467 Check device numbers when creating a list of modified files for
2468 incremental archiving. This is the default. @xref{device numbers},
2469 for a detailed description.
2471 @opsummary{checkpoint}
2472 @item --checkpoint[=@var{number}]
2474 This option directs @command{tar} to print periodic checkpoint
2475 messages as it reads through the archive. It is intended for when you
2476 want a visual indication that @command{tar} is still running, but
2477 don't want to see @option{--verbose} output. You can also instruct
2478 @command{tar} to execute a list of actions on each checkpoint, see
2479 @option{--checkpoint-action} below. For a detailed description, see
2482 @opsummary{checkpoint-action}
2483 @item --checkpoint-action=@var{action}
2484 Instruct @command{tar} to execute an action upon hitting a
2485 breakpoint. Here we give only a brief outline. @xref{checkpoints},
2486 for a complete description.
2488 The @var{action} argument can be one of the following:
2492 Produce an audible bell on the console.
2496 Print a single dot on the standard listing stream.
2499 Display a textual message on the standard error, with the status and
2500 number of the checkpoint. This is the default.
2502 @item echo=@var{string}
2503 Display @var{string} on the standard error. Before output, the string
2504 is subject to meta-character expansion.
2506 @item exec=@var{command}
2507 Execute the given @var{command}.
2509 @item sleep=@var{time}
2510 Wait for @var{time} seconds.
2512 @item ttyout=@var{string}
2513 Output @var{string} on the current console (@file{/dev/tty}).
2516 Several @option{--checkpoint-action} options can be specified. The
2517 supplied actions will be executed in order of their appearance in the
2520 Using @option{--checkpoint-action} without @option{--checkpoint}
2521 assumes default checkpoint frequency of one checkpoint per 10 records.
2523 @opsummary{check-links}
2526 If this option was given, @command{tar} will check the number of links
2527 dumped for each processed file. If this number does not match the
2528 total number of hard links for the file, a warning message will be
2529 output @footnote{Earlier versions of @GNUTAR{} understood @option{-l} as a
2530 synonym for @option{--one-file-system}. The current semantics, which
2531 complies to UNIX98, was introduced with version
2532 1.15.91. @xref{Changes}, for more information.}.
2536 @opsummary{compress}
2537 @opsummary{uncompress}
2542 @command{tar} will use the @command{compress} program when reading or
2543 writing the archive. This allows you to directly act on archives
2544 while saving space. @xref{gzip}.
2546 @opsummary{confirmation}
2547 @item --confirmation
2549 (See @option{--interactive}.) @xref{interactive}.
2551 @opsummary{delay-directory-restore}
2552 @item --delay-directory-restore
2554 Delay setting modification times and permissions of extracted
2555 directories until the end of extraction. @xref{Directory Modification Times and Permissions}.
2557 @opsummary{dereference}
2561 When creating a @command{tar} archive, @command{tar} will archive the
2562 file that a symbolic link points to, rather than archiving the
2563 symlink. @xref{dereference}.
2565 @opsummary{directory}
2566 @item --directory=@var{dir}
2569 When this option is specified, @command{tar} will change its current directory
2570 to @var{dir} before performing any operations. When this option is used
2571 during archive creation, it is order sensitive. @xref{directory}.
2574 @item --exclude=@var{pattern}
2576 When performing operations, @command{tar} will skip files that match
2577 @var{pattern}. @xref{exclude}.
2579 @opsummary{exclude-backups}
2580 @item --exclude-backups
2581 Exclude backup and lock files. @xref{exclude,, exclude-backups}.
2583 @opsummary{exclude-from}
2584 @item --exclude-from=@var{file}
2585 @itemx -X @var{file}
2587 Similar to @option{--exclude}, except @command{tar} will use the list of
2588 patterns in the file @var{file}. @xref{exclude}.
2590 @opsummary{exclude-caches}
2591 @item --exclude-caches
2593 Exclude from dump any directory containing a valid cache directory
2594 tag file, but still dump the directory node and the tag file itself.
2596 @xref{exclude,, exclude-caches}.
2598 @opsummary{exclude-caches-under}
2599 @item --exclude-caches-under
2601 Exclude from dump any directory containing a valid cache directory
2602 tag file, but still dump the directory node itself.
2606 @opsummary{exclude-caches-all}
2607 @item --exclude-caches-all
2609 Exclude from dump any directory containing a valid cache directory
2610 tag file. @xref{exclude}.
2612 @opsummary{exclude-tag}
2613 @item --exclude-tag=@var{file}
2615 Exclude from dump any directory containing file named @var{file}, but
2616 dump the directory node and @var{file} itself. @xref{exclude,, exclude-tag}.
2618 @opsummary{exclude-tag-under}
2619 @item --exclude-tag-under=@var{file}
2621 Exclude from dump the contents of any directory containing file
2622 named @var{file}, but dump the directory node itself. @xref{exclude,,
2625 @opsummary{exclude-tag-all}
2626 @item --exclude-tag-all=@var{file}
2628 Exclude from dump any directory containing file named @var{file}.
2629 @xref{exclude,,exclude-tag-all}.
2631 @opsummary{exclude-vcs}
2634 Exclude from dump directories and files, that are internal for some
2635 widely used version control systems.
2637 @xref{exclude,,exclude-vcs}.
2640 @item --file=@var{archive}
2641 @itemx -f @var{archive}
2643 @command{tar} will use the file @var{archive} as the @command{tar} archive it
2644 performs operations on, rather than @command{tar}'s compilation dependent
2645 default. @xref{file tutorial}.
2647 @opsummary{files-from}
2648 @item --files-from=@var{file}
2649 @itemx -T @var{file}
2651 @command{tar} will use the contents of @var{file} as a list of archive members
2652 or files to operate on, in addition to those specified on the
2653 command-line. @xref{files}.
2655 @opsummary{force-local}
2658 Forces @command{tar} to interpret the file name given to @option{--file}
2659 as a local file, even if it looks like a remote tape drive name.
2660 @xref{local and remote archives}.
2663 @item --format=@var{format}
2664 @itemx -H @var{format}
2666 Selects output archive format. @var{Format} may be one of the
2671 Creates an archive that is compatible with Unix V7 @command{tar}.
2674 Creates an archive that is compatible with GNU @command{tar} version
2678 Creates archive in GNU tar 1.13 format. Basically it is the same as
2679 @samp{oldgnu} with the only difference in the way it handles long
2683 Creates a @acronym{POSIX.1-1988} compatible archive.
2686 Creates a @acronym{POSIX.1-2001 archive}.
2690 @xref{Formats}, for a detailed discussion of these formats.
2692 @opsummary{full-time}
2694 This option instructs @command{tar} to print file times to their full
2695 resolution. Usually this means 1-second resolution, but that depends
2696 on the underlying file system. The @option{--full-time} option takes
2697 effect only when detailed output (verbosity level 2 or higher) has
2698 been requested using the @option{--verbose} option, e.g., when listing
2699 or extracting archives:
2702 $ @kbd{tar -t -v --full-time -f archive.tar}
2706 or, when creating an archive:
2709 $ @kbd{tar -c -vv --full-time -f archive.tar .}
2712 Notice, thar when creating the archive you need to specify
2713 @option{--verbose} twice to get a detailed output (@pxref{verbose
2717 @item --group=@var{group}
2719 Files added to the @command{tar} archive will have a group @acronym{ID} of @var{group},
2720 rather than the group from the source file. @var{group} is first decoded
2721 as a group symbolic name, but if this interpretation fails, it has to be
2722 a decimal numeric group @acronym{ID}. @xref{override}.
2724 Also see the comments for the @option{--owner=@var{user}} option.
2734 This option tells @command{tar} to read or write archives through
2735 @command{gzip}, allowing @command{tar} to directly operate on several
2736 kinds of compressed archives transparently. @xref{gzip}.
2738 @opsummary{hard-dereference}
2739 @item --hard-dereference
2740 When creating an archive, dereference hard links and store the files
2741 they refer to, instead of creating usual hard link members.
2749 @command{tar} will print out a short message summarizing the operations and
2750 options to @command{tar} and exit. @xref{help}.
2752 @opsummary{ignore-case}
2754 Ignore case when matching member or file names with
2755 patterns. @xref{controlling pattern-matching}.
2757 @opsummary{ignore-command-error}
2758 @item --ignore-command-error
2759 Ignore exit codes of subprocesses. @xref{Writing to an External Program}.
2761 @opsummary{ignore-failed-read}
2762 @item --ignore-failed-read
2764 Do not exit unsuccessfully merely because an unreadable file was encountered.
2767 @opsummary{ignore-zeros}
2768 @item --ignore-zeros
2771 With this option, @command{tar} will ignore zeroed blocks in the
2772 archive, which normally signals EOF. @xref{Reading}.
2774 @opsummary{incremental}
2778 Informs @command{tar} that it is working with an old
2779 @acronym{GNU}-format incremental backup archive. It is intended
2780 primarily for backwards compatibility only. @xref{Incremental Dumps},
2781 for a detailed discussion of incremental archives.
2783 @opsummary{index-file}
2784 @item --index-file=@var{file}
2786 Send verbose output to @var{file} instead of to standard output.
2788 @opsummary{info-script}
2789 @opsummary{new-volume-script}
2790 @item --info-script=@var{script-file}
2791 @itemx --new-volume-script=@var{script-file}
2792 @itemx -F @var{script-file}
2794 When @command{tar} is performing multi-tape backups, @var{script-file} is run
2795 at the end of each tape. If @var{script-file} exits with nonzero status,
2796 @command{tar} fails immediately. @xref{info-script}, for a detailed
2797 discussion of @var{script-file}.
2799 @opsummary{interactive}
2801 @itemx --confirmation
2804 Specifies that @command{tar} should ask the user for confirmation before
2805 performing potentially destructive options, such as overwriting files.
2808 @opsummary{keep-newer-files}
2809 @item --keep-newer-files
2811 Do not replace existing files that are newer than their archive copies
2812 when extracting files from an archive.
2814 @opsummary{keep-old-files}
2815 @item --keep-old-files
2818 Do not overwrite existing files when extracting files from an archive.
2819 @xref{Keep Old Files}.
2822 @item --label=@var{name}
2823 @itemx -V @var{name}
2825 When creating an archive, instructs @command{tar} to write @var{name}
2826 as a name record in the archive. When extracting or listing archives,
2827 @command{tar} will only operate on archives that have a label matching
2828 the pattern specified in @var{name}. @xref{Tape Files}.
2831 @item --level=@var{n}
2832 Force incremental backup of level @var{n}. As of @GNUTAR version
2833 @value{VERSION}, the option @option{--level=0} truncates the snapshot
2834 file, thereby forcing the level 0 dump. Other values of @var{n} are
2835 effectively ignored. @xref{--level=0}, for details and examples.
2837 The use of this option is valid only in conjunction with the
2838 @option{--listed-incremental} option. @xref{Incremental Dumps},
2839 for a detailed description.
2841 @opsummary{listed-incremental}
2842 @item --listed-incremental=@var{snapshot-file}
2843 @itemx -g @var{snapshot-file}
2845 During a @option{--create} operation, specifies that the archive that
2846 @command{tar} creates is a new @acronym{GNU}-format incremental
2847 backup, using @var{snapshot-file} to determine which files to backup.
2848 With other operations, informs @command{tar} that the archive is in
2849 incremental format. @xref{Incremental Dumps}.
2854 This option tells @command{tar} to read or write archives through
2855 @command{lzip}. @xref{gzip}.
2860 This option tells @command{tar} to read or write archives through
2861 @command{lzma}. @xref{gzip}.
2865 This option tells @command{tar} to read or write archives through
2866 @command{lzop}. @xref{gzip}.
2869 @item --mode=@var{permissions}
2871 When adding files to an archive, @command{tar} will use
2872 @var{permissions} for the archive members, rather than the permissions
2873 from the files. @var{permissions} can be specified either as an octal
2874 number or as symbolic permissions, like with
2875 @command{chmod}. @xref{override}.
2878 @item --mtime=@var{date}
2880 When adding files to an archive, @command{tar} will use @var{date} as
2881 the modification time of members when creating archives, instead of
2882 their actual modification times. The value of @var{date} can be
2883 either a textual date representation (@pxref{Date input formats}) or a
2884 name of the existing file, starting with @samp{/} or @samp{.}. In the
2885 latter case, the modification time of that file is used. @xref{override}.
2887 @opsummary{multi-volume}
2888 @item --multi-volume
2891 Informs @command{tar} that it should create or otherwise operate on a
2892 multi-volume @command{tar} archive. @xref{Using Multiple Tapes}.
2894 @opsummary{new-volume-script}
2895 @item --new-volume-script
2897 (see @option{--info-script})
2900 @item --newer=@var{date}
2901 @itemx --after-date=@var{date}
2904 When creating an archive, @command{tar} will only add files that have changed
2905 since @var{date}. If @var{date} begins with @samp{/} or @samp{.}, it
2906 is taken to be the name of a file whose data modification time specifies
2907 the date. @xref{after}.
2909 @opsummary{newer-mtime}
2910 @item --newer-mtime=@var{date}
2912 Like @option{--newer}, but add only files whose
2913 contents have changed (as opposed to just @option{--newer}, which will
2914 also back up files for which any status information has
2915 changed). @xref{after}.
2917 @opsummary{no-anchored}
2919 An exclude pattern can match any subsequence of the name's components.
2920 @xref{controlling pattern-matching}.
2922 @opsummary{no-auto-compress}
2923 @item --no-auto-compress
2925 Disables automatic compressed format recognition based on the archive
2926 suffix. @xref{--auto-compress}. @xref{gzip}.
2928 @opsummary{no-check-device}
2929 @item --no-check-device
2930 Do not check device numbers when creating a list of modified files
2931 for incremental archiving. @xref{device numbers}, for
2932 a detailed description.
2934 @opsummary{no-delay-directory-restore}
2935 @item --no-delay-directory-restore
2937 Modification times and permissions of extracted
2938 directories are set when all files from this directory have been
2939 extracted. This is the default.
2940 @xref{Directory Modification Times and Permissions}.
2942 @opsummary{no-ignore-case}
2943 @item --no-ignore-case
2944 Use case-sensitive matching.
2945 @xref{controlling pattern-matching}.
2947 @opsummary{no-ignore-command-error}
2948 @item --no-ignore-command-error
2949 Print warnings about subprocesses that terminated with a nonzero exit
2950 code. @xref{Writing to an External Program}.
2955 If the @option{--null} option was given previously, this option
2956 cancels its effect, so that any following @option{--files-from}
2957 options will expect their file lists to be newline-terminated.
2959 @opsummary{no-overwrite-dir}
2960 @item --no-overwrite-dir
2962 Preserve metadata of existing directories when extracting files
2963 from an archive. @xref{Overwrite Old Files}.
2965 @opsummary{no-quote-chars}
2966 @item --no-quote-chars=@var{string}
2967 Remove characters listed in @var{string} from the list of quoted
2968 characters set by the previous @option{--quote-chars} option
2969 (@pxref{quoting styles}).
2971 @opsummary{no-recursion}
2972 @item --no-recursion
2974 With this option, @command{tar} will not recurse into directories.
2977 @opsummary{no-same-owner}
2978 @item --no-same-owner
2981 When extracting an archive, do not attempt to preserve the owner
2982 specified in the @command{tar} archive. This the default behavior
2985 @opsummary{no-same-permissions}
2986 @item --no-same-permissions
2988 When extracting an archive, subtract the user's umask from files from
2989 the permissions specified in the archive. This is the default behavior
2995 The archive media does not support seeks to arbitrary
2996 locations. Usually @command{tar} determines automatically whether
2997 the archive can be seeked or not. Use this option to disable this
3000 @opsummary{no-unquote}
3002 Treat all input file or member names literally, do not interpret
3003 escape sequences. @xref{input name quoting}.
3005 @opsummary{no-wildcards}
3006 @item --no-wildcards
3007 Do not use wildcards.
3008 @xref{controlling pattern-matching}.
3010 @opsummary{no-wildcards-match-slash}
3011 @item --no-wildcards-match-slash
3012 Wildcards do not match @samp{/}.
3013 @xref{controlling pattern-matching}.
3018 When @command{tar} is using the @option{--files-from} option, this option
3019 instructs @command{tar} to expect file names terminated with @acronym{NUL}, so
3020 @command{tar} can correctly work with file names that contain newlines.
3023 @opsummary{numeric-owner}
3024 @item --numeric-owner
3026 This option will notify @command{tar} that it should use numeric user
3027 and group IDs when creating a @command{tar} file, rather than names.
3031 The function of this option depends on the action @command{tar} is
3032 performing. When extracting files, @option{-o} is a synonym for
3033 @option{--no-same-owner}, i.e., it prevents @command{tar} from
3034 restoring ownership of files being extracted.
3036 When creating an archive, it is a synonym for
3037 @option{--old-archive}. This behavior is for compatibility
3038 with previous versions of @GNUTAR{}, and will be
3039 removed in future releases.
3041 @xref{Changes}, for more information.
3043 @opsummary{occurrence}
3044 @item --occurrence[=@var{number}]
3046 This option can be used in conjunction with one of the subcommands
3047 @option{--delete}, @option{--diff}, @option{--extract} or
3048 @option{--list} when a list of files is given either on the command
3049 line or via @option{-T} option.
3051 This option instructs @command{tar} to process only the @var{number}th
3052 occurrence of each named file. @var{Number} defaults to 1, so
3055 tar -x -f archive.tar --occurrence filename
3059 will extract the first occurrence of the member @file{filename} from @file{archive.tar}
3060 and will terminate without scanning to the end of the archive.
3062 @opsummary{old-archive}
3064 Synonym for @option{--format=v7}.
3066 @opsummary{one-file-system}
3067 @item --one-file-system
3068 Used when creating an archive. Prevents @command{tar} from recursing into
3069 directories that are on different file systems from the current
3072 @opsummary{overwrite}
3075 Overwrite existing files and directory metadata when extracting files
3076 from an archive. @xref{Overwrite Old Files}.
3078 @opsummary{overwrite-dir}
3079 @item --overwrite-dir
3081 Overwrite the metadata of existing directories when extracting files
3082 from an archive. @xref{Overwrite Old Files}.
3085 @item --owner=@var{user}
3087 Specifies that @command{tar} should use @var{user} as the owner of members
3088 when creating archives, instead of the user associated with the source
3089 file. @var{user} is first decoded as a user symbolic name, but if
3090 this interpretation fails, it has to be a decimal numeric user @acronym{ID}.
3093 This option does not affect extraction from archives.
3095 @opsummary{pax-option}
3096 @item --pax-option=@var{keyword-list}
3097 This option enables creation of the archive in @acronym{POSIX.1-2001}
3098 format (@pxref{posix}) and modifies the way @command{tar} handles the
3099 extended header keywords. @var{Keyword-list} is a comma-separated
3100 list of keyword options. @xref{PAX keywords}, for a detailed
3103 @opsummary{portability}
3105 @itemx --old-archive
3106 Synonym for @option{--format=v7}.
3110 Same as @option{--format=posix}.
3112 @opsummary{preserve}
3115 Synonymous with specifying both @option{--preserve-permissions} and
3116 @option{--same-order}. @xref{Setting Access Permissions}.
3118 @opsummary{preserve-order}
3119 @item --preserve-order
3121 (See @option{--same-order}; @pxref{Reading}.)
3123 @opsummary{preserve-permissions}
3124 @opsummary{same-permissions}
3125 @item --preserve-permissions
3126 @itemx --same-permissions
3129 When @command{tar} is extracting an archive, it normally subtracts the
3130 users' umask from the permissions specified in the archive and uses
3131 that number as the permissions to create the destination file.
3132 Specifying this option instructs @command{tar} that it should use the
3133 permissions directly from the archive. @xref{Setting Access Permissions}.
3135 @opsummary{quote-chars}
3136 @item --quote-chars=@var{string}
3137 Always quote characters from @var{string}, even if the selected
3138 quoting style would not quote them (@pxref{quoting styles}).
3140 @opsummary{quoting-style}
3141 @item --quoting-style=@var{style}
3142 Set quoting style to use when printing member and file names
3143 (@pxref{quoting styles}). Valid @var{style} values are:
3144 @code{literal}, @code{shell}, @code{shell-always}, @code{c},
3145 @code{escape}, @code{locale}, and @code{clocale}. Default quoting
3146 style is @code{escape}, unless overridden while configuring the
3149 @opsummary{read-full-records}
3150 @item --read-full-records
3153 Specifies that @command{tar} should reblock its input, for reading
3154 from pipes on systems with buggy implementations. @xref{Reading}.
3156 @opsummary{record-size}
3157 @item --record-size=@var{size}[@var{suf}]
3159 Instructs @command{tar} to use @var{size} bytes per record when accessing the
3160 archive. The argument can be suffixed with a @dfn{size suffix}, e.g.
3161 @option{--record-size=10K} for 10 Kilobytes. @xref{size-suffixes},
3162 for a list of valid suffixes. @xref{Blocking Factor}, for a detailed
3163 description of this option.
3165 @opsummary{recursion}
3168 With this option, @command{tar} recurses into directories (default).
3171 @opsummary{recursive-unlink}
3172 @item --recursive-unlink
3175 directory hierarchies before extracting directories of the same name
3176 from the archive. @xref{Recursive Unlink}.
3178 @opsummary{remove-files}
3179 @item --remove-files
3181 Directs @command{tar} to remove the source file from the file system after
3182 appending it to an archive. @xref{remove files}.
3184 @opsummary{restrict}
3187 Disable use of some potentially harmful @command{tar} options.
3188 Currently this option disables shell invocation from multi-volume menu
3189 (@pxref{Using Multiple Tapes}).
3191 @opsummary{rmt-command}
3192 @item --rmt-command=@var{cmd}
3194 Notifies @command{tar} that it should use @var{cmd} instead of
3195 the default @file{/usr/libexec/rmt} (@pxref{Remote Tape Server}).
3197 @opsummary{rsh-command}
3198 @item --rsh-command=@var{cmd}
3200 Notifies @command{tar} that is should use @var{cmd} to communicate with remote
3201 devices. @xref{Device}.
3203 @opsummary{same-order}
3205 @itemx --preserve-order
3208 This option is an optimization for @command{tar} when running on machines with
3209 small amounts of memory. It informs @command{tar} that the list of file
3210 arguments has already been sorted to match the order of files in the
3211 archive. @xref{Reading}.
3213 @opsummary{same-owner}
3216 When extracting an archive, @command{tar} will attempt to preserve the owner
3217 specified in the @command{tar} archive with this option present.
3218 This is the default behavior for the superuser; this option has an
3219 effect only for ordinary users. @xref{Attributes}.
3221 @opsummary{same-permissions}
3222 @item --same-permissions
3224 (See @option{--preserve-permissions}; @pxref{Setting Access Permissions}.)
3230 Assume that the archive media supports seeks to arbitrary
3231 locations. Usually @command{tar} determines automatically whether
3232 the archive can be seeked or not. This option is intended for use
3233 in cases when such recognition fails. It takes effect only if the
3234 archive is open for reading (e.g. with @option{--list} or
3235 @option{--extract} options).
3237 @opsummary{show-defaults}
3238 @item --show-defaults
3240 Displays the default options used by @command{tar} and exits
3241 successfully. This option is intended for use in shell scripts.
3242 Here is an example of what you can see using this option:
3245 $ @kbd{tar --show-defaults}
3246 --format=gnu -f- -b20 --quoting-style=escape
3247 --rmt-command=/usr/libexec/rmt --rsh-command=/usr/bin/rsh
3251 Notice, that this option outputs only one line. The example output
3252 above has been split to fit page boundaries.
3254 @opsummary{show-omitted-dirs}
3255 @item --show-omitted-dirs
3257 Instructs @command{tar} to mention the directories it is skipping when
3258 operating on a @command{tar} archive. @xref{show-omitted-dirs}.
3260 @opsummary{show-transformed-names}
3261 @opsummary{show-stored-names}
3262 @item --show-transformed-names
3263 @itemx --show-stored-names
3265 Display file or member names after applying any transformations
3266 (@pxref{transform}). In particular, when used in conjunction with one of
3267 the archive creation operations it instructs @command{tar} to list the
3268 member names stored in the archive, as opposed to the actual file
3269 names. @xref{listing member and file names}.
3275 Invokes a @acronym{GNU} extension when adding files to an archive that handles
3276 sparse files efficiently. @xref{sparse}.
3278 @opsummary{sparse-version}
3279 @item --sparse-version=@var{version}
3281 Specifies the @dfn{format version} to use when archiving sparse
3282 files. Implies @option{--sparse}. @xref{sparse}. For the description
3283 of the supported sparse formats, @xref{Sparse Formats}.
3285 @opsummary{starting-file}
3286 @item --starting-file=@var{name}
3287 @itemx -K @var{name}
3289 This option affects extraction only; @command{tar} will skip extracting
3290 files in the archive until it finds one that matches @var{name}.
3293 @opsummary{strip-components}
3294 @item --strip-components=@var{number}
3295 Strip given @var{number} of leading components from file names before
3296 extraction. For example, if archive @file{archive.tar} contained
3297 @file{/some/file/name}, then running
3300 tar --extract --file archive.tar --strip-components=2
3304 would extract this file to file @file{name}.
3307 @item --suffix=@var{suffix}
3309 Alters the suffix @command{tar} uses when backing up files from the default
3310 @samp{~}. @xref{backup}.
3312 @opsummary{tape-length}
3313 @item --tape-length=@var{num}[@var{suf}]
3314 @itemx -L @var{num}[@var{suf}]
3316 Specifies the length of tapes that @command{tar} is writing as being
3317 @w{@var{num} x 1024} bytes long. If optional @var{suf} is given, it
3318 specifies a multiplicative factor to be used instead of 1024. For
3319 example, @samp{-L2M} means 2 megabytes. @xref{size-suffixes}, for a
3320 list of allowed suffixes. @xref{Using Multiple Tapes}, for a detailed
3321 discussion of this option.
3323 @opsummary{test-label}
3326 Reads the volume label. If an argument is specified, test whether it
3327 matches the volume label. @xref{--test-label option}.
3329 @opsummary{to-command}
3330 @item --to-command=@var{command}
3332 During extraction @command{tar} will pipe extracted files to the
3333 standard input of @var{command}. @xref{Writing to an External Program}.
3335 @opsummary{to-stdout}
3339 During extraction, @command{tar} will extract files to stdout rather
3340 than to the file system. @xref{Writing to Standard Output}.
3343 @item --totals[=@var{signo}]
3345 Displays the total number of bytes transferred when processing an
3346 archive. If an argument is given, these data are displayed on
3347 request, when signal @var{signo} is delivered to @command{tar}.
3354 Sets the data modification time of extracted files to the extraction time,
3355 rather than the data modification time stored in the archive.
3356 @xref{Data Modification Times}.
3358 @opsummary{transform}
3360 @item --transform=@var{sed-expr}
3361 @itemx --xform=@var{sed-expr}
3362 Transform file or member names using @command{sed} replacement expression
3363 @var{sed-expr}. For example,
3366 $ @kbd{tar cf archive.tar --transform 's,^\./,usr/,' .}
3370 will add to @file{archive} files from the current working directory,
3371 replacing initial @samp{./} prefix with @samp{usr/}. For the detailed
3372 discussion, @xref{transform}.
3374 To see transformed member names in verbose listings, use
3375 @option{--show-transformed-names} option
3376 (@pxref{show-transformed-names}).
3378 @opsummary{uncompress}
3381 (See @option{--compress}, @pxref{gzip})
3386 (See @option{--gzip}, @pxref{gzip})
3388 @opsummary{unlink-first}
3389 @item --unlink-first
3392 Directs @command{tar} to remove the corresponding file from the file
3393 system before extracting it from the archive. @xref{Unlink First}.
3397 Enable unquoting input file or member names (default). @xref{input
3400 @opsummary{use-compress-program}
3401 @item --use-compress-program=@var{prog}
3402 @itemx -I=@var{prog}
3404 Instructs @command{tar} to access the archive through @var{prog}, which is
3405 presumed to be a compression program of some sort. @xref{gzip}.
3410 Display file modification dates in @acronym{UTC}. This option implies
3417 Specifies that @command{tar} should be more verbose about the
3418 operations it is performing. This option can be specified multiple
3419 times for some operations to increase the amount of information displayed.
3426 Verifies that the archive was correctly written when creating an
3427 archive. @xref{verify}.
3432 Print information about the program's name, version, origin and legal
3433 status, all on standard output, and then exit successfully.
3436 @opsummary{volno-file}
3437 @item --volno-file=@var{file}
3439 Used in conjunction with @option{--multi-volume}. @command{tar} will
3440 keep track of which volume of a multi-volume archive it is working in
3441 @var{file}. @xref{volno-file}.
3444 @item --warning=@var{keyword}
3446 Enable or disable warning messages identified by @var{keyword}. The
3447 messages are suppressed if @var{keyword} is prefixed with @samp{no-}.
3450 @opsummary{wildcards}
3452 Use wildcards when matching member names with patterns.
3453 @xref{controlling pattern-matching}.
3455 @opsummary{wildcards-match-slash}
3456 @item --wildcards-match-slash
3457 Wildcards match @samp{/}.
3458 @xref{controlling pattern-matching}.
3463 Use @command{xz} for compressing or decompressing the archives. @xref{gzip}.
3467 @node Short Option Summary
3468 @subsection Short Options Cross Reference
3470 Here is an alphabetized list of all of the short option forms, matching
3471 them with the equivalent long option.
3473 @multitable @columnfractions 0.20 0.80
3474 @headitem Short Option @tab Reference
3476 @item -A @tab @ref{--concatenate}.
3478 @item -B @tab @ref{--read-full-records}.
3480 @item -C @tab @ref{--directory}.
3482 @item -F @tab @ref{--info-script}.
3484 @item -G @tab @ref{--incremental}.
3486 @item -J @tab @ref{--xz}.
3488 @item -K @tab @ref{--starting-file}.
3490 @item -L @tab @ref{--tape-length}.
3492 @item -M @tab @ref{--multi-volume}.
3494 @item -N @tab @ref{--newer}.
3496 @item -O @tab @ref{--to-stdout}.
3498 @item -P @tab @ref{--absolute-names}.
3500 @item -R @tab @ref{--block-number}.
3502 @item -S @tab @ref{--sparse}.
3504 @item -T @tab @ref{--files-from}.
3506 @item -U @tab @ref{--unlink-first}.
3508 @item -V @tab @ref{--label}.
3510 @item -W @tab @ref{--verify}.
3512 @item -X @tab @ref{--exclude-from}.
3514 @item -Z @tab @ref{--compress}.
3516 @item -b @tab @ref{--blocking-factor}.
3518 @item -c @tab @ref{--create}.
3520 @item -d @tab @ref{--compare}.
3522 @item -f @tab @ref{--file}.
3524 @item -g @tab @ref{--listed-incremental}.
3526 @item -h @tab @ref{--dereference}.
3528 @item -i @tab @ref{--ignore-zeros}.
3530 @item -j @tab @ref{--bzip2}.
3532 @item -k @tab @ref{--keep-old-files}.
3534 @item -l @tab @ref{--check-links}.
3536 @item -m @tab @ref{--touch}.
3538 @item -o @tab When creating, @ref{--no-same-owner}, when extracting ---
3539 @ref{--portability}.
3541 The latter usage is deprecated. It is retained for compatibility with
3542 the earlier versions of @GNUTAR{}. In future releases
3543 @option{-o} will be equivalent to @option{--no-same-owner} only.
3545 @item -p @tab @ref{--preserve-permissions}.
3547 @item -r @tab @ref{--append}.
3549 @item -s @tab @ref{--same-order}.
3551 @item -t @tab @ref{--list}.
3553 @item -u @tab @ref{--update}.
3555 @item -v @tab @ref{--verbose}.
3557 @item -w @tab @ref{--interactive}.
3559 @item -x @tab @ref{--extract}.
3561 @item -z @tab @ref{--gzip}.
3566 @section @GNUTAR{} documentation
3568 @cindex Getting program version number
3570 @cindex Version of the @command{tar} program
3571 Being careful, the first thing is really checking that you are using
3572 @GNUTAR{}, indeed. The @option{--version} option
3573 causes @command{tar} to print information about its name, version,
3574 origin and legal status, all on standard output, and then exit
3575 successfully. For example, @w{@samp{tar --version}} might print:
3578 tar (GNU tar) @value{VERSION}
3579 Copyright (C) 2010 Free Software Foundation, Inc.
3580 Copyright (C) 2010 Free Software Foundation, Inc.
3581 License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>.
3582 This is free software: you are free to change and redistribute it.
3583 There is NO WARRANTY, to the extent permitted by law.
3585 Written by John Gilmore and Jay Fenlason.
3589 The first occurrence of @samp{tar} in the result above is the program
3590 name in the package (for example, @command{rmt} is another program),
3591 while the second occurrence of @samp{tar} is the name of the package
3592 itself, containing possibly many programs. The package is currently
3593 named @samp{tar}, after the name of the main program it
3594 contains@footnote{There are plans to merge the @command{cpio} and
3595 @command{tar} packages into a single one which would be called
3596 @code{paxutils}. So, who knows if, one of this days, the
3597 @option{--version} would not output @w{@samp{tar (@acronym{GNU}
3600 @cindex Obtaining help
3601 @cindex Listing all @command{tar} options
3602 @xopindex{help, introduction}
3603 Another thing you might want to do is checking the spelling or meaning
3604 of some particular @command{tar} option, without resorting to this
3605 manual, for once you have carefully read it. @GNUTAR{}
3606 has a short help feature, triggerable through the
3607 @option{--help} option. By using this option, @command{tar} will
3608 print a usage message listing all available options on standard
3609 output, then exit successfully, without doing anything else and
3610 ignoring all other options. Even if this is only a brief summary, it
3611 may be several screens long. So, if you are not using some kind of
3612 scrollable window, you might prefer to use something like:
3615 $ @kbd{tar --help | less}
3619 presuming, here, that you like using @command{less} for a pager. Other
3620 popular pagers are @command{more} and @command{pg}. If you know about some
3621 @var{keyword} which interests you and do not want to read all the
3622 @option{--help} output, another common idiom is doing:
3625 tar --help | grep @var{keyword}
3629 for getting only the pertinent lines. Notice, however, that some
3630 @command{tar} options have long description lines and the above
3631 command will list only the first of them.
3633 The exact look of the option summary displayed by @kbd{tar --help} is
3634 configurable. @xref{Configuring Help Summary}, for a detailed description.
3637 If you only wish to check the spelling of an option, running @kbd{tar
3638 --usage} may be a better choice. This will display a terse list of
3639 @command{tar} options without accompanying explanations.
3641 The short help output is quite succinct, and you might have to get
3642 back to the full documentation for precise points. If you are reading
3643 this paragraph, you already have the @command{tar} manual in some
3644 form. This manual is available in a variety of forms from
3645 @url{http://www.gnu.org/software/tar/manual}. It may be printed out of the @GNUTAR{}
3646 distribution, provided you have @TeX{} already installed somewhere,
3647 and a laser printer around. Just configure the distribution, execute
3648 the command @w{@samp{make dvi}}, then print @file{doc/tar.dvi} the
3649 usual way (contact your local guru to know how). If @GNUTAR{}
3650 has been conveniently installed at your place, this
3651 manual is also available in interactive, hypertextual form as an Info
3652 file. Just call @w{@samp{info tar}} or, if you do not have the
3653 @command{info} program handy, use the Info reader provided within
3654 @acronym{GNU} Emacs, calling @samp{tar} from the main Info menu.
3656 There is currently no @code{man} page for @GNUTAR{}.
3657 If you observe such a @code{man} page on the system you are running,
3658 either it does not belong to @GNUTAR{}, or it has not
3659 been produced by @acronym{GNU}. Some package maintainers convert
3660 @kbd{tar --help} output to a man page, using @command{help2man}. In
3661 any case, please bear in mind that the authoritative source of
3662 information about @GNUTAR{} is this Texinfo documentation.
3665 @section Obtaining @GNUTAR{} default values
3667 @opindex show-defaults
3668 @GNUTAR{} has some predefined defaults that are used when you do not
3669 explicitly specify another values. To obtain a list of such
3670 defaults, use @option{--show-defaults} option. This will output the
3671 values in the form of @command{tar} command line options:
3675 $ @kbd{tar --show-defaults}
3676 --format=gnu -f- -b20 --quoting-style=escape
3677 --rmt-command=/etc/rmt --rsh-command=/usr/bin/rsh
3682 Notice, that this option outputs only one line. The example output above
3683 has been split to fit page boundaries.
3686 The above output shows that this version of @GNUTAR{} defaults to
3687 using @samp{gnu} archive format (@pxref{Formats}), it uses standard
3688 output as the archive, if no @option{--file} option has been given
3689 (@pxref{file tutorial}), the default blocking factor is 20
3690 (@pxref{Blocking Factor}). It also shows the default locations where
3691 @command{tar} will look for @command{rmt} and @command{rsh} binaries.
3694 @section Checking @command{tar} progress
3696 Typically, @command{tar} performs most operations without reporting any
3697 information to the user except error messages. When using @command{tar}
3698 with many options, particularly ones with complicated or
3699 difficult-to-predict behavior, it is possible to make serious mistakes.
3700 @command{tar} provides several options that make observing @command{tar}
3701 easier. These options cause @command{tar} to print information as it
3702 progresses in its job, and you might want to use them just for being
3703 more careful about what is going on, or merely for entertaining
3704 yourself. If you have encountered a problem when operating on an
3705 archive, however, you may need more information than just an error
3706 message in order to solve the problem. The following options can be
3707 helpful diagnostic tools.
3709 @cindex Verbose operation
3711 Normally, the @option{--list} (@option{-t}) command to list an archive
3712 prints just the file names (one per line) and the other commands are
3713 silent. When used with most operations, the @option{--verbose}
3714 (@option{-v}) option causes @command{tar} to print the name of each
3715 file or archive member as it is processed. This and the other options
3716 which make @command{tar} print status information can be useful in
3717 monitoring @command{tar}.
3719 With @option{--create} or @option{--extract}, @option{--verbose} used
3720 once just prints the names of the files or members as they are processed.
3721 Using it twice causes @command{tar} to print a longer listing
3722 (@xref{verbose member listing}, for the description) for each member.
3723 Since @option{--list} already prints the names of the members,
3724 @option{--verbose} used once with @option{--list} causes @command{tar}
3725 to print an @samp{ls -l} type listing of the files in the archive.
3726 The following examples both extract members with long list output:
3729 $ @kbd{tar --extract --file=archive.tar --verbose --verbose}
3730 $ @kbd{tar xvvf archive.tar}
3733 Verbose output appears on the standard output except when an archive is
3734 being written to the standard output, as with @samp{tar --create
3735 --file=- --verbose} (@samp{tar cfv -}, or even @samp{tar cv}---if the
3736 installer let standard output be the default archive). In that case
3737 @command{tar} writes verbose output to the standard error stream.
