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 1.1 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 License".
50 (a) The FSF's Back-Cover Text is: ``You have the freedom to
51 copy and modify this GNU manual. Buying copies from the FSF
52 supports it in developing GNU and promoting software freedom.''
56 @dircategory Archiving
58 * Tar: (tar). Making tape (or disk) archives.
61 @dircategory Individual utilities
63 * tar: (tar)tar invocation. Invoking @GNUTAR{}.
66 @shorttitlepage @acronym{GNU} @command{tar}
69 @title @acronym{GNU} tar: an archiver tool
70 @subtitle @value{RENDITION} @value{VERSION}, @value{UPDATED}
71 @author John Gilmore, Jay Fenlason et al.
74 @vskip 0pt plus 1filll
80 @top @acronym{GNU} tar: an archiver tool
85 @cindex archiving files
87 The first part of this master menu lists the major nodes in this Info
88 document. The rest of the menu lists all the lower level nodes.
91 @c The master menu goes here.
93 @c NOTE: To update it from within Emacs, make sure mastermenu.el is
94 @c loaded and run texinfo-master-menu.
95 @c To update it from the command line, run
106 * Date input formats::
113 * Configuring Help Summary::
114 * Fixing Snapshot Files::
117 * Free Software Needs Free Documentation::
118 * Copying This Manual::
119 * Index of Command Line Options::
123 --- The Detailed Node Listing ---
127 * Book Contents:: What this Book Contains
128 * Definitions:: Some Definitions
129 * What tar Does:: What @command{tar} Does
130 * Naming tar Archives:: How @command{tar} Archives are Named
131 * Authors:: @GNUTAR{} Authors
132 * Reports:: Reporting bugs or suggestions
134 Tutorial Introduction to @command{tar}
137 * stylistic conventions::
138 * basic tar options:: Basic @command{tar} Operations and Options
139 * frequent operations::
140 * Two Frequent Options::
141 * create:: How to Create Archives
142 * list:: How to List Archives
143 * extract:: How to Extract Members from an Archive
146 Two Frequently Used Options
152 How to Create Archives
154 * prepare for examples::
155 * Creating the archive::
164 How to Extract Members from an Archive
166 * extracting archives::
169 * extracting untrusted archives::
175 * using tar options::
185 The Three Option Styles
187 * Long Options:: Long Option Style
188 * Short Options:: Short Option Style
189 * Old Options:: Old Option Style
190 * Mixing:: Mixing Option Styles
192 All @command{tar} Options
194 * Operation Summary::
196 * Short Option Summary::
208 Advanced @GNUTAR{} Operations
217 How to Add Files to Existing Archives: @option{--append}
219 * appending files:: Appending Files to an Archive
226 Options Used by @option{--create}
228 * override:: Overriding File Metadata.
229 * Ignore Failed Read::
231 Options Used by @option{--extract}
233 * Reading:: Options to Help Read Archives
234 * Writing:: Changing How @command{tar} Writes Files
235 * Scarce:: Coping with Scarce Resources
237 Options to Help Read Archives
239 * read full records::
242 Changing How @command{tar} Writes Files
244 * Dealing with Old Files::
245 * Overwrite Old Files::
250 * Data Modification Times::
251 * Setting Access Permissions::
252 * Directory Modification Times and Permissions::
253 * Writing to Standard Output::
254 * Writing to an External Program::
257 Coping with Scarce Resources
262 Performing Backups and Restoring Files
264 * Full Dumps:: Using @command{tar} to Perform Full Dumps
265 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
266 * Backup Levels:: Levels of Backups
267 * Backup Parameters:: Setting Parameters for Backups and Restoration
268 * Scripted Backups:: Using the Backup Scripts
269 * Scripted Restoration:: Using the Restore Script
271 Setting Parameters for Backups and Restoration
273 * General-Purpose Variables::
274 * Magnetic Tape Control::
276 * backup-specs example:: An Example Text of @file{Backup-specs}
278 Choosing Files and Names for @command{tar}
280 * file:: Choosing the Archive's Name
281 * Selecting Archive Members::
282 * files:: Reading Names from a File
283 * exclude:: Excluding Some Files
284 * wildcards:: Wildcards Patterns and Matching
285 * quoting styles:: Ways of Quoting Special Characters in Names
286 * transform:: Modifying File and Member Names
287 * after:: Operating Only on New Files
288 * recurse:: Descending into Directories
289 * one:: Crossing File System Boundaries
291 Reading Names from a File
297 * problems with exclude::
299 Wildcards Patterns and Matching
301 * controlling pattern-matching::
303 Crossing File System Boundaries
305 * directory:: Changing Directory
306 * absolute:: Absolute File Names
310 * General date syntax:: Common rules.
311 * Calendar date items:: 19 Dec 1994.
312 * Time of day items:: 9:20pm.
313 * Time zone items:: @sc{est}, @sc{pdt}, @sc{gmt}.
314 * Day of week items:: Monday and others.
315 * Relative items in date strings:: next tuesday, 2 years ago.
316 * Pure numbers in date strings:: 19931219, 1440.
317 * Seconds since the Epoch:: @@1078100502.
318 * Specifying time zone rules:: TZ="America/New_York", TZ="UTC0".
319 * Authors of get_date:: Bellovin, Eggert, Salz, Berets, et al.
321 Controlling the Archive Format
323 * Compression:: Using Less Space through Compression
324 * Attributes:: Handling File Attributes
325 * Portability:: Making @command{tar} Archives More Portable
326 * cpio:: Comparison of @command{tar} and @command{cpio}
328 Using Less Space through Compression
330 * gzip:: Creating and Reading Compressed Archives
331 * sparse:: Archiving Sparse Files
333 Creating and Reading Compressed Archives
335 * lbzip2:: Using lbzip2 with @GNUTAR{}.
337 Making @command{tar} Archives More Portable
339 * Portable Names:: Portable Names
340 * dereference:: Symbolic Links
341 * hard links:: Hard Links
342 * old:: Old V7 Archives
343 * ustar:: Ustar Archives
344 * gnu:: GNU and old GNU format archives.
345 * posix:: @acronym{POSIX} archives
346 * Checksumming:: Checksumming Problems
347 * Large or Negative Values:: Large files, negative time stamps, etc.
348 * Other Tars:: How to Extract GNU-Specific Data Using
349 Other @command{tar} Implementations
351 @GNUTAR{} and @acronym{POSIX} @command{tar}
353 * PAX keywords:: Controlling Extended Header Keywords.
355 How to Extract GNU-Specific Data Using Other @command{tar} Implementations
357 * Split Recovery:: Members Split Between Volumes
358 * Sparse Recovery:: Sparse Members
360 Tapes and Other Archive Media
362 * Device:: Device selection and switching
363 * Remote Tape Server::
364 * Common Problems and Solutions::
365 * Blocking:: Blocking
366 * Many:: Many archives on one tape
367 * Using Multiple Tapes:: Using Multiple Tapes
368 * label:: Including a Label in the Archive
374 * Format Variations:: Format Variations
375 * Blocking Factor:: The Blocking Factor of an Archive
377 Many Archives on One Tape
379 * Tape Positioning:: Tape Positions and Tape Marks
380 * mt:: The @command{mt} Utility
384 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
385 * Tape Files:: Tape Files
386 * Tarcat:: Concatenate Volumes into a Single Archive
391 * Standard:: Basic Tar Format
392 * Extensions:: @acronym{GNU} Extensions to the Archive Format
393 * Sparse Formats:: Storing Sparse Files
400 * PAX 0:: PAX Format, Versions 0.0 and 0.1
401 * PAX 1:: PAX Format, Version 1.0
405 * Generate Mode:: File Generation Mode.
406 * Status Mode:: File Status Mode.
407 * Exec Mode:: Synchronous Execution mode.
411 * GNU Free Documentation License:: License for copying this manual
417 @chapter Introduction
420 and manipulates @dfn{archives} which are actually collections of
421 many other files; the program provides users with an organized and
422 systematic method for controlling a large amount of data.
423 The name ``tar'' originally came from the phrase ``Tape ARchive'', but
424 archives need not (and these days, typically do not) reside on tapes.
427 * Book Contents:: What this Book Contains
428 * Definitions:: Some Definitions
429 * What tar Does:: What @command{tar} Does
430 * Naming tar Archives:: How @command{tar} Archives are Named
431 * Authors:: @GNUTAR{} Authors
432 * Reports:: Reporting bugs or suggestions
436 @section What this Book Contains
438 The first part of this chapter introduces you to various terms that will
439 recur throughout the book. It also tells you who has worked on @GNUTAR{}
440 and its documentation, and where you should send bug reports
443 The second chapter is a tutorial (@pxref{Tutorial}) which provides a
444 gentle introduction for people who are new to using @command{tar}. It is
445 meant to be self-contained, not requiring any reading from subsequent
446 chapters to make sense. It moves from topic to topic in a logical,
447 progressive order, building on information already explained.
449 Although the tutorial is paced and structured to allow beginners to
450 learn how to use @command{tar}, it is not intended solely for beginners.
451 The tutorial explains how to use the three most frequently used
452 operations (@samp{create}, @samp{list}, and @samp{extract}) as well as
453 two frequently used options (@samp{file} and @samp{verbose}). The other
454 chapters do not refer to the tutorial frequently; however, if a section
455 discusses something which is a complex variant of a basic concept, there
456 may be a cross-reference to that basic concept. (The entire book,
457 including the tutorial, assumes that the reader understands some basic
458 concepts of using a Unix-type operating system; @pxref{Tutorial}.)
460 The third chapter presents the remaining five operations, and
461 information about using @command{tar} options and option syntax.
463 The other chapters are meant to be used as a reference. Each chapter
464 presents everything that needs to be said about a specific topic.
466 One of the chapters (@pxref{Date input formats}) exists in its
467 entirety in other @acronym{GNU} manuals, and is mostly self-contained.
468 In addition, one section of this manual (@pxref{Standard}) contains a
469 big quote which is taken directly from @command{tar} sources.
471 In general, we give both long and short (abbreviated) option names
472 at least once in each section where the relevant option is covered, so
473 that novice readers will become familiar with both styles. (A few
474 options have no short versions, and the relevant sections will
478 @section Some Definitions
482 The @command{tar} program is used to create and manipulate @command{tar}
483 archives. An @dfn{archive} is a single file which contains the contents
484 of many files, while still identifying the names of the files, their
485 owner(s), and so forth. (In addition, archives record access
486 permissions, user and group, size in bytes, and data modification time.
487 Some archives also record the file names in each archived directory, as
488 well as other file and directory information.) You can use @command{tar}
489 to @dfn{create} a new archive in a specified directory.
492 @cindex archive member
495 The files inside an archive are called @dfn{members}. Within this
496 manual, we use the term @dfn{file} to refer only to files accessible in
497 the normal ways (by @command{ls}, @command{cat}, and so forth), and the term
498 @dfn{member} to refer only to the members of an archive. Similarly, a
499 @dfn{file name} is the name of a file, as it resides in the file system,
500 and a @dfn{member name} is the name of an archive member within the
505 The term @dfn{extraction} refers to the process of copying an archive
506 member (or multiple members) into a file in the file system. Extracting
507 all the members of an archive is often called @dfn{extracting the
508 archive}. The term @dfn{unpack} can also be used to refer to the
509 extraction of many or all the members of an archive. Extracting an
510 archive does not destroy the archive's structure, just as creating an
511 archive does not destroy the copies of the files that exist outside of
512 the archive. You may also @dfn{list} the members in a given archive
513 (this is often thought of as ``printing'' them to the standard output,
514 or the command line), or @dfn{append} members to a pre-existing archive.
515 All of these operations can be performed using @command{tar}.
518 @section What @command{tar} Does
521 The @command{tar} program provides the ability to create @command{tar}
522 archives, as well as various other kinds of manipulation. For example,
523 you can use @command{tar} on previously created archives to extract files,
524 to store additional files, or to update or list files which were already
527 Initially, @command{tar} archives were used to store files conveniently on
528 magnetic tape. The name @command{tar} comes from this use; it stands for
529 @code{t}ape @code{ar}chiver. Despite the utility's name, @command{tar} can
530 direct its output to available devices, files, or other programs (using
531 pipes). @command{tar} may even access remote devices or files (as archives).
533 You can use @command{tar} archives in many ways. We want to stress a few
534 of them: storage, backup, and transportation.
536 @FIXME{the following table entries need a bit of work.}
539 Often, @command{tar} archives are used to store related files for
540 convenient file transfer over a network. For example, the
541 @acronym{GNU} Project distributes its software bundled into
542 @command{tar} archives, so that all the files relating to a particular
543 program (or set of related programs) can be transferred as a single
546 A magnetic tape can store several files in sequence. However, the tape
547 has no names for these files; it only knows their relative position on
548 the tape. One way to store several files on one tape and retain their
549 names is by creating a @command{tar} archive. Even when the basic transfer
550 mechanism can keep track of names, as FTP can, the nuisance of handling
551 multiple files, directories, and multiple links makes @command{tar}
554 Archive files are also used for long-term storage. You can think of
555 this as transportation from the present into the future. (It is a
556 science-fiction idiom that you can move through time as well as in
557 space; the idea here is that @command{tar} can be used to move archives in
558 all dimensions, even time!)
561 Because the archive created by @command{tar} is capable of preserving
562 file information and directory structure, @command{tar} is commonly
563 used for performing full and incremental backups of disks. A backup
564 puts a collection of files (possibly pertaining to many users and
565 projects) together on a disk or a tape. This guards against
566 accidental destruction of the information in those files.
567 @GNUTAR{} has special features that allow it to be
568 used to make incremental and full dumps of all the files in a
572 You can create an archive on one system, transfer it to another system,
573 and extract the contents there. This allows you to transport a group of
574 files from one system to another.
577 @node Naming tar Archives
578 @section How @command{tar} Archives are Named
580 Conventionally, @command{tar} archives are given names ending with
581 @samp{.tar}. This is not necessary for @command{tar} to operate properly,
582 but this manual follows that convention in order to accustom readers to
583 it and to make examples more clear.
588 Often, people refer to @command{tar} archives as ``@command{tar} files,'' and
589 archive members as ``files'' or ``entries''. For people familiar with
590 the operation of @command{tar}, this causes no difficulty. However, in
591 this manual, we consistently refer to ``archives'' and ``archive
592 members'' to make learning to use @command{tar} easier for novice users.
595 @section @GNUTAR{} Authors
597 @GNUTAR{} was originally written by John Gilmore,
598 and modified by many people. The @acronym{GNU} enhancements were
599 written by Jay Fenlason, then Joy Kendall, and the whole package has
600 been further maintained by Thomas Bushnell, n/BSG, Fran@,{c}ois
601 Pinard, Paul Eggert, and finally Sergey Poznyakoff with the help of
602 numerous and kind users.
604 We wish to stress that @command{tar} is a collective work, and owes much to
605 all those people who reported problems, offered solutions and other
606 insights, or shared their thoughts and suggestions. An impressive, yet
607 partial list of those contributors can be found in the @file{THANKS}
608 file from the @GNUTAR{} distribution.
610 @FIXME{i want all of these names mentioned, Absolutely. BUT, i'm not
611 sure i want to spell out the history in this detail, at least not for
612 the printed book. i'm just not sure it needs to be said this way.
613 i'll think about it.}
615 @FIXME{History is more important, and surely more interesting, than
616 actual names. Quoting names without history would be meaningless. FP}
618 Jay Fenlason put together a draft of a @GNUTAR{}
619 manual, borrowing notes from the original man page from John Gilmore.
620 This was withdrawn in version 1.11. Thomas Bushnell, n/BSG and Amy
621 Gorin worked on a tutorial and manual for @GNUTAR{}.
622 Fran@,{c}ois Pinard put version 1.11.8 of the manual together by
623 taking information from all these sources and merging them. Melissa
624 Weisshaus finally edited and redesigned the book to create version
625 1.12. The book for versions from 1.14 up to @value{VERSION} were edited
626 by the current maintainer, Sergey Poznyakoff.
628 For version 1.12, Daniel Hagerty contributed a great deal of technical
629 consulting. In particular, he is the primary author of @ref{Backups}.
631 In July, 2003 @GNUTAR{} was put on CVS at savannah.gnu.org
632 (see @url{http://savannah.gnu.org/projects/tar}), and
633 active development and maintenance work has started
634 again. Currently @GNUTAR{} is being maintained by Paul Eggert, Sergey
635 Poznyakoff and Jeff Bailey.
637 Support for @acronym{POSIX} archives was added by Sergey Poznyakoff.
640 @section Reporting bugs or suggestions
643 @cindex reporting bugs
644 If you find problems or have suggestions about this program or manual,
645 please report them to @file{bug-tar@@gnu.org}.
647 When reporting a bug, please be sure to include as much detail as
648 possible, in order to reproduce it. @FIXME{Be more specific, I'd
649 like to make this node as detailed as 'Bug reporting' node in Emacs
653 @chapter Tutorial Introduction to @command{tar}
655 This chapter guides you through some basic examples of three @command{tar}
656 operations: @option{--create}, @option{--list}, and @option{--extract}. If
657 you already know how to use some other version of @command{tar}, then you
658 may not need to read this chapter. This chapter omits most complicated
659 details about how @command{tar} works.
663 * stylistic conventions::
664 * basic tar options:: Basic @command{tar} Operations and Options
665 * frequent operations::
666 * Two Frequent Options::
667 * create:: How to Create Archives
668 * list:: How to List Archives
669 * extract:: How to Extract Members from an Archive
674 @section Assumptions this Tutorial Makes
676 This chapter is paced to allow beginners to learn about @command{tar}
677 slowly. At the same time, we will try to cover all the basic aspects of
678 these three operations. In order to accomplish both of these tasks, we
679 have made certain assumptions about your knowledge before reading this
680 manual, and the hardware you will be using:
684 Before you start to work through this tutorial, you should understand
685 what the terms ``archive'' and ``archive member'' mean
686 (@pxref{Definitions}). In addition, you should understand something
687 about how Unix-type operating systems work, and you should know how to
688 use some basic utilities. For example, you should know how to create,
689 list, copy, rename, edit, and delete files and directories; how to
690 change between directories; and how to figure out where you are in the
691 file system. You should have some basic understanding of directory
692 structure and how files are named according to which directory they are
693 in. You should understand concepts such as standard output and standard
694 input, what various definitions of the term @samp{argument} mean, and the
695 differences between relative and absolute file names. @FIXME{and what
699 This manual assumes that you are working from your own home directory
700 (unless we state otherwise). In this tutorial, you will create a
701 directory to practice @command{tar} commands in. When we show file names,
702 we will assume that those names are relative to your home directory.
703 For example, my home directory is @file{/home/fsf/melissa}. All of
704 my examples are in a subdirectory of the directory named by that file
705 name; the subdirectory is called @file{practice}.
708 In general, we show examples of archives which exist on (or can be
709 written to, or worked with from) a directory on a hard disk. In most
710 cases, you could write those archives to, or work with them on any other
711 device, such as a tape drive. However, some of the later examples in
712 the tutorial and next chapter will not work on tape drives.
713 Additionally, working with tapes is much more complicated than working
714 with hard disks. For these reasons, the tutorial does not cover working
715 with tape drives. @xref{Media}, for complete information on using
716 @command{tar} archives with tape drives.
718 @FIXME{this is a cop out. need to add some simple tape drive info.}
721 @node stylistic conventions
722 @section Stylistic Conventions
724 In the examples, @samp{$} represents a typical shell prompt. It
725 precedes lines you should type; to make this more clear, those lines are
726 shown in @kbd{this font}, as opposed to lines which represent the
727 computer's response; those lines are shown in @code{this font}, or
728 sometimes @samp{like this}.
730 @c When we have lines which are too long to be
731 @c displayed in any other way, we will show them like this:
733 @node basic tar options
734 @section Basic @command{tar} Operations and Options
736 @command{tar} can take a wide variety of arguments which specify and define
737 the actions it will have on the particular set of files or the archive.
738 The main types of arguments to @command{tar} fall into one of two classes:
739 operations, and options.
741 Some arguments fall into a class called @dfn{operations}; exactly one of
742 these is both allowed and required for any instance of using @command{tar};
743 you may @emph{not} specify more than one. People sometimes speak of
744 @dfn{operating modes}. You are in a particular operating mode when you
745 have specified the operation which specifies it; there are eight
746 operations in total, and thus there are eight operating modes.
748 The other arguments fall into the class known as @dfn{options}. You are
749 not required to specify any options, and you are allowed to specify more
750 than one at a time (depending on the way you are using @command{tar} at
751 that time). Some options are used so frequently, and are so useful for
752 helping you type commands more carefully that they are effectively
753 ``required''. We will discuss them in this chapter.
755 You can write most of the @command{tar} operations and options in any
756 of three forms: long (mnemonic) form, short form, and old style. Some
757 of the operations and options have no short or ``old'' forms; however,
758 the operations and options which we will cover in this tutorial have
759 corresponding abbreviations. We will indicate those abbreviations
760 appropriately to get you used to seeing them. Note, that the ``old
761 style'' option forms exist in @GNUTAR{} for compatibility with Unix
762 @command{tar}. In this book we present a full discussion of this way
763 of writing options and operations (@pxref{Old Options}), and we discuss
764 the other two styles of writing options (@xref{Long Options}, and
765 @pxref{Short Options}).
767 In the examples and in the text of this tutorial, we usually use the
768 long forms of operations and options; but the ``short'' forms produce
769 the same result and can make typing long @command{tar} commands easier.
770 For example, instead of typing
773 @kbd{tar --create --verbose --file=afiles.tar apple angst aspic}
779 @kbd{tar -c -v -f afiles.tar apple angst aspic}
785 @kbd{tar -cvf afiles.tar apple angst aspic}
789 For more information on option syntax, see @ref{Advanced tar}. In
790 discussions in the text, when we name an option by its long form, we
791 also give the corresponding short option in parentheses.
793 The term, ``option'', can be confusing at times, since ``operations''
794 are often lumped in with the actual, @emph{optional} ``options'' in certain
795 general class statements. For example, we just talked about ``short and
796 long forms of options and operations''. However, experienced @command{tar}
797 users often refer to these by shorthand terms such as, ``short and long
798 options''. This term assumes that the ``operations'' are included, also.
799 Context will help you determine which definition of ``options'' to use.
801 Similarly, the term ``command'' can be confusing, as it is often used in
802 two different ways. People sometimes refer to @command{tar} ``commands''.
803 A @command{tar} @dfn{command} is the entire command line of user input
804 which tells @command{tar} what to do --- including the operation, options,
805 and any arguments (file names, pipes, other commands, etc.). However,
806 you will also sometimes hear the term ``the @command{tar} command''. When
807 the word ``command'' is used specifically like this, a person is usually
808 referring to the @command{tar} @emph{operation}, not the whole line.
809 Again, use context to figure out which of the meanings the speaker
812 @node frequent operations
813 @section The Three Most Frequently Used Operations
815 Here are the three most frequently used operations (both short and long
816 forms), as well as a brief description of their meanings. The rest of
817 this chapter will cover how to use these operations in detail. We will
818 present the rest of the operations in the next chapter.
823 Create a new @command{tar} archive.
826 List the contents of an archive.
829 Extract one or more members from an archive.
832 @node Two Frequent Options
833 @section Two Frequently Used Options
835 To understand how to run @command{tar} in the three operating modes listed
836 previously, you also need to understand how to use two of the options to
837 @command{tar}: @option{--file} (which takes an archive file as an argument)
838 and @option{--verbose}. (You are usually not @emph{required} to specify
839 either of these options when you run @command{tar}, but they can be very
840 useful in making things more clear and helping you avoid errors.)
849 @unnumberedsubsec The @option{--file} Option
852 @xopindex{file, tutorial}
853 @item --file=@var{archive-name}
854 @itemx -f @var{archive-name}
855 Specify the name of an archive file.
858 You can specify an argument for the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option whenever you
859 use @command{tar}; this option determines the name of the archive file
860 that @command{tar} will work on.
863 If you don't specify this argument, then @command{tar} will examine
864 the environment variable @env{TAPE}. If it is set, its value will be
865 used as the archive name. Otherwise, @command{tar} will use the
866 default archive, determined at compile time. Usually it is
867 standard output or some physical tape drive attached to your machine
868 (you can verify what the default is by running @kbd{tar
869 --show-defaults}, @pxref{defaults}). If there is no tape drive
870 attached, or the default is not meaningful, then @command{tar} will
871 print an error message. The error message might look roughly like one
875 tar: can't open /dev/rmt8 : No such device or address
876 tar: can't open /dev/rsmt0 : I/O error
880 To avoid confusion, we recommend that you always specify an archive file
881 name by using @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) when writing your @command{tar} commands.
882 For more information on using the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option, see
885 @node verbose tutorial
886 @unnumberedsubsec The @option{--verbose} Option
889 @xopindex{verbose, introduced}
892 Show the files being worked on as @command{tar} is running.
895 @option{--verbose} (@option{-v}) shows details about the results of running
896 @command{tar}. This can be especially useful when the results might not be
897 obvious. For example, if you want to see the progress of @command{tar} as
898 it writes files into the archive, you can use the @option{--verbose}
899 option. In the beginning, you may find it useful to use
900 @option{--verbose} at all times; when you are more accustomed to
901 @command{tar}, you will likely want to use it at certain times but not at
902 others. We will use @option{--verbose} at times to help make something
903 clear, and we will give many examples both using and not using
904 @option{--verbose} to show the differences.
906 Each instance of @option{--verbose} on the command line increases the
907 verbosity level by one, so if you need more details on the output,
910 When reading archives (@option{--list}, @option{--extract},
911 @option{--diff}), @command{tar} by default prints only the names of
912 the members being extracted. Using @option{--verbose} will show a full,
913 @command{ls} style member listing.
915 In contrast, when writing archives (@option{--create}, @option{--append},
916 @option{--update}), @command{tar} does not print file names by
917 default. So, a single @option{--verbose} option shows the file names
918 being added to the archive, while two @option{--verbose} options
919 enable the full listing.
921 For example, to create an archive in verbose mode:
924 $ @kbd{tar -cvf afiles.tar apple angst aspic}
931 Creating the same archive with the verbosity level 2 could give:
934 $ @kbd{tar -cvvf afiles.tar apple angst aspic}
935 -rw-r--r-- gray/staff 62373 2006-06-09 12:06 apple
936 -rw-r--r-- gray/staff 11481 2006-06-09 12:06 angst
937 -rw-r--r-- gray/staff 23152 2006-06-09 12:06 aspic
941 This works equally well using short or long forms of options. Using
942 long forms, you would simply write out the mnemonic form of the option
946 $ @kbd{tar --create --verbose --verbose @dots{}}
950 Note that you must double the hyphens properly each time.
952 Later in the tutorial, we will give examples using @w{@option{--verbose
955 @anchor{verbose member listing}
956 The full output consists of six fields:
959 @item File type and permissions in symbolic form.
960 These are displayed in the same format as the first column of
961 @command{ls -l} output (@pxref{What information is listed,
962 format=verbose, Verbose listing, fileutils, GNU file utilities}).
964 @item Owner name and group separated by a slash character.
965 If these data are not available (for example, when listing a @samp{v7} format
966 archive), numeric @acronym{ID} values are printed instead.
968 @item Size of the file, in bytes.
970 @item File modification date in ISO 8601 format.
972 @item File modification time.
975 If the name contains any special characters (white space, newlines,
976 etc.) these are displayed in an unambiguous form using so called
977 @dfn{quoting style}. For the detailed discussion of available styles
978 and on how to use them, see @ref{quoting styles}.
980 Depending on the file type, the name can be followed by some
981 additional information, described in the following table:
984 @item -> @var{link-name}
985 The file or archive member is a @dfn{symbolic link} and
986 @var{link-name} is the name of file it links to.
988 @item link to @var{link-name}
989 The file or archive member is a @dfn{hard link} and @var{link-name} is
990 the name of file it links to.
993 The archive member is an old GNU format long link. You will normally
997 The archive member is an old GNU format long name. You will normally
1000 @item --Volume Header--
1001 The archive member is a GNU @dfn{volume header} (@pxref{Tape Files}).
1003 @item --Continued at byte @var{n}--
1004 Encountered only at the beginning of a multi-volume archive
1005 (@pxref{Using Multiple Tapes}). This archive member is a continuation
1006 from the previous volume. The number @var{n} gives the offset where
1007 the original file was split.
1009 @item unknown file type @var{c}
1010 An archive member of unknown type. @var{c} is the type character from
1011 the archive header. If you encounter such a message, it means that
1012 either your archive contains proprietary member types @GNUTAR{} is not
1013 able to handle, or the archive is corrupted.
1018 For example, here is an archive listing containing most of the special
1019 suffixes explained above:
1023 V--------- 0/0 1536 2006-06-09 13:07 MyVolume--Volume Header--
1024 -rw-r--r-- gray/staff 456783 2006-06-09 12:06 aspic--Continued at
1026 -rw-r--r-- gray/staff 62373 2006-06-09 12:06 apple
1027 lrwxrwxrwx gray/staff 0 2006-06-09 13:01 angst -> apple
1028 -rw-r--r-- gray/staff 35793 2006-06-09 12:06 blues
1029 hrw-r--r-- gray/staff 0 2006-06-09 12:06 music link to blues
1037 @unnumberedsubsec Getting Help: Using the @option{--help} Option
1043 The @option{--help} option to @command{tar} prints out a very brief list of
1044 all operations and option available for the current version of
1045 @command{tar} available on your system.
1049 @section How to Create Archives
1052 @cindex Creation of the archive
1053 @cindex Archive, creation of
1054 One of the basic operations of @command{tar} is @option{--create} (@option{-c}), which
1055 you use to create a @command{tar} archive. We will explain
1056 @option{--create} first because, in order to learn about the other
1057 operations, you will find it useful to have an archive available to
1060 To make this easier, in this section you will first create a directory
1061 containing three files. Then, we will show you how to create an
1062 @emph{archive} (inside the new directory). Both the directory, and
1063 the archive are specifically for you to practice on. The rest of this
1064 chapter and the next chapter will show many examples using this
1065 directory and the files you will create: some of those files may be
1066 other directories and other archives.
1068 The three files you will archive in this example are called
1069 @file{blues}, @file{folk}, and @file{jazz}. The archive is called
1070 @file{collection.tar}.
1072 This section will proceed slowly, detailing how to use @option{--create}
1073 in @code{verbose} mode, and showing examples using both short and long
1074 forms. In the rest of the tutorial, and in the examples in the next
1075 chapter, we will proceed at a slightly quicker pace. This section
1076 moves more slowly to allow beginning users to understand how
1077 @command{tar} works.
1080 * prepare for examples::
1081 * Creating the archive::
1087 @node prepare for examples
1088 @subsection Preparing a Practice Directory for Examples
1090 To follow along with this and future examples, create a new directory
1091 called @file{practice} containing files called @file{blues}, @file{folk}
1092 and @file{jazz}. The files can contain any information you like:
1093 ideally, they should contain information which relates to their names,
1094 and be of different lengths. Our examples assume that @file{practice}
1095 is a subdirectory of your home directory.
1097 Now @command{cd} to the directory named @file{practice}; @file{practice}
1098 is now your @dfn{working directory}. (@emph{Please note}: Although
1099 the full file name of this directory is
1100 @file{/@var{homedir}/practice}, in our examples we will refer to
1101 this directory as @file{practice}; the @var{homedir} is presumed.)
1103 In general, you should check that the files to be archived exist where
1104 you think they do (in the working directory) by running @command{ls}.
1105 Because you just created the directory and the files and have changed to
1106 that directory, you probably don't need to do that this time.
1108 It is very important to make sure there isn't already a file in the
1109 working directory with the archive name you intend to use (in this case,
1110 @samp{collection.tar}), or that you don't care about its contents.
1111 Whenever you use @samp{create}, @command{tar} will erase the current
1112 contents of the file named by @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) if it exists. @command{tar}
1113 will not tell you if you are about to overwrite an archive unless you
1114 specify an option which does this (@pxref{backup}, for the
1115 information on how to do so). To add files to an existing archive,
1116 you need to use a different option, such as @option{--append} (@option{-r}); see
1117 @ref{append} for information on how to do this.
1119 @node Creating the archive
1120 @subsection Creating the Archive
1122 @xopindex{create, introduced}
1123 To place the files @file{blues}, @file{folk}, and @file{jazz} into an
1124 archive named @file{collection.tar}, use the following command:
1127 $ @kbd{tar --create --file=collection.tar blues folk jazz}
1130 The order of the arguments is not very important, @emph{when using long
1131 option forms}. You could also say:
1134 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1138 However, you can see that this order is harder to understand; this is
1139 why we will list the arguments in the order that makes the commands
1140 easiest to understand (and we encourage you to do the same when you use
1141 @command{tar}, to avoid errors).
1143 Note that the sequence
1144 @option{--file=@-collection.tar} is considered to be @emph{one} argument.
1145 If you substituted any other string of characters for
1146 @kbd{collection.tar}, then that string would become the name of the
1147 archive file you create.
1149 The order of the options becomes more important when you begin to use
1150 short forms. With short forms, if you type commands in the wrong order
1151 (even if you type them correctly in all other ways), you may end up with
1152 results you don't expect. For this reason, it is a good idea to get
1153 into the habit of typing options in the order that makes inherent sense.
1154 @xref{short create}, for more information on this.
1156 In this example, you type the command as shown above: @option{--create}
1157 is the operation which creates the new archive
1158 (@file{collection.tar}), and @option{--file} is the option which lets
1159 you give it the name you chose. The files, @file{blues}, @file{folk},
1160 and @file{jazz}, are now members of the archive, @file{collection.tar}
1161 (they are @dfn{file name arguments} to the @option{--create} operation.
1162 @xref{Choosing}, for the detailed discussion on these.) Now that they are
1163 in the archive, they are called @emph{archive members}, not files.
1164 (@pxref{Definitions,members}).
1166 When you create an archive, you @emph{must} specify which files you
1167 want placed in the archive. If you do not specify any archive
1168 members, @GNUTAR{} will complain.
1170 If you now list the contents of the working directory (@command{ls}), you will
1171 find the archive file listed as well as the files you saw previously:
1174 blues folk jazz collection.tar
1178 Creating the archive @samp{collection.tar} did not destroy the copies of
1179 the files in the directory.
1181 Keep in mind that if you don't indicate an operation, @command{tar} will not
1182 run and will prompt you for one. If you don't name any files, @command{tar}
1183 will complain. You must have write access to the working directory,
1184 or else you will not be able to create an archive in that directory.
1186 @emph{Caution}: Do not attempt to use @option{--create} (@option{-c}) to add files to
1187 an existing archive; it will delete the archive and write a new one.
1188 Use @option{--append} (@option{-r}) instead. @xref{append}.
1190 @node create verbose
1191 @subsection Running @option{--create} with @option{--verbose}
1193 @xopindex{create, using with @option{--verbose}}
1194 @xopindex{verbose, using with @option{--create}}
1195 If you include the @option{--verbose} (@option{-v}) option on the command line,
1196 @command{tar} will list the files it is acting on as it is working. In
1197 verbose mode, the @code{create} example above would appear as:
1200 $ @kbd{tar --create --verbose --file=collection.tar blues folk jazz}
1206 This example is just like the example we showed which did not use
1207 @option{--verbose}, except that @command{tar} generated the remaining
1209 lines (note the different font styles).
1215 In the rest of the examples in this chapter, we will frequently use
1216 @code{verbose} mode so we can show actions or @command{tar} responses that
1217 you would otherwise not see, and which are important for you to
1221 @subsection Short Forms with @samp{create}
1223 As we said before, the @option{--create} (@option{-c}) operation is one of the most
1224 basic uses of @command{tar}, and you will use it countless times.
1225 Eventually, you will probably want to use abbreviated (or ``short'')
1226 forms of options. A full discussion of the three different forms that
1227 options can take appears in @ref{Styles}; for now, here is what the
1228 previous example (including the @option{--verbose} (@option{-v}) option) looks like
1229 using short option forms:
1232 $ @kbd{tar -cvf collection.tar blues folk jazz}
1239 As you can see, the system responds the same no matter whether you use
1240 long or short option forms.
1242 @FIXME{i don't like how this is worded:} One difference between using
1243 short and long option forms is that, although the exact placement of
1244 arguments following options is no more specific when using short forms,
1245 it is easier to become confused and make a mistake when using short
1246 forms. For example, suppose you attempted the above example in the
1250 $ @kbd{tar -cfv collection.tar blues folk jazz}
1254 In this case, @command{tar} will make an archive file called @file{v},
1255 containing the files @file{blues}, @file{folk}, and @file{jazz}, because
1256 the @samp{v} is the closest ``file name'' to the @option{-f} option, and
1257 is thus taken to be the chosen archive file name. @command{tar} will try
1258 to add a file called @file{collection.tar} to the @file{v} archive file;
1259 if the file @file{collection.tar} did not already exist, @command{tar} will
1260 report an error indicating that this file does not exist. If the file
1261 @file{collection.tar} does already exist (e.g., from a previous command
1262 you may have run), then @command{tar} will add this file to the archive.
1263 Because the @option{-v} option did not get registered, @command{tar} will not
1264 run under @samp{verbose} mode, and will not report its progress.
1266 The end result is that you may be quite confused about what happened,
1267 and possibly overwrite a file. To illustrate this further, we will show
1268 you how an example we showed previously would look using short forms.
1273 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1277 is confusing as it is. When shown using short forms, however, it
1278 becomes much more so:
1281 $ @kbd{tar blues -c folk -f collection.tar jazz}
1285 It would be very easy to put the wrong string of characters
1286 immediately following the @option{-f}, but doing that could sacrifice
1289 For this reason, we recommend that you pay very careful attention to
1290 the order of options and placement of file and archive names,
1291 especially when using short option forms. Not having the option name
1292 written out mnemonically can affect how well you remember which option
1293 does what, and therefore where different names have to be placed.
1296 @subsection Archiving Directories
1298 @cindex Archiving Directories
1299 @cindex Directories, Archiving
1300 You can archive a directory by specifying its directory name as a
1301 file name argument to @command{tar}. The files in the directory will be
1302 archived relative to the working directory, and the directory will be
1303 re-created along with its contents when the archive is extracted.
1305 To archive a directory, first move to its superior directory. If you
1306 have followed the previous instructions in this tutorial, you should
1315 This will put you into the directory which contains @file{practice},
1316 i.e., your home directory. Once in the superior directory, you can
1317 specify the subdirectory, @file{practice}, as a file name argument. To
1318 store @file{practice} in the new archive file @file{music.tar}, type:
1321 $ @kbd{tar --create --verbose --file=music.tar practice}
1325 @command{tar} should output:
1332 practice/collection.tar
1335 Note that the archive thus created is not in the subdirectory
1336 @file{practice}, but rather in the current working directory---the
1337 directory from which @command{tar} was invoked. Before trying to archive a
1338 directory from its superior directory, you should make sure you have
1339 write access to the superior directory itself, not only the directory
1340 you are trying archive with @command{tar}. For example, you will probably
1341 not be able to store your home directory in an archive by invoking
1342 @command{tar} from the root directory; @xref{absolute}. (Note
1343 also that @file{collection.tar}, the original archive file, has itself
1344 been archived. @command{tar} will accept any file as a file to be
1345 archived, regardless of its content. When @file{music.tar} is
1346 extracted, the archive file @file{collection.tar} will be re-written
1347 into the file system).
1349 If you give @command{tar} a command such as
1352 $ @kbd{tar --create --file=foo.tar .}
1356 @command{tar} will report @samp{tar: ./foo.tar is the archive; not
1357 dumped}. This happens because @command{tar} creates the archive
1358 @file{foo.tar} in the current directory before putting any files into
1359 it. Then, when @command{tar} attempts to add all the files in the
1360 directory @file{.} to the archive, it notices that the file
1361 @file{./foo.tar} is the same as the archive @file{foo.tar}, and skips
1362 it. (It makes no sense to put an archive into itself.) @GNUTAR{}
1363 will continue in this case, and create the archive
1364 normally, except for the exclusion of that one file. (@emph{Please
1365 note:} Other implementations of @command{tar} may not be so clever;
1366 they will enter an infinite loop when this happens, so you should not
1367 depend on this behavior unless you are certain you are running
1368 @GNUTAR{}. In general, it is wise to always place the archive outside
1369 of the directory being dumped.)
1372 @section How to List Archives
1375 Frequently, you will find yourself wanting to determine exactly what a
1376 particular archive contains. You can use the @option{--list}
1377 (@option{-t}) operation to get the member names as they currently
1378 appear in the archive, as well as various attributes of the files at
1379 the time they were archived. For example, you can examine the archive
1380 @file{collection.tar} that you created in the last section with the
1384 $ @kbd{tar --list --file=collection.tar}
1388 The output of @command{tar} would then be:
1397 The archive @file{bfiles.tar} would list as follows:
1406 Be sure to use a @option{--file=@var{archive-name}} (@option{-f
1407 @var{archive-name}}) option just as with @option{--create}
1408 (@option{-c}) to specify the name of the archive.
1410 @xopindex{list, using with @option{--verbose}}
1411 @xopindex{verbose, using with @option{--list}}
1412 If you use the @option{--verbose} (@option{-v}) option with
1413 @option{--list}, then @command{tar} will print out a listing
1414 reminiscent of @w{@samp{ls -l}}, showing owner, file size, and so
1415 forth. This output is described in detail in @ref{verbose member listing}.
1417 If you had used @option{--verbose} (@option{-v}) mode, the example
1418 above would look like:
1421 $ @kbd{tar --list --verbose --file=collection.tar folk}
1422 -rw-r--r-- myself user 62 1990-05-23 10:55 folk
1425 @cindex listing member and file names
1426 @anchor{listing member and file names}
1427 It is important to notice that the output of @kbd{tar --list
1428 --verbose} does not necessarily match that produced by @kbd{tar
1429 --create --verbose} while creating the archive. It is because
1430 @GNUTAR{}, unless told explicitly not to do so, removes some directory
1431 prefixes from file names before storing them in the archive
1432 (@xref{absolute}, for more information). In other
1433 words, in verbose mode @GNUTAR{} shows @dfn{file names} when creating
1434 an archive and @dfn{member names} when listing it. Consider this
1439 $ @kbd{tar --create --verbose --file archive /etc/mail}
1440 tar: Removing leading `/' from member names
1442 /etc/mail/sendmail.cf
1444 $ @kbd{tar --test --file archive}
1446 etc/mail/sendmail.cf
1451 @opindex show-stored-names
1452 This default behavior can sometimes be inconvenient. You can force
1453 @GNUTAR{} show member names when creating archive by supplying
1454 @option{--show-stored-names} option.
1457 @item --show-stored-names
1458 Print member (as opposed to @emph{file}) names when creating the archive.
1461 @cindex File name arguments, using @option{--list} with
1462 @xopindex{list, using with file name arguments}
1463 You can specify one or more individual member names as arguments when
1464 using @samp{list}. In this case, @command{tar} will only list the
1465 names of members you identify. For example, @w{@kbd{tar --list
1466 --file=afiles.tar apple}} would only print @file{apple}.
1468 Because @command{tar} preserves file names, these must be specified as
1469 they appear in the archive (i.e., relative to the directory from which
1470 the archive was created). Therefore, it is essential when specifying
1471 member names to @command{tar} that you give the exact member names.
1472 For example, @w{@kbd{tar --list --file=bfiles.tar birds}} would produce an
1473 error message something like @samp{tar: birds: Not found in archive},
1474 because there is no member named @file{birds}, only one named
1475 @file{./birds}. While the names @file{birds} and @file{./birds} name
1476 the same file, @emph{member} names by default are compared verbatim.
1478 However, @w{@kbd{tar --list --file=bfiles.tar baboon}} would respond
1479 with @file{baboon}, because this exact member name is in the archive file
1480 @file{bfiles.tar}. If you are not sure of the exact file name,
1481 use @dfn{globbing patterns}, for example:
1484 $ @kbd{tar --list --file=bfiles.tar --wildcards '*b*'}
1488 will list all members whose name contains @samp{b}. @xref{wildcards},
1489 for a detailed discussion of globbing patterns and related
1490 @command{tar} command line options.
1497 @unnumberedsubsec Listing the Contents of a Stored Directory
1499 To get information about the contents of an archived directory,
1500 use the directory name as a file name argument in conjunction with
1501 @option{--list} (@option{-t}). To find out file attributes, include the
1502 @option{--verbose} (@option{-v}) option.
1504 For example, to find out about files in the directory @file{practice}, in
1505 the archive file @file{music.tar}, type:
1508 $ @kbd{tar --list --verbose --file=music.tar practice}
1511 @command{tar} responds:
1514 drwxrwxrwx myself user 0 1990-05-31 21:49 practice/
1515 -rw-r--r-- myself user 42 1990-05-21 13:29 practice/blues
1516 -rw-r--r-- myself user 62 1990-05-23 10:55 practice/folk
1517 -rw-r--r-- myself user 40 1990-05-21 13:30 practice/jazz
1518 -rw-r--r-- myself user 10240 1990-05-31 21:49 practice/collection.tar
1521 When you use a directory name as a file name argument, @command{tar} acts on
1522 all the files (including sub-directories) in that directory.
1525 @section How to Extract Members from an Archive
1527 @cindex Retrieving files from an archive
1528 @cindex Resurrecting files from an archive
1531 Creating an archive is only half the job---there is no point in storing
1532 files in an archive if you can't retrieve them. The act of retrieving
1533 members from an archive so they can be used and manipulated as
1534 unarchived files again is called @dfn{extraction}. To extract files
1535 from an archive, use the @option{--extract} (@option{--get} or
1536 @option{-x}) operation. As with @option{--create}, specify the name
1537 of the archive with @option{--file} (@option{-f}) option. Extracting
1538 an archive does not modify the archive in any way; you can extract it
1539 multiple times if you want or need to.
1541 Using @option{--extract}, you can extract an entire archive, or specific
1542 files. The files can be directories containing other files, or not. As
1543 with @option{--create} (@option{-c}) and @option{--list} (@option{-t}), you may use the short or the
1544 long form of the operation without affecting the performance.
1547 * extracting archives::
1548 * extracting files::
1550 * extracting untrusted archives::
1551 * failing commands::
1554 @node extracting archives
1555 @subsection Extracting an Entire Archive
1557 To extract an entire archive, specify the archive file name only, with
1558 no individual file names as arguments. For example,
1561 $ @kbd{tar -xvf collection.tar}
1568 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
1569 -rw-r--r-- me user 21 1996-09-23 16:44 blues
1570 -rw-r--r-- me user 20 1996-09-23 16:44 folk
1573 @node extracting files
1574 @subsection Extracting Specific Files
1576 To extract specific archive members, give their exact member names as
1577 arguments, as printed by @option{--list} (@option{-t}). If you had
1578 mistakenly deleted one of the files you had placed in the archive
1579 @file{collection.tar} earlier (say, @file{blues}), you can extract it
1580 from the archive without changing the archive's structure. Its
1581 contents will be identical to the original file @file{blues} that you
1584 First, make sure you are in the @file{practice} directory, and list the
1585 files in the directory. Now, delete the file, @samp{blues}, and list
1586 the files in the directory again.
1588 You can now extract the member @file{blues} from the archive file
1589 @file{collection.tar} like this:
1592 $ @kbd{tar --extract --file=collection.tar blues}
1596 If you list the files in the directory again, you will see that the file
1597 @file{blues} has been restored, with its original permissions, data
1598 modification times, and owner.@footnote{This is only accidentally
1599 true, but not in general. Whereas modification times are always
1600 restored, in most cases, one has to be root for restoring the owner,
1601 and use a special option for restoring permissions. Here, it just
1602 happens that the restoring user is also the owner of the archived
1603 members, and that the current @code{umask} is compatible with original
1604 permissions.} (These parameters will be identical to those which
1605 the file had when you originally placed it in the archive; any changes
1606 you may have made before deleting the file from the file system,
1607 however, will @emph{not} have been made to the archive member.) The
1608 archive file, @samp{collection.tar}, is the same as it was before you
1609 extracted @samp{blues}. You can confirm this by running @command{tar} with
1610 @option{--list} (@option{-t}).
1612 Remember that as with other operations, specifying the exact member
1613 name is important. @w{@kbd{tar --extract --file=bfiles.tar birds}}
1614 will fail, because there is no member named @file{birds}. To extract
1615 the member named @file{./birds}, you must specify @w{@kbd{tar
1616 --extract --file=bfiles.tar ./birds}}. If you don't remember the
1617 exact member names, use @option{--list} (@option{-t}) option
1618 (@pxref{list}). You can also extract those members that match a
1619 specific @dfn{globbing pattern}. For example, to extract from
1620 @file{bfiles.tar} all files that begin with @samp{b}, no matter their
1621 directory prefix, you could type:
1624 $ @kbd{tar -x -f bfiles.tar --wildcards --no-anchored 'b*'}
1628 Here, @option{--wildcards} instructs @command{tar} to treat
1629 command line arguments as globbing patterns and @option{--no-anchored}
1630 informs it that the patterns apply to member names after any @samp{/}
1631 delimiter. The use of globbing patterns is discussed in detail in
1634 You can extract a file to standard output by combining the above options
1635 with the @option{--to-stdout} (@option{-O}) option (@pxref{Writing to Standard
1638 If you give the @option{--verbose} option, then @option{--extract}
1639 will print the names of the archive members as it extracts them.
1642 @subsection Extracting Files that are Directories
1644 Extracting directories which are members of an archive is similar to
1645 extracting other files. The main difference to be aware of is that if
1646 the extracted directory has the same name as any directory already in
1647 the working directory, then files in the extracted directory will be
1648 placed into the directory of the same name. Likewise, if there are
1649 files in the pre-existing directory with the same names as the members
1650 which you extract, the files from the extracted archive will replace
1651 the files already in the working directory (and possible
1652 subdirectories). This will happen regardless of whether or not the
1653 files in the working directory were more recent than those extracted
1654 (there exist, however, special options that alter this behavior
1657 However, if a file was stored with a directory name as part of its file
1658 name, and that directory does not exist under the working directory when
1659 the file is extracted, @command{tar} will create the directory.
1661 We can demonstrate how to use @option{--extract} to extract a directory
1662 file with an example. Change to the @file{practice} directory if you
1663 weren't there, and remove the files @file{folk} and @file{jazz}. Then,
1664 go back to the parent directory and extract the archive
1665 @file{music.tar}. You may either extract the entire archive, or you may
1666 extract only the files you just deleted. To extract the entire archive,
1667 don't give any file names as arguments after the archive name
1668 @file{music.tar}. To extract only the files you deleted, use the
1672 $ @kbd{tar -xvf music.tar practice/folk practice/jazz}
1678 If you were to specify two @option{--verbose} (@option{-v}) options, @command{tar}
1679 would have displayed more detail about the extracted files, as shown
1680 in the example below:
1683 $ @kbd{tar -xvvf music.tar practice/folk practice/jazz}
1684 -rw-r--r-- me user 28 1996-10-18 16:31 practice/jazz
1685 -rw-r--r-- me user 20 1996-09-23 16:44 practice/folk
1689 Because you created the directory with @file{practice} as part of the
1690 file names of each of the files by archiving the @file{practice}
1691 directory as @file{practice}, you must give @file{practice} as part
1692 of the file names when you extract those files from the archive.
1694 @node extracting untrusted archives
1695 @subsection Extracting Archives from Untrusted Sources
1697 Extracting files from archives can overwrite files that already exist.
1698 If you receive an archive from an untrusted source, you should make a
1699 new directory and extract into that directory, so that you don't have
1700 to worry about the extraction overwriting one of your existing files.
1701 For example, if @file{untrusted.tar} came from somewhere else on the
1702 Internet, and you don't necessarily trust its contents, you can
1703 extract it as follows:
1706 $ @kbd{mkdir newdir}
1708 $ @kbd{tar -xvf ../untrusted.tar}
1711 It is also a good practice to examine contents of the archive
1712 before extracting it, using @option{--list} (@option{-t}) option, possibly combined
1713 with @option{--verbose} (@option{-v}).
1715 @node failing commands
1716 @subsection Commands That Will Fail
1718 Here are some sample commands you might try which will not work, and why
1721 If you try to use this command,
1724 $ @kbd{tar -xvf music.tar folk jazz}
1728 you will get the following response:
1731 tar: folk: Not found in archive
1732 tar: jazz: Not found in archive
1736 This is because these files were not originally @emph{in} the parent
1737 directory @file{..}, where the archive is located; they were in the
1738 @file{practice} directory, and their file names reflect this:
1741 $ @kbd{tar -tvf music.tar}
1747 @FIXME{make sure the above works when going through the examples in
1751 Likewise, if you try to use this command,
1754 $ @kbd{tar -tvf music.tar folk jazz}
1758 you would get a similar response. Members with those names are not in the
1759 archive. You must use the correct member names, or wildcards, in order
1760 to extract the files from the archive.
1762 If you have forgotten the correct names of the files in the archive,
1763 use @w{@kbd{tar --list --verbose}} to list them correctly.
1765 @FIXME{more examples, here? hag thinks it's a good idea.}
1768 @section Going Further Ahead in this Manual
1771 @FIXME{need to write up a node here about the things that are going to
1772 be in the rest of the manual.}
1774 @node tar invocation
1775 @chapter Invoking @GNUTAR{}
1777 This chapter is about how one invokes the @GNUTAR{}
1778 command, from the command synopsis (@pxref{Synopsis}). There are
1779 numerous options, and many styles for writing them. One mandatory
1780 option specifies the operation @command{tar} should perform
1781 (@pxref{Operation Summary}), other options are meant to detail how
1782 this operation should be performed (@pxref{Option Summary}).
1783 Non-option arguments are not always interpreted the same way,
1784 depending on what the operation is.
1786 You will find in this chapter everything about option styles and rules for
1787 writing them (@pxref{Styles}). On the other hand, operations and options
1788 are fully described elsewhere, in other chapters. Here, you will find
1789 only synthetic descriptions for operations and options, together with
1790 pointers to other parts of the @command{tar} manual.
1792 Some options are so special they are fully described right in this
1793 chapter. They have the effect of inhibiting the normal operation of
1794 @command{tar} or else, they globally alter the amount of feedback the user
1795 receives about what is going on. These are the @option{--help} and
1796 @option{--version} (@pxref{help}), @option{--verbose} (@pxref{verbose})
1797 and @option{--interactive} options (@pxref{interactive}).
1801 * using tar options::
1813 @section General Synopsis of @command{tar}
1815 The @GNUTAR{} program is invoked as either one of:
1818 @kbd{tar @var{option}@dots{} [@var{name}]@dots{}}
1819 @kbd{tar @var{letter}@dots{} [@var{argument}]@dots{} [@var{option}]@dots{} [@var{name}]@dots{}}
1822 The second form is for when old options are being used.
1824 You can use @command{tar} to store files in an archive, to extract them from
1825 an archive, and to do other types of archive manipulation. The primary
1826 argument to @command{tar}, which is called the @dfn{operation}, specifies
1827 which action to take. The other arguments to @command{tar} are either
1828 @dfn{options}, which change the way @command{tar} performs an operation,
1829 or file names or archive members, which specify the files or members
1830 @command{tar} is to act on.
1832 You can actually type in arguments in any order, even if in this manual
1833 the options always precede the other arguments, to make examples easier
1834 to understand. Further, the option stating the main operation mode
1835 (the @command{tar} main command) is usually given first.
1837 Each @var{name} in the synopsis above is interpreted as an archive member
1838 name when the main command is one of @option{--compare}
1839 (@option{--diff}, @option{-d}), @option{--delete}, @option{--extract}
1840 (@option{--get}, @option{-x}), @option{--list} (@option{-t}) or
1841 @option{--update} (@option{-u}). When naming archive members, you
1842 must give the exact name of the member in the archive, as it is
1843 printed by @option{--list}. For @option{--append} (@option{-r}) and
1844 @option{--create} (@option{-c}), these @var{name} arguments specify
1845 the names of either files or directory hierarchies to place in the archive.
1846 These files or hierarchies should already exist in the file system,
1847 prior to the execution of the @command{tar} command.
1849 @command{tar} interprets relative file names as being relative to the
1850 working directory. @command{tar} will make all file names relative
1851 (by removing leading slashes when archiving or restoring files),
1852 unless you specify otherwise (using the @option{--absolute-names}
1853 option). @xref{absolute}, for more information about
1854 @option{--absolute-names}.
1856 If you give the name of a directory as either a file name or a member
1857 name, then @command{tar} acts recursively on all the files and directories
1858 beneath that directory. For example, the name @file{/} identifies all
1859 the files in the file system to @command{tar}.
1861 The distinction between file names and archive member names is especially
1862 important when shell globbing is used, and sometimes a source of confusion
1863 for newcomers. @xref{wildcards}, for more information about globbing.
1864 The problem is that shells may only glob using existing files in the
1865 file system. Only @command{tar} itself may glob on archive members, so when
1866 needed, you must ensure that wildcard characters reach @command{tar} without
1867 being interpreted by the shell first. Using a backslash before @samp{*}
1868 or @samp{?}, or putting the whole argument between quotes, is usually
1869 sufficient for this.
1871 Even if @var{name}s are often specified on the command line, they
1872 can also be read from a text file in the file system, using the
1873 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}}) option.
1875 If you don't use any file name arguments, @option{--append} (@option{-r}),
1876 @option{--delete} and @option{--concatenate} (@option{--catenate},
1877 @option{-A}) will do nothing, while @option{--create} (@option{-c})
1878 will usually yield a diagnostic and inhibit @command{tar} execution.
1879 The other operations of @command{tar} (@option{--list},
1880 @option{--extract}, @option{--compare}, and @option{--update})
1881 will act on the entire contents of the archive.
1884 @cindex return status
1885 Besides successful exits, @GNUTAR{} may fail for
1886 many reasons. Some reasons correspond to bad usage, that is, when the
1887 @command{tar} command line is improperly written. Errors may be
1888 encountered later, while processing the archive or the files. Some
1889 errors are recoverable, in which case the failure is delayed until
1890 @command{tar} has completed all its work. Some errors are such that
1891 it would be not meaningful, or at least risky, to continue processing:
1892 @command{tar} then aborts processing immediately. All abnormal exits,
1893 whether immediate or delayed, should always be clearly diagnosed on
1894 @code{stderr}, after a line stating the nature of the error.
1896 Possible exit codes of @GNUTAR{} are summarized in the following
1901 @samp{Successful termination}.
1904 @samp{Some files differ}. If tar was invoked with @option{--compare}
1905 (@option{--diff}, @option{-d}) command line option, this means that
1906 some files in the archive differ from their disk counterparts
1907 (@pxref{compare}). If tar was given @option{--create},
1908 @option{--append} or @option{--update} option, this exit code means
1909 that some files were changed while being archived and so the resulting
1910 archive does not contain the exact copy of the file set.
1913 @samp{Fatal error}. This means that some fatal, unrecoverable error
1917 If @command{tar} has invoked a subprocess and that subprocess exited with a
1918 nonzero exit code, @command{tar} exits with that code as well.
1919 This can happen, for example, if @command{tar} was given some
1920 compression option (@pxref{gzip}) and the external compressor program
1921 failed. Another example is @command{rmt} failure during backup to the
1922 remote device (@pxref{Remote Tape Server}).
1924 @node using tar options
1925 @section Using @command{tar} Options
1927 @GNUTAR{} has a total of eight operating modes which
1928 allow you to perform a variety of tasks. You are required to choose
1929 one operating mode each time you employ the @command{tar} program by
1930 specifying one, and only one operation as an argument to the
1931 @command{tar} command (the corresponding options may be found
1932 at @ref{frequent operations} and @ref{Operations}). Depending on
1933 circumstances, you may also wish to customize how the chosen operating
1934 mode behaves. For example, you may wish to change the way the output
1935 looks, or the format of the files that you wish to archive may require
1936 you to do something special in order to make the archive look right.
1938 You can customize and control @command{tar}'s performance by running
1939 @command{tar} with one or more options (such as @option{--verbose}
1940 (@option{-v}), which we used in the tutorial). As we said in the
1941 tutorial, @dfn{options} are arguments to @command{tar} which are (as
1942 their name suggests) optional. Depending on the operating mode, you
1943 may specify one or more options. Different options will have different
1944 effects, but in general they all change details of the operation, such
1945 as archive format, archive name, or level of user interaction. Some
1946 options make sense with all operating modes, while others are
1947 meaningful only with particular modes. You will likely use some
1948 options frequently, while you will only use others infrequently, or
1949 not at all. (A full list of options is available in @pxref{All Options}.)
1951 @vrindex TAR_OPTIONS, environment variable
1952 @anchor{TAR_OPTIONS}
1953 The @env{TAR_OPTIONS} environment variable specifies default options to
1954 be placed in front of any explicit options. For example, if
1955 @code{TAR_OPTIONS} is @samp{-v --unlink-first}, @command{tar} behaves as
1956 if the two options @option{-v} and @option{--unlink-first} had been
1957 specified before any explicit options. Option specifications are
1958 separated by whitespace. A backslash escapes the next character, so it
1959 can be used to specify an option containing whitespace or a backslash.
1961 Note that @command{tar} options are case sensitive. For example, the
1962 options @option{-T} and @option{-t} are different; the first requires an
1963 argument for stating the name of a file providing a list of @var{name}s,
1964 while the second does not require an argument and is another way to
1965 write @option{--list} (@option{-t}).
1967 In addition to the eight operations, there are many options to
1968 @command{tar}, and three different styles for writing both: long (mnemonic)
1969 form, short form, and old style. These styles are discussed below.
1970 Both the options and the operations can be written in any of these three
1973 @FIXME{menu at end of this node. need to think of an actual outline
1974 for this chapter; probably do that after stuff from chapter 4 is
1978 @section The Three Option Styles
1980 There are three styles for writing operations and options to the command
1981 line invoking @command{tar}. The different styles were developed at
1982 different times during the history of @command{tar}. These styles will be
1983 presented below, from the most recent to the oldest.
1985 Some options must take an argument@footnote{For example, @option{--file}
1986 (@option{-f}) takes the name of an archive file as an argument. If
1987 you do not supply an archive file name, @command{tar} will use a
1988 default, but this can be confusing; thus, we recommend that you always
1989 supply a specific archive file name.}. Where you @emph{place} the
1990 arguments generally depends on which style of options you choose. We
1991 will detail specific information relevant to each option style in the
1992 sections on the different option styles, below. The differences are
1993 subtle, yet can often be very important; incorrect option placement
1994 can cause you to overwrite a number of important files. We urge you
1995 to note these differences, and only use the option style(s) which
1996 makes the most sense to you until you feel comfortable with the others.
1998 Some options @emph{may} take an argument. Such options may have at
1999 most long and short forms, they do not have old style equivalent. The
2000 rules for specifying an argument for such options are stricter than
2001 those for specifying mandatory arguments. Please, pay special
2005 * Long Options:: Long Option Style
2006 * Short Options:: Short Option Style
2007 * Old Options:: Old Option Style
2008 * Mixing:: Mixing Option Styles
2012 @subsection Long Option Style
2014 @cindex long options
2015 @cindex options, long style
2016 @cindex options, GNU style
2017 @cindex options, mnemonic names
2018 Each option has at least one @dfn{long} (or @dfn{mnemonic}) name starting with two
2019 dashes in a row, e.g., @option{--list}. The long names are more clear than
2020 their corresponding short or old names. It sometimes happens that a
2021 single long option has many different names which are
2022 synonymous, such as @option{--compare} and @option{--diff}. In addition,
2023 long option names can be given unique abbreviations. For example,
2024 @option{--cre} can be used in place of @option{--create} because there is no
2025 other long option which begins with @samp{cre}. (One way to find
2026 this out is by trying it and seeing what happens; if a particular
2027 abbreviation could represent more than one option, @command{tar} will tell
2028 you that that abbreviation is ambiguous and you'll know that that
2029 abbreviation won't work. You may also choose to run @samp{tar --help}
2030 to see a list of options. Be aware that if you run @command{tar} with a
2031 unique abbreviation for the long name of an option you didn't want to
2032 use, you are stuck; @command{tar} will perform the command as ordered.)
2034 Long options are meant to be obvious and easy to remember, and their
2035 meanings are generally easier to discern than those of their
2036 corresponding short options (see below). For example:
2039 $ @kbd{tar --create --verbose --blocking-factor=20 --file=/dev/rmt0}
2043 gives a fairly good set of hints about what the command does, even
2044 for those not fully acquainted with @command{tar}.
2046 @cindex arguments to long options
2047 @cindex long options with mandatory arguments
2048 Long options which require arguments take those arguments
2049 immediately following the option name. There are two ways of
2050 specifying a mandatory argument. It can be separated from the
2051 option name either by an equal sign, or by any amount of
2052 white space characters. For example, the @option{--file} option (which
2053 tells the name of the @command{tar} archive) is given a file such as
2054 @file{archive.tar} as argument by using any of the following notations:
2055 @option{--file=archive.tar} or @option{--file archive.tar}.
2057 @cindex optional arguments to long options
2058 @cindex long options with optional arguments
2059 In contrast, optional arguments must always be introduced using
2060 an equal sign. For example, the @option{--backup} option takes
2061 an optional argument specifying backup type. It must be used
2062 as @option{--backup=@var{backup-type}}.
2065 @subsection Short Option Style
2067 @cindex short options
2068 @cindex options, short style
2069 @cindex options, traditional
2070 Most options also have a @dfn{short option} name. Short options start with
2071 a single dash, and are followed by a single character, e.g., @option{-t}
2072 (which is equivalent to @option{--list}). The forms are absolutely
2073 identical in function; they are interchangeable.
2075 The short option names are faster to type than long option names.
2077 @cindex arguments to short options
2078 @cindex short options with mandatory arguments
2079 Short options which require arguments take their arguments immediately
2080 following the option, usually separated by white space. It is also
2081 possible to stick the argument right after the short option name, using
2082 no intervening space. For example, you might write @w{@option{-f
2083 archive.tar}} or @option{-farchive.tar} instead of using
2084 @option{--file=archive.tar}. Both @option{--file=@var{archive-name}} and
2085 @w{@option{-f @var{archive-name}}} denote the option which indicates a
2086 specific archive, here named @file{archive.tar}.
2088 @cindex optional arguments to short options
2089 @cindex short options with optional arguments
2090 Short options which take optional arguments take their arguments
2091 immediately following the option letter, @emph{without any intervening
2092 white space characters}.
2094 Short options' letters may be clumped together, but you are not
2095 required to do this (as compared to old options; see below). When
2096 short options are clumped as a set, use one (single) dash for them
2097 all, e.g., @w{@samp{@command{tar} -cvf}}. Only the last option in
2098 such a set is allowed to have an argument@footnote{Clustering many
2099 options, the last of which has an argument, is a rather opaque way to
2100 write options. Some wonder if @acronym{GNU} @code{getopt} should not
2101 even be made helpful enough for considering such usages as invalid.}.
2103 When the options are separated, the argument for each option which requires
2104 an argument directly follows that option, as is usual for Unix programs.
2108 $ @kbd{tar -c -v -b 20 -f /dev/rmt0}
2111 If you reorder short options' locations, be sure to move any arguments
2112 that belong to them. If you do not move the arguments properly, you may
2113 end up overwriting files.
2116 @subsection Old Option Style
2117 @cindex options, old style
2118 @cindex old option style
2120 Like short options, @dfn{old options} are single letters. However, old options
2121 must be written together as a single clumped set, without spaces separating
2122 them or dashes preceding them@footnote{Beware that if you precede options
2123 with a dash, you are announcing the short option style instead of the
2124 old option style; short options are decoded differently.}. This set
2125 of letters must be the first to appear on the command line, after the
2126 @command{tar} program name and some white space; old options cannot appear
2127 anywhere else. The letter of an old option is exactly the same letter as
2128 the corresponding short option. For example, the old option @samp{t} is
2129 the same as the short option @option{-t}, and consequently, the same as the
2130 long option @option{--list}. So for example, the command @w{@samp{tar
2131 cv}} specifies the option @option{-v} in addition to the operation @option{-c}.
2133 @cindex arguments to old options
2134 @cindex old options with mandatory arguments
2135 When options that need arguments are given together with the command,
2136 all the associated arguments follow, in the same order as the options.
2137 Thus, the example given previously could also be written in the old
2141 $ @kbd{tar cvbf 20 /dev/rmt0}
2145 Here, @samp{20} is the argument of @option{-b} and @samp{/dev/rmt0} is
2146 the argument of @option{-f}.
2148 On the other hand, this old style syntax makes it difficult to match
2149 option letters with their corresponding arguments, and is often
2150 confusing. In the command @w{@samp{tar cvbf 20 /dev/rmt0}}, for example,
2151 @samp{20} is the argument for @option{-b}, @samp{/dev/rmt0} is the
2152 argument for @option{-f}, and @option{-v} does not have a corresponding
2153 argument. Even using short options like in @w{@samp{tar -c -v -b 20 -f
2154 /dev/rmt0}} is clearer, putting all arguments next to the option they
2157 If you want to reorder the letters in the old option argument, be
2158 sure to reorder any corresponding argument appropriately.
2160 This old way of writing @command{tar} options can surprise even experienced
2161 users. For example, the two commands:
2164 @kbd{tar cfz archive.tar.gz file}
2165 @kbd{tar -cfz archive.tar.gz file}
2169 are quite different. The first example uses @file{archive.tar.gz} as
2170 the value for option @samp{f} and recognizes the option @samp{z}. The
2171 second example, however, uses @file{z} as the value for option
2172 @samp{f} --- probably not what was intended.
2174 Old options are kept for compatibility with old versions of @command{tar}.
2176 This second example could be corrected in many ways, among which the
2177 following are equivalent:
2180 @kbd{tar -czf archive.tar.gz file}
2181 @kbd{tar -cf archive.tar.gz -z file}
2182 @kbd{tar cf archive.tar.gz -z file}
2185 @cindex option syntax, traditional
2186 As far as we know, all @command{tar} programs, @acronym{GNU} and
2187 non-@acronym{GNU}, support old options. @GNUTAR{}
2188 supports them not only for historical reasons, but also because many
2189 people are used to them. For compatibility with Unix @command{tar},
2190 the first argument is always treated as containing command and option
2191 letters even if it doesn't start with @samp{-}. Thus, @samp{tar c} is
2192 equivalent to @w{@samp{tar -c}:} both of them specify the
2193 @option{--create} (@option{-c}) command to create an archive.
2196 @subsection Mixing Option Styles
2198 @cindex options, mixing different styles
2199 All three styles may be intermixed in a single @command{tar} command,
2200 so long as the rules for each style are fully
2201 respected@footnote{Before @GNUTAR{} version 1.11.6,
2202 a bug prevented intermixing old style options with long options in
2203 some cases.}. Old style options and either of the modern styles of
2204 options may be mixed within a single @command{tar} command. However,
2205 old style options must be introduced as the first arguments only,
2206 following the rule for old options (old options must appear directly
2207 after the @command{tar} command and some white space). Modern options
2208 may be given only after all arguments to the old options have been
2209 collected. If this rule is not respected, a modern option might be
2210 falsely interpreted as the value of the argument to one of the old
2213 For example, all the following commands are wholly equivalent, and
2214 illustrate the many combinations and orderings of option styles.
2217 @kbd{tar --create --file=archive.tar}
2218 @kbd{tar --create -f archive.tar}
2219 @kbd{tar --create -farchive.tar}
2220 @kbd{tar --file=archive.tar --create}
2221 @kbd{tar --file=archive.tar -c}
2222 @kbd{tar -c --file=archive.tar}
2223 @kbd{tar -c -f archive.tar}
2224 @kbd{tar -c -farchive.tar}
2225 @kbd{tar -cf archive.tar}
2226 @kbd{tar -cfarchive.tar}
2227 @kbd{tar -f archive.tar --create}
2228 @kbd{tar -f archive.tar -c}
2229 @kbd{tar -farchive.tar --create}
2230 @kbd{tar -farchive.tar -c}
2231 @kbd{tar c --file=archive.tar}
2232 @kbd{tar c -f archive.tar}
2233 @kbd{tar c -farchive.tar}
2234 @kbd{tar cf archive.tar}
2235 @kbd{tar f archive.tar --create}
2236 @kbd{tar f archive.tar -c}
2237 @kbd{tar fc archive.tar}
2240 On the other hand, the following commands are @emph{not} equivalent to
2244 @kbd{tar -f -c archive.tar}
2245 @kbd{tar -fc archive.tar}
2246 @kbd{tar -fcarchive.tar}
2247 @kbd{tar -farchive.tarc}
2248 @kbd{tar cfarchive.tar}
2252 These last examples mean something completely different from what the
2253 user intended (judging based on the example in the previous set which
2254 uses long options, whose intent is therefore very clear). The first
2255 four specify that the @command{tar} archive would be a file named
2256 @option{-c}, @samp{c}, @samp{carchive.tar} or @samp{archive.tarc},
2257 respectively. The first two examples also specify a single non-option,
2258 @var{name} argument having the value @samp{archive.tar}. The last
2259 example contains only old style option letters (repeating option
2260 @samp{c} twice), not all of which are meaningful (eg., @samp{.},
2261 @samp{h}, or @samp{i}), with no argument value. @FIXME{not sure i liked
2262 the first sentence of this paragraph..}
2265 @section All @command{tar} Options
2267 The coming manual sections contain an alphabetical listing of all
2268 @command{tar} operations and options, with brief descriptions and
2269 cross-references to more in-depth explanations in the body of the manual.
2270 They also contain an alphabetically arranged table of the short option
2271 forms with their corresponding long option. You can use this table as
2272 a reference for deciphering @command{tar} commands in scripts.
2275 * Operation Summary::
2277 * Short Option Summary::
2280 @node Operation Summary
2281 @subsection Operations
2289 Appends files to the end of the archive. @xref{append}.
2291 @opsummary{catenate}
2295 Same as @option{--concatenate}. @xref{concatenate}.
2301 Compares archive members with their counterparts in the file
2302 system, and reports differences in file size, mode, owner,
2303 modification date and contents. @xref{compare}.
2305 @opsummary{concatenate}
2309 Appends other @command{tar} archives to the end of the archive.
2316 Creates a new @command{tar} archive. @xref{create}.
2321 Deletes members from the archive. Don't try this on an archive on a
2322 tape! @xref{delete}.
2328 Same @option{--compare}. @xref{compare}.
2334 Extracts members from the archive into the file system. @xref{extract}.
2340 Same as @option{--extract}. @xref{extract}.
2346 Lists the members in an archive. @xref{list}.
2352 Adds files to the end of the archive, but only if they are newer than
2353 their counterparts already in the archive, or if they do not already
2354 exist in the archive. @xref{update}.
2358 @node Option Summary
2359 @subsection @command{tar} Options
2363 @opsummary{absolute-names}
2364 @item --absolute-names
2367 Normally when creating an archive, @command{tar} strips an initial
2368 @samp{/} from member names. This option disables that behavior.
2371 @opsummary{after-date}
2374 (See @option{--newer}, @pxref{after})
2376 @opsummary{anchored}
2378 A pattern must match an initial subsequence of the name's components.
2379 @xref{controlling pattern-matching}.
2381 @opsummary{atime-preserve}
2382 @item --atime-preserve
2383 @itemx --atime-preserve=replace
2384 @itemx --atime-preserve=system
2386 Attempt to preserve the access time of files when reading them. This
2387 option currently is effective only on files that you own, unless you
2388 have superuser privileges.
2390 @option{--atime-preserve=replace} remembers the access time of a file
2391 before reading it, and then restores the access time afterwards. This
2392 may cause problems if other programs are reading the file at the same
2393 time, as the times of their accesses will be lost. On most platforms
2394 restoring the access time also requires @command{tar} to restore the
2395 data modification time too, so this option may also cause problems if
2396 other programs are writing the file at the same time (@command{tar} attempts
2397 to detect this situation, but cannot do so reliably due to race
2398 conditions). Worse, on most platforms restoring the access time also
2399 updates the status change time, which means that this option is
2400 incompatible with incremental backups.
2402 @option{--atime-preserve=system} avoids changing time stamps on files,
2403 without interfering with time stamp updates
2404 caused by other programs, so it works better with incremental backups.
2405 However, it requires a special @code{O_NOATIME} option from the
2406 underlying operating and file system implementation, and it also requires
2407 that searching directories does not update their access times. As of
2408 this writing (November 2005) this works only with Linux, and only with
2409 Linux kernels 2.6.8 and later. Worse, there is currently no reliable
2410 way to know whether this feature actually works. Sometimes
2411 @command{tar} knows that it does not work, and if you use
2412 @option{--atime-preserve=system} then @command{tar} complains and
2413 exits right away. But other times @command{tar} might think that the
2414 option works when it actually does not.
2416 Currently @option{--atime-preserve} with no operand defaults to
2417 @option{--atime-preserve=replace}, but this may change in the future
2418 as support for @option{--atime-preserve=system} improves.
2420 If your operating or file system does not support
2421 @option{--atime-preserve=@-system}, you might be able to preserve access
2422 times reliably by using the @command{mount} command. For example,
2423 you can mount the file system read-only, or access the file system via
2424 a read-only loopback mount, or use the @samp{noatime} mount option
2425 available on some systems. However, mounting typically requires
2426 superuser privileges and can be a pain to manage.
2428 @opsummary{auto-compress}
2429 @item --auto-compress
2432 During a @option{--create} operation, enables automatic compressed
2433 format recognition based on the archive suffix. The effect of this
2434 option is cancelled by @option{--no-auto-compress}. @xref{gzip}.
2437 @item --backup=@var{backup-type}
2439 Rather than deleting files from the file system, @command{tar} will
2440 back them up using simple or numbered backups, depending upon
2441 @var{backup-type}. @xref{backup}.
2443 @opsummary{block-number}
2444 @item --block-number
2447 With this option present, @command{tar} prints error messages for read errors
2448 with the block number in the archive file. @xref{block-number}.