3739 If @option{--index-file=@var{file}} is specified, @command{tar} sends
3740 verbose output to @var{file} rather than to standard output or standard
3744 @cindex Obtaining total status information
3746 The @option{--totals} option causes @command{tar} to print on the
3747 standard error the total amount of bytes transferred when processing
3748 an archive. When creating or appending to an archive, this option
3749 prints the number of bytes written to the archive and the average
3750 speed at which they have been written, e.g.:
3754 $ @kbd{tar -c -f archive.tar --totals /home}
3755 Total bytes written: 7924664320 (7.4GiB, 85MiB/s)
3759 When reading an archive, this option displays the number of bytes
3764 $ @kbd{tar -x -f archive.tar --totals}
3765 Total bytes read: 7924664320 (7.4GiB, 95MiB/s)
3769 Finally, when deleting from an archive, the @option{--totals} option
3770 displays both numbers plus number of bytes removed from the archive:
3774 $ @kbd{tar --delete -f foo.tar --totals --wildcards '*~'}
3775 Total bytes read: 9543680 (9.2MiB, 201MiB/s)
3776 Total bytes written: 3829760 (3.7MiB, 81MiB/s)
3777 Total bytes deleted: 1474048
3781 You can also obtain this information on request. When
3782 @option{--totals} is used with an argument, this argument is
3783 interpreted as a symbolic name of a signal, upon delivery of which the
3784 statistics is to be printed:
3787 @item --totals=@var{signo}
3788 Print statistics upon delivery of signal @var{signo}. Valid arguments
3789 are: @code{SIGHUP}, @code{SIGQUIT}, @code{SIGINT}, @code{SIGUSR1} and
3790 @code{SIGUSR2}. Shortened names without @samp{SIG} prefix are also
3794 Both forms of @option{--totals} option can be used simultaneously.
3795 Thus, @kbd{tar -x --totals --totals=USR1} instructs @command{tar} to
3796 extract all members from its default archive and print statistics
3797 after finishing the extraction, as well as when receiving signal
3800 @anchor{Progress information}
3801 @cindex Progress information
3802 The @option{--checkpoint} option prints an occasional message
3803 as @command{tar} reads or writes the archive. It is designed for
3804 those who don't need the more detailed (and voluminous) output of
3805 @option{--block-number} (@option{-R}), but do want visual confirmation
3806 that @command{tar} is actually making forward progress. By default it
3807 prints a message each 10 records read or written. This can be changed
3808 by giving it a numeric argument after an equal sign:
3811 $ @kbd{tar -c --checkpoint=1000} /var
3812 tar: Write checkpoint 1000
3813 tar: Write checkpoint 2000
3814 tar: Write checkpoint 3000
3817 This example shows the default checkpoint message used by
3818 @command{tar}. If you place a dot immediately after the equal
3819 sign, it will print a @samp{.} at each checkpoint@footnote{This is
3820 actually a shortcut for @option{--checkpoint=@var{n}
3821 --checkpoint-action=dot}. @xref{checkpoints, dot}.}. For example:
3824 $ @kbd{tar -c --checkpoint=.1000} /var
3828 The @option{--checkpoint} option provides a flexible mechanism for
3829 executing arbitrary actions upon hitting checkpoints, see the next
3830 section (@pxref{checkpoints}), for more information on it.
3832 @opindex show-omitted-dirs
3833 @anchor{show-omitted-dirs}
3834 The @option{--show-omitted-dirs} option, when reading an archive---with
3835 @option{--list} or @option{--extract}, for example---causes a message
3836 to be printed for each directory in the archive which is skipped.
3837 This happens regardless of the reason for skipping: the directory might
3838 not have been named on the command line (implicitly or explicitly),
3839 it might be excluded by the use of the
3840 @option{--exclude=@var{pattern}} option, or some other reason.
3842 @opindex block-number
3843 @cindex Block number where error occurred
3844 @anchor{block-number}
3845 If @option{--block-number} (@option{-R}) is used, @command{tar} prints, along with
3846 every message it would normally produce, the block number within the
3847 archive where the message was triggered. Also, supplementary messages
3848 are triggered when reading blocks full of NULs, or when hitting end of
3849 file on the archive. As of now, if the archive is properly terminated
3850 with a NUL block, the reading of the file may stop before end of file
3851 is met, so the position of end of file will not usually show when
3852 @option{--block-number} (@option{-R}) is used. Note that @GNUTAR{}
3853 drains the archive before exiting when reading the
3854 archive from a pipe.
3856 @cindex Error message, block number of
3857 This option is especially useful when reading damaged archives, since
3858 it helps pinpoint the damaged sections. It can also be used with
3859 @option{--list} (@option{-t}) when listing a file-system backup tape, allowing you to
3860 choose among several backup tapes when retrieving a file later, in
3861 favor of the tape where the file appears earliest (closest to the
3862 front of the tape). @xref{backup}.
3865 @section Checkpoints
3866 @cindex checkpoints, defined
3868 @opindex checkpoint-action
3870 A @dfn{checkpoint} is a moment of time before writing @var{n}th record to
3871 the archive (a @dfn{write checkpoint}), or before reading @var{n}th record
3872 from the archive (a @dfn{read checkpoint}). Checkpoints allow to
3873 periodically execute arbitrary actions.
3875 The checkpoint facility is enabled using the following option:
3878 @xopindex{checkpoint, defined}
3879 @item --checkpoint[=@var{n}]
3880 Schedule checkpoints before writing or reading each @var{n}th record.
3881 The default value for @var{n} is 10.
3884 A list of arbitrary @dfn{actions} can be executed at each checkpoint.
3885 These actions include: pausing, displaying textual messages, and
3886 executing arbitrary external programs. Actions are defined using
3887 the @option{--checkpoint-action} option.
3890 @xopindex{checkpoint-action, defined}
3891 @item --checkpoint-action=@var{action}
3892 Execute an @var{action} at each checkpoint.
3895 @cindex @code{echo}, checkpoint action
3896 The simplest value of @var{action} is @samp{echo}. It instructs
3897 @command{tar} to display the default message on the standard error
3898 stream upon arriving at each checkpoint. The default message is (in
3899 @acronym{POSIX} locale) @samp{Write checkpoint @var{n}}, for write
3900 checkpoints, and @samp{Read checkpoint @var{n}}, for read checkpoints.
3901 Here, @var{n} represents ordinal number of the checkpoint.
3903 In another locales, translated versions of this message are used.
3905 This is the default action, so running:
3908 $ @kbd{tar -c --checkpoint=1000 --checkpoint-action=echo} /var
3915 $ @kbd{tar -c --checkpoint=1000} /var
3918 The @samp{echo} action also allows to supply a customized message.
3919 You do so by placing an equals sign and the message right after it,
3923 --checkpoint-action="echo=Hit %s checkpoint #%u"
3926 The @samp{%s} and @samp{%u} in the above example are
3927 @dfn{meta-characters}. The @samp{%s} meta-character is replaced with
3928 the @dfn{type} of the checkpoint: @samp{write} or
3929 @samp{read} (or a corresponding translated version in locales other
3930 than @acronym{POSIX}). The @samp{%u} meta-character is replaced with
3931 the ordinal number of the checkpoint. Thus, the above example could
3932 produce the following output when used with the @option{--create}
3936 tar: Hit write checkpoint #10
3937 tar: Hit write checkpoint #20
3938 tar: Hit write checkpoint #30
3941 Aside from meta-character expansion, the message string is subject to
3942 @dfn{unquoting}, during which the backslash @dfn{escape sequences} are
3943 replaced with their corresponding @acronym{ASCII} characters
3944 (@pxref{escape sequences}). E.g. the following action will produce an
3945 audible bell and the message described above at each checkpoint:
3948 --checkpoint-action='echo=\aHit %s checkpoint #%u'
3951 @cindex @code{bell}, checkpoint action
3952 There is also a special action which produces an audible signal:
3953 @samp{bell}. It is not equivalent to @samp{echo='\a'}, because
3954 @samp{bell} sends the bell directly to the console (@file{/dev/tty}),
3955 whereas @samp{echo='\a'} sends it to the standard error.
3957 @cindex @code{ttyout}, checkpoint action
3958 The @samp{ttyout=@var{string}} action outputs @var{string} to
3959 @file{/dev/tty}, so it can be used even if the standard output is
3960 redirected elsewhere. The @var{string} is subject to the same
3961 modifications as with @samp{echo} action. In contrast to the latter,
3962 @samp{ttyout} does not prepend @command{tar} executable name to the
3963 string, nor does it output a newline after it. For example, the
3964 following action will print the checkpoint message at the same screen
3965 line, overwriting any previous message:
3968 --checkpoint-action="ttyout=\rHit %s checkpoint #%u"
3971 @cindex @code{dot}, checkpoint action
3972 Another available checkpoint action is @samp{dot} (or @samp{.}). It
3973 instructs @command{tar} to print a single dot on the standard listing
3977 $ @kbd{tar -c --checkpoint=1000 --checkpoint-action=dot} /var
3981 For compatibility with previous @GNUTAR{} versions, this action can
3982 be abbreviated by placing a dot in front of the checkpoint frequency,
3983 as shown in the previous section.
3985 @cindex @code{sleep}, checkpoint action
3986 Yet another action, @samp{sleep}, pauses @command{tar} for a specified
3987 amount of seconds. The following example will stop for 30 seconds at each
3991 $ @kbd{tar -c --checkpoint=1000 --checkpoint-action=sleep=30}
3994 @cindex @code{exec}, checkpoint action
3995 Finally, the @code{exec} action executes a given external program.
3999 $ @kbd{tar -c --checkpoint=1000 --checkpoint-action=exec=/sbin/cpoint}
4002 This program is executed using @command{/bin/sh -c}, with no
4003 additional arguments. Its exit code is ignored. It gets a copy of
4004 @command{tar}'s environment plus the following variables:
4007 @vrindex TAR_VERSION, checkpoint script environment
4009 @GNUTAR{} version number.
4011 @vrindex TAR_ARCHIVE, checkpoint script environment
4013 The name of the archive @command{tar} is processing.
4015 @vrindex TAR_BLOCKING_FACTOR, checkpoint script environment
4016 @item TAR_BLOCKING_FACTOR
4017 Current blocking factor (@pxref{Blocking}).
4019 @vrindex TAR_CHECKPOINT, checkpoint script environment
4020 @item TAR_CHECKPOINT
4021 Number of the checkpoint.
4023 @vrindex TAR_SUBCOMMAND, checkpoint script environment
4024 @item TAR_SUBCOMMAND
4025 A short option describing the operation @command{tar} is executing.
4026 @xref{Operations}, for a complete list of subcommand options.
4028 @vrindex TAR_FORMAT, checkpoint script environment
4030 Format of the archive being processed. @xref{Formats}, for a complete
4031 list of archive format names.
4034 Any number of actions can be defined, by supplying several
4035 @option{--checkpoint-action} options in the command line. For
4036 example, the command below displays two messages, pauses
4037 execution for 30 seconds and executes the @file{/sbin/cpoint} script:
4041 $ @kbd{tar -c -f arc.tar \
4042 --checkpoint-action='\aecho=Hit %s checkpoint #%u' \
4043 --checkpoint-action='echo=Sleeping for 30 seconds' \
4044 --checkpoint-action='sleep=30' \
4045 --checkpoint-action='exec=/sbin/cpoint'}
4049 This example also illustrates the fact that
4050 @option{--checkpoint-action} can be used without
4051 @option{--checkpoint}. In this case, the default checkpoint frequency
4052 (at each 10th record) is assumed.
4055 @section Controlling Warning Messages
4057 Sometimes, while performing the requested task, @GNUTAR{} notices
4058 some conditions that are not exactly errors, but which the user
4059 should be aware of. When this happens, @command{tar} issues a
4060 @dfn{warning message} describing the condition. Warning messages
4061 are output to the standard error and they do not affect the exit
4062 code of @command{tar} command.
4064 @xopindex{warning, explained}
4065 @GNUTAR{} allows the user to suppress some or all of its warning
4069 @item --warning=@var{keyword}
4070 Control display of the warning messages identified by @var{keyword}.
4071 If @var{keyword} starts with the prefix @samp{no-}, such messages are
4072 suppressed. Otherwise, they are enabled.
4074 Multiple @option{--warning} messages accumulate.
4076 The tables below list allowed values for @var{keyword} along with the
4077 warning messages they control.
4080 @subheading Keywords controlling @command{tar} operation
4084 Enable all warning messages. This is the default.
4087 Disable all warning messages.
4088 @kwindex filename-with-nuls
4089 @cindex @samp{file name read contains nul character}, warning message
4090 @item filename-with-nuls
4091 @samp{%s: file name read contains nul character}
4092 @kwindex alone-zero-block
4093 @cindex @samp{A lone zero block at}, warning message
4094 @item alone-zero-block
4095 @samp{A lone zero block at %s}
4098 @subheading Keywords applicable for @command{tar --create}
4101 @cindex @samp{contains a cache directory tag}, warning message
4103 @samp{%s: contains a cache directory tag %s; %s}
4104 @kwindex file-shrank
4105 @cindex @samp{File shrank by %s bytes}, warning message
4107 @samp{%s: File shrank by %s bytes; padding with zeros}
4109 @cindex @samp{file is on a different filesystem}, warning message
4111 @samp{%s: file is on a different filesystem; not dumped}
4112 @kwindex file-ignored
4113 @cindex @samp{Unknown file type; file ignored}, warning message
4114 @cindex @samp{socket ignored}, warning message
4115 @cindex @samp{door ignored}, warning message
4117 @samp{%s: Unknown file type; file ignored}
4118 @*@samp{%s: socket ignored}
4119 @*@samp{%s: door ignored}
4120 @kwindex file-unchanged
4121 @cindex @samp{file is unchanged; not dumped}, warning message
4122 @item file-unchanged
4123 @samp{%s: file is unchanged; not dumped}
4124 @kwindex ignore-archive
4125 @cindex @samp{file is the archive; not dumped}, warning message
4126 @kwindex ignore-archive
4127 @cindex @samp{file is the archive; not dumped}, warning message
4128 @item ignore-archive
4129 @samp{%s: file is the archive; not dumped}
4130 @kwindex file-removed
4131 @cindex @samp{File removed before we read it}, warning message
4133 @samp{%s: File removed before we read it}
4134 @kwindex file-changed
4135 @cindex @samp{file changed as we read it}, warning message
4137 @samp{%s: file changed as we read it}
4140 @subheading Keywords applicable for @command{tar --extract}
4143 @cindex @samp{implausibly old time stamp %s}, warning message
4144 @cindex @samp{time stamp %s is %s s in the future}, warning message
4146 @samp{%s: implausibly old time stamp %s}
4147 @*@samp{%s: time stamp %s is %s s in the future}
4148 @kwindex contiguous-cast
4149 @cindex @samp{Extracting contiguous files as regular files}, warning message
4150 @item contiguous-cast
4151 @samp{Extracting contiguous files as regular files}
4152 @kwindex symlink-cast
4153 @cindex @samp{Attempting extraction of symbolic links as hard links}, warning message
4155 @samp{Attempting extraction of symbolic links as hard links}
4156 @kwindex unknown-cast
4157 @cindex @samp{Unknown file type `%c', extracted as normal file}, warning message
4159 @samp{%s: Unknown file type `%c', extracted as normal file}
4160 @kwindex ignore-newer
4161 @cindex @samp{Current %s is newer or same age}, warning message
4163 @samp{Current %s is newer or same age}
4164 @kwindex unknown-keyword
4165 @cindex @samp{Ignoring unknown extended header keyword `%s'}, warning message
4166 @item unknown-keyword
4167 @samp{Ignoring unknown extended header keyword `%s'}
4170 @subheading Keywords controlling incremental extraction:
4172 @kwindex rename-directory
4173 @cindex @samp{%s: Directory has been renamed from %s}, warning message
4174 @cindex @samp{%s: Directory has been renamed}, warning message
4175 @item rename-directory
4176 @samp{%s: Directory has been renamed from %s}
4177 @*@samp{%s: Directory has been renamed}
4178 @kwindex new-directory
4179 @cindex @samp{%s: Directory is new}, warning message
4181 @samp{%s: Directory is new}
4183 @cindex @samp{%s: directory is on a different device: not purging}, warning message
4185 @samp{%s: directory is on a different device: not purging}
4186 @kwindex bad-dumpdir
4187 @cindex @samp{Malformed dumpdir: 'X' never used}, warning message
4189 @samp{Malformed dumpdir: 'X' never used}
4193 @section Asking for Confirmation During Operations
4194 @cindex Interactive operation
4196 Typically, @command{tar} carries out a command without stopping for
4197 further instructions. In some situations however, you may want to
4198 exclude some files and archive members from the operation (for instance
4199 if disk or storage space is tight). You can do this by excluding
4200 certain files automatically (@pxref{Choosing}), or by performing
4201 an operation interactively, using the @option{--interactive} (@option{-w}) option.
4202 @command{tar} also accepts @option{--confirmation} for this option.
4204 @opindex interactive
4205 When the @option{--interactive} (@option{-w}) option is specified, before
4206 reading, writing, or deleting files, @command{tar} first prints a message
4207 for each such file, telling what operation it intends to take, then asks
4208 for confirmation on the terminal. The actions which require
4209 confirmation include adding a file to the archive, extracting a file
4210 from the archive, deleting a file from the archive, and deleting a file
4211 from disk. To confirm the action, you must type a line of input
4212 beginning with @samp{y}. If your input line begins with anything other
4213 than @samp{y}, @command{tar} skips that file.
4215 If @command{tar} is reading the archive from the standard input,
4216 @command{tar} opens the file @file{/dev/tty} to support the interactive
4219 Verbose output is normally sent to standard output, separate from
4220 other error messages. However, if the archive is produced directly
4221 on standard output, then verbose output is mixed with errors on
4222 @code{stderr}. Producing the archive on standard output may be used
4223 as a way to avoid using disk space, when the archive is soon to be
4224 consumed by another process reading it, say. Some people felt the need
4225 of producing an archive on stdout, still willing to segregate between
4226 verbose output and error output. A possible approach would be using a
4227 named pipe to receive the archive, and having the consumer process to
4228 read from that named pipe. This has the advantage of letting standard
4229 output free to receive verbose output, all separate from errors.
4232 @chapter @GNUTAR{} Operations
4245 @section Basic @GNUTAR{} Operations
4247 The basic @command{tar} operations, @option{--create} (@option{-c}),
4248 @option{--list} (@option{-t}) and @option{--extract} (@option{--get},
4249 @option{-x}), are currently presented and described in the tutorial
4250 chapter of this manual. This section provides some complementary notes
4251 for these operations.
4254 @xopindex{create, complementary notes}
4258 Creating an empty archive would have some kind of elegance. One can
4259 initialize an empty archive and later use @option{--append}
4260 (@option{-r}) for adding all members. Some applications would not
4261 welcome making an exception in the way of adding the first archive
4262 member. On the other hand, many people reported that it is
4263 dangerously too easy for @command{tar} to destroy a magnetic tape with
4264 an empty archive@footnote{This is well described in @cite{Unix-haters
4265 Handbook}, by Simson Garfinkel, Daniel Weise & Steven Strassmann, IDG
4266 Books, ISBN 1-56884-203-1.}. The two most common errors are:
4270 Mistakingly using @code{create} instead of @code{extract}, when the
4271 intent was to extract the full contents of an archive. This error
4272 is likely: keys @kbd{c} and @kbd{x} are right next to each other on
4273 the QWERTY keyboard. Instead of being unpacked, the archive then
4274 gets wholly destroyed. When users speak about @dfn{exploding} an
4275 archive, they usually mean something else :-).
4278 Forgetting the argument to @code{file}, when the intent was to create
4279 an archive with a single file in it. This error is likely because a
4280 tired user can easily add the @kbd{f} key to the cluster of option
4281 letters, by the mere force of habit, without realizing the full
4282 consequence of doing so. The usual consequence is that the single
4283 file, which was meant to be saved, is rather destroyed.
4286 So, recognizing the likelihood and the catastrophic nature of these
4287 errors, @GNUTAR{} now takes some distance from elegance, and
4288 cowardly refuses to create an archive when @option{--create} option is
4289 given, there are no arguments besides options, and
4290 @option{--files-from} (@option{-T}) option is @emph{not} used. To get
4291 around the cautiousness of @GNUTAR{} and nevertheless create an
4292 archive with nothing in it, one may still use, as the value for the
4293 @option{--files-from} option, a file with no names in it, as shown in
4294 the following commands:
4297 @kbd{tar --create --file=empty-archive.tar --files-from=/dev/null}
4298 @kbd{tar cfT empty-archive.tar /dev/null}
4301 @xopindex{extract, complementary notes}
4306 A socket is stored, within a @GNUTAR{} archive, as a pipe.
4308 @item @option{--list} (@option{-t})
4310 @GNUTAR{} now shows dates as @samp{1996-08-30},
4311 while it used to show them as @samp{Aug 30 1996}. Preferably,
4312 people should get used to ISO 8601 dates. Local American dates should
4313 be made available again with full date localization support, once
4314 ready. In the meantime, programs not being localizable for dates
4315 should prefer international dates, that's really the way to go.
4317 Look up @url{http://www.cl.cam.ac.uk/@/~mgk25/@/iso-time.html} if you
4318 are curious, it contains a detailed explanation of the ISO 8601 standard.
4323 @section Advanced @GNUTAR{} Operations
4325 Now that you have learned the basics of using @GNUTAR{}, you may want
4326 to learn about further ways in which @command{tar} can help you.
4328 This chapter presents five, more advanced operations which you probably
4329 won't use on a daily basis, but which serve more specialized functions.
4330 We also explain the different styles of options and why you might want
4331 to use one or another, or a combination of them in your @command{tar}
4332 commands. Additionally, this chapter includes options which allow you to
4333 define the output from @command{tar} more carefully, and provide help and
4334 error correction in special circumstances.
4336 @FIXME{check this after the chapter is actually revised to make sure
4337 it still introduces the info in the chapter correctly : ).}
4349 @subsection The Five Advanced @command{tar} Operations
4351 @cindex basic operations
4352 In the last chapter, you learned about the first three operations to
4353 @command{tar}. This chapter presents the remaining five operations to
4354 @command{tar}: @option{--append}, @option{--update}, @option{--concatenate},
4355 @option{--delete}, and @option{--compare}.
4357 You are not likely to use these operations as frequently as those
4358 covered in the last chapter; however, since they perform specialized
4359 functions, they are quite useful when you do need to use them. We
4360 will give examples using the same directory and files that you created
4361 in the last chapter. As you may recall, the directory is called
4362 @file{practice}, the files are @samp{jazz}, @samp{blues}, @samp{folk},
4363 and the two archive files you created are
4364 @samp{collection.tar} and @samp{music.tar}.
4366 We will also use the archive files @samp{afiles.tar} and
4367 @samp{bfiles.tar}. The archive @samp{afiles.tar} contains the members @samp{apple},
4368 @samp{angst}, and @samp{aspic}; @samp{bfiles.tar} contains the members
4369 @samp{./birds}, @samp{baboon}, and @samp{./box}.
4371 Unless we state otherwise, all practicing you do and examples you follow
4372 in this chapter will take place in the @file{practice} directory that
4373 you created in the previous chapter; see @ref{prepare for examples}.
4374 (Below in this section, we will remind you of the state of the examples
4375 where the last chapter left them.)
4377 The five operations that we will cover in this chapter are:
4382 Add new entries to an archive that already exists.
4385 Add more recent copies of archive members to the end of an archive, if
4390 Add one or more pre-existing archives to the end of another archive.
4392 Delete items from an archive (does not work on tapes).
4396 Compare archive members to their counterparts in the file system.
4400 @subsection How to Add Files to Existing Archives: @option{--append}
4402 @cindex appending files to existing archive
4404 If you want to add files to an existing archive, you don't need to
4405 create a new archive; you can use @option{--append} (@option{-r}).
4406 The archive must already exist in order to use @option{--append}. (A
4407 related operation is the @option{--update} operation; you can use this
4408 to add newer versions of archive members to an existing archive. To learn how to
4409 do this with @option{--update}, @pxref{update}.)
4411 If you use @option{--append} to add a file that has the same name as an
4412 archive member to an archive containing that archive member, then the
4413 old member is not deleted. What does happen, however, is somewhat
4414 complex. @command{tar} @emph{allows} you to have infinite number of files
4415 with the same name. Some operations treat these same-named members no
4416 differently than any other set of archive members: for example, if you
4417 view an archive with @option{--list} (@option{-t}), you will see all
4418 of those members listed, with their data modification times, owners, etc.
4420 Other operations don't deal with these members as perfectly as you might
4421 prefer; if you were to use @option{--extract} to extract the archive,
4422 only the most recently added copy of a member with the same name as
4423 other members would end up in the working directory. This is because
4424 @option{--extract} extracts an archive in the order the members appeared
4425 in the archive; the most recently archived members will be extracted
4426 last. Additionally, an extracted member will @emph{replace} a file of
4427 the same name which existed in the directory already, and @command{tar}
4428 will not prompt you about this@footnote{Unless you give it
4429 @option{--keep-old-files} option, or the disk copy is newer than
4430 the one in the archive and you invoke @command{tar} with
4431 @option{--keep-newer-files} option.}. Thus, only the most recently archived
4432 member will end up being extracted, as it will replace the one
4433 extracted before it, and so on.
4435 @cindex extracting @var{n}th copy of the file
4436 @xopindex{occurrence, described}
4437 There exists a special option that allows you to get around this
4438 behavior and extract (or list) only a particular copy of the file.
4439 This is @option{--occurrence} option. If you run @command{tar} with
4440 this option, it will extract only the first copy of the file. You
4441 may also give this option an argument specifying the number of
4442 copy to be extracted. Thus, for example if the archive
4443 @file{archive.tar} contained three copies of file @file{myfile}, then
4447 tar --extract --file archive.tar --occurrence=2 myfile
4451 would extract only the second copy. @xref{Option
4452 Summary,---occurrence}, for the description of @option{--occurrence}
4455 @FIXME{ hag -- you might want to incorporate some of the above into the
4456 MMwtSN node; not sure. i didn't know how to make it simpler...
4458 There are a few ways to get around this. Xref to Multiple Members
4459 with the Same Name, maybe.}
4461 @cindex Members, replacing with other members
4462 @cindex Replacing members with other members
4463 @xopindex{delete, using before --append}
4464 If you want to replace an archive member, use @option{--delete} to
4465 delete the member you want to remove from the archive, and then use
4466 @option{--append} to add the member you want to be in the archive. Note
4467 that you can not change the order of the archive; the most recently
4468 added member will still appear last. In this sense, you cannot truly
4469 ``replace'' one member with another. (Replacing one member with another
4470 will not work on certain types of media, such as tapes; see @ref{delete}
4471 and @ref{Media}, for more information.)
4474 * appending files:: Appending Files to an Archive
4478 @node appending files
4479 @subsubsection Appending Files to an Archive
4480 @cindex Adding files to an Archive
4481 @cindex Appending files to an Archive
4482 @cindex Archives, Appending files to
4485 The simplest way to add a file to an already existing archive is the
4486 @option{--append} (@option{-r}) operation, which writes specified
4487 files into the archive whether or not they are already among the
4490 When you use @option{--append}, you @emph{must} specify file name
4491 arguments, as there is no default. If you specify a file that already
4492 exists in the archive, another copy of the file will be added to the
4493 end of the archive. As with other operations, the member names of the
4494 newly added files will be exactly the same as their names given on the
4495 command line. The @option{--verbose} (@option{-v}) option will print
4496 out the names of the files as they are written into the archive.
4498 @option{--append} cannot be performed on some tape drives, unfortunately,
4499 due to deficiencies in the formats those tape drives use. The archive
4500 must be a valid @command{tar} archive, or else the results of using this
4501 operation will be unpredictable. @xref{Media}.
4503 To demonstrate using @option{--append} to add a file to an archive,
4504 create a file called @file{rock} in the @file{practice} directory.
4505 Make sure you are in the @file{practice} directory. Then, run the
4506 following @command{tar} command to add @file{rock} to
4507 @file{collection.tar}:
4510 $ @kbd{tar --append --file=collection.tar rock}
4514 If you now use the @option{--list} (@option{-t}) operation, you will see that
4515 @file{rock} has been added to the archive:
4518 $ @kbd{tar --list --file=collection.tar}
4519 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
4520 -rw-r--r-- me user 21 1996-09-23 16:44 blues
4521 -rw-r--r-- me user 20 1996-09-23 16:44 folk
4522 -rw-r--r-- me user 20 1996-09-23 16:44 rock
4526 @subsubsection Multiple Members with the Same Name
4527 @cindex members, multiple
4528 @cindex multiple members
4530 You can use @option{--append} (@option{-r}) to add copies of files
4531 which have been updated since the archive was created. (However, we
4532 do not recommend doing this since there is another @command{tar}
4533 option called @option{--update}; @xref{update}, for more information.
4534 We describe this use of @option{--append} here for the sake of
4535 completeness.) When you extract the archive, the older version will
4536 be effectively lost. This works because files are extracted from an
4537 archive in the order in which they were archived. Thus, when the
4538 archive is extracted, a file archived later in time will replace a
4539 file of the same name which was archived earlier, even though the
4540 older version of the file will remain in the archive unless you delete
4541 all versions of the file.
4543 Supposing you change the file @file{blues} and then append the changed
4544 version to @file{collection.tar}. As you saw above, the original
4545 @file{blues} is in the archive @file{collection.tar}. If you change the
4546 file and append the new version of the file to the archive, there will
4547 be two copies in the archive. When you extract the archive, the older
4548 version of the file will be extracted first, and then replaced by the
4549 newer version when it is extracted.
4551 You can append the new, changed copy of the file @file{blues} to the
4552 archive in this way:
4555 $ @kbd{tar --append --verbose --file=collection.tar blues}
4560 Because you specified the @option{--verbose} option, @command{tar} has
4561 printed the name of the file being appended as it was acted on. Now
4562 list the contents of the archive:
4565 $ @kbd{tar --list --verbose --file=collection.tar}
4566 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
4567 -rw-r--r-- me user 21 1996-09-23 16:44 blues
4568 -rw-r--r-- me user 20 1996-09-23 16:44 folk
4569 -rw-r--r-- me user 20 1996-09-23 16:44 rock
4570 -rw-r--r-- me user 58 1996-10-24 18:30 blues
4574 The newest version of @file{blues} is now at the end of the archive
4575 (note the different creation dates and file sizes). If you extract
4576 the archive, the older version of the file @file{blues} will be
4577 replaced by the newer version. You can confirm this by extracting
4578 the archive and running @samp{ls} on the directory.
4580 If you wish to extract the first occurrence of the file @file{blues}
4581 from the archive, use @option{--occurrence} option, as shown in
4582 the following example:
4585 $ @kbd{tar --extract -vv --occurrence --file=collection.tar blues}
4586 -rw-r--r-- me user 21 1996-09-23 16:44 blues
4589 @xref{Writing}, for more information on @option{--extract} and
4590 see @ref{Option Summary, --occurrence}, for a description of
4591 @option{--occurrence} option.
4594 @subsection Updating an Archive
4595 @cindex Updating an archive
4598 In the previous section, you learned how to use @option{--append} to
4599 add a file to an existing archive. A related operation is
4600 @option{--update} (@option{-u}). The @option{--update} operation
4601 updates a @command{tar} archive by comparing the date of the specified
4602 archive members against the date of the file with the same name. If
4603 the file has been modified more recently than the archive member, then
4604 the newer version of the file is added to the archive (as with
4607 Unfortunately, you cannot use @option{--update} with magnetic tape drives.
4608 The operation will fail.
4610 @FIXME{other examples of media on which --update will fail? need to ask
4611 charles and/or mib/thomas/dave shevett..}
4613 Both @option{--update} and @option{--append} work by adding to the end
4614 of the archive. When you extract a file from the archive, only the
4615 version stored last will wind up in the file system, unless you use
4616 the @option{--backup} option. @xref{multiple}, for a detailed discussion.
4623 @subsubsection How to Update an Archive Using @option{--update}
4626 You must use file name arguments with the @option{--update}
4627 (@option{-u}) operation. If you don't specify any files,
4628 @command{tar} won't act on any files and won't tell you that it didn't
4629 do anything (which may end up confusing you).
4631 @c note: the above parenthetical added because in fact, this
4632 @c behavior just confused the author. :-)
4634 To see the @option{--update} option at work, create a new file,
4635 @file{classical}, in your practice directory, and some extra text to the
4636 file @file{blues}, using any text editor. Then invoke @command{tar} with
4637 the @samp{update} operation and the @option{--verbose} (@option{-v})
4638 option specified, using the names of all the files in the @file{practice}
4639 directory as file name arguments:
4642 $ @kbd{tar --update -v -f collection.tar blues folk rock classical}
4649 Because we have specified verbose mode, @command{tar} prints out the names
4650 of the files it is working on, which in this case are the names of the
4651 files that needed to be updated. If you run @samp{tar --list} and look
4652 at the archive, you will see @file{blues} and @file{classical} at its
4653 end. There will be a total of two versions of the member @samp{blues};
4654 the one at the end will be newer and larger, since you added text before
4657 The reason @command{tar} does not overwrite the older file when updating
4658 it is because writing to the middle of a section of tape is a difficult
4659 process. Tapes are not designed to go backward. @xref{Media}, for more
4660 information about tapes.
4662 @option{--update} (@option{-u}) is not suitable for performing backups for two
4663 reasons: it does not change directory content entries, and it
4664 lengthens the archive every time it is used. The @GNUTAR{}
4665 options intended specifically for backups are more
4666 efficient. If you need to run backups, please consult @ref{Backups}.
4669 @subsection Combining Archives with @option{--concatenate}
4671 @cindex Adding archives to an archive
4672 @cindex Concatenating Archives
4673 @opindex concatenate
4675 @c @cindex @option{-A} described
4676 Sometimes it may be convenient to add a second archive onto the end of
4677 an archive rather than adding individual files to the archive. To add
4678 one or more archives to the end of another archive, you should use the
4679 @option{--concatenate} (@option{--catenate}, @option{-A}) operation.
4681 To use @option{--concatenate}, give the first archive with
4682 @option{--file} option and name the rest of archives to be
4683 concatenated on the command line. The members, and their member
4684 names, will be copied verbatim from those archives to the first
4685 one@footnote{This can cause multiple members to have the same name. For
4686 information on how this affects reading the archive, see @ref{multiple}.}.
4687 The new, concatenated archive will be called by the same name as the
4688 one given with the @option{--file} option. As usual, if you omit
4689 @option{--file}, @command{tar} will use the value of the environment
4690 variable @env{TAPE}, or, if this has not been set, the default archive name.
4692 @FIXME{There is no way to specify a new name...}
4694 To demonstrate how @option{--concatenate} works, create two small archives
4695 called @file{bluesrock.tar} and @file{folkjazz.tar}, using the relevant
4696 files from @file{practice}:
4699 $ @kbd{tar -cvf bluesrock.tar blues rock}
4702 $ @kbd{tar -cvf folkjazz.tar folk jazz}
4708 If you like, You can run @samp{tar --list} to make sure the archives
4709 contain what they are supposed to:
4712 $ @kbd{tar -tvf bluesrock.tar}
4713 -rw-r--r-- melissa user 105 1997-01-21 19:42 blues
4714 -rw-r--r-- melissa user 33 1997-01-20 15:34 rock
4715 $ @kbd{tar -tvf jazzfolk.tar}
4716 -rw-r--r-- melissa user 20 1996-09-23 16:44 folk
4717 -rw-r--r-- melissa user 65 1997-01-30 14:15 jazz
4720 We can concatenate these two archives with @command{tar}:
4724 $ @kbd{tar --concatenate --file=bluesrock.tar jazzfolk.tar}
4727 If you now list the contents of the @file{bluesrock.tar}, you will see
4728 that now it also contains the archive members of @file{jazzfolk.tar}:
4731 $ @kbd{tar --list --file=bluesrock.tar}
4738 When you use @option{--concatenate}, the source and target archives must
4739 already exist and must have been created using compatible format
4740 parameters. Notice, that @command{tar} does not check whether the
4741 archives it concatenates have compatible formats, it does not
4742 even check if the files are really tar archives.
4744 Like @option{--append} (@option{-r}), this operation cannot be performed on some
4745 tape drives, due to deficiencies in the formats those tape drives use.
4747 @cindex @code{concatenate} vs @command{cat}
4748 @cindex @command{cat} vs @code{concatenate}
4749 It may seem more intuitive to you to want or try to use @command{cat} to
4750 concatenate two archives instead of using the @option{--concatenate}
4751 operation; after all, @command{cat} is the utility for combining files.
4753 However, @command{tar} archives incorporate an end-of-file marker which
4754 must be removed if the concatenated archives are to be read properly as
4755 one archive. @option{--concatenate} removes the end-of-archive marker
4756 from the target archive before each new archive is appended. If you use
4757 @command{cat} to combine the archives, the result will not be a valid
4758 @command{tar} format archive. If you need to retrieve files from an
4759 archive that was added to using the @command{cat} utility, use the
4760 @option{--ignore-zeros} (@option{-i}) option. @xref{Ignore Zeros}, for further
4761 information on dealing with archives improperly combined using the
4762 @command{cat} shell utility.
4765 @subsection Removing Archive Members Using @option{--delete}
4766 @cindex Deleting files from an archive
4767 @cindex Removing files from an archive
4770 You can remove members from an archive by using the @option{--delete}
4771 option. Specify the name of the archive with @option{--file}
4772 (@option{-f}) and then specify the names of the members to be deleted;
4773 if you list no member names, nothing will be deleted. The
4774 @option{--verbose} option will cause @command{tar} to print the names
4775 of the members as they are deleted. As with @option{--extract}, you
4776 must give the exact member names when using @samp{tar --delete}.
4777 @option{--delete} will remove all versions of the named file from the
4778 archive. The @option{--delete} operation can run very slowly.
4780 Unlike other operations, @option{--delete} has no short form.
4782 @cindex Tapes, using @option{--delete} and
4783 @cindex Deleting from tape archives
4784 This operation will rewrite the archive. You can only use
4785 @option{--delete} on an archive if the archive device allows you to
4786 write to any point on the media, such as a disk; because of this, it
4787 does not work on magnetic tapes. Do not try to delete an archive member
4788 from a magnetic tape; the action will not succeed, and you will be
4789 likely to scramble the archive and damage your tape. There is no safe
4790 way (except by completely re-writing the archive) to delete files from
4791 most kinds of magnetic tape. @xref{Media}.
4793 To delete all versions of the file @file{blues} from the archive
4794 @file{collection.tar} in the @file{practice} directory, make sure you
4795 are in that directory, and then,
4798 $ @kbd{tar --list --file=collection.tar}
4803 $ @kbd{tar --delete --file=collection.tar blues}
4804 $ @kbd{tar --list --file=collection.tar}
4810 @FIXME{Check if the above listing is actually produced after running
4811 all the examples on collection.tar.}
4813 The @option{--delete} option has been reported to work properly when
4814 @command{tar} acts as a filter from @code{stdin} to @code{stdout}.
4817 @subsection Comparing Archive Members with the File System
4818 @cindex Verifying the currency of an archive
4821 The @option{--compare} (@option{-d}), or @option{--diff} operation compares
4822 specified archive members against files with the same names, and then
4823 reports differences in file size, mode, owner, modification date and
4824 contents. You should @emph{only} specify archive member names, not file
4825 names. If you do not name any members, then @command{tar} will compare the
4826 entire archive. If a file is represented in the archive but does not
4827 exist in the file system, @command{tar} reports a difference.
4829 You have to specify the record size of the archive when modifying an
4830 archive with a non-default record size.
4832 @command{tar} ignores files in the file system that do not have
4833 corresponding members in the archive.
4835 The following example compares the archive members @file{rock},
4836 @file{blues} and @file{funk} in the archive @file{bluesrock.tar} with
4837 files of the same name in the file system. (Note that there is no file,
4838 @file{funk}; @command{tar} will report an error message.)
4841 $ @kbd{tar --compare --file=bluesrock.tar rock blues funk}
4844 tar: funk not found in archive
4847 The spirit behind the @option{--compare} (@option{--diff},
4848 @option{-d}) option is to check whether the archive represents the
4849 current state of files on disk, more than validating the integrity of
4850 the archive media. For this latter goal, see @ref{verify}.
4852 @node create options
4853 @section Options Used by @option{--create}
4855 @xopindex{create, additional options}
4856 The previous chapter described the basics of how to use
4857 @option{--create} (@option{-c}) to create an archive from a set of files.
4858 @xref{create}. This section described advanced options to be used with
4862 * override:: Overriding File Metadata.
4863 * Ignore Failed Read::
4867 @subsection Overriding File Metadata
4869 As described above, a @command{tar} archive keeps, for each member it contains,
4870 its @dfn{metadata}, such as modification time, mode and ownership of
4871 the file. @GNUTAR{} allows to replace these data with other values
4872 when adding files to the archive. The options described in this
4873 section affect creation of archives of any type. For POSIX archives,
4874 see also @ref{PAX keywords}, for additional ways of controlling
4875 metadata, stored in the archive.
4879 @item --mode=@var{permissions}
4881 When adding files to an archive, @command{tar} will use
4882 @var{permissions} for the archive members, rather than the permissions
4883 from the files. @var{permissions} can be specified either as an octal
4884 number or as symbolic permissions, like with
4885 @command{chmod} (@xref{File permissions, Permissions, File
4886 permissions, fileutils, @acronym{GNU} file utilities}. This reference
4887 also has useful information for those not being overly familiar with
4888 the UNIX permission system). Using latter syntax allows for
4889 more flexibility. For example, the value @samp{a+rw} adds read and write
4890 permissions for everybody, while retaining executable bits on directories
4891 or on any other file already marked as executable:
4894 $ @kbd{tar -c -f archive.tar --mode='a+rw' .}
4897 @item --mtime=@var{date}
4900 When adding files to an archive, @command{tar} will use @var{date} as
4901 the modification time of members when creating archives, instead of
4902 their actual modification times. The argument @var{date} can be
4903 either a textual date representation in almost arbitrary format
4904 (@pxref{Date input formats}) or a name of an existing file, starting
4905 with @samp{/} or @samp{.}. In the latter case, the modification time
4906 of that file will be used.