2450 @opsummary{blocking-factor}
2451 @item --blocking-factor=@var{blocking}
2452 @itemx -b @var{blocking}
2454 Sets the blocking factor @command{tar} uses to @var{blocking} x 512 bytes per
2455 record. @xref{Blocking Factor}.
2461 This option tells @command{tar} to read or write archives through
2462 @code{bzip2}. @xref{gzip}.
2464 @opsummary{check-device}
2465 @item --check-device
2466 Check device numbers when creating a list of modified files for
2467 incremental archiving. This is the default. @xref{device numbers},
2468 for a detailed description.
2470 @opsummary{checkpoint}
2471 @item --checkpoint[=@var{number}]
2473 This option directs @command{tar} to print periodic checkpoint
2474 messages as it reads through the archive. It is intended for when you
2475 want a visual indication that @command{tar} is still running, but
2476 don't want to see @option{--verbose} output. You can also instruct
2477 @command{tar} to execute a list of actions on each checkpoint, see
2478 @option{--checkpoint-action} below. For a detailed description, see
2481 @opsummary{checkpoint-action}
2482 @item --checkpoint-action=@var{action}
2483 Instruct @command{tar} to execute an action upon hitting a
2484 breakpoint. Here we give only a brief outline. @xref{checkpoints},
2485 for a complete description.
2487 The @var{action} argument can be one of the following:
2491 Produce an audible bell on the console.
2495 Print a single dot on the standard listing stream.
2498 Display a textual message on the standard error, with the status and
2499 number of the checkpoint. This is the default.
2501 @item echo=@var{string}
2502 Display @var{string} on the standard error. Before output, the string
2503 is subject to meta-character expansion.
2505 @item exec=@var{command}
2506 Execute the given @var{command}.
2508 @item sleep=@var{time}
2509 Wait for @var{time} seconds.
2511 @item ttyout=@var{string}
2512 Output @var{string} on the current console (@file{/dev/tty}).
2515 Several @option{--checkpoint-action} options can be specified. The
2516 supplied actions will be executed in order of their appearance in the
2519 Using @option{--checkpoint-action} without @option{--checkpoint}
2520 assumes default checkpoint frequency of one checkpoint per 10 records.
2522 @opsummary{check-links}
2525 If this option was given, @command{tar} will check the number of links
2526 dumped for each processed file. If this number does not match the
2527 total number of hard links for the file, a warning message will be
2528 output @footnote{Earlier versions of @GNUTAR{} understood @option{-l} as a
2529 synonym for @option{--one-file-system}. The current semantics, which
2530 complies to UNIX98, was introduced with version
2531 1.15.91. @xref{Changes}, for more information.}.
2535 @opsummary{compress}
2536 @opsummary{uncompress}
2541 @command{tar} will use the @command{compress} program when reading or
2542 writing the archive. This allows you to directly act on archives
2543 while saving space. @xref{gzip}.
2545 @opsummary{confirmation}
2546 @item --confirmation
2548 (See @option{--interactive}.) @xref{interactive}.
2550 @opsummary{delay-directory-restore}
2551 @item --delay-directory-restore
2553 Delay setting modification times and permissions of extracted
2554 directories until the end of extraction. @xref{Directory Modification Times and Permissions}.
2556 @opsummary{dereference}
2560 When creating a @command{tar} archive, @command{tar} will archive the
2561 file that a symbolic link points to, rather than archiving the
2562 symlink. @xref{dereference}.
2564 @opsummary{directory}
2565 @item --directory=@var{dir}
2568 When this option is specified, @command{tar} will change its current directory
2569 to @var{dir} before performing any operations. When this option is used
2570 during archive creation, it is order sensitive. @xref{directory}.
2573 @item --exclude=@var{pattern}
2575 When performing operations, @command{tar} will skip files that match
2576 @var{pattern}. @xref{exclude}.
2578 @opsummary{exclude-backups}
2579 @item --exclude-backups
2580 Exclude backup and lock files. @xref{exclude,, exclude-backups}.
2582 @opsummary{exclude-from}
2583 @item --exclude-from=@var{file}
2584 @itemx -X @var{file}
2586 Similar to @option{--exclude}, except @command{tar} will use the list of
2587 patterns in the file @var{file}. @xref{exclude}.
2589 @opsummary{exclude-caches}
2590 @item --exclude-caches
2592 Exclude from dump any directory containing a valid cache directory
2593 tag file, but still dump the directory node and the tag file itself.
2595 @xref{exclude,, exclude-caches}.
2597 @opsummary{exclude-caches-under}
2598 @item --exclude-caches-under
2600 Exclude from dump any directory containing a valid cache directory
2601 tag file, but still dump the directory node itself.
2605 @opsummary{exclude-caches-all}
2606 @item --exclude-caches-all
2608 Exclude from dump any directory containing a valid cache directory
2609 tag file. @xref{exclude}.
2611 @opsummary{exclude-tag}
2612 @item --exclude-tag=@var{file}
2614 Exclude from dump any directory containing file named @var{file}, but
2615 dump the directory node and @var{file} itself. @xref{exclude,, exclude-tag}.
2617 @opsummary{exclude-tag-under}
2618 @item --exclude-tag-under=@var{file}
2620 Exclude from dump the contents of any directory containing file
2621 named @var{file}, but dump the directory node itself. @xref{exclude,,
2624 @opsummary{exclude-tag-all}
2625 @item --exclude-tag-all=@var{file}
2627 Exclude from dump any directory containing file named @var{file}.
2628 @xref{exclude,,exclude-tag-all}.
2630 @opsummary{exclude-vcs}
2633 Exclude from dump directories and files, that are internal for some
2634 widely used version control systems.
2636 @xref{exclude,,exclude-vcs}.
2639 @item --file=@var{archive}
2640 @itemx -f @var{archive}
2642 @command{tar} will use the file @var{archive} as the @command{tar} archive it
2643 performs operations on, rather than @command{tar}'s compilation dependent
2644 default. @xref{file tutorial}.
2646 @opsummary{files-from}
2647 @item --files-from=@var{file}
2648 @itemx -T @var{file}
2650 @command{tar} will use the contents of @var{file} as a list of archive members
2651 or files to operate on, in addition to those specified on the
2652 command-line. @xref{files}.
2654 @opsummary{force-local}
2657 Forces @command{tar} to interpret the file name given to @option{--file}
2658 as a local file, even if it looks like a remote tape drive name.
2659 @xref{local and remote archives}.
2662 @item --format=@var{format}
2663 @itemx -H @var{format}
2665 Selects output archive format. @var{Format} may be one of the
2670 Creates an archive that is compatible with Unix V7 @command{tar}.
2673 Creates an archive that is compatible with GNU @command{tar} version
2677 Creates archive in GNU tar 1.13 format. Basically it is the same as
2678 @samp{oldgnu} with the only difference in the way it handles long
2682 Creates a @acronym{POSIX.1-1988} compatible archive.
2685 Creates a @acronym{POSIX.1-2001 archive}.
2689 @xref{Formats}, for a detailed discussion of these formats.
2691 @opsummary{full-time}
2693 This option instructs @command{tar} to print file times to their full
2694 resolution. Usually this means 1-second resolution, but that depends
2695 on the underlying file system. The @option{--full-time} option takes
2696 effect only when detailed output (verbosity level 2 or higher) has
2697 been requested using the @option{--verbose} option, e.g., when listing
2698 or extracting archives:
2701 $ @kbd{tar -t -v --full-time -f archive.tar}
2705 or, when creating an archive:
2708 $ @kbd{tar -c -vv --full-time -f archive.tar .}
2711 Notice, thar when creating the archive you need to specify
2712 @option{--verbose} twice to get a detailed output (@pxref{verbose
2716 @item --group=@var{group}
2718 Files added to the @command{tar} archive will have a group @acronym{ID} of @var{group},
2719 rather than the group from the source file. @var{group} is first decoded
2720 as a group symbolic name, but if this interpretation fails, it has to be
2721 a decimal numeric group @acronym{ID}. @xref{override}.
2723 Also see the comments for the @option{--owner=@var{user}} option.
2733 This option tells @command{tar} to read or write archives through
2734 @command{gzip}, allowing @command{tar} to directly operate on several
2735 kinds of compressed archives transparently. @xref{gzip}.
2737 @opsummary{hard-dereference}
2738 @item --hard-dereference
2739 When creating an archive, dereference hard links and store the files
2740 they refer to, instead of creating usual hard link members.
2748 @command{tar} will print out a short message summarizing the operations and
2749 options to @command{tar} and exit. @xref{help}.
2751 @opsummary{ignore-case}
2753 Ignore case when matching member or file names with
2754 patterns. @xref{controlling pattern-matching}.
2756 @opsummary{ignore-command-error}
2757 @item --ignore-command-error
2758 Ignore exit codes of subprocesses. @xref{Writing to an External Program}.
2760 @opsummary{ignore-failed-read}
2761 @item --ignore-failed-read
2763 Do not exit unsuccessfully merely because an unreadable file was encountered.
2766 @opsummary{ignore-zeros}
2767 @item --ignore-zeros
2770 With this option, @command{tar} will ignore zeroed blocks in the
2771 archive, which normally signals EOF. @xref{Reading}.
2773 @opsummary{incremental}
2777 Informs @command{tar} that it is working with an old
2778 @acronym{GNU}-format incremental backup archive. It is intended
2779 primarily for backwards compatibility only. @xref{Incremental Dumps},
2780 for a detailed discussion of incremental archives.
2782 @opsummary{index-file}
2783 @item --index-file=@var{file}
2785 Send verbose output to @var{file} instead of to standard output.
2787 @opsummary{info-script}
2788 @opsummary{new-volume-script}
2789 @item --info-script=@var{script-file}
2790 @itemx --new-volume-script=@var{script-file}
2791 @itemx -F @var{script-file}
2793 When @command{tar} is performing multi-tape backups, @var{script-file} is run
2794 at the end of each tape. If @var{script-file} exits with nonzero status,
2795 @command{tar} fails immediately. @xref{info-script}, for a detailed
2796 discussion of @var{script-file}.
2798 @opsummary{interactive}
2800 @itemx --confirmation
2803 Specifies that @command{tar} should ask the user for confirmation before
2804 performing potentially destructive options, such as overwriting files.
2807 @opsummary{keep-newer-files}
2808 @item --keep-newer-files
2810 Do not replace existing files that are newer than their archive copies
2811 when extracting files from an archive.
2813 @opsummary{keep-old-files}
2814 @item --keep-old-files
2817 Do not overwrite existing files when extracting files from an archive.
2818 @xref{Keep Old Files}.
2821 @item --label=@var{name}
2822 @itemx -V @var{name}
2824 When creating an archive, instructs @command{tar} to write @var{name}
2825 as a name record in the archive. When extracting or listing archives,
2826 @command{tar} will only operate on archives that have a label matching
2827 the pattern specified in @var{name}. @xref{Tape Files}.
2830 @item --level=@var{n}
2831 Force incremental backup of level @var{n}. As of @GNUTAR version
2832 @value{VERSION}, the option @option{--level=0} truncates the snapshot
2833 file, thereby forcing the level 0 dump. Other values of @var{n} are
2834 effectively ignored. @xref{--level=0}, for details and examples.
2836 The use of this option is valid only in conjunction with the
2837 @option{--listed-incremental} option. @xref{Incremental Dumps},
2838 for a detailed description.
2840 @opsummary{listed-incremental}
2841 @item --listed-incremental=@var{snapshot-file}
2842 @itemx -g @var{snapshot-file}
2844 During a @option{--create} operation, specifies that the archive that
2845 @command{tar} creates is a new @acronym{GNU}-format incremental
2846 backup, using @var{snapshot-file} to determine which files to backup.
2847 With other operations, informs @command{tar} that the archive is in
2848 incremental format. @xref{Incremental Dumps}.
2853 This option tells @command{tar} to read or write archives through
2854 @command{lzip}. @xref{gzip}.
2859 This option tells @command{tar} to read or write archives through
2860 @command{lzma}. @xref{gzip}.
2864 This option tells @command{tar} to read or write archives through
2865 @command{lzop}. @xref{gzip}.
2868 @item --mode=@var{permissions}
2870 When adding files to an archive, @command{tar} will use
2871 @var{permissions} for the archive members, rather than the permissions
2872 from the files. @var{permissions} can be specified either as an octal
2873 number or as symbolic permissions, like with
2874 @command{chmod}. @xref{override}.
2877 @item --mtime=@var{date}
2879 When adding files to an archive, @command{tar} will use @var{date} as
2880 the modification time of members when creating archives, instead of
2881 their actual modification times. The value of @var{date} can be
2882 either a textual date representation (@pxref{Date input formats}) or a
2883 name of the existing file, starting with @samp{/} or @samp{.}. In the
2884 latter case, the modification time of that file is used. @xref{override}.
2886 @opsummary{multi-volume}
2887 @item --multi-volume
2890 Informs @command{tar} that it should create or otherwise operate on a
2891 multi-volume @command{tar} archive. @xref{Using Multiple Tapes}.
2893 @opsummary{new-volume-script}
2894 @item --new-volume-script
2896 (see @option{--info-script})
2899 @item --newer=@var{date}
2900 @itemx --after-date=@var{date}
2903 When creating an archive, @command{tar} will only add files that have changed
2904 since @var{date}. If @var{date} begins with @samp{/} or @samp{.}, it
2905 is taken to be the name of a file whose data modification time specifies
2906 the date. @xref{after}.
2908 @opsummary{newer-mtime}
2909 @item --newer-mtime=@var{date}
2911 Like @option{--newer}, but add only files whose
2912 contents have changed (as opposed to just @option{--newer}, which will
2913 also back up files for which any status information has
2914 changed). @xref{after}.
2916 @opsummary{no-anchored}
2918 An exclude pattern can match any subsequence of the name's components.
2919 @xref{controlling pattern-matching}.
2921 @opsummary{no-auto-compress}
2922 @item --no-auto-compress
2924 Disables automatic compressed format recognition based on the archive
2925 suffix. @xref{--auto-compress}. @xref{gzip}.
2927 @opsummary{no-check-device}
2928 @item --no-check-device
2929 Do not check device numbers when creating a list of modified files
2930 for incremental archiving. @xref{device numbers}, for
2931 a detailed description.
2933 @opsummary{no-delay-directory-restore}
2934 @item --no-delay-directory-restore
2936 Modification times and permissions of extracted
2937 directories are set when all files from this directory have been
2938 extracted. This is the default.
2939 @xref{Directory Modification Times and Permissions}.
2941 @opsummary{no-ignore-case}
2942 @item --no-ignore-case
2943 Use case-sensitive matching.
2944 @xref{controlling pattern-matching}.
2946 @opsummary{no-ignore-command-error}
2947 @item --no-ignore-command-error
2948 Print warnings about subprocesses that terminated with a nonzero exit
2949 code. @xref{Writing to an External Program}.
2954 If the @option{--null} option was given previously, this option
2955 cancels its effect, so that any following @option{--files-from}
2956 options will expect their file lists to be newline-terminated.
2958 @opsummary{no-overwrite-dir}
2959 @item --no-overwrite-dir
2961 Preserve metadata of existing directories when extracting files
2962 from an archive. @xref{Overwrite Old Files}.
2964 @opsummary{no-quote-chars}
2965 @item --no-quote-chars=@var{string}
2966 Remove characters listed in @var{string} from the list of quoted
2967 characters set by the previous @option{--quote-chars} option
2968 (@pxref{quoting styles}).
2970 @opsummary{no-recursion}
2971 @item --no-recursion
2973 With this option, @command{tar} will not recurse into directories.
2976 @opsummary{no-same-owner}
2977 @item --no-same-owner
2980 When extracting an archive, do not attempt to preserve the owner
2981 specified in the @command{tar} archive. This the default behavior
2984 @opsummary{no-same-permissions}
2985 @item --no-same-permissions
2987 When extracting an archive, subtract the user's umask from files from
2988 the permissions specified in the archive. This is the default behavior
2994 The archive media does not support seeks to arbitrary
2995 locations. Usually @command{tar} determines automatically whether
2996 the archive can be seeked or not. Use this option to disable this
2999 @opsummary{no-unquote}
3001 Treat all input file or member names literally, do not interpret
3002 escape sequences. @xref{input name quoting}.
3004 @opsummary{no-wildcards}
3005 @item --no-wildcards
3006 Do not use wildcards.
3007 @xref{controlling pattern-matching}.
3009 @opsummary{no-wildcards-match-slash}
3010 @item --no-wildcards-match-slash
3011 Wildcards do not match @samp{/}.
3012 @xref{controlling pattern-matching}.
3017 When @command{tar} is using the @option{--files-from} option, this option
3018 instructs @command{tar} to expect file names terminated with @acronym{NUL}, so
3019 @command{tar} can correctly work with file names that contain newlines.
3022 @opsummary{numeric-owner}
3023 @item --numeric-owner
3025 This option will notify @command{tar} that it should use numeric user
3026 and group IDs when creating a @command{tar} file, rather than names.
3030 The function of this option depends on the action @command{tar} is
3031 performing. When extracting files, @option{-o} is a synonym for
3032 @option{--no-same-owner}, i.e., it prevents @command{tar} from
3033 restoring ownership of files being extracted.
3035 When creating an archive, it is a synonym for
3036 @option{--old-archive}. This behavior is for compatibility
3037 with previous versions of @GNUTAR{}, and will be
3038 removed in future releases.
3040 @xref{Changes}, for more information.
3042 @opsummary{occurrence}
3043 @item --occurrence[=@var{number}]
3045 This option can be used in conjunction with one of the subcommands
3046 @option{--delete}, @option{--diff}, @option{--extract} or
3047 @option{--list} when a list of files is given either on the command
3048 line or via @option{-T} option.
3050 This option instructs @command{tar} to process only the @var{number}th
3051 occurrence of each named file. @var{Number} defaults to 1, so
3054 tar -x -f archive.tar --occurrence filename
3058 will extract the first occurrence of the member @file{filename} from @file{archive.tar}
3059 and will terminate without scanning to the end of the archive.
3061 @opsummary{old-archive}
3063 Synonym for @option{--format=v7}.
3065 @opsummary{one-file-system}
3066 @item --one-file-system
3067 Used when creating an archive. Prevents @command{tar} from recursing into
3068 directories that are on different file systems from the current
3071 @opsummary{overwrite}
3074 Overwrite existing files and directory metadata when extracting files
3075 from an archive. @xref{Overwrite Old Files}.
3077 @opsummary{overwrite-dir}
3078 @item --overwrite-dir
3080 Overwrite the metadata of existing directories when extracting files
3081 from an archive. @xref{Overwrite Old Files}.
3084 @item --owner=@var{user}
3086 Specifies that @command{tar} should use @var{user} as the owner of members
3087 when creating archives, instead of the user associated with the source
3088 file. @var{user} is first decoded as a user symbolic name, but if
3089 this interpretation fails, it has to be a decimal numeric user @acronym{ID}.
3092 This option does not affect extraction from archives.
3094 @opsummary{pax-option}
3095 @item --pax-option=@var{keyword-list}
3096 This option enables creation of the archive in @acronym{POSIX.1-2001}
3097 format (@pxref{posix}) and modifies the way @command{tar} handles the
3098 extended header keywords. @var{Keyword-list} is a comma-separated
3099 list of keyword options. @xref{PAX keywords}, for a detailed
3102 @opsummary{portability}
3104 @itemx --old-archive
3105 Synonym for @option{--format=v7}.
3109 Same as @option{--format=posix}.
3111 @opsummary{preserve}
3114 Synonymous with specifying both @option{--preserve-permissions} and
3115 @option{--same-order}. @xref{Setting Access Permissions}.
3117 @opsummary{preserve-order}
3118 @item --preserve-order
3120 (See @option{--same-order}; @pxref{Reading}.)
3122 @opsummary{preserve-permissions}
3123 @opsummary{same-permissions}
3124 @item --preserve-permissions
3125 @itemx --same-permissions
3128 When @command{tar} is extracting an archive, it normally subtracts the
3129 users' umask from the permissions specified in the archive and uses
3130 that number as the permissions to create the destination file.
3131 Specifying this option instructs @command{tar} that it should use the
3132 permissions directly from the archive. @xref{Setting Access Permissions}.
3134 @opsummary{quote-chars}
3135 @item --quote-chars=@var{string}
3136 Always quote characters from @var{string}, even if the selected
3137 quoting style would not quote them (@pxref{quoting styles}).
3139 @opsummary{quoting-style}
3140 @item --quoting-style=@var{style}
3141 Set quoting style to use when printing member and file names
3142 (@pxref{quoting styles}). Valid @var{style} values are:
3143 @code{literal}, @code{shell}, @code{shell-always}, @code{c},
3144 @code{escape}, @code{locale}, and @code{clocale}. Default quoting
3145 style is @code{escape}, unless overridden while configuring the
3148 @opsummary{read-full-records}
3149 @item --read-full-records
3152 Specifies that @command{tar} should reblock its input, for reading
3153 from pipes on systems with buggy implementations. @xref{Reading}.
3155 @opsummary{record-size}
3156 @item --record-size=@var{size}
3158 Instructs @command{tar} to use @var{size} bytes per record when accessing the
3159 archive. @xref{Blocking Factor}.
3161 @opsummary{recursion}
3164 With this option, @command{tar} recurses into directories (default).
3167 @opsummary{recursive-unlink}
3168 @item --recursive-unlink
3171 directory hierarchies before extracting directories of the same name
3172 from the archive. @xref{Recursive Unlink}.
3174 @opsummary{remove-files}
3175 @item --remove-files
3177 Directs @command{tar} to remove the source file from the file system after
3178 appending it to an archive. @xref{remove files}.
3180 @opsummary{restrict}
3183 Disable use of some potentially harmful @command{tar} options.
3184 Currently this option disables shell invocation from multi-volume menu
3185 (@pxref{Using Multiple Tapes}).
3187 @opsummary{rmt-command}
3188 @item --rmt-command=@var{cmd}
3190 Notifies @command{tar} that it should use @var{cmd} instead of
3191 the default @file{/usr/libexec/rmt} (@pxref{Remote Tape Server}).
3193 @opsummary{rsh-command}
3194 @item --rsh-command=@var{cmd}
3196 Notifies @command{tar} that is should use @var{cmd} to communicate with remote
3197 devices. @xref{Device}.
3199 @opsummary{same-order}
3201 @itemx --preserve-order
3204 This option is an optimization for @command{tar} when running on machines with
3205 small amounts of memory. It informs @command{tar} that the list of file
3206 arguments has already been sorted to match the order of files in the
3207 archive. @xref{Reading}.
3209 @opsummary{same-owner}
3212 When extracting an archive, @command{tar} will attempt to preserve the owner
3213 specified in the @command{tar} archive with this option present.
3214 This is the default behavior for the superuser; this option has an
3215 effect only for ordinary users. @xref{Attributes}.
3217 @opsummary{same-permissions}
3218 @item --same-permissions
3220 (See @option{--preserve-permissions}; @pxref{Setting Access Permissions}.)
3226 Assume that the archive media supports seeks to arbitrary
3227 locations. Usually @command{tar} determines automatically whether
3228 the archive can be seeked or not. This option is intended for use
3229 in cases when such recognition fails. It takes effect only if the
3230 archive is open for reading (e.g. with @option{--list} or
3231 @option{--extract} options).
3233 @opsummary{show-defaults}
3234 @item --show-defaults
3236 Displays the default options used by @command{tar} and exits
3237 successfully. This option is intended for use in shell scripts.
3238 Here is an example of what you can see using this option:
3241 $ @kbd{tar --show-defaults}
3242 --format=gnu -f- -b20 --quoting-style=escape
3243 --rmt-command=/usr/libexec/rmt --rsh-command=/usr/bin/rsh
3247 Notice, that this option outputs only one line. The example output
3248 above has been split to fit page boundaries.
3250 @opsummary{show-omitted-dirs}
3251 @item --show-omitted-dirs
3253 Instructs @command{tar} to mention the directories it is skipping when
3254 operating on a @command{tar} archive. @xref{show-omitted-dirs}.
3256 @opsummary{show-transformed-names}
3257 @opsummary{show-stored-names}
3258 @item --show-transformed-names
3259 @itemx --show-stored-names
3261 Display file or member names after applying any transformations
3262 (@pxref{transform}). In particular, when used in conjunction with one of
3263 the archive creation operations it instructs @command{tar} to list the
3264 member names stored in the archive, as opposed to the actual file
3265 names. @xref{listing member and file names}.
3271 Invokes a @acronym{GNU} extension when adding files to an archive that handles
3272 sparse files efficiently. @xref{sparse}.
3274 @opsummary{sparse-version}
3275 @item --sparse-version=@var{version}
3277 Specifies the @dfn{format version} to use when archiving sparse
3278 files. Implies @option{--sparse}. @xref{sparse}. For the description
3279 of the supported sparse formats, @xref{Sparse Formats}.
3281 @opsummary{starting-file}
3282 @item --starting-file=@var{name}
3283 @itemx -K @var{name}
3285 This option affects extraction only; @command{tar} will skip extracting
3286 files in the archive until it finds one that matches @var{name}.
3289 @opsummary{strip-components}
3290 @item --strip-components=@var{number}
3291 Strip given @var{number} of leading components from file names before
3292 extraction. For example, if archive @file{archive.tar} contained
3293 @file{/some/file/name}, then running
3296 tar --extract --file archive.tar --strip-components=2
3300 would extract this file to file @file{name}.
3303 @item --suffix=@var{suffix}
3305 Alters the suffix @command{tar} uses when backing up files from the default
3306 @samp{~}. @xref{backup}.
3308 @opsummary{tape-length}
3309 @item --tape-length=@var{num}
3312 Specifies the length of tapes that @command{tar} is writing as being
3313 @w{@var{num} x 1024} bytes long. @xref{Using Multiple Tapes}.
3315 @opsummary{test-label}
3318 Reads the volume label. If an argument is specified, test whether it
3319 matches the volume label. @xref{--test-label option}.
3321 @opsummary{to-command}
3322 @item --to-command=@var{command}
3324 During extraction @command{tar} will pipe extracted files to the
3325 standard input of @var{command}. @xref{Writing to an External Program}.
3327 @opsummary{to-stdout}
3331 During extraction, @command{tar} will extract files to stdout rather
3332 than to the file system. @xref{Writing to Standard Output}.
3335 @item --totals[=@var{signo}]
3337 Displays the total number of bytes transferred when processing an
3338 archive. If an argument is given, these data are displayed on
3339 request, when signal @var{signo} is delivered to @command{tar}.
3346 Sets the data modification time of extracted files to the extraction time,
3347 rather than the data modification time stored in the archive.
3348 @xref{Data Modification Times}.
3350 @opsummary{transform}
3352 @item --transform=@var{sed-expr}
3353 @itemx --xform=@var{sed-expr}
3354 Transform file or member names using @command{sed} replacement expression
3355 @var{sed-expr}. For example,
3358 $ @kbd{tar cf archive.tar --transform 's,^\./,usr/,' .}
3362 will add to @file{archive} files from the current working directory,
3363 replacing initial @samp{./} prefix with @samp{usr/}. For the detailed
3364 discussion, @xref{transform}.
3366 To see transformed member names in verbose listings, use
3367 @option{--show-transformed-names} option
3368 (@pxref{show-transformed-names}).
3370 @opsummary{uncompress}
3373 (See @option{--compress}, @pxref{gzip})
3378 (See @option{--gzip}, @pxref{gzip})
3380 @opsummary{unlink-first}
3381 @item --unlink-first
3384 Directs @command{tar} to remove the corresponding file from the file
3385 system before extracting it from the archive. @xref{Unlink First}.
3389 Enable unquoting input file or member names (default). @xref{input
3392 @opsummary{use-compress-program}
3393 @item --use-compress-program=@var{prog}
3394 @itemx -I=@var{prog}
3396 Instructs @command{tar} to access the archive through @var{prog}, which is
3397 presumed to be a compression program of some sort. @xref{gzip}.
3402 Display file modification dates in @acronym{UTC}. This option implies
3409 Specifies that @command{tar} should be more verbose about the
3410 operations it is performing. This option can be specified multiple
3411 times for some operations to increase the amount of information displayed.
3418 Verifies that the archive was correctly written when creating an
3419 archive. @xref{verify}.
3424 Print information about the program's name, version, origin and legal
3425 status, all on standard output, and then exit successfully.
3428 @opsummary{volno-file}
3429 @item --volno-file=@var{file}
3431 Used in conjunction with @option{--multi-volume}. @command{tar} will
3432 keep track of which volume of a multi-volume archive it is working in
3433 @var{file}. @xref{volno-file}.
3436 @item --warning=@var{keyword}
3438 Enable or disable warning messages identified by @var{keyword}. The
3439 messages are suppressed if @var{keyword} is prefixed with @samp{no-}.
3442 @opsummary{wildcards}
3444 Use wildcards when matching member names with patterns.
3445 @xref{controlling pattern-matching}.
3447 @opsummary{wildcards-match-slash}
3448 @item --wildcards-match-slash
3449 Wildcards match @samp{/}.
3450 @xref{controlling pattern-matching}.
3455 Use @command{xz} for compressing or decompressing the archives. @xref{gzip}.
3459 @node Short Option Summary
3460 @subsection Short Options Cross Reference
3462 Here is an alphabetized list of all of the short option forms, matching
3463 them with the equivalent long option.
3465 @multitable @columnfractions 0.20 0.80
3466 @headitem Short Option @tab Reference
3468 @item -A @tab @ref{--concatenate}.
3470 @item -B @tab @ref{--read-full-records}.
3472 @item -C @tab @ref{--directory}.
3474 @item -F @tab @ref{--info-script}.
3476 @item -G @tab @ref{--incremental}.
3478 @item -J @tab @ref{--xz}.
3480 @item -K @tab @ref{--starting-file}.
3482 @item -L @tab @ref{--tape-length}.
3484 @item -M @tab @ref{--multi-volume}.
3486 @item -N @tab @ref{--newer}.
3488 @item -O @tab @ref{--to-stdout}.
3490 @item -P @tab @ref{--absolute-names}.
3492 @item -R @tab @ref{--block-number}.
3494 @item -S @tab @ref{--sparse}.
3496 @item -T @tab @ref{--files-from}.
3498 @item -U @tab @ref{--unlink-first}.
3500 @item -V @tab @ref{--label}.
3502 @item -W @tab @ref{--verify}.
3504 @item -X @tab @ref{--exclude-from}.
3506 @item -Z @tab @ref{--compress}.
3508 @item -b @tab @ref{--blocking-factor}.
3510 @item -c @tab @ref{--create}.
3512 @item -d @tab @ref{--compare}.
3514 @item -f @tab @ref{--file}.
3516 @item -g @tab @ref{--listed-incremental}.
3518 @item -h @tab @ref{--dereference}.
3520 @item -i @tab @ref{--ignore-zeros}.
3522 @item -j @tab @ref{--bzip2}.
3524 @item -k @tab @ref{--keep-old-files}.
3526 @item -l @tab @ref{--check-links}.
3528 @item -m @tab @ref{--touch}.
3530 @item -o @tab When creating, @ref{--no-same-owner}, when extracting ---
3531 @ref{--portability}.
3533 The latter usage is deprecated. It is retained for compatibility with
3534 the earlier versions of @GNUTAR{}. In future releases
3535 @option{-o} will be equivalent to @option{--no-same-owner} only.
3537 @item -p @tab @ref{--preserve-permissions}.
3539 @item -r @tab @ref{--append}.
3541 @item -s @tab @ref{--same-order}.
3543 @item -t @tab @ref{--list}.
3545 @item -u @tab @ref{--update}.
3547 @item -v @tab @ref{--verbose}.
3549 @item -w @tab @ref{--interactive}.
3551 @item -x @tab @ref{--extract}.
3553 @item -z @tab @ref{--gzip}.
3558 @section @GNUTAR{} documentation
3560 @cindex Getting program version number
3562 @cindex Version of the @command{tar} program
3563 Being careful, the first thing is really checking that you are using
3564 @GNUTAR{}, indeed. The @option{--version} option
3565 causes @command{tar} to print information about its name, version,
3566 origin and legal status, all on standard output, and then exit
3567 successfully. For example, @w{@samp{tar --version}} might print:
3570 tar (GNU tar) @value{VERSION}
3571 Copyright (C) 2010 Free Software Foundation, Inc.
3572 Copyright (C) 2010 Free Software Foundation, Inc.
3573 License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>.
3574 This is free software: you are free to change and redistribute it.
3575 There is NO WARRANTY, to the extent permitted by law.
3577 Written by John Gilmore and Jay Fenlason.
3581 The first occurrence of @samp{tar} in the result above is the program
3582 name in the package (for example, @command{rmt} is another program),
3583 while the second occurrence of @samp{tar} is the name of the package
3584 itself, containing possibly many programs. The package is currently
3585 named @samp{tar}, after the name of the main program it
3586 contains@footnote{There are plans to merge the @command{cpio} and
3587 @command{tar} packages into a single one which would be called
3588 @code{paxutils}. So, who knows if, one of this days, the
3589 @option{--version} would not output @w{@samp{tar (@acronym{GNU}
3592 @cindex Obtaining help
3593 @cindex Listing all @command{tar} options
3594 @xopindex{help, introduction}
3595 Another thing you might want to do is checking the spelling or meaning
3596 of some particular @command{tar} option, without resorting to this
3597 manual, for once you have carefully read it. @GNUTAR{}
3598 has a short help feature, triggerable through the
3599 @option{--help} option. By using this option, @command{tar} will
3600 print a usage message listing all available options on standard
3601 output, then exit successfully, without doing anything else and
3602 ignoring all other options. Even if this is only a brief summary, it
3603 may be several screens long. So, if you are not using some kind of
3604 scrollable window, you might prefer to use something like:
3607 $ @kbd{tar --help | less}
3611 presuming, here, that you like using @command{less} for a pager. Other
3612 popular pagers are @command{more} and @command{pg}. If you know about some
3613 @var{keyword} which interests you and do not want to read all the
3614 @option{--help} output, another common idiom is doing:
3617 tar --help | grep @var{keyword}
3621 for getting only the pertinent lines. Notice, however, that some
3622 @command{tar} options have long description lines and the above
3623 command will list only the first of them.
3625 The exact look of the option summary displayed by @kbd{tar --help} is
3626 configurable. @xref{Configuring Help Summary}, for a detailed description.
3629 If you only wish to check the spelling of an option, running @kbd{tar
3630 --usage} may be a better choice. This will display a terse list of
3631 @command{tar} options without accompanying explanations.
3633 The short help output is quite succinct, and you might have to get
3634 back to the full documentation for precise points. If you are reading
3635 this paragraph, you already have the @command{tar} manual in some
3636 form. This manual is available in a variety of forms from
3637 @url{http://www.gnu.org/software/tar/manual}. It may be printed out of the @GNUTAR{}
3638 distribution, provided you have @TeX{} already installed somewhere,
3639 and a laser printer around. Just configure the distribution, execute
3640 the command @w{@samp{make dvi}}, then print @file{doc/tar.dvi} the
3641 usual way (contact your local guru to know how). If @GNUTAR{}
3642 has been conveniently installed at your place, this
3643 manual is also available in interactive, hypertextual form as an Info
3644 file. Just call @w{@samp{info tar}} or, if you do not have the
3645 @command{info} program handy, use the Info reader provided within
3646 @acronym{GNU} Emacs, calling @samp{tar} from the main Info menu.
3648 There is currently no @code{man} page for @GNUTAR{}.
3649 If you observe such a @code{man} page on the system you are running,
3650 either it does not belong to @GNUTAR{}, or it has not
3651 been produced by @acronym{GNU}. Some package maintainers convert
3652 @kbd{tar --help} output to a man page, using @command{help2man}. In
3653 any case, please bear in mind that the authoritative source of
3654 information about @GNUTAR{} is this Texinfo documentation.
3657 @section Obtaining @GNUTAR{} default values
3659 @opindex show-defaults
3660 @GNUTAR{} has some predefined defaults that are used when you do not
3661 explicitly specify another values. To obtain a list of such
3662 defaults, use @option{--show-defaults} option. This will output the
3663 values in the form of @command{tar} command line options:
3667 $ @kbd{tar --show-defaults}
3668 --format=gnu -f- -b20 --quoting-style=escape
3669 --rmt-command=/etc/rmt --rsh-command=/usr/bin/rsh
3674 Notice, that this option outputs only one line. The example output above
3675 has been split to fit page boundaries.
3678 The above output shows that this version of @GNUTAR{} defaults to
3679 using @samp{gnu} archive format (@pxref{Formats}), it uses standard
3680 output as the archive, if no @option{--file} option has been given
3681 (@pxref{file tutorial}), the default blocking factor is 20
3682 (@pxref{Blocking Factor}). It also shows the default locations where
3683 @command{tar} will look for @command{rmt} and @command{rsh} binaries.
3686 @section Checking @command{tar} progress
3688 Typically, @command{tar} performs most operations without reporting any
3689 information to the user except error messages. When using @command{tar}
3690 with many options, particularly ones with complicated or
3691 difficult-to-predict behavior, it is possible to make serious mistakes.
3692 @command{tar} provides several options that make observing @command{tar}
3693 easier. These options cause @command{tar} to print information as it
3694 progresses in its job, and you might want to use them just for being
3695 more careful about what is going on, or merely for entertaining
3696 yourself. If you have encountered a problem when operating on an
3697 archive, however, you may need more information than just an error
3698 message in order to solve the problem. The following options can be
3699 helpful diagnostic tools.
3701 @cindex Verbose operation
3703 Normally, the @option{--list} (@option{-t}) command to list an archive
3704 prints just the file names (one per line) and the other commands are
3705 silent. When used with most operations, the @option{--verbose}
3706 (@option{-v}) option causes @command{tar} to print the name of each
3707 file or archive member as it is processed. This and the other options
3708 which make @command{tar} print status information can be useful in
3709 monitoring @command{tar}.
3711 With @option{--create} or @option{--extract}, @option{--verbose} used
3712 once just prints the names of the files or members as they are processed.
3713 Using it twice causes @command{tar} to print a longer listing
3714 (@xref{verbose member listing}, for the description) for each member.
3715 Since @option{--list} already prints the names of the members,
3716 @option{--verbose} used once with @option{--list} causes @command{tar}
3717 to print an @samp{ls -l} type listing of the files in the archive.
3718 The following examples both extract members with long list output:
3721 $ @kbd{tar --extract --file=archive.tar --verbose --verbose}
3722 $ @kbd{tar xvvf archive.tar}
3725 Verbose output appears on the standard output except when an archive is
3726 being written to the standard output, as with @samp{tar --create
3727 --file=- --verbose} (@samp{tar cfv -}, or even @samp{tar cv}---if the
3728 installer let standard output be the default archive). In that case
3729 @command{tar} writes verbose output to the standard error stream.