4908 The following example will set the modification date to 00:00:00,
4912 $ @kbd{tar -c -f archive.tar --mtime='1970-01-01' .}
4916 When used with @option{--verbose} (@pxref{verbose tutorial}) @GNUTAR{}
4917 will try to convert the specified date back to its textual
4918 representation and compare it with the one given with
4919 @option{--mtime} options. If the two dates differ, @command{tar} will
4920 print a warning saying what date it will use. This is to help user
4921 ensure he is using the right date.
4926 $ @kbd{tar -c -f archive.tar -v --mtime=yesterday .}
4927 tar: Option --mtime: Treating date `yesterday' as 2006-06-20
4932 @item --owner=@var{user}
4935 Specifies that @command{tar} should use @var{user} as the owner of members
4936 when creating archives, instead of the user associated with the source
4937 file. The argument @var{user} can be either an existing user symbolic
4938 name, or a decimal numeric user @acronym{ID}.
4940 There is no value indicating a missing number, and @samp{0} usually means
4941 @code{root}. Some people like to force @samp{0} as the value to offer in
4942 their distributions for the owner of files, because the @code{root} user is
4943 anonymous anyway, so that might as well be the owner of anonymous
4944 archives. For example:
4947 $ @kbd{tar -c -f archive.tar --owner=0 .}
4954 $ @kbd{tar -c -f archive.tar --owner=root .}
4957 @item --group=@var{group}
4960 Files added to the @command{tar} archive will have a group @acronym{ID} of @var{group},
4961 rather than the group from the source file. The argument @var{group}
4962 can be either an existing group symbolic name, or a decimal numeric group @acronym{ID}.
4965 @node Ignore Failed Read
4966 @subsection Ignore Fail Read
4969 @item --ignore-failed-read
4970 @opindex ignore-failed-read
4971 Do not exit with nonzero on unreadable files or directories.
4974 @node extract options
4975 @section Options Used by @option{--extract}
4976 @cindex options for use with @option{--extract}
4978 @xopindex{extract, additional options}
4979 The previous chapter showed how to use @option{--extract} to extract
4980 an archive into the file system. Various options cause @command{tar} to
4981 extract more information than just file contents, such as the owner,
4982 the permissions, the modification date, and so forth. This section
4983 presents options to be used with @option{--extract} when certain special
4984 considerations arise. You may review the information presented in
4985 @ref{extract} for more basic information about the
4986 @option{--extract} operation.
4989 * Reading:: Options to Help Read Archives
4990 * Writing:: Changing How @command{tar} Writes Files
4991 * Scarce:: Coping with Scarce Resources
4995 @subsection Options to Help Read Archives
4996 @cindex Options when reading archives
4998 @cindex Reading incomplete records
4999 @cindex Records, incomplete
5000 @opindex read-full-records
5001 Normally, @command{tar} will request data in full record increments from
5002 an archive storage device. If the device cannot return a full record,
5003 @command{tar} will report an error. However, some devices do not always
5004 return full records, or do not require the last record of an archive to
5005 be padded out to the next record boundary. To keep reading until you
5006 obtain a full record, or to accept an incomplete record if it contains
5007 an end-of-archive marker, specify the @option{--read-full-records} (@option{-B}) option
5008 in conjunction with the @option{--extract} or @option{--list} operations.
5011 The @option{--read-full-records} (@option{-B}) option is turned on by default when
5012 @command{tar} reads an archive from standard input, or from a remote
5013 machine. This is because on @acronym{BSD} Unix systems, attempting to read a
5014 pipe returns however much happens to be in the pipe, even if it is
5015 less than was requested. If this option were not enabled, @command{tar}
5016 would fail as soon as it read an incomplete record from the pipe.
5018 If you're not sure of the blocking factor of an archive, you can
5019 read the archive by specifying @option{--read-full-records} (@option{-B}) and
5020 @option{--blocking-factor=@var{512-size}} (@option{-b
5021 @var{512-size}}), using a blocking factor larger than what the archive
5022 uses. This lets you avoid having to determine the blocking factor
5023 of an archive. @xref{Blocking Factor}.
5026 * read full records::
5030 @node read full records
5031 @unnumberedsubsubsec Reading Full Records
5033 @FIXME{need sentence or so of intro here}
5036 @opindex read-full-records
5037 @item --read-full-records
5039 Use in conjunction with @option{--extract} (@option{--get},
5040 @option{-x}) to read an archive which contains incomplete records, or
5041 one which has a blocking factor less than the one specified.
5045 @unnumberedsubsubsec Ignoring Blocks of Zeros
5047 @cindex End-of-archive blocks, ignoring
5048 @cindex Ignoring end-of-archive blocks
5049 @opindex ignore-zeros
5050 Normally, @command{tar} stops reading when it encounters a block of zeros
5051 between file entries (which usually indicates the end of the archive).
5052 @option{--ignore-zeros} (@option{-i}) allows @command{tar} to
5053 completely read an archive which contains a block of zeros before the
5054 end (i.e., a damaged archive, or one that was created by concatenating
5055 several archives together).
5057 The @option{--ignore-zeros} (@option{-i}) option is turned off by default because many
5058 versions of @command{tar} write garbage after the end-of-archive entry,
5059 since that part of the media is never supposed to be read. @GNUTAR{}
5060 does not write after the end of an archive, but seeks to
5061 maintain compatibility among archiving utilities.
5064 @item --ignore-zeros
5066 To ignore blocks of zeros (i.e., end-of-archive entries) which may be
5067 encountered while reading an archive. Use in conjunction with
5068 @option{--extract} or @option{--list}.
5072 @subsection Changing How @command{tar} Writes Files
5075 @FIXME{Introductory paragraph}
5078 * Dealing with Old Files::
5079 * Overwrite Old Files::
5081 * Keep Newer Files::
5083 * Recursive Unlink::
5084 * Data Modification Times::
5085 * Setting Access Permissions::
5086 * Directory Modification Times and Permissions::
5087 * Writing to Standard Output::
5088 * Writing to an External Program::
5092 @node Dealing with Old Files
5093 @unnumberedsubsubsec Options Controlling the Overwriting of Existing Files
5095 @xopindex{overwrite-dir, introduced}
5096 When extracting files, if @command{tar} discovers that the extracted
5097 file already exists, it normally replaces the file by removing it before
5098 extracting it, to prevent confusion in the presence of hard or symbolic
5099 links. (If the existing file is a symbolic link, it is removed, not
5100 followed.) However, if a directory cannot be removed because it is
5101 nonempty, @command{tar} normally overwrites its metadata (ownership,
5102 permission, etc.). The @option{--overwrite-dir} option enables this
5103 default behavior. To be more cautious and preserve the metadata of
5104 such a directory, use the @option{--no-overwrite-dir} option.
5106 @cindex Overwriting old files, prevention
5107 @xopindex{keep-old-files, introduced}
5108 To be even more cautious and prevent existing files from being replaced, use
5109 the @option{--keep-old-files} (@option{-k}) option. It causes @command{tar} to refuse
5110 to replace or update a file that already exists, i.e., a file with the
5111 same name as an archive member prevents extraction of that archive
5112 member. Instead, it reports an error.
5114 @xopindex{overwrite, introduced}
5115 To be more aggressive about altering existing files, use the
5116 @option{--overwrite} option. It causes @command{tar} to overwrite
5117 existing files and to follow existing symbolic links when extracting.
5119 @cindex Protecting old files
5120 Some people argue that @GNUTAR{} should not hesitate
5121 to overwrite files with other files when extracting. When extracting
5122 a @command{tar} archive, they expect to see a faithful copy of the
5123 state of the file system when the archive was created. It is debatable
5124 that this would always be a proper behavior. For example, suppose one
5125 has an archive in which @file{usr/local} is a link to
5126 @file{usr/local2}. Since then, maybe the site removed the link and
5127 renamed the whole hierarchy from @file{/usr/local2} to
5128 @file{/usr/local}. Such things happen all the time. I guess it would
5129 not be welcome at all that @GNUTAR{} removes the
5130 whole hierarchy just to make room for the link to be reinstated
5131 (unless it @emph{also} simultaneously restores the full
5132 @file{/usr/local2}, of course!) @GNUTAR{} is indeed
5133 able to remove a whole hierarchy to reestablish a symbolic link, for
5134 example, but @emph{only if} @option{--recursive-unlink} is specified
5135 to allow this behavior. In any case, single files are silently
5138 @xopindex{unlink-first, introduced}
5139 Finally, the @option{--unlink-first} (@option{-U}) option can improve performance in
5140 some cases by causing @command{tar} to remove files unconditionally
5141 before extracting them.
5143 @node Overwrite Old Files
5144 @unnumberedsubsubsec Overwrite Old Files
5149 Overwrite existing files and directory metadata when extracting files
5152 This causes @command{tar} to write extracted files into the file system without
5153 regard to the files already on the system; i.e., files with the same
5154 names as archive members are overwritten when the archive is extracted.
5155 It also causes @command{tar} to extract the ownership, permissions,
5156 and time stamps onto any preexisting files or directories.
5157 If the name of a corresponding file name is a symbolic link, the file
5158 pointed to by the symbolic link will be overwritten instead of the
5159 symbolic link itself (if this is possible). Moreover, special devices,
5160 empty directories and even symbolic links are automatically removed if
5161 they are in the way of extraction.
5163 Be careful when using the @option{--overwrite} option, particularly when
5164 combined with the @option{--absolute-names} (@option{-P}) option, as this combination
5165 can change the contents, ownership or permissions of any file on your
5166 system. Also, many systems do not take kindly to overwriting files that
5167 are currently being executed.
5169 @opindex overwrite-dir
5170 @item --overwrite-dir
5171 Overwrite the metadata of directories when extracting files from an
5172 archive, but remove other files before extracting.
5175 @node Keep Old Files
5176 @unnumberedsubsubsec Keep Old Files
5179 @opindex keep-old-files
5180 @item --keep-old-files
5182 Do not replace existing files from archive. The
5183 @option{--keep-old-files} (@option{-k}) option prevents @command{tar}
5184 from replacing existing files with files with the same name from the
5185 archive. The @option{--keep-old-files} option is meaningless with
5186 @option{--list} (@option{-t}). Prevents @command{tar} from replacing
5187 files in the file system during extraction.
5190 @node Keep Newer Files
5191 @unnumberedsubsubsec Keep Newer Files
5194 @opindex keep-newer-files
5195 @item --keep-newer-files
5196 Do not replace existing files that are newer than their archive
5197 copies. This option is meaningless with @option{--list} (@option{-t}).
5201 @unnumberedsubsubsec Unlink First
5204 @opindex unlink-first
5205 @item --unlink-first
5207 Remove files before extracting over them.
5208 This can make @command{tar} run a bit faster if you know in advance
5209 that the extracted files all need to be removed. Normally this option
5210 slows @command{tar} down slightly, so it is disabled by default.
5213 @node Recursive Unlink
5214 @unnumberedsubsubsec Recursive Unlink
5217 @opindex recursive-unlink
5218 @item --recursive-unlink
5219 When this option is specified, try removing files and directory hierarchies
5220 before extracting over them. @emph{This is a dangerous option!}
5223 If you specify the @option{--recursive-unlink} option,
5224 @command{tar} removes @emph{anything} that keeps you from extracting a file
5225 as far as current permissions will allow it. This could include removal
5226 of the contents of a full directory hierarchy.
5228 @node Data Modification Times
5229 @unnumberedsubsubsec Setting Data Modification Times
5231 @cindex Data modification times of extracted files
5232 @cindex Modification times of extracted files
5233 Normally, @command{tar} sets the data modification times of extracted
5234 files to the corresponding times recorded for the files in the archive, but
5235 limits the permissions of extracted files by the current @code{umask}
5238 To set the data modification times of extracted files to the time when
5239 the files were extracted, use the @option{--touch} (@option{-m}) option in
5240 conjunction with @option{--extract} (@option{--get}, @option{-x}).
5246 Sets the data modification time of extracted archive members to the time
5247 they were extracted, not the time recorded for them in the archive.
5248 Use in conjunction with @option{--extract} (@option{--get}, @option{-x}).
5251 @node Setting Access Permissions
5252 @unnumberedsubsubsec Setting Access Permissions
5254 @cindex Permissions of extracted files
5255 @cindex Modes of extracted files
5256 To set the modes (access permissions) of extracted files to those
5257 recorded for those files in the archive, use @option{--same-permissions}
5258 in conjunction with the @option{--extract} (@option{--get},
5259 @option{-x}) operation.
5262 @opindex preserve-permissions
5263 @opindex same-permissions
5264 @item --preserve-permissions
5265 @itemx --same-permissions
5266 @c @itemx --ignore-umask
5268 Set modes of extracted archive members to those recorded in the
5269 archive, instead of current umask settings. Use in conjunction with
5270 @option{--extract} (@option{--get}, @option{-x}).
5273 @node Directory Modification Times and Permissions
5274 @unnumberedsubsubsec Directory Modification Times and Permissions
5276 After successfully extracting a file member, @GNUTAR{} normally
5277 restores its permissions and modification times, as described in the
5278 previous sections. This cannot be done for directories, because
5279 after extracting a directory @command{tar} will almost certainly
5280 extract files into that directory and this will cause the directory
5281 modification time to be updated. Moreover, restoring that directory
5282 permissions may not permit file creation within it. Thus, restoring
5283 directory permissions and modification times must be delayed at least
5284 until all files have been extracted into that directory. @GNUTAR{}
5285 restores directories using the following approach.
5287 The extracted directories are created with the mode specified in the
5288 archive, as modified by the umask of the user, which gives sufficient
5289 permissions to allow file creation. The meta-information about the
5290 directory is recorded in the temporary list of directories. When
5291 preparing to extract next archive member, @GNUTAR{} checks if the
5292 directory prefix of this file contains the remembered directory. If
5293 it does not, the program assumes that all files have been extracted
5294 into that directory, restores its modification time and permissions
5295 and removes its entry from the internal list. This approach allows
5296 to correctly restore directory meta-information in the majority of
5297 cases, while keeping memory requirements sufficiently small. It is
5298 based on the fact, that most @command{tar} archives use the predefined
5299 order of members: first the directory, then all the files and
5300 subdirectories in that directory.
5302 However, this is not always true. The most important exception are
5303 incremental archives (@pxref{Incremental Dumps}). The member order in
5304 an incremental archive is reversed: first all directory members are
5305 stored, followed by other (non-directory) members. So, when extracting
5306 from incremental archives, @GNUTAR{} alters the above procedure. It
5307 remembers all restored directories, and restores their meta-data
5308 only after the entire archive has been processed. Notice, that you do
5309 not need to specify any special options for that, as @GNUTAR{}
5310 automatically detects archives in incremental format.
5312 There may be cases, when such processing is required for normal archives
5313 too. Consider the following example:
5317 $ @kbd{tar --no-recursion -cvf archive \
5318 foo foo/file1 bar bar/file foo/file2}
5327 During the normal operation, after encountering @file{bar}
5328 @GNUTAR{} will assume that all files from the directory @file{foo}
5329 were already extracted and will therefore restore its timestamp and
5330 permission bits. However, after extracting @file{foo/file2} the
5331 directory timestamp will be offset again.
5333 To correctly restore directory meta-information in such cases, use
5334 the @option{--delay-directory-restore} command line option:
5337 @opindex delay-directory-restore
5338 @item --delay-directory-restore
5339 Delays restoring of the modification times and permissions of extracted
5340 directories until the end of extraction. This way, correct
5341 meta-information is restored even if the archive has unusual member
5344 @opindex no-delay-directory-restore
5345 @item --no-delay-directory-restore
5346 Cancel the effect of the previous @option{--delay-directory-restore}.
5347 Use this option if you have used @option{--delay-directory-restore} in
5348 @env{TAR_OPTIONS} variable (@pxref{TAR_OPTIONS}) and wish to
5349 temporarily disable it.
5352 @node Writing to Standard Output
5353 @unnumberedsubsubsec Writing to Standard Output
5355 @cindex Writing extracted files to standard output
5356 @cindex Standard output, writing extracted files to
5357 To write the extracted files to the standard output, instead of
5358 creating the files on the file system, use @option{--to-stdout} (@option{-O}) in
5359 conjunction with @option{--extract} (@option{--get}, @option{-x}). This option is useful if you are
5360 extracting files to send them through a pipe, and do not need to
5361 preserve them in the file system. If you extract multiple members,
5362 they appear on standard output concatenated, in the order they are
5363 found in the archive.
5369 Writes files to the standard output. Use only in conjunction with
5370 @option{--extract} (@option{--get}, @option{-x}). When this option is
5371 used, instead of creating the files specified, @command{tar} writes
5372 the contents of the files extracted to its standard output. This may
5373 be useful if you are only extracting the files in order to send them
5374 through a pipe. This option is meaningless with @option{--list}
5378 This can be useful, for example, if you have a tar archive containing
5379 a big file and don't want to store the file on disk before processing
5380 it. You can use a command like this:
5383 tar -xOzf foo.tgz bigfile | process
5386 or even like this if you want to process the concatenation of the files:
5389 tar -xOzf foo.tgz bigfile1 bigfile2 | process
5392 However, @option{--to-command} may be more convenient for use with
5393 multiple files. See the next section.
5395 @node Writing to an External Program
5396 @unnumberedsubsubsec Writing to an External Program
5398 You can instruct @command{tar} to send the contents of each extracted
5399 file to the standard input of an external program:
5403 @item --to-command=@var{command}
5404 Extract files and pipe their contents to the standard input of
5405 @var{command}. When this option is used, instead of creating the
5406 files specified, @command{tar} invokes @var{command} and pipes the
5407 contents of the files to its standard output. The @var{command} may
5408 contain command line arguments. The program is executed via
5409 @code{sh -c}. Notice, that @var{command} is executed once for each regular file
5410 extracted. Non-regular files (directories, etc.) are ignored when this
5414 The command can obtain the information about the file it processes
5415 from the following environment variables:
5418 @vrindex TAR_FILETYPE, to-command environment
5420 Type of the file. It is a single letter with the following meaning:
5422 @multitable @columnfractions 0.10 0.90
5423 @item f @tab Regular file
5424 @item d @tab Directory
5425 @item l @tab Symbolic link
5426 @item h @tab Hard link
5427 @item b @tab Block device
5428 @item c @tab Character device
5431 Currently only regular files are supported.
5433 @vrindex TAR_MODE, to-command environment
5435 File mode, an octal number.
5437 @vrindex TAR_FILENAME, to-command environment
5439 The name of the file.
5441 @vrindex TAR_REALNAME, to-command environment
5443 Name of the file as stored in the archive.
5445 @vrindex TAR_UNAME, to-command environment
5447 Name of the file owner.
5449 @vrindex TAR_GNAME, to-command environment
5451 Name of the file owner group.
5453 @vrindex TAR_ATIME, to-command environment
5455 Time of last access. It is a decimal number, representing seconds
5456 since the Epoch. If the archive provides times with nanosecond
5457 precision, the nanoseconds are appended to the timestamp after a
5460 @vrindex TAR_MTIME, to-command environment
5462 Time of last modification.
5464 @vrindex TAR_CTIME, to-command environment
5466 Time of last status change.
5468 @vrindex TAR_SIZE, to-command environment
5472 @vrindex TAR_UID, to-command environment
5474 UID of the file owner.
5476 @vrindex TAR_GID, to-command environment
5478 GID of the file owner.
5481 Additionally, the following variables contain information about
5482 tar mode and the archive being processed:
5485 @vrindex TAR_VERSION, to-command environment
5487 @GNUTAR{} version number.
5489 @vrindex TAR_ARCHIVE, to-command environment
5491 The name of the archive @command{tar} is processing.
5493 @vrindex TAR_BLOCKING_FACTOR, to-command environment
5494 @item TAR_BLOCKING_FACTOR
5495 Current blocking factor (@pxref{Blocking}).
5497 @vrindex TAR_VOLUME, to-command environment
5499 Ordinal number of the volume @command{tar} is processing.
5501 @vrindex TAR_FORMAT, to-command environment
5503 Format of the archive being processed. @xref{Formats}, for a complete
5504 list of archive format names.
5507 If @var{command} exits with a non-0 status, @command{tar} will print
5508 an error message similar to the following:
5511 tar: 2345: Child returned status 1
5514 Here, @samp{2345} is the PID of the finished process.
5516 If this behavior is not wanted, use @option{--ignore-command-error}:
5519 @opindex ignore-command-error
5520 @item --ignore-command-error
5521 Ignore exit codes of subprocesses. Notice that if the program
5522 exits on signal or otherwise terminates abnormally, the error message
5523 will be printed even if this option is used.
5525 @opindex no-ignore-command-error
5526 @item --no-ignore-command-error
5527 Cancel the effect of any previous @option{--ignore-command-error}
5528 option. This option is useful if you have set
5529 @option{--ignore-command-error} in @env{TAR_OPTIONS}
5530 (@pxref{TAR_OPTIONS}) and wish to temporarily cancel it.
5534 @unnumberedsubsubsec Removing Files
5536 @FIXME{The section is too terse. Something more to add? An example,
5540 @opindex remove-files
5541 @item --remove-files
5542 Remove files after adding them to the archive.
5546 @subsection Coping with Scarce Resources
5549 @cindex Small memory
5550 @cindex Running out of space
5558 @unnumberedsubsubsec Starting File
5561 @opindex starting-file
5562 @item --starting-file=@var{name}
5563 @itemx -K @var{name}
5564 Starts an operation in the middle of an archive. Use in conjunction
5565 with @option{--extract} (@option{--get}, @option{-x}) or @option{--list} (@option{-t}).
5568 @cindex Middle of the archive, starting in the
5569 If a previous attempt to extract files failed due to lack of disk
5570 space, you can use @option{--starting-file=@var{name}} (@option{-K
5571 @var{name}}) to start extracting only after member @var{name} of the
5572 archive. This assumes, of course, that there is now free space, or
5573 that you are now extracting into a different file system. (You could
5574 also choose to suspend @command{tar}, remove unnecessary files from
5575 the file system, and then resume the same @command{tar} operation.
5576 In this case, @option{--starting-file} is not necessary.) See also
5577 @ref{interactive}, and @ref{exclude}.
5580 @unnumberedsubsubsec Same Order
5583 @cindex Large lists of file names on small machines
5585 @opindex preserve-order
5587 @itemx --preserve-order
5589 To process large lists of file names on machines with small amounts of
5590 memory. Use in conjunction with @option{--compare} (@option{--diff},
5591 @option{-d}), @option{--list} (@option{-t}) or @option{--extract}
5592 (@option{--get}, @option{-x}).
5595 The @option{--same-order} (@option{--preserve-order}, @option{-s}) option tells @command{tar} that the list of file
5596 names to be listed or extracted is sorted in the same order as the
5597 files in the archive. This allows a large list of names to be used,
5598 even on a small machine that would not otherwise be able to hold all
5599 the names in memory at the same time. Such a sorted list can easily be
5600 created by running @samp{tar -t} on the archive and editing its output.
5602 This option is probably never needed on modern computer systems.
5605 @section Backup options
5607 @cindex backup options
5609 @GNUTAR{} offers options for making backups of files
5610 before writing new versions. These options control the details of
5611 these backups. They may apply to the archive itself before it is
5612 created or rewritten, as well as individual extracted members. Other
5613 @acronym{GNU} programs (@command{cp}, @command{install}, @command{ln},
5614 and @command{mv}, for example) offer similar options.
5616 Backup options may prove unexpectedly useful when extracting archives
5617 containing many members having identical name, or when extracting archives
5618 on systems having file name limitations, making different members appear
5619 as having similar names through the side-effect of name truncation.
5620 @FIXME{This is true only if we have a good scheme for truncated backup names,
5621 which I'm not sure at all: I suspect work is needed in this area.}
5622 When any existing file is backed up before being overwritten by extraction,
5623 then clashing files are automatically be renamed to be unique, and the
5624 true name is kept for only the last file of a series of clashing files.
5625 By using verbose mode, users may track exactly what happens.
5627 At the detail level, some decisions are still experimental, and may
5628 change in the future, we are waiting comments from our users. So, please
5629 do not learn to depend blindly on the details of the backup features.
5630 For example, currently, directories themselves are never renamed through
5631 using these options, so, extracting a file over a directory still has
5632 good chances to fail. Also, backup options apply to created archives,
5633 not only to extracted members. For created archives, backups will not
5634 be attempted when the archive is a block or character device, or when it
5635 refers to a remote file.
5637 For the sake of simplicity and efficiency, backups are made by renaming old
5638 files prior to creation or extraction, and not by copying. The original
5639 name is restored if the file creation fails. If a failure occurs after a
5640 partial extraction of a file, both the backup and the partially extracted
5644 @item --backup[=@var{method}]
5646 @vindex VERSION_CONTROL
5648 Back up files that are about to be overwritten or removed.
5649 Without this option, the original versions are destroyed.
5651 Use @var{method} to determine the type of backups made.
5652 If @var{method} is not specified, use the value of the @env{VERSION_CONTROL}
5653 environment variable. And if @env{VERSION_CONTROL} is not set,
5654 use the @samp{existing} method.
5656 @vindex version-control @r{Emacs variable}
5657 This option corresponds to the Emacs variable @samp{version-control};
5658 the same values for @var{method} are accepted as in Emacs. This option
5659 also allows more descriptive names. The valid @var{method}s are:
5664 @cindex numbered @r{backup method}
5665 Always make numbered backups.
5669 @cindex existing @r{backup method}
5670 Make numbered backups of files that already have them, simple backups
5675 @cindex simple @r{backup method}
5676 Always make simple backups.
5680 @item --suffix=@var{suffix}
5682 @cindex backup suffix
5683 @vindex SIMPLE_BACKUP_SUFFIX
5684 Append @var{suffix} to each backup file made with @option{--backup}. If this
5685 option is not specified, the value of the @env{SIMPLE_BACKUP_SUFFIX}
5686 environment variable is used. And if @env{SIMPLE_BACKUP_SUFFIX} is not
5687 set, the default is @samp{~}, just as in Emacs.
5692 @section Notable @command{tar} Usages
5695 @FIXME{Using Unix file linking capability to recreate directory
5696 structures---linking files into one subdirectory and then
5697 @command{tar}ring that directory.}
5699 @FIXME{Nice hairy example using absolute-names, newer, etc.}
5702 You can easily use archive files to transport a group of files from
5703 one system to another: put all relevant files into an archive on one
5704 computer system, transfer the archive to another system, and extract
5705 the contents there. The basic transfer medium might be magnetic tape,
5706 Internet FTP, or even electronic mail (though you must encode the
5707 archive with @command{uuencode} in order to transport it properly by
5708 mail). Both machines do not have to use the same operating system, as
5709 long as they both support the @command{tar} program.
5711 For example, here is how you might copy a directory's contents from
5712 one disk to another, while preserving the dates, modes, owners and
5713 link-structure of all the files therein. In this case, the transfer
5714 medium is a @dfn{pipe}:
5717 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xf -)}
5721 You can avoid subshells by using @option{-C} option:
5724 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xf -}
5728 The command also works using long option forms:
5732 $ @kbd{(cd sourcedir; tar --create --file=- . ) \
5733 | (cd targetdir; tar --extract --file=-)}
5742 $ @kbd{tar --directory sourcedir --create --file=- . \
5743 | tar --directory targetdir --extract --file=-}
5748 This is one of the easiest methods to transfer a @command{tar} archive.
5751 @section Looking Ahead: The Rest of this Manual
5753 You have now seen how to use all eight of the operations available to
5754 @command{tar}, and a number of the possible options. The next chapter
5755 explains how to choose and change file and archive names, how to use
5756 files to store names of other files which you can then call as
5757 arguments to @command{tar} (this can help you save time if you expect to
5758 archive the same list of files a number of times), and so forth.
5759 @FIXME{in case it's not obvious, i'm making this up in some sense
5760 based on my limited memory of what the next chapter *really* does. i
5761 just wanted to flesh out this final section a little bit so i'd
5762 remember to stick it in here. :-)}
5764 If there are too many files to conveniently list on the command line,
5765 you can list the names in a file, and @command{tar} will read that file.
5768 There are various ways of causing @command{tar} to skip over some files,
5769 and not archive them. @xref{Choosing}.
5772 @chapter Performing Backups and Restoring Files
5775 @GNUTAR{} is distributed along with the scripts for performing backups
5776 and restores. Even if there is a good chance those scripts may be
5777 satisfying to you, they are not the only scripts or methods available for doing
5778 backups and restore. You may well create your own, or use more
5779 sophisticated packages dedicated to that purpose.
5781 Some users are enthusiastic about @code{Amanda} (The Advanced Maryland
5782 Automatic Network Disk Archiver), a backup system developed by James
5783 da Silva @file{jds@@cs.umd.edu} and available on many Unix systems.
5784 This is free software, and it is available from @uref{http://www.amanda.org}.
5788 Here is a possible plan for a future documentation about the backuping
5789 scripts which are provided within the @GNUTAR{}
5795 @item what are dumps
5796 @item different levels of dumps
5798 @item full dump = dump everything
5799 @item level 1, level 2 dumps etc
5800 A level @var{n} dump dumps everything changed since the last level
5803 @item how to use scripts for dumps (ie, the concept)
5805 @item scripts to run after editing backup specs (details)
5807 @item Backup Specs, what is it.
5809 @item how to customize
5810 @item actual text of script [/sp/dump/backup-specs]
5814 @item rsh doesn't work
5815 @item rtape isn't installed
5818 @item the @option{--incremental} option of tar
5821 @item write protection
5822 @item types of media, different sizes and types, useful for different things
5823 @item files and tape marks
5824 one tape mark between files, two at end.
5825 @item positioning the tape
5826 MT writes two at end of write,
5827 backspaces over one when writing again.
5833 This chapter documents both the provided shell scripts and @command{tar}
5834 options which are more specific to usage as a backup tool.
5836 To @dfn{back up} a file system means to create archives that contain
5837 all the files in that file system. Those archives can then be used to
5838 restore any or all of those files (for instance if a disk crashes or a
5839 file is accidentally deleted). File system @dfn{backups} are also
5843 * Full Dumps:: Using @command{tar} to Perform Full Dumps
5844 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
5845 * Backup Levels:: Levels of Backups
5846 * Backup Parameters:: Setting Parameters for Backups and Restoration
5847 * Scripted Backups:: Using the Backup Scripts
5848 * Scripted Restoration:: Using the Restore Script
5852 @section Using @command{tar} to Perform Full Dumps
5858 @cindex corrupted archives
5859 Full dumps should only be made when no other people or programs
5860 are modifying files in the file system. If files are modified while
5861 @command{tar} is making the backup, they may not be stored properly in
5862 the archive, in which case you won't be able to restore them if you
5863 have to. (Files not being modified are written with no trouble, and do
5864 not corrupt the entire archive.)
5866 You will want to use the @option{--label=@var{archive-label}}
5867 (@option{-V @var{archive-label}}) option to give the archive a
5868 volume label, so you can tell what this archive is even if the label
5869 falls off the tape, or anything like that.
5871 Unless the file system you are dumping is guaranteed to fit on
5872 one volume, you will need to use the @option{--multi-volume} (@option{-M}) option.
5873 Make sure you have enough tapes on hand to complete the backup.
5875 If you want to dump each file system separately you will need to use
5876 the @option{--one-file-system} option to prevent
5877 @command{tar} from crossing file system boundaries when storing
5880 The @option{--incremental} (@option{-G}) (@pxref{Incremental Dumps})
5881 option is not needed, since this is a complete copy of everything in
5882 the file system, and a full restore from this backup would only be
5883 done onto a completely
5886 Unless you are in a hurry, and trust the @command{tar} program (and your
5887 tapes), it is a good idea to use the @option{--verify} (@option{-W})
5888 option, to make sure your files really made it onto the dump properly.
5889 This will also detect cases where the file was modified while (or just
5890 after) it was being archived. Not all media (notably cartridge tapes)
5891 are capable of being verified, unfortunately.
5893 @node Incremental Dumps
5894 @section Using @command{tar} to Perform Incremental Dumps
5896 @dfn{Incremental backup} is a special form of @GNUTAR{} archive that
5897 stores additional metadata so that exact state of the file system
5898 can be restored when extracting the archive.
5900 @GNUTAR{} currently offers two options for handling incremental
5901 backups: @option{--listed-incremental=@var{snapshot-file}} (@option{-g
5902 @var{snapshot-file}}) and @option{--incremental} (@option{-G}).
5904 @xopindex{listed-incremental, described}
5905 The option @option{--listed-incremental} instructs tar to operate on
5906 an incremental archive with additional metadata stored in a standalone
5907 file, called a @dfn{snapshot file}. The purpose of this file is to help
5908 determine which files have been changed, added or deleted since the
5909 last backup, so that the next incremental backup will contain only
5910 modified files. The name of the snapshot file is given as an argument
5914 @item --listed-incremental=@var{file}
5915 @itemx -g @var{file}
5916 Handle incremental backups with snapshot data in @var{file}.
5919 To create an incremental backup, you would use
5920 @option{--listed-incremental} together with @option{--create}
5921 (@pxref{create}). For example:
5924 $ @kbd{tar --create \
5925 --file=archive.1.tar \
5926 --listed-incremental=/var/log/usr.snar \
5930 This will create in @file{archive.1.tar} an incremental backup of
5931 the @file{/usr} file system, storing additional metadata in the file
5932 @file{/var/log/usr.snar}. If this file does not exist, it will be
5933 created. The created archive will then be a @dfn{level 0 backup};
5934 please see the next section for more on backup levels.
5936 Otherwise, if the file @file{/var/log/usr.snar} exists, it
5937 determines which files are modified. In this case only these files will be
5938 stored in the archive. Suppose, for example, that after running the
5939 above command, you delete file @file{/usr/doc/old} and create
5940 directory @file{/usr/local/db} with the following contents:
5943 $ @kbd{ls /usr/local/db}
5948 Some time later you create another incremental backup. You will
5952 $ @kbd{tar --create \
5953 --file=archive.2.tar \
5954 --listed-incremental=/var/log/usr.snar \
5956 tar: usr/local/db: Directory is new
5963 The created archive @file{archive.2.tar} will contain only these
5964 three members. This archive is called a @dfn{level 1 backup}. Notice
5965 that @file{/var/log/usr.snar} will be updated with the new data, so if
5966 you plan to create more @samp{level 1} backups, it is necessary to
5967 create a working copy of the snapshot file before running
5968 @command{tar}. The above example will then be modified as follows:
5971 $ @kbd{cp /var/log/usr.snar /var/log/usr.snar-1}
5972 $ @kbd{tar --create \
5973 --file=archive.2.tar \
5974 --listed-incremental=/var/log/usr.snar-1 \
5979 @xopindex{level, described}
5980 You can force @samp{level 0} backups either by removing the snapshot
5981 file before running @command{tar}, or by supplying the
5982 @option{--level=0} option, e.g.:
5985 $ @kbd{tar --create \
5986 --file=archive.2.tar \
5987 --listed-incremental=/var/log/usr.snar-0 \
5992 Incremental dumps depend crucially on time stamps, so the results are
5993 unreliable if you modify a file's time stamps during dumping (e.g.,
5994 with the @option{--atime-preserve=replace} option), or if you set the clock
5997 @anchor{device numbers}
5998 @cindex Device numbers, using in incremental backups
5999 Metadata stored in snapshot files include device numbers, which,
6000 obviously are supposed to be non-volatile values. However, it turns
6001 out that @acronym{NFS} devices have undependable values when an automounter
6002 gets in the picture. This can lead to a great deal of spurious
6003 redumping in incremental dumps, so it is somewhat useless to compare
6004 two @acronym{NFS} devices numbers over time. The solution implemented
6005 currently is to consider all @acronym{NFS} devices as being equal
6006 when it comes to comparing directories; this is fairly gross, but
6007 there does not seem to be a better way to go.
6009 Apart from using @acronym{NFS}, there are a number of cases where
6010 relying on device numbers can cause spurious redumping of unmodified
6011 files. For example, this occurs when archiving @acronym{LVM} snapshot
6012 volumes. To avoid this, use @option{--no-check-device} option:
6015 @xopindex{no-check-device, described}
6016 @item --no-check-device
6017 Do not rely on device numbers when preparing a list of changed files
6018 for an incremental dump.
6020 @xopindex{check-device, described}
6021 @item --check-device
6022 Use device numbers when preparing a list of changed files
6023 for an incremental dump. This is the default behavior. The purpose
6024 of this option is to undo the effect of the @option{--no-check-device}
6025 if it was given in @env{TAR_OPTIONS} environment variable
6026 (@pxref{TAR_OPTIONS}).
6029 There is also another way to cope with changing device numbers. It is
6030 described in detail in @ref{Fixing Snapshot Files}.
6032 Note that incremental archives use @command{tar} extensions and may
6033 not be readable by non-@acronym{GNU} versions of the @command{tar} program.
6035 @xopindex{listed-incremental, using with @option{--extract}}
6036 @xopindex{extract, using with @option{--listed-incremental}}
6037 To extract from the incremental dumps, use
6038 @option{--listed-incremental} together with @option{--extract}
6039 option (@pxref{extracting files}). In this case, @command{tar} does
6040 not need to access snapshot file, since all the data necessary for
6041 extraction are stored in the archive itself. So, when extracting, you
6042 can give whatever argument to @option{--listed-incremental}, the usual
6043 practice is to use @option{--listed-incremental=/dev/null}.
6044 Alternatively, you can use @option{--incremental}, which needs no
6045 arguments. In general, @option{--incremental} (@option{-G}) can be
6046 used as a shortcut for @option{--listed-incremental} when listing or
6047 extracting incremental backups (for more information regarding this
6048 option, @pxref{incremental-op}).
6050 When extracting from the incremental backup @GNUTAR{} attempts to
6051 restore the exact state the file system had when the archive was
6052 created. In particular, it will @emph{delete} those files in the file
6053 system that did not exist in their directories when the archive was
6054 created. If you have created several levels of incremental files,
6055 then in order to restore the exact contents the file system had when
6056 the last level was created, you will need to restore from all backups
6057 in turn. Continuing our example, to restore the state of @file{/usr}
6058 file system, one would do@footnote{Notice, that since both archives
6059 were created without @option{-P} option (@pxref{absolute}), these
6060 commands should be run from the root file system.}:
6063 $ @kbd{tar --extract \
6064 --listed-incremental=/dev/null \
6065 --file archive.1.tar}
6066 $ @kbd{tar --extract \
6067 --listed-incremental=/dev/null \
6068 --file archive.2.tar}
6071 To list the contents of an incremental archive, use @option{--list}
6072 (@pxref{list}), as usual. To obtain more information about the
6073 archive, use @option{--listed-incremental} or @option{--incremental}
6074 combined with two @option{--verbose} options@footnote{Two
6075 @option{--verbose} options were selected to avoid breaking usual
6076 verbose listing output (@option{--list --verbose}) when using in
6079 @xopindex{incremental, using with @option{--list}}
6080 @xopindex{listed-incremental, using with @option{--list}}
6081 @xopindex{list, using with @option{--incremental}}
6082 @xopindex{list, using with @option{--listed-incremental}}
6083 Versions of @GNUTAR{} up to 1.15.1 used to dump verbatim binary
6084 contents of the DUMPDIR header (with terminating nulls) when
6085 @option{--incremental} or @option{--listed-incremental} option was
6086 given, no matter what the verbosity level. This behavior, and,
6087 especially, the binary output it produced were considered inconvenient
6088 and were changed in version 1.16.}:
6091 @kbd{tar --list --incremental --verbose --verbose archive.tar}
6094 This command will print, for each directory in the archive, the list
6095 of files in that directory at the time the archive was created. This
6096 information is put out in a format which is both human-readable and
6097 unambiguous for a program: each file name is printed as
6104 where @var{x} is a letter describing the status of the file: @samp{Y}
6105 if the file is present in the archive, @samp{N} if the file is not
6106 included in the archive, or a @samp{D} if the file is a directory (and
6107 is included in the archive). @xref{Dumpdir}, for the detailed
6108 description of dumpdirs and status codes. Each such
6109 line is terminated by a newline character. The last line is followed
6110 by an additional newline to indicate the end of the data.