3731 If @option{--index-file=@var{file}} is specified, @command{tar} sends
3732 verbose output to @var{file} rather than to standard output or standard
3736 @cindex Obtaining total status information
3738 The @option{--totals} option causes @command{tar} to print on the
3739 standard error the total amount of bytes transferred when processing
3740 an archive. When creating or appending to an archive, this option
3741 prints the number of bytes written to the archive and the average
3742 speed at which they have been written, e.g.:
3746 $ @kbd{tar -c -f archive.tar --totals /home}
3747 Total bytes written: 7924664320 (7.4GiB, 85MiB/s)
3751 When reading an archive, this option displays the number of bytes
3756 $ @kbd{tar -x -f archive.tar --totals}
3757 Total bytes read: 7924664320 (7.4GiB, 95MiB/s)
3761 Finally, when deleting from an archive, the @option{--totals} option
3762 displays both numbers plus number of bytes removed from the archive:
3766 $ @kbd{tar --delete -f foo.tar --totals --wildcards '*~'}
3767 Total bytes read: 9543680 (9.2MiB, 201MiB/s)
3768 Total bytes written: 3829760 (3.7MiB, 81MiB/s)
3769 Total bytes deleted: 1474048
3773 You can also obtain this information on request. When
3774 @option{--totals} is used with an argument, this argument is
3775 interpreted as a symbolic name of a signal, upon delivery of which the
3776 statistics is to be printed:
3779 @item --totals=@var{signo}
3780 Print statistics upon delivery of signal @var{signo}. Valid arguments
3781 are: @code{SIGHUP}, @code{SIGQUIT}, @code{SIGINT}, @code{SIGUSR1} and
3782 @code{SIGUSR2}. Shortened names without @samp{SIG} prefix are also
3786 Both forms of @option{--totals} option can be used simultaneously.
3787 Thus, @kbd{tar -x --totals --totals=USR1} instructs @command{tar} to
3788 extract all members from its default archive and print statistics
3789 after finishing the extraction, as well as when receiving signal
3792 @anchor{Progress information}
3793 @cindex Progress information
3794 The @option{--checkpoint} option prints an occasional message
3795 as @command{tar} reads or writes the archive. It is designed for
3796 those who don't need the more detailed (and voluminous) output of
3797 @option{--block-number} (@option{-R}), but do want visual confirmation
3798 that @command{tar} is actually making forward progress. By default it
3799 prints a message each 10 records read or written. This can be changed
3800 by giving it a numeric argument after an equal sign:
3803 $ @kbd{tar -c --checkpoint=1000} /var
3804 tar: Write checkpoint 1000
3805 tar: Write checkpoint 2000
3806 tar: Write checkpoint 3000
3809 This example shows the default checkpoint message used by
3810 @command{tar}. If you place a dot immediately after the equal
3811 sign, it will print a @samp{.} at each checkpoint@footnote{This is
3812 actually a shortcut for @option{--checkpoint=@var{n}
3813 --checkpoint-action=dot}. @xref{checkpoints, dot}.}. For example:
3816 $ @kbd{tar -c --checkpoint=.1000} /var
3820 The @option{--checkpoint} option provides a flexible mechanism for
3821 executing arbitrary actions upon hitting checkpoints, see the next
3822 section (@pxref{checkpoints}), for more information on it.
3824 @opindex show-omitted-dirs
3825 @anchor{show-omitted-dirs}
3826 The @option{--show-omitted-dirs} option, when reading an archive---with
3827 @option{--list} or @option{--extract}, for example---causes a message
3828 to be printed for each directory in the archive which is skipped.
3829 This happens regardless of the reason for skipping: the directory might
3830 not have been named on the command line (implicitly or explicitly),
3831 it might be excluded by the use of the
3832 @option{--exclude=@var{pattern}} option, or some other reason.
3834 @opindex block-number
3835 @cindex Block number where error occurred
3836 @anchor{block-number}
3837 If @option{--block-number} (@option{-R}) is used, @command{tar} prints, along with
3838 every message it would normally produce, the block number within the
3839 archive where the message was triggered. Also, supplementary messages
3840 are triggered when reading blocks full of NULs, or when hitting end of
3841 file on the archive. As of now, if the archive is properly terminated
3842 with a NUL block, the reading of the file may stop before end of file
3843 is met, so the position of end of file will not usually show when
3844 @option{--block-number} (@option{-R}) is used. Note that @GNUTAR{}
3845 drains the archive before exiting when reading the
3846 archive from a pipe.
3848 @cindex Error message, block number of
3849 This option is especially useful when reading damaged archives, since
3850 it helps pinpoint the damaged sections. It can also be used with
3851 @option{--list} (@option{-t}) when listing a file-system backup tape, allowing you to
3852 choose among several backup tapes when retrieving a file later, in
3853 favor of the tape where the file appears earliest (closest to the
3854 front of the tape). @xref{backup}.
3857 @section Checkpoints
3858 @cindex checkpoints, defined
3860 @opindex checkpoint-action
3862 A @dfn{checkpoint} is a moment of time before writing @var{n}th record to
3863 the archive (a @dfn{write checkpoint}), or before reading @var{n}th record
3864 from the archive (a @dfn{read checkpoint}). Checkpoints allow to
3865 periodically execute arbitrary actions.
3867 The checkpoint facility is enabled using the following option:
3870 @xopindex{checkpoint, defined}
3871 @item --checkpoint[=@var{n}]
3872 Schedule checkpoints before writing or reading each @var{n}th record.
3873 The default value for @var{n} is 10.
3876 A list of arbitrary @dfn{actions} can be executed at each checkpoint.
3877 These actions include: pausing, displaying textual messages, and
3878 executing arbitrary external programs. Actions are defined using
3879 the @option{--checkpoint-action} option.
3882 @xopindex{checkpoint-action, defined}
3883 @item --checkpoint-action=@var{action}
3884 Execute an @var{action} at each checkpoint.
3887 @cindex @code{echo}, checkpoint action
3888 The simplest value of @var{action} is @samp{echo}. It instructs
3889 @command{tar} to display the default message on the standard error
3890 stream upon arriving at each checkpoint. The default message is (in
3891 @acronym{POSIX} locale) @samp{Write checkpoint @var{n}}, for write
3892 checkpoints, and @samp{Read checkpoint @var{n}}, for read checkpoints.
3893 Here, @var{n} represents ordinal number of the checkpoint.
3895 In another locales, translated versions of this message are used.
3897 This is the default action, so running:
3900 $ @kbd{tar -c --checkpoint=1000 --checkpoint-action=echo} /var
3907 $ @kbd{tar -c --checkpoint=1000} /var
3910 The @samp{echo} action also allows to supply a customized message.
3911 You do so by placing an equals sign and the message right after it,
3915 --checkpoint-action="echo=Hit %s checkpoint #%u"
3918 The @samp{%s} and @samp{%u} in the above example are
3919 @dfn{meta-characters}. The @samp{%s} meta-character is replaced with
3920 the @dfn{type} of the checkpoint: @samp{write} or
3921 @samp{read} (or a corresponding translated version in locales other
3922 than @acronym{POSIX}). The @samp{%u} meta-character is replaced with
3923 the ordinal number of the checkpoint. Thus, the above example could
3924 produce the following output when used with the @option{--create}
3928 tar: Hit write checkpoint #10
3929 tar: Hit write checkpoint #20
3930 tar: Hit write checkpoint #30
3933 Aside from meta-character expansion, the message string is subject to
3934 @dfn{unquoting}, during which the backslash @dfn{escape sequences} are
3935 replaced with their corresponding @acronym{ASCII} characters
3936 (@pxref{escape sequences}). E.g. the following action will produce an
3937 audible bell and the message described above at each checkpoint:
3940 --checkpoint-action='echo=\aHit %s checkpoint #%u'
3943 @cindex @code{bell}, checkpoint action
3944 There is also a special action which produces an audible signal:
3945 @samp{bell}. It is not equivalent to @samp{echo='\a'}, because
3946 @samp{bell} sends the bell directly to the console (@file{/dev/tty}),
3947 whereas @samp{echo='\a'} sends it to the standard error.
3949 @cindex @code{ttyout}, checkpoint action
3950 The @samp{ttyout=@var{string}} action outputs @var{string} to
3951 @file{/dev/tty}, so it can be used even if the standard output is
3952 redirected elsewhere. The @var{string} is subject to the same
3953 modifications as with @samp{echo} action. In contrast to the latter,
3954 @samp{ttyout} does not prepend @command{tar} executable name to the
3955 string, nor does it output a newline after it. For example, the
3956 following action will print the checkpoint message at the same screen
3957 line, overwriting any previous message:
3960 --checkpoint-action="ttyout=\rHit %s checkpoint #%u"
3963 @cindex @code{dot}, checkpoint action
3964 Another available checkpoint action is @samp{dot} (or @samp{.}). It
3965 instructs @command{tar} to print a single dot on the standard listing
3969 $ @kbd{tar -c --checkpoint=1000 --checkpoint-action=dot} /var
3973 For compatibility with previous @GNUTAR{} versions, this action can
3974 be abbreviated by placing a dot in front of the checkpoint frequency,
3975 as shown in the previous section.
3977 @cindex @code{sleep}, checkpoint action
3978 Yet another action, @samp{sleep}, pauses @command{tar} for a specified
3979 amount of seconds. The following example will stop for 30 seconds at each
3983 $ @kbd{tar -c --checkpoint=1000 --checkpoint-action=sleep=30}
3986 @cindex @code{exec}, checkpoint action
3987 Finally, the @code{exec} action executes a given external program.
3991 $ @kbd{tar -c --checkpoint=1000 --checkpoint-action=exec=/sbin/cpoint}
3994 This program is executed using @command{/bin/sh -c}, with no
3995 additional arguments. Its exit code is ignored. It gets a copy of
3996 @command{tar}'s environment plus the following variables:
3999 @vrindex TAR_VERSION, checkpoint script environment
4001 @GNUTAR{} version number.
4003 @vrindex TAR_ARCHIVE, checkpoint script environment
4005 The name of the archive @command{tar} is processing.
4007 @vrindex TAR_BLOCKING_FACTOR, checkpoint script environment
4008 @item TAR_BLOCKING_FACTOR
4009 Current blocking factor (@pxref{Blocking}).
4011 @vrindex TAR_CHECKPOINT, checkpoint script environment
4012 @item TAR_CHECKPOINT
4013 Number of the checkpoint.
4015 @vrindex TAR_SUBCOMMAND, checkpoint script environment
4016 @item TAR_SUBCOMMAND
4017 A short option describing the operation @command{tar} is executing.
4018 @xref{Operations}, for a complete list of subcommand options.
4020 @vrindex TAR_FORMAT, checkpoint script environment
4022 Format of the archive being processed. @xref{Formats}, for a complete
4023 list of archive format names.
4026 Any number of actions can be defined, by supplying several
4027 @option{--checkpoint-action} options in the command line. For
4028 example, the command below displays two messages, pauses
4029 execution for 30 seconds and executes the @file{/sbin/cpoint} script:
4033 $ @kbd{tar -c -f arc.tar \
4034 --checkpoint-action='\aecho=Hit %s checkpoint #%u' \
4035 --checkpoint-action='echo=Sleeping for 30 seconds' \
4036 --checkpoint-action='sleep=30' \
4037 --checkpoint-action='exec=/sbin/cpoint'}
4041 This example also illustrates the fact that
4042 @option{--checkpoint-action} can be used without
4043 @option{--checkpoint}. In this case, the default checkpoint frequency
4044 (at each 10th record) is assumed.
4047 @section Controlling Warning Messages
4049 Sometimes, while performing the requested task, @GNUTAR{} notices
4050 some conditions that are not exactly errors, but which the user
4051 should be aware of. When this happens, @command{tar} issues a
4052 @dfn{warning message} describing the condition. Warning messages
4053 are output to the standard error and they do not affect the exit
4054 code of @command{tar} command.
4056 @xopindex{warning, explained}
4057 @GNUTAR{} allows the user to suppress some or all of its warning
4061 @item --warning=@var{keyword}
4062 Control display of the warning messages identified by @var{keyword}.
4063 If @var{keyword} starts with the prefix @samp{no-}, such messages are
4064 suppressed. Otherwise, they are enabled.
4066 Multiple @option{--warning} messages accumulate.
4068 The tables below list allowed values for @var{keyword} along with the
4069 warning messages they control.
4072 @subheading Keywords controlling @command{tar} operation
4076 Enable all warning messages. This is the default.
4079 Disable all warning messages.
4080 @kwindex filename-with-nuls
4081 @cindex @samp{file name read contains nul character}, warning message
4082 @item filename-with-nuls
4083 @samp{%s: file name read contains nul character}
4084 @kwindex alone-zero-block
4085 @cindex @samp{A lone zero block at}, warning message
4086 @item alone-zero-block
4087 @samp{A lone zero block at %s}
4090 @subheading Keywords applicable for @command{tar --create}
4093 @cindex @samp{contains a cache directory tag}, warning message
4095 @samp{%s: contains a cache directory tag %s; %s}
4096 @kwindex file-shrank
4097 @cindex @samp{File shrank by %s bytes}, warning message
4099 @samp{%s: File shrank by %s bytes; padding with zeros}
4101 @cindex @samp{file is on a different filesystem}, warning message
4103 @samp{%s: file is on a different filesystem; not dumped}
4104 @kwindex file-ignored
4105 @cindex @samp{Unknown file type; file ignored}, warning message
4106 @cindex @samp{socket ignored}, warning message
4107 @cindex @samp{door ignored}, warning message
4109 @samp{%s: Unknown file type; file ignored}
4110 @*@samp{%s: socket ignored}
4111 @*@samp{%s: door ignored}
4112 @kwindex file-unchanged
4113 @cindex @samp{file is unchanged; not dumped}, warning message
4114 @item file-unchanged
4115 @samp{%s: file is unchanged; not dumped}
4116 @kwindex ignore-archive
4117 @cindex @samp{file is the archive; not dumped}, warning message
4118 @kwindex ignore-archive
4119 @cindex @samp{file is the archive; not dumped}, warning message
4120 @item ignore-archive
4121 @samp{%s: file is the archive; not dumped}
4122 @kwindex file-removed
4123 @cindex @samp{File removed before we read it}, warning message
4125 @samp{%s: File removed before we read it}
4126 @kwindex file-changed
4127 @cindex @samp{file changed as we read it}, warning message
4129 @samp{%s: file changed as we read it}
4132 @subheading Keywords applicable for @command{tar --extract}
4135 @cindex @samp{implausibly old time stamp %s}, warning message
4136 @cindex @samp{time stamp %s is %s s in the future}, warning message
4138 @samp{%s: implausibly old time stamp %s}
4139 @*@samp{%s: time stamp %s is %s s in the future}
4140 @kwindex contiguous-cast
4141 @cindex @samp{Extracting contiguous files as regular files}, warning message
4142 @item contiguous-cast
4143 @samp{Extracting contiguous files as regular files}
4144 @kwindex symlink-cast
4145 @cindex @samp{Attempting extraction of symbolic links as hard links}, warning message
4147 @samp{Attempting extraction of symbolic links as hard links}
4148 @kwindex unknown-cast
4149 @cindex @samp{Unknown file type `%c', extracted as normal file}, warning message
4151 @samp{%s: Unknown file type `%c', extracted as normal file}
4152 @kwindex ignore-newer
4153 @cindex @samp{Current %s is newer or same age}, warning message
4155 @samp{Current %s is newer or same age}
4156 @kwindex unknown-keyword
4157 @cindex @samp{Ignoring unknown extended header keyword `%s'}, warning message
4158 @item unknown-keyword
4159 @samp{Ignoring unknown extended header keyword `%s'}
4162 @subheading Keywords controlling incremental extraction:
4164 @kwindex rename-directory
4165 @cindex @samp{%s: Directory has been renamed from %s}, warning message
4166 @cindex @samp{%s: Directory has been renamed}, warning message
4167 @item rename-directory
4168 @samp{%s: Directory has been renamed from %s}
4169 @*@samp{%s: Directory has been renamed}
4170 @kwindex new-directory
4171 @cindex @samp{%s: Directory is new}, warning message
4173 @samp{%s: Directory is new}
4175 @cindex @samp{%s: directory is on a different device: not purging}, warning message
4177 @samp{%s: directory is on a different device: not purging}
4178 @kwindex bad-dumpdir
4179 @cindex @samp{Malformed dumpdir: 'X' never used}, warning message
4181 @samp{Malformed dumpdir: 'X' never used}
4185 @section Asking for Confirmation During Operations
4186 @cindex Interactive operation
4188 Typically, @command{tar} carries out a command without stopping for
4189 further instructions. In some situations however, you may want to
4190 exclude some files and archive members from the operation (for instance
4191 if disk or storage space is tight). You can do this by excluding
4192 certain files automatically (@pxref{Choosing}), or by performing
4193 an operation interactively, using the @option{--interactive} (@option{-w}) option.
4194 @command{tar} also accepts @option{--confirmation} for this option.
4196 @opindex interactive
4197 When the @option{--interactive} (@option{-w}) option is specified, before
4198 reading, writing, or deleting files, @command{tar} first prints a message
4199 for each such file, telling what operation it intends to take, then asks
4200 for confirmation on the terminal. The actions which require
4201 confirmation include adding a file to the archive, extracting a file
4202 from the archive, deleting a file from the archive, and deleting a file
4203 from disk. To confirm the action, you must type a line of input
4204 beginning with @samp{y}. If your input line begins with anything other
4205 than @samp{y}, @command{tar} skips that file.
4207 If @command{tar} is reading the archive from the standard input,
4208 @command{tar} opens the file @file{/dev/tty} to support the interactive
4211 Verbose output is normally sent to standard output, separate from
4212 other error messages. However, if the archive is produced directly
4213 on standard output, then verbose output is mixed with errors on
4214 @code{stderr}. Producing the archive on standard output may be used
4215 as a way to avoid using disk space, when the archive is soon to be
4216 consumed by another process reading it, say. Some people felt the need
4217 of producing an archive on stdout, still willing to segregate between
4218 verbose output and error output. A possible approach would be using a
4219 named pipe to receive the archive, and having the consumer process to
4220 read from that named pipe. This has the advantage of letting standard
4221 output free to receive verbose output, all separate from errors.
4224 @chapter @GNUTAR{} Operations
4237 @section Basic @GNUTAR{} Operations
4239 The basic @command{tar} operations, @option{--create} (@option{-c}),
4240 @option{--list} (@option{-t}) and @option{--extract} (@option{--get},
4241 @option{-x}), are currently presented and described in the tutorial
4242 chapter of this manual. This section provides some complementary notes
4243 for these operations.
4246 @xopindex{create, complementary notes}
4250 Creating an empty archive would have some kind of elegance. One can
4251 initialize an empty archive and later use @option{--append}
4252 (@option{-r}) for adding all members. Some applications would not
4253 welcome making an exception in the way of adding the first archive
4254 member. On the other hand, many people reported that it is
4255 dangerously too easy for @command{tar} to destroy a magnetic tape with
4256 an empty archive@footnote{This is well described in @cite{Unix-haters
4257 Handbook}, by Simson Garfinkel, Daniel Weise & Steven Strassmann, IDG
4258 Books, ISBN 1-56884-203-1.}. The two most common errors are:
4262 Mistakingly using @code{create} instead of @code{extract}, when the
4263 intent was to extract the full contents of an archive. This error
4264 is likely: keys @kbd{c} and @kbd{x} are right next to each other on
4265 the QWERTY keyboard. Instead of being unpacked, the archive then
4266 gets wholly destroyed. When users speak about @dfn{exploding} an
4267 archive, they usually mean something else :-).
4270 Forgetting the argument to @code{file}, when the intent was to create
4271 an archive with a single file in it. This error is likely because a
4272 tired user can easily add the @kbd{f} key to the cluster of option
4273 letters, by the mere force of habit, without realizing the full
4274 consequence of doing so. The usual consequence is that the single
4275 file, which was meant to be saved, is rather destroyed.
4278 So, recognizing the likelihood and the catastrophic nature of these
4279 errors, @GNUTAR{} now takes some distance from elegance, and
4280 cowardly refuses to create an archive when @option{--create} option is
4281 given, there are no arguments besides options, and
4282 @option{--files-from} (@option{-T}) option is @emph{not} used. To get
4283 around the cautiousness of @GNUTAR{} and nevertheless create an
4284 archive with nothing in it, one may still use, as the value for the
4285 @option{--files-from} option, a file with no names in it, as shown in
4286 the following commands:
4289 @kbd{tar --create --file=empty-archive.tar --files-from=/dev/null}
4290 @kbd{tar cfT empty-archive.tar /dev/null}
4293 @xopindex{extract, complementary notes}
4298 A socket is stored, within a @GNUTAR{} archive, as a pipe.
4300 @item @option{--list} (@option{-t})
4302 @GNUTAR{} now shows dates as @samp{1996-08-30},
4303 while it used to show them as @samp{Aug 30 1996}. Preferably,
4304 people should get used to ISO 8601 dates. Local American dates should
4305 be made available again with full date localization support, once
4306 ready. In the meantime, programs not being localizable for dates
4307 should prefer international dates, that's really the way to go.
4309 Look up @url{http://www.cl.cam.ac.uk/@/~mgk25/@/iso-time.html} if you
4310 are curious, it contains a detailed explanation of the ISO 8601 standard.
4315 @section Advanced @GNUTAR{} Operations
4317 Now that you have learned the basics of using @GNUTAR{}, you may want
4318 to learn about further ways in which @command{tar} can help you.
4320 This chapter presents five, more advanced operations which you probably
4321 won't use on a daily basis, but which serve more specialized functions.
4322 We also explain the different styles of options and why you might want
4323 to use one or another, or a combination of them in your @command{tar}
4324 commands. Additionally, this chapter includes options which allow you to
4325 define the output from @command{tar} more carefully, and provide help and
4326 error correction in special circumstances.
4328 @FIXME{check this after the chapter is actually revised to make sure
4329 it still introduces the info in the chapter correctly : ).}
4341 @subsection The Five Advanced @command{tar} Operations
4343 @cindex basic operations
4344 In the last chapter, you learned about the first three operations to
4345 @command{tar}. This chapter presents the remaining five operations to
4346 @command{tar}: @option{--append}, @option{--update}, @option{--concatenate},
4347 @option{--delete}, and @option{--compare}.
4349 You are not likely to use these operations as frequently as those
4350 covered in the last chapter; however, since they perform specialized
4351 functions, they are quite useful when you do need to use them. We
4352 will give examples using the same directory and files that you created
4353 in the last chapter. As you may recall, the directory is called
4354 @file{practice}, the files are @samp{jazz}, @samp{blues}, @samp{folk},
4355 and the two archive files you created are
4356 @samp{collection.tar} and @samp{music.tar}.
4358 We will also use the archive files @samp{afiles.tar} and
4359 @samp{bfiles.tar}. The archive @samp{afiles.tar} contains the members @samp{apple},
4360 @samp{angst}, and @samp{aspic}; @samp{bfiles.tar} contains the members
4361 @samp{./birds}, @samp{baboon}, and @samp{./box}.
4363 Unless we state otherwise, all practicing you do and examples you follow
4364 in this chapter will take place in the @file{practice} directory that
4365 you created in the previous chapter; see @ref{prepare for examples}.
4366 (Below in this section, we will remind you of the state of the examples
4367 where the last chapter left them.)
4369 The five operations that we will cover in this chapter are:
4374 Add new entries to an archive that already exists.
4377 Add more recent copies of archive members to the end of an archive, if
4382 Add one or more pre-existing archives to the end of another archive.
4384 Delete items from an archive (does not work on tapes).
4388 Compare archive members to their counterparts in the file system.
4392 @subsection How to Add Files to Existing Archives: @option{--append}
4394 @cindex appending files to existing archive
4396 If you want to add files to an existing archive, you don't need to
4397 create a new archive; you can use @option{--append} (@option{-r}).
4398 The archive must already exist in order to use @option{--append}. (A
4399 related operation is the @option{--update} operation; you can use this
4400 to add newer versions of archive members to an existing archive. To learn how to
4401 do this with @option{--update}, @pxref{update}.)
4403 If you use @option{--append} to add a file that has the same name as an
4404 archive member to an archive containing that archive member, then the
4405 old member is not deleted. What does happen, however, is somewhat
4406 complex. @command{tar} @emph{allows} you to have infinite number of files
4407 with the same name. Some operations treat these same-named members no
4408 differently than any other set of archive members: for example, if you
4409 view an archive with @option{--list} (@option{-t}), you will see all
4410 of those members listed, with their data modification times, owners, etc.
4412 Other operations don't deal with these members as perfectly as you might
4413 prefer; if you were to use @option{--extract} to extract the archive,
4414 only the most recently added copy of a member with the same name as
4415 other members would end up in the working directory. This is because
4416 @option{--extract} extracts an archive in the order the members appeared
4417 in the archive; the most recently archived members will be extracted
4418 last. Additionally, an extracted member will @emph{replace} a file of
4419 the same name which existed in the directory already, and @command{tar}
4420 will not prompt you about this@footnote{Unless you give it
4421 @option{--keep-old-files} option, or the disk copy is newer than
4422 the one in the archive and you invoke @command{tar} with
4423 @option{--keep-newer-files} option.}. Thus, only the most recently archived
4424 member will end up being extracted, as it will replace the one
4425 extracted before it, and so on.
4427 @cindex extracting @var{n}th copy of the file
4428 @xopindex{occurrence, described}
4429 There exists a special option that allows you to get around this
4430 behavior and extract (or list) only a particular copy of the file.
4431 This is @option{--occurrence} option. If you run @command{tar} with
4432 this option, it will extract only the first copy of the file. You
4433 may also give this option an argument specifying the number of
4434 copy to be extracted. Thus, for example if the archive
4435 @file{archive.tar} contained three copies of file @file{myfile}, then
4439 tar --extract --file archive.tar --occurrence=2 myfile
4443 would extract only the second copy. @xref{Option
4444 Summary,---occurrence}, for the description of @option{--occurrence}
4447 @FIXME{ hag -- you might want to incorporate some of the above into the
4448 MMwtSN node; not sure. i didn't know how to make it simpler...
4450 There are a few ways to get around this. Xref to Multiple Members
4451 with the Same Name, maybe.}
4453 @cindex Members, replacing with other members
4454 @cindex Replacing members with other members
4455 @xopindex{delete, using before --append}
4456 If you want to replace an archive member, use @option{--delete} to
4457 delete the member you want to remove from the archive, and then use
4458 @option{--append} to add the member you want to be in the archive. Note
4459 that you can not change the order of the archive; the most recently
4460 added member will still appear last. In this sense, you cannot truly
4461 ``replace'' one member with another. (Replacing one member with another
4462 will not work on certain types of media, such as tapes; see @ref{delete}
4463 and @ref{Media}, for more information.)
4466 * appending files:: Appending Files to an Archive
4470 @node appending files
4471 @subsubsection Appending Files to an Archive
4472 @cindex Adding files to an Archive
4473 @cindex Appending files to an Archive
4474 @cindex Archives, Appending files to
4477 The simplest way to add a file to an already existing archive is the
4478 @option{--append} (@option{-r}) operation, which writes specified
4479 files into the archive whether or not they are already among the
4482 When you use @option{--append}, you @emph{must} specify file name
4483 arguments, as there is no default. If you specify a file that already
4484 exists in the archive, another copy of the file will be added to the
4485 end of the archive. As with other operations, the member names of the
4486 newly added files will be exactly the same as their names given on the
4487 command line. The @option{--verbose} (@option{-v}) option will print
4488 out the names of the files as they are written into the archive.
4490 @option{--append} cannot be performed on some tape drives, unfortunately,
4491 due to deficiencies in the formats those tape drives use. The archive
4492 must be a valid @command{tar} archive, or else the results of using this
4493 operation will be unpredictable. @xref{Media}.
4495 To demonstrate using @option{--append} to add a file to an archive,
4496 create a file called @file{rock} in the @file{practice} directory.
4497 Make sure you are in the @file{practice} directory. Then, run the
4498 following @command{tar} command to add @file{rock} to
4499 @file{collection.tar}:
4502 $ @kbd{tar --append --file=collection.tar rock}
4506 If you now use the @option{--list} (@option{-t}) operation, you will see that
4507 @file{rock} has been added to the archive:
4510 $ @kbd{tar --list --file=collection.tar}
4511 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
4512 -rw-r--r-- me user 21 1996-09-23 16:44 blues
4513 -rw-r--r-- me user 20 1996-09-23 16:44 folk
4514 -rw-r--r-- me user 20 1996-09-23 16:44 rock
4518 @subsubsection Multiple Members with the Same Name
4519 @cindex members, multiple
4520 @cindex multiple members
4522 You can use @option{--append} (@option{-r}) to add copies of files
4523 which have been updated since the archive was created. (However, we
4524 do not recommend doing this since there is another @command{tar}
4525 option called @option{--update}; @xref{update}, for more information.
4526 We describe this use of @option{--append} here for the sake of
4527 completeness.) When you extract the archive, the older version will
4528 be effectively lost. This works because files are extracted from an
4529 archive in the order in which they were archived. Thus, when the
4530 archive is extracted, a file archived later in time will replace a
4531 file of the same name which was archived earlier, even though the
4532 older version of the file will remain in the archive unless you delete
4533 all versions of the file.
4535 Supposing you change the file @file{blues} and then append the changed
4536 version to @file{collection.tar}. As you saw above, the original
4537 @file{blues} is in the archive @file{collection.tar}. If you change the
4538 file and append the new version of the file to the archive, there will
4539 be two copies in the archive. When you extract the archive, the older
4540 version of the file will be extracted first, and then replaced by the
4541 newer version when it is extracted.
4543 You can append the new, changed copy of the file @file{blues} to the
4544 archive in this way:
4547 $ @kbd{tar --append --verbose --file=collection.tar blues}
4552 Because you specified the @option{--verbose} option, @command{tar} has
4553 printed the name of the file being appended as it was acted on. Now
4554 list the contents of the archive:
4557 $ @kbd{tar --list --verbose --file=collection.tar}
4558 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
4559 -rw-r--r-- me user 21 1996-09-23 16:44 blues
4560 -rw-r--r-- me user 20 1996-09-23 16:44 folk
4561 -rw-r--r-- me user 20 1996-09-23 16:44 rock
4562 -rw-r--r-- me user 58 1996-10-24 18:30 blues
4566 The newest version of @file{blues} is now at the end of the archive
4567 (note the different creation dates and file sizes). If you extract
4568 the archive, the older version of the file @file{blues} will be
4569 replaced by the newer version. You can confirm this by extracting
4570 the archive and running @samp{ls} on the directory.
4572 If you wish to extract the first occurrence of the file @file{blues}
4573 from the archive, use @option{--occurrence} option, as shown in
4574 the following example:
4577 $ @kbd{tar --extract -vv --occurrence --file=collection.tar blues}
4578 -rw-r--r-- me user 21 1996-09-23 16:44 blues
4581 @xref{Writing}, for more information on @option{--extract} and
4582 see @ref{Option Summary, --occurrence}, for a description of
4583 @option{--occurrence} option.
4586 @subsection Updating an Archive
4587 @cindex Updating an archive
4590 In the previous section, you learned how to use @option{--append} to
4591 add a file to an existing archive. A related operation is
4592 @option{--update} (@option{-u}). The @option{--update} operation
4593 updates a @command{tar} archive by comparing the date of the specified
4594 archive members against the date of the file with the same name. If
4595 the file has been modified more recently than the archive member, then
4596 the newer version of the file is added to the archive (as with
4599 Unfortunately, you cannot use @option{--update} with magnetic tape drives.
4600 The operation will fail.
4602 @FIXME{other examples of media on which --update will fail? need to ask
4603 charles and/or mib/thomas/dave shevett..}
4605 Both @option{--update} and @option{--append} work by adding to the end
4606 of the archive. When you extract a file from the archive, only the
4607 version stored last will wind up in the file system, unless you use
4608 the @option{--backup} option. @xref{multiple}, for a detailed discussion.
4615 @subsubsection How to Update an Archive Using @option{--update}
4618 You must use file name arguments with the @option{--update}
4619 (@option{-u}) operation. If you don't specify any files,
4620 @command{tar} won't act on any files and won't tell you that it didn't
4621 do anything (which may end up confusing you).
4623 @c note: the above parenthetical added because in fact, this
4624 @c behavior just confused the author. :-)
4626 To see the @option{--update} option at work, create a new file,
4627 @file{classical}, in your practice directory, and some extra text to the
4628 file @file{blues}, using any text editor. Then invoke @command{tar} with
4629 the @samp{update} operation and the @option{--verbose} (@option{-v})
4630 option specified, using the names of all the files in the @file{practice}
4631 directory as file name arguments:
4634 $ @kbd{tar --update -v -f collection.tar blues folk rock classical}
4641 Because we have specified verbose mode, @command{tar} prints out the names
4642 of the files it is working on, which in this case are the names of the
4643 files that needed to be updated. If you run @samp{tar --list} and look
4644 at the archive, you will see @file{blues} and @file{classical} at its
4645 end. There will be a total of two versions of the member @samp{blues};
4646 the one at the end will be newer and larger, since you added text before
4649 The reason @command{tar} does not overwrite the older file when updating
4650 it is because writing to the middle of a section of tape is a difficult
4651 process. Tapes are not designed to go backward. @xref{Media}, for more
4652 information about tapes.
4654 @option{--update} (@option{-u}) is not suitable for performing backups for two
4655 reasons: it does not change directory content entries, and it
4656 lengthens the archive every time it is used. The @GNUTAR{}
4657 options intended specifically for backups are more
4658 efficient. If you need to run backups, please consult @ref{Backups}.
4661 @subsection Combining Archives with @option{--concatenate}
4663 @cindex Adding archives to an archive
4664 @cindex Concatenating Archives
4665 @opindex concatenate
4667 @c @cindex @option{-A} described
4668 Sometimes it may be convenient to add a second archive onto the end of
4669 an archive rather than adding individual files to the archive. To add
4670 one or more archives to the end of another archive, you should use the
4671 @option{--concatenate} (@option{--catenate}, @option{-A}) operation.
4673 To use @option{--concatenate}, give the first archive with
4674 @option{--file} option and name the rest of archives to be
4675 concatenated on the command line. The members, and their member
4676 names, will be copied verbatim from those archives to the first
4677 one@footnote{This can cause multiple members to have the same name. For
4678 information on how this affects reading the archive, see @ref{multiple}.}.
4679 The new, concatenated archive will be called by the same name as the
4680 one given with the @option{--file} option. As usual, if you omit
4681 @option{--file}, @command{tar} will use the value of the environment
4682 variable @env{TAPE}, or, if this has not been set, the default archive name.
4684 @FIXME{There is no way to specify a new name...}
4686 To demonstrate how @option{--concatenate} works, create two small archives
4687 called @file{bluesrock.tar} and @file{folkjazz.tar}, using the relevant
4688 files from @file{practice}:
4691 $ @kbd{tar -cvf bluesrock.tar blues rock}
4694 $ @kbd{tar -cvf folkjazz.tar folk jazz}
4700 If you like, You can run @samp{tar --list} to make sure the archives
4701 contain what they are supposed to:
4704 $ @kbd{tar -tvf bluesrock.tar}
4705 -rw-r--r-- melissa user 105 1997-01-21 19:42 blues
4706 -rw-r--r-- melissa user 33 1997-01-20 15:34 rock
4707 $ @kbd{tar -tvf jazzfolk.tar}
4708 -rw-r--r-- melissa user 20 1996-09-23 16:44 folk
4709 -rw-r--r-- melissa user 65 1997-01-30 14:15 jazz
4712 We can concatenate these two archives with @command{tar}:
4716 $ @kbd{tar --concatenate --file=bluesrock.tar jazzfolk.tar}
4719 If you now list the contents of the @file{bluesrock.tar}, you will see
4720 that now it also contains the archive members of @file{jazzfolk.tar}:
4723 $ @kbd{tar --list --file=bluesrock.tar}
4730 When you use @option{--concatenate}, the source and target archives must
4731 already exist and must have been created using compatible format
4732 parameters. Notice, that @command{tar} does not check whether the
4733 archives it concatenates have compatible formats, it does not
4734 even check if the files are really tar archives.
4736 Like @option{--append} (@option{-r}), this operation cannot be performed on some
4737 tape drives, due to deficiencies in the formats those tape drives use.
4739 @cindex @code{concatenate} vs @command{cat}
4740 @cindex @command{cat} vs @code{concatenate}
4741 It may seem more intuitive to you to want or try to use @command{cat} to
4742 concatenate two archives instead of using the @option{--concatenate}
4743 operation; after all, @command{cat} is the utility for combining files.
4745 However, @command{tar} archives incorporate an end-of-file marker which
4746 must be removed if the concatenated archives are to be read properly as
4747 one archive. @option{--concatenate} removes the end-of-archive marker
4748 from the target archive before each new archive is appended. If you use
4749 @command{cat} to combine the archives, the result will not be a valid
4750 @command{tar} format archive. If you need to retrieve files from an
4751 archive that was added to using the @command{cat} utility, use the
4752 @option{--ignore-zeros} (@option{-i}) option. @xref{Ignore Zeros}, for further
4753 information on dealing with archives improperly combined using the
4754 @command{cat} shell utility.
4757 @subsection Removing Archive Members Using @option{--delete}
4758 @cindex Deleting files from an archive
4759 @cindex Removing files from an archive
4762 You can remove members from an archive by using the @option{--delete}
4763 option. Specify the name of the archive with @option{--file}
4764 (@option{-f}) and then specify the names of the members to be deleted;
4765 if you list no member names, nothing will be deleted. The
4766 @option{--verbose} option will cause @command{tar} to print the names
4767 of the members as they are deleted. As with @option{--extract}, you
4768 must give the exact member names when using @samp{tar --delete}.
4769 @option{--delete} will remove all versions of the named file from the
4770 archive. The @option{--delete} operation can run very slowly.
4772 Unlike other operations, @option{--delete} has no short form.
4774 @cindex Tapes, using @option{--delete} and
4775 @cindex Deleting from tape archives
4776 This operation will rewrite the archive. You can only use
4777 @option{--delete} on an archive if the archive device allows you to
4778 write to any point on the media, such as a disk; because of this, it
4779 does not work on magnetic tapes. Do not try to delete an archive member
4780 from a magnetic tape; the action will not succeed, and you will be
4781 likely to scramble the archive and damage your tape. There is no safe
4782 way (except by completely re-writing the archive) to delete files from
4783 most kinds of magnetic tape. @xref{Media}.
4785 To delete all versions of the file @file{blues} from the archive
4786 @file{collection.tar} in the @file{practice} directory, make sure you
4787 are in that directory, and then,
4790 $ @kbd{tar --list --file=collection.tar}
4795 $ @kbd{tar --delete --file=collection.tar blues}
4796 $ @kbd{tar --list --file=collection.tar}
4802 @FIXME{Check if the above listing is actually produced after running
4803 all the examples on collection.tar.}
4805 The @option{--delete} option has been reported to work properly when
4806 @command{tar} acts as a filter from @code{stdin} to @code{stdout}.