6112 @anchor{incremental-op}The option @option{--incremental} (@option{-G})
6113 gives the same behavior as @option{--listed-incremental} when used
6114 with @option{--list} and @option{--extract} options. When used with
6115 @option{--create} option, it creates an incremental archive without
6116 creating snapshot file. Thus, it is impossible to create several
6117 levels of incremental backups with @option{--incremental} option.
6120 @section Levels of Backups
6122 An archive containing all the files in the file system is called a
6123 @dfn{full backup} or @dfn{full dump}. You could insure your data by
6124 creating a full dump every day. This strategy, however, would waste a
6125 substantial amount of archive media and user time, as unchanged files
6126 are daily re-archived.
6128 It is more efficient to do a full dump only occasionally. To back up
6129 files between full dumps, you can use @dfn{incremental dumps}. A @dfn{level
6130 one} dump archives all the files that have changed since the last full
6133 A typical dump strategy would be to perform a full dump once a week,
6134 and a level one dump once a day. This means some versions of files
6135 will in fact be archived more than once, but this dump strategy makes
6136 it possible to restore a file system to within one day of accuracy by
6137 only extracting two archives---the last weekly (full) dump and the
6138 last daily (level one) dump. The only information lost would be in
6139 files changed or created since the last daily backup. (Doing dumps
6140 more than once a day is usually not worth the trouble.)
6142 @GNUTAR{} comes with scripts you can use to do full
6143 and level-one (actually, even level-two and so on) dumps. Using
6144 scripts (shell programs) to perform backups and restoration is a
6145 convenient and reliable alternative to typing out file name lists
6146 and @command{tar} commands by hand.
6148 Before you use these scripts, you need to edit the file
6149 @file{backup-specs}, which specifies parameters used by the backup
6150 scripts and by the restore script. This file is usually located
6151 in @file{/etc/backup} directory. @xref{Backup Parameters}, for its
6152 detailed description. Once the backup parameters are set, you can
6153 perform backups or restoration by running the appropriate script.
6155 The name of the backup script is @code{backup}. The name of the
6156 restore script is @code{restore}. The following sections describe
6157 their use in detail.
6159 @emph{Please Note:} The backup and restoration scripts are
6160 designed to be used together. While it is possible to restore files by
6161 hand from an archive which was created using a backup script, and to create
6162 an archive by hand which could then be extracted using the restore script,
6163 it is easier to use the scripts. @xref{Incremental Dumps}, before
6164 making such an attempt.
6166 @node Backup Parameters
6167 @section Setting Parameters for Backups and Restoration
6169 The file @file{backup-specs} specifies backup parameters for the
6170 backup and restoration scripts provided with @command{tar}. You must
6171 edit @file{backup-specs} to fit your system configuration and schedule
6172 before using these scripts.
6174 Syntactically, @file{backup-specs} is a shell script, containing
6175 mainly variable assignments. However, any valid shell construct
6176 is allowed in this file. Particularly, you may wish to define
6177 functions within that script (e.g., see @code{RESTORE_BEGIN} below).
6178 For more information about shell script syntax, please refer to
6179 @url{http://www.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html#ta
6180 g_02, the definition of the Shell Command Language}. See also
6181 @ref{Top,,Bash Features,bashref,Bash Reference Manual}.
6183 The shell variables controlling behavior of @code{backup} and
6184 @code{restore} are described in the following subsections.
6187 * General-Purpose Variables::
6188 * Magnetic Tape Control::
6190 * backup-specs example:: An Example Text of @file{Backup-specs}
6193 @node General-Purpose Variables
6194 @subsection General-Purpose Variables
6196 @defvr {Backup variable} ADMINISTRATOR
6197 The user name of the backup administrator. @code{Backup} scripts
6198 sends a backup report to this address.
6201 @defvr {Backup variable} BACKUP_HOUR
6202 The hour at which the backups are done. This can be a number from 0
6203 to 23, or the time specification in form @var{hours}:@var{minutes},
6204 or the string @samp{now}.
6206 This variable is used by @code{backup}. Its value may be overridden
6207 using @option{--time} option (@pxref{Scripted Backups}).
6210 @defvr {Backup variable} TAPE_FILE
6212 The device @command{tar} writes the archive to. If @var{TAPE_FILE}
6213 is a remote archive (@pxref{remote-dev}), backup script will suppose
6214 that your @command{mt} is able to access remote devices. If @var{RSH}
6215 (@pxref{RSH}) is set, @option{--rsh-command} option will be added to
6216 invocations of @command{mt}.
6219 @defvr {Backup variable} BLOCKING
6221 The blocking factor @command{tar} will use when writing the dump archive.
6222 @xref{Blocking Factor}.
6225 @defvr {Backup variable} BACKUP_DIRS
6227 A list of file systems to be dumped (for @code{backup}), or restored
6228 (for @code{restore}). You can include any directory
6229 name in the list --- subdirectories on that file system will be
6230 included, regardless of how they may look to other networked machines.
6231 Subdirectories on other file systems will be ignored.
6233 The host name specifies which host to run @command{tar} on, and should
6234 normally be the host that actually contains the file system. However,
6235 the host machine must have @GNUTAR{} installed, and
6236 must be able to access the directory containing the backup scripts and
6237 their support files using the same file name that is used on the
6238 machine where the scripts are run (i.e., what @command{pwd} will print
6239 when in that directory on that machine). If the host that contains
6240 the file system does not have this capability, you can specify another
6241 host as long as it can access the file system through @acronym{NFS}.
6243 If the list of file systems is very long you may wish to put it
6244 in a separate file. This file is usually named
6245 @file{/etc/backup/dirs}, but this name may be overridden in
6246 @file{backup-specs} using @code{DIRLIST} variable.
6249 @defvr {Backup variable} DIRLIST
6251 The name of the file that contains a list of file systems to backup
6252 or restore. By default it is @file{/etc/backup/dirs}.
6255 @defvr {Backup variable} BACKUP_FILES
6257 A list of individual files to be dumped (for @code{backup}), or restored
6258 (for @code{restore}). These should be accessible from the machine on
6259 which the backup script is run.
6261 If the list of individual files is very long you may wish to store it
6262 in a separate file. This file is usually named
6263 @file{/etc/backup/files}, but this name may be overridden in
6264 @file{backup-specs} using @code{FILELIST} variable.
6267 @defvr {Backup variable} FILELIST
6269 The name of the file that contains a list of individual files to backup
6270 or restore. By default it is @file{/etc/backup/files}.
6273 @defvr {Backup variable} MT
6275 Full file name of @command{mt} binary.
6278 @defvr {Backup variable} RSH
6280 Full file name of @command{rsh} binary or its equivalent. You may wish to
6281 set it to @code{ssh}, to improve security. In this case you will have
6282 to use public key authentication.
6285 @defvr {Backup variable} RSH_COMMAND
6287 Full file name of @command{rsh} binary on remote machines. This will
6288 be passed via @option{--rsh-command} option to the remote invocation
6292 @defvr {Backup variable} VOLNO_FILE
6294 Name of temporary file to hold volume numbers. This needs to be accessible
6295 by all the machines which have file systems to be dumped.
6298 @defvr {Backup variable} XLIST
6300 Name of @dfn{exclude file list}. An @dfn{exclude file list} is a file
6301 located on the remote machine and containing the list of files to
6302 be excluded from the backup. Exclude file lists are searched in
6303 /etc/tar-backup directory. A common use for exclude file lists
6304 is to exclude files containing security-sensitive information
6305 (e.g., @file{/etc/shadow} from backups).
6307 This variable affects only @code{backup}.
6310 @defvr {Backup variable} SLEEP_TIME
6312 Time to sleep between dumps of any two successive file systems
6314 This variable affects only @code{backup}.
6317 @defvr {Backup variable} DUMP_REMIND_SCRIPT
6319 Script to be run when it's time to insert a new tape in for the next
6320 volume. Administrators may want to tailor this script for their site.
6321 If this variable isn't set, @GNUTAR{} will display its built-in
6322 prompt, and will expect confirmation from the console. For the
6323 description of the default prompt, see @ref{change volume prompt}.
6327 @defvr {Backup variable} SLEEP_MESSAGE
6329 Message to display on the terminal while waiting for dump time. Usually
6330 this will just be some literal text.
6333 @defvr {Backup variable} TAR
6335 Full file name of the @GNUTAR{} executable. If this is not set, backup
6336 scripts will search @command{tar} in the current shell path.
6339 @node Magnetic Tape Control
6340 @subsection Magnetic Tape Control
6342 Backup scripts access tape device using special @dfn{hook functions}.
6343 These functions take a single argument --- the name of the tape
6344 device. Their names are kept in the following variables:
6346 @defvr {Backup variable} MT_BEGIN
6347 The name of @dfn{begin} function. This function is called before
6348 accessing the drive. By default it retensions the tape:
6354 mt -f "$1" retension
6359 @defvr {Backup variable} MT_REWIND
6360 The name of @dfn{rewind} function. The default definition is as
6373 @defvr {Backup variable} MT_OFFLINE
6374 The name of the function switching the tape off line. By default
6375 it is defined as follows:
6378 MT_OFFLINE=mt_offline
6386 @defvr {Backup variable} MT_STATUS
6387 The name of the function used to obtain the status of the archive device,
6388 including error count. Default definition:
6400 @subsection User Hooks
6402 @dfn{User hooks} are shell functions executed before and after
6403 each @command{tar} invocation. Thus, there are @dfn{backup
6404 hooks}, which are executed before and after dumping each file
6405 system, and @dfn{restore hooks}, executed before and
6406 after restoring a file system. Each user hook is a shell function
6407 taking four arguments:
6409 @deffn {User Hook Function} hook @var{level} @var{host} @var{fs} @var{fsname}
6414 Current backup or restore level.
6417 Name or IP address of the host machine being dumped or restored.
6420 Full file name of the file system being dumped or restored.
6423 File system name with directory separators replaced with colons. This
6424 is useful, e.g., for creating unique files.
6428 Following variables keep the names of user hook functions:
6430 @defvr {Backup variable} DUMP_BEGIN
6431 Dump begin function. It is executed before dumping the file system.
6434 @defvr {Backup variable} DUMP_END
6435 Executed after dumping the file system.
6438 @defvr {Backup variable} RESTORE_BEGIN
6439 Executed before restoring the file system.
6442 @defvr {Backup variable} RESTORE_END
6443 Executed after restoring the file system.
6446 @node backup-specs example
6447 @subsection An Example Text of @file{Backup-specs}
6449 The following is an example of @file{backup-specs}:
6452 # site-specific parameters for file system backup.
6454 ADMINISTRATOR=friedman
6456 TAPE_FILE=/dev/nrsmt0
6458 # Use @code{ssh} instead of the less secure @code{rsh}
6460 RSH_COMMAND=/usr/bin/ssh
6462 # Override MT_STATUS function:
6468 # Disable MT_OFFLINE function
6485 apple-gunkies:/com/mailer/gnu
6486 apple-gunkies:/com/archive/gnu"
6488 BACKUP_FILES="/com/mailer/aliases /com/mailer/league*[a-z]"
6492 @node Scripted Backups
6493 @section Using the Backup Scripts
6495 The syntax for running a backup script is:
6498 backup --level=@var{level} --time=@var{time}
6501 The @option{--level} option requests the dump level. Thus, to produce
6502 a full dump, specify @code{--level=0} (this is the default, so
6503 @option{--level} may be omitted if its value is
6504 @code{0})@footnote{For backward compatibility, the @code{backup} will also
6505 try to deduce the requested dump level from the name of the
6506 script itself. If the name consists of a string @samp{level-}
6507 followed by a single decimal digit, that digit is taken as
6508 the dump level number. Thus, you may create a link from @code{backup}
6509 to @code{level-1} and then run @code{level-1} whenever you need to
6510 create a level one dump.}.
6512 The @option{--time} option determines when should the backup be
6513 run. @var{Time} may take three forms:
6516 @item @var{hh}:@var{mm}
6518 The dump must be run at @var{hh} hours @var{mm} minutes.
6522 The dump must be run at @var{hh} hours.
6526 The dump must be run immediately.
6529 You should start a script with a tape or disk mounted. Once you
6530 start a script, it prompts you for new tapes or disks as it
6531 needs them. Media volumes don't have to correspond to archive
6532 files --- a multi-volume archive can be started in the middle of a
6533 tape that already contains the end of another multi-volume archive.
6534 The @code{restore} script prompts for media by its archive volume,
6535 so to avoid an error message you should keep track of which tape
6536 (or disk) contains which volume of the archive (@pxref{Scripted
6539 The backup scripts write two files on the file system. The first is a
6540 record file in @file{/etc/tar-backup/}, which is used by the scripts
6541 to store and retrieve information about which files were dumped. This
6542 file is not meant to be read by humans, and should not be deleted by
6543 them. @xref{Snapshot Files}, for a more detailed explanation of this
6546 The second file is a log file containing the names of the file systems
6547 and files dumped, what time the backup was made, and any error
6548 messages that were generated, as well as how much space was left in
6549 the media volume after the last volume of the archive was written.
6550 You should check this log file after every backup. The file name is
6551 @file{log-@var{mm-dd-yyyy}-level-@var{n}}, where @var{mm-dd-yyyy}
6552 represents current date, and @var{n} represents current dump level number.
6554 The script also prints the name of each system being dumped to the
6557 Following is the full list of options accepted by @code{backup}
6561 @item -l @var{level}
6562 @itemx --level=@var{level}
6563 Do backup level @var{level} (default 0).
6567 Force backup even if today's log file already exists.
6569 @item -v[@var{level}]
6570 @itemx --verbose[=@var{level}]
6571 Set verbosity level. The higher the level is, the more debugging
6572 information will be output during execution. Default @var{level}
6573 is 100, which means the highest debugging level.
6575 @item -t @var{start-time}
6576 @itemx --time=@var{start-time}
6577 Wait till @var{time}, then do backup.
6581 Display short help message and exit.
6585 Display information about the program's name, version, origin and legal
6586 status, all on standard output, and then exit successfully.
6590 @node Scripted Restoration
6591 @section Using the Restore Script
6593 To restore files that were archived using a scripted backup, use the
6594 @code{restore} script. Its usage is quite straightforward. In the
6595 simplest form, invoke @code{restore --all}, it will
6596 then restore all the file systems and files specified in
6597 @file{backup-specs} (@pxref{General-Purpose Variables,BACKUP_DIRS}).
6599 You may select the file systems (and/or files) to restore by
6600 giving @code{restore} a list of @dfn{patterns} in its command
6601 line. For example, running
6608 will restore all file systems on the machine @samp{albert}. A more
6609 complicated example:
6612 restore 'albert:*' '*:/var'
6616 This command will restore all file systems on the machine @samp{albert}
6617 as well as @file{/var} file system on all machines.
6619 By default @code{restore} will start restoring files from the lowest
6620 available dump level (usually zero) and will continue through
6621 all available dump levels. There may be situations where such a
6622 thorough restore is not necessary. For example, you may wish to
6623 restore only files from the recent level one backup. To do so,
6624 use @option{--level} option, as shown in the example below:
6630 The full list of options accepted by @code{restore} follows:
6635 Restore all file systems and files specified in @file{backup-specs}.
6637 @item -l @var{level}
6638 @itemx --level=@var{level}
6639 Start restoring from the given backup level, instead of the default 0.
6641 @item -v[@var{level}]
6642 @itemx --verbose[=@var{level}]
6643 Set verbosity level. The higher the level is, the more debugging
6644 information will be output during execution. Default @var{level}
6645 is 100, which means the highest debugging level.
6649 Display short help message and exit.
6653 Display information about the program's name, version, origin and legal
6654 status, all on standard output, and then exit successfully.
6657 You should start the restore script with the media containing the
6658 first volume of the archive mounted. The script will prompt for other
6659 volumes as they are needed. If the archive is on tape, you don't need
6660 to rewind the tape to to its beginning---if the tape head is
6661 positioned past the beginning of the archive, the script will rewind
6662 the tape as needed. @xref{Tape Positioning}, for a discussion of tape
6666 @strong{Warning:} The script will delete files from the active file
6667 system if they were not in the file system when the archive was made.
6670 @xref{Incremental Dumps}, for an explanation of how the script makes
6674 @chapter Choosing Files and Names for @command{tar}
6676 Certain options to @command{tar} enable you to specify a name for your
6677 archive. Other options let you decide which files to include or exclude
6678 from the archive, based on when or whether files were modified, whether
6679 the file names do or don't match specified patterns, or whether files
6680 are in specified directories.
6682 This chapter discusses these options in detail.
6685 * file:: Choosing the Archive's Name
6686 * Selecting Archive Members::
6687 * files:: Reading Names from a File
6688 * exclude:: Excluding Some Files
6689 * wildcards:: Wildcards Patterns and Matching
6690 * quoting styles:: Ways of Quoting Special Characters in Names
6691 * transform:: Modifying File and Member Names
6692 * after:: Operating Only on New Files
6693 * recurse:: Descending into Directories
6694 * one:: Crossing File System Boundaries
6698 @section Choosing and Naming Archive Files
6700 @cindex Naming an archive
6701 @cindex Archive Name
6702 @cindex Choosing an archive file
6703 @cindex Where is the archive?
6705 By default, @command{tar} uses an archive file name that was compiled when
6706 it was built on the system; usually this name refers to some physical
6707 tape drive on the machine. However, the person who installed @command{tar}
6708 on the system may not have set the default to a meaningful value as far as
6709 most users are concerned. As a result, you will usually want to tell
6710 @command{tar} where to find (or create) the archive. The
6711 @option{--file=@var{archive-name}} (@option{-f @var{archive-name}})
6712 option allows you to either specify or name a file to use as the archive
6713 instead of the default archive file location.
6716 @xopindex{file, short description}
6717 @item --file=@var{archive-name}
6718 @itemx -f @var{archive-name}
6719 Name the archive to create or operate on. Use in conjunction with
6723 For example, in this @command{tar} command,
6726 $ @kbd{tar -cvf collection.tar blues folk jazz}
6730 @file{collection.tar} is the name of the archive. It must directly
6731 follow the @option{-f} option, since whatever directly follows @option{-f}
6732 @emph{will} end up naming the archive. If you neglect to specify an
6733 archive name, you may end up overwriting a file in the working directory
6734 with the archive you create since @command{tar} will use this file's name
6735 for the archive name.
6737 An archive can be saved as a file in the file system, sent through a
6738 pipe or over a network, or written to an I/O device such as a tape,
6739 floppy disk, or CD write drive.
6741 @cindex Writing new archives
6742 @cindex Archive creation
6743 If you do not name the archive, @command{tar} uses the value of the
6744 environment variable @env{TAPE} as the file name for the archive. If
6745 that is not available, @command{tar} uses a default, compiled-in archive
6746 name, usually that for tape unit zero (i.e., @file{/dev/tu00}).
6748 @cindex Standard input and output
6749 @cindex tar to standard input and output
6750 If you use @file{-} as an @var{archive-name}, @command{tar} reads the
6751 archive from standard input (when listing or extracting files), or
6752 writes it to standard output (when creating an archive). If you use
6753 @file{-} as an @var{archive-name} when modifying an archive,
6754 @command{tar} reads the original archive from its standard input and
6755 writes the entire new archive to its standard output.
6757 The following example is a convenient way of copying directory
6758 hierarchy from @file{sourcedir} to @file{targetdir}.
6761 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xpf -)}
6764 The @option{-C} option allows to avoid using subshells:
6767 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xpf -}
6770 In both examples above, the leftmost @command{tar} invocation archives
6771 the contents of @file{sourcedir} to the standard output, while the
6772 rightmost one reads this archive from its standard input and
6773 extracts it. The @option{-p} option tells it to restore permissions
6774 of the extracted files.
6776 @cindex Remote devices
6777 @cindex tar to a remote device
6779 To specify an archive file on a device attached to a remote machine,
6783 @kbd{--file=@var{hostname}:/@var{dev}/@var{file-name}}
6787 @command{tar} will set up the remote connection, if possible, and
6788 prompt you for a username and password. If you use
6789 @option{--file=@@@var{hostname}:/@var{dev}/@var{file-name}}, @command{tar}
6790 will attempt to set up the remote connection using your username
6791 as the username on the remote machine.
6793 @cindex Local and remote archives
6794 @anchor{local and remote archives}
6795 If the archive file name includes a colon (@samp{:}), then it is assumed
6796 to be a file on another machine. If the archive file is
6797 @samp{@var{user}@@@var{host}:@var{file}}, then @var{file} is used on the
6798 host @var{host}. The remote host is accessed using the @command{rsh}
6799 program, with a username of @var{user}. If the username is omitted
6800 (along with the @samp{@@} sign), then your user name will be used.
6801 (This is the normal @command{rsh} behavior.) It is necessary for the
6802 remote machine, in addition to permitting your @command{rsh} access, to
6803 have the @file{rmt} program installed (this command is included in
6804 the @GNUTAR{} distribution and by default is installed under
6805 @file{@var{prefix}/libexec/rmt}, where @var{prefix} means your
6806 installation prefix). If you need to use a file whose name includes a
6807 colon, then the remote tape drive behavior
6808 can be inhibited by using the @option{--force-local} option.
6810 When the archive is being created to @file{/dev/null}, @GNUTAR{}
6811 tries to minimize input and output operations. The Amanda backup
6812 system, when used with @GNUTAR{}, has an initial sizing pass which
6815 @node Selecting Archive Members
6816 @section Selecting Archive Members
6817 @cindex Specifying files to act on
6818 @cindex Specifying archive members
6820 @dfn{File Name arguments} specify which files in the file system
6821 @command{tar} operates on, when creating or adding to an archive, or which
6822 archive members @command{tar} operates on, when reading or deleting from
6823 an archive. @xref{Operations}.
6825 To specify file names, you can include them as the last arguments on
6826 the command line, as follows:
6828 @kbd{tar} @var{operation} [@var{option1} @var{option2} @dots{}] [@var{file name-1} @var{file name-2} @dots{}]
6831 If a file name begins with dash (@samp{-}), precede it with
6832 @option{--add-file} option to prevent it from being treated as an
6835 @anchor{input name quoting}
6836 By default @GNUTAR{} attempts to @dfn{unquote} each file or member
6837 name, replacing @dfn{escape sequences} according to the following
6840 @multitable @columnfractions 0.20 0.60
6841 @headitem Escape @tab Replaced with
6842 @item \a @tab Audible bell (@acronym{ASCII} 7)
6843 @item \b @tab Backspace (@acronym{ASCII} 8)
6844 @item \f @tab Form feed (@acronym{ASCII} 12)
6845 @item \n @tab New line (@acronym{ASCII} 10)
6846 @item \r @tab Carriage return (@acronym{ASCII} 13)
6847 @item \t @tab Horizontal tabulation (@acronym{ASCII} 9)
6848 @item \v @tab Vertical tabulation (@acronym{ASCII} 11)
6849 @item \? @tab @acronym{ASCII} 127
6850 @item \@var{n} @tab @acronym{ASCII} @var{n} (@var{n} should be an octal number
6854 A backslash followed by any other symbol is retained.
6856 This default behavior is controlled by the following command line
6862 Enable unquoting input file or member names (default).
6866 Disable unquoting input file or member names.
6869 If you specify a directory name as a file name argument, all the files
6870 in that directory are operated on by @command{tar}.
6872 If you do not specify files, @command{tar} behavior differs depending
6873 on the operation mode as described below:
6875 When @command{tar} is invoked with @option{--create} (@option{-c}),
6876 @command{tar} will stop immediately, reporting the following:
6880 $ @kbd{tar cf a.tar}
6881 tar: Cowardly refusing to create an empty archive
6882 Try `tar --help' or `tar --usage' for more information.
6886 If you specify either @option{--list} (@option{-t}) or
6887 @option{--extract} (@option{--get}, @option{-x}), @command{tar}
6888 operates on all the archive members in the archive.
6890 If run with @option{--diff} option, tar will compare the archive with
6891 the contents of the current working directory.
6893 If you specify any other operation, @command{tar} does nothing.
6895 By default, @command{tar} takes file names from the command line. However,
6896 there are other ways to specify file or member names, or to modify the
6897 manner in which @command{tar} selects the files or members upon which to
6898 operate. In general, these methods work both for specifying the names
6899 of files and archive members.
6902 @section Reading Names from a File
6904 @cindex Reading file names from a file
6905 @cindex Lists of file names
6906 @cindex File Name arguments, alternatives
6907 @cindex @command{find}, using with @command{tar}
6908 Instead of giving the names of files or archive members on the command
6909 line, you can put the names into a file, and then use the
6910 @option{--files-from=@var{file-of-names}} (@option{-T
6911 @var{file-of-names}}) option to @command{tar}. Give the name of the
6912 file which contains the list of files to include as the argument to
6913 @option{--files-from}. In the list, the file names should be separated by
6914 newlines. You will frequently use this option when you have generated
6915 the list of files to archive with the @command{find} utility.
6919 @item --files-from=@var{file-name}
6920 @itemx -T @var{file-name}
6921 Get names to extract or create from file @var{file-name}.
6924 If you give a single dash as a file name for @option{--files-from}, (i.e.,
6925 you specify either @code{--files-from=-} or @code{-T -}), then the file
6926 names are read from standard input.
6928 Unless you are running @command{tar} with @option{--create}, you can not use
6929 both @code{--files-from=-} and @code{--file=-} (@code{-f -}) in the same
6932 Any number of @option{-T} options can be given in the command line.
6934 The following example shows how to use @command{find} to generate a list of
6935 files smaller than 400K in length and put that list into a file
6936 called @file{small-files}. You can then use the @option{-T} option to
6937 @command{tar} to specify the files from that file, @file{small-files}, to
6938 create the archive @file{little.tgz}. (The @option{-z} option to
6939 @command{tar} compresses the archive with @command{gzip}; @pxref{gzip} for
6943 $ @kbd{find . -size -400 -print > small-files}
6944 $ @kbd{tar -c -v -z -T small-files -f little.tgz}
6948 In the file list given by @option{-T} option, any file name beginning
6949 with @samp{-} character is considered a @command{tar} option and is
6950 processed accordingly@footnote{Versions of @GNUTAR{} up to 1.15.1
6951 recognized only @option{-C} option in file lists, and only if the
6952 option and its argument occupied two consecutive lines.}. For example,
6953 the common use of this feature is to change to another directory by
6954 specifying @option{-C} option:
6964 $ @kbd{tar -c -f foo.tar --files-from list}
6969 In this example, @command{tar} will first switch to @file{/etc}
6970 directory and add files @file{passwd} and @file{hosts} to the
6971 archive. Then it will change to @file{/lib} directory and will archive
6972 the file @file{libc.a}. Thus, the resulting archive @file{foo.tar} will
6977 $ @kbd{tar tf foo.tar}
6985 @xopindex{directory, using in @option{--files-from} argument}
6986 Notice that the option parsing algorithm used with @option{-T} is
6987 stricter than the one used by shell. Namely, when specifying option
6988 arguments, you should observe the following rules:
6992 When using short (single-letter) option form, its argument must
6993 immediately follow the option letter, without any intervening
6994 whitespace. For example: @code{-Cdir}.
6997 When using long option form, the option argument must be separated
6998 from the option by a single equal sign. No whitespace is allowed on
6999 any side of the equal sign. For example: @code{--directory=dir}.
7002 For both short and long option forms, the option argument can be given
7003 on the next line after the option name, e.g.:
7024 If you happen to have a file whose name starts with @samp{-},
7025 precede it with @option{--add-file} option to prevent it from
7026 being recognized as an option. For example: @code{--add-file=--my-file}.
7033 @subsection @code{NUL}-Terminated File Names
7035 @cindex File names, terminated by @code{NUL}
7036 @cindex @code{NUL}-terminated file names
7037 The @option{--null} option causes
7038 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}})
7039 to read file names terminated by a @code{NUL} instead of a newline, so
7040 files whose names contain newlines can be archived using
7041 @option{--files-from}.
7044 @xopindex{null, described}
7046 Only consider @code{NUL}-terminated file names, instead of files that
7047 terminate in a newline.
7049 @xopindex{no-null, described}
7051 Undo the effect of any previous @option{--null} option.
7054 The @option{--null} option is just like the one in @acronym{GNU}
7055 @command{xargs} and @command{cpio}, and is useful with the
7056 @option{-print0} predicate of @acronym{GNU} @command{find}. In
7057 @command{tar}, @option{--null} also disables special handling for
7058 file names that begin with dash.
7060 This example shows how to use @command{find} to generate a list of files
7061 larger than 800K in length and put that list into a file called
7062 @file{long-files}. The @option{-print0} option to @command{find} is just
7063 like @option{-print}, except that it separates files with a @code{NUL}
7064 rather than with a newline. You can then run @command{tar} with both the
7065 @option{--null} and @option{-T} options to specify that @command{tar} gets the
7066 files from that file, @file{long-files}, to create the archive
7067 @file{big.tgz}. The @option{--null} option to @command{tar} will cause
7068 @command{tar} to recognize the @code{NUL} separator between files.
7071 $ @kbd{find . -size +800 -print0 > long-files}
7072 $ @kbd{tar -c -v --null --files-from=long-files --file=big.tar}
7075 The @option{--no-null} option can be used if you need to read both
7076 @code{NUL}-terminated and newline-terminated files on the same command line.
7077 For example, if @file{flist} is a newline-terminated file, then the
7078 following command can be used to combine it with the above command:
7082 $ @kbd{find . -size +800 -print0 |
7083 tar -c -f big.tar --null -T - --no-null -T flist}
7087 This example uses short options for typographic reasons, to avoid
7090 @GNUTAR is able to automatically detect @code{NUL}-terminated file lists, so
7091 it is safe to use them even without the @option{--null} option. In
7092 this case @command{tar} will print a warning and continue reading such
7093 a file as if @option{--null} were actually given:
7097 $ @kbd{find . -size +800 -print0 | tar -c -f big.tar -T -}
7098 tar: -: file name read contains nul character
7102 The null terminator, however, remains in effect only for this
7103 particular file, any following @option{-T} options will assume
7104 newline termination. Of course, the null autodetection applies
7105 to these eventual surplus @option{-T} options as well.
7108 @section Excluding Some Files
7110 @cindex File names, excluding files by
7111 @cindex Excluding files by name and pattern
7112 @cindex Excluding files by file system
7114 @opindex exclude-from
7115 To avoid operating on files whose names match a particular pattern,
7116 use the @option{--exclude} or @option{--exclude-from} options.
7120 @item --exclude=@var{pattern}
7121 Causes @command{tar} to ignore files that match the @var{pattern}.
7125 The @option{--exclude=@var{pattern}} option prevents any file or
7126 member whose name matches the shell wildcard (@var{pattern}) from
7128 For example, to create an archive with all the contents of the directory
7129 @file{src} except for files whose names end in @file{.o}, use the
7130 command @samp{tar -cf src.tar --exclude='*.o' src}.
7132 You may give multiple @option{--exclude} options.
7135 @opindex exclude-from
7136 @item --exclude-from=@var{file}
7137 @itemx -X @var{file}
7138 Causes @command{tar} to ignore files that match the patterns listed in
7142 @findex exclude-from
7143 Use the @option{--exclude-from} option to read a
7144 list of patterns, one per line, from @var{file}; @command{tar} will
7145 ignore files matching those patterns. Thus if @command{tar} is
7146 called as @w{@samp{tar -c -X foo .}} and the file @file{foo} contains a
7147 single line @file{*.o}, no files whose names end in @file{.o} will be
7148 added to the archive.
7150 Notice, that lines from @var{file} are read verbatim. One of the
7151 frequent errors is leaving some extra whitespace after a file name,
7152 which is difficult to catch using text editors.
7154 However, empty lines are OK.
7157 @cindex version control system, excluding files
7158 @cindex VCS, excluding files
7159 @cindex SCCS, excluding files
7160 @cindex RCS, excluding files
7161 @cindex CVS, excluding files
7162 @cindex SVN, excluding files
7163 @cindex git, excluding files
7164 @cindex Bazaar, excluding files
7165 @cindex Arch, excluding files
7166 @cindex Mercurial, excluding files
7167 @cindex Darcs, excluding files
7168 @opindex exclude-vcs
7170 Exclude files and directories used by following version control
7171 systems: @samp{CVS}, @samp{RCS}, @samp{SCCS}, @samp{SVN}, @samp{Arch},
7172 @samp{Bazaar}, @samp{Mercurial}, and @samp{Darcs}.
7174 As of version @value{VERSION}, the following files are excluded:
7177 @item @file{CVS/}, and everything under it
7178 @item @file{RCS/}, and everything under it
7179 @item @file{SCCS/}, and everything under it
7180 @item @file{.git/}, and everything under it
7181 @item @file{.gitignore}
7182 @item @file{.cvsignore}
7183 @item @file{.svn/}, and everything under it
7184 @item @file{.arch-ids/}, and everything under it
7185 @item @file{@{arch@}/}, and everything under it
7186 @item @file{=RELEASE-ID}
7187 @item @file{=meta-update}
7188 @item @file{=update}
7190 @item @file{.bzrignore}
7191 @item @file{.bzrtags}
7193 @item @file{.hgignore}
7194 @item @file{.hgrags}
7198 @opindex exclude-backups
7199 @item --exclude-backups
7200 Exclude backup and lock files. This option causes exclusion of files
7201 that match the following shell globbing patterns:
7211 @findex exclude-caches
7212 When creating an archive, the @option{--exclude-caches} option family
7213 causes @command{tar} to exclude all directories that contain a @dfn{cache
7214 directory tag}. A cache directory tag is a short file with the
7215 well-known name @file{CACHEDIR.TAG} and having a standard header
7216 specified in @url{http://www.brynosaurus.com/cachedir/spec.html}.
7217 Various applications write cache directory tags into directories they
7218 use to hold regenerable, non-precious data, so that such data can be
7219 more easily excluded from backups.
7221 There are three @samp{exclude-caches} options, each providing a different
7222 exclusion semantics:
7225 @opindex exclude-caches
7226 @item --exclude-caches
7227 Do not archive the contents of the directory, but archive the
7228 directory itself and the @file{CACHEDIR.TAG} file.
7230 @opindex exclude-caches-under
7231 @item --exclude-caches-under
7232 Do not archive the contents of the directory, nor the
7233 @file{CACHEDIR.TAG} file, archive only the directory itself.
7235 @opindex exclude-caches-all
7236 @item --exclude-caches-all
7237 Omit directories containing @file{CACHEDIR.TAG} file entirely.
7241 Another option family, @option{--exclude-tag}, provides a generalization of
7242 this concept. It takes a single argument, a file name to look for.
7243 Any directory that contains this file will be excluded from the dump.
7244 Similarly to @samp{exclude-caches}, there are three options in this
7248 @opindex exclude-tag
7249 @item --exclude-tag=@var{file}
7250 Do not dump the contents of the directory, but dump the
7251 directory itself and the @var{file}.
7253 @opindex exclude-tag-under
7254 @item --exclude-tag-under=@var{file}
7255 Do not dump the contents of the directory, nor the
7256 @var{file}, archive only the directory itself.
7258 @opindex exclude-tag-all
7259 @item --exclude-tag-all=@var{file}
7260 Omit directories containing @var{file} file entirely.
7263 Multiple @option{--exclude-tag*} options can be given.
7265 For example, given this directory:
7280 The @option{--exclude-tag} will produce the following:
7283 $ @kbd{tar -cf archive.tar --exclude-tag=tagfile -v dir}
7288 tar: dir/folk/: contains a cache directory tag tagfile;
7293 Both the @file{dir/folk} directory and its tagfile are preserved in
7294 the archive, however the rest of files in this directory are not.
7296 Now, using the @option{--exclude-tag-under} option will exclude
7297 @file{tagfile} from the dump, while still preserving the directory
7298 itself, as shown in this example:
7301 $ @kbd{tar -cf archive.tar --exclude-tag-under=tagfile -v dir}
7306 ./tar: dir/folk/: contains a cache directory tag tagfile;
7310 Finally, using @option{--exclude-tag-all} omits the @file{dir/folk}
7314 $ @kbd{tar -cf archive.tar --exclude-tag-all=tagfile -v dir}
7318 ./tar: dir/folk/: contains a cache directory tag tagfile;
7319 directory not dumped
7323 * problems with exclude::
7326 @node problems with exclude
7327 @unnumberedsubsec Problems with Using the @code{exclude} Options
7329 @xopindex{exclude, potential problems with}
7330 Some users find @samp{exclude} options confusing. Here are some common
7335 The main operating mode of @command{tar} does not act on a file name
7336 explicitly listed on the command line, if one of its file name
7337 components is excluded. In the example above, if
7338 you create an archive and exclude files that end with @samp{*.o}, but
7339 explicitly name the file @samp{dir.o/foo} after all the options have been
7340 listed, @samp{dir.o/foo} will be excluded from the archive.
7343 You can sometimes confuse the meanings of @option{--exclude} and
7344 @option{--exclude-from}. Be careful: use @option{--exclude} when files
7345 to be excluded are given as a pattern on the command line. Use
7346 @option{--exclude-from} to introduce the name of a file which contains
7347 a list of patterns, one per line; each of these patterns can exclude
7348 zero, one, or many files.
7351 When you use @option{--exclude=@var{pattern}}, be sure to quote the
7352 @var{pattern} parameter, so @GNUTAR{} sees wildcard characters
7353 like @samp{*}. If you do not do this, the shell might expand the
7354 @samp{*} itself using files at hand, so @command{tar} might receive a
7355 list of files instead of one pattern, or none at all, making the
7356 command somewhat illegal. This might not correspond to what you want.
7361 $ @kbd{tar -c -f @var{archive.tar} --exclude '*.o' @var{directory}}
7369 $ @kbd{tar -c -f @var{archive.tar} --exclude *.o @var{directory}}
7373 You must use use shell syntax, or globbing, rather than @code{regexp}
7374 syntax, when using exclude options in @command{tar}. If you try to use
7375 @code{regexp} syntax to describe files to be excluded, your command
7379 @FIXME{The change in semantics must have occurred before 1.11,
7380 so I doubt if it is worth mentioning at all. Anyway, should at
7381 least specify in which version the semantics changed.}
7382 In earlier versions of @command{tar}, what is now the
7383 @option{--exclude-from} option was called @option{--exclude} instead.
7384 Now, @option{--exclude} applies to patterns listed on the command
7385 line and @option{--exclude-from} applies to patterns listed in a
7391 @section Wildcards Patterns and Matching
7393 @dfn{Globbing} is the operation by which @dfn{wildcard} characters,
7394 @samp{*} or @samp{?} for example, are replaced and expanded into all
7395 existing files matching the given pattern. @GNUTAR{} can use wildcard
7396 patterns for matching (or globbing) archive members when extracting
7397 from or listing an archive. Wildcard patterns are also used for
7398 verifying volume labels of @command{tar} archives. This section has the
7399 purpose of explaining wildcard syntax for @command{tar}.
7401 @FIXME{the next few paragraphs need work.}
7403 A @var{pattern} should be written according to shell syntax, using wildcard
7404 characters to effect globbing. Most characters in the pattern stand
7405 for themselves in the matched string, and case is significant: @samp{a}
7406 will match only @samp{a}, and not @samp{A}. The character @samp{?} in the
7407 pattern matches any single character in the matched string. The character
7408 @samp{*} in the pattern matches zero, one, or more single characters in
7409 the matched string. The character @samp{\} says to take the following
7410 character of the pattern @emph{literally}; it is useful when one needs to
7411 match the @samp{?}, @samp{*}, @samp{[} or @samp{\} characters, themselves.
7413 The character @samp{[}, up to the matching @samp{]}, introduces a character
7414 class. A @dfn{character class} is a list of acceptable characters
7415 for the next single character of the matched string. For example,
7416 @samp{[abcde]} would match any of the first five letters of the alphabet.
7417 Note that within a character class, all of the ``special characters''
7418 listed above other than @samp{\} lose their special meaning; for example,
7419 @samp{[-\\[*?]]} would match any of the characters, @samp{-}, @samp{\},
7420 @samp{[}, @samp{*}, @samp{?}, or @samp{]}. (Due to parsing constraints,
7421 the characters @samp{-} and @samp{]} must either come @emph{first} or
7422 @emph{last} in a character class.)
7424 @cindex Excluding characters from a character class
7425 @cindex Character class, excluding characters from
7426 If the first character of the class after the opening @samp{[}
7427 is @samp{!} or @samp{^}, then the meaning of the class is reversed.
7428 Rather than listing character to match, it lists those characters which
7429 are @emph{forbidden} as the next single character of the matched string.