4809 @subsection Comparing Archive Members with the File System
4810 @cindex Verifying the currency of an archive
4813 The @option{--compare} (@option{-d}), or @option{--diff} operation compares
4814 specified archive members against files with the same names, and then
4815 reports differences in file size, mode, owner, modification date and
4816 contents. You should @emph{only} specify archive member names, not file
4817 names. If you do not name any members, then @command{tar} will compare the
4818 entire archive. If a file is represented in the archive but does not
4819 exist in the file system, @command{tar} reports a difference.
4821 You have to specify the record size of the archive when modifying an
4822 archive with a non-default record size.
4824 @command{tar} ignores files in the file system that do not have
4825 corresponding members in the archive.
4827 The following example compares the archive members @file{rock},
4828 @file{blues} and @file{funk} in the archive @file{bluesrock.tar} with
4829 files of the same name in the file system. (Note that there is no file,
4830 @file{funk}; @command{tar} will report an error message.)
4833 $ @kbd{tar --compare --file=bluesrock.tar rock blues funk}
4836 tar: funk not found in archive
4839 The spirit behind the @option{--compare} (@option{--diff},
4840 @option{-d}) option is to check whether the archive represents the
4841 current state of files on disk, more than validating the integrity of
4842 the archive media. For this latter goal, see @ref{verify}.
4844 @node create options
4845 @section Options Used by @option{--create}
4847 @xopindex{create, additional options}
4848 The previous chapter described the basics of how to use
4849 @option{--create} (@option{-c}) to create an archive from a set of files.
4850 @xref{create}. This section described advanced options to be used with
4854 * override:: Overriding File Metadata.
4855 * Ignore Failed Read::
4859 @subsection Overriding File Metadata
4861 As described above, a @command{tar} archive keeps, for each member it contains,
4862 its @dfn{metadata}, such as modification time, mode and ownership of
4863 the file. @GNUTAR{} allows to replace these data with other values
4864 when adding files to the archive. The options described in this
4865 section affect creation of archives of any type. For POSIX archives,
4866 see also @ref{PAX keywords}, for additional ways of controlling
4867 metadata, stored in the archive.
4871 @item --mode=@var{permissions}
4873 When adding files to an archive, @command{tar} will use
4874 @var{permissions} for the archive members, rather than the permissions
4875 from the files. @var{permissions} can be specified either as an octal
4876 number or as symbolic permissions, like with
4877 @command{chmod} (@xref{File permissions, Permissions, File
4878 permissions, fileutils, @acronym{GNU} file utilities}. This reference
4879 also has useful information for those not being overly familiar with
4880 the UNIX permission system). Using latter syntax allows for
4881 more flexibility. For example, the value @samp{a+rw} adds read and write
4882 permissions for everybody, while retaining executable bits on directories
4883 or on any other file already marked as executable:
4886 $ @kbd{tar -c -f archive.tar --mode='a+rw' .}
4889 @item --mtime=@var{date}
4892 When adding files to an archive, @command{tar} will use @var{date} as
4893 the modification time of members when creating archives, instead of
4894 their actual modification times. The argument @var{date} can be
4895 either a textual date representation in almost arbitrary format
4896 (@pxref{Date input formats}) or a name of an existing file, starting
4897 with @samp{/} or @samp{.}. In the latter case, the modification time
4898 of that file will be used.
4900 The following example will set the modification date to 00:00:00,
4904 $ @kbd{tar -c -f archive.tar --mtime='1970-01-01' .}
4908 When used with @option{--verbose} (@pxref{verbose tutorial}) @GNUTAR{}
4909 will try to convert the specified date back to its textual
4910 representation and compare it with the one given with
4911 @option{--mtime} options. If the two dates differ, @command{tar} will
4912 print a warning saying what date it will use. This is to help user
4913 ensure he is using the right date.
4918 $ @kbd{tar -c -f archive.tar -v --mtime=yesterday .}
4919 tar: Option --mtime: Treating date `yesterday' as 2006-06-20
4924 @item --owner=@var{user}
4927 Specifies that @command{tar} should use @var{user} as the owner of members
4928 when creating archives, instead of the user associated with the source
4929 file. The argument @var{user} can be either an existing user symbolic
4930 name, or a decimal numeric user @acronym{ID}.
4932 There is no value indicating a missing number, and @samp{0} usually means
4933 @code{root}. Some people like to force @samp{0} as the value to offer in
4934 their distributions for the owner of files, because the @code{root} user is
4935 anonymous anyway, so that might as well be the owner of anonymous
4936 archives. For example:
4939 $ @kbd{tar -c -f archive.tar --owner=0 .}
4946 $ @kbd{tar -c -f archive.tar --owner=root .}
4949 @item --group=@var{group}
4952 Files added to the @command{tar} archive will have a group @acronym{ID} of @var{group},
4953 rather than the group from the source file. The argument @var{group}
4954 can be either an existing group symbolic name, or a decimal numeric group @acronym{ID}.
4957 @node Ignore Failed Read
4958 @subsection Ignore Fail Read
4961 @item --ignore-failed-read
4962 @opindex ignore-failed-read
4963 Do not exit with nonzero on unreadable files or directories.
4966 @node extract options
4967 @section Options Used by @option{--extract}
4968 @cindex options for use with @option{--extract}
4970 @xopindex{extract, additional options}
4971 The previous chapter showed how to use @option{--extract} to extract
4972 an archive into the file system. Various options cause @command{tar} to
4973 extract more information than just file contents, such as the owner,
4974 the permissions, the modification date, and so forth. This section
4975 presents options to be used with @option{--extract} when certain special
4976 considerations arise. You may review the information presented in
4977 @ref{extract} for more basic information about the
4978 @option{--extract} operation.
4981 * Reading:: Options to Help Read Archives
4982 * Writing:: Changing How @command{tar} Writes Files
4983 * Scarce:: Coping with Scarce Resources
4987 @subsection Options to Help Read Archives
4988 @cindex Options when reading archives
4990 @cindex Reading incomplete records
4991 @cindex Records, incomplete
4992 @opindex read-full-records
4993 Normally, @command{tar} will request data in full record increments from
4994 an archive storage device. If the device cannot return a full record,
4995 @command{tar} will report an error. However, some devices do not always
4996 return full records, or do not require the last record of an archive to
4997 be padded out to the next record boundary. To keep reading until you
4998 obtain a full record, or to accept an incomplete record if it contains
4999 an end-of-archive marker, specify the @option{--read-full-records} (@option{-B}) option
5000 in conjunction with the @option{--extract} or @option{--list} operations.
5003 The @option{--read-full-records} (@option{-B}) option is turned on by default when
5004 @command{tar} reads an archive from standard input, or from a remote
5005 machine. This is because on @acronym{BSD} Unix systems, attempting to read a
5006 pipe returns however much happens to be in the pipe, even if it is
5007 less than was requested. If this option were not enabled, @command{tar}
5008 would fail as soon as it read an incomplete record from the pipe.
5010 If you're not sure of the blocking factor of an archive, you can
5011 read the archive by specifying @option{--read-full-records} (@option{-B}) and
5012 @option{--blocking-factor=@var{512-size}} (@option{-b
5013 @var{512-size}}), using a blocking factor larger than what the archive
5014 uses. This lets you avoid having to determine the blocking factor
5015 of an archive. @xref{Blocking Factor}.
5018 * read full records::
5022 @node read full records
5023 @unnumberedsubsubsec Reading Full Records
5025 @FIXME{need sentence or so of intro here}
5028 @opindex read-full-records
5029 @item --read-full-records
5031 Use in conjunction with @option{--extract} (@option{--get},
5032 @option{-x}) to read an archive which contains incomplete records, or
5033 one which has a blocking factor less than the one specified.
5037 @unnumberedsubsubsec Ignoring Blocks of Zeros
5039 @cindex End-of-archive blocks, ignoring
5040 @cindex Ignoring end-of-archive blocks
5041 @opindex ignore-zeros
5042 Normally, @command{tar} stops reading when it encounters a block of zeros
5043 between file entries (which usually indicates the end of the archive).
5044 @option{--ignore-zeros} (@option{-i}) allows @command{tar} to
5045 completely read an archive which contains a block of zeros before the
5046 end (i.e., a damaged archive, or one that was created by concatenating
5047 several archives together).
5049 The @option{--ignore-zeros} (@option{-i}) option is turned off by default because many
5050 versions of @command{tar} write garbage after the end-of-archive entry,
5051 since that part of the media is never supposed to be read. @GNUTAR{}
5052 does not write after the end of an archive, but seeks to
5053 maintain compatibility among archiving utilities.
5056 @item --ignore-zeros
5058 To ignore blocks of zeros (i.e., end-of-archive entries) which may be
5059 encountered while reading an archive. Use in conjunction with
5060 @option{--extract} or @option{--list}.
5064 @subsection Changing How @command{tar} Writes Files
5067 @FIXME{Introductory paragraph}
5070 * Dealing with Old Files::
5071 * Overwrite Old Files::
5073 * Keep Newer Files::
5075 * Recursive Unlink::
5076 * Data Modification Times::
5077 * Setting Access Permissions::
5078 * Directory Modification Times and Permissions::
5079 * Writing to Standard Output::
5080 * Writing to an External Program::
5084 @node Dealing with Old Files
5085 @unnumberedsubsubsec Options Controlling the Overwriting of Existing Files
5087 @xopindex{overwrite-dir, introduced}
5088 When extracting files, if @command{tar} discovers that the extracted
5089 file already exists, it normally replaces the file by removing it before
5090 extracting it, to prevent confusion in the presence of hard or symbolic
5091 links. (If the existing file is a symbolic link, it is removed, not
5092 followed.) However, if a directory cannot be removed because it is
5093 nonempty, @command{tar} normally overwrites its metadata (ownership,
5094 permission, etc.). The @option{--overwrite-dir} option enables this
5095 default behavior. To be more cautious and preserve the metadata of
5096 such a directory, use the @option{--no-overwrite-dir} option.
5098 @cindex Overwriting old files, prevention
5099 @xopindex{keep-old-files, introduced}
5100 To be even more cautious and prevent existing files from being replaced, use
5101 the @option{--keep-old-files} (@option{-k}) option. It causes @command{tar} to refuse
5102 to replace or update a file that already exists, i.e., a file with the
5103 same name as an archive member prevents extraction of that archive
5104 member. Instead, it reports an error.
5106 @xopindex{overwrite, introduced}
5107 To be more aggressive about altering existing files, use the
5108 @option{--overwrite} option. It causes @command{tar} to overwrite
5109 existing files and to follow existing symbolic links when extracting.
5111 @cindex Protecting old files
5112 Some people argue that @GNUTAR{} should not hesitate
5113 to overwrite files with other files when extracting. When extracting
5114 a @command{tar} archive, they expect to see a faithful copy of the
5115 state of the file system when the archive was created. It is debatable
5116 that this would always be a proper behavior. For example, suppose one
5117 has an archive in which @file{usr/local} is a link to
5118 @file{usr/local2}. Since then, maybe the site removed the link and
5119 renamed the whole hierarchy from @file{/usr/local2} to
5120 @file{/usr/local}. Such things happen all the time. I guess it would
5121 not be welcome at all that @GNUTAR{} removes the
5122 whole hierarchy just to make room for the link to be reinstated
5123 (unless it @emph{also} simultaneously restores the full
5124 @file{/usr/local2}, of course!) @GNUTAR{} is indeed
5125 able to remove a whole hierarchy to reestablish a symbolic link, for
5126 example, but @emph{only if} @option{--recursive-unlink} is specified
5127 to allow this behavior. In any case, single files are silently
5130 @xopindex{unlink-first, introduced}
5131 Finally, the @option{--unlink-first} (@option{-U}) option can improve performance in
5132 some cases by causing @command{tar} to remove files unconditionally
5133 before extracting them.
5135 @node Overwrite Old Files
5136 @unnumberedsubsubsec Overwrite Old Files
5141 Overwrite existing files and directory metadata when extracting files
5144 This causes @command{tar} to write extracted files into the file system without
5145 regard to the files already on the system; i.e., files with the same
5146 names as archive members are overwritten when the archive is extracted.
5147 It also causes @command{tar} to extract the ownership, permissions,
5148 and time stamps onto any preexisting files or directories.
5149 If the name of a corresponding file name is a symbolic link, the file
5150 pointed to by the symbolic link will be overwritten instead of the
5151 symbolic link itself (if this is possible). Moreover, special devices,
5152 empty directories and even symbolic links are automatically removed if
5153 they are in the way of extraction.
5155 Be careful when using the @option{--overwrite} option, particularly when
5156 combined with the @option{--absolute-names} (@option{-P}) option, as this combination
5157 can change the contents, ownership or permissions of any file on your
5158 system. Also, many systems do not take kindly to overwriting files that
5159 are currently being executed.
5161 @opindex overwrite-dir
5162 @item --overwrite-dir
5163 Overwrite the metadata of directories when extracting files from an
5164 archive, but remove other files before extracting.
5167 @node Keep Old Files
5168 @unnumberedsubsubsec Keep Old Files
5171 @opindex keep-old-files
5172 @item --keep-old-files
5174 Do not replace existing files from archive. The
5175 @option{--keep-old-files} (@option{-k}) option prevents @command{tar}
5176 from replacing existing files with files with the same name from the
5177 archive. The @option{--keep-old-files} option is meaningless with
5178 @option{--list} (@option{-t}). Prevents @command{tar} from replacing
5179 files in the file system during extraction.
5182 @node Keep Newer Files
5183 @unnumberedsubsubsec Keep Newer Files
5186 @opindex keep-newer-files
5187 @item --keep-newer-files
5188 Do not replace existing files that are newer than their archive
5189 copies. This option is meaningless with @option{--list} (@option{-t}).
5193 @unnumberedsubsubsec Unlink First
5196 @opindex unlink-first
5197 @item --unlink-first
5199 Remove files before extracting over them.
5200 This can make @command{tar} run a bit faster if you know in advance
5201 that the extracted files all need to be removed. Normally this option
5202 slows @command{tar} down slightly, so it is disabled by default.
5205 @node Recursive Unlink
5206 @unnumberedsubsubsec Recursive Unlink
5209 @opindex recursive-unlink
5210 @item --recursive-unlink
5211 When this option is specified, try removing files and directory hierarchies
5212 before extracting over them. @emph{This is a dangerous option!}
5215 If you specify the @option{--recursive-unlink} option,
5216 @command{tar} removes @emph{anything} that keeps you from extracting a file
5217 as far as current permissions will allow it. This could include removal
5218 of the contents of a full directory hierarchy.
5220 @node Data Modification Times
5221 @unnumberedsubsubsec Setting Data Modification Times
5223 @cindex Data modification times of extracted files
5224 @cindex Modification times of extracted files
5225 Normally, @command{tar} sets the data modification times of extracted
5226 files to the corresponding times recorded for the files in the archive, but
5227 limits the permissions of extracted files by the current @code{umask}
5230 To set the data modification times of extracted files to the time when
5231 the files were extracted, use the @option{--touch} (@option{-m}) option in
5232 conjunction with @option{--extract} (@option{--get}, @option{-x}).
5238 Sets the data modification time of extracted archive members to the time
5239 they were extracted, not the time recorded for them in the archive.
5240 Use in conjunction with @option{--extract} (@option{--get}, @option{-x}).
5243 @node Setting Access Permissions
5244 @unnumberedsubsubsec Setting Access Permissions
5246 @cindex Permissions of extracted files
5247 @cindex Modes of extracted files
5248 To set the modes (access permissions) of extracted files to those
5249 recorded for those files in the archive, use @option{--same-permissions}
5250 in conjunction with the @option{--extract} (@option{--get},
5251 @option{-x}) operation.
5254 @opindex preserve-permissions
5255 @opindex same-permissions
5256 @item --preserve-permissions
5257 @itemx --same-permissions
5258 @c @itemx --ignore-umask
5260 Set modes of extracted archive members to those recorded in the
5261 archive, instead of current umask settings. Use in conjunction with
5262 @option{--extract} (@option{--get}, @option{-x}).
5265 @node Directory Modification Times and Permissions
5266 @unnumberedsubsubsec Directory Modification Times and Permissions
5268 After successfully extracting a file member, @GNUTAR{} normally
5269 restores its permissions and modification times, as described in the
5270 previous sections. This cannot be done for directories, because
5271 after extracting a directory @command{tar} will almost certainly
5272 extract files into that directory and this will cause the directory
5273 modification time to be updated. Moreover, restoring that directory
5274 permissions may not permit file creation within it. Thus, restoring
5275 directory permissions and modification times must be delayed at least
5276 until all files have been extracted into that directory. @GNUTAR{}
5277 restores directories using the following approach.
5279 The extracted directories are created with the mode specified in the
5280 archive, as modified by the umask of the user, which gives sufficient
5281 permissions to allow file creation. The meta-information about the
5282 directory is recorded in the temporary list of directories. When
5283 preparing to extract next archive member, @GNUTAR{} checks if the
5284 directory prefix of this file contains the remembered directory. If
5285 it does not, the program assumes that all files have been extracted
5286 into that directory, restores its modification time and permissions
5287 and removes its entry from the internal list. This approach allows
5288 to correctly restore directory meta-information in the majority of
5289 cases, while keeping memory requirements sufficiently small. It is
5290 based on the fact, that most @command{tar} archives use the predefined
5291 order of members: first the directory, then all the files and
5292 subdirectories in that directory.
5294 However, this is not always true. The most important exception are
5295 incremental archives (@pxref{Incremental Dumps}). The member order in
5296 an incremental archive is reversed: first all directory members are
5297 stored, followed by other (non-directory) members. So, when extracting
5298 from incremental archives, @GNUTAR{} alters the above procedure. It
5299 remembers all restored directories, and restores their meta-data
5300 only after the entire archive has been processed. Notice, that you do
5301 not need to specify any special options for that, as @GNUTAR{}
5302 automatically detects archives in incremental format.
5304 There may be cases, when such processing is required for normal archives
5305 too. Consider the following example:
5309 $ @kbd{tar --no-recursion -cvf archive \
5310 foo foo/file1 bar bar/file foo/file2}
5319 During the normal operation, after encountering @file{bar}
5320 @GNUTAR{} will assume that all files from the directory @file{foo}
5321 were already extracted and will therefore restore its timestamp and
5322 permission bits. However, after extracting @file{foo/file2} the
5323 directory timestamp will be offset again.
5325 To correctly restore directory meta-information in such cases, use
5326 the @option{--delay-directory-restore} command line option:
5329 @opindex delay-directory-restore
5330 @item --delay-directory-restore
5331 Delays restoring of the modification times and permissions of extracted
5332 directories until the end of extraction. This way, correct
5333 meta-information is restored even if the archive has unusual member
5336 @opindex no-delay-directory-restore
5337 @item --no-delay-directory-restore
5338 Cancel the effect of the previous @option{--delay-directory-restore}.
5339 Use this option if you have used @option{--delay-directory-restore} in
5340 @env{TAR_OPTIONS} variable (@pxref{TAR_OPTIONS}) and wish to
5341 temporarily disable it.
5344 @node Writing to Standard Output
5345 @unnumberedsubsubsec Writing to Standard Output
5347 @cindex Writing extracted files to standard output
5348 @cindex Standard output, writing extracted files to
5349 To write the extracted files to the standard output, instead of
5350 creating the files on the file system, use @option{--to-stdout} (@option{-O}) in
5351 conjunction with @option{--extract} (@option{--get}, @option{-x}). This option is useful if you are
5352 extracting files to send them through a pipe, and do not need to
5353 preserve them in the file system. If you extract multiple members,
5354 they appear on standard output concatenated, in the order they are
5355 found in the archive.
5361 Writes files to the standard output. Use only in conjunction with
5362 @option{--extract} (@option{--get}, @option{-x}). When this option is
5363 used, instead of creating the files specified, @command{tar} writes
5364 the contents of the files extracted to its standard output. This may
5365 be useful if you are only extracting the files in order to send them
5366 through a pipe. This option is meaningless with @option{--list}
5370 This can be useful, for example, if you have a tar archive containing
5371 a big file and don't want to store the file on disk before processing
5372 it. You can use a command like this:
5375 tar -xOzf foo.tgz bigfile | process
5378 or even like this if you want to process the concatenation of the files:
5381 tar -xOzf foo.tgz bigfile1 bigfile2 | process
5384 However, @option{--to-command} may be more convenient for use with
5385 multiple files. See the next section.
5387 @node Writing to an External Program
5388 @unnumberedsubsubsec Writing to an External Program
5390 You can instruct @command{tar} to send the contents of each extracted
5391 file to the standard input of an external program:
5395 @item --to-command=@var{command}
5396 Extract files and pipe their contents to the standard input of
5397 @var{command}. When this option is used, instead of creating the
5398 files specified, @command{tar} invokes @var{command} and pipes the
5399 contents of the files to its standard output. The @var{command} may
5400 contain command line arguments. The program is executed via
5401 @code{sh -c}. Notice, that @var{command} is executed once for each regular file
5402 extracted. Non-regular files (directories, etc.) are ignored when this
5406 The command can obtain the information about the file it processes
5407 from the following environment variables:
5410 @vrindex TAR_FILETYPE, to-command environment
5412 Type of the file. It is a single letter with the following meaning:
5414 @multitable @columnfractions 0.10 0.90
5415 @item f @tab Regular file
5416 @item d @tab Directory
5417 @item l @tab Symbolic link
5418 @item h @tab Hard link
5419 @item b @tab Block device
5420 @item c @tab Character device
5423 Currently only regular files are supported.
5425 @vrindex TAR_MODE, to-command environment
5427 File mode, an octal number.
5429 @vrindex TAR_FILENAME, to-command environment
5431 The name of the file.
5433 @vrindex TAR_REALNAME, to-command environment
5435 Name of the file as stored in the archive.
5437 @vrindex TAR_UNAME, to-command environment
5439 Name of the file owner.
5441 @vrindex TAR_GNAME, to-command environment
5443 Name of the file owner group.
5445 @vrindex TAR_ATIME, to-command environment
5447 Time of last access. It is a decimal number, representing seconds
5448 since the Epoch. If the archive provides times with nanosecond
5449 precision, the nanoseconds are appended to the timestamp after a
5452 @vrindex TAR_MTIME, to-command environment
5454 Time of last modification.
5456 @vrindex TAR_CTIME, to-command environment
5458 Time of last status change.
5460 @vrindex TAR_SIZE, to-command environment
5464 @vrindex TAR_UID, to-command environment
5466 UID of the file owner.
5468 @vrindex TAR_GID, to-command environment
5470 GID of the file owner.
5473 Additionally, the following variables contain information about
5474 tar mode and the archive being processed:
5477 @vrindex TAR_VERSION, to-command environment
5479 @GNUTAR{} version number.
5481 @vrindex TAR_ARCHIVE, to-command environment
5483 The name of the archive @command{tar} is processing.
5485 @vrindex TAR_BLOCKING_FACTOR, to-command environment
5486 @item TAR_BLOCKING_FACTOR
5487 Current blocking factor (@pxref{Blocking}).
5489 @vrindex TAR_VOLUME, to-command environment
5491 Ordinal number of the volume @command{tar} is processing.
5493 @vrindex TAR_FORMAT, to-command environment
5495 Format of the archive being processed. @xref{Formats}, for a complete
5496 list of archive format names.
5499 If @var{command} exits with a non-0 status, @command{tar} will print
5500 an error message similar to the following:
5503 tar: 2345: Child returned status 1
5506 Here, @samp{2345} is the PID of the finished process.
5508 If this behavior is not wanted, use @option{--ignore-command-error}:
5511 @opindex ignore-command-error
5512 @item --ignore-command-error
5513 Ignore exit codes of subprocesses. Notice that if the program
5514 exits on signal or otherwise terminates abnormally, the error message
5515 will be printed even if this option is used.
5517 @opindex no-ignore-command-error
5518 @item --no-ignore-command-error
5519 Cancel the effect of any previous @option{--ignore-command-error}
5520 option. This option is useful if you have set
5521 @option{--ignore-command-error} in @env{TAR_OPTIONS}
5522 (@pxref{TAR_OPTIONS}) and wish to temporarily cancel it.
5526 @unnumberedsubsubsec Removing Files
5528 @FIXME{The section is too terse. Something more to add? An example,
5532 @opindex remove-files
5533 @item --remove-files
5534 Remove files after adding them to the archive.
5538 @subsection Coping with Scarce Resources
5541 @cindex Small memory
5542 @cindex Running out of space
5550 @unnumberedsubsubsec Starting File
5553 @opindex starting-file
5554 @item --starting-file=@var{name}
5555 @itemx -K @var{name}
5556 Starts an operation in the middle of an archive. Use in conjunction
5557 with @option{--extract} (@option{--get}, @option{-x}) or @option{--list} (@option{-t}).
5560 @cindex Middle of the archive, starting in the
5561 If a previous attempt to extract files failed due to lack of disk
5562 space, you can use @option{--starting-file=@var{name}} (@option{-K
5563 @var{name}}) to start extracting only after member @var{name} of the
5564 archive. This assumes, of course, that there is now free space, or
5565 that you are now extracting into a different file system. (You could
5566 also choose to suspend @command{tar}, remove unnecessary files from
5567 the file system, and then resume the same @command{tar} operation.
5568 In this case, @option{--starting-file} is not necessary.) See also
5569 @ref{interactive}, and @ref{exclude}.
5572 @unnumberedsubsubsec Same Order
5575 @cindex Large lists of file names on small machines
5577 @opindex preserve-order
5579 @itemx --preserve-order
5581 To process large lists of file names on machines with small amounts of
5582 memory. Use in conjunction with @option{--compare} (@option{--diff},
5583 @option{-d}), @option{--list} (@option{-t}) or @option{--extract}
5584 (@option{--get}, @option{-x}).
5587 The @option{--same-order} (@option{--preserve-order}, @option{-s}) option tells @command{tar} that the list of file
5588 names to be listed or extracted is sorted in the same order as the
5589 files in the archive. This allows a large list of names to be used,
5590 even on a small machine that would not otherwise be able to hold all
5591 the names in memory at the same time. Such a sorted list can easily be
5592 created by running @samp{tar -t} on the archive and editing its output.
5594 This option is probably never needed on modern computer systems.
5597 @section Backup options
5599 @cindex backup options
5601 @GNUTAR{} offers options for making backups of files
5602 before writing new versions. These options control the details of
5603 these backups. They may apply to the archive itself before it is
5604 created or rewritten, as well as individual extracted members. Other
5605 @acronym{GNU} programs (@command{cp}, @command{install}, @command{ln},
5606 and @command{mv}, for example) offer similar options.
5608 Backup options may prove unexpectedly useful when extracting archives
5609 containing many members having identical name, or when extracting archives
5610 on systems having file name limitations, making different members appear
5611 as having similar names through the side-effect of name truncation.
5612 @FIXME{This is true only if we have a good scheme for truncated backup names,
5613 which I'm not sure at all: I suspect work is needed in this area.}
5614 When any existing file is backed up before being overwritten by extraction,
5615 then clashing files are automatically be renamed to be unique, and the
5616 true name is kept for only the last file of a series of clashing files.
5617 By using verbose mode, users may track exactly what happens.
5619 At the detail level, some decisions are still experimental, and may
5620 change in the future, we are waiting comments from our users. So, please
5621 do not learn to depend blindly on the details of the backup features.
5622 For example, currently, directories themselves are never renamed through
5623 using these options, so, extracting a file over a directory still has
5624 good chances to fail. Also, backup options apply to created archives,
5625 not only to extracted members. For created archives, backups will not
5626 be attempted when the archive is a block or character device, or when it
5627 refers to a remote file.
5629 For the sake of simplicity and efficiency, backups are made by renaming old
5630 files prior to creation or extraction, and not by copying. The original
5631 name is restored if the file creation fails. If a failure occurs after a
5632 partial extraction of a file, both the backup and the partially extracted
5636 @item --backup[=@var{method}]
5638 @vindex VERSION_CONTROL
5640 Back up files that are about to be overwritten or removed.
5641 Without this option, the original versions are destroyed.
5643 Use @var{method} to determine the type of backups made.
5644 If @var{method} is not specified, use the value of the @env{VERSION_CONTROL}
5645 environment variable. And if @env{VERSION_CONTROL} is not set,
5646 use the @samp{existing} method.
5648 @vindex version-control @r{Emacs variable}
5649 This option corresponds to the Emacs variable @samp{version-control};
5650 the same values for @var{method} are accepted as in Emacs. This option
5651 also allows more descriptive names. The valid @var{method}s are:
5656 @cindex numbered @r{backup method}
5657 Always make numbered backups.
5661 @cindex existing @r{backup method}
5662 Make numbered backups of files that already have them, simple backups
5667 @cindex simple @r{backup method}
5668 Always make simple backups.
5672 @item --suffix=@var{suffix}
5674 @cindex backup suffix
5675 @vindex SIMPLE_BACKUP_SUFFIX
5676 Append @var{suffix} to each backup file made with @option{--backup}. If this
5677 option is not specified, the value of the @env{SIMPLE_BACKUP_SUFFIX}
5678 environment variable is used. And if @env{SIMPLE_BACKUP_SUFFIX} is not
5679 set, the default is @samp{~}, just as in Emacs.
5684 @section Notable @command{tar} Usages
5687 @FIXME{Using Unix file linking capability to recreate directory
5688 structures---linking files into one subdirectory and then
5689 @command{tar}ring that directory.}
5691 @FIXME{Nice hairy example using absolute-names, newer, etc.}
5694 You can easily use archive files to transport a group of files from
5695 one system to another: put all relevant files into an archive on one
5696 computer system, transfer the archive to another system, and extract
5697 the contents there. The basic transfer medium might be magnetic tape,
5698 Internet FTP, or even electronic mail (though you must encode the
5699 archive with @command{uuencode} in order to transport it properly by
5700 mail). Both machines do not have to use the same operating system, as
5701 long as they both support the @command{tar} program.
5703 For example, here is how you might copy a directory's contents from
5704 one disk to another, while preserving the dates, modes, owners and
5705 link-structure of all the files therein. In this case, the transfer
5706 medium is a @dfn{pipe}:
5709 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xf -)}
5713 You can avoid subshells by using @option{-C} option:
5716 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xf -}
5720 The command also works using long option forms:
5724 $ @kbd{(cd sourcedir; tar --create --file=- . ) \
5725 | (cd targetdir; tar --extract --file=-)}
5734 $ @kbd{tar --directory sourcedir --create --file=- . \
5735 | tar --directory targetdir --extract --file=-}
5740 This is one of the easiest methods to transfer a @command{tar} archive.
5743 @section Looking Ahead: The Rest of this Manual
5745 You have now seen how to use all eight of the operations available to
5746 @command{tar}, and a number of the possible options. The next chapter
5747 explains how to choose and change file and archive names, how to use
5748 files to store names of other files which you can then call as
5749 arguments to @command{tar} (this can help you save time if you expect to
5750 archive the same list of files a number of times), and so forth.
5751 @FIXME{in case it's not obvious, i'm making this up in some sense
5752 based on my limited memory of what the next chapter *really* does. i
5753 just wanted to flesh out this final section a little bit so i'd
5754 remember to stick it in here. :-)}
5756 If there are too many files to conveniently list on the command line,
5757 you can list the names in a file, and @command{tar} will read that file.
5760 There are various ways of causing @command{tar} to skip over some files,
5761 and not archive them. @xref{Choosing}.
5764 @chapter Performing Backups and Restoring Files
5767 @GNUTAR{} is distributed along with the scripts for performing backups
5768 and restores. Even if there is a good chance those scripts may be
5769 satisfying to you, they are not the only scripts or methods available for doing
5770 backups and restore. You may well create your own, or use more
5771 sophisticated packages dedicated to that purpose.
5773 Some users are enthusiastic about @code{Amanda} (The Advanced Maryland
5774 Automatic Network Disk Archiver), a backup system developed by James
5775 da Silva @file{jds@@cs.umd.edu} and available on many Unix systems.
5776 This is free software, and it is available from @uref{http://www.amanda.org}.
5780 Here is a possible plan for a future documentation about the backuping
5781 scripts which are provided within the @GNUTAR{}
5787 @item what are dumps
5788 @item different levels of dumps
5790 @item full dump = dump everything
5791 @item level 1, level 2 dumps etc
5792 A level @var{n} dump dumps everything changed since the last level
5795 @item how to use scripts for dumps (ie, the concept)
5797 @item scripts to run after editing backup specs (details)
5799 @item Backup Specs, what is it.
5801 @item how to customize
5802 @item actual text of script [/sp/dump/backup-specs]
5806 @item rsh doesn't work
5807 @item rtape isn't installed
5810 @item the @option{--incremental} option of tar
5813 @item write protection
5814 @item types of media, different sizes and types, useful for different things
5815 @item files and tape marks
5816 one tape mark between files, two at end.
5817 @item positioning the tape
5818 MT writes two at end of write,
5819 backspaces over one when writing again.
5825 This chapter documents both the provided shell scripts and @command{tar}
5826 options which are more specific to usage as a backup tool.
5828 To @dfn{back up} a file system means to create archives that contain
5829 all the files in that file system. Those archives can then be used to
5830 restore any or all of those files (for instance if a disk crashes or a
5831 file is accidentally deleted). File system @dfn{backups} are also
5835 * Full Dumps:: Using @command{tar} to Perform Full Dumps
5836 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
5837 * Backup Levels:: Levels of Backups
5838 * Backup Parameters:: Setting Parameters for Backups and Restoration
5839 * Scripted Backups:: Using the Backup Scripts
5840 * Scripted Restoration:: Using the Restore Script
5844 @section Using @command{tar} to Perform Full Dumps
5850 @cindex corrupted archives
5851 Full dumps should only be made when no other people or programs
5852 are modifying files in the file system. If files are modified while
5853 @command{tar} is making the backup, they may not be stored properly in
5854 the archive, in which case you won't be able to restore them if you
5855 have to. (Files not being modified are written with no trouble, and do
5856 not corrupt the entire archive.)
5858 You will want to use the @option{--label=@var{archive-label}}
5859 (@option{-V @var{archive-label}}) option to give the archive a
5860 volume label, so you can tell what this archive is even if the label
5861 falls off the tape, or anything like that.
5863 Unless the file system you are dumping is guaranteed to fit on
5864 one volume, you will need to use the @option{--multi-volume} (@option{-M}) option.
5865 Make sure you have enough tapes on hand to complete the backup.
5867 If you want to dump each file system separately you will need to use
5868 the @option{--one-file-system} option to prevent
5869 @command{tar} from crossing file system boundaries when storing
5872 The @option{--incremental} (@option{-G}) (@pxref{Incremental Dumps})
5873 option is not needed, since this is a complete copy of everything in
5874 the file system, and a full restore from this backup would only be
5875 done onto a completely
5878 Unless you are in a hurry, and trust the @command{tar} program (and your
5879 tapes), it is a good idea to use the @option{--verify} (@option{-W})
5880 option, to make sure your files really made it onto the dump properly.
5881 This will also detect cases where the file was modified while (or just
5882 after) it was being archived. Not all media (notably cartridge tapes)
5883 are capable of being verified, unfortunately.
5885 @node Incremental Dumps
5886 @section Using @command{tar} to Perform Incremental Dumps
5888 @dfn{Incremental backup} is a special form of @GNUTAR{} archive that
5889 stores additional metadata so that exact state of the file system
5890 can be restored when extracting the archive.
5892 @GNUTAR{} currently offers two options for handling incremental
5893 backups: @option{--listed-incremental=@var{snapshot-file}} (@option{-g
5894 @var{snapshot-file}}) and @option{--incremental} (@option{-G}).
5896 @xopindex{listed-incremental, described}
5897 The option @option{--listed-incremental} instructs tar to operate on
5898 an incremental archive with additional metadata stored in a standalone
5899 file, called a @dfn{snapshot file}. The purpose of this file is to help
5900 determine which files have been changed, added or deleted since the
5901 last backup, so that the next incremental backup will contain only
5902 modified files. The name of the snapshot file is given as an argument
5906 @item --listed-incremental=@var{file}
5907 @itemx -g @var{file}
5908 Handle incremental backups with snapshot data in @var{file}.
5911 To create an incremental backup, you would use
5912 @option{--listed-incremental} together with @option{--create}
5913 (@pxref{create}). For example:
5916 $ @kbd{tar --create \
5917 --file=archive.1.tar \
5918 --listed-incremental=/var/log/usr.snar \
5922 This will create in @file{archive.1.tar} an incremental backup of
5923 the @file{/usr} file system, storing additional metadata in the file
5924 @file{/var/log/usr.snar}. If this file does not exist, it will be
5925 created. The created archive will then be a @dfn{level 0 backup};
5926 please see the next section for more on backup levels.
5928 Otherwise, if the file @file{/var/log/usr.snar} exists, it
5929 determines which files are modified. In this case only these files will be
5930 stored in the archive. Suppose, for example, that after running the
5931 above command, you delete file @file{/usr/doc/old} and create
5932 directory @file{/usr/local/db} with the following contents:
5935 $ @kbd{ls /usr/local/db}
5940 Some time later you create another incremental backup. You will
5944 $ @kbd{tar --create \
5945 --file=archive.2.tar \
5946 --listed-incremental=/var/log/usr.snar \
5948 tar: usr/local/db: Directory is new
5955 The created archive @file{archive.2.tar} will contain only these
5956 three members. This archive is called a @dfn{level 1 backup}. Notice
5957 that @file{/var/log/usr.snar} will be updated with the new data, so if
5958 you plan to create more @samp{level 1} backups, it is necessary to
5959 create a working copy of the snapshot file before running
5960 @command{tar}. The above example will then be modified as follows:
5963 $ @kbd{cp /var/log/usr.snar /var/log/usr.snar-1}
5964 $ @kbd{tar --create \
5965 --file=archive.2.tar \
5966 --listed-incremental=/var/log/usr.snar-1 \
5971 @xopindex{level, described}
5972 You can force @samp{level 0} backups either by removing the snapshot
5973 file before running @command{tar}, or by supplying the
5974 @option{--level=0} option, e.g.:
5977 $ @kbd{tar --create \
5978 --file=archive.2.tar \
5979 --listed-incremental=/var/log/usr.snar-0 \
5984 Incremental dumps depend crucially on time stamps, so the results are
5985 unreliable if you modify a file's time stamps during dumping (e.g.,
5986 with the @option{--atime-preserve=replace} option), or if you set the clock
5989 @anchor{device numbers}
5990 @cindex Device numbers, using in incremental backups
5991 Metadata stored in snapshot files include device numbers, which,
5992 obviously are supposed to be non-volatile values. However, it turns
5993 out that @acronym{NFS} devices have undependable values when an automounter
5994 gets in the picture. This can lead to a great deal of spurious
5995 redumping in incremental dumps, so it is somewhat useless to compare
5996 two @acronym{NFS} devices numbers over time. The solution implemented
5997 currently is to consider all @acronym{NFS} devices as being equal
5998 when it comes to comparing directories; this is fairly gross, but
5999 there does not seem to be a better way to go.
6001 Apart from using @acronym{NFS}, there are a number of cases where
6002 relying on device numbers can cause spurious redumping of unmodified
6003 files. For example, this occurs when archiving @acronym{LVM} snapshot
6004 volumes. To avoid this, use @option{--no-check-device} option:
6007 @xopindex{no-check-device, described}
6008 @item --no-check-device
6009 Do not rely on device numbers when preparing a list of changed files
6010 for an incremental dump.
6012 @xopindex{check-device, described}
6013 @item --check-device
6014 Use device numbers when preparing a list of changed files
6015 for an incremental dump. This is the default behavior. The purpose
6016 of this option is to undo the effect of the @option{--no-check-device}
6017 if it was given in @env{TAR_OPTIONS} environment variable
6018 (@pxref{TAR_OPTIONS}).
6021 There is also another way to cope with changing device numbers. It is
6022 described in detail in @ref{Fixing Snapshot Files}.
6024 Note that incremental archives use @command{tar} extensions and may
6025 not be readable by non-@acronym{GNU} versions of the @command{tar} program.
6027 @xopindex{listed-incremental, using with @option{--extract}}
6028 @xopindex{extract, using with @option{--listed-incremental}}
6029 To extract from the incremental dumps, use
6030 @option{--listed-incremental} together with @option{--extract}
6031 option (@pxref{extracting files}). In this case, @command{tar} does
6032 not need to access snapshot file, since all the data necessary for
6033 extraction are stored in the archive itself. So, when extracting, you
6034 can give whatever argument to @option{--listed-incremental}, the usual
6035 practice is to use @option{--listed-incremental=/dev/null}.
6036 Alternatively, you can use @option{--incremental}, which needs no
6037 arguments. In general, @option{--incremental} (@option{-G}) can be
6038 used as a shortcut for @option{--listed-incremental} when listing or
6039 extracting incremental backups (for more information regarding this
6040 option, @pxref{incremental-op}).
6042 When extracting from the incremental backup @GNUTAR{} attempts to
6043 restore the exact state the file system had when the archive was
6044 created. In particular, it will @emph{delete} those files in the file
6045 system that did not exist in their directories when the archive was
6046 created. If you have created several levels of incremental files,
6047 then in order to restore the exact contents the file system had when
6048 the last level was created, you will need to restore from all backups
6049 in turn. Continuing our example, to restore the state of @file{/usr}
6050 file system, one would do@footnote{Notice, that since both archives
6051 were created without @option{-P} option (@pxref{absolute}), these
6052 commands should be run from the root file system.}:
6055 $ @kbd{tar --extract \
6056 --listed-incremental=/dev/null \
6057 --file archive.1.tar}
6058 $ @kbd{tar --extract \
6059 --listed-incremental=/dev/null \
6060 --file archive.2.tar}
6063 To list the contents of an incremental archive, use @option{--list}
6064 (@pxref{list}), as usual. To obtain more information about the
6065 archive, use @option{--listed-incremental} or @option{--incremental}
6066 combined with two @option{--verbose} options@footnote{Two
6067 @option{--verbose} options were selected to avoid breaking usual
6068 verbose listing output (@option{--list --verbose}) when using in
6071 @xopindex{incremental, using with @option{--list}}
6072 @xopindex{listed-incremental, using with @option{--list}}
6073 @xopindex{list, using with @option{--incremental}}
6074 @xopindex{list, using with @option{--listed-incremental}}
6075 Versions of @GNUTAR{} up to 1.15.1 used to dump verbatim binary
6076 contents of the DUMPDIR header (with terminating nulls) when
6077 @option{--incremental} or @option{--listed-incremental} option was
6078 given, no matter what the verbosity level. This behavior, and,
6079 especially, the binary output it produced were considered inconvenient
6080 and were changed in version 1.16.}:
6083 @kbd{tar --list --incremental --verbose --verbose archive.tar}
6086 This command will print, for each directory in the archive, the list
6087 of files in that directory at the time the archive was created. This
6088 information is put out in a format which is both human-readable and
6089 unambiguous for a program: each file name is printed as
6096 where @var{x} is a letter describing the status of the file: @samp{Y}
6097 if the file is present in the archive, @samp{N} if the file is not
6098 included in the archive, or a @samp{D} if the file is a directory (and
6099 is included in the archive). @xref{Dumpdir}, for the detailed
6100 description of dumpdirs and status codes. Each such
6101 line is terminated by a newline character. The last line is followed
6102 by an additional newline to indicate the end of the data.
6104 @anchor{incremental-op}The option @option{--incremental} (@option{-G})
6105 gives the same behavior as @option{--listed-incremental} when used
6106 with @option{--list} and @option{--extract} options. When used with
6107 @option{--create} option, it creates an incremental archive without
6108 creating snapshot file. Thus, it is impossible to create several
6109 levels of incremental backups with @option{--incremental} option.
6112 @section Levels of Backups
6114 An archive containing all the files in the file system is called a
6115 @dfn{full backup} or @dfn{full dump}. You could insure your data by
6116 creating a full dump every day. This strategy, however, would waste a
6117 substantial amount of archive media and user time, as unchanged files
6118 are daily re-archived.
6120 It is more efficient to do a full dump only occasionally. To back up
6121 files between full dumps, you can use @dfn{incremental dumps}. A @dfn{level
6122 one} dump archives all the files that have changed since the last full
6125 A typical dump strategy would be to perform a full dump once a week,
6126 and a level one dump once a day. This means some versions of files
6127 will in fact be archived more than once, but this dump strategy makes
6128 it possible to restore a file system to within one day of accuracy by
6129 only extracting two archives---the last weekly (full) dump and the
6130 last daily (level one) dump. The only information lost would be in
6131 files changed or created since the last daily backup. (Doing dumps
6132 more than once a day is usually not worth the trouble.)
6134 @GNUTAR{} comes with scripts you can use to do full
6135 and level-one (actually, even level-two and so on) dumps. Using
6136 scripts (shell programs) to perform backups and restoration is a
6137 convenient and reliable alternative to typing out file name lists
6138 and @command{tar} commands by hand.
6140 Before you use these scripts, you need to edit the file
6141 @file{backup-specs}, which specifies parameters used by the backup
6142 scripts and by the restore script. This file is usually located
6143 in @file{/etc/backup} directory. @xref{Backup Parameters}, for its
6144 detailed description. Once the backup parameters are set, you can
6145 perform backups or restoration by running the appropriate script.
6147 The name of the backup script is @code{backup}. The name of the
6148 restore script is @code{restore}. The following sections describe
6149 their use in detail.
6151 @emph{Please Note:} The backup and restoration scripts are
6152 designed to be used together. While it is possible to restore files by
6153 hand from an archive which was created using a backup script, and to create
6154 an archive by hand which could then be extracted using the restore script,
6155 it is easier to use the scripts. @xref{Incremental Dumps}, before
6156 making such an attempt.
6158 @node Backup Parameters
6159 @section Setting Parameters for Backups and Restoration
6161 The file @file{backup-specs} specifies backup parameters for the
6162 backup and restoration scripts provided with @command{tar}. You must
6163 edit @file{backup-specs} to fit your system configuration and schedule
6164 before using these scripts.
6166 Syntactically, @file{backup-specs} is a shell script, containing
6167 mainly variable assignments. However, any valid shell construct
6168 is allowed in this file. Particularly, you may wish to define
6169 functions within that script (e.g., see @code{RESTORE_BEGIN} below).
6170 For more information about shell script syntax, please refer to
6171 @url{http://www.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html#ta
6172 g_02, the definition of the Shell Command Language}. See also
6173 @ref{Top,,Bash Features,bashref,Bash Reference Manual}.
6175 The shell variables controlling behavior of @code{backup} and
6176 @code{restore} are described in the following subsections.
6179 * General-Purpose Variables::
6180 * Magnetic Tape Control::
6182 * backup-specs example:: An Example Text of @file{Backup-specs}
6185 @node General-Purpose Variables
6186 @subsection General-Purpose Variables
6188 @defvr {Backup variable} ADMINISTRATOR
6189 The user name of the backup administrator. @code{Backup} scripts
6190 sends a backup report to this address.
6193 @defvr {Backup variable} BACKUP_HOUR
6194 The hour at which the backups are done. This can be a number from 0
6195 to 23, or the time specification in form @var{hours}:@var{minutes},
6196 or the string @samp{now}.
6198 This variable is used by @code{backup}. Its value may be overridden
6199 using @option{--time} option (@pxref{Scripted Backups}).
6202 @defvr {Backup variable} TAPE_FILE
6204 The device @command{tar} writes the archive to. If @var{TAPE_FILE}
6205 is a remote archive (@pxref{remote-dev}), backup script will suppose
6206 that your @command{mt} is able to access remote devices. If @var{RSH}
6207 (@pxref{RSH}) is set, @option{--rsh-command} option will be added to
6208 invocations of @command{mt}.
6211 @defvr {Backup variable} BLOCKING
6213 The blocking factor @command{tar} will use when writing the dump archive.
6214 @xref{Blocking Factor}.
6217 @defvr {Backup variable} BACKUP_DIRS
6219 A list of file systems to be dumped (for @code{backup}), or restored
6220 (for @code{restore}). You can include any directory
6221 name in the list --- subdirectories on that file system will be
6222 included, regardless of how they may look to other networked machines.
6223 Subdirectories on other file systems will be ignored.
6225 The host name specifies which host to run @command{tar} on, and should
6226 normally be the host that actually contains the file system. However,
6227 the host machine must have @GNUTAR{} installed, and
6228 must be able to access the directory containing the backup scripts and
6229 their support files using the same file name that is used on the
6230 machine where the scripts are run (i.e., what @command{pwd} will print
6231 when in that directory on that machine). If the host that contains
6232 the file system does not have this capability, you can specify another
6233 host as long as it can access the file system through @acronym{NFS}.
6235 If the list of file systems is very long you may wish to put it
6236 in a separate file. This file is usually named
6237 @file{/etc/backup/dirs}, but this name may be overridden in
6238 @file{backup-specs} using @code{DIRLIST} variable.
6241 @defvr {Backup variable} DIRLIST
6243 The name of the file that contains a list of file systems to backup
6244 or restore. By default it is @file{/etc/backup/dirs}.
6247 @defvr {Backup variable} BACKUP_FILES
6249 A list of individual files to be dumped (for @code{backup}), or restored
6250 (for @code{restore}). These should be accessible from the machine on
6251 which the backup script is run.
6253 If the list of individual files is very long you may wish to store it
6254 in a separate file. This file is usually named
6255 @file{/etc/backup/files}, but this name may be overridden in
6256 @file{backup-specs} using @code{FILELIST} variable.
6259 @defvr {Backup variable} FILELIST
6261 The name of the file that contains a list of individual files to backup
6262 or restore. By default it is @file{/etc/backup/files}.
6265 @defvr {Backup variable} MT
6267 Full file name of @command{mt} binary.
6270 @defvr {Backup variable} RSH
6272 Full file name of @command{rsh} binary or its equivalent. You may wish to
6273 set it to @code{ssh}, to improve security. In this case you will have
6274 to use public key authentication.
6277 @defvr {Backup variable} RSH_COMMAND
6279 Full file name of @command{rsh} binary on remote machines. This will
6280 be passed via @option{--rsh-command} option to the remote invocation
6284 @defvr {Backup variable} VOLNO_FILE
6286 Name of temporary file to hold volume numbers. This needs to be accessible
6287 by all the machines which have file systems to be dumped.
6290 @defvr {Backup variable} XLIST
6292 Name of @dfn{exclude file list}. An @dfn{exclude file list} is a file
6293 located on the remote machine and containing the list of files to
6294 be excluded from the backup. Exclude file lists are searched in
6295 /etc/tar-backup directory. A common use for exclude file lists
6296 is to exclude files containing security-sensitive information
6297 (e.g., @file{/etc/shadow} from backups).
6299 This variable affects only @code{backup}.
6302 @defvr {Backup variable} SLEEP_TIME
6304 Time to sleep between dumps of any two successive file systems
6306 This variable affects only @code{backup}.
6309 @defvr {Backup variable} DUMP_REMIND_SCRIPT
6311 Script to be run when it's time to insert a new tape in for the next
6312 volume. Administrators may want to tailor this script for their site.
6313 If this variable isn't set, @GNUTAR{} will display its built-in
6314 prompt, and will expect confirmation from the console. For the
6315 description of the default prompt, see @ref{change volume prompt}.
6319 @defvr {Backup variable} SLEEP_MESSAGE
6321 Message to display on the terminal while waiting for dump time. Usually
6322 this will just be some literal text.
6325 @defvr {Backup variable} TAR
6327 Full file name of the @GNUTAR{} executable. If this is not set, backup
6328 scripts will search @command{tar} in the current shell path.
6331 @node Magnetic Tape Control
6332 @subsection Magnetic Tape Control
6334 Backup scripts access tape device using special @dfn{hook functions}.
6335 These functions take a single argument --- the name of the tape
6336 device. Their names are kept in the following variables:
6338 @defvr {Backup variable} MT_BEGIN
6339 The name of @dfn{begin} function. This function is called before
6340 accessing the drive. By default it retensions the tape:
6346 mt -f "$1" retension
6351 @defvr {Backup variable} MT_REWIND
6352 The name of @dfn{rewind} function. The default definition is as
6365 @defvr {Backup variable} MT_OFFLINE
6366 The name of the function switching the tape off line. By default
6367 it is defined as follows:
6370 MT_OFFLINE=mt_offline
6378 @defvr {Backup variable} MT_STATUS
6379 The name of the function used to obtain the status of the archive device,
6380 including error count. Default definition:
6392 @subsection User Hooks
6394 @dfn{User hooks} are shell functions executed before and after
6395 each @command{tar} invocation. Thus, there are @dfn{backup
6396 hooks}, which are executed before and after dumping each file
6397 system, and @dfn{restore hooks}, executed before and
6398 after restoring a file system. Each user hook is a shell function
6399 taking four arguments:
6401 @deffn {User Hook Function} hook @var{level} @var{host} @var{fs} @var{fsname}
6406 Current backup or restore level.
6409 Name or IP address of the host machine being dumped or restored.
6412 Full file name of the file system being dumped or restored.
6415 File system name with directory separators replaced with colons. This
6416 is useful, e.g., for creating unique files.
6420 Following variables keep the names of user hook functions:
6422 @defvr {Backup variable} DUMP_BEGIN
6423 Dump begin function. It is executed before dumping the file system.
6426 @defvr {Backup variable} DUMP_END
6427 Executed after dumping the file system.
6430 @defvr {Backup variable} RESTORE_BEGIN
6431 Executed before restoring the file system.
6434 @defvr {Backup variable} RESTORE_END
6435 Executed after restoring the file system.
6438 @node backup-specs example
6439 @subsection An Example Text of @file{Backup-specs}
6441 The following is an example of @file{backup-specs}:
6444 # site-specific parameters for file system backup.
6446 ADMINISTRATOR=friedman
6448 TAPE_FILE=/dev/nrsmt0
6450 # Use @code{ssh} instead of the less secure @code{rsh}
6452 RSH_COMMAND=/usr/bin/ssh
6454 # Override MT_STATUS function:
6460 # Disable MT_OFFLINE function
6477 apple-gunkies:/com/mailer/gnu
6478 apple-gunkies:/com/archive/gnu"
6480 BACKUP_FILES="/com/mailer/aliases /com/mailer/league*[a-z]"
6484 @node Scripted Backups
6485 @section Using the Backup Scripts
6487 The syntax for running a backup script is:
6490 backup --level=@var{level} --time=@var{time}
6493 The @option{--level} option requests the dump level. Thus, to produce
6494 a full dump, specify @code{--level=0} (this is the default, so
6495 @option{--level} may be omitted if its value is
6496 @code{0})@footnote{For backward compatibility, the @code{backup} will also
6497 try to deduce the requested dump level from the name of the
6498 script itself. If the name consists of a string @samp{level-}
6499 followed by a single decimal digit, that digit is taken as
6500 the dump level number. Thus, you may create a link from @code{backup}
6501 to @code{level-1} and then run @code{level-1} whenever you need to
6502 create a level one dump.}.
6504 The @option{--time} option determines when should the backup be
6505 run. @var{Time} may take three forms:
6508 @item @var{hh}:@var{mm}
6510 The dump must be run at @var{hh} hours @var{mm} minutes.
6514 The dump must be run at @var{hh} hours.
6518 The dump must be run immediately.
6521 You should start a script with a tape or disk mounted. Once you
6522 start a script, it prompts you for new tapes or disks as it
6523 needs them. Media volumes don't have to correspond to archive
6524 files --- a multi-volume archive can be started in the middle of a
6525 tape that already contains the end of another multi-volume archive.
6526 The @code{restore} script prompts for media by its archive volume,
6527 so to avoid an error message you should keep track of which tape
6528 (or disk) contains which volume of the archive (@pxref{Scripted
6531 The backup scripts write two files on the file system. The first is a
6532 record file in @file{/etc/tar-backup/}, which is used by the scripts
6533 to store and retrieve information about which files were dumped. This
6534 file is not meant to be read by humans, and should not be deleted by
6535 them. @xref{Snapshot Files}, for a more detailed explanation of this
6538 The second file is a log file containing the names of the file systems
6539 and files dumped, what time the backup was made, and any error
6540 messages that were generated, as well as how much space was left in
6541 the media volume after the last volume of the archive was written.
6542 You should check this log file after every backup. The file name is
6543 @file{log-@var{mm-dd-yyyy}-level-@var{n}}, where @var{mm-dd-yyyy}
6544 represents current date, and @var{n} represents current dump level number.
6546 The script also prints the name of each system being dumped to the
6549 Following is the full list of options accepted by @code{backup}
6553 @item -l @var{level}
6554 @itemx --level=@var{level}
6555 Do backup level @var{level} (default 0).
6559 Force backup even if today's log file already exists.
6561 @item -v[@var{level}]
6562 @itemx --verbose[=@var{level}]
6563 Set verbosity level. The higher the level is, the more debugging
6564 information will be output during execution. Default @var{level}
6565 is 100, which means the highest debugging level.
6567 @item -t @var{start-time}
6568 @itemx --time=@var{start-time}
6569 Wait till @var{time}, then do backup.
6573 Display short help message and exit.
6577 Display information about the program's name, version, origin and legal
6578 status, all on standard output, and then exit successfully.
6582 @node Scripted Restoration
6583 @section Using the Restore Script
6585 To restore files that were archived using a scripted backup, use the
6586 @code{restore} script. Its usage is quite straightforward. In the
6587 simplest form, invoke @code{restore --all}, it will
6588 then restore all the file systems and files specified in
6589 @file{backup-specs} (@pxref{General-Purpose Variables,BACKUP_DIRS}).
6591 You may select the file systems (and/or files) to restore by
6592 giving @code{restore} a list of @dfn{patterns} in its command
6593 line. For example, running
6600 will restore all file systems on the machine @samp{albert}. A more
6601 complicated example:
6604 restore 'albert:*' '*:/var'
6608 This command will restore all file systems on the machine @samp{albert}
6609 as well as @file{/var} file system on all machines.
6611 By default @code{restore} will start restoring files from the lowest
6612 available dump level (usually zero) and will continue through
6613 all available dump levels. There may be situations where such a
6614 thorough restore is not necessary. For example, you may wish to
6615 restore only files from the recent level one backup. To do so,
6616 use @option{--level} option, as shown in the example below:
6622 The full list of options accepted by @code{restore} follows:
6627 Restore all file systems and files specified in @file{backup-specs}.
6629 @item -l @var{level}
6630 @itemx --level=@var{level}
6631 Start restoring from the given backup level, instead of the default 0.
6633 @item -v[@var{level}]
6634 @itemx --verbose[=@var{level}]
6635 Set verbosity level. The higher the level is, the more debugging
6636 information will be output during execution. Default @var{level}
6637 is 100, which means the highest debugging level.
6641 Display short help message and exit.
6645 Display information about the program's name, version, origin and legal
6646 status, all on standard output, and then exit successfully.
6649 You should start the restore script with the media containing the
6650 first volume of the archive mounted. The script will prompt for other
6651 volumes as they are needed. If the archive is on tape, you don't need
6652 to rewind the tape to to its beginning---if the tape head is
6653 positioned past the beginning of the archive, the script will rewind
6654 the tape as needed. @xref{Tape Positioning}, for a discussion of tape
6658 @strong{Warning:} The script will delete files from the active file
6659 system if they were not in the file system when the archive was made.
6662 @xref{Incremental Dumps}, for an explanation of how the script makes
6666 @chapter Choosing Files and Names for @command{tar}
6668 Certain options to @command{tar} enable you to specify a name for your
6669 archive. Other options let you decide which files to include or exclude
6670 from the archive, based on when or whether files were modified, whether
6671 the file names do or don't match specified patterns, or whether files
6672 are in specified directories.
6674 This chapter discusses these options in detail.
6677 * file:: Choosing the Archive's Name
6678 * Selecting Archive Members::
6679 * files:: Reading Names from a File
6680 * exclude:: Excluding Some Files
6681 * wildcards:: Wildcards Patterns and Matching
6682 * quoting styles:: Ways of Quoting Special Characters in Names
6683 * transform:: Modifying File and Member Names
6684 * after:: Operating Only on New Files
6685 * recurse:: Descending into Directories
6686 * one:: Crossing File System Boundaries
6690 @section Choosing and Naming Archive Files
6692 @cindex Naming an archive
6693 @cindex Archive Name
6694 @cindex Choosing an archive file
6695 @cindex Where is the archive?
6697 By default, @command{tar} uses an archive file name that was compiled when
6698 it was built on the system; usually this name refers to some physical
6699 tape drive on the machine. However, the person who installed @command{tar}
6700 on the system may not have set the default to a meaningful value as far as
6701 most users are concerned. As a result, you will usually want to tell
6702 @command{tar} where to find (or create) the archive. The
6703 @option{--file=@var{archive-name}} (@option{-f @var{archive-name}})
6704 option allows you to either specify or name a file to use as the archive
6705 instead of the default archive file location.
6708 @xopindex{file, short description}
6709 @item --file=@var{archive-name}
6710 @itemx -f @var{archive-name}
6711 Name the archive to create or operate on. Use in conjunction with
6715 For example, in this @command{tar} command,
6718 $ @kbd{tar -cvf collection.tar blues folk jazz}
6722 @file{collection.tar} is the name of the archive. It must directly
6723 follow the @option{-f} option, since whatever directly follows @option{-f}
6724 @emph{will} end up naming the archive. If you neglect to specify an
6725 archive name, you may end up overwriting a file in the working directory
6726 with the archive you create since @command{tar} will use this file's name
6727 for the archive name.
6729 An archive can be saved as a file in the file system, sent through a
6730 pipe or over a network, or written to an I/O device such as a tape,
6731 floppy disk, or CD write drive.
6733 @cindex Writing new archives
6734 @cindex Archive creation
6735 If you do not name the archive, @command{tar} uses the value of the
6736 environment variable @env{TAPE} as the file name for the archive. If
6737 that is not available, @command{tar} uses a default, compiled-in archive
6738 name, usually that for tape unit zero (i.e., @file{/dev/tu00}).
6740 @cindex Standard input and output
6741 @cindex tar to standard input and output
6742 If you use @file{-} as an @var{archive-name}, @command{tar} reads the
6743 archive from standard input (when listing or extracting files), or
6744 writes it to standard output (when creating an archive). If you use
6745 @file{-} as an @var{archive-name} when modifying an archive,
6746 @command{tar} reads the original archive from its standard input and
6747 writes the entire new archive to its standard output.
6749 The following example is a convenient way of copying directory
6750 hierarchy from @file{sourcedir} to @file{targetdir}.
6753 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xpf -)}
6756 The @option{-C} option allows to avoid using subshells:
6759 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xpf -}
6762 In both examples above, the leftmost @command{tar} invocation archives
6763 the contents of @file{sourcedir} to the standard output, while the
6764 rightmost one reads this archive from its standard input and
6765 extracts it. The @option{-p} option tells it to restore permissions
6766 of the extracted files.
6768 @cindex Remote devices
6769 @cindex tar to a remote device
6771 To specify an archive file on a device attached to a remote machine,
6775 @kbd{--file=@var{hostname}:/@var{dev}/@var{file-name}}
6779 @command{tar} will set up the remote connection, if possible, and
6780 prompt you for a username and password. If you use
6781 @option{--file=@@@var{hostname}:/@var{dev}/@var{file-name}}, @command{tar}
6782 will attempt to set up the remote connection using your username
6783 as the username on the remote machine.
6785 @cindex Local and remote archives
6786 @anchor{local and remote archives}
6787 If the archive file name includes a colon (@samp{:}), then it is assumed
6788 to be a file on another machine. If the archive file is
6789 @samp{@var{user}@@@var{host}:@var{file}}, then @var{file} is used on the
6790 host @var{host}. The remote host is accessed using the @command{rsh}
6791 program, with a username of @var{user}. If the username is omitted
6792 (along with the @samp{@@} sign), then your user name will be used.
6793 (This is the normal @command{rsh} behavior.) It is necessary for the
6794 remote machine, in addition to permitting your @command{rsh} access, to
6795 have the @file{rmt} program installed (this command is included in
6796 the @GNUTAR{} distribution and by default is installed under
6797 @file{@var{prefix}/libexec/rmt}, where @var{prefix} means your
6798 installation prefix). If you need to use a file whose name includes a
6799 colon, then the remote tape drive behavior
6800 can be inhibited by using the @option{--force-local} option.
6802 When the archive is being created to @file{/dev/null}, @GNUTAR{}
6803 tries to minimize input and output operations. The Amanda backup
6804 system, when used with @GNUTAR{}, has an initial sizing pass which
6807 @node Selecting Archive Members
6808 @section Selecting Archive Members
6809 @cindex Specifying files to act on
6810 @cindex Specifying archive members
6812 @dfn{File Name arguments} specify which files in the file system
6813 @command{tar} operates on, when creating or adding to an archive, or which
6814 archive members @command{tar} operates on, when reading or deleting from
6815 an archive. @xref{Operations}.
6817 To specify file names, you can include them as the last arguments on
6818 the command line, as follows:
6820 @kbd{tar} @var{operation} [@var{option1} @var{option2} @dots{}] [@var{file name-1} @var{file name-2} @dots{}]
6823 If a file name begins with dash (@samp{-}), precede it with
6824 @option{--add-file} option to prevent it from being treated as an
6827 @anchor{input name quoting}
6828 By default @GNUTAR{} attempts to @dfn{unquote} each file or member
6829 name, replacing @dfn{escape sequences} according to the following
6832 @multitable @columnfractions 0.20 0.60
6833 @headitem Escape @tab Replaced with
6834 @item \a @tab Audible bell (@acronym{ASCII} 7)
6835 @item \b @tab Backspace (@acronym{ASCII} 8)
6836 @item \f @tab Form feed (@acronym{ASCII} 12)
6837 @item \n @tab New line (@acronym{ASCII} 10)
6838 @item \r @tab Carriage return (@acronym{ASCII} 13)
6839 @item \t @tab Horizontal tabulation (@acronym{ASCII} 9)
6840 @item \v @tab Vertical tabulation (@acronym{ASCII} 11)
6841 @item \? @tab @acronym{ASCII} 127
6842 @item \@var{n} @tab @acronym{ASCII} @var{n} (@var{n} should be an octal number
6846 A backslash followed by any other symbol is retained.
6848 This default behavior is controlled by the following command line
6854 Enable unquoting input file or member names (default).
6858 Disable unquoting input file or member names.
6861 If you specify a directory name as a file name argument, all the files
6862 in that directory are operated on by @command{tar}.
6864 If you do not specify files, @command{tar} behavior differs depending
6865 on the operation mode as described below:
6867 When @command{tar} is invoked with @option{--create} (@option{-c}),
6868 @command{tar} will stop immediately, reporting the following:
6872 $ @kbd{tar cf a.tar}
6873 tar: Cowardly refusing to create an empty archive
6874 Try `tar --help' or `tar --usage' for more information.
6878 If you specify either @option{--list} (@option{-t}) or
6879 @option{--extract} (@option{--get}, @option{-x}), @command{tar}
6880 operates on all the archive members in the archive.
6882 If run with @option{--diff} option, tar will compare the archive with
6883 the contents of the current working directory.
6885 If you specify any other operation, @command{tar} does nothing.
6887 By default, @command{tar} takes file names from the command line. However,
6888 there are other ways to specify file or member names, or to modify the
6889 manner in which @command{tar} selects the files or members upon which to
6890 operate. In general, these methods work both for specifying the names
6891 of files and archive members.
6894 @section Reading Names from a File
6896 @cindex Reading file names from a file
6897 @cindex Lists of file names
6898 @cindex File Name arguments, alternatives
6899 @cindex @command{find}, using with @command{tar}
6900 Instead of giving the names of files or archive members on the command
6901 line, you can put the names into a file, and then use the
6902 @option{--files-from=@var{file-of-names}} (@option{-T
6903 @var{file-of-names}}) option to @command{tar}. Give the name of the
6904 file which contains the list of files to include as the argument to
6905 @option{--files-from}. In the list, the file names should be separated by
6906 newlines. You will frequently use this option when you have generated
6907 the list of files to archive with the @command{find} utility.
6911 @item --files-from=@var{file-name}
6912 @itemx -T @var{file-name}
6913 Get names to extract or create from file @var{file-name}.
6916 If you give a single dash as a file name for @option{--files-from}, (i.e.,
6917 you specify either @code{--files-from=-} or @code{-T -}), then the file
6918 names are read from standard input.
6920 Unless you are running @command{tar} with @option{--create}, you can not use
6921 both @code{--files-from=-} and @code{--file=-} (@code{-f -}) in the same
6924 Any number of @option{-T} options can be given in the command line.
6926 The following example shows how to use @command{find} to generate a list of
6927 files smaller than 400K in length and put that list into a file
6928 called @file{small-files}. You can then use the @option{-T} option to
6929 @command{tar} to specify the files from that file, @file{small-files}, to
6930 create the archive @file{little.tgz}. (The @option{-z} option to
6931 @command{tar} compresses the archive with @command{gzip}; @pxref{gzip} for
6935 $ @kbd{find . -size -400 -print > small-files}
6936 $ @kbd{tar -c -v -z -T small-files -f little.tgz}
6940 In the file list given by @option{-T} option, any file name beginning
6941 with @samp{-} character is considered a @command{tar} option and is
6942 processed accordingly@footnote{Versions of @GNUTAR{} up to 1.15.1
6943 recognized only @option{-C} option in file lists, and only if the
6944 option and its argument occupied two consecutive lines.}. For example,
6945 the common use of this feature is to change to another directory by
6946 specifying @option{-C} option:
6956 $ @kbd{tar -c -f foo.tar --files-from list}
6961 In this example, @command{tar} will first switch to @file{/etc}
6962 directory and add files @file{passwd} and @file{hosts} to the
6963 archive. Then it will change to @file{/lib} directory and will archive
6964 the file @file{libc.a}. Thus, the resulting archive @file{foo.tar} will
6969 $ @kbd{tar tf foo.tar}
6977 @xopindex{directory, using in @option{--files-from} argument}
6978 Notice that the option parsing algorithm used with @option{-T} is
6979 stricter than the one used by shell. Namely, when specifying option
6980 arguments, you should observe the following rules:
6984 When using short (single-letter) option form, its argument must
6985 immediately follow the option letter, without any intervening
6986 whitespace. For example: @code{-Cdir}.
6989 When using long option form, the option argument must be separated
6990 from the option by a single equal sign. No whitespace is allowed on
6991 any side of the equal sign. For example: @code{--directory=dir}.
6994 For both short and long option forms, the option argument can be given
6995 on the next line after the option name, e.g.:
7016 If you happen to have a file whose name starts with @samp{-},
7017 precede it with @option{--add-file} option to prevent it from
7018 being recognized as an option. For example: @code{--add-file=--my-file}.
7025 @subsection @code{NUL}-Terminated File Names
7027 @cindex File names, terminated by @code{NUL}
7028 @cindex @code{NUL}-terminated file names
7029 The @option{--null} option causes
7030 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}})
7031 to read file names terminated by a @code{NUL} instead of a newline, so
7032 files whose names contain newlines can be archived using
7033 @option{--files-from}.
7036 @xopindex{null, described}
7038 Only consider @code{NUL}-terminated file names, instead of files that
7039 terminate in a newline.
7041 @xopindex{no-null, described}
7043 Undo the effect of any previous @option{--null} option.
7046 The @option{--null} option is just like the one in @acronym{GNU}
7047 @command{xargs} and @command{cpio}, and is useful with the
7048 @option{-print0} predicate of @acronym{GNU} @command{find}. In
7049 @command{tar}, @option{--null} also disables special handling for
7050 file names that begin with dash.
7052 This example shows how to use @command{find} to generate a list of files
7053 larger than 800K in length and put that list into a file called
7054 @file{long-files}. The @option{-print0} option to @command{find} is just
7055 like @option{-print}, except that it separates files with a @code{NUL}
7056 rather than with a newline. You can then run @command{tar} with both the
7057 @option{--null} and @option{-T} options to specify that @command{tar} gets the
7058 files from that file, @file{long-files}, to create the archive
7059 @file{big.tgz}. The @option{--null} option to @command{tar} will cause
7060 @command{tar} to recognize the @code{NUL} separator between files.
7063 $ @kbd{find . -size +800 -print0 > long-files}
7064 $ @kbd{tar -c -v --null --files-from=long-files --file=big.tar}
7067 The @option{--no-null} option can be used if you need to read both
7068 @code{NUL}-terminated and newline-terminated files on the same command line.
7069 For example, if @file{flist} is a newline-terminated file, then the
7070 following command can be used to combine it with the above command:
7074 $ @kbd{find . -size +800 -print0 |
7075 tar -c -f big.tar --null -T - --no-null -T flist}
7079 This example uses short options for typographic reasons, to avoid
7082 @GNUTAR is able to automatically detect @code{NUL}-terminated file lists, so
7083 it is safe to use them even without the @option{--null} option. In
7084 this case @command{tar} will print a warning and continue reading such
7085 a file as if @option{--null} were actually given:
7089 $ @kbd{find . -size +800 -print0 | tar -c -f big.tar -T -}
7090 tar: -: file name read contains nul character
7094 The null terminator, however, remains in effect only for this
7095 particular file, any following @option{-T} options will assume
7096 newline termination. Of course, the null autodetection applies
7097 to these eventual surplus @option{-T} options as well.
7100 @section Excluding Some Files
7102 @cindex File names, excluding files by
7103 @cindex Excluding files by name and pattern
7104 @cindex Excluding files by file system
7106 @opindex exclude-from
7107 To avoid operating on files whose names match a particular pattern,
7108 use the @option{--exclude} or @option{--exclude-from} options.
7112 @item --exclude=@var{pattern}
7113 Causes @command{tar} to ignore files that match the @var{pattern}.
7117 The @option{--exclude=@var{pattern}} option prevents any file or
7118 member whose name matches the shell wildcard (@var{pattern}) from
7120 For example, to create an archive with all the contents of the directory
7121 @file{src} except for files whose names end in @file{.o}, use the
7122 command @samp{tar -cf src.tar --exclude='*.o' src}.
7124 You may give multiple @option{--exclude} options.
7127 @opindex exclude-from
7128 @item --exclude-from=@var{file}
7129 @itemx -X @var{file}
7130 Causes @command{tar} to ignore files that match the patterns listed in
7134 @findex exclude-from
7135 Use the @option{--exclude-from} option to read a
7136 list of patterns, one per line, from @var{file}; @command{tar} will
7137 ignore files matching those patterns. Thus if @command{tar} is
7138 called as @w{@samp{tar -c -X foo .}} and the file @file{foo} contains a
7139 single line @file{*.o}, no files whose names end in @file{.o} will be
7140 added to the archive.
7142 Notice, that lines from @var{file} are read verbatim. One of the
7143 frequent errors is leaving some extra whitespace after a file name,
7144 which is difficult to catch using text editors.
7146 However, empty lines are OK.
7149 @cindex version control system, excluding files
7150 @cindex VCS, excluding files
7151 @cindex SCCS, excluding files
7152 @cindex RCS, excluding files
7153 @cindex CVS, excluding files
7154 @cindex SVN, excluding files
7155 @cindex git, excluding files
7156 @cindex Bazaar, excluding files
7157 @cindex Arch, excluding files
7158 @cindex Mercurial, excluding files
7159 @cindex Darcs, excluding files
7160 @opindex exclude-vcs
7162 Exclude files and directories used by following version control
7163 systems: @samp{CVS}, @samp{RCS}, @samp{SCCS}, @samp{SVN}, @samp{Arch},
7164 @samp{Bazaar}, @samp{Mercurial}, and @samp{Darcs}.
7166 As of version @value{VERSION}, the following files are excluded:
7169 @item @file{CVS/}, and everything under it
7170 @item @file{RCS/}, and everything under it
7171 @item @file{SCCS/}, and everything under it
7172 @item @file{.git/}, and everything under it
7173 @item @file{.gitignore}
7174 @item @file{.cvsignore}
7175 @item @file{.svn/}, and everything under it
7176 @item @file{.arch-ids/}, and everything under it
7177 @item @file{@{arch@}/}, and everything under it
7178 @item @file{=RELEASE-ID}
7179 @item @file{=meta-update}
7180 @item @file{=update}
7182 @item @file{.bzrignore}
7183 @item @file{.bzrtags}
7185 @item @file{.hgignore}
7186 @item @file{.hgrags}
7190 @opindex exclude-backups
7191 @item --exclude-backups
7192 Exclude backup and lock files. This option causes exclusion of files
7193 that match the following shell globbing patterns:
7203 @findex exclude-caches
7204 When creating an archive, the @option{--exclude-caches} option family
7205 causes @command{tar} to exclude all directories that contain a @dfn{cache
7206 directory tag}. A cache directory tag is a short file with the
7207 well-known name @file{CACHEDIR.TAG} and having a standard header
7208 specified in @url{http://www.brynosaurus.com/cachedir/spec.html}.
7209 Various applications write cache directory tags into directories they
7210 use to hold regenerable, non-precious data, so that such data can be
7211 more easily excluded from backups.
7213 There are three @samp{exclude-caches} options, each providing a different
7214 exclusion semantics:
7217 @opindex exclude-caches
7218 @item --exclude-caches
7219 Do not archive the contents of the directory, but archive the
7220 directory itself and the @file{CACHEDIR.TAG} file.
7222 @opindex exclude-caches-under
7223 @item --exclude-caches-under
7224 Do not archive the contents of the directory, nor the
7225 @file{CACHEDIR.TAG} file, archive only the directory itself.
7227 @opindex exclude-caches-all
7228 @item --exclude-caches-all
7229 Omit directories containing @file{CACHEDIR.TAG} file entirely.
7233 Another option family, @option{--exclude-tag}, provides a generalization of
7234 this concept. It takes a single argument, a file name to look for.
7235 Any directory that contains this file will be excluded from the dump.
7236 Similarly to @samp{exclude-caches}, there are three options in this
7240 @opindex exclude-tag
7241 @item --exclude-tag=@var{file}
7242 Do not dump the contents of the directory, but dump the
7243 directory itself and the @var{file}.