7431 Other characters of the class stand for themselves. The special
7432 construction @samp{[@var{a}-@var{e}]}, using an hyphen between two
7433 letters, is meant to represent all characters between @var{a} and
7436 @FIXME{need to add a sentence or so here to make this clear for those
7437 who don't have dan around.}
7439 Periods (@samp{.}) or forward slashes (@samp{/}) are not considered
7440 special for wildcard matches. However, if a pattern completely matches
7441 a directory prefix of a matched string, then it matches the full matched
7442 string: thus, excluding a directory also excludes all the files beneath it.
7445 * controlling pattern-matching::
7448 @node controlling pattern-matching
7449 @unnumberedsubsec Controlling Pattern-Matching
7451 For the purposes of this section, we call @dfn{exclusion members} all
7452 member names obtained while processing @option{--exclude} and
7453 @option{--exclude-from} options, and @dfn{inclusion members} those
7454 member names that were given in the command line or read from the file
7455 specified with @option{--files-from} option.
7457 These two pairs of member lists are used in the following operations:
7458 @option{--diff}, @option{--extract}, @option{--list},
7461 There are no inclusion members in create mode (@option{--create} and
7462 @option{--append}), since in this mode the names obtained from the
7463 command line refer to @emph{files}, not archive members.
7465 By default, inclusion members are compared with archive members
7466 literally @footnote{Notice that earlier @GNUTAR{} versions used
7467 globbing for inclusion members, which contradicted to UNIX98
7468 specification and was not documented. @xref{Changes}, for more
7469 information on this and other changes.} and exclusion members are
7470 treated as globbing patterns. For example:
7474 $ @kbd{tar tf foo.tar}
7479 # @i{Member names are used verbatim:}
7480 $ @kbd{tar -xf foo.tar -v '[remarks]'}
7482 # @i{Exclude member names are globbed:}
7483 $ @kbd{tar -xf foo.tar -v --exclude '*.c'}
7489 This behavior can be altered by using the following options:
7494 Treat all member names as wildcards.
7496 @opindex no-wildcards
7497 @item --no-wildcards
7498 Treat all member names as literal strings.
7501 Thus, to extract files whose names end in @samp{.c}, you can use:
7504 $ @kbd{tar -xf foo.tar -v --wildcards '*.c'}
7510 Notice quoting of the pattern to prevent the shell from interpreting
7513 The effect of @option{--wildcards} option is canceled by
7514 @option{--no-wildcards}. This can be used to pass part of
7515 the command line arguments verbatim and other part as globbing
7516 patterns. For example, the following invocation:
7519 $ @kbd{tar -xf foo.tar --wildcards '*.txt' --no-wildcards '[remarks]'}
7523 instructs @command{tar} to extract from @file{foo.tar} all files whose
7524 names end in @samp{.txt} and the file named @file{[remarks]}.
7526 Normally, a pattern matches a name if an initial subsequence of the
7527 name's components matches the pattern, where @samp{*}, @samp{?}, and
7528 @samp{[...]} are the usual shell wildcards, @samp{\} escapes wildcards,
7529 and wildcards can match @samp{/}.
7531 Other than optionally stripping leading @samp{/} from names
7532 (@pxref{absolute}), patterns and names are used as-is. For
7533 example, trailing @samp{/} is not trimmed from a user-specified name
7534 before deciding whether to exclude it.
7536 However, this matching procedure can be altered by the options listed
7537 below. These options accumulate. For example:
7540 --ignore-case --exclude='makefile' --no-ignore-case ---exclude='readme'
7544 ignores case when excluding @samp{makefile}, but not when excluding
7549 @opindex no-anchored
7551 @itemx --no-anchored
7552 If anchored, a pattern must match an initial subsequence
7553 of the name's components. Otherwise, the pattern can match any
7554 subsequence. Default is @option{--no-anchored} for exclusion members
7555 and @option{--anchored} inclusion members.
7557 @opindex ignore-case
7558 @opindex no-ignore-case
7560 @itemx --no-ignore-case
7561 When ignoring case, upper-case patterns match lower-case names and vice versa.
7562 When not ignoring case (the default), matching is case-sensitive.
7564 @opindex wildcards-match-slash
7565 @opindex no-wildcards-match-slash
7566 @item --wildcards-match-slash
7567 @itemx --no-wildcards-match-slash
7568 When wildcards match slash (the default for exclusion members), a
7569 wildcard like @samp{*} in the pattern can match a @samp{/} in the
7570 name. Otherwise, @samp{/} is matched only by @samp{/}.
7574 The @option{--recursion} and @option{--no-recursion} options
7575 (@pxref{recurse}) also affect how member patterns are interpreted. If
7576 recursion is in effect, a pattern matches a name if it matches any of
7577 the name's parent directories.
7579 The following table summarizes pattern-matching default values:
7581 @multitable @columnfractions .3 .7
7582 @headitem Members @tab Default settings
7583 @item Inclusion @tab @option{--no-wildcards --anchored --no-wildcards-match-slash}
7584 @item Exclusion @tab @option{--wildcards --no-anchored --wildcards-match-slash}
7587 @node quoting styles
7588 @section Quoting Member Names
7590 When displaying member names, @command{tar} takes care to avoid
7591 ambiguities caused by certain characters. This is called @dfn{name
7592 quoting}. The characters in question are:
7595 @item Non-printable control characters:
7596 @anchor{escape sequences}
7597 @multitable @columnfractions 0.20 0.10 0.60
7598 @headitem Character @tab @acronym{ASCII} @tab Character name
7599 @item \a @tab 7 @tab Audible bell
7600 @item \b @tab 8 @tab Backspace
7601 @item \f @tab 12 @tab Form feed
7602 @item \n @tab 10 @tab New line
7603 @item \r @tab 13 @tab Carriage return
7604 @item \t @tab 9 @tab Horizontal tabulation
7605 @item \v @tab 11 @tab Vertical tabulation
7608 @item Space (@acronym{ASCII} 32)
7610 @item Single and double quotes (@samp{'} and @samp{"})
7612 @item Backslash (@samp{\})
7615 The exact way @command{tar} uses to quote these characters depends on
7616 the @dfn{quoting style}. The default quoting style, called
7617 @dfn{escape} (see below), uses backslash notation to represent control
7618 characters, space and backslash. Using this quoting style, control
7619 characters are represented as listed in column @samp{Character} in the
7620 above table, a space is printed as @samp{\ } and a backslash as @samp{\\}.
7622 @GNUTAR{} offers seven distinct quoting styles, which can be selected
7623 using @option{--quoting-style} option:
7626 @item --quoting-style=@var{style}
7627 @opindex quoting-style
7629 Sets quoting style. Valid values for @var{style} argument are:
7630 literal, shell, shell-always, c, escape, locale, clocale.
7633 These styles are described in detail below. To illustrate their
7634 effect, we will use an imaginary tar archive @file{arch.tar}
7635 containing the following members:
7639 # 1. Contains horizontal tabulation character.
7641 # 2. Contains newline character
7644 # 3. Contains a space
7646 # 4. Contains double quotes
7648 # 5. Contains single quotes
7650 # 6. Contains a backslash character:
7655 Here is how usual @command{ls} command would have listed them, if they
7656 had existed in the current working directory:
7674 No quoting, display each character as is:
7678 $ @kbd{tar tf arch.tar --quoting-style=literal}
7691 Display characters the same way Bourne shell does:
7692 control characters, except @samp{\t} and @samp{\n}, are printed using
7693 backslash escapes, @samp{\t} and @samp{\n} are printed as is, and a
7694 single quote is printed as @samp{\'}. If a name contains any quoted
7695 characters, it is enclosed in single quotes. In particular, if a name
7696 contains single quotes, it is printed as several single-quoted strings:
7700 $ @kbd{tar tf arch.tar --quoting-style=shell}
7703 './a'\''single'\''quote'
7713 Same as @samp{shell}, but the names are always enclosed in single
7718 $ @kbd{tar tf arch.tar --quoting-style=shell-always}
7721 './a'\''single'\''quote'
7731 Use the notation of the C programming language. All names are
7732 enclosed in double quotes. Control characters are quoted using
7733 backslash notations, double quotes are represented as @samp{\"},
7734 backslash characters are represented as @samp{\\}. Single quotes and
7735 spaces are not quoted:
7739 $ @kbd{tar tf arch.tar --quoting-style=c}
7743 "./a\"double\"quote"
7751 Control characters are printed using backslash notation, a space is
7752 printed as @samp{\ } and a backslash as @samp{\\}. This is the
7753 default quoting style, unless it was changed when configured the
7758 $ @kbd{tar tf arch.tar --quoting-style=escape}
7770 Control characters, single quote and backslash are printed using
7771 backslash notation. All names are quoted using left and right
7772 quotation marks, appropriate to the current locale. If it does not
7773 define quotation marks, use @samp{`} as left and @samp{'} as right
7774 quotation marks. Any occurrences of the right quotation mark in a
7775 name are escaped with @samp{\}, for example:
7781 $ @kbd{tar tf arch.tar --quoting-style=locale}
7784 `./a\'single\'quote'
7793 Same as @samp{locale}, but @samp{"} is used for both left and right
7794 quotation marks, if not provided by the currently selected locale:
7798 $ @kbd{tar tf arch.tar --quoting-style=clocale}
7802 "./a\"double\"quote"
7810 You can specify which characters should be quoted in addition to those
7811 implied by the current quoting style:
7814 @item --quote-chars=@var{string}
7815 Always quote characters from @var{string}, even if the selected
7816 quoting style would not quote them.
7819 For example, using @samp{escape} quoting (compare with the usual
7820 escape listing above):
7824 $ @kbd{tar tf arch.tar --quoting-style=escape --quote-chars=' "'}
7835 To disable quoting of such additional characters, use the following
7839 @item --no-quote-chars=@var{string}
7840 Remove characters listed in @var{string} from the list of quoted
7841 characters set by the previous @option{--quote-chars} option.
7844 This option is particularly useful if you have added
7845 @option{--quote-chars} to your @env{TAR_OPTIONS} (@pxref{TAR_OPTIONS})
7846 and wish to disable it for the current invocation.
7848 Note, that @option{--no-quote-chars} does @emph{not} disable those
7849 characters that are quoted by default in the selected quoting style.
7852 @section Modifying File and Member Names
7854 @command{Tar} archives contain detailed information about files stored
7855 in them and full file names are part of that information. When
7856 storing a file to an archive, its file name is recorded in it,
7857 along with the actual file contents. When restoring from an archive,
7858 a file is created on disk with exactly the same name as that stored
7859 in the archive. In the majority of cases this is the desired behavior
7860 of a file archiver. However, there are some cases when it is not.
7862 First of all, it is often unsafe to extract archive members with
7863 absolute file names or those that begin with a @file{../}. @GNUTAR{}
7864 takes special precautions when extracting such names and provides a
7865 special option for handling them, which is described in
7868 Secondly, you may wish to extract file names without some leading
7869 directory components, or with otherwise modified names. In other
7870 cases it is desirable to store files under differing names in the
7873 @GNUTAR{} provides several options for these needs.
7876 @opindex strip-components
7877 @item --strip-components=@var{number}
7878 Strip given @var{number} of leading components from file names before
7882 For example, suppose you have archived whole @file{/usr} hierarchy to
7883 a tar archive named @file{usr.tar}. Among other files, this archive
7884 contains @file{usr/include/stdlib.h}, which you wish to extract to
7885 the current working directory. To do so, you type:
7888 $ @kbd{tar -xf usr.tar --strip=2 usr/include/stdlib.h}
7891 The option @option{--strip=2} instructs @command{tar} to strip the
7892 two leading components (@file{usr/} and @file{include/}) off the file
7895 If you add the @option{--verbose} (@option{-v}) option to the invocation
7896 above, you will note that the verbose listing still contains the
7897 full file name, with the two removed components still in place. This
7898 can be inconvenient, so @command{tar} provides a special option for
7899 altering this behavior:
7901 @anchor{show-transformed-names}
7903 @opindex show-transformed-names
7904 @item --show-transformed-names
7905 Display file or member names with all requested transformations
7914 $ @kbd{tar -xf usr.tar -v --strip=2 usr/include/stdlib.h}
7915 usr/include/stdlib.h
7916 $ @kbd{tar -xf usr.tar -v --strip=2 --show-transformed usr/include/stdlib.h}
7921 Notice that in both cases the file @file{stdlib.h} is extracted to the
7922 current working directory, @option{--show-transformed-names} affects
7923 only the way its name is displayed.
7925 This option is especially useful for verifying whether the invocation
7926 will have the desired effect. Thus, before running
7929 $ @kbd{tar -x --strip=@var{n}}
7933 it is often advisable to run
7936 $ @kbd{tar -t -v --show-transformed --strip=@var{n}}
7940 to make sure the command will produce the intended results.
7942 In case you need to apply more complex modifications to the file name,
7943 @GNUTAR{} provides a general-purpose transformation option:
7948 @item --transform=@var{expression}
7949 @itemx --xform=@var{expression}
7950 Modify file names using supplied @var{expression}.
7954 The @var{expression} is a @command{sed}-like replace expression of the
7958 s/@var{regexp}/@var{replace}/[@var{flags}]
7962 where @var{regexp} is a @dfn{regular expression}, @var{replace} is a
7963 replacement for each file name part that matches @var{regexp}. Both
7964 @var{regexp} and @var{replace} are described in detail in
7965 @ref{The "s" Command, The "s" Command, The `s' Command, sed, GNU sed}.
7967 Any delimiter can be used in lieu of @samp{/}, the only requirement being
7968 that it be used consistently throughout the expression. For example,
7969 the following two expressions are equivalent:
7978 Changing delimiters is often useful when the @var{regex} contains
7979 slashes. For example, it is more convenient to write @code{s,/,-,} than
7982 As in @command{sed}, you can give several replace expressions,
7983 separated by a semicolon.
7985 Supported @var{flags} are:
7989 Apply the replacement to @emph{all} matches to the @var{regexp}, not
7993 Use case-insensitive matching.
7996 @var{regexp} is an @dfn{extended regular expression} (@pxref{Extended
7997 regexps, Extended regular expressions, Extended regular expressions,
8001 Only replace the @var{number}th match of the @var{regexp}.
8003 Note: the @acronym{POSIX} standard does not specify what should happen
8004 when you mix the @samp{g} and @var{number} modifiers. @GNUTAR{}
8005 follows the GNU @command{sed} implementation in this regard, so
8006 the interaction is defined to be: ignore matches before the
8007 @var{number}th, and then match and replace all matches from the
8012 In addition, several @dfn{transformation scope} flags are supported,
8013 that control to what files transformations apply. These are:
8017 Apply transformation to regular archive members.
8020 Do not apply transformation to regular archive members.
8023 Apply transformation to symbolic link targets.
8026 Do not apply transformation to symbolic link targets.
8029 Apply transformation to hard link targets.
8032 Do not apply transformation to hard link targets.
8035 Default is @samp{rsh}, which means to apply tranformations to both archive
8036 members and targets of symbolic and hard links.
8038 Default scope flags can also be changed using @samp{flags=} statement
8039 in the transform expression. The flags set this way remain in force
8040 until next @samp{flags=} statement or end of expression, whichever
8041 occurs first. For example:
8044 --transform 'flags=S;s|^|/usr/local/|'
8047 Here are several examples of @option{--transform} usage:
8050 @item Extract @file{usr/} hierarchy into @file{usr/local/}:
8053 $ @kbd{tar --transform='s,usr/,usr/local/,' -x -f arch.tar}
8056 @item Strip two leading directory components (equivalent to
8057 @option{--strip-components=2}):
8060 $ @kbd{tar --transform='s,/*[^/]*/[^/]*/,,' -x -f arch.tar}
8063 @item Convert each file name to lower case:
8066 $ @kbd{tar --transform 's/.*/\L&/' -x -f arch.tar}
8069 @item Prepend @file{/prefix/} to each file name:
8072 $ @kbd{tar --transform 's,^,/prefix/,' -x -f arch.tar}
8075 @item Archive the @file{/lib} directory, prepending @samp{/usr/local}
8076 to each archive member:
8079 $ @kbd{tar --transform 's,^,/usr/local/,S' -c -f arch.tar /lib}
8083 Notice the use of flags in the last example. The @file{/lib}
8084 directory often contains many symbolic links to files within it.
8085 It may look, for example, like this:
8089 drwxr-xr-x root/root 0 2008-07-08 16:20 /lib/
8090 -rwxr-xr-x root/root 1250840 2008-05-25 07:44 /lib/libc-2.3.2.so
8091 lrwxrwxrwx root/root 0 2008-06-24 17:12 /lib/libc.so.6 -> libc-2.3.2.so
8095 Using the expression @samp{s,^,/usr/local/,} would mean adding
8096 @samp{/usr/local} to both regular archive members and to link
8097 targets. In this case, @file{/lib/libc.so.6} would become:
8100 /usr/local/lib/libc.so.6 -> /usr/local/libc-2.3.2.so
8103 This is definitely not desired. To avoid this, the @samp{S} flag
8104 is used, which excludes symbolic link targets from filename
8105 transformations. The result is:
8108 $ @kbd{tar --transform 's,^,/usr/local/,S', -c -v -f arch.tar \
8109 --show-transformed /lib}
8110 drwxr-xr-x root/root 0 2008-07-08 16:20 /usr/local/lib/
8111 -rwxr-xr-x root/root 1250840 2008-05-25 07:44 /usr/local/lib/libc-2.3.2.so
8112 lrwxrwxrwx root/root 0 2008-06-24 17:12 /usr/local/lib/libc.so.6 \
8116 Unlike @option{--strip-components}, @option{--transform} can be used
8117 in any @GNUTAR{} operation mode. For example, the following command
8118 adds files to the archive while replacing the leading @file{usr/}
8119 component with @file{var/}:
8122 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' /}
8125 To test @option{--transform} effect we suggest using
8126 @option{--show-transformed-names} option:
8129 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' \
8130 --verbose --show-transformed-names /}
8133 If both @option{--strip-components} and @option{--transform} are used
8134 together, then @option{--transform} is applied first, and the required
8135 number of components is then stripped from its result.
8137 You can use as many @option{--transform} options in a single command
8138 line as you want. The specified expressions will then be applied in
8139 order of their appearance. For example, the following two invocations
8143 $ @kbd{tar -cf arch.tar --transform='s,/usr/var,/var/' \
8144 --transform='s,/usr/local,/usr/,'}
8145 $ @kbd{tar -cf arch.tar \
8146 --transform='s,/usr/var,/var/;s,/usr/local,/usr/,'}
8150 @section Operating Only on New Files
8152 @cindex Excluding file by age
8153 @cindex Data Modification time, excluding files by
8154 @cindex Modification time, excluding files by
8155 @cindex Age, excluding files by
8156 The @option{--after-date=@var{date}} (@option{--newer=@var{date}},
8157 @option{-N @var{date}}) option causes @command{tar} to only work on
8158 files whose data modification or status change times are newer than
8159 the @var{date} given. If @var{date} starts with @samp{/} or @samp{.},
8160 it is taken to be a file name; the data modification time of that file
8161 is used as the date. If you use this option when creating or appending
8162 to an archive, the archive will only include new files. If you use
8163 @option{--after-date} when extracting an archive, @command{tar} will
8164 only extract files newer than the @var{date} you specify.
8166 If you only want @command{tar} to make the date comparison based on
8167 modification of the file's data (rather than status
8168 changes), then use the @option{--newer-mtime=@var{date}} option.
8170 @cindex --after-date and --update compared
8171 @cindex --newer-mtime and --update compared
8172 You may use these options with any operation. Note that these options
8173 differ from the @option{--update} (@option{-u}) operation in that they
8174 allow you to specify a particular date against which @command{tar} can
8175 compare when deciding whether or not to archive the files.
8180 @item --after-date=@var{date}
8181 @itemx --newer=@var{date}
8182 @itemx -N @var{date}
8183 Only store files newer than @var{date}.
8185 Acts on files only if their data modification or status change times are
8186 later than @var{date}. Use in conjunction with any operation.
8188 If @var{date} starts with @samp{/} or @samp{.}, it is taken to be a file
8189 name; the data modification time of that file is used as the date.
8191 @opindex newer-mtime
8192 @item --newer-mtime=@var{date}
8193 Acts like @option{--after-date}, but only looks at data modification times.
8196 These options limit @command{tar} to operate only on files which have
8197 been modified after the date specified. A file's status is considered to have
8198 changed if its contents have been modified, or if its owner,
8199 permissions, and so forth, have been changed. (For more information on
8200 how to specify a date, see @ref{Date input formats}; remember that the
8201 entire date argument must be quoted if it contains any spaces.)
8203 Gurus would say that @option{--after-date} tests both the data
8204 modification time (@code{mtime}, the time the contents of the file
8205 were last modified) and the status change time (@code{ctime}, the time
8206 the file's status was last changed: owner, permissions, etc.@:)
8207 fields, while @option{--newer-mtime} tests only the @code{mtime}
8210 To be precise, @option{--after-date} checks @emph{both} @code{mtime} and
8211 @code{ctime} and processes the file if either one is more recent than
8212 @var{date}, while @option{--newer-mtime} only checks @code{mtime} and
8213 disregards @code{ctime}. Neither does it use @code{atime} (the last time the
8214 contents of the file were looked at).
8216 Date specifiers can have embedded spaces. Because of this, you may need
8217 to quote date arguments to keep the shell from parsing them as separate
8218 arguments. For example, the following command will add to the archive
8219 all the files modified less than two days ago:
8222 $ @kbd{tar -cf foo.tar --newer-mtime '2 days ago'}
8225 When any of these options is used with the option @option{--verbose}
8226 (@pxref{verbose tutorial}) @GNUTAR{} will try to convert the specified
8227 date back to its textual representation and compare that with the
8228 one given with the option. If the two dates differ, @command{tar} will
8229 print a warning saying what date it will use. This is to help user
8230 ensure he is using the right date. For example:
8234 $ @kbd{tar -c -f archive.tar --after-date='10 days ago' .}
8235 tar: Option --after-date: Treating date `10 days ago' as 2006-06-11
8241 @strong{Please Note:} @option{--after-date} and @option{--newer-mtime}
8242 should not be used for incremental backups. @xref{Incremental Dumps},
8243 for proper way of creating incremental backups.
8247 @section Descending into Directories
8248 @cindex Avoiding recursion in directories
8249 @cindex Descending directories, avoiding
8250 @cindex Directories, avoiding recursion
8251 @cindex Recursion in directories, avoiding
8253 Usually, @command{tar} will recursively explore all directories (either
8254 those given on the command line or through the @option{--files-from}
8255 option) for the various files they contain. However, you may not always
8256 want @command{tar} to act this way.
8258 @opindex no-recursion
8259 @cindex @command{find}, using with @command{tar}
8260 The @option{--no-recursion} option inhibits @command{tar}'s recursive descent
8261 into specified directories. If you specify @option{--no-recursion}, you can
8262 use the @command{find} (@pxref{Top,, find, find, GNU Find Manual})
8263 utility for hunting through levels of directories to
8264 construct a list of file names which you could then pass to @command{tar}.
8265 @command{find} allows you to be more selective when choosing which files to
8266 archive; see @ref{files}, for more information on using @command{find} with
8270 @item --no-recursion
8271 Prevents @command{tar} from recursively descending directories.
8275 Requires @command{tar} to recursively descend directories.
8276 This is the default.
8279 When you use @option{--no-recursion}, @GNUTAR{} grabs
8280 directory entries themselves, but does not descend on them
8281 recursively. Many people use @command{find} for locating files they
8282 want to back up, and since @command{tar} @emph{usually} recursively
8283 descends on directories, they have to use the @samp{@w{-not -type d}}
8284 test in their @command{find} invocation (@pxref{Type, Type, Type test,
8285 find, Finding Files}), as they usually do not want all the files in a
8286 directory. They then use the @option{--files-from} option to archive
8287 the files located via @command{find}.
8289 The problem when restoring files archived in this manner is that the
8290 directories themselves are not in the archive; so the
8291 @option{--same-permissions} (@option{--preserve-permissions},
8292 @option{-p}) option does not affect them---while users might really
8293 like it to. Specifying @option{--no-recursion} is a way to tell
8294 @command{tar} to grab only the directory entries given to it, adding
8295 no new files on its own. To summarize, if you use @command{find} to
8296 create a list of files to be stored in an archive, use it as follows:
8300 $ @kbd{find @var{dir} @var{tests} | \
8301 tar -cf @var{archive} -T - --no-recursion}
8305 The @option{--no-recursion} option also applies when extracting: it
8306 causes @command{tar} to extract only the matched directory entries, not
8307 the files under those directories.
8309 The @option{--no-recursion} option also affects how globbing patterns
8310 are interpreted (@pxref{controlling pattern-matching}).
8312 The @option{--no-recursion} and @option{--recursion} options apply to
8313 later options and operands, and can be overridden by later occurrences
8314 of @option{--no-recursion} and @option{--recursion}. For example:
8317 $ @kbd{tar -cf jams.tar --no-recursion grape --recursion grape/concord}
8321 creates an archive with one entry for @file{grape}, and the recursive
8322 contents of @file{grape/concord}, but no entries under @file{grape}
8323 other than @file{grape/concord}.
8326 @section Crossing File System Boundaries
8327 @cindex File system boundaries, not crossing
8329 @command{tar} will normally automatically cross file system boundaries in
8330 order to archive files which are part of a directory tree. You can
8331 change this behavior by running @command{tar} and specifying
8332 @option{--one-file-system}. This option only affects files that are
8333 archived because they are in a directory that is being archived;
8334 @command{tar} will still archive files explicitly named on the command line
8335 or through @option{--files-from}, regardless of where they reside.
8338 @opindex one-file-system
8339 @item --one-file-system
8340 Prevents @command{tar} from crossing file system boundaries when
8341 archiving. Use in conjunction with any write operation.
8344 The @option{--one-file-system} option causes @command{tar} to modify its
8345 normal behavior in archiving the contents of directories. If a file in
8346 a directory is not on the same file system as the directory itself, then
8347 @command{tar} will not archive that file. If the file is a directory
8348 itself, @command{tar} will not archive anything beneath it; in other words,
8349 @command{tar} will not cross mount points.
8351 This option is useful for making full or incremental archival backups of
8352 a file system. If this option is used in conjunction with
8353 @option{--verbose} (@option{-v}), files that are excluded are
8354 mentioned by name on the standard error.
8357 * directory:: Changing Directory
8358 * absolute:: Absolute File Names
8362 @subsection Changing the Working Directory
8364 @FIXME{need to read over this node now for continuity; i've switched
8365 things around some.}
8367 @cindex Changing directory mid-stream
8368 @cindex Directory, changing mid-stream
8369 @cindex Working directory, specifying
8370 To change the working directory in the middle of a list of file names,
8371 either on the command line or in a file specified using
8372 @option{--files-from} (@option{-T}), use @option{--directory} (@option{-C}).
8373 This will change the working directory to the specified directory
8374 after that point in the list.
8378 @item --directory=@var{directory}
8379 @itemx -C @var{directory}
8380 Changes the working directory in the middle of a command line.
8386 $ @kbd{tar -c -f jams.tar grape prune -C food cherry}
8390 will place the files @file{grape} and @file{prune} from the current
8391 directory into the archive @file{jams.tar}, followed by the file
8392 @file{cherry} from the directory @file{food}. This option is especially
8393 useful when you have several widely separated files that you want to
8394 store in the same archive.
8396 Note that the file @file{cherry} is recorded in the archive under the
8397 precise name @file{cherry}, @emph{not} @file{food/cherry}. Thus, the
8398 archive will contain three files that all appear to have come from the
8399 same directory; if the archive is extracted with plain @samp{tar
8400 --extract}, all three files will be written in the current directory.
8402 Contrast this with the command,
8405 $ @kbd{tar -c -f jams.tar grape prune -C food red/cherry}
8409 which records the third file in the archive under the name
8410 @file{red/cherry} so that, if the archive is extracted using
8411 @samp{tar --extract}, the third file will be written in a subdirectory
8414 You can use the @option{--directory} option to make the archive
8415 independent of the original name of the directory holding the files.
8416 The following command places the files @file{/etc/passwd},
8417 @file{/etc/hosts}, and @file{/lib/libc.a} into the archive
8421 $ @kbd{tar -c -f foo.tar -C /etc passwd hosts -C /lib libc.a}
8425 However, the names of the archive members will be exactly what they were
8426 on the command line: @file{passwd}, @file{hosts}, and @file{libc.a}.
8427 They will not appear to be related by file name to the original
8428 directories where those files were located.
8430 Note that @option{--directory} options are interpreted consecutively. If
8431 @option{--directory} specifies a relative file name, it is interpreted
8432 relative to the then current directory, which might not be the same as
8433 the original current working directory of @command{tar}, due to a previous
8434 @option{--directory} option.
8436 When using @option{--files-from} (@pxref{files}), you can put various
8437 @command{tar} options (including @option{-C}) in the file list. Notice,
8438 however, that in this case the option and its argument may not be
8439 separated by whitespace. If you use short option, its argument must
8440 either follow the option letter immediately, without any intervening
8441 whitespace, or occupy the next line. Otherwise, if you use long
8442 option, separate its argument by an equal sign.
8444 For instance, the file list for the above example will be:
8457 To use it, you would invoke @command{tar} as follows:
8460 $ @kbd{tar -c -f foo.tar --files-from list}
8463 The interpretation of @option{--directory} is disabled by
8464 @option{--null} option.
8467 @subsection Absolute File Names
8468 @cindex absolute file names
8469 @cindex file names, absolute
8471 By default, @GNUTAR{} drops a leading @samp{/} on
8472 input or output, and complains about file names containing a @file{..}
8473 component. There is an option that turns off this behavior:
8476 @opindex absolute-names
8477 @item --absolute-names
8479 Do not strip leading slashes from file names, and permit file names
8480 containing a @file{..} file name component.
8483 When @command{tar} extracts archive members from an archive, it strips any
8484 leading slashes (@samp{/}) from the member name. This causes absolute
8485 member names in the archive to be treated as relative file names. This
8486 allows you to have such members extracted wherever you want, instead of
8487 being restricted to extracting the member in the exact directory named
8488 in the archive. For example, if the archive member has the name
8489 @file{/etc/passwd}, @command{tar} will extract it as if the name were
8490 really @file{etc/passwd}.
8492 File names containing @file{..} can cause problems when extracting, so
8493 @command{tar} normally warns you about such files when creating an
8494 archive, and rejects attempts to extracts such files.
8496 Other @command{tar} programs do not do this. As a result, if you
8497 create an archive whose member names start with a slash, they will be
8498 difficult for other people with a non-@GNUTAR{}
8499 program to use. Therefore, @GNUTAR{} also strips
8500 leading slashes from member names when putting members into the
8501 archive. For example, if you ask @command{tar} to add the file
8502 @file{/bin/ls} to an archive, it will do so, but the member name will
8503 be @file{bin/ls}@footnote{A side effect of this is that when
8504 @option{--create} is used with @option{--verbose} the resulting output
8505 is not, generally speaking, the same as the one you'd get running
8506 @kbd{tar --list} command. This may be important if you use some
8507 scripts for comparing both outputs. @xref{listing member and file names},
8508 for the information on how to handle this case.}.
8510 If you use the @option{--absolute-names} (@option{-P}) option,
8511 @command{tar} will do none of these transformations.
8513 To archive or extract files relative to the root directory, specify
8514 the @option{--absolute-names} (@option{-P}) option.
8516 Normally, @command{tar} acts on files relative to the working
8517 directory---ignoring superior directory names when archiving, and
8518 ignoring leading slashes when extracting.
8520 When you specify @option{--absolute-names} (@option{-P}),
8521 @command{tar} stores file names including all superior directory
8522 names, and preserves leading slashes. If you only invoked
8523 @command{tar} from the root directory you would never need the
8524 @option{--absolute-names} option, but using this option
8525 may be more convenient than switching to root.
8527 @FIXME{Should be an example in the tutorial/wizardry section using this
8528 to transfer files between systems.}
8531 @item --absolute-names
8532 Preserves full file names (including superior directory names) when
8533 archiving files. Preserves leading slash when extracting files.
8537 @command{tar} prints out a message about removing the @samp{/} from
8538 file names. This message appears once per @GNUTAR{}
8539 invocation. It represents something which ought to be told; ignoring
8540 what it means can cause very serious surprises, later.
8542 Some people, nevertheless, do not want to see this message. Wanting to
8543 play really dangerously, one may of course redirect @command{tar} standard
8544 error to the sink. For example, under @command{sh}:
8547 $ @kbd{tar -c -f archive.tar /home 2> /dev/null}
8551 Another solution, both nicer and simpler, would be to change to
8552 the @file{/} directory first, and then avoid absolute notation.
8556 $ @kbd{tar -c -f archive.tar -C / home}
8559 @include getdate.texi
8562 @chapter Controlling the Archive Format
8564 @cindex Tar archive formats
8565 Due to historical reasons, there are several formats of tar archives.
8566 All of them are based on the same principles, but have some subtle
8567 differences that often make them incompatible with each other.
8569 GNU tar is able to create and handle archives in a variety of formats.
8570 The most frequently used formats are (in alphabetical order):
8574 Format used by @GNUTAR{} versions up to 1.13.25. This format derived
8575 from an early @acronym{POSIX} standard, adding some improvements such as
8576 sparse file handling and incremental archives. Unfortunately these
8577 features were implemented in a way incompatible with other archive
8580 Archives in @samp{gnu} format are able to hold file names of unlimited
8584 Format used by @GNUTAR{} of versions prior to 1.12.
8587 Archive format, compatible with the V7 implementation of tar. This
8588 format imposes a number of limitations. The most important of them
8592 @item The maximum length of a file name is limited to 99 characters.
8593 @item The maximum length of a symbolic link is limited to 99 characters.
8594 @item It is impossible to store special files (block and character
8595 devices, fifos etc.)
8596 @item Maximum value of user or group @acronym{ID} is limited to 2097151 (7777777
8598 @item V7 archives do not contain symbolic ownership information (user
8599 and group name of the file owner).
8602 This format has traditionally been used by Automake when producing
8603 Makefiles. This practice will change in the future, in the meantime,
8604 however this means that projects containing file names more than 99
8605 characters long will not be able to use @GNUTAR{} @value{VERSION} and
8606 Automake prior to 1.9.
8609 Archive format defined by @acronym{POSIX.1-1988} specification. It stores
8610 symbolic ownership information. It is also able to store
8611 special files. However, it imposes several restrictions as well:
8614 @item The maximum length of a file name is limited to 256 characters,
8615 provided that the file name can be split at a directory separator in
8616 two parts, first of them being at most 155 bytes long. So, in most
8617 cases the maximum file name length will be shorter than 256
8619 @item The maximum length of a symbolic link name is limited to
8621 @item Maximum size of a file the archive is able to accommodate
8623 @item Maximum value of UID/GID is 2097151.
8624 @item Maximum number of bits in device major and minor numbers is 21.
8628 Format used by J@"org Schilling @command{star}
8629 implementation. @GNUTAR{} is able to read @samp{star} archives but
8630 currently does not produce them.
8633 Archive format defined by @acronym{POSIX.1-2001} specification. This is the
8634 most flexible and feature-rich format. It does not impose any
8635 restrictions on file sizes or file name lengths. This format is quite
8636 recent, so not all tar implementations are able to handle it properly.
8637 However, this format is designed in such a way that any tar
8638 implementation able to read @samp{ustar} archives will be able to read
8639 most @samp{posix} archives as well, with the only exception that any
8640 additional information (such as long file names etc.) will in such
8641 case be extracted as plain text files along with the files it refers to.
8643 This archive format will be the default format for future versions
8648 The following table summarizes the limitations of each of these
8651 @multitable @columnfractions .10 .20 .20 .20 .20
8652 @headitem Format @tab UID @tab File Size @tab File Name @tab Devn
8653 @item gnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
8654 @item oldgnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
8655 @item v7 @tab 2097151 @tab 8GB @tab 99 @tab n/a
8656 @item ustar @tab 2097151 @tab 8GB @tab 256 @tab 21
8657 @item posix @tab Unlimited @tab Unlimited @tab Unlimited @tab Unlimited
8660 The default format for @GNUTAR{} is defined at compilation
8661 time. You may check it by running @command{tar --help}, and examining
8662 the last lines of its output. Usually, @GNUTAR{} is configured
8663 to create archives in @samp{gnu} format, however, future version will
8664 switch to @samp{posix}.
8667 * Compression:: Using Less Space through Compression
8668 * Attributes:: Handling File Attributes
8669 * Portability:: Making @command{tar} Archives More Portable
8670 * cpio:: Comparison of @command{tar} and @command{cpio}
8674 @section Using Less Space through Compression
8677 * gzip:: Creating and Reading Compressed Archives
8678 * sparse:: Archiving Sparse Files
8682 @subsection Creating and Reading Compressed Archives
8683 @cindex Compressed archives
8684 @cindex Storing archives in compressed format
8692 @GNUTAR{} is able to create and read compressed archives. It supports
8693 a wide variety of compression programs, namely: @command{gzip},
8694 @command{bzip2}, @command{lzip}, @command{lzma}, @command{lzop},
8695 @command{xz} and traditional @command{compress}. The latter is
8696 supported mostly for backward compatibility, and we recommend
8697 against using it, because it is by far less effective than the other
8698 compression programs@footnote{It also had patent problems in the past.}.
8700 Creating a compressed archive is simple: you just specify a
8701 @dfn{compression option} along with the usual archive creation
8702 commands. The compression option is @option{-z} (@option{--gzip}) to
8703 create a @command{gzip} compressed archive, @option{-j}
8704 (@option{--bzip2}) to create a @command{bzip2} compressed archive,
8705 @option{--lzip} to create an @asis{lzip} compressed archive,
8706 @option{-J} (@option{--xz}) to create an @asis{XZ} archive,
8707 @option{--lzma} to create an @asis{LZMA} compressed
8708 archive, @option{--lzop} to create an @asis{LSOP} archive, and
8709 @option{-Z} (@option{--compress}) to use @command{compress} program.
8713 $ @kbd{tar cfz archive.tar.gz .}
8716 You can also let @GNUTAR{} select the compression program based on
8717 the suffix of the archive file name. This is done using
8718 @option{--auto-compress} (@option{-a}) command line option. For
8719 example, the following invocation will use @command{bzip2} for
8723 $ @kbd{tar cfa archive.tar.bz2 .}
8727 whereas the following one will use @command{lzma}:
8730 $ @kbd{tar cfa archive.tar.lzma .}
8733 For a complete list of file name suffixes recognized by @GNUTAR{},
8734 see @ref{auto-compress}.
8736 Reading compressed archive is even simpler: you don't need to specify
8737 any additional options as @GNUTAR{} recognizes its format
8738 automatically. Thus, the following commands will list and extract the
8739 archive created in previous example:
8742 # List the compressed archive
8743 $ @kbd{tar tf archive.tar.gz}
8744 # Extract the compressed archive
8745 $ @kbd{tar xf archive.tar.gz}
8748 The format recognition algorithm is based on @dfn{signatures}, a
8749 special byte sequences in the beginning of file, that are specific for
8750 certain compression formats. If this approach fails, @command{tar}
8751 falls back to using archive name suffix to determine its format
8752 (@pxref{auto-compress}, for a list of recognized suffixes).
8754 The only case when you have to specify a decompression option while
8755 reading the archive is when reading from a pipe or from a tape drive
8756 that does not support random access. However, in this case @GNUTAR{}
8757 will indicate which option you should use. For example:
8760 $ @kbd{cat archive.tar.gz | tar tf -}
8761 tar: Archive is compressed. Use -z option
8762 tar: Error is not recoverable: exiting now
8765 If you see such diagnostics, just add the suggested option to the
8766 invocation of @GNUTAR{}:
8769 $ @kbd{cat archive.tar.gz | tar tfz -}
8772 Notice also, that there are several restrictions on operations on
8773 compressed archives. First of all, compressed archives cannot be
8774 modified, i.e., you cannot update (@option{--update}, alias @option{-u})
8775 them or delete (@option{--delete}) members from them or
8776 add (@option{--append}, alias @option{-r}) members to them. Likewise, you
8777 cannot append another @command{tar} archive to a compressed archive using
8778 @option{--concatenate} (@option{-A}). Secondly, multi-volume
8779 archives cannot be compressed.
8781 The following options allow to select a particular compressor program:
8789 Filter the archive through @command{gzip}.
8794 Filter the archive through @code{xz}.
8798 Filter the archive through @code{bzip2}.
8802 Filter the archive through @command{lzip}.
8806 Filter the archive through @command{lzma}.
8810 Filter the archive through @command{lzop}.
8817 Filter the archive through @command{compress}.
8820 When any of these options is given, @GNUTAR{} searches the compressor
8821 binary in the current path and invokes it. The name of the compressor
8822 program is specified at compilation time using a corresponding
8823 @option{--with-@var{compname}} option to @command{configure}, e.g.
8824 @option{--with-bzip2} to select a specific @command{bzip2} binary.
8825 @xref{lbzip2}, for a detailed discussion.
8827 The output produced by @command{tar --help} shows the actual
8828 compressor names along with each of these options.
8830 You can use any of these options on physical devices (tape drives,
8831 etc.) and remote files as well as on normal files; data to or from
8832 such devices or remote files is reblocked by another copy of the
8833 @command{tar} program to enforce the specified (or default) record
8834 size. The default compression parameters are used. Most compression
8835 programs allow to override these by setting a program-specific
8836 environment variable. For example, when using @command{gzip} you can
8837 use @env{GZIP} as in the example below:
8840 $ @kbd{GZIP=--best tar cfz archive.tar.gz subdir}
8844 Another way would be to use the @option{-I} option instead (see
8848 $ @kbd{tar -cf archive.tar.gz -I 'gzip --best' subdir}
8852 Finally, the third, traditional, way to achieve the same result is to
8856 $ @kbd{tar cf - subdir | gzip --best -c - > archive.tar.gz}
8859 @cindex corrupted archives
8860 About corrupted compressed archives: compressed files have no
8861 redundancy, for maximum compression. The adaptive nature of the
8862 compression scheme means that the compression tables are implicitly
8863 spread all over the archive. If you lose a few blocks, the dynamic
8864 construction of the compression tables becomes unsynchronized, and there
8865 is little chance that you could recover later in the archive.
8867 Another compression options provide a better control over creating
8868 compressed archives. These are:
8871 @anchor{auto-compress}
8872 @opindex auto-compress
8873 @item --auto-compress
8875 Select a compression program to use by the archive file name
8876 suffix. The following suffixes are recognized:
8878 @multitable @columnfractions 0.3 0.6
8879 @headitem Suffix @tab Compression program
8880 @item @samp{.gz} @tab @command{gzip}
8881 @item @samp{.tgz} @tab @command{gzip}
8882 @item @samp{.taz} @tab @command{gzip}
8883 @item @samp{.Z} @tab @command{compress}
8884 @item @samp{.taZ} @tab @command{compress}
8885 @item @samp{.bz2} @tab @command{bzip2}
8886 @item @samp{.tz2} @tab @command{bzip2}
8887 @item @samp{.tbz2} @tab @command{bzip2}
8888 @item @samp{.tbz} @tab @command{bzip2}
8889 @item @samp{.lz} @tab @command{lzip}
8890 @item @samp{.lzma} @tab @command{lzma}
8891 @item @samp{.tlz} @tab @command{lzma}
8892 @item @samp{.lzo} @tab @command{lzop}
8893 @item @samp{.xz} @tab @command{xz}
8896 @opindex use-compress-program
8897 @item --use-compress-program=@var{prog}
8898 @itemx -I=@var{prog}
8899 Use external compression program @var{prog}. Use this option if you
8900 are not happy with the compression program associated with the suffix
8901 at compile time or if you have a compression program that @GNUTAR{}
8902 does not support. There are two requirements to which @var{prog}
8905 First, when called without options, it should read data from standard
8906 input, compress it and output it on standard output.
8908 Secondly, if called with @option{-d} argument, it should do exactly
8909 the opposite, i.e., read the compressed data from the standard input
8910 and produce uncompressed data on the standard output.
8913 @cindex gpg, using with tar
8914 @cindex gnupg, using with tar
8915 @cindex Using encrypted archives
8916 The @option{--use-compress-program} option, in particular, lets you
8917 implement your own filters, not necessarily dealing with
8918 compression/decompression. For example, suppose you wish to implement
8919 PGP encryption on top of compression, using @command{gpg} (@pxref{Top,
8920 gpg, gpg ---- encryption and signing tool, gpg, GNU Privacy Guard
8921 Manual}). The following script does that:
8927 -d) gpg --decrypt - | gzip -d -c;;
8928 '') gzip -c | gpg -s;;
8929 *) echo "Unknown option $1">&2; exit 1;;
8934 Suppose you name it @file{gpgz} and save it somewhere in your
8935 @env{PATH}. Then the following command will create a compressed
8936 archive signed with your private key:
8939 $ @kbd{tar -cf foo.tar.gpgz -Igpgz .}
8943 Likewise, the command below will list its contents:
8946 $ @kbd{tar -tf foo.tar.gpgz -Igpgz .}
8950 The above is based on the following discussion:
8952 I have one question, or maybe it's a suggestion if there isn't a way
8953 to do it now. I would like to use @option{--gzip}, but I'd also like
8954 the output to be fed through a program like @acronym{GNU}
8955 @command{ecc} (actually, right now that's @samp{exactly} what I'd like
8956 to use :-)), basically adding ECC protection on top of compression.
8957 It seems as if this should be quite easy to do, but I can't work out
8958 exactly how to go about it. Of course, I can pipe the standard output
8959 of @command{tar} through @command{ecc}, but then I lose (though I
8960 haven't started using it yet, I confess) the ability to have
8961 @command{tar} use @command{rmt} for it's I/O (I think).
8963 I think the most straightforward thing would be to let me specify a
8964 general set of filters outboard of compression (preferably ordered,
8965 so the order can be automatically reversed on input operations, and
8966 with the options they require specifiable), but beggars shouldn't be
8967 choosers and anything you decide on would be fine with me.
8969 By the way, I like @command{ecc} but if (as the comments say) it can't
8970 deal with loss of block sync, I'm tempted to throw some time at adding
8971 that capability. Supposing I were to actually do such a thing and
8972 get it (apparently) working, do you accept contributed changes to
8973 utilities like that? (Leigh Clayton @file{loc@@soliton.com}, May 1995).
8975 Isn't that exactly the role of the
8976 @option{--use-compress-prog=@var{program}} option?
8977 I never tried it myself, but I suspect you may want to write a
8978 @var{prog} script or program able to filter stdin to stdout to
8979 way you want. It should recognize the @option{-d} option, for when
8980 extraction is needed rather than creation.
8982 It has been reported that if one writes compressed data (through the
8983 @option{--gzip} or @option{--compress} options) to a DLT and tries to use
8984 the DLT compression mode, the data will actually get bigger and one will
8985 end up with less space on the tape.
8989 * lbzip2:: Using lbzip2 with @GNUTAR{}.
8993 @subsubsection Using lbzip2 with @GNUTAR{}.
8995 @cindex Laszlo Ersek
8996 @command{Lbzip2} is a multithreaded utility for handling
8997 @samp{bzip2} compression, written by Laszlo Ersek. It makes use of
8998 multiple processors to speed up its operation and in general works
8999 considerably faster than @command{bzip2}. For a detailed description
9000 of @command{lbzip2} see @uref{http://freshmeat.net/@/projects/@/lbzip2} and
9001 @uref{http://www.linuxinsight.com/@/lbzip2-parallel-bzip2-utility.html,
9002 lbzip2: parallel bzip2 utility}.
9004 Recent versions of @command{lbzip2} are mostly command line compatible
9005 with @command{bzip2}, which makes it possible to automatically invoke
9006 it via the @option{--bzip2} @GNUTAR{} command line option. To do so,
9007 @GNUTAR{} must be configured with the @option{--with-bzip2} command
9008 line option, like this:
9011 $ @kbd{./configure --with-bzip2=lbzip2 [@var{other-options}]}
9014 Once configured and compiled this way, @command{tar --help} will show the
9019 $ @kbd{tar --help | grep -- --bzip2}
9020 -j, --bzip2 filter the archive through lbzip2
9025 which means that running @command{tar --bzip2} will invoke @command{lbzip2}.
9028 @subsection Archiving Sparse Files
9029 @cindex Sparse Files
9031 Files in the file system occasionally have @dfn{holes}. A @dfn{hole}
9032 in a file is a section of the file's contents which was never written.
9033 The contents of a hole reads as all zeros. On many operating systems,
9034 actual disk storage is not allocated for holes, but they are counted
9035 in the length of the file. If you archive such a file, @command{tar}
9036 could create an archive longer than the original. To have @command{tar}
9037 attempt to recognize the holes in a file, use @option{--sparse}
9038 (@option{-S}). When you use this option, then, for any file using
9039 less disk space than would be expected from its length, @command{tar}
9040 searches the file for consecutive stretches of zeros. It then records
9041 in the archive for the file where the consecutive stretches of zeros
9042 are, and only archives the ``real contents'' of the file. On
9043 extraction (using @option{--sparse} is not needed on extraction) any
9044 such files have holes created wherever the continuous stretches of zeros
9045 were found. Thus, if you use @option{--sparse}, @command{tar} archives
9046 won't take more space than the original.
9052 This option instructs @command{tar} to test each file for sparseness
9053 before attempting to archive it. If the file is found to be sparse it
9054 is treated specially, thus allowing to decrease the amount of space
9055 used by its image in the archive.
9057 This option is meaningful only when creating or updating archives. It
9058 has no effect on extraction.
9061 Consider using @option{--sparse} when performing file system backups,
9062 to avoid archiving the expanded forms of files stored sparsely in the
9065 Even if your system has no sparse files currently, some may be
9066 created in the future. If you use @option{--sparse} while making file
9067 system backups as a matter of course, you can be assured the archive
9068 will never take more space on the media than the files take on disk
9069 (otherwise, archiving a disk filled with sparse files might take
9070 hundreds of tapes). @xref{Incremental Dumps}.
9072 However, be aware that @option{--sparse} option presents a serious
9073 drawback. Namely, in order to determine if the file is sparse
9074 @command{tar} has to read it before trying to archive it, so in total
9075 the file is read @strong{twice}. So, always bear in mind that the
9076 time needed to process all files with this option is roughly twice
9077 the time needed to archive them without it.
9078 @FIXME{A technical note:
9080 Programs like @command{dump} do not have to read the entire file; by
9081 examining the file system directly, they can determine in advance
9082 exactly where the holes are and thus avoid reading through them. The
9083 only data it need read are the actual allocated data blocks.
9084 @GNUTAR{} uses a more portable and straightforward
9085 archiving approach, it would be fairly difficult that it does
9086 otherwise. Elizabeth Zwicky writes to @file{comp.unix.internals}, on
9090 What I did say is that you cannot tell the difference between a hole and an
9091 equivalent number of nulls without reading raw blocks. @code{st_blocks} at
9092 best tells you how many holes there are; it doesn't tell you @emph{where}.
9093 Just as programs may, conceivably, care what @code{st_blocks} is (care
9094 to name one that does?), they may also care where the holes are (I have
9095 no examples of this one either, but it's equally imaginable).
9097 I conclude from this that good archivers are not portable. One can
9098 arguably conclude that if you want a portable program, you can in good
9099 conscience restore files with as many holes as possible, since you can't
9104 @cindex sparse formats, defined
9105 When using @samp{POSIX} archive format, @GNUTAR{} is able to store
9106 sparse files using in three distinct ways, called @dfn{sparse
9107 formats}. A sparse format is identified by its @dfn{number},
9108 consisting, as usual of two decimal numbers, delimited by a dot. By
9109 default, format @samp{1.0} is used. If, for some reason, you wish to
9110 use an earlier format, you can select it using
9111 @option{--sparse-version} option.
9114 @opindex sparse-version
9115 @item --sparse-version=@var{version}
9117 Select the format to store sparse files in. Valid @var{version} values
9118 are: @samp{0.0}, @samp{0.1} and @samp{1.0}. @xref{Sparse Formats},
9119 for a detailed description of each format.
9122 Using @option{--sparse-format} option implies @option{--sparse}.
9125 @section Handling File Attributes
9126 @cindex atrributes, files
9127 @cindex file attributes
9129 When @command{tar} reads files, it updates their access times. To
9130 avoid this, use the @option{--atime-preserve[=METHOD]} option, which can either
9131 reset the access time retroactively or avoid changing it in the first
9135 @opindex atime-preserve
9136 @item --atime-preserve
9137 @itemx --atime-preserve=replace
9138 @itemx --atime-preserve=system
9139 Preserve the access times of files that are read. This works only for
9140 files that you own, unless you have superuser privileges.
9142 @option{--atime-preserve=replace} works on most systems, but it also
9143 restores the data modification time and updates the status change
9144 time. Hence it doesn't interact with incremental dumps nicely
9145 (@pxref{Incremental Dumps}), and it can set access or data modification times
9146 incorrectly if other programs access the file while @command{tar} is
9149 @option{--atime-preserve=system} avoids changing the access time in
9150 the first place, if the operating system supports this.
9151 Unfortunately, this may or may not work on any given operating system
9152 or file system. If @command{tar} knows for sure it won't work, it
9153 complains right away.
9155 Currently @option{--atime-preserve} with no operand defaults to
9156 @option{--atime-preserve=replace}, but this is intended to change to
9157 @option{--atime-preserve=system} when the latter is better-supported.
9162 Do not extract data modification time.
9164 When this option is used, @command{tar} leaves the data modification times
9165 of the files it extracts as the times when the files were extracted,
9166 instead of setting it to the times recorded in the archive.
9168 This option is meaningless with @option{--list} (@option{-t}).
9172 Create extracted files with the same ownership they have in the
9175 This is the default behavior for the superuser,
9176 so this option is meaningful only for non-root users, when @command{tar}
9177 is executed on those systems able to give files away. This is
9178 considered as a security flaw by many people, at least because it
9179 makes quite difficult to correctly account users for the disk space
9180 they occupy. Also, the @code{suid} or @code{sgid} attributes of
9181 files are easily and silently lost when files are given away.
9183 When writing an archive, @command{tar} writes the user @acronym{ID} and user name
9184 separately. If it can't find a user name (because the user @acronym{ID} is not
9185 in @file{/etc/passwd}), then it does not write one. When restoring,
9186 it tries to look the name (if one was written) up in
9187 @file{/etc/passwd}. If it fails, then it uses the user @acronym{ID} stored in
9188 the archive instead.
9190 @opindex no-same-owner
9191 @item --no-same-owner
9193 Do not attempt to restore ownership when extracting. This is the
9194 default behavior for ordinary users, so this option has an effect
9195 only for the superuser.
9197 @opindex numeric-owner
9198 @item --numeric-owner
9199 The @option{--numeric-owner} option allows (ANSI) archives to be written
9200 without user/group name information or such information to be ignored
9201 when extracting. It effectively disables the generation and/or use
9202 of user/group name information. This option forces extraction using
9203 the numeric ids from the archive, ignoring the names.
9205 This is useful in certain circumstances, when restoring a backup from
9206 an emergency floppy with different passwd/group files for example.
9207 It is otherwise impossible to extract files with the right ownerships
9208 if the password file in use during the extraction does not match the
9209 one belonging to the file system(s) being extracted. This occurs,
9210 for example, if you are restoring your files after a major crash and
9211 had booted from an emergency floppy with no password file or put your
9212 disk into another machine to do the restore.
9214 The numeric ids are @emph{always} saved into @command{tar} archives.
9215 The identifying names are added at create time when provided by the
9216 system, unless @option{--format=oldgnu} is used. Numeric ids could be
9217 used when moving archives between a collection of machines using
9218 a centralized management for attribution of numeric ids to users
9219 and groups. This is often made through using the NIS capabilities.
9221 When making a @command{tar} file for distribution to other sites, it
9222 is sometimes cleaner to use a single owner for all files in the
9223 distribution, and nicer to specify the write permission bits of the
9224 files as stored in the archive independently of their actual value on
9225 the file system. The way to prepare a clean distribution is usually
9226 to have some Makefile rule creating a directory, copying all needed
9227 files in that directory, then setting ownership and permissions as
9228 wanted (there are a lot of possible schemes), and only then making a
9229 @command{tar} archive out of this directory, before cleaning
9230 everything out. Of course, we could add a lot of options to
9231 @GNUTAR{} for fine tuning permissions and ownership.
9232 This is not the good way, I think. @GNUTAR{} is
9233 already crowded with options and moreover, the approach just explained
9234 gives you a great deal of control already.
9236 @xopindex{same-permissions, short description}
9237 @xopindex{preserve-permissions, short description}
9239 @itemx --same-permissions
9240 @itemx --preserve-permissions
9241 Extract all protection information.
9243 This option causes @command{tar} to set the modes (access permissions) of
9244 extracted files exactly as recorded in the archive. If this option
9245 is not used, the current @code{umask} setting limits the permissions
9246 on extracted files. This option is by default enabled when
9247 @command{tar} is executed by a superuser.
9250 This option is meaningless with @option{--list} (@option{-t}).
9254 Same as both @option{--same-permissions} and @option{--same-order}.
9256 This option is deprecated, and will be removed in @GNUTAR{} version 1.23.
9261 @section Making @command{tar} Archives More Portable
9263 Creating a @command{tar} archive on a particular system that is meant to be
9264 useful later on many other machines and with other versions of @command{tar}
9265 is more challenging than you might think. @command{tar} archive formats
9266 have been evolving since the first versions of Unix. Many such formats
9267 are around, and are not always compatible with each other. This section
9268 discusses a few problems, and gives some advice about making @command{tar}
9269 archives more portable.
9271 One golden rule is simplicity. For example, limit your @command{tar}
9272 archives to contain only regular files and directories, avoiding
9273 other kind of special files. Do not attempt to save sparse files or
9274 contiguous files as such. Let's discuss a few more problems, in turn.
9276 @FIXME{Discuss GNU extensions (incremental backups, multi-volume
9277 archives and archive labels) in GNU and PAX formats.}
9280 * Portable Names:: Portable Names
9281 * dereference:: Symbolic Links
9282 * hard links:: Hard Links
9283 * old:: Old V7 Archives
9284 * ustar:: Ustar Archives
9285 * gnu:: GNU and old GNU format archives.
9286 * posix:: @acronym{POSIX} archives
9287 * Checksumming:: Checksumming Problems
9288 * Large or Negative Values:: Large files, negative time stamps, etc.
9289 * Other Tars:: How to Extract GNU-Specific Data Using
9290 Other @command{tar} Implementations
9293 @node Portable Names
9294 @subsection Portable Names
9296 Use portable file and member names. A name is portable if it contains
9297 only @acronym{ASCII} letters and digits, @samp{/}, @samp{.}, @samp{_}, and
9298 @samp{-}; it cannot be empty, start with @samp{-} or @samp{//}, or
9299 contain @samp{/-}. Avoid deep directory nesting. For portability to
9300 old Unix hosts, limit your file name components to 14 characters or
9303 If you intend to have your @command{tar} archives to be read under
9304 MSDOS, you should not rely on case distinction for file names, and you
9305 might use the @acronym{GNU} @command{doschk} program for helping you
9306 further diagnosing illegal MSDOS names, which are even more limited
9310 @subsection Symbolic Links
9311 @cindex File names, using symbolic links
9312 @cindex Symbolic link as file name
9314 @opindex dereference
9315 Normally, when @command{tar} archives a symbolic link, it writes a
9316 block to the archive naming the target of the link. In that way, the
9317 @command{tar} archive is a faithful record of the file system contents.
9318 @option{--dereference} (@option{-h}) is used with @option{--create} (@option{-c}), and causes
9319 @command{tar} to archive the files symbolic links point to, instead of
9320 the links themselves. When this option is used, when @command{tar}
9321 encounters a symbolic link, it will archive the linked-to file,
9322 instead of simply recording the presence of a symbolic link.
9324 The name under which the file is stored in the file system is not
9325 recorded in the archive. To record both the symbolic link name and
9326 the file name in the system, archive the file under both names. If
9327 all links were recorded automatically by @command{tar}, an extracted file
9328 might be linked to a file name that no longer exists in the file
9331 If a linked-to file is encountered again by @command{tar} while creating
9332 the same archive, an entire second copy of it will be stored. (This
9333 @emph{might} be considered a bug.)
9335 So, for portable archives, do not archive symbolic links as such,
9336 and use @option{--dereference} (@option{-h}): many systems do not support
9337 symbolic links, and moreover, your distribution might be unusable if
9338 it contains unresolved symbolic links.
9341 @subsection Hard Links
9342 @cindex File names, using hard links
9343 @cindex hard links, dereferencing
9344 @cindex dereferencing hard links
9346 Normally, when @command{tar} archives a hard link, it writes a
9347 block to the archive naming the target of the link (a @samp{1} type
9348 block). In that way, the actual file contents is stored in file only
9349 once. For example, consider the following two files:
9354 -rw-r--r-- 2 gray staff 4 2007-10-30 15:11 one
9355 -rw-r--r-- 2 gray staff 4 2007-10-30 15:11 jeden
9359 Here, @file{jeden} is a link to @file{one}. When archiving this
9360 directory with a verbose level 2, you will get an output similar to
9364 $ tar cfvv ../archive.tar .
9365 drwxr-xr-x gray/staff 0 2007-10-30 15:13 ./
9366 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./jeden
9367 hrw-r--r-- gray/staff 0 2007-10-30 15:11 ./one link to ./jeden
9370 The last line shows that, instead of storing two copies of the file,
9371 @command{tar} stored it only once, under the name @file{jeden}, and
9372 stored file @file{one} as a hard link to this file.
9374 It may be important to know that all hard links to the given file are
9375 stored in the archive. For example, this may be necessary for exact
9376 reproduction of the file system. The following option does that:
9379 @xopindex{check-links, described}
9382 Check the number of links dumped for each processed file. If this
9383 number does not match the total number of hard links for the file, print
9387 For example, trying to archive only file @file{jeden} with this option
9388 produces the following diagnostics:
9391 $ tar -c -f ../archive.tar -l jeden
9392 tar: Missing links to `jeden'.
9395 Although creating special records for hard links helps keep a faithful
9396 record of the file system contents and makes archives more compact, it
9397 may present some difficulties when extracting individual members from
9398 the archive. For example, trying to extract file @file{one} from the
9399 archive created in previous examples produces, in the absense of file
9403 $ tar xf archive.tar ./one
9404 tar: ./one: Cannot hard link to `./jeden': No such file or directory
9405 tar: Error exit delayed from previous errors
9408 The reason for this behavior is that @command{tar} cannot seek back in
9409 the archive to the previous member (in this case, @file{one}), to
9410 extract it@footnote{There are plans to fix this in future releases.}.
9411 If you wish to avoid such problems at the cost of a bigger archive,
9412 use the following option:
9415 @xopindex{hard-dereference, described}
9416 @item --hard-dereference
9417 Dereference hard links and store the files they refer to.
9420 For example, trying this option on our two sample files, we get two
9421 copies in the archive, each of which can then be extracted
9422 independently of the other:
9426 $ tar -c -vv -f ../archive.tar --hard-dereference .
9427 drwxr-xr-x gray/staff 0 2007-10-30 15:13 ./
9428 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./jeden
9429 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./one
9434 @subsection Old V7 Archives
9435 @cindex Format, old style
9436 @cindex Old style format
9437 @cindex Old style archives
9438 @cindex v7 archive format
9440 Certain old versions of @command{tar} cannot handle additional
9441 information recorded by newer @command{tar} programs. To create an
9442 archive in V7 format (not ANSI), which can be read by these old
9443 versions, specify the @option{--format=v7} option in
9444 conjunction with the @option{--create} (@option{-c}) (@command{tar} also
9445 accepts @option{--portability} or @option{--old-archive} for this
9446 option). When you specify it,
9447 @command{tar} leaves out information about directories, pipes, fifos,
9448 contiguous files, and device files, and specifies file ownership by
9449 group and user IDs instead of group and user names.
9451 When updating an archive, do not use @option{--format=v7}
9452 unless the archive was created using this option.
9454 In most cases, a @emph{new} format archive can be read by an @emph{old}
9455 @command{tar} program without serious trouble, so this option should
9456 seldom be needed. On the other hand, most modern @command{tar}s are
9457 able to read old format archives, so it might be safer for you to
9458 always use @option{--format=v7} for your distributions. Notice,
9459 however, that @samp{ustar} format is a better alternative, as it is
9460 free from many of @samp{v7}'s drawbacks.
9463 @subsection Ustar Archive Format
9465 @cindex ustar archive format
9466 Archive format defined by @acronym{POSIX}.1-1988 specification is called
9467 @code{ustar}. Although it is more flexible than the V7 format, it
9468 still has many restrictions (@pxref{Formats,ustar}, for the detailed
9469 description of @code{ustar} format). Along with V7 format,
9470 @code{ustar} format is a good choice for archives intended to be read
9471 with other implementations of @command{tar}.
9473 To create archive in @code{ustar} format, use @option{--format=ustar}
9474 option in conjunction with the @option{--create} (@option{-c}).
9477 @subsection @acronym{GNU} and old @GNUTAR{} format
9479 @cindex GNU archive format
9480 @cindex Old GNU archive format
9481 @GNUTAR{} was based on an early draft of the
9482 @acronym{POSIX} 1003.1 @code{ustar} standard. @acronym{GNU} extensions to
9483 @command{tar}, such as the support for file names longer than 100
9484 characters, use portions of the @command{tar} header record which were
9485 specified in that @acronym{POSIX} draft as unused. Subsequent changes in
9486 @acronym{POSIX} have allocated the same parts of the header record for
9487 other purposes. As a result, @GNUTAR{} format is
9488 incompatible with the current @acronym{POSIX} specification, and with
9489 @command{tar} programs that follow it.
9491 In the majority of cases, @command{tar} will be configured to create
9492 this format by default. This will change in future releases, since
9493 we plan to make @samp{POSIX} format the default.
9495 To force creation a @GNUTAR{} archive, use option
9496 @option{--format=gnu}.
9499 @subsection @GNUTAR{} and @acronym{POSIX} @command{tar}
9501 @cindex POSIX archive format
9502 @cindex PAX archive format
9503 Starting from version 1.14 @GNUTAR{} features full support for
9504 @acronym{POSIX.1-2001} archives.
9506 A @acronym{POSIX} conformant archive will be created if @command{tar}
9507 was given @option{--format=posix} (@option{--format=pax}) option. No
9508 special option is required to read and extract from a @acronym{POSIX}
9512 * PAX keywords:: Controlling Extended Header Keywords.
9516 @subsubsection Controlling Extended Header Keywords
9520 @item --pax-option=@var{keyword-list}
9521 Handle keywords in @acronym{PAX} extended headers. This option is
9522 equivalent to @option{-o} option of the @command{pax} utility.
9525 @var{Keyword-list} is a comma-separated
9526 list of keyword options, each keyword option taking one of
9527 the following forms:
9530 @item delete=@var{pattern}
9531 When used with one of archive-creation commands,
9532 this option instructs @command{tar} to omit from extended header records
9533 that it produces any keywords matching the string @var{pattern}.
9535 When used in extract or list mode, this option instructs tar
9536 to ignore any keywords matching the given @var{pattern} in the extended
9537 header records. In both cases, matching is performed using the pattern
9538 matching notation described in @acronym{POSIX 1003.2}, 3.13
9539 (@pxref{wildcards}). For example:
9542 --pax-option delete=security.*
9545 would suppress security-related information.
9547 @item exthdr.name=@var{string}
9549 This keyword allows user control over the name that is written into the
9550 ustar header blocks for the extended headers. The name is obtained
9551 from @var{string} after making the following substitutions:
9553 @multitable @columnfractions .25 .55
9554 @headitem Meta-character @tab Replaced By
9555 @item %d @tab The directory name of the file, equivalent to the
9556 result of the @command{dirname} utility on the translated file name.
9557 @item %f @tab The name of the file with the directory information
9558 stripped, equivalent to the result of the @command{basename} utility
9559 on the translated file name.
9560 @item %p @tab The process @acronym{ID} of the @command{tar} process.
9561 @item %% @tab A @samp{%} character.
9564 Any other @samp{%} characters in @var{string} produce undefined
9567 If no option @samp{exthdr.name=string} is specified, @command{tar}
9568 will use the following default value:
9574 @item exthdr.mtime=@var{value}
9576 This keyword defines the value of the @samp{mtime} field that
9577 is written into the ustar header blocks for the extended headers.
9578 By default, the @samp{mtime} field is set to the modification time
9579 of the archive member described by that extended headers.
9581 @item globexthdr.name=@var{string}
9582 This keyword allows user control over the name that is written into
9583 the ustar header blocks for global extended header records. The name
9584 is obtained from the contents of @var{string}, after making
9585 the following substitutions:
9587 @multitable @columnfractions .25 .55
9588 @headitem Meta-character @tab Replaced By
9589 @item %n @tab An integer that represents the
9590 sequence number of the global extended header record in the archive,
9592 @item %p @tab The process @acronym{ID} of the @command{tar} process.
9593 @item %% @tab A @samp{%} character.
9596 Any other @samp{%} characters in @var{string} produce undefined results.
9598 If no option @samp{globexthdr.name=string} is specified, @command{tar}
9599 will use the following default value:
9602 $TMPDIR/GlobalHead.%p.%n
9606 where @samp{$TMPDIR} represents the value of the @var{TMPDIR}
9607 environment variable. If @var{TMPDIR} is not set, @command{tar}
9610 @item globexthdr.mtime=@var{value}
9612 This keyword defines the value of the @samp{mtime} field that
9613 is written into the ustar header blocks for the global extended headers.
9614 By default, the @samp{mtime} field is set to the time when
9615 @command{tar} was invoked.
9617 @item @var{keyword}=@var{value}
9618 When used with one of archive-creation commands, these keyword/value pairs
9619 will be included at the beginning of the archive in a global extended
9620 header record. When used with one of archive-reading commands,
9621 @command{tar} will behave as if it has encountered these keyword/value
9622 pairs at the beginning of the archive in a global extended header
9625 @item @var{keyword}:=@var{value}
9626 When used with one of archive-creation commands, these keyword/value pairs
9627 will be included as records at the beginning of an extended header for
9628 each file. This is effectively equivalent to @var{keyword}=@var{value}
9629 form except that it creates no global extended header records.
9631 When used with one of archive-reading commands, @command{tar} will
9632 behave as if these keyword/value pairs were included as records at the
9633 end of each extended header; thus, they will override any global or
9634 file-specific extended header record keywords of the same names.
9635 For example, in the command:
9638 tar --format=posix --create \
9639 --file archive --pax-option gname:=user .
9642 the group name will be forced to a new value for all files
9643 stored in the archive.
9646 In any of the forms described above, the @var{value} may be
9647 a string enclosed in curly braces. In that case, the string
9648 between the braces is understood either as a textual time
9649 representation, as described in @ref{Date input formats}, or a name of
9650 the existing file, starting with @samp{/} or @samp{.}. In the latter
9651 case, the modification time of that file is used.
9653 For example, to set all modification times to the current date, you
9654 use the following option:
9657 --pax-option='mtime:=@{now@}'
9660 Note quoting of the option's argument.
9662 @cindex archives, binary equivalent
9663 @cindex binary equivalent archives, creating
9664 As another example, here is the option that ensures that any two
9665 archives created using it, will be binary equivalent if they have the
9669 --pax-option=exthdr.name=%d/PaxHeaders/%f,atime:=0
9673 @subsection Checksumming Problems
9675 SunOS and HP-UX @command{tar} fail to accept archives created using
9676 @GNUTAR{} and containing non-@acronym{ASCII} file names, that
9677 is, file names having characters with the eight bit set, because they
9678 use signed checksums, while @GNUTAR{} uses unsigned
9679 checksums while creating archives, as per @acronym{POSIX} standards. On
9680 reading, @GNUTAR{} computes both checksums and
9681 accepts any. It is somewhat worrying that a lot of people may go
9682 around doing backup of their files using faulty (or at least
9683 non-standard) software, not learning about it until it's time to
9684 restore their missing files with an incompatible file extractor, or
9687 @GNUTAR{} computes checksums both ways, and accept
9688 any on read, so @acronym{GNU} tar can read Sun tapes even with their
9689 wrong checksums. @GNUTAR{} produces the standard
9690 checksum, however, raising incompatibilities with Sun. That is to
9691 say, @GNUTAR{} has not been modified to
9692 @emph{produce} incorrect archives to be read by buggy @command{tar}'s.
9693 I've been told that more recent Sun @command{tar} now read standard
9694 archives, so maybe Sun did a similar patch, after all?
9696 The story seems to be that when Sun first imported @command{tar}
9697 sources on their system, they recompiled it without realizing that
9698 the checksums were computed differently, because of a change in
9699 the default signing of @code{char}'s in their compiler. So they
9700 started computing checksums wrongly. When they later realized their
9701 mistake, they merely decided to stay compatible with it, and with
9702 themselves afterwards. Presumably, but I do not really know, HP-UX
9703 has chosen that their @command{tar} archives to be compatible with Sun's.
9704 The current standards do not favor Sun @command{tar} format. In any
9705 case, it now falls on the shoulders of SunOS and HP-UX users to get
9706 a @command{tar} able to read the good archives they receive.
9708 @node Large or Negative Values
9709 @subsection Large or Negative Values
9710 @cindex large values
9711 @cindex future time stamps
9712 @cindex negative time stamps
9715 The above sections suggest to use @samp{oldest possible} archive
9716 format if in doubt. However, sometimes it is not possible. If you
9717 attempt to archive a file whose metadata cannot be represented using
9718 required format, @GNUTAR{} will print error message and ignore such a
9719 file. You will than have to switch to a format that is able to
9720 handle such values. The format summary table (@pxref{Formats}) will
9723 In particular, when trying to archive files larger than 8GB or with
9724 timestamps not in the range 1970-01-01 00:00:00 through 2242-03-16
9725 12:56:31 @sc{utc}, you will have to chose between @acronym{GNU} and
9726 @acronym{POSIX} archive formats. When considering which format to
9727 choose, bear in mind that the @acronym{GNU} format uses
9728 two's-complement base-256 notation to store values that do not fit
9729 into standard @acronym{ustar} range. Such archives can generally be
9730 read only by a @GNUTAR{} implementation. Moreover, they sometimes
9731 cannot be correctly restored on another hosts even by @GNUTAR{}. For
9732 example, using two's complement representation for negative time
9733 stamps that assumes a signed 32-bit @code{time_t} generates archives
9734 that are not portable to hosts with differing @code{time_t}
9737 On the other hand, @acronym{POSIX} archives, generally speaking, can
9738 be extracted by any tar implementation that understands older
9739 @acronym{ustar} format. The only exception are files larger than 8GB.
9741 @FIXME{Describe how @acronym{POSIX} archives are extracted by non
9745 @subsection How to Extract GNU-Specific Data Using Other @command{tar} Implementations
9747 In previous sections you became acquainted with various quirks
9748 necessary to make your archives portable. Sometimes you may need to
9749 extract archives containing GNU-specific members using some
9750 third-party @command{tar} implementation or an older version of
9751 @GNUTAR{}. Of course your best bet is to have @GNUTAR{} installed,
9752 but if it is for some reason impossible, this section will explain
9753 how to cope without it.
9755 When we speak about @dfn{GNU-specific} members we mean two classes of
9756 them: members split between the volumes of a multi-volume archive and
9757 sparse members. You will be able to always recover such members if
9758 the archive is in PAX format. In addition split members can be
9759 recovered from archives in old GNU format. The following subsections
9760 describe the required procedures in detail.
9763 * Split Recovery:: Members Split Between Volumes
9764 * Sparse Recovery:: Sparse Members
9767 @node Split Recovery
9768 @subsubsection Extracting Members Split Between Volumes
9770 @cindex Mutli-volume archives, extracting using non-GNU tars
9771 If a member is split between several volumes of an old GNU format archive
9772 most third party @command{tar} implementation will fail to extract
9773 it. To extract it, use @command{tarcat} program (@pxref{Tarcat}).
9774 This program is available from
9775 @uref{http://www.gnu.org/@/software/@/tar/@/utils/@/tarcat.html, @GNUTAR{}
9776 home page}. It concatenates several archive volumes into a single
9777 valid archive. For example, if you have three volumes named from
9778 @file{vol-1.tar} to @file{vol-3.tar}, you can do the following to
9779 extract them using a third-party @command{tar}:
9782 $ @kbd{tarcat vol-1.tar vol-2.tar vol-3.tar | tar xf -}
9785 @cindex Mutli-volume archives in PAX format, extracting using non-GNU tars
9786 You could use this approach for most (although not all) PAX
9787 format archives as well. However, extracting split members from a PAX
9788 archive is a much easier task, because PAX volumes are constructed in
9789 such a way that each part of a split member is extracted to a
9790 different file by @command{tar} implementations that are not aware of
9791 GNU extensions. More specifically, the very first part retains its
9792 original name, and all subsequent parts are named using the pattern:
9795 %d/GNUFileParts.%p/%f.%n
9799 where symbols preceeded by @samp{%} are @dfn{macro characters} that
9800 have the following meaning:
9802 @multitable @columnfractions .25 .55
9803 @headitem Meta-character @tab Replaced By
9804 @item %d @tab The directory name of the file, equivalent to the
9805 result of the @command{dirname} utility on its full name.
9806 @item %f @tab The file name of the file, equivalent to the result
9807 of the @command{basename} utility on its full name.
9808 @item %p @tab The process @acronym{ID} of the @command{tar} process that
9809 created the archive.
9810 @item %n @tab Ordinal number of this particular part.
9813 For example, if the file @file{var/longfile} was split during archive
9814 creation between three volumes, and the creator @command{tar} process
9815 had process @acronym{ID} @samp{27962}, then the member names will be:
9819 var/GNUFileParts.27962/longfile.1
9820 var/GNUFileParts.27962/longfile.2
9823 When you extract your archive using a third-party @command{tar}, these
9824 files will be created on your disk, and the only thing you will need
9825 to do to restore your file in its original form is concatenate them in
9826 the proper order, for example:
9831 $ @kbd{cat GNUFileParts.27962/longfile.1 \
9832 GNUFileParts.27962/longfile.2 >> longfile}
9833 $ rm -f GNUFileParts.27962
9837 Notice, that if the @command{tar} implementation you use supports PAX
9838 format archives, it will probably emit warnings about unknown keywords
9839 during extraction. They will look like this:
9844 Unknown extended header keyword 'GNU.volume.filename' ignored.
9845 Unknown extended header keyword 'GNU.volume.size' ignored.
9846 Unknown extended header keyword 'GNU.volume.offset' ignored.
9851 You can safely ignore these warnings.
9853 If your @command{tar} implementation is not PAX-aware, you will get
9854 more warnings and more files generated on your disk, e.g.:
9858 $ @kbd{tar xf vol-1.tar}
9859 var/PaxHeaders.27962/longfile: Unknown file type 'x', extracted as
9861 Unexpected EOF in archive
9862 $ @kbd{tar xf vol-2.tar}
9863 tmp/GlobalHead.27962.1: Unknown file type 'g', extracted as normal file
9864 GNUFileParts.27962/PaxHeaders.27962/sparsefile.1: Unknown file type
9865 'x', extracted as normal file
9869 Ignore these warnings. The @file{PaxHeaders.*} directories created
9870 will contain files with @dfn{extended header keywords} describing the
9871 extracted files. You can delete them, unless they describe sparse
9872 members. Read further to learn more about them.
9874 @node Sparse Recovery
9875 @subsubsection Extracting Sparse Members
9877 @cindex sparse files, extracting with non-GNU tars
9878 Any @command{tar} implementation will be able to extract sparse members from a
9879 PAX archive. However, the extracted files will be @dfn{condensed},
9880 i.e., any zero blocks will be removed from them. When we restore such
9881 a condensed file to its original form, by adding zero blocks (or
9882 @dfn{holes}) back to their original locations, we call this process
9883 @dfn{expanding} a compressed sparse file.
9886 To expand a file, you will need a simple auxiliary program called
9887 @command{xsparse}. It is available in source form from
9888 @uref{http://www.gnu.org/@/software/@/tar/@/utils/@/xsparse.html, @GNUTAR{}
9891 @cindex sparse files v.1.0, extracting with non-GNU tars
9892 Let's begin with archive members in @dfn{sparse format
9893 version 1.0}@footnote{@xref{PAX 1}.}, which are the easiest to expand.
9894 The condensed file will contain both file map and file data, so no
9895 additional data will be needed to restore it. If the original file
9896 name was @file{@var{dir}/@var{name}}, then the condensed file will be
9897 named @file{@var{dir}/@/GNUSparseFile.@var{n}/@/@var{name}}, where
9898 @var{n} is a decimal number@footnote{Technically speaking, @var{n} is a
9899 @dfn{process @acronym{ID}} of the @command{tar} process which created the
9900 archive (@pxref{PAX keywords}).}.