7245 @opindex exclude-tag-under
7246 @item --exclude-tag-under=@var{file}
7247 Do not dump the contents of the directory, nor the
7248 @var{file}, archive only the directory itself.
7250 @opindex exclude-tag-all
7251 @item --exclude-tag-all=@var{file}
7252 Omit directories containing @var{file} file entirely.
7255 Multiple @option{--exclude-tag*} options can be given.
7257 For example, given this directory:
7272 The @option{--exclude-tag} will produce the following:
7275 $ @kbd{tar -cf archive.tar --exclude-tag=tagfile -v dir}
7280 tar: dir/folk/: contains a cache directory tag tagfile;
7285 Both the @file{dir/folk} directory and its tagfile are preserved in
7286 the archive, however the rest of files in this directory are not.
7288 Now, using the @option{--exclude-tag-under} option will exclude
7289 @file{tagfile} from the dump, while still preserving the directory
7290 itself, as shown in this example:
7293 $ @kbd{tar -cf archive.tar --exclude-tag-under=tagfile -v dir}
7298 ./tar: dir/folk/: contains a cache directory tag tagfile;
7302 Finally, using @option{--exclude-tag-all} omits the @file{dir/folk}
7306 $ @kbd{tar -cf archive.tar --exclude-tag-all=tagfile -v dir}
7310 ./tar: dir/folk/: contains a cache directory tag tagfile;
7311 directory not dumped
7315 * problems with exclude::
7318 @node problems with exclude
7319 @unnumberedsubsec Problems with Using the @code{exclude} Options
7321 @xopindex{exclude, potential problems with}
7322 Some users find @samp{exclude} options confusing. Here are some common
7327 The main operating mode of @command{tar} does not act on a file name
7328 explicitly listed on the command line, if one of its file name
7329 components is excluded. In the example above, if
7330 you create an archive and exclude files that end with @samp{*.o}, but
7331 explicitly name the file @samp{dir.o/foo} after all the options have been
7332 listed, @samp{dir.o/foo} will be excluded from the archive.
7335 You can sometimes confuse the meanings of @option{--exclude} and
7336 @option{--exclude-from}. Be careful: use @option{--exclude} when files
7337 to be excluded are given as a pattern on the command line. Use
7338 @option{--exclude-from} to introduce the name of a file which contains
7339 a list of patterns, one per line; each of these patterns can exclude
7340 zero, one, or many files.
7343 When you use @option{--exclude=@var{pattern}}, be sure to quote the
7344 @var{pattern} parameter, so @GNUTAR{} sees wildcard characters
7345 like @samp{*}. If you do not do this, the shell might expand the
7346 @samp{*} itself using files at hand, so @command{tar} might receive a
7347 list of files instead of one pattern, or none at all, making the
7348 command somewhat illegal. This might not correspond to what you want.
7353 $ @kbd{tar -c -f @var{archive.tar} --exclude '*.o' @var{directory}}
7361 $ @kbd{tar -c -f @var{archive.tar} --exclude *.o @var{directory}}
7365 You must use use shell syntax, or globbing, rather than @code{regexp}
7366 syntax, when using exclude options in @command{tar}. If you try to use
7367 @code{regexp} syntax to describe files to be excluded, your command
7371 @FIXME{The change in semantics must have occurred before 1.11,
7372 so I doubt if it is worth mentioning at all. Anyway, should at
7373 least specify in which version the semantics changed.}
7374 In earlier versions of @command{tar}, what is now the
7375 @option{--exclude-from} option was called @option{--exclude} instead.
7376 Now, @option{--exclude} applies to patterns listed on the command
7377 line and @option{--exclude-from} applies to patterns listed in a
7383 @section Wildcards Patterns and Matching
7385 @dfn{Globbing} is the operation by which @dfn{wildcard} characters,
7386 @samp{*} or @samp{?} for example, are replaced and expanded into all
7387 existing files matching the given pattern. @GNUTAR{} can use wildcard
7388 patterns for matching (or globbing) archive members when extracting
7389 from or listing an archive. Wildcard patterns are also used for
7390 verifying volume labels of @command{tar} archives. This section has the
7391 purpose of explaining wildcard syntax for @command{tar}.
7393 @FIXME{the next few paragraphs need work.}
7395 A @var{pattern} should be written according to shell syntax, using wildcard
7396 characters to effect globbing. Most characters in the pattern stand
7397 for themselves in the matched string, and case is significant: @samp{a}
7398 will match only @samp{a}, and not @samp{A}. The character @samp{?} in the
7399 pattern matches any single character in the matched string. The character
7400 @samp{*} in the pattern matches zero, one, or more single characters in
7401 the matched string. The character @samp{\} says to take the following
7402 character of the pattern @emph{literally}; it is useful when one needs to
7403 match the @samp{?}, @samp{*}, @samp{[} or @samp{\} characters, themselves.
7405 The character @samp{[}, up to the matching @samp{]}, introduces a character
7406 class. A @dfn{character class} is a list of acceptable characters
7407 for the next single character of the matched string. For example,
7408 @samp{[abcde]} would match any of the first five letters of the alphabet.
7409 Note that within a character class, all of the ``special characters''
7410 listed above other than @samp{\} lose their special meaning; for example,
7411 @samp{[-\\[*?]]} would match any of the characters, @samp{-}, @samp{\},
7412 @samp{[}, @samp{*}, @samp{?}, or @samp{]}. (Due to parsing constraints,
7413 the characters @samp{-} and @samp{]} must either come @emph{first} or
7414 @emph{last} in a character class.)
7416 @cindex Excluding characters from a character class
7417 @cindex Character class, excluding characters from
7418 If the first character of the class after the opening @samp{[}
7419 is @samp{!} or @samp{^}, then the meaning of the class is reversed.
7420 Rather than listing character to match, it lists those characters which
7421 are @emph{forbidden} as the next single character of the matched string.
7423 Other characters of the class stand for themselves. The special
7424 construction @samp{[@var{a}-@var{e}]}, using an hyphen between two
7425 letters, is meant to represent all characters between @var{a} and
7428 @FIXME{need to add a sentence or so here to make this clear for those
7429 who don't have dan around.}
7431 Periods (@samp{.}) or forward slashes (@samp{/}) are not considered
7432 special for wildcard matches. However, if a pattern completely matches
7433 a directory prefix of a matched string, then it matches the full matched
7434 string: thus, excluding a directory also excludes all the files beneath it.
7437 * controlling pattern-matching::
7440 @node controlling pattern-matching
7441 @unnumberedsubsec Controlling Pattern-Matching
7443 For the purposes of this section, we call @dfn{exclusion members} all
7444 member names obtained while processing @option{--exclude} and
7445 @option{--exclude-from} options, and @dfn{inclusion members} those
7446 member names that were given in the command line or read from the file
7447 specified with @option{--files-from} option.
7449 These two pairs of member lists are used in the following operations:
7450 @option{--diff}, @option{--extract}, @option{--list},
7453 There are no inclusion members in create mode (@option{--create} and
7454 @option{--append}), since in this mode the names obtained from the
7455 command line refer to @emph{files}, not archive members.
7457 By default, inclusion members are compared with archive members
7458 literally @footnote{Notice that earlier @GNUTAR{} versions used
7459 globbing for inclusion members, which contradicted to UNIX98
7460 specification and was not documented. @xref{Changes}, for more
7461 information on this and other changes.} and exclusion members are
7462 treated as globbing patterns. For example:
7466 $ @kbd{tar tf foo.tar}
7471 # @i{Member names are used verbatim:}
7472 $ @kbd{tar -xf foo.tar -v '[remarks]'}
7474 # @i{Exclude member names are globbed:}
7475 $ @kbd{tar -xf foo.tar -v --exclude '*.c'}
7481 This behavior can be altered by using the following options:
7486 Treat all member names as wildcards.
7488 @opindex no-wildcards
7489 @item --no-wildcards
7490 Treat all member names as literal strings.
7493 Thus, to extract files whose names end in @samp{.c}, you can use:
7496 $ @kbd{tar -xf foo.tar -v --wildcards '*.c'}
7502 Notice quoting of the pattern to prevent the shell from interpreting
7505 The effect of @option{--wildcards} option is canceled by
7506 @option{--no-wildcards}. This can be used to pass part of
7507 the command line arguments verbatim and other part as globbing
7508 patterns. For example, the following invocation:
7511 $ @kbd{tar -xf foo.tar --wildcards '*.txt' --no-wildcards '[remarks]'}
7515 instructs @command{tar} to extract from @file{foo.tar} all files whose
7516 names end in @samp{.txt} and the file named @file{[remarks]}.
7518 Normally, a pattern matches a name if an initial subsequence of the
7519 name's components matches the pattern, where @samp{*}, @samp{?}, and
7520 @samp{[...]} are the usual shell wildcards, @samp{\} escapes wildcards,
7521 and wildcards can match @samp{/}.
7523 Other than optionally stripping leading @samp{/} from names
7524 (@pxref{absolute}), patterns and names are used as-is. For
7525 example, trailing @samp{/} is not trimmed from a user-specified name
7526 before deciding whether to exclude it.
7528 However, this matching procedure can be altered by the options listed
7529 below. These options accumulate. For example:
7532 --ignore-case --exclude='makefile' --no-ignore-case ---exclude='readme'
7536 ignores case when excluding @samp{makefile}, but not when excluding
7541 @opindex no-anchored
7543 @itemx --no-anchored
7544 If anchored, a pattern must match an initial subsequence
7545 of the name's components. Otherwise, the pattern can match any
7546 subsequence. Default is @option{--no-anchored} for exclusion members
7547 and @option{--anchored} inclusion members.
7549 @opindex ignore-case
7550 @opindex no-ignore-case
7552 @itemx --no-ignore-case
7553 When ignoring case, upper-case patterns match lower-case names and vice versa.
7554 When not ignoring case (the default), matching is case-sensitive.
7556 @opindex wildcards-match-slash
7557 @opindex no-wildcards-match-slash
7558 @item --wildcards-match-slash
7559 @itemx --no-wildcards-match-slash
7560 When wildcards match slash (the default for exclusion members), a
7561 wildcard like @samp{*} in the pattern can match a @samp{/} in the
7562 name. Otherwise, @samp{/} is matched only by @samp{/}.
7566 The @option{--recursion} and @option{--no-recursion} options
7567 (@pxref{recurse}) also affect how member patterns are interpreted. If
7568 recursion is in effect, a pattern matches a name if it matches any of
7569 the name's parent directories.
7571 The following table summarizes pattern-matching default values:
7573 @multitable @columnfractions .3 .7
7574 @headitem Members @tab Default settings
7575 @item Inclusion @tab @option{--no-wildcards --anchored --no-wildcards-match-slash}
7576 @item Exclusion @tab @option{--wildcards --no-anchored --wildcards-match-slash}
7579 @node quoting styles
7580 @section Quoting Member Names
7582 When displaying member names, @command{tar} takes care to avoid
7583 ambiguities caused by certain characters. This is called @dfn{name
7584 quoting}. The characters in question are:
7587 @item Non-printable control characters:
7588 @anchor{escape sequences}
7589 @multitable @columnfractions 0.20 0.10 0.60
7590 @headitem Character @tab @acronym{ASCII} @tab Character name
7591 @item \a @tab 7 @tab Audible bell
7592 @item \b @tab 8 @tab Backspace
7593 @item \f @tab 12 @tab Form feed
7594 @item \n @tab 10 @tab New line
7595 @item \r @tab 13 @tab Carriage return
7596 @item \t @tab 9 @tab Horizontal tabulation
7597 @item \v @tab 11 @tab Vertical tabulation
7600 @item Space (@acronym{ASCII} 32)
7602 @item Single and double quotes (@samp{'} and @samp{"})
7604 @item Backslash (@samp{\})
7607 The exact way @command{tar} uses to quote these characters depends on
7608 the @dfn{quoting style}. The default quoting style, called
7609 @dfn{escape} (see below), uses backslash notation to represent control
7610 characters, space and backslash. Using this quoting style, control
7611 characters are represented as listed in column @samp{Character} in the
7612 above table, a space is printed as @samp{\ } and a backslash as @samp{\\}.
7614 @GNUTAR{} offers seven distinct quoting styles, which can be selected
7615 using @option{--quoting-style} option:
7618 @item --quoting-style=@var{style}
7619 @opindex quoting-style
7621 Sets quoting style. Valid values for @var{style} argument are:
7622 literal, shell, shell-always, c, escape, locale, clocale.
7625 These styles are described in detail below. To illustrate their
7626 effect, we will use an imaginary tar archive @file{arch.tar}
7627 containing the following members:
7631 # 1. Contains horizontal tabulation character.
7633 # 2. Contains newline character
7636 # 3. Contains a space
7638 # 4. Contains double quotes
7640 # 5. Contains single quotes
7642 # 6. Contains a backslash character:
7647 Here is how usual @command{ls} command would have listed them, if they
7648 had existed in the current working directory:
7666 No quoting, display each character as is:
7670 $ @kbd{tar tf arch.tar --quoting-style=literal}
7683 Display characters the same way Bourne shell does:
7684 control characters, except @samp{\t} and @samp{\n}, are printed using
7685 backslash escapes, @samp{\t} and @samp{\n} are printed as is, and a
7686 single quote is printed as @samp{\'}. If a name contains any quoted
7687 characters, it is enclosed in single quotes. In particular, if a name
7688 contains single quotes, it is printed as several single-quoted strings:
7692 $ @kbd{tar tf arch.tar --quoting-style=shell}
7695 './a'\''single'\''quote'
7705 Same as @samp{shell}, but the names are always enclosed in single
7710 $ @kbd{tar tf arch.tar --quoting-style=shell-always}
7713 './a'\''single'\''quote'
7723 Use the notation of the C programming language. All names are
7724 enclosed in double quotes. Control characters are quoted using
7725 backslash notations, double quotes are represented as @samp{\"},
7726 backslash characters are represented as @samp{\\}. Single quotes and
7727 spaces are not quoted:
7731 $ @kbd{tar tf arch.tar --quoting-style=c}
7735 "./a\"double\"quote"
7743 Control characters are printed using backslash notation, a space is
7744 printed as @samp{\ } and a backslash as @samp{\\}. This is the
7745 default quoting style, unless it was changed when configured the
7750 $ @kbd{tar tf arch.tar --quoting-style=escape}
7762 Control characters, single quote and backslash are printed using
7763 backslash notation. All names are quoted using left and right
7764 quotation marks, appropriate to the current locale. If it does not
7765 define quotation marks, use @samp{`} as left and @samp{'} as right
7766 quotation marks. Any occurrences of the right quotation mark in a
7767 name are escaped with @samp{\}, for example:
7773 $ @kbd{tar tf arch.tar --quoting-style=locale}
7776 `./a\'single\'quote'
7785 Same as @samp{locale}, but @samp{"} is used for both left and right
7786 quotation marks, if not provided by the currently selected locale:
7790 $ @kbd{tar tf arch.tar --quoting-style=clocale}
7794 "./a\"double\"quote"
7802 You can specify which characters should be quoted in addition to those
7803 implied by the current quoting style:
7806 @item --quote-chars=@var{string}
7807 Always quote characters from @var{string}, even if the selected
7808 quoting style would not quote them.
7811 For example, using @samp{escape} quoting (compare with the usual
7812 escape listing above):
7816 $ @kbd{tar tf arch.tar --quoting-style=escape --quote-chars=' "'}
7827 To disable quoting of such additional characters, use the following
7831 @item --no-quote-chars=@var{string}
7832 Remove characters listed in @var{string} from the list of quoted
7833 characters set by the previous @option{--quote-chars} option.
7836 This option is particularly useful if you have added
7837 @option{--quote-chars} to your @env{TAR_OPTIONS} (@pxref{TAR_OPTIONS})
7838 and wish to disable it for the current invocation.
7840 Note, that @option{--no-quote-chars} does @emph{not} disable those
7841 characters that are quoted by default in the selected quoting style.
7844 @section Modifying File and Member Names
7846 @command{Tar} archives contain detailed information about files stored
7847 in them and full file names are part of that information. When
7848 storing a file to an archive, its file name is recorded in it,
7849 along with the actual file contents. When restoring from an archive,
7850 a file is created on disk with exactly the same name as that stored
7851 in the archive. In the majority of cases this is the desired behavior
7852 of a file archiver. However, there are some cases when it is not.
7854 First of all, it is often unsafe to extract archive members with
7855 absolute file names or those that begin with a @file{../}. @GNUTAR{}
7856 takes special precautions when extracting such names and provides a
7857 special option for handling them, which is described in
7860 Secondly, you may wish to extract file names without some leading
7861 directory components, or with otherwise modified names. In other
7862 cases it is desirable to store files under differing names in the
7865 @GNUTAR{} provides several options for these needs.
7868 @opindex strip-components
7869 @item --strip-components=@var{number}
7870 Strip given @var{number} of leading components from file names before
7874 For example, suppose you have archived whole @file{/usr} hierarchy to
7875 a tar archive named @file{usr.tar}. Among other files, this archive
7876 contains @file{usr/include/stdlib.h}, which you wish to extract to
7877 the current working directory. To do so, you type:
7880 $ @kbd{tar -xf usr.tar --strip=2 usr/include/stdlib.h}
7883 The option @option{--strip=2} instructs @command{tar} to strip the
7884 two leading components (@file{usr/} and @file{include/}) off the file
7887 If you add the @option{--verbose} (@option{-v}) option to the invocation
7888 above, you will note that the verbose listing still contains the
7889 full file name, with the two removed components still in place. This
7890 can be inconvenient, so @command{tar} provides a special option for
7891 altering this behavior:
7893 @anchor{show-transformed-names}
7895 @opindex show-transformed-names
7896 @item --show-transformed-names
7897 Display file or member names with all requested transformations
7906 $ @kbd{tar -xf usr.tar -v --strip=2 usr/include/stdlib.h}
7907 usr/include/stdlib.h
7908 $ @kbd{tar -xf usr.tar -v --strip=2 --show-transformed usr/include/stdlib.h}
7913 Notice that in both cases the file @file{stdlib.h} is extracted to the
7914 current working directory, @option{--show-transformed-names} affects
7915 only the way its name is displayed.
7917 This option is especially useful for verifying whether the invocation
7918 will have the desired effect. Thus, before running
7921 $ @kbd{tar -x --strip=@var{n}}
7925 it is often advisable to run
7928 $ @kbd{tar -t -v --show-transformed --strip=@var{n}}
7932 to make sure the command will produce the intended results.
7934 In case you need to apply more complex modifications to the file name,
7935 @GNUTAR{} provides a general-purpose transformation option:
7940 @item --transform=@var{expression}
7941 @itemx --xform=@var{expression}
7942 Modify file names using supplied @var{expression}.
7946 The @var{expression} is a @command{sed}-like replace expression of the
7950 s/@var{regexp}/@var{replace}/[@var{flags}]
7954 where @var{regexp} is a @dfn{regular expression}, @var{replace} is a
7955 replacement for each file name part that matches @var{regexp}. Both
7956 @var{regexp} and @var{replace} are described in detail in
7957 @ref{The "s" Command, The "s" Command, The `s' Command, sed, GNU sed}.
7959 Any delimiter can be used in lieu of @samp{/}, the only requirement being
7960 that it be used consistently throughout the expression. For example,
7961 the following two expressions are equivalent:
7970 Changing delimiters is often useful when the @var{regex} contains
7971 slashes. For example, it is more convenient to write @code{s,/,-,} than
7974 As in @command{sed}, you can give several replace expressions,
7975 separated by a semicolon.
7977 Supported @var{flags} are:
7981 Apply the replacement to @emph{all} matches to the @var{regexp}, not
7985 Use case-insensitive matching.
7988 @var{regexp} is an @dfn{extended regular expression} (@pxref{Extended
7989 regexps, Extended regular expressions, Extended regular expressions,
7993 Only replace the @var{number}th match of the @var{regexp}.
7995 Note: the @acronym{POSIX} standard does not specify what should happen
7996 when you mix the @samp{g} and @var{number} modifiers. @GNUTAR{}
7997 follows the GNU @command{sed} implementation in this regard, so
7998 the interaction is defined to be: ignore matches before the
7999 @var{number}th, and then match and replace all matches from the
8004 In addition, several @dfn{transformation scope} flags are supported,
8005 that control to what files transformations apply. These are:
8009 Apply transformation to regular archive members.
8012 Do not apply transformation to regular archive members.
8015 Apply transformation to symbolic link targets.
8018 Do not apply transformation to symbolic link targets.
8021 Apply transformation to hard link targets.
8024 Do not apply transformation to hard link targets.
8027 Default is @samp{rsh}, which means to apply tranformations to both archive
8028 members and targets of symbolic and hard links.
8030 Default scope flags can also be changed using @samp{flags=} statement
8031 in the transform expression. The flags set this way remain in force
8032 until next @samp{flags=} statement or end of expression, whichever
8033 occurs first. For example:
8036 --transform 'flags=S;s|^|/usr/local/|'
8039 Here are several examples of @option{--transform} usage:
8042 @item Extract @file{usr/} hierarchy into @file{usr/local/}:
8045 $ @kbd{tar --transform='s,usr/,usr/local/,' -x -f arch.tar}
8048 @item Strip two leading directory components (equivalent to
8049 @option{--strip-components=2}):
8052 $ @kbd{tar --transform='s,/*[^/]*/[^/]*/,,' -x -f arch.tar}
8055 @item Convert each file name to lower case:
8058 $ @kbd{tar --transform 's/.*/\L&/' -x -f arch.tar}
8061 @item Prepend @file{/prefix/} to each file name:
8064 $ @kbd{tar --transform 's,^,/prefix/,' -x -f arch.tar}
8067 @item Archive the @file{/lib} directory, prepending @samp{/usr/local}
8068 to each archive member:
8071 $ @kbd{tar --transform 's,^,/usr/local/,S' -c -f arch.tar /lib}
8075 Notice the use of flags in the last example. The @file{/lib}
8076 directory often contains many symbolic links to files within it.
8077 It may look, for example, like this:
8081 drwxr-xr-x root/root 0 2008-07-08 16:20 /lib/
8082 -rwxr-xr-x root/root 1250840 2008-05-25 07:44 /lib/libc-2.3.2.so
8083 lrwxrwxrwx root/root 0 2008-06-24 17:12 /lib/libc.so.6 -> libc-2.3.2.so
8087 Using the expression @samp{s,^,/usr/local/,} would mean adding
8088 @samp{/usr/local} to both regular archive members and to link
8089 targets. In this case, @file{/lib/libc.so.6} would become:
8092 /usr/local/lib/libc.so.6 -> /usr/local/libc-2.3.2.so
8095 This is definitely not desired. To avoid this, the @samp{S} flag
8096 is used, which excludes symbolic link targets from filename
8097 transformations. The result is:
8100 $ @kbd{tar --transform 's,^,/usr/local/,S', -c -v -f arch.tar \
8101 --show-transformed /lib}
8102 drwxr-xr-x root/root 0 2008-07-08 16:20 /usr/local/lib/
8103 -rwxr-xr-x root/root 1250840 2008-05-25 07:44 /usr/local/lib/libc-2.3.2.so
8104 lrwxrwxrwx root/root 0 2008-06-24 17:12 /usr/local/lib/libc.so.6 \
8108 Unlike @option{--strip-components}, @option{--transform} can be used
8109 in any @GNUTAR{} operation mode. For example, the following command
8110 adds files to the archive while replacing the leading @file{usr/}
8111 component with @file{var/}:
8114 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' /}
8117 To test @option{--transform} effect we suggest using
8118 @option{--show-transformed-names} option:
8121 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' \
8122 --verbose --show-transformed-names /}
8125 If both @option{--strip-components} and @option{--transform} are used
8126 together, then @option{--transform} is applied first, and the required
8127 number of components is then stripped from its result.
8129 You can use as many @option{--transform} options in a single command
8130 line as you want. The specified expressions will then be applied in
8131 order of their appearance. For example, the following two invocations
8135 $ @kbd{tar -cf arch.tar --transform='s,/usr/var,/var/' \
8136 --transform='s,/usr/local,/usr/,'}
8137 $ @kbd{tar -cf arch.tar \
8138 --transform='s,/usr/var,/var/;s,/usr/local,/usr/,'}
8142 @section Operating Only on New Files
8144 @cindex Excluding file by age
8145 @cindex Data Modification time, excluding files by
8146 @cindex Modification time, excluding files by
8147 @cindex Age, excluding files by
8148 The @option{--after-date=@var{date}} (@option{--newer=@var{date}},
8149 @option{-N @var{date}}) option causes @command{tar} to only work on
8150 files whose data modification or status change times are newer than
8151 the @var{date} given. If @var{date} starts with @samp{/} or @samp{.},
8152 it is taken to be a file name; the data modification time of that file
8153 is used as the date. If you use this option when creating or appending
8154 to an archive, the archive will only include new files. If you use
8155 @option{--after-date} when extracting an archive, @command{tar} will
8156 only extract files newer than the @var{date} you specify.
8158 If you only want @command{tar} to make the date comparison based on
8159 modification of the file's data (rather than status
8160 changes), then use the @option{--newer-mtime=@var{date}} option.
8162 @cindex --after-date and --update compared
8163 @cindex --newer-mtime and --update compared
8164 You may use these options with any operation. Note that these options
8165 differ from the @option{--update} (@option{-u}) operation in that they
8166 allow you to specify a particular date against which @command{tar} can
8167 compare when deciding whether or not to archive the files.
8172 @item --after-date=@var{date}
8173 @itemx --newer=@var{date}
8174 @itemx -N @var{date}
8175 Only store files newer than @var{date}.
8177 Acts on files only if their data modification or status change times are
8178 later than @var{date}. Use in conjunction with any operation.
8180 If @var{date} starts with @samp{/} or @samp{.}, it is taken to be a file
8181 name; the data modification time of that file is used as the date.
8183 @opindex newer-mtime
8184 @item --newer-mtime=@var{date}
8185 Acts like @option{--after-date}, but only looks at data modification times.
8188 These options limit @command{tar} to operate only on files which have
8189 been modified after the date specified. A file's status is considered to have
8190 changed if its contents have been modified, or if its owner,
8191 permissions, and so forth, have been changed. (For more information on
8192 how to specify a date, see @ref{Date input formats}; remember that the
8193 entire date argument must be quoted if it contains any spaces.)
8195 Gurus would say that @option{--after-date} tests both the data
8196 modification time (@code{mtime}, the time the contents of the file
8197 were last modified) and the status change time (@code{ctime}, the time
8198 the file's status was last changed: owner, permissions, etc.@:)
8199 fields, while @option{--newer-mtime} tests only the @code{mtime}
8202 To be precise, @option{--after-date} checks @emph{both} @code{mtime} and
8203 @code{ctime} and processes the file if either one is more recent than
8204 @var{date}, while @option{--newer-mtime} only checks @code{mtime} and
8205 disregards @code{ctime}. Neither does it use @code{atime} (the last time the
8206 contents of the file were looked at).
8208 Date specifiers can have embedded spaces. Because of this, you may need
8209 to quote date arguments to keep the shell from parsing them as separate
8210 arguments. For example, the following command will add to the archive
8211 all the files modified less than two days ago:
8214 $ @kbd{tar -cf foo.tar --newer-mtime '2 days ago'}
8217 When any of these options is used with the option @option{--verbose}
8218 (@pxref{verbose tutorial}) @GNUTAR{} will try to convert the specified
8219 date back to its textual representation and compare that with the
8220 one given with the option. If the two dates differ, @command{tar} will
8221 print a warning saying what date it will use. This is to help user
8222 ensure he is using the right date. For example:
8226 $ @kbd{tar -c -f archive.tar --after-date='10 days ago' .}
8227 tar: Option --after-date: Treating date `10 days ago' as 2006-06-11
8233 @strong{Please Note:} @option{--after-date} and @option{--newer-mtime}
8234 should not be used for incremental backups. @xref{Incremental Dumps},
8235 for proper way of creating incremental backups.
8239 @section Descending into Directories
8240 @cindex Avoiding recursion in directories
8241 @cindex Descending directories, avoiding
8242 @cindex Directories, avoiding recursion
8243 @cindex Recursion in directories, avoiding
8245 Usually, @command{tar} will recursively explore all directories (either
8246 those given on the command line or through the @option{--files-from}
8247 option) for the various files they contain. However, you may not always
8248 want @command{tar} to act this way.
8250 @opindex no-recursion
8251 @cindex @command{find}, using with @command{tar}
8252 The @option{--no-recursion} option inhibits @command{tar}'s recursive descent
8253 into specified directories. If you specify @option{--no-recursion}, you can
8254 use the @command{find} (@pxref{Top,, find, find, GNU Find Manual})
8255 utility for hunting through levels of directories to
8256 construct a list of file names which you could then pass to @command{tar}.
8257 @command{find} allows you to be more selective when choosing which files to
8258 archive; see @ref{files}, for more information on using @command{find} with
8262 @item --no-recursion
8263 Prevents @command{tar} from recursively descending directories.
8267 Requires @command{tar} to recursively descend directories.
8268 This is the default.
8271 When you use @option{--no-recursion}, @GNUTAR{} grabs
8272 directory entries themselves, but does not descend on them
8273 recursively. Many people use @command{find} for locating files they
8274 want to back up, and since @command{tar} @emph{usually} recursively
8275 descends on directories, they have to use the @samp{@w{-not -type d}}
8276 test in their @command{find} invocation (@pxref{Type, Type, Type test,
8277 find, Finding Files}), as they usually do not want all the files in a
8278 directory. They then use the @option{--files-from} option to archive
8279 the files located via @command{find}.
8281 The problem when restoring files archived in this manner is that the
8282 directories themselves are not in the archive; so the
8283 @option{--same-permissions} (@option{--preserve-permissions},
8284 @option{-p}) option does not affect them---while users might really
8285 like it to. Specifying @option{--no-recursion} is a way to tell
8286 @command{tar} to grab only the directory entries given to it, adding
8287 no new files on its own. To summarize, if you use @command{find} to
8288 create a list of files to be stored in an archive, use it as follows:
8292 $ @kbd{find @var{dir} @var{tests} | \
8293 tar -cf @var{archive} -T - --no-recursion}
8297 The @option{--no-recursion} option also applies when extracting: it
8298 causes @command{tar} to extract only the matched directory entries, not
8299 the files under those directories.
8301 The @option{--no-recursion} option also affects how globbing patterns
8302 are interpreted (@pxref{controlling pattern-matching}).
8304 The @option{--no-recursion} and @option{--recursion} options apply to
8305 later options and operands, and can be overridden by later occurrences
8306 of @option{--no-recursion} and @option{--recursion}. For example:
8309 $ @kbd{tar -cf jams.tar --no-recursion grape --recursion grape/concord}
8313 creates an archive with one entry for @file{grape}, and the recursive
8314 contents of @file{grape/concord}, but no entries under @file{grape}
8315 other than @file{grape/concord}.
8318 @section Crossing File System Boundaries
8319 @cindex File system boundaries, not crossing
8321 @command{tar} will normally automatically cross file system boundaries in
8322 order to archive files which are part of a directory tree. You can
8323 change this behavior by running @command{tar} and specifying
8324 @option{--one-file-system}. This option only affects files that are
8325 archived because they are in a directory that is being archived;
8326 @command{tar} will still archive files explicitly named on the command line
8327 or through @option{--files-from}, regardless of where they reside.
8330 @opindex one-file-system
8331 @item --one-file-system
8332 Prevents @command{tar} from crossing file system boundaries when
8333 archiving. Use in conjunction with any write operation.
8336 The @option{--one-file-system} option causes @command{tar} to modify its
8337 normal behavior in archiving the contents of directories. If a file in
8338 a directory is not on the same file system as the directory itself, then
8339 @command{tar} will not archive that file. If the file is a directory
8340 itself, @command{tar} will not archive anything beneath it; in other words,
8341 @command{tar} will not cross mount points.
8343 This option is useful for making full or incremental archival backups of
8344 a file system. If this option is used in conjunction with
8345 @option{--verbose} (@option{-v}), files that are excluded are
8346 mentioned by name on the standard error.
8349 * directory:: Changing Directory
8350 * absolute:: Absolute File Names
8354 @subsection Changing the Working Directory
8356 @FIXME{need to read over this node now for continuity; i've switched
8357 things around some.}
8359 @cindex Changing directory mid-stream
8360 @cindex Directory, changing mid-stream
8361 @cindex Working directory, specifying
8362 To change the working directory in the middle of a list of file names,
8363 either on the command line or in a file specified using
8364 @option{--files-from} (@option{-T}), use @option{--directory} (@option{-C}).
8365 This will change the working directory to the specified directory
8366 after that point in the list.
8370 @item --directory=@var{directory}
8371 @itemx -C @var{directory}
8372 Changes the working directory in the middle of a command line.
8378 $ @kbd{tar -c -f jams.tar grape prune -C food cherry}
8382 will place the files @file{grape} and @file{prune} from the current
8383 directory into the archive @file{jams.tar}, followed by the file
8384 @file{cherry} from the directory @file{food}. This option is especially
8385 useful when you have several widely separated files that you want to
8386 store in the same archive.
8388 Note that the file @file{cherry} is recorded in the archive under the
8389 precise name @file{cherry}, @emph{not} @file{food/cherry}. Thus, the
8390 archive will contain three files that all appear to have come from the
8391 same directory; if the archive is extracted with plain @samp{tar
8392 --extract}, all three files will be written in the current directory.
8394 Contrast this with the command,
8397 $ @kbd{tar -c -f jams.tar grape prune -C food red/cherry}
8401 which records the third file in the archive under the name
8402 @file{red/cherry} so that, if the archive is extracted using
8403 @samp{tar --extract}, the third file will be written in a subdirectory
8406 You can use the @option{--directory} option to make the archive
8407 independent of the original name of the directory holding the files.
8408 The following command places the files @file{/etc/passwd},
8409 @file{/etc/hosts}, and @file{/lib/libc.a} into the archive
8413 $ @kbd{tar -c -f foo.tar -C /etc passwd hosts -C /lib libc.a}
8417 However, the names of the archive members will be exactly what they were
8418 on the command line: @file{passwd}, @file{hosts}, and @file{libc.a}.
8419 They will not appear to be related by file name to the original
8420 directories where those files were located.
8422 Note that @option{--directory} options are interpreted consecutively. If
8423 @option{--directory} specifies a relative file name, it is interpreted
8424 relative to the then current directory, which might not be the same as
8425 the original current working directory of @command{tar}, due to a previous
8426 @option{--directory} option.
8428 When using @option{--files-from} (@pxref{files}), you can put various
8429 @command{tar} options (including @option{-C}) in the file list. Notice,
8430 however, that in this case the option and its argument may not be
8431 separated by whitespace. If you use short option, its argument must
8432 either follow the option letter immediately, without any intervening
8433 whitespace, or occupy the next line. Otherwise, if you use long
8434 option, separate its argument by an equal sign.
8436 For instance, the file list for the above example will be:
8449 To use it, you would invoke @command{tar} as follows:
8452 $ @kbd{tar -c -f foo.tar --files-from list}
8455 The interpretation of @option{--directory} is disabled by
8456 @option{--null} option.
8459 @subsection Absolute File Names
8460 @cindex absolute file names
8461 @cindex file names, absolute
8463 By default, @GNUTAR{} drops a leading @samp{/} on
8464 input or output, and complains about file names containing a @file{..}
8465 component. There is an option that turns off this behavior:
8468 @opindex absolute-names
8469 @item --absolute-names
8471 Do not strip leading slashes from file names, and permit file names
8472 containing a @file{..} file name component.
8475 When @command{tar} extracts archive members from an archive, it strips any
8476 leading slashes (@samp{/}) from the member name. This causes absolute
8477 member names in the archive to be treated as relative file names. This
8478 allows you to have such members extracted wherever you want, instead of
8479 being restricted to extracting the member in the exact directory named
8480 in the archive. For example, if the archive member has the name
8481 @file{/etc/passwd}, @command{tar} will extract it as if the name were
8482 really @file{etc/passwd}.
8484 File names containing @file{..} can cause problems when extracting, so
8485 @command{tar} normally warns you about such files when creating an
8486 archive, and rejects attempts to extracts such files.
8488 Other @command{tar} programs do not do this. As a result, if you
8489 create an archive whose member names start with a slash, they will be
8490 difficult for other people with a non-@GNUTAR{}
8491 program to use. Therefore, @GNUTAR{} also strips
8492 leading slashes from member names when putting members into the
8493 archive. For example, if you ask @command{tar} to add the file
8494 @file{/bin/ls} to an archive, it will do so, but the member name will
8495 be @file{bin/ls}@footnote{A side effect of this is that when
8496 @option{--create} is used with @option{--verbose} the resulting output
8497 is not, generally speaking, the same as the one you'd get running
8498 @kbd{tar --list} command. This may be important if you use some
8499 scripts for comparing both outputs. @xref{listing member and file names},
8500 for the information on how to handle this case.}.
8502 If you use the @option{--absolute-names} (@option{-P}) option,
8503 @command{tar} will do none of these transformations.
8505 To archive or extract files relative to the root directory, specify
8506 the @option{--absolute-names} (@option{-P}) option.
8508 Normally, @command{tar} acts on files relative to the working
8509 directory---ignoring superior directory names when archiving, and
8510 ignoring leading slashes when extracting.
8512 When you specify @option{--absolute-names} (@option{-P}),
8513 @command{tar} stores file names including all superior directory
8514 names, and preserves leading slashes. If you only invoked
8515 @command{tar} from the root directory you would never need the
8516 @option{--absolute-names} option, but using this option
8517 may be more convenient than switching to root.
8519 @FIXME{Should be an example in the tutorial/wizardry section using this
8520 to transfer files between systems.}
8523 @item --absolute-names
8524 Preserves full file names (including superior directory names) when
8525 archiving files. Preserves leading slash when extracting files.
8529 @command{tar} prints out a message about removing the @samp{/} from
8530 file names. This message appears once per @GNUTAR{}
8531 invocation. It represents something which ought to be told; ignoring
8532 what it means can cause very serious surprises, later.
8534 Some people, nevertheless, do not want to see this message. Wanting to
8535 play really dangerously, one may of course redirect @command{tar} standard
8536 error to the sink. For example, under @command{sh}:
8539 $ @kbd{tar -c -f archive.tar /home 2> /dev/null}
8543 Another solution, both nicer and simpler, would be to change to
8544 the @file{/} directory first, and then avoid absolute notation.
8548 $ @kbd{tar -c -f archive.tar -C / home}
8551 @include getdate.texi
8554 @chapter Controlling the Archive Format
8556 @cindex Tar archive formats
8557 Due to historical reasons, there are several formats of tar archives.
8558 All of them are based on the same principles, but have some subtle
8559 differences that often make them incompatible with each other.
8561 GNU tar is able to create and handle archives in a variety of formats.
8562 The most frequently used formats are (in alphabetical order):
8566 Format used by @GNUTAR{} versions up to 1.13.25. This format derived
8567 from an early @acronym{POSIX} standard, adding some improvements such as
8568 sparse file handling and incremental archives. Unfortunately these
8569 features were implemented in a way incompatible with other archive
8572 Archives in @samp{gnu} format are able to hold file names of unlimited
8576 Format used by @GNUTAR{} of versions prior to 1.12.
8579 Archive format, compatible with the V7 implementation of tar. This
8580 format imposes a number of limitations. The most important of them
8584 @item The maximum length of a file name is limited to 99 characters.