9902 To expand a version 1.0 file, run @command{xsparse} as follows:
9905 $ @kbd{xsparse @file{cond-file}}
9909 where @file{cond-file} is the name of the condensed file. The utility
9910 will deduce the name for the resulting expanded file using the
9911 following algorithm:
9914 @item If @file{cond-file} does not contain any directories,
9915 @file{../cond-file} will be used;
9917 @item If @file{cond-file} has the form
9918 @file{@var{dir}/@var{t}/@var{name}}, where both @var{t} and @var{name}
9919 are simple names, with no @samp{/} characters in them, the output file
9920 name will be @file{@var{dir}/@var{name}}.
9922 @item Otherwise, if @file{cond-file} has the form
9923 @file{@var{dir}/@var{name}}, the output file name will be
9927 In the unlikely case when this algorithm does not suit your needs,
9928 you can explicitly specify output file name as a second argument to
9932 $ @kbd{xsparse @file{cond-file} @file{out-file}}
9935 It is often a good idea to run @command{xsparse} in @dfn{dry run} mode
9936 first. In this mode, the command does not actually expand the file,
9937 but verbosely lists all actions it would be taking to do so. The dry
9938 run mode is enabled by @option{-n} command line argument:
9942 $ @kbd{xsparse -n /home/gray/GNUSparseFile.6058/sparsefile}
9943 Reading v.1.0 sparse map
9944 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
9945 `/home/gray/sparsefile'
9950 To actually expand the file, you would run:
9953 $ @kbd{xsparse /home/gray/GNUSparseFile.6058/sparsefile}
9957 The program behaves the same way all UNIX utilities do: it will keep
9958 quiet unless it has simething important to tell you (e.g. an error
9959 condition or something). If you wish it to produce verbose output,
9960 similar to that from the dry run mode, use @option{-v} option:
9964 $ @kbd{xsparse -v /home/gray/GNUSparseFile.6058/sparsefile}
9965 Reading v.1.0 sparse map
9966 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
9967 `/home/gray/sparsefile'
9972 Additionally, if your @command{tar} implementation has extracted the
9973 @dfn{extended headers} for this file, you can instruct @command{xstar}
9974 to use them in order to verify the integrity of the expanded file.
9975 The option @option{-x} sets the name of the extended header file to
9976 use. Continuing our example:
9980 $ @kbd{xsparse -v -x /home/gray/PaxHeaders.6058/sparsefile \
9981 /home/gray/GNUSparseFile.6058/sparsefile}
9982 Reading extended header file
9983 Found variable GNU.sparse.major = 1
9984 Found variable GNU.sparse.minor = 0
9985 Found variable GNU.sparse.name = sparsefile
9986 Found variable GNU.sparse.realsize = 217481216
9987 Reading v.1.0 sparse map
9988 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
9989 `/home/gray/sparsefile'
9994 @anchor{extracting sparse v.0.x}
9995 @cindex sparse files v.0.1, extracting with non-GNU tars
9996 @cindex sparse files v.0.0, extracting with non-GNU tars
9997 An @dfn{extended header} is a special @command{tar} archive header
9998 that precedes an archive member and contains a set of
9999 @dfn{variables}, describing the member properties that cannot be
10000 stored in the standard @code{ustar} header. While optional for
10001 expanding sparse version 1.0 members, the use of extended headers is
10002 mandatory when expanding sparse members in older sparse formats: v.0.0
10003 and v.0.1 (The sparse formats are described in detail in @ref{Sparse
10004 Formats}.) So, for these formats, the question is: how to obtain
10005 extended headers from the archive?
10007 If you use a @command{tar} implementation that does not support PAX
10008 format, extended headers for each member will be extracted as a
10009 separate file. If we represent the member name as
10010 @file{@var{dir}/@var{name}}, then the extended header file will be
10011 named @file{@var{dir}/@/PaxHeaders.@var{n}/@/@var{name}}, where
10012 @var{n} is an integer number.
10014 Things become more difficult if your @command{tar} implementation
10015 does support PAX headers, because in this case you will have to
10016 manually extract the headers. We recommend the following algorithm:
10020 Consult the documentation of your @command{tar} implementation for an
10021 option that prints @dfn{block numbers} along with the archive
10022 listing (analogous to @GNUTAR{}'s @option{-R} option). For example,
10023 @command{star} has @option{-block-number}.
10026 Obtain verbose listing using the @samp{block number} option, and
10027 find block numbers of the sparse member in question and the member
10028 immediately following it. For example, running @command{star} on our
10033 $ @kbd{star -t -v -block-number -f arc.tar}
10035 star: Unknown extended header keyword 'GNU.sparse.size' ignored.
10036 star: Unknown extended header keyword 'GNU.sparse.numblocks' ignored.
10037 star: Unknown extended header keyword 'GNU.sparse.name' ignored.
10038 star: Unknown extended header keyword 'GNU.sparse.map' ignored.
10039 block 56: 425984 -rw-r--r-- gray/users Jun 25 14:46 2006 GNUSparseFile.28124/sparsefile
10040 block 897: 65391 -rw-r--r-- gray/users Jun 24 20:06 2006 README
10046 (as usual, ignore the warnings about unknown keywords.)
10049 Let @var{size} be the size of the sparse member, @var{Bs} be its block number
10050 and @var{Bn} be the block number of the next member.
10054 @var{N} = @var{Bs} - @var{Bn} - @var{size}/512 - 2
10058 This number gives the size of the extended header part in tar @dfn{blocks}.
10059 In our example, this formula gives: @code{897 - 56 - 425984 / 512 - 2
10063 Use @command{dd} to extract the headers:
10066 @kbd{dd if=@var{archive} of=@var{hname} bs=512 skip=@var{Bs} count=@var{N}}
10070 where @var{archive} is the archive name, @var{hname} is a name of the
10071 file to store the extended header in, @var{Bs} and @var{N} are
10072 computed in previous steps.
10074 In our example, this command will be
10077 $ @kbd{dd if=arc.tar of=xhdr bs=512 skip=56 count=7}
10081 Finally, you can expand the condensed file, using the obtained header:
10085 $ @kbd{xsparse -v -x xhdr GNUSparseFile.6058/sparsefile}
10086 Reading extended header file
10087 Found variable GNU.sparse.size = 217481216
10088 Found variable GNU.sparse.numblocks = 208
10089 Found variable GNU.sparse.name = sparsefile
10090 Found variable GNU.sparse.map = 0,2048,1050624,2048,@dots{}
10091 Expanding file `GNUSparseFile.28124/sparsefile' to `sparsefile'
10097 @section Comparison of @command{tar} and @command{cpio}
10100 @FIXME{Reorganize the following material}
10102 The @command{cpio} archive formats, like @command{tar}, do have maximum
10103 file name lengths. The binary and old @acronym{ASCII} formats have a maximum file
10104 length of 256, and the new @acronym{ASCII} and @acronym{CRC ASCII} formats have a max
10105 file length of 1024. @acronym{GNU} @command{cpio} can read and write archives
10106 with arbitrary file name lengths, but other @command{cpio} implementations
10107 may crash unexplainedly trying to read them.
10109 @command{tar} handles symbolic links in the form in which it comes in @acronym{BSD};
10110 @command{cpio} doesn't handle symbolic links in the form in which it comes
10111 in System V prior to SVR4, and some vendors may have added symlinks
10112 to their system without enhancing @command{cpio} to know about them.
10113 Others may have enhanced it in a way other than the way I did it
10114 at Sun, and which was adopted by AT&T (and which is, I think, also
10115 present in the @command{cpio} that Berkeley picked up from AT&T and put
10116 into a later @acronym{BSD} release---I think I gave them my changes).
10118 (SVR4 does some funny stuff with @command{tar}; basically, its @command{cpio}
10119 can handle @command{tar} format input, and write it on output, and it
10120 probably handles symbolic links. They may not have bothered doing
10121 anything to enhance @command{tar} as a result.)
10123 @command{cpio} handles special files; traditional @command{tar} doesn't.
10125 @command{tar} comes with V7, System III, System V, and @acronym{BSD} source;
10126 @command{cpio} comes only with System III, System V, and later @acronym{BSD}
10127 (4.3-tahoe and later).
10129 @command{tar}'s way of handling multiple hard links to a file can handle
10130 file systems that support 32-bit i-numbers (e.g., the @acronym{BSD} file system);
10131 @command{cpio}s way requires you to play some games (in its ``binary''
10132 format, i-numbers are only 16 bits, and in its ``portable @acronym{ASCII}'' format,
10133 they're 18 bits---it would have to play games with the "file system @acronym{ID}"
10134 field of the header to make sure that the file system @acronym{ID}/i-number pairs
10135 of different files were always different), and I don't know which
10136 @command{cpio}s, if any, play those games. Those that don't might get
10137 confused and think two files are the same file when they're not, and
10138 make hard links between them.
10140 @command{tar}s way of handling multiple hard links to a file places only
10141 one copy of the link on the tape, but the name attached to that copy
10142 is the @emph{only} one you can use to retrieve the file; @command{cpio}s
10143 way puts one copy for every link, but you can retrieve it using any
10147 What type of check sum (if any) is used, and how is this calculated.
10150 See the attached manual pages for @command{tar} and @command{cpio} format.
10151 @command{tar} uses a checksum which is the sum of all the bytes in the
10152 @command{tar} header for a file; @command{cpio} uses no checksum.
10155 If anyone knows why @command{cpio} was made when @command{tar} was present
10159 It wasn't. @command{cpio} first showed up in PWB/UNIX 1.0; no
10160 generally-available version of UNIX had @command{tar} at the time. I don't
10161 know whether any version that was generally available @emph{within AT&T}
10162 had @command{tar}, or, if so, whether the people within AT&T who did
10163 @command{cpio} knew about it.
10165 On restore, if there is a corruption on a tape @command{tar} will stop at
10166 that point, while @command{cpio} will skip over it and try to restore the
10169 The main difference is just in the command syntax and header format.
10171 @command{tar} is a little more tape-oriented in that everything is blocked
10172 to start on a record boundary.
10175 Is there any differences between the ability to recover crashed
10176 archives between the two of them. (Is there any chance of recovering
10177 crashed archives at all.)
10180 Theoretically it should be easier under @command{tar} since the blocking
10181 lets you find a header with some variation of @samp{dd skip=@var{nn}}.
10182 However, modern @command{cpio}'s and variations have an option to just
10183 search for the next file header after an error with a reasonable chance
10184 of resyncing. However, lots of tape driver software won't allow you to
10185 continue past a media error which should be the only reason for getting
10186 out of sync unless a file changed sizes while you were writing the
10190 If anyone knows why @command{cpio} was made when @command{tar} was present
10191 at the unix scene, please tell me about this too.
10194 Probably because it is more media efficient (by not blocking everything
10195 and using only the space needed for the headers where @command{tar}
10196 always uses 512 bytes per file header) and it knows how to archive
10199 You might want to look at the freely available alternatives. The
10200 major ones are @command{afio}, @GNUTAR{}, and
10201 @command{pax}, each of which have their own extensions with some
10202 backwards compatibility.
10204 Sparse files were @command{tar}red as sparse files (which you can
10205 easily test, because the resulting archive gets smaller, and
10206 @acronym{GNU} @command{cpio} can no longer read it).
10209 @chapter Tapes and Other Archive Media
10212 A few special cases about tape handling warrant more detailed
10213 description. These special cases are discussed below.
10215 Many complexities surround the use of @command{tar} on tape drives. Since
10216 the creation and manipulation of archives located on magnetic tape was
10217 the original purpose of @command{tar}, it contains many features making
10218 such manipulation easier.
10220 Archives are usually written on dismountable media---tape cartridges,
10221 mag tapes, or floppy disks.
10223 The amount of data a tape or disk holds depends not only on its size,
10224 but also on how it is formatted. A 2400 foot long reel of mag tape
10225 holds 40 megabytes of data when formatted at 1600 bits per inch. The
10226 physically smaller EXABYTE tape cartridge holds 2.3 gigabytes.
10228 Magnetic media are re-usable---once the archive on a tape is no longer
10229 needed, the archive can be erased and the tape or disk used over.
10230 Media quality does deteriorate with use, however. Most tapes or disks
10231 should be discarded when they begin to produce data errors. EXABYTE
10232 tape cartridges should be discarded when they generate an @dfn{error
10233 count} (number of non-usable bits) of more than 10k.
10235 Magnetic media are written and erased using magnetic fields, and
10236 should be protected from such fields to avoid damage to stored data.
10237 Sticking a floppy disk to a filing cabinet using a magnet is probably
10241 * Device:: Device selection and switching
10242 * Remote Tape Server::
10243 * Common Problems and Solutions::
10244 * Blocking:: Blocking
10245 * Many:: Many archives on one tape
10246 * Using Multiple Tapes:: Using Multiple Tapes
10247 * label:: Including a Label in the Archive
10249 * Write Protection::
10253 @section Device Selection and Switching
10257 @item -f [@var{hostname}:]@var{file}
10258 @itemx --file=[@var{hostname}:]@var{file}
10259 Use archive file or device @var{file} on @var{hostname}.
10262 This option is used to specify the file name of the archive @command{tar}
10265 If the file name is @samp{-}, @command{tar} reads the archive from standard
10266 input (when listing or extracting), or writes it to standard output
10267 (when creating). If the @samp{-} file name is given when updating an
10268 archive, @command{tar} will read the original archive from its standard
10269 input, and will write the entire new archive to its standard output.
10271 If the file name contains a @samp{:}, it is interpreted as
10272 @samp{hostname:file name}. If the @var{hostname} contains an @dfn{at}
10273 sign (@samp{@@}), it is treated as @samp{user@@hostname:file name}. In
10274 either case, @command{tar} will invoke the command @command{rsh} (or
10275 @command{remsh}) to start up an @command{/usr/libexec/rmt} on the remote
10276 machine. If you give an alternate login name, it will be given to the
10278 Naturally, the remote machine must have an executable
10279 @command{/usr/libexec/rmt}. This program is free software from the
10280 University of California, and a copy of the source code can be found
10281 with the sources for @command{tar}; it's compiled and installed by default.
10282 The exact path to this utility is determined when configuring the package.
10283 It is @file{@var{prefix}/libexec/rmt}, where @var{prefix} stands for
10284 your installation prefix. This location may also be overridden at
10285 runtime by using the @option{--rmt-command=@var{command}} option (@xref{Option Summary,
10286 ---rmt-command}, for detailed description of this option. @xref{Remote
10287 Tape Server}, for the description of @command{rmt} command).
10289 If this option is not given, but the environment variable @env{TAPE}
10290 is set, its value is used; otherwise, old versions of @command{tar}
10291 used a default archive name (which was picked when @command{tar} was
10292 compiled). The default is normally set up to be the @dfn{first} tape
10293 drive or other transportable I/O medium on the system.
10295 Starting with version 1.11.5, @GNUTAR{} uses
10296 standard input and standard output as the default device, and I will
10297 not try anymore supporting automatic device detection at installation
10298 time. This was failing really in too many cases, it was hopeless.
10299 This is now completely left to the installer to override standard
10300 input and standard output for default device, if this seems
10301 preferable. Further, I think @emph{most} actual usages of
10302 @command{tar} are done with pipes or disks, not really tapes,
10303 cartridges or diskettes.
10305 Some users think that using standard input and output is running
10306 after trouble. This could lead to a nasty surprise on your screen if
10307 you forget to specify an output file name---especially if you are going
10308 through a network or terminal server capable of buffering large amounts
10309 of output. We had so many bug reports in that area of configuring
10310 default tapes automatically, and so many contradicting requests, that
10311 we finally consider the problem to be portably intractable. We could
10312 of course use something like @samp{/dev/tape} as a default, but this
10313 is @emph{also} running after various kind of trouble, going from hung
10314 processes to accidental destruction of real tapes. After having seen
10315 all this mess, using standard input and output as a default really
10316 sounds like the only clean choice left, and a very useful one too.
10318 @GNUTAR{} reads and writes archive in records, I
10319 suspect this is the main reason why block devices are preferred over
10320 character devices. Most probably, block devices are more efficient
10321 too. The installer could also check for @samp{DEFTAPE} in
10322 @file{<sys/mtio.h>}.
10325 @xopindex{force-local, short description}
10326 @item --force-local
10327 Archive file is local even if it contains a colon.
10329 @opindex rsh-command
10330 @item --rsh-command=@var{command}
10331 Use remote @var{command} instead of @command{rsh}. This option exists
10332 so that people who use something other than the standard @command{rsh}
10333 (e.g., a Kerberized @command{rsh}) can access a remote device.
10335 When this command is not used, the shell command found when
10336 the @command{tar} program was installed is used instead. This is
10337 the first found of @file{/usr/ucb/rsh}, @file{/usr/bin/remsh},
10338 @file{/usr/bin/rsh}, @file{/usr/bsd/rsh} or @file{/usr/bin/nsh}.
10339 The installer may have overridden this by defining the environment
10340 variable @env{RSH} @emph{at installation time}.
10343 Specify drive and density.
10345 @xopindex{multi-volume, short description}
10347 @itemx --multi-volume
10348 Create/list/extract multi-volume archive.
10350 This option causes @command{tar} to write a @dfn{multi-volume} archive---one
10351 that may be larger than will fit on the medium used to hold it.
10352 @xref{Multi-Volume Archives}.
10354 @xopindex{tape-length, short description}
10356 @itemx --tape-length=@var{size}[@var{suf}]
10357 Change tape after writing @var{size} units of data. Unless @var{suf} is
10358 given, @var{size} is treated as kilobytes, i.e. @samp{@var{size} x
10359 1024} bytes. The following suffixes alter this behavior:
10361 @float Table, size-suffixes
10362 @caption{Size Suffixes}
10363 @multitable @columnfractions 0.2 0.3 0.3
10364 @headitem Suffix @tab Units @tab Byte Equivalent
10365 @item b @tab Blocks @tab @var{size} x 512
10366 @item B @tab Kilobytes @tab @var{size} x 1024
10367 @item c @tab Bytes @tab @var{size}
10368 @item G @tab Gigabytes @tab @var{size} x 1024^3
10369 @item K @tab Kilobytes @tab @var{size} x 1024
10370 @item k @tab Kilobytes @tab @var{size} x 1024
10371 @item M @tab Megabytes @tab @var{size} x 1024^2
10372 @item P @tab Petabytes @tab @var{size} x 1024^5
10373 @item T @tab Terabytes @tab @var{size} x 1024^4
10374 @item w @tab Words @tab @var{size} x 2
10378 This option might be useful when your tape drivers do not properly
10379 detect end of physical tapes. By being slightly conservative on the
10380 maximum tape length, you might avoid the problem entirely.
10382 @xopindex{info-script, short description}
10383 @xopindex{new-volume-script, short description}
10384 @item -F @var{file}
10385 @itemx --info-script=@var{file}
10386 @itemx --new-volume-script=@var{file}
10387 Execute @file{file} at end of each tape. This implies
10388 @option{--multi-volume} (@option{-M}). @xref{info-script}, for a detailed
10389 description of this option.
10392 @node Remote Tape Server
10393 @section Remote Tape Server
10395 @cindex remote tape drive
10397 In order to access the tape drive on a remote machine, @command{tar}
10398 uses the remote tape server written at the University of California at
10399 Berkeley. The remote tape server must be installed as
10400 @file{@var{prefix}/libexec/rmt} on any machine whose tape drive you
10401 want to use. @command{tar} calls @command{rmt} by running an
10402 @command{rsh} or @command{remsh} to the remote machine, optionally
10403 using a different login name if one is supplied.
10405 A copy of the source for the remote tape server is provided. It is
10406 Copyright @copyright{} 1983 by the Regents of the University of
10407 California, but can be freely distributed. It is compiled and
10408 installed by default.
10410 @cindex absolute file names
10411 Unless you use the @option{--absolute-names} (@option{-P}) option,
10412 @GNUTAR{} will not allow you to create an archive that contains
10413 absolute file names (a file name beginning with @samp{/}.) If you try,
10414 @command{tar} will automatically remove the leading @samp{/} from the
10415 file names it stores in the archive. It will also type a warning
10416 message telling you what it is doing.
10418 When reading an archive that was created with a different
10419 @command{tar} program, @GNUTAR{} automatically
10420 extracts entries in the archive which have absolute file names as if
10421 the file names were not absolute. This is an important feature. A
10422 visitor here once gave a @command{tar} tape to an operator to restore;
10423 the operator used Sun @command{tar} instead of @GNUTAR{},
10424 and the result was that it replaced large portions of
10425 our @file{/bin} and friends with versions from the tape; needless to
10426 say, we were unhappy about having to recover the file system from
10429 For example, if the archive contained a file @file{/usr/bin/computoy},
10430 @GNUTAR{} would extract the file to @file{usr/bin/computoy},
10431 relative to the current directory. If you want to extract the files in
10432 an archive to the same absolute names that they had when the archive
10433 was created, you should do a @samp{cd /} before extracting the files
10434 from the archive, or you should either use the @option{--absolute-names}
10435 option, or use the command @samp{tar -C / @dots{}}.
10437 @cindex Ultrix 3.1 and write failure
10438 Some versions of Unix (Ultrix 3.1 is known to have this problem),
10439 can claim that a short write near the end of a tape succeeded,
10440 when it actually failed. This will result in the -M option not
10441 working correctly. The best workaround at the moment is to use a
10442 significantly larger blocking factor than the default 20.
10444 In order to update an archive, @command{tar} must be able to backspace the
10445 archive in order to reread or rewrite a record that was just read (or
10446 written). This is currently possible only on two kinds of files: normal
10447 disk files (or any other file that can be backspaced with @samp{lseek}),
10448 and industry-standard 9-track magnetic tape (or any other kind of tape
10449 that can be backspaced with the @code{MTIOCTOP} @code{ioctl}).
10451 This means that the @option{--append}, @option{--concatenate}, and
10452 @option{--delete} commands will not work on any other kind of file.
10453 Some media simply cannot be backspaced, which means these commands and
10454 options will never be able to work on them. These non-backspacing
10455 media include pipes and cartridge tape drives.
10457 Some other media can be backspaced, and @command{tar} will work on them
10458 once @command{tar} is modified to do so.
10460 Archives created with the @option{--multi-volume}, @option{--label}, and
10461 @option{--incremental} (@option{-G}) options may not be readable by other version
10462 of @command{tar}. In particular, restoring a file that was split over
10463 a volume boundary will require some careful work with @command{dd}, if
10464 it can be done at all. Other versions of @command{tar} may also create
10465 an empty file whose name is that of the volume header. Some versions
10466 of @command{tar} may create normal files instead of directories archived
10467 with the @option{--incremental} (@option{-G}) option.
10469 @node Common Problems and Solutions
10470 @section Some Common Problems and their Solutions
10475 errors from system:
10477 no such file or directory
10480 errors from @command{tar}:
10481 directory checksum error
10482 header format error
10484 errors from media/system:
10496 @dfn{Block} and @dfn{record} terminology is rather confused, and it
10497 is also confusing to the expert reader. On the other hand, readers
10498 who are new to the field have a fresh mind, and they may safely skip
10499 the next two paragraphs, as the remainder of this manual uses those
10500 two terms in a quite consistent way.
10502 John Gilmore, the writer of the public domain @command{tar} from which
10503 @GNUTAR{} was originally derived, wrote (June 1995):
10506 The nomenclature of tape drives comes from IBM, where I believe
10507 they were invented for the IBM 650 or so. On IBM mainframes, what
10508 is recorded on tape are tape blocks. The logical organization of
10509 data is into records. There are various ways of putting records into
10510 blocks, including @code{F} (fixed sized records), @code{V} (variable
10511 sized records), @code{FB} (fixed blocked: fixed size records, @var{n}
10512 to a block), @code{VB} (variable size records, @var{n} to a block),
10513 @code{VSB} (variable spanned blocked: variable sized records that can
10514 occupy more than one block), etc. The @code{JCL} @samp{DD RECFORM=}
10515 parameter specified this to the operating system.
10517 The Unix man page on @command{tar} was totally confused about this.
10518 When I wrote @code{PD TAR}, I used the historically correct terminology
10519 (@command{tar} writes data records, which are grouped into blocks).
10520 It appears that the bogus terminology made it into @acronym{POSIX} (no surprise
10521 here), and now Fran@,{c}ois has migrated that terminology back
10522 into the source code too.
10525 The term @dfn{physical block} means the basic transfer chunk from or
10526 to a device, after which reading or writing may stop without anything
10527 being lost. In this manual, the term @dfn{block} usually refers to
10528 a disk physical block, @emph{assuming} that each disk block is 512
10529 bytes in length. It is true that some disk devices have different
10530 physical blocks, but @command{tar} ignore these differences in its own
10531 format, which is meant to be portable, so a @command{tar} block is always
10532 512 bytes in length, and @dfn{block} always mean a @command{tar} block.
10533 The term @dfn{logical block} often represents the basic chunk of
10534 allocation of many disk blocks as a single entity, which the operating
10535 system treats somewhat atomically; this concept is only barely used
10538 The term @dfn{physical record} is another way to speak of a physical
10539 block, those two terms are somewhat interchangeable. In this manual,
10540 the term @dfn{record} usually refers to a tape physical block,
10541 @emph{assuming} that the @command{tar} archive is kept on magnetic tape.
10542 It is true that archives may be put on disk or used with pipes,
10543 but nevertheless, @command{tar} tries to read and write the archive one
10544 @dfn{record} at a time, whatever the medium in use. One record is made
10545 up of an integral number of blocks, and this operation of putting many
10546 disk blocks into a single tape block is called @dfn{reblocking}, or
10547 more simply, @dfn{blocking}. The term @dfn{logical record} refers to
10548 the logical organization of many characters into something meaningful
10549 to the application. The term @dfn{unit record} describes a small set
10550 of characters which are transmitted whole to or by the application,
10551 and often refers to a line of text. Those two last terms are unrelated
10552 to what we call a @dfn{record} in @GNUTAR{}.
10554 When writing to tapes, @command{tar} writes the contents of the archive
10555 in chunks known as @dfn{records}. To change the default blocking
10556 factor, use the @option{--blocking-factor=@var{512-size}} (@option{-b
10557 @var{512-size}}) option. Each record will then be composed of
10558 @var{512-size} blocks. (Each @command{tar} block is 512 bytes.
10559 @xref{Standard}.) Each file written to the archive uses at least one
10560 full record. As a result, using a larger record size can result in
10561 more wasted space for small files. On the other hand, a larger record
10562 size can often be read and written much more efficiently.
10564 Further complicating the problem is that some tape drives ignore the
10565 blocking entirely. For these, a larger record size can still improve
10566 performance (because the software layers above the tape drive still
10567 honor the blocking), but not as dramatically as on tape drives that
10570 When reading an archive, @command{tar} can usually figure out the
10571 record size on itself. When this is the case, and a non-standard
10572 record size was used when the archive was created, @command{tar} will
10573 print a message about a non-standard blocking factor, and then operate
10574 normally. On some tape devices, however, @command{tar} cannot figure
10575 out the record size itself. On most of those, you can specify a
10576 blocking factor (with @option{--blocking-factor}) larger than the
10577 actual blocking factor, and then use the @option{--read-full-records}
10578 (@option{-B}) option. (If you specify a blocking factor with
10579 @option{--blocking-factor} and don't use the
10580 @option{--read-full-records} option, then @command{tar} will not
10581 attempt to figure out the recording size itself.) On some devices,
10582 you must always specify the record size exactly with
10583 @option{--blocking-factor} when reading, because @command{tar} cannot
10584 figure it out. In any case, use @option{--list} (@option{-t}) before
10585 doing any extractions to see whether @command{tar} is reading the archive
10588 @command{tar} blocks are all fixed size (512 bytes), and its scheme for
10589 putting them into records is to put a whole number of them (one or
10590 more) into each record. @command{tar} records are all the same size;
10591 at the end of the file there's a block containing all zeros, which
10592 is how you tell that the remainder of the last record(s) are garbage.
10594 In a standard @command{tar} file (no options), the block size is 512
10595 and the record size is 10240, for a blocking factor of 20. What the
10596 @option{--blocking-factor} option does is sets the blocking factor,
10597 changing the record size while leaving the block size at 512 bytes.
10598 20 was fine for ancient 800 or 1600 bpi reel-to-reel tape drives;
10599 most tape drives these days prefer much bigger records in order to
10600 stream and not waste tape. When writing tapes for myself, some tend
10601 to use a factor of the order of 2048, say, giving a record size of
10602 around one megabyte.
10604 If you use a blocking factor larger than 20, older @command{tar}
10605 programs might not be able to read the archive, so we recommend this
10606 as a limit to use in practice. @GNUTAR{}, however,
10607 will support arbitrarily large record sizes, limited only by the
10608 amount of virtual memory or the physical characteristics of the tape
10612 * Format Variations:: Format Variations
10613 * Blocking Factor:: The Blocking Factor of an Archive
10616 @node Format Variations
10617 @subsection Format Variations
10618 @cindex Format Parameters
10619 @cindex Format Options
10620 @cindex Options, archive format specifying
10621 @cindex Options, format specifying
10624 Format parameters specify how an archive is written on the archive
10625 media. The best choice of format parameters will vary depending on
10626 the type and number of files being archived, and on the media used to
10629 To specify format parameters when accessing or creating an archive,
10630 you can use the options described in the following sections.
10631 If you do not specify any format parameters, @command{tar} uses
10632 default parameters. You cannot modify a compressed archive.
10633 If you create an archive with the @option{--blocking-factor} option
10634 specified (@pxref{Blocking Factor}), you must specify that
10635 blocking-factor when operating on the archive. @xref{Formats}, for other
10636 examples of format parameter considerations.
10638 @node Blocking Factor
10639 @subsection The Blocking Factor of an Archive
10640 @cindex Blocking Factor
10641 @cindex Record Size
10642 @cindex Number of blocks per record
10643 @cindex Number of bytes per record
10644 @cindex Bytes per record
10645 @cindex Blocks per record
10648 @opindex blocking-factor
10649 The data in an archive is grouped into blocks, which are 512 bytes.
10650 Blocks are read and written in whole number multiples called
10651 @dfn{records}. The number of blocks in a record (i.e., the size of a
10652 record in units of 512 bytes) is called the @dfn{blocking factor}.
10653 The @option{--blocking-factor=@var{512-size}} (@option{-b
10654 @var{512-size}}) option specifies the blocking factor of an archive.
10655 The default blocking factor is typically 20 (i.e., 10240 bytes), but
10656 can be specified at installation. To find out the blocking factor of
10657 an existing archive, use @samp{tar --list --file=@var{archive-name}}.
10658 This may not work on some devices.
10660 Records are separated by gaps, which waste space on the archive media.
10661 If you are archiving on magnetic tape, using a larger blocking factor
10662 (and therefore larger records) provides faster throughput and allows you
10663 to fit more data on a tape (because there are fewer gaps). If you are
10664 archiving on cartridge, a very large blocking factor (say 126 or more)
10665 greatly increases performance. A smaller blocking factor, on the other
10666 hand, may be useful when archiving small files, to avoid archiving lots
10667 of nulls as @command{tar} fills out the archive to the end of the record.
10668 In general, the ideal record size depends on the size of the
10669 inter-record gaps on the tape you are using, and the average size of the
10670 files you are archiving. @xref{create}, for information on
10673 @FIXME{Need example of using a cartridge with blocking factor=126 or more.}
10675 Archives with blocking factors larger than 20 cannot be read
10676 by very old versions of @command{tar}, or by some newer versions
10677 of @command{tar} running on old machines with small address spaces.
10678 With @GNUTAR{}, the blocking factor of an archive is limited
10679 only by the maximum record size of the device containing the archive,
10680 or by the amount of available virtual memory.
10682 Also, on some systems, not using adequate blocking factors, as sometimes
10683 imposed by the device drivers, may yield unexpected diagnostics. For
10684 example, this has been reported:
10687 Cannot write to /dev/dlt: Invalid argument
10691 In such cases, it sometimes happen that the @command{tar} bundled by
10692 the system is aware of block size idiosyncrasies, while @GNUTAR{}
10693 requires an explicit specification for the block size,
10694 which it cannot guess. This yields some people to consider
10695 @GNUTAR{} is misbehaving, because by comparison,
10696 @cite{the bundle @command{tar} works OK}. Adding @w{@kbd{-b 256}},
10697 for example, might resolve the problem.
10699 If you use a non-default blocking factor when you create an archive, you
10700 must specify the same blocking factor when you modify that archive. Some
10701 archive devices will also require you to specify the blocking factor when
10702 reading that archive, however this is not typically the case. Usually, you
10703 can use @option{--list} (@option{-t}) without specifying a blocking factor---@command{tar}
10704 reports a non-default record size and then lists the archive members as
10705 it would normally. To extract files from an archive with a non-standard
10706 blocking factor (particularly if you're not sure what the blocking factor
10707 is), you can usually use the @option{--read-full-records} (@option{-B}) option while
10708 specifying a blocking factor larger then the blocking factor of the archive
10709 (i.e., @samp{tar --extract --read-full-records --blocking-factor=300}).
10710 @xref{list}, for more information on the @option{--list} (@option{-t})
10711 operation. @xref{Reading}, for a more detailed explanation of that option.
10714 @item --blocking-factor=@var{number}
10715 @itemx -b @var{number}
10716 Specifies the blocking factor of an archive. Can be used with any
10717 operation, but is usually not necessary with @option{--list} (@option{-t}).
10723 @item -b @var{blocks}
10724 @itemx --blocking-factor=@var{blocks}
10725 Set record size to @math{@var{blocks}*512} bytes.
10727 This option is used to specify a @dfn{blocking factor} for the archive.
10728 When reading or writing the archive, @command{tar}, will do reads and writes
10729 of the archive in records of @math{@var{block}*512} bytes. This is true
10730 even when the archive is compressed. Some devices requires that all
10731 write operations be a multiple of a certain size, and so, @command{tar}
10732 pads the archive out to the next record boundary.
10734 The default blocking factor is set when @command{tar} is compiled, and is
10735 typically 20. Blocking factors larger than 20 cannot be read by very
10736 old versions of @command{tar}, or by some newer versions of @command{tar}
10737 running on old machines with small address spaces.
10739 With a magnetic tape, larger records give faster throughput and fit
10740 more data on a tape (because there are fewer inter-record gaps).
10741 If the archive is in a disk file or a pipe, you may want to specify
10742 a smaller blocking factor, since a large one will result in a large
10743 number of null bytes at the end of the archive.
10745 When writing cartridge or other streaming tapes, a much larger
10746 blocking factor (say 126 or more) will greatly increase performance.
10747 However, you must specify the same blocking factor when reading or
10748 updating the archive.
10750 Apparently, Exabyte drives have a physical block size of 8K bytes.
10751 If we choose our blocksize as a multiple of 8k bytes, then the problem
10752 seems to disappear. Id est, we are using block size of 112 right
10753 now, and we haven't had the problem since we switched@dots{}
10755 With @GNUTAR{} the blocking factor is limited only
10756 by the maximum record size of the device containing the archive, or by
10757 the amount of available virtual memory.
10759 However, deblocking or reblocking is virtually avoided in a special
10760 case which often occurs in practice, but which requires all the
10761 following conditions to be simultaneously true:
10764 the archive is subject to a compression option,
10766 the archive is not handled through standard input or output, nor
10767 redirected nor piped,
10769 the archive is directly handled to a local disk, instead of any special
10772 @option{--blocking-factor} is not explicitly specified on the @command{tar}
10776 If the output goes directly to a local disk, and not through
10777 stdout, then the last write is not extended to a full record size.
10778 Otherwise, reblocking occurs. Here are a few other remarks on this
10784 @command{gzip} will complain about trailing garbage if asked to
10785 uncompress a compressed archive on tape, there is an option to turn
10786 the message off, but it breaks the regularity of simply having to use
10787 @samp{@var{prog} -d} for decompression. It would be nice if gzip was
10788 silently ignoring any number of trailing zeros. I'll ask Jean-loup
10789 Gailly, by sending a copy of this message to him.
10792 @command{compress} does not show this problem, but as Jean-loup pointed
10793 out to Michael, @samp{compress -d} silently adds garbage after
10794 the result of decompression, which tar ignores because it already
10795 recognized its end-of-file indicator. So this bug may be safely
10799 @samp{gzip -d -q} will be silent about the trailing zeros indeed,
10800 but will still return an exit status of 2 which tar reports in turn.
10801 @command{tar} might ignore the exit status returned, but I hate doing
10802 that, as it weakens the protection @command{tar} offers users against
10803 other possible problems at decompression time. If @command{gzip} was
10804 silently skipping trailing zeros @emph{and} also avoiding setting the
10805 exit status in this innocuous case, that would solve this situation.
10808 @command{tar} should become more solid at not stopping to read a pipe at
10809 the first null block encountered. This inelegantly breaks the pipe.
10810 @command{tar} should rather drain the pipe out before exiting itself.
10813 @xopindex{ignore-zeros, short description}
10815 @itemx --ignore-zeros
10816 Ignore blocks of zeros in archive (means EOF).
10818 The @option{--ignore-zeros} (@option{-i}) option causes @command{tar} to ignore blocks
10819 of zeros in the archive. Normally a block of zeros indicates the
10820 end of the archive, but when reading a damaged archive, or one which
10821 was created by concatenating several archives together, this option
10822 allows @command{tar} to read the entire archive. This option is not on
10823 by default because many versions of @command{tar} write garbage after
10826 Note that this option causes @command{tar} to read to the end of the
10827 archive file, which may sometimes avoid problems when multiple files
10828 are stored on a single physical tape.
10830 @xopindex{read-full-records, short description}
10832 @itemx --read-full-records
10833 Reblock as we read (for reading 4.2@acronym{BSD} pipes).
10835 If @option{--read-full-records} is used, @command{tar}
10836 will not panic if an attempt to read a record from the archive does
10837 not return a full record. Instead, @command{tar} will keep reading
10838 until it has obtained a full
10841 This option is turned on by default when @command{tar} is reading
10842 an archive from standard input, or from a remote machine. This is
10843 because on @acronym{BSD} Unix systems, a read of a pipe will return however
10844 much happens to be in the pipe, even if it is less than @command{tar}
10845 requested. If this option was not used, @command{tar} would fail as
10846 soon as it read an incomplete record from the pipe.
10848 This option is also useful with the commands for updating an archive.
10854 @FIXME{Appropriate options should be moved here from elsewhere.}
10856 @cindex blocking factor
10857 @cindex tape blocking
10859 When handling various tapes or cartridges, you have to take care of
10860 selecting a proper blocking, that is, the number of disk blocks you
10861 put together as a single tape block on the tape, without intervening
10862 tape gaps. A @dfn{tape gap} is a small landing area on the tape
10863 with no information on it, used for decelerating the tape to a
10864 full stop, and for later regaining the reading or writing speed.
10865 When the tape driver starts reading a record, the record has to
10866 be read whole without stopping, as a tape gap is needed to stop the
10867 tape motion without losing information.
10869 @cindex Exabyte blocking
10870 @cindex DAT blocking
10871 Using higher blocking (putting more disk blocks per tape block) will use
10872 the tape more efficiently as there will be less tape gaps. But reading
10873 such tapes may be more difficult for the system, as more memory will be
10874 required to receive at once the whole record. Further, if there is a
10875 reading error on a huge record, this is less likely that the system will
10876 succeed in recovering the information. So, blocking should not be too
10877 low, nor it should be too high. @command{tar} uses by default a blocking of
10878 20 for historical reasons, and it does not really matter when reading or
10879 writing to disk. Current tape technology would easily accommodate higher
10880 blockings. Sun recommends a blocking of 126 for Exabytes and 96 for DATs.
10881 We were told that for some DLT drives, the blocking should be a multiple
10882 of 4Kb, preferably 64Kb (@w{@kbd{-b 128}}) or 256 for decent performance.
10883 Other manufacturers may use different recommendations for the same tapes.
10884 This might also depends of the buffering techniques used inside modern
10885 tape controllers. Some imposes a minimum blocking, or a maximum blocking.
10886 Others request blocking to be some exponent of two.
10888 So, there is no fixed rule for blocking. But blocking at read time
10889 should ideally be the same as blocking used at write time. At one place
10890 I know, with a wide variety of equipment, they found it best to use a
10891 blocking of 32 to guarantee that their tapes are fully interchangeable.
10893 I was also told that, for recycled tapes, prior erasure (by the same
10894 drive unit that will be used to create the archives) sometimes lowers
10895 the error rates observed at rewriting time.
10897 I might also use @option{--number-blocks} instead of
10898 @option{--block-number}, so @option{--block} will then expand to
10899 @option{--blocking-factor} unambiguously.
10902 @section Many Archives on One Tape
10904 @FIXME{Appropriate options should be moved here from elsewhere.}
10906 @findex ntape @r{device}
10907 Most tape devices have two entries in the @file{/dev} directory, or
10908 entries that come in pairs, which differ only in the minor number for
10909 this device. Let's take for example @file{/dev/tape}, which often
10910 points to the only or usual tape device of a given system. There might
10911 be a corresponding @file{/dev/nrtape} or @file{/dev/ntape}. The simpler
10912 name is the @emph{rewinding} version of the device, while the name
10913 having @samp{nr} in it is the @emph{no rewinding} version of the same
10916 A rewinding tape device will bring back the tape to its beginning point
10917 automatically when this device is opened or closed. Since @command{tar}
10918 opens the archive file before using it and closes it afterwards, this
10919 means that a simple:
10922 $ @kbd{tar cf /dev/tape @var{directory}}
10926 will reposition the tape to its beginning both prior and after saving
10927 @var{directory} contents to it, thus erasing prior tape contents and
10928 making it so that any subsequent write operation will destroy what has
10931 @cindex tape positioning
10932 So, a rewinding device is normally meant to hold one and only one file.
10933 If you want to put more than one @command{tar} archive on a given tape, you
10934 will need to avoid using the rewinding version of the tape device. You
10935 will also have to pay special attention to tape positioning. Errors in
10936 positioning may overwrite the valuable data already on your tape. Many
10937 people, burnt by past experiences, will only use rewinding devices and
10938 limit themselves to one file per tape, precisely to avoid the risk of
10939 such errors. Be fully aware that writing at the wrong position on a
10940 tape loses all information past this point and most probably until the
10941 end of the tape, and this destroyed information @emph{cannot} be
10944 To save @var{directory-1} as a first archive at the beginning of a
10945 tape, and leave that tape ready for a second archive, you should use:
10948 $ @kbd{mt -f /dev/nrtape rewind}
10949 $ @kbd{tar cf /dev/nrtape @var{directory-1}}
10953 @dfn{Tape marks} are special magnetic patterns written on the tape
10954 media, which are later recognizable by the reading hardware. These
10955 marks are used after each file, when there are many on a single tape.
10956 An empty file (that is to say, two tape marks in a row) signal the
10957 logical end of the tape, after which no file exist. Usually,
10958 non-rewinding tape device drivers will react to the close request issued
10959 by @command{tar} by first writing two tape marks after your archive, and by
10960 backspacing over one of these. So, if you remove the tape at that time
10961 from the tape drive, it is properly terminated. But if you write
10962 another file at the current position, the second tape mark will be
10963 erased by the new information, leaving only one tape mark between files.
10965 So, you may now save @var{directory-2} as a second archive after the
10966 first on the same tape by issuing the command:
10969 $ @kbd{tar cf /dev/nrtape @var{directory-2}}
10973 and so on for all the archives you want to put on the same tape.
10975 Another usual case is that you do not write all the archives the same
10976 day, and you need to remove and store the tape between two archive
10977 sessions. In general, you must remember how many files are already
10978 saved on your tape. Suppose your tape already has 16 files on it, and
10979 that you are ready to write the 17th. You have to take care of skipping
10980 the first 16 tape marks before saving @var{directory-17}, say, by using
10984 $ @kbd{mt -f /dev/nrtape rewind}
10985 $ @kbd{mt -f /dev/nrtape fsf 16}
10986 $ @kbd{tar cf /dev/nrtape @var{directory-17}}
10989 In all the previous examples, we put aside blocking considerations, but
10990 you should do the proper things for that as well. @xref{Blocking}.
10993 * Tape Positioning:: Tape Positions and Tape Marks
10994 * mt:: The @command{mt} Utility
10997 @node Tape Positioning
10998 @subsection Tape Positions and Tape Marks
11001 Just as archives can store more than one file from the file system,
11002 tapes can store more than one archive file. To keep track of where
11003 archive files (or any other type of file stored on tape) begin and
11004 end, tape archive devices write magnetic @dfn{tape marks} on the
11005 archive media. Tape drives write one tape mark between files,
11006 two at the end of all the file entries.
11008 If you think of data as a series of records "rrrr"'s, and tape marks as
11009 "*"'s, a tape might look like the following:
11012 rrrr*rrrrrr*rrrrr*rr*rrrrr**-------------------------
11015 Tape devices read and write tapes using a read/write @dfn{tape
11016 head}---a physical part of the device which can only access one
11017 point on the tape at a time. When you use @command{tar} to read or
11018 write archive data from a tape device, the device will begin reading
11019 or writing from wherever on the tape the tape head happens to be,
11020 regardless of which archive or what part of the archive the tape
11021 head is on. Before writing an archive, you should make sure that no
11022 data on the tape will be overwritten (unless it is no longer needed).
11023 Before reading an archive, you should make sure the tape head is at
11024 the beginning of the archive you want to read. You can do it manually
11025 via @code{mt} utility (@pxref{mt}). The @code{restore} script does
11026 that automatically (@pxref{Scripted Restoration}).
11028 If you want to add new archive file entries to a tape, you should
11029 advance the tape to the end of the existing file entries, backspace
11030 over the last tape mark, and write the new archive file. If you were
11031 to add two archives to the example above, the tape might look like the
11035 rrrr*rrrrrr*rrrrr*rr*rrrrr*rrr*rrrr**----------------
11039 @subsection The @command{mt} Utility
11042 @FIXME{Is it true that this only works on non-block devices?
11043 should explain the difference, (fixed or variable).}
11044 @xref{Blocking Factor}.
11046 You can use the @command{mt} utility to advance or rewind a tape past a
11047 specified number of archive files on the tape. This will allow you
11048 to move to the beginning of an archive before extracting or reading
11049 it, or to the end of all the archives before writing a new one.
11050 @FIXME{Why isn't there an "advance 'til you find two tape marks
11053 The syntax of the @command{mt} command is:
11056 @kbd{mt [-f @var{tapename}] @var{operation} [@var{number}]}
11059 where @var{tapename} is the name of the tape device, @var{number} is
11060 the number of times an operation is performed (with a default of one),
11061 and @var{operation} is one of the following:
11063 @FIXME{is there any use for record operations?}
11068 Writes @var{number} tape marks at the current position on the tape.
11071 Moves tape position forward @var{number} files.
11074 Moves tape position back @var{number} files.
11077 Rewinds the tape. (Ignores @var{number}.)
11081 Rewinds the tape and takes the tape device off-line. (Ignores @var{number}.)
11084 Prints status information about the tape unit.
11088 If you don't specify a @var{tapename}, @command{mt} uses the environment
11089 variable @env{TAPE}; if @env{TAPE} is not set, @command{mt} will use
11090 the default device specified in your @file{sys/mtio.h} file
11091 (@code{DEFTAPE} variable). If this is not defined, the program will
11092 display a descriptive error message and exit with code 1.
11094 @command{mt} returns a 0 exit status when the operation(s) were
11095 successful, 1 if the command was unrecognized, and 2 if an operation
11098 @node Using Multiple Tapes
11099 @section Using Multiple Tapes
11101 Often you might want to write a large archive, one larger than will fit
11102 on the actual tape you are using. In such a case, you can run multiple
11103 @command{tar} commands, but this can be inconvenient, particularly if you
11104 are using options like @option{--exclude=@var{pattern}} or dumping entire file systems.
11105 Therefore, @command{tar} provides a special mode for creating
11106 multi-volume archives.
11108 @dfn{Multi-volume} archive is a single @command{tar} archive, stored
11109 on several media volumes of fixed size. Although in this section we will
11110 often call @samp{volume} a @dfn{tape}, there is absolutely no
11111 requirement for multi-volume archives to be stored on tapes. Instead,
11112 they can use whatever media type the user finds convenient, they can
11113 even be located on files.
11115 When creating a multi-volume archive, @GNUTAR{} continues to fill
11116 current volume until it runs out of space, then it switches to
11117 next volume (usually the operator is queried to replace the tape on
11118 this point), and continues working on the new volume. This operation
11119 continues until all requested files are dumped. If @GNUTAR{} detects
11120 end of media while dumping a file, such a file is archived in split
11121 form. Some very big files can even be split across several volumes.
11123 Each volume is itself a valid @GNUTAR{} archive, so it can be read
11124 without any special options. Consequently any file member residing
11125 entirely on one volume can be extracted or otherwise operated upon
11126 without needing the other volume. Sure enough, to extract a split
11127 member you would need all volumes its parts reside on.
11129 Multi-volume archives suffer from several limitations. In particular,
11130 they cannot be compressed.
11132 @GNUTAR{} is able to create multi-volume archives of two formats
11133 (@pxref{Formats}): @samp{GNU} and @samp{POSIX}.
11136 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
11137 * Tape Files:: Tape Files
11138 * Tarcat:: Concatenate Volumes into a Single Archive
11142 @node Multi-Volume Archives
11143 @subsection Archives Longer than One Tape or Disk
11144 @cindex Multi-volume archives
11146 @opindex multi-volume
11147 To create an archive that is larger than will fit on a single unit of
11148 the media, use the @option{--multi-volume} (@option{-M}) option in conjunction with
11149 the @option{--create} option (@pxref{create}). A @dfn{multi-volume}
11150 archive can be manipulated like any other archive (provided the
11151 @option{--multi-volume} option is specified), but is stored on more
11152 than one tape or file.
11154 When you specify @option{--multi-volume}, @command{tar} does not report an
11155 error when it comes to the end of an archive volume (when reading), or
11156 the end of the media (when writing). Instead, it prompts you to load
11157 a new storage volume. If the archive is on a magnetic tape, you
11158 should change tapes when you see the prompt; if the archive is on a
11159 floppy disk, you should change disks; etc.
11162 @item --multi-volume
11164 Creates a multi-volume archive, when used in conjunction with
11165 @option{--create} (@option{-c}). To perform any other operation on a multi-volume
11166 archive, specify @option{--multi-volume} in conjunction with that
11171 $ @kbd{tar --create --multi-volume --file=/dev/tape @var{files}}
11175 The method @command{tar} uses to detect end of tape is not perfect, and
11176 fails on some operating systems or on some devices. If @command{tar}
11177 cannot detect the end of the tape itself, you can use
11178 @option{--tape-length} option to inform it about the capacity of the
11181 @anchor{tape-length}
11183 @opindex tape-length
11184 @item --tape-length=@var{size}[@var{suf}]
11185 @itemx -L @var{size}[@var{suf}]
11186 Set maximum length of a volume. The @var{suf}, if given, specifies
11187 units in which @var{size} is expressed, e.g. @samp{2M} mean 2
11188 megabytes (@pxref{size-suffixes}, for a list of allowed size
11189 suffixes). Without @var{suf}, units of 1024 bytes (kilobyte) are
11192 This option selects @option{--multi-volume} automatically. For example:
11195 $ @kbd{tar --create --tape-length=41943040 --file=/dev/tape @var{files}}
11199 or, which is equivalent:
11202 $ @kbd{tar --create --tape-length=4G --file=/dev/tape @var{files}}
11206 @anchor{change volume prompt}
11207 When @GNUTAR{} comes to the end of a storage media, it asks you to
11208 change the volume. The built-in prompt for POSIX locale
11209 is@footnote{If you run @GNUTAR{} under a different locale, the
11210 translation to the locale's language will be used.}:
11213 Prepare volume #@var{n} for `@var{archive}' and hit return:
11217 where @var{n} is the ordinal number of the volume to be created and
11218 @var{archive} is archive file or device name.
11220 When prompting for a new tape, @command{tar} accepts any of the following
11225 Request @command{tar} to explain possible responses.
11227 Request @command{tar} to exit immediately.
11228 @item n @var{file-name}
11229 Request @command{tar} to write the next volume on the file @var{file-name}.
11231 Request @command{tar} to run a subshell. This option can be disabled
11232 by giving @option{--restrict} command line option to
11233 @command{tar}@footnote{@xref{--restrict}, for more information about
11236 Request @command{tar} to begin writing the next volume.
11239 (You should only type @samp{y} after you have changed the tape;
11240 otherwise @command{tar} will write over the volume it just finished.)
11242 @cindex Volume number file
11244 @anchor{volno-file}
11245 @opindex volno-file
11246 The volume number used by @command{tar} in its tape-changing prompt
11247 can be changed; if you give the
11248 @option{--volno-file=@var{file-of-number}} option, then
11249 @var{file-of-number} should be an non-existing file to be created, or
11250 else, a file already containing a decimal number. That number will be
11251 used as the volume number of the first volume written. When
11252 @command{tar} is finished, it will rewrite the file with the
11253 now-current volume number. (This does not change the volume number
11254 written on a tape label, as per @ref{label}, it @emph{only} affects
11255 the number used in the prompt.)
11257 @cindex End-of-archive info script
11258 @cindex Info script
11259 @anchor{info-script}
11260 @opindex info-script
11261 @opindex new-volume-script
11262 If you want more elaborate behavior than this, you can write a special
11263 @dfn{new volume script}, that will be responsible for changing the
11264 volume, and instruct @command{tar} to use it instead of its normal
11265 prompting procedure:
11268 @item --info-script=@var{script-name}
11269 @itemx --new-volume-script=@var{script-name}
11270 @itemx -F @var{script-name}
11271 Specify the full name of the volume script to use. The script can be
11272 used to eject cassettes, or to broadcast messages such as
11273 @samp{Someone please come change my tape} when performing unattended
11277 The @var{script-name} is executed without any command line
11278 arguments. It inherits @command{tar}'s shell environment.
11279 Additional data is passed to it via the following
11280 environment variables:
11283 @vrindex TAR_VERSION, info script environment variable
11285 @GNUTAR{} version number.
11287 @vrindex TAR_ARCHIVE, info script environment variable
11289 The name of the archive @command{tar} is processing.
11291 @vrindex TAR_BLOCKING_FACTOR, info script environment variable
11292 @item TAR_BLOCKING_FACTOR
11293 Current blocking factor (@pxref{Blocking}).
11295 @vrindex TAR_VOLUME, info script environment variable
11297 Ordinal number of the volume @command{tar} is about to start.
11299 @vrindex TAR_SUBCOMMAND, info script environment variable
11300 @item TAR_SUBCOMMAND
11301 A short option describing the operation @command{tar} is executing.
11302 @xref{Operations}, for a complete list of subcommand options.
11304 @vrindex TAR_FORMAT, info script environment variable
11306 Format of the archive being processed. @xref{Formats}, for a complete
11307 list of archive format names.
11309 @vrindex TAR_FD, info script environment variable
11311 File descriptor which can be used to communicate the new volume
11312 name to @command{tar}.
11315 The volume script can instruct @command{tar} to use new archive name,
11316 by writing in to file descriptor @env{$TAR_FD} (see below for an example).
11318 If the info script fails, @command{tar} exits; otherwise, it begins
11319 writing the next volume.
11321 If you want @command{tar} to cycle through a series of files or tape
11322 drives, there are three approaches to choose from. First of all, you
11323 can give @command{tar} multiple @option{--file} options. In this case
11324 the specified files will be used, in sequence, as the successive
11325 volumes of the archive. Only when the first one in the sequence needs
11326 to be used again will @command{tar} prompt for a tape change (or run
11327 the info script). For example, suppose someone has two tape drives on
11328 a system named @file{/dev/tape0} and @file{/dev/tape1}. For having
11329 @GNUTAR{} to switch to the second drive when it needs to write the
11330 second tape, and then back to the first tape, etc., just do either of:
11333 $ @kbd{tar --create --multi-volume --file=/dev/tape0 --file=/dev/tape1 @var{files}}
11334 $ @kbd{tar cMff /dev/tape0 /dev/tape1 @var{files}}
11337 The second method is to use the @samp{n} response to the tape-change
11340 Finally, the most flexible approach is to use a volume script, that
11341 writes new archive name to the file descriptor @env{$TAR_FD}. For example, the
11342 following volume script will create a series of archive files, named
11343 @file{@var{archive}-@var{vol}}, where @var{archive} is the name of the
11344 archive being created (as given by @option{--file} option) and
11345 @var{vol} is the ordinal number of the archive being created:
11350 echo Preparing volume $TAR_VOLUME of $TAR_ARCHIVE.
11352 name=`expr $TAR_ARCHIVE : '\(.*\)-.*'`
11353 case $TAR_SUBCOMMAND in
11355 -d|-x|-t) test -r $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME || exit 1
11360 echo $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME >&$TAR_FD
11364 The same script can be used while listing, comparing or extracting
11365 from the created archive. For example:
11369 # @r{Create a multi-volume archive:}
11370 $ @kbd{tar -c -L1024 -f archive.tar -F new-volume .}
11371 # @r{Extract from the created archive:}
11372 $ @kbd{tar -x -f archive.tar -F new-volume .}
11377 Notice, that the first command had to use @option{-L} option, since
11378 otherwise @GNUTAR{} will end up writing everything to file
11379 @file{archive.tar}.
11381 You can read each individual volume of a multi-volume archive as if it
11382 were an archive by itself. For example, to list the contents of one
11383 volume, use @option{--list}, without @option{--multi-volume} specified.
11384 To extract an archive member from one volume (assuming it is described
11385 that volume), use @option{--extract}, again without
11386 @option{--multi-volume}.
11388 If an archive member is split across volumes (i.e., its entry begins on
11389 one volume of the media and ends on another), you need to specify
11390 @option{--multi-volume} to extract it successfully. In this case, you
11391 should load the volume where the archive member starts, and use
11392 @samp{tar --extract --multi-volume}---@command{tar} will prompt for later
11393 volumes as it needs them. @xref{extracting archives}, for more
11394 information about extracting archives.
11396 Multi-volume archives can be modified like any other archive. To add
11397 files to a multi-volume archive, you need to only mount the last
11398 volume of the archive media (and new volumes, if needed). For all
11399 other operations, you need to use the entire archive.
11401 If a multi-volume archive was labeled using
11402 @option{--label=@var{archive-label}} (@pxref{label}) when it was
11403 created, @command{tar} will not automatically label volumes which are
11404 added later. To label subsequent volumes, specify
11405 @option{--label=@var{archive-label}} again in conjunction with the
11406 @option{--append}, @option{--update} or @option{--concatenate} operation.
11408 Notice that multi-volume support is a GNU extension and the archives
11409 created in this mode should be read only using @GNUTAR{}. If you
11410 absolutely have to process such archives using a third-party @command{tar}
11411 implementation, read @ref{Split Recovery}.
11414 @subsection Tape Files
11415 @cindex labeling archives
11419 To give the archive a name which will be recorded in it, use the
11420 @option{--label=@var{volume-label}} (@option{-V @var{volume-label}})
11421 option. This will write a special block identifying
11422 @var{volume-label} as the name of the archive to the front of the
11423 archive which will be displayed when the archive is listed with
11424 @option{--list}. If you are creating a multi-volume archive with
11425 @option{--multi-volume} (@pxref{Using Multiple Tapes}), then the
11426 volume label will have @samp{Volume @var{nnn}} appended to the name
11427 you give, where @var{nnn} is the number of the volume of the archive.
11428 If you use the @option{--label=@var{volume-label}} option when
11429 reading an archive, it checks to make sure the label on the tape
11430 matches the one you gave. @xref{label}.
11432 When @command{tar} writes an archive to tape, it creates a single
11433 tape file. If multiple archives are written to the same tape, one
11434 after the other, they each get written as separate tape files. When
11435 extracting, it is necessary to position the tape at the right place
11436 before running @command{tar}. To do this, use the @command{mt} command.
11437 For more information on the @command{mt} command and on the organization
11438 of tapes into a sequence of tape files, see @ref{mt}.
11440 People seem to often do:
11443 @kbd{--label="@var{some-prefix} `date +@var{some-format}`"}
11446 or such, for pushing a common date in all volumes or an archive set.
11449 @subsection Concatenate Volumes into a Single Archive
11452 Sometimes it is necessary to convert existing @GNUTAR{} multi-volume
11453 archive to a single @command{tar} archive. Simply concatenating all
11454 volumes into one will not work, since each volume carries an additional
11455 information at the beginning. @GNUTAR{} is shipped with the shell
11456 script @command{tarcat} designed for this purpose.
11458 The script takes a list of files comprising a multi-volume archive
11459 and creates the resulting archive at the standard output. For example:
11462 @kbd{tarcat vol.1 vol.2 vol.3 | tar tf -}
11465 The script implements a simple heuristics to determine the format of
11466 the first volume file and to decide how to process the rest of the
11467 files. However, it makes no attempt to verify whether the files are
11468 given in order or even if they are valid @command{tar} archives.
11469 It uses @command{dd} and does not filter its standard error, so you
11470 will usually see lots of spurious messages.
11472 @FIXME{The script is not installed. Should we install it?}
11475 @section Including a Label in the Archive
11476 @cindex Labeling an archive
11477 @cindex Labels on the archive media
11478 @cindex Labeling multi-volume archives
11481 To avoid problems caused by misplaced paper labels on the archive
11482 media, you can include a @dfn{label} entry --- an archive member which
11483 contains the name of the archive --- in the archive itself. Use the
11484 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
11485 option@footnote{Until version 1.10, that option was called
11486 @option{--volume}, but is not available under that name anymore.} in
11487 conjunction with the @option{--create} operation to include a label
11488 entry in the archive as it is being created.
11491 @item --label=@var{archive-label}
11492 @itemx -V @var{archive-label}
11493 Includes an @dfn{archive-label} at the beginning of the archive when
11494 the archive is being created, when used in conjunction with the
11495 @option{--create} operation. Checks to make sure the archive label
11496 matches the one specified (when used in conjunction with any other
11500 If you create an archive using both
11501 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
11502 and @option{--multi-volume} (@option{-M}), each volume of the archive
11503 will have an archive label of the form @samp{@var{archive-label}
11504 Volume @var{n}}, where @var{n} is 1 for the first volume, 2 for the
11505 next, and so on. @xref{Using Multiple Tapes}, for information on
11506 creating multiple volume archives.
11508 @cindex Volume label, listing
11509 @cindex Listing volume label
11510 The volume label will be displayed by @option{--list} along with
11511 the file contents. If verbose display is requested, it will also be
11512 explicitly marked as in the example below:
11516 $ @kbd{tar --verbose --list --file=iamanarchive}
11517 V--------- 0 0 0 1992-03-07 12:01 iamalabel--Volume Header--
11518 -rw-r--r-- ringo user 40 1990-05-21 13:30 iamafilename
11522 @opindex test-label
11523 @anchor{--test-label option}
11524 However, @option{--list} option will cause listing entire
11525 contents of the archive, which may be undesirable (for example, if the
11526 archive is stored on a tape). You can request checking only the volume
11527 label by specifying @option{--test-label} option. This option reads only the
11528 first block of an archive, so it can be used with slow storage
11529 devices. For example:
11533 $ @kbd{tar --test-label --file=iamanarchive}
11538 If @option{--test-label} is used with one or more command line
11539 arguments, @command{tar} compares the volume label with each
11540 argument. It exits with code 0 if a match is found, and with code 1
11541 otherwise@footnote{Note that @GNUTAR{} versions up to 1.23 indicated
11542 mismatch with an exit code 2 and printed a spurious diagnostics on
11543 stderr.}. No output is displayed, unless you also used the
11544 @option{--verbose} option. For example:
11548 $ @kbd{tar --test-label --file=iamanarchive 'iamalabel'}
11550 $ @kbd{tar --test-label --file=iamanarchive 'alabel'}
11555 When used with the @option{--verbose} option, @command{tar}
11556 prints the actual volume label (if any), and a verbose diagnostics in
11557 case of a mismatch:
11561 $ @kbd{tar --test-label --verbose --file=iamanarchive 'iamalabel'}
11564 $ @kbd{tar --test-label --verbose --file=iamanarchive 'alabel'}
11566 tar: Archive label mismatch
11571 If you request any operation, other than @option{--create}, along
11572 with using @option{--label} option, @command{tar} will first check if
11573 the archive label matches the one specified and will refuse to proceed
11574 if it does not. Use this as a safety precaution to avoid accidentally
11575 overwriting existing archives. For example, if you wish to add files
11576 to @file{archive}, presumably labeled with string @samp{My volume},
11581 $ @kbd{tar -rf archive --label 'My volume' .}
11582 tar: Archive not labeled to match `My volume'
11587 in case its label does not match. This will work even if
11588 @file{archive} is not labeled at all.
11590 Similarly, @command{tar} will refuse to list or extract the
11591 archive if its label doesn't match the @var{archive-label}
11592 specified. In those cases, @var{archive-label} argument is interpreted
11593 as a globbing-style pattern which must match the actual magnetic
11594 volume label. @xref{exclude}, for a precise description of how match
11595 is attempted@footnote{Previous versions of @command{tar} used full
11596 regular expression matching, or before that, only exact string
11597 matching, instead of wildcard matchers. We decided for the sake of
11598 simplicity to use a uniform matching device through
11599 @command{tar}.}. If the switch @option{--multi-volume} (@option{-M}) is being used,
11600 the volume label matcher will also suffix @var{archive-label} by
11601 @w{@samp{ Volume [1-9]*}} if the initial match fails, before giving
11602 up. Since the volume numbering is automatically added in labels at
11603 creation time, it sounded logical to equally help the user taking care
11604 of it when the archive is being read.
11606 You can also use @option{--label} to get a common information on
11607 all tapes of a series. For having this information different in each
11608 series created through a single script used on a regular basis, just
11609 manage to get some date string as part of the label. For example:
11613 $ @kbd{tar cfMV /dev/tape "Daily backup for `date +%Y-%m-%d`"}
11614 $ @kbd{tar --create --file=/dev/tape --multi-volume \
11615 --label="Daily backup for `date +%Y-%m-%d`"}
11619 Some more notes about volume labels:
11622 @item Each label has its own date and time, which corresponds
11623 to the time when @GNUTAR{} initially attempted to write it,
11624 often soon after the operator launches @command{tar} or types the
11625 carriage return telling that the next tape is ready.
11627 @item Comparing date labels to get an idea of tape throughput is
11628 unreliable. It gives correct results only if the delays for rewinding
11629 tapes and the operator switching them were negligible, which is
11630 usually not the case.
11634 @section Verifying Data as It is Stored
11635 @cindex Verifying a write operation
11636 @cindex Double-checking a write operation
11641 @opindex verify, short description
11642 Attempt to verify the archive after writing.
11645 This option causes @command{tar} to verify the archive after writing it.
11646 Each volume is checked after it is written, and any discrepancies
11647 are recorded on the standard error output.
11649 Verification requires that the archive be on a back-space-able medium.
11650 This means pipes, some cartridge tape drives, and some other devices
11651 cannot be verified.
11653 You can insure the accuracy of an archive by comparing files in the
11654 system with archive members. @command{tar} can compare an archive to the
11655 file system as the archive is being written, to verify a write
11656 operation, or can compare a previously written archive, to insure that
11659 @xopindex{verify, using with @option{--create}}
11660 @xopindex{create, using with @option{--verify}}
11661 To check for discrepancies in an archive immediately after it is
11662 written, use the @option{--verify} (@option{-W}) option in conjunction with
11663 the @option{--create} operation. When this option is
11664 specified, @command{tar} checks archive members against their counterparts
11665 in the file system, and reports discrepancies on the standard error.
11667 To verify an archive, you must be able to read it from before the end
11668 of the last written entry. This option is useful for detecting data
11669 errors on some tapes. Archives written to pipes, some cartridge tape
11670 drives, and some other devices cannot be verified.
11672 One can explicitly compare an already made archive with the file
11673 system by using the @option{--compare} (@option{--diff}, @option{-d})
11674 option, instead of using the more automatic @option{--verify} option.
11677 Note that these two options have a slightly different intent. The
11678 @option{--compare} option checks how identical are the logical contents of some
11679 archive with what is on your disks, while the @option{--verify} option is
11680 really for checking if the physical contents agree and if the recording
11681 media itself is of dependable quality. So, for the @option{--verify}
11682 operation, @command{tar} tries to defeat all in-memory cache pertaining to
11683 the archive, while it lets the speed optimization undisturbed for the
11684 @option{--compare} option. If you nevertheless use @option{--compare} for
11685 media verification, you may have to defeat the in-memory cache yourself,
11686 maybe by opening and reclosing the door latch of your recording unit,
11687 forcing some doubt in your operating system about the fact this is really
11688 the same volume as the one just written or read.
11690 The @option{--verify} option would not be necessary if drivers were indeed
11691 able to detect dependably all write failures. This sometimes require many
11692 magnetic heads, some able to read after the writes occurred. One would
11693 not say that drivers unable to detect all cases are necessarily flawed,
11694 as long as programming is concerned.
11696 The @option{--verify} (@option{-W}) option will not work in
11697 conjunction with the @option{--multi-volume} (@option{-M}) option or
11698 the @option{--append} (@option{-r}), @option{--update} (@option{-u})
11699 and @option{--delete} operations. @xref{Operations}, for more
11700 information on these operations.
11702 Also, since @command{tar} normally strips leading @samp{/} from file
11703 names (@pxref{absolute}), a command like @samp{tar --verify -cf
11704 /tmp/foo.tar /etc} will work as desired only if the working directory is
11705 @file{/}, as @command{tar} uses the archive's relative member names
11706 (e.g., @file{etc/motd}) when verifying the archive.
11708 @node Write Protection
11709 @section Write Protection
11711 Almost all tapes and diskettes, and in a few rare cases, even disks can
11712 be @dfn{write protected}, to protect data on them from being changed.
11713 Once an archive is written, you should write protect the media to prevent
11714 the archive from being accidentally overwritten or deleted. (This will
11715 protect the archive from being changed with a tape or floppy drive---it
11716 will not protect it from magnet fields or other physical hazards.)
11718 The write protection device itself is usually an integral part of the
11719 physical media, and can be a two position (write enabled/write
11720 disabled) switch, a notch which can be popped out or covered, a ring
11721 which can be removed from the center of a tape reel, or some other
11722 changeable feature.
11727 This appendix lists some important user-visible changes between
11728 version @GNUTAR{} @value{VERSION} and previous versions. An up-to-date
11729 version of this document is available at
11730 @uref{http://www.gnu.org/@/software/@/tar/manual/changes.html,the
11731 @GNUTAR{} documentation page}.
11734 @item Use of globbing patterns when listing and extracting.
11736 Previous versions of GNU tar assumed shell-style globbing when
11737 extracting from or listing an archive. For example:
11740 $ @kbd{tar xf foo.tar '*.c'}
11743 would extract all files whose names end in @samp{.c}. This behavior
11744 was not documented and was incompatible with traditional tar
11745 implementations. Therefore, starting from version 1.15.91, GNU tar
11746 no longer uses globbing by default. For example, the above invocation
11747 is now interpreted as a request to extract from the archive the file
11750 To facilitate transition to the new behavior for those users who got
11751 used to the previous incorrect one, @command{tar} will print a warning
11752 if it finds out that a requested member was not found in the archive
11753 and its name looks like a globbing pattern. For example:
11756 $ @kbd{tar xf foo.tar '*.c'}
11757 tar: Pattern matching characters used in file names. Please,
11758 tar: use --wildcards to enable pattern matching, or --no-wildcards to
11759 tar: suppress this warning.
11760 tar: *.c: Not found in archive
11761 tar: Error exit delayed from previous errors
11764 To treat member names as globbing patterns, use the @option{--wildcards} option.
11765 If you want to tar to mimic the behavior of versions prior to 1.15.91,
11766 add this option to your @env{TAR_OPTIONS} variable.
11768 @xref{wildcards}, for the detailed discussion of the use of globbing
11769 patterns by @GNUTAR{}.
11771 @item Use of short option @option{-o}.
11773 Earlier versions of @GNUTAR{} understood @option{-o} command line
11774 option as a synonym for @option{--old-archive}.
11776 @GNUTAR{} starting from version 1.13.90 understands this option as
11777 a synonym for @option{--no-same-owner}. This is compatible with
11778 UNIX98 @command{tar} implementations.
11780 However, to facilitate transition, @option{-o} option retains its
11781 old semantics when it is used with one of archive-creation commands.
11782 Users are encouraged to use @option{--format=oldgnu} instead.
11784 It is especially important, since versions of @acronym{GNU} Automake
11785 up to and including 1.8.4 invoke tar with this option to produce
11786 distribution tarballs. @xref{Formats,v7}, for the detailed discussion
11787 of this issue and its implications.
11789 @xref{Options, tar-formats, Changing Automake's Behavior,
11790 automake, GNU Automake}, for a description on how to use various
11791 archive formats with @command{automake}.
11793 Future versions of @GNUTAR{} will understand @option{-o} only as a
11794 synonym for @option{--no-same-owner}.
11796 @item Use of short option @option{-l}
11798 Earlier versions of @GNUTAR{} understood @option{-l} option as a
11799 synonym for @option{--one-file-system}. Since such usage contradicted
11800 to UNIX98 specification and harmed compatibility with other
11801 implementations, it was declared deprecated in version 1.14. However,
11802 to facilitate transition to its new semantics, it was supported by
11803 versions 1.15 and 1.15.90. The present use of @option{-l} as a short
11804 variant of @option{--check-links} was introduced in version 1.15.91.
11806 @item Use of options @option{--portability} and @option{--old-archive}
11808 These options are deprecated. Please use @option{--format=v7} instead.
11810 @item Use of option @option{--posix}
11812 This option is deprecated. Please use @option{--format=posix} instead.
11815 @node Configuring Help Summary
11816 @appendix Configuring Help Summary
11818 Running @kbd{tar --help} displays the short @command{tar} option
11819 summary (@pxref{help}). This summary is organized by @dfn{groups} of
11820 semantically close options. The options within each group are printed
11821 in the following order: a short option, eventually followed by a list
11822 of corresponding long option names, followed by a short description of
11823 the option. For example, here is an excerpt from the actual @kbd{tar
11827 Main operation mode:
11829 -A, --catenate, --concatenate append tar files to an archive
11830 -c, --create create a new archive
11831 -d, --diff, --compare find differences between archive and
11833 --delete delete from the archive
11836 @vrindex ARGP_HELP_FMT, environment variable
11837 The exact visual representation of the help output is configurable via
11838 @env{ARGP_HELP_FMT} environment variable. The value of this variable
11839 is a comma-separated list of @dfn{format variable} assignments. There
11840 are two kinds of format variables. An @dfn{offset variable} keeps the
11841 offset of some part of help output text from the leftmost column on
11842 the screen. A @dfn{boolean} variable is a flag that toggles some
11843 output feature on or off. Depending on the type of the corresponding
11844 variable, there are two kinds of assignments:
11847 @item Offset assignment
11849 The assignment to an offset variable has the following syntax:
11852 @var{variable}=@var{value}
11856 where @var{variable} is the variable name, and @var{value} is a
11857 numeric value to be assigned to the variable.
11859 @item Boolean assignment
11861 To assign @code{true} value to a variable, simply put this variable name. To
11862 assign @code{false} value, prefix the variable name with @samp{no-}. For
11867 # Assign @code{true} value:
11869 # Assign @code{false} value:
11875 Following variables are declared:
11877 @deftypevr {Help Output} boolean dup-args
11878 If true, arguments for an option are shown with both short and long
11879 options, even when a given option has both forms, for example:
11882 -f ARCHIVE, --file=ARCHIVE use archive file or device ARCHIVE
11885 If false, then if an option has both short and long forms, the
11886 argument is only shown with the long one, for example:
11889 -f, --file=ARCHIVE use archive file or device ARCHIVE
11893 and a message indicating that the argument is applicable to both
11894 forms is printed below the options. This message can be disabled
11895 using @code{dup-args-note} (see below).
11897 The default is false.
11900 @deftypevr {Help Output} boolean dup-args-note
11901 If this variable is true, which is the default, the following notice
11902 is displayed at the end of the help output:
11905 Mandatory or optional arguments to long options are also mandatory or
11906 optional for any corresponding short options.
11909 Setting @code{no-dup-args-note} inhibits this message. Normally, only one of
11910 variables @code{dup-args} or @code{dup-args-note} should be set.
11913 @deftypevr {Help Output} offset short-opt-col
11914 Column in which short options start. Default is 2.
11918 $ @kbd{tar --help|grep ARCHIVE}
11919 -f, --file=ARCHIVE use archive file or device ARCHIVE
11920 $ @kbd{ARGP_HELP_FMT=short-opt-col=6 tar --help|grep ARCHIVE}
11921 -f, --file=ARCHIVE use archive file or device ARCHIVE
11926 @deftypevr {Help Output} offset long-opt-col
11927 Column in which long options start. Default is 6. For example:
11931 $ @kbd{tar --help|grep ARCHIVE}
11932 -f, --file=ARCHIVE use archive file or device ARCHIVE
11933 $ @kbd{ARGP_HELP_FMT=long-opt-col=16 tar --help|grep ARCHIVE}
11934 -f, --file=ARCHIVE use archive file or device ARCHIVE
11939 @deftypevr {Help Output} offset doc-opt-col
11940 Column in which @dfn{doc options} start. A doc option isn't actually
11941 an option, but rather an arbitrary piece of documentation that is
11942 displayed in much the same manner as the options. For example, in
11943 the description of @option{--format} option:
11947 -H, --format=FORMAT create archive of the given format.
11949 FORMAT is one of the following:
11951 gnu GNU tar 1.13.x format
11952 oldgnu GNU format as per tar <= 1.12
11953 pax POSIX 1003.1-2001 (pax) format
11955 ustar POSIX 1003.1-1988 (ustar) format
11956 v7 old V7 tar format
11961 the format names are doc options. Thus, if you set
11962 @kbd{ARGP_HELP_FMT=doc-opt-col=6} the above part of the help output
11963 will look as follows:
11967 -H, --format=FORMAT create archive of the given format.
11969 FORMAT is one of the following:
11971 gnu GNU tar 1.13.x format
11972 oldgnu GNU format as per tar <= 1.12
11973 pax POSIX 1003.1-2001 (pax) format
11975 ustar POSIX 1003.1-1988 (ustar) format
11976 v7 old V7 tar format
11981 @deftypevr {Help Output} offset opt-doc-col
11982 Column in which option description starts. Default is 29.
11986 $ @kbd{tar --help|grep ARCHIVE}
11987 -f, --file=ARCHIVE use archive file or device ARCHIVE
11988 $ @kbd{ARGP_HELP_FMT=opt-doc-col=19 tar --help|grep ARCHIVE}
11989 -f, --file=ARCHIVE use archive file or device ARCHIVE
11990 $ @kbd{ARGP_HELP_FMT=opt-doc-col=9 tar --help|grep ARCHIVE}
11992 use archive file or device ARCHIVE
11997 Notice, that the description starts on a separate line if
11998 @code{opt-doc-col} value is too small.
12001 @deftypevr {Help Output} offset header-col
12002 Column in which @dfn{group headers} are printed. A group header is a
12003 descriptive text preceding an option group. For example, in the
12007 Main operation mode:
12009 -A, --catenate, --concatenate append tar files to
12011 -c, --create create a new archive
12014 @samp{Main operation mode:} is the group header.
12016 The default value is 1.
12019 @deftypevr {Help Output} offset usage-indent
12020 Indentation of wrapped usage lines. Affects @option{--usage}
12021 output. Default is 12.
12024 @deftypevr {Help Output} offset rmargin
12025 Right margin of the text output. Used for wrapping.
12028 @node Fixing Snapshot Files
12029 @appendix Fixing Snapshot Files
12030 @include tar-snapshot-edit.texi
12032 @node Tar Internals
12033 @appendix Tar Internals
12034 @include intern.texi
12038 @include genfile.texi
12040 @node Free Software Needs Free Documentation
12041 @appendix Free Software Needs Free Documentation
12042 @include freemanuals.texi
12044 @node GNU Free Documentation License
12045 @appendix GNU Free Documentation License
12049 @node Index of Command Line Options
12050 @appendix Index of Command Line Options
12052 This appendix contains an index of all @GNUTAR{} long command line
12053 options. The options are listed without the preceding double-dash.
12054 For a cross-reference of short command line options, see
12055 @ref{Short Option Summary}.
12068 @c Local variables:
12069 @c texinfo-column-for-description: 32