8585 @item The maximum length of a symbolic link is limited to 99 characters.
8586 @item It is impossible to store special files (block and character
8587 devices, fifos etc.)
8588 @item Maximum value of user or group @acronym{ID} is limited to 2097151 (7777777
8590 @item V7 archives do not contain symbolic ownership information (user
8591 and group name of the file owner).
8594 This format has traditionally been used by Automake when producing
8595 Makefiles. This practice will change in the future, in the meantime,
8596 however this means that projects containing file names more than 99
8597 characters long will not be able to use @GNUTAR{} @value{VERSION} and
8598 Automake prior to 1.9.
8601 Archive format defined by @acronym{POSIX.1-1988} specification. It stores
8602 symbolic ownership information. It is also able to store
8603 special files. However, it imposes several restrictions as well:
8606 @item The maximum length of a file name is limited to 256 characters,
8607 provided that the file name can be split at a directory separator in
8608 two parts, first of them being at most 155 bytes long. So, in most
8609 cases the maximum file name length will be shorter than 256
8611 @item The maximum length of a symbolic link name is limited to
8613 @item Maximum size of a file the archive is able to accommodate
8615 @item Maximum value of UID/GID is 2097151.
8616 @item Maximum number of bits in device major and minor numbers is 21.
8620 Format used by J@"org Schilling @command{star}
8621 implementation. @GNUTAR{} is able to read @samp{star} archives but
8622 currently does not produce them.
8625 Archive format defined by @acronym{POSIX.1-2001} specification. This is the
8626 most flexible and feature-rich format. It does not impose any
8627 restrictions on file sizes or file name lengths. This format is quite
8628 recent, so not all tar implementations are able to handle it properly.
8629 However, this format is designed in such a way that any tar
8630 implementation able to read @samp{ustar} archives will be able to read
8631 most @samp{posix} archives as well, with the only exception that any
8632 additional information (such as long file names etc.) will in such
8633 case be extracted as plain text files along with the files it refers to.
8635 This archive format will be the default format for future versions
8640 The following table summarizes the limitations of each of these
8643 @multitable @columnfractions .10 .20 .20 .20 .20
8644 @headitem Format @tab UID @tab File Size @tab File Name @tab Devn
8645 @item gnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
8646 @item oldgnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
8647 @item v7 @tab 2097151 @tab 8GB @tab 99 @tab n/a
8648 @item ustar @tab 2097151 @tab 8GB @tab 256 @tab 21
8649 @item posix @tab Unlimited @tab Unlimited @tab Unlimited @tab Unlimited
8652 The default format for @GNUTAR{} is defined at compilation
8653 time. You may check it by running @command{tar --help}, and examining
8654 the last lines of its output. Usually, @GNUTAR{} is configured
8655 to create archives in @samp{gnu} format, however, future version will
8656 switch to @samp{posix}.
8659 * Compression:: Using Less Space through Compression
8660 * Attributes:: Handling File Attributes
8661 * Portability:: Making @command{tar} Archives More Portable
8662 * cpio:: Comparison of @command{tar} and @command{cpio}
8666 @section Using Less Space through Compression
8669 * gzip:: Creating and Reading Compressed Archives
8670 * sparse:: Archiving Sparse Files
8674 @subsection Creating and Reading Compressed Archives
8675 @cindex Compressed archives
8676 @cindex Storing archives in compressed format
8684 @GNUTAR{} is able to create and read compressed archives. It supports
8685 a wide variety of compression programs, namely: @command{gzip},
8686 @command{bzip2}, @command{lzip}, @command{lzma}, @command{lzop},
8687 @command{xz} and traditional @command{compress}. The latter is
8688 supported mostly for backward compatibility, and we recommend
8689 against using it, because it is by far less effective than the other
8690 compression programs@footnote{It also had patent problems in the past.}.
8692 Creating a compressed archive is simple: you just specify a
8693 @dfn{compression option} along with the usual archive creation
8694 commands. The compression option is @option{-z} (@option{--gzip}) to
8695 create a @command{gzip} compressed archive, @option{-j}
8696 (@option{--bzip2}) to create a @command{bzip2} compressed archive,
8697 @option{--lzip} to create an @asis{lzip} compressed archive,
8698 @option{-J} (@option{--xz}) to create an @asis{XZ} archive,
8699 @option{--lzma} to create an @asis{LZMA} compressed
8700 archive, @option{--lzop} to create an @asis{LSOP} archive, and
8701 @option{-Z} (@option{--compress}) to use @command{compress} program.
8705 $ @kbd{tar cfz archive.tar.gz .}
8708 You can also let @GNUTAR{} select the compression program based on
8709 the suffix of the archive file name. This is done using
8710 @option{--auto-compress} (@option{-a}) command line option. For
8711 example, the following invocation will use @command{bzip2} for
8715 $ @kbd{tar cfa archive.tar.bz2 .}
8719 whereas the following one will use @command{lzma}:
8722 $ @kbd{tar cfa archive.tar.lzma .}
8725 For a complete list of file name suffixes recognized by @GNUTAR{},
8726 see @ref{auto-compress}.
8728 Reading compressed archive is even simpler: you don't need to specify
8729 any additional options as @GNUTAR{} recognizes its format
8730 automatically. Thus, the following commands will list and extract the
8731 archive created in previous example:
8734 # List the compressed archive
8735 $ @kbd{tar tf archive.tar.gz}
8736 # Extract the compressed archive
8737 $ @kbd{tar xf archive.tar.gz}
8740 The format recognition algorithm is based on @dfn{signatures}, a
8741 special byte sequences in the beginning of file, that are specific for
8742 certain compression formats. If this approach fails, @command{tar}
8743 falls back to using archive name suffix to determine its format
8744 (@pxref{auto-compress}, for a list of recognized suffixes).
8746 The only case when you have to specify a decompression option while
8747 reading the archive is when reading from a pipe or from a tape drive
8748 that does not support random access. However, in this case @GNUTAR{}
8749 will indicate which option you should use. For example:
8752 $ @kbd{cat archive.tar.gz | tar tf -}
8753 tar: Archive is compressed. Use -z option
8754 tar: Error is not recoverable: exiting now
8757 If you see such diagnostics, just add the suggested option to the
8758 invocation of @GNUTAR{}:
8761 $ @kbd{cat archive.tar.gz | tar tfz -}
8764 Notice also, that there are several restrictions on operations on
8765 compressed archives. First of all, compressed archives cannot be
8766 modified, i.e., you cannot update (@option{--update}, alias @option{-u})
8767 them or delete (@option{--delete}) members from them or
8768 add (@option{--append}, alias @option{-r}) members to them. Likewise, you
8769 cannot append another @command{tar} archive to a compressed archive using
8770 @option{--concatenate} (@option{-A}). Secondly, multi-volume
8771 archives cannot be compressed.
8773 The following options allow to select a particular compressor program:
8781 Filter the archive through @command{gzip}.
8786 Filter the archive through @code{xz}.
8790 Filter the archive through @code{bzip2}.
8794 Filter the archive through @command{lzip}.
8798 Filter the archive through @command{lzma}.
8802 Filter the archive through @command{lzop}.
8809 Filter the archive through @command{compress}.
8812 When any of these options is given, @GNUTAR{} searches the compressor
8813 binary in the current path and invokes it. The name of the compressor
8814 program is specified at compilation time using a corresponding
8815 @option{--with-@var{compname}} option to @command{configure}, e.g.
8816 @option{--with-bzip2} to select a specific @command{bzip2} binary.
8817 @xref{lbzip2}, for a detailed discussion.
8819 The output produced by @command{tar --help} shows the actual
8820 compressor names along with each of these options.
8822 You can use any of these options on physical devices (tape drives,
8823 etc.) and remote files as well as on normal files; data to or from
8824 such devices or remote files is reblocked by another copy of the
8825 @command{tar} program to enforce the specified (or default) record
8826 size. The default compression parameters are used. Most compression
8827 programs allow to override these by setting a program-specific
8828 environment variable. For example, when using @command{gzip} you can
8829 use @env{GZIP} as in the example below:
8832 $ @kbd{GZIP=--best tar cfz archive.tar.gz subdir}
8836 Another way would be to use the @option{-I} option instead (see
8840 $ @kbd{tar -cf archive.tar.gz -I 'gzip --best' subdir}
8844 Finally, the third, traditional, way to achieve the same result is to
8848 $ @kbd{tar cf - subdir | gzip --best -c - > archive.tar.gz}
8851 @cindex corrupted archives
8852 About corrupted compressed archives: compressed files have no
8853 redundancy, for maximum compression. The adaptive nature of the
8854 compression scheme means that the compression tables are implicitly
8855 spread all over the archive. If you lose a few blocks, the dynamic
8856 construction of the compression tables becomes unsynchronized, and there
8857 is little chance that you could recover later in the archive.
8859 Another compression options provide a better control over creating
8860 compressed archives. These are:
8863 @anchor{auto-compress}
8864 @opindex auto-compress
8865 @item --auto-compress
8867 Select a compression program to use by the archive file name
8868 suffix. The following suffixes are recognized:
8870 @multitable @columnfractions 0.3 0.6
8871 @headitem Suffix @tab Compression program
8872 @item @samp{.gz} @tab @command{gzip}
8873 @item @samp{.tgz} @tab @command{gzip}
8874 @item @samp{.taz} @tab @command{gzip}
8875 @item @samp{.Z} @tab @command{compress}
8876 @item @samp{.taZ} @tab @command{compress}
8877 @item @samp{.bz2} @tab @command{bzip2}
8878 @item @samp{.tz2} @tab @command{bzip2}
8879 @item @samp{.tbz2} @tab @command{bzip2}
8880 @item @samp{.tbz} @tab @command{bzip2}
8881 @item @samp{.lz} @tab @command{lzip}
8882 @item @samp{.lzma} @tab @command{lzma}
8883 @item @samp{.tlz} @tab @command{lzma}
8884 @item @samp{.lzo} @tab @command{lzop}
8885 @item @samp{.xz} @tab @command{xz}
8888 @opindex use-compress-program
8889 @item --use-compress-program=@var{prog}
8890 @itemx -I=@var{prog}
8891 Use external compression program @var{prog}. Use this option if you
8892 are not happy with the compression program associated with the suffix
8893 at compile time or if you have a compression program that @GNUTAR{}
8894 does not support. There are two requirements to which @var{prog}
8897 First, when called without options, it should read data from standard
8898 input, compress it and output it on standard output.
8900 Secondly, if called with @option{-d} argument, it should do exactly
8901 the opposite, i.e., read the compressed data from the standard input
8902 and produce uncompressed data on the standard output.
8905 @cindex gpg, using with tar
8906 @cindex gnupg, using with tar
8907 @cindex Using encrypted archives
8908 The @option{--use-compress-program} option, in particular, lets you
8909 implement your own filters, not necessarily dealing with
8910 compression/decompression. For example, suppose you wish to implement
8911 PGP encryption on top of compression, using @command{gpg} (@pxref{Top,
8912 gpg, gpg ---- encryption and signing tool, gpg, GNU Privacy Guard
8913 Manual}). The following script does that:
8919 -d) gpg --decrypt - | gzip -d -c;;
8920 '') gzip -c | gpg -s;;
8921 *) echo "Unknown option $1">&2; exit 1;;
8926 Suppose you name it @file{gpgz} and save it somewhere in your
8927 @env{PATH}. Then the following command will create a compressed
8928 archive signed with your private key:
8931 $ @kbd{tar -cf foo.tar.gpgz -Igpgz .}
8935 Likewise, the command below will list its contents:
8938 $ @kbd{tar -tf foo.tar.gpgz -Igpgz .}
8942 The above is based on the following discussion:
8944 I have one question, or maybe it's a suggestion if there isn't a way
8945 to do it now. I would like to use @option{--gzip}, but I'd also like
8946 the output to be fed through a program like @acronym{GNU}
8947 @command{ecc} (actually, right now that's @samp{exactly} what I'd like
8948 to use :-)), basically adding ECC protection on top of compression.
8949 It seems as if this should be quite easy to do, but I can't work out
8950 exactly how to go about it. Of course, I can pipe the standard output
8951 of @command{tar} through @command{ecc}, but then I lose (though I
8952 haven't started using it yet, I confess) the ability to have
8953 @command{tar} use @command{rmt} for it's I/O (I think).
8955 I think the most straightforward thing would be to let me specify a
8956 general set of filters outboard of compression (preferably ordered,
8957 so the order can be automatically reversed on input operations, and
8958 with the options they require specifiable), but beggars shouldn't be
8959 choosers and anything you decide on would be fine with me.
8961 By the way, I like @command{ecc} but if (as the comments say) it can't
8962 deal with loss of block sync, I'm tempted to throw some time at adding
8963 that capability. Supposing I were to actually do such a thing and
8964 get it (apparently) working, do you accept contributed changes to
8965 utilities like that? (Leigh Clayton @file{loc@@soliton.com}, May 1995).
8967 Isn't that exactly the role of the
8968 @option{--use-compress-prog=@var{program}} option?
8969 I never tried it myself, but I suspect you may want to write a
8970 @var{prog} script or program able to filter stdin to stdout to
8971 way you want. It should recognize the @option{-d} option, for when
8972 extraction is needed rather than creation.
8974 It has been reported that if one writes compressed data (through the
8975 @option{--gzip} or @option{--compress} options) to a DLT and tries to use
8976 the DLT compression mode, the data will actually get bigger and one will
8977 end up with less space on the tape.
8981 * lbzip2:: Using lbzip2 with @GNUTAR{}.
8985 @subsubsection Using lbzip2 with @GNUTAR{}.
8987 @cindex Laszlo Ersek
8988 @command{Lbzip2} is a multithreaded utility for handling
8989 @samp{bzip2} compression, written by Laszlo Ersek. It makes use of
8990 multiple processors to speed up its operation and in general works
8991 considerably faster than @command{bzip2}. For a detailed description
8992 of @command{lbzip2} see @uref{http://freshmeat.net/@/projects/@/lbzip2} and
8993 @uref{http://www.linuxinsight.com/@/lbzip2-parallel-bzip2-utility.html,
8994 lbzip2: parallel bzip2 utility}.
8996 Recent versions of @command{lbzip2} are mostly command line compatible
8997 with @command{bzip2}, which makes it possible to automatically invoke
8998 it via the @option{--bzip2} @GNUTAR{} command line option. To do so,
8999 @GNUTAR{} must be configured with the @option{--with-bzip2} command
9000 line option, like this:
9003 $ @kbd{./configure --with-bzip2=lbzip2 [@var{other-options}]}
9006 Once configured and compiled this way, @command{tar --help} will show the
9011 $ @kbd{tar --help | grep -- --bzip2}
9012 -j, --bzip2 filter the archive through lbzip2
9017 which means that running @command{tar --bzip2} will invoke @command{lbzip2}.
9020 @subsection Archiving Sparse Files
9021 @cindex Sparse Files
9023 Files in the file system occasionally have @dfn{holes}. A @dfn{hole}
9024 in a file is a section of the file's contents which was never written.
9025 The contents of a hole reads as all zeros. On many operating systems,
9026 actual disk storage is not allocated for holes, but they are counted
9027 in the length of the file. If you archive such a file, @command{tar}
9028 could create an archive longer than the original. To have @command{tar}
9029 attempt to recognize the holes in a file, use @option{--sparse}
9030 (@option{-S}). When you use this option, then, for any file using
9031 less disk space than would be expected from its length, @command{tar}
9032 searches the file for consecutive stretches of zeros. It then records
9033 in the archive for the file where the consecutive stretches of zeros
9034 are, and only archives the ``real contents'' of the file. On
9035 extraction (using @option{--sparse} is not needed on extraction) any
9036 such files have holes created wherever the continuous stretches of zeros
9037 were found. Thus, if you use @option{--sparse}, @command{tar} archives
9038 won't take more space than the original.
9044 This option instructs @command{tar} to test each file for sparseness
9045 before attempting to archive it. If the file is found to be sparse it
9046 is treated specially, thus allowing to decrease the amount of space
9047 used by its image in the archive.
9049 This option is meaningful only when creating or updating archives. It
9050 has no effect on extraction.
9053 Consider using @option{--sparse} when performing file system backups,
9054 to avoid archiving the expanded forms of files stored sparsely in the
9057 Even if your system has no sparse files currently, some may be
9058 created in the future. If you use @option{--sparse} while making file
9059 system backups as a matter of course, you can be assured the archive
9060 will never take more space on the media than the files take on disk
9061 (otherwise, archiving a disk filled with sparse files might take
9062 hundreds of tapes). @xref{Incremental Dumps}.
9064 However, be aware that @option{--sparse} option presents a serious
9065 drawback. Namely, in order to determine if the file is sparse
9066 @command{tar} has to read it before trying to archive it, so in total
9067 the file is read @strong{twice}. So, always bear in mind that the
9068 time needed to process all files with this option is roughly twice
9069 the time needed to archive them without it.
9070 @FIXME{A technical note:
9072 Programs like @command{dump} do not have to read the entire file; by
9073 examining the file system directly, they can determine in advance
9074 exactly where the holes are and thus avoid reading through them. The
9075 only data it need read are the actual allocated data blocks.
9076 @GNUTAR{} uses a more portable and straightforward
9077 archiving approach, it would be fairly difficult that it does
9078 otherwise. Elizabeth Zwicky writes to @file{comp.unix.internals}, on
9082 What I did say is that you cannot tell the difference between a hole and an
9083 equivalent number of nulls without reading raw blocks. @code{st_blocks} at
9084 best tells you how many holes there are; it doesn't tell you @emph{where}.
9085 Just as programs may, conceivably, care what @code{st_blocks} is (care
9086 to name one that does?), they may also care where the holes are (I have
9087 no examples of this one either, but it's equally imaginable).
9089 I conclude from this that good archivers are not portable. One can
9090 arguably conclude that if you want a portable program, you can in good
9091 conscience restore files with as many holes as possible, since you can't
9096 @cindex sparse formats, defined
9097 When using @samp{POSIX} archive format, @GNUTAR{} is able to store
9098 sparse files using in three distinct ways, called @dfn{sparse
9099 formats}. A sparse format is identified by its @dfn{number},
9100 consisting, as usual of two decimal numbers, delimited by a dot. By
9101 default, format @samp{1.0} is used. If, for some reason, you wish to
9102 use an earlier format, you can select it using
9103 @option{--sparse-version} option.
9106 @opindex sparse-version
9107 @item --sparse-version=@var{version}
9109 Select the format to store sparse files in. Valid @var{version} values
9110 are: @samp{0.0}, @samp{0.1} and @samp{1.0}. @xref{Sparse Formats},
9111 for a detailed description of each format.
9114 Using @option{--sparse-format} option implies @option{--sparse}.
9117 @section Handling File Attributes
9118 @cindex atrributes, files
9119 @cindex file attributes
9121 When @command{tar} reads files, it updates their access times. To
9122 avoid this, use the @option{--atime-preserve[=METHOD]} option, which can either
9123 reset the access time retroactively or avoid changing it in the first
9127 @opindex atime-preserve
9128 @item --atime-preserve
9129 @itemx --atime-preserve=replace
9130 @itemx --atime-preserve=system
9131 Preserve the access times of files that are read. This works only for
9132 files that you own, unless you have superuser privileges.
9134 @option{--atime-preserve=replace} works on most systems, but it also
9135 restores the data modification time and updates the status change
9136 time. Hence it doesn't interact with incremental dumps nicely
9137 (@pxref{Incremental Dumps}), and it can set access or data modification times
9138 incorrectly if other programs access the file while @command{tar} is
9141 @option{--atime-preserve=system} avoids changing the access time in
9142 the first place, if the operating system supports this.
9143 Unfortunately, this may or may not work on any given operating system
9144 or file system. If @command{tar} knows for sure it won't work, it
9145 complains right away.
9147 Currently @option{--atime-preserve} with no operand defaults to
9148 @option{--atime-preserve=replace}, but this is intended to change to
9149 @option{--atime-preserve=system} when the latter is better-supported.
9154 Do not extract data modification time.
9156 When this option is used, @command{tar} leaves the data modification times
9157 of the files it extracts as the times when the files were extracted,
9158 instead of setting it to the times recorded in the archive.
9160 This option is meaningless with @option{--list} (@option{-t}).
9164 Create extracted files with the same ownership they have in the
9167 This is the default behavior for the superuser,
9168 so this option is meaningful only for non-root users, when @command{tar}
9169 is executed on those systems able to give files away. This is
9170 considered as a security flaw by many people, at least because it
9171 makes quite difficult to correctly account users for the disk space
9172 they occupy. Also, the @code{suid} or @code{sgid} attributes of
9173 files are easily and silently lost when files are given away.
9175 When writing an archive, @command{tar} writes the user @acronym{ID} and user name
9176 separately. If it can't find a user name (because the user @acronym{ID} is not
9177 in @file{/etc/passwd}), then it does not write one. When restoring,
9178 it tries to look the name (if one was written) up in
9179 @file{/etc/passwd}. If it fails, then it uses the user @acronym{ID} stored in
9180 the archive instead.
9182 @opindex no-same-owner
9183 @item --no-same-owner
9185 Do not attempt to restore ownership when extracting. This is the
9186 default behavior for ordinary users, so this option has an effect
9187 only for the superuser.
9189 @opindex numeric-owner
9190 @item --numeric-owner
9191 The @option{--numeric-owner} option allows (ANSI) archives to be written
9192 without user/group name information or such information to be ignored
9193 when extracting. It effectively disables the generation and/or use
9194 of user/group name information. This option forces extraction using
9195 the numeric ids from the archive, ignoring the names.
9197 This is useful in certain circumstances, when restoring a backup from
9198 an emergency floppy with different passwd/group files for example.
9199 It is otherwise impossible to extract files with the right ownerships
9200 if the password file in use during the extraction does not match the
9201 one belonging to the file system(s) being extracted. This occurs,
9202 for example, if you are restoring your files after a major crash and
9203 had booted from an emergency floppy with no password file or put your
9204 disk into another machine to do the restore.
9206 The numeric ids are @emph{always} saved into @command{tar} archives.
9207 The identifying names are added at create time when provided by the
9208 system, unless @option{--format=oldgnu} is used. Numeric ids could be
9209 used when moving archives between a collection of machines using
9210 a centralized management for attribution of numeric ids to users
9211 and groups. This is often made through using the NIS capabilities.
9213 When making a @command{tar} file for distribution to other sites, it
9214 is sometimes cleaner to use a single owner for all files in the
9215 distribution, and nicer to specify the write permission bits of the
9216 files as stored in the archive independently of their actual value on
9217 the file system. The way to prepare a clean distribution is usually
9218 to have some Makefile rule creating a directory, copying all needed
9219 files in that directory, then setting ownership and permissions as
9220 wanted (there are a lot of possible schemes), and only then making a
9221 @command{tar} archive out of this directory, before cleaning
9222 everything out. Of course, we could add a lot of options to
9223 @GNUTAR{} for fine tuning permissions and ownership.
9224 This is not the good way, I think. @GNUTAR{} is
9225 already crowded with options and moreover, the approach just explained
9226 gives you a great deal of control already.
9228 @xopindex{same-permissions, short description}
9229 @xopindex{preserve-permissions, short description}
9231 @itemx --same-permissions
9232 @itemx --preserve-permissions
9233 Extract all protection information.
9235 This option causes @command{tar} to set the modes (access permissions) of
9236 extracted files exactly as recorded in the archive. If this option
9237 is not used, the current @code{umask} setting limits the permissions
9238 on extracted files. This option is by default enabled when
9239 @command{tar} is executed by a superuser.
9242 This option is meaningless with @option{--list} (@option{-t}).
9246 Same as both @option{--same-permissions} and @option{--same-order}.
9248 This option is deprecated, and will be removed in @GNUTAR{} version 1.23.
9253 @section Making @command{tar} Archives More Portable
9255 Creating a @command{tar} archive on a particular system that is meant to be
9256 useful later on many other machines and with other versions of @command{tar}
9257 is more challenging than you might think. @command{tar} archive formats
9258 have been evolving since the first versions of Unix. Many such formats
9259 are around, and are not always compatible with each other. This section
9260 discusses a few problems, and gives some advice about making @command{tar}
9261 archives more portable.
9263 One golden rule is simplicity. For example, limit your @command{tar}
9264 archives to contain only regular files and directories, avoiding
9265 other kind of special files. Do not attempt to save sparse files or
9266 contiguous files as such. Let's discuss a few more problems, in turn.
9268 @FIXME{Discuss GNU extensions (incremental backups, multi-volume
9269 archives and archive labels) in GNU and PAX formats.}
9272 * Portable Names:: Portable Names
9273 * dereference:: Symbolic Links
9274 * hard links:: Hard Links
9275 * old:: Old V7 Archives
9276 * ustar:: Ustar Archives
9277 * gnu:: GNU and old GNU format archives.
9278 * posix:: @acronym{POSIX} archives
9279 * Checksumming:: Checksumming Problems
9280 * Large or Negative Values:: Large files, negative time stamps, etc.
9281 * Other Tars:: How to Extract GNU-Specific Data Using
9282 Other @command{tar} Implementations
9285 @node Portable Names
9286 @subsection Portable Names
9288 Use portable file and member names. A name is portable if it contains
9289 only @acronym{ASCII} letters and digits, @samp{/}, @samp{.}, @samp{_}, and
9290 @samp{-}; it cannot be empty, start with @samp{-} or @samp{//}, or
9291 contain @samp{/-}. Avoid deep directory nesting. For portability to
9292 old Unix hosts, limit your file name components to 14 characters or
9295 If you intend to have your @command{tar} archives to be read under
9296 MSDOS, you should not rely on case distinction for file names, and you
9297 might use the @acronym{GNU} @command{doschk} program for helping you
9298 further diagnosing illegal MSDOS names, which are even more limited
9302 @subsection Symbolic Links
9303 @cindex File names, using symbolic links
9304 @cindex Symbolic link as file name
9306 @opindex dereference
9307 Normally, when @command{tar} archives a symbolic link, it writes a
9308 block to the archive naming the target of the link. In that way, the
9309 @command{tar} archive is a faithful record of the file system contents.
9310 @option{--dereference} (@option{-h}) is used with @option{--create} (@option{-c}), and causes
9311 @command{tar} to archive the files symbolic links point to, instead of
9312 the links themselves. When this option is used, when @command{tar}
9313 encounters a symbolic link, it will archive the linked-to file,
9314 instead of simply recording the presence of a symbolic link.
9316 The name under which the file is stored in the file system is not
9317 recorded in the archive. To record both the symbolic link name and
9318 the file name in the system, archive the file under both names. If
9319 all links were recorded automatically by @command{tar}, an extracted file
9320 might be linked to a file name that no longer exists in the file
9323 If a linked-to file is encountered again by @command{tar} while creating
9324 the same archive, an entire second copy of it will be stored. (This
9325 @emph{might} be considered a bug.)
9327 So, for portable archives, do not archive symbolic links as such,
9328 and use @option{--dereference} (@option{-h}): many systems do not support
9329 symbolic links, and moreover, your distribution might be unusable if
9330 it contains unresolved symbolic links.
9333 @subsection Hard Links
9334 @cindex File names, using hard links
9335 @cindex hard links, dereferencing
9336 @cindex dereferencing hard links
9338 Normally, when @command{tar} archives a hard link, it writes a
9339 block to the archive naming the target of the link (a @samp{1} type
9340 block). In that way, the actual file contents is stored in file only
9341 once. For example, consider the following two files:
9346 -rw-r--r-- 2 gray staff 4 2007-10-30 15:11 one
9347 -rw-r--r-- 2 gray staff 4 2007-10-30 15:11 jeden
9351 Here, @file{jeden} is a link to @file{one}. When archiving this
9352 directory with a verbose level 2, you will get an output similar to
9356 $ tar cfvv ../archive.tar .
9357 drwxr-xr-x gray/staff 0 2007-10-30 15:13 ./
9358 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./jeden
9359 hrw-r--r-- gray/staff 0 2007-10-30 15:11 ./one link to ./jeden
9362 The last line shows that, instead of storing two copies of the file,
9363 @command{tar} stored it only once, under the name @file{jeden}, and
9364 stored file @file{one} as a hard link to this file.
9366 It may be important to know that all hard links to the given file are
9367 stored in the archive. For example, this may be necessary for exact
9368 reproduction of the file system. The following option does that:
9371 @xopindex{check-links, described}
9374 Check the number of links dumped for each processed file. If this
9375 number does not match the total number of hard links for the file, print
9379 For example, trying to archive only file @file{jeden} with this option
9380 produces the following diagnostics:
9383 $ tar -c -f ../archive.tar -l jeden
9384 tar: Missing links to `jeden'.
9387 Although creating special records for hard links helps keep a faithful
9388 record of the file system contents and makes archives more compact, it
9389 may present some difficulties when extracting individual members from
9390 the archive. For example, trying to extract file @file{one} from the
9391 archive created in previous examples produces, in the absense of file
9395 $ tar xf archive.tar ./one
9396 tar: ./one: Cannot hard link to `./jeden': No such file or directory
9397 tar: Error exit delayed from previous errors
9400 The reason for this behavior is that @command{tar} cannot seek back in
9401 the archive to the previous member (in this case, @file{one}), to
9402 extract it@footnote{There are plans to fix this in future releases.}.
9403 If you wish to avoid such problems at the cost of a bigger archive,
9404 use the following option:
9407 @xopindex{hard-dereference, described}
9408 @item --hard-dereference
9409 Dereference hard links and store the files they refer to.
9412 For example, trying this option on our two sample files, we get two
9413 copies in the archive, each of which can then be extracted
9414 independently of the other:
9418 $ tar -c -vv -f ../archive.tar --hard-dereference .
9419 drwxr-xr-x gray/staff 0 2007-10-30 15:13 ./
9420 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./jeden
9421 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./one
9426 @subsection Old V7 Archives
9427 @cindex Format, old style
9428 @cindex Old style format
9429 @cindex Old style archives
9430 @cindex v7 archive format
9432 Certain old versions of @command{tar} cannot handle additional
9433 information recorded by newer @command{tar} programs. To create an
9434 archive in V7 format (not ANSI), which can be read by these old
9435 versions, specify the @option{--format=v7} option in
9436 conjunction with the @option{--create} (@option{-c}) (@command{tar} also
9437 accepts @option{--portability} or @option{--old-archive} for this
9438 option). When you specify it,
9439 @command{tar} leaves out information about directories, pipes, fifos,
9440 contiguous files, and device files, and specifies file ownership by
9441 group and user IDs instead of group and user names.
9443 When updating an archive, do not use @option{--format=v7}
9444 unless the archive was created using this option.
9446 In most cases, a @emph{new} format archive can be read by an @emph{old}
9447 @command{tar} program without serious trouble, so this option should
9448 seldom be needed. On the other hand, most modern @command{tar}s are
9449 able to read old format archives, so it might be safer for you to
9450 always use @option{--format=v7} for your distributions. Notice,
9451 however, that @samp{ustar} format is a better alternative, as it is
9452 free from many of @samp{v7}'s drawbacks.
9455 @subsection Ustar Archive Format
9457 @cindex ustar archive format
9458 Archive format defined by @acronym{POSIX}.1-1988 specification is called
9459 @code{ustar}. Although it is more flexible than the V7 format, it
9460 still has many restrictions (@pxref{Formats,ustar}, for the detailed
9461 description of @code{ustar} format). Along with V7 format,
9462 @code{ustar} format is a good choice for archives intended to be read
9463 with other implementations of @command{tar}.
9465 To create archive in @code{ustar} format, use @option{--format=ustar}
9466 option in conjunction with the @option{--create} (@option{-c}).
9469 @subsection @acronym{GNU} and old @GNUTAR{} format
9471 @cindex GNU archive format
9472 @cindex Old GNU archive format
9473 @GNUTAR{} was based on an early draft of the
9474 @acronym{POSIX} 1003.1 @code{ustar} standard. @acronym{GNU} extensions to
9475 @command{tar}, such as the support for file names longer than 100
9476 characters, use portions of the @command{tar} header record which were
9477 specified in that @acronym{POSIX} draft as unused. Subsequent changes in
9478 @acronym{POSIX} have allocated the same parts of the header record for
9479 other purposes. As a result, @GNUTAR{} format is
9480 incompatible with the current @acronym{POSIX} specification, and with
9481 @command{tar} programs that follow it.
9483 In the majority of cases, @command{tar} will be configured to create
9484 this format by default. This will change in future releases, since
9485 we plan to make @samp{POSIX} format the default.
9487 To force creation a @GNUTAR{} archive, use option
9488 @option{--format=gnu}.
9491 @subsection @GNUTAR{} and @acronym{POSIX} @command{tar}
9493 @cindex POSIX archive format
9494 @cindex PAX archive format
9495 Starting from version 1.14 @GNUTAR{} features full support for
9496 @acronym{POSIX.1-2001} archives.
9498 A @acronym{POSIX} conformant archive will be created if @command{tar}
9499 was given @option{--format=posix} (@option{--format=pax}) option. No
9500 special option is required to read and extract from a @acronym{POSIX}
9504 * PAX keywords:: Controlling Extended Header Keywords.
9508 @subsubsection Controlling Extended Header Keywords
9512 @item --pax-option=@var{keyword-list}
9513 Handle keywords in @acronym{PAX} extended headers. This option is
9514 equivalent to @option{-o} option of the @command{pax} utility.
9517 @var{Keyword-list} is a comma-separated
9518 list of keyword options, each keyword option taking one of
9519 the following forms:
9522 @item delete=@var{pattern}
9523 When used with one of archive-creation commands,
9524 this option instructs @command{tar} to omit from extended header records
9525 that it produces any keywords matching the string @var{pattern}.
9527 When used in extract or list mode, this option instructs tar
9528 to ignore any keywords matching the given @var{pattern} in the extended
9529 header records. In both cases, matching is performed using the pattern
9530 matching notation described in @acronym{POSIX 1003.2}, 3.13
9531 (@pxref{wildcards}). For example:
9534 --pax-option delete=security.*
9537 would suppress security-related information.
9539 @item exthdr.name=@var{string}
9541 This keyword allows user control over the name that is written into the
9542 ustar header blocks for the extended headers. The name is obtained
9543 from @var{string} after making the following substitutions:
9545 @multitable @columnfractions .25 .55
9546 @headitem Meta-character @tab Replaced By
9547 @item %d @tab The directory name of the file, equivalent to the
9548 result of the @command{dirname} utility on the translated file name.
9549 @item %f @tab The name of the file with the directory information
9550 stripped, equivalent to the result of the @command{basename} utility
9551 on the translated file name.
9552 @item %p @tab The process @acronym{ID} of the @command{tar} process.
9553 @item %% @tab A @samp{%} character.
9556 Any other @samp{%} characters in @var{string} produce undefined
9559 If no option @samp{exthdr.name=string} is specified, @command{tar}
9560 will use the following default value:
9566 @item exthdr.mtime=@var{value}
9568 This keyword defines the value of the @samp{mtime} field that
9569 is written into the ustar header blocks for the extended headers.
9570 By default, the @samp{mtime} field is set to the modification time
9571 of the archive member described by that extended headers.
9573 @item globexthdr.name=@var{string}
9574 This keyword allows user control over the name that is written into
9575 the ustar header blocks for global extended header records. The name
9576 is obtained from the contents of @var{string}, after making
9577 the following substitutions:
9579 @multitable @columnfractions .25 .55
9580 @headitem Meta-character @tab Replaced By
9581 @item %n @tab An integer that represents the
9582 sequence number of the global extended header record in the archive,
9584 @item %p @tab The process @acronym{ID} of the @command{tar} process.
9585 @item %% @tab A @samp{%} character.
9588 Any other @samp{%} characters in @var{string} produce undefined results.
9590 If no option @samp{globexthdr.name=string} is specified, @command{tar}
9591 will use the following default value:
9594 $TMPDIR/GlobalHead.%p.%n
9598 where @samp{$TMPDIR} represents the value of the @var{TMPDIR}
9599 environment variable. If @var{TMPDIR} is not set, @command{tar}
9602 @item globexthdr.mtime=@var{value}
9604 This keyword defines the value of the @samp{mtime} field that
9605 is written into the ustar header blocks for the global extended headers.
9606 By default, the @samp{mtime} field is set to the time when
9607 @command{tar} was invoked.
9609 @item @var{keyword}=@var{value}
9610 When used with one of archive-creation commands, these keyword/value pairs
9611 will be included at the beginning of the archive in a global extended
9612 header record. When used with one of archive-reading commands,
9613 @command{tar} will behave as if it has encountered these keyword/value
9614 pairs at the beginning of the archive in a global extended header
9617 @item @var{keyword}:=@var{value}
9618 When used with one of archive-creation commands, these keyword/value pairs
9619 will be included as records at the beginning of an extended header for
9620 each file. This is effectively equivalent to @var{keyword}=@var{value}
9621 form except that it creates no global extended header records.
9623 When used with one of archive-reading commands, @command{tar} will
9624 behave as if these keyword/value pairs were included as records at the
9625 end of each extended header; thus, they will override any global or
9626 file-specific extended header record keywords of the same names.
9627 For example, in the command:
9630 tar --format=posix --create \
9631 --file archive --pax-option gname:=user .
9634 the group name will be forced to a new value for all files
9635 stored in the archive.
9638 In any of the forms described above, the @var{value} may be
9639 a string enclosed in curly braces. In that case, the string
9640 between the braces is understood either as a textual time
9641 representation, as described in @ref{Date input formats}, or a name of
9642 the existing file, starting with @samp{/} or @samp{.}. In the latter
9643 case, the modification time of that file is used.
9645 For example, to set all modification times to the current date, you
9646 use the following option:
9649 --pax-option='mtime:=@{now@}'
9652 Note quoting of the option's argument.
9654 @cindex archives, binary equivalent
9655 @cindex binary equivalent archives, creating
9656 As another example, here is the option that ensures that any two
9657 archives created using it, will be binary equivalent if they have the
9661 --pax-option=exthdr.name=%d/PaxHeaders/%f,atime:=0
9665 @subsection Checksumming Problems
9667 SunOS and HP-UX @command{tar} fail to accept archives created using
9668 @GNUTAR{} and containing non-@acronym{ASCII} file names, that
9669 is, file names having characters with the eight bit set, because they
9670 use signed checksums, while @GNUTAR{} uses unsigned
9671 checksums while creating archives, as per @acronym{POSIX} standards. On
9672 reading, @GNUTAR{} computes both checksums and
9673 accepts any. It is somewhat worrying that a lot of people may go
9674 around doing backup of their files using faulty (or at least
9675 non-standard) software, not learning about it until it's time to
9676 restore their missing files with an incompatible file extractor, or
9679 @GNUTAR{} computes checksums both ways, and accept
9680 any on read, so @acronym{GNU} tar can read Sun tapes even with their
9681 wrong checksums. @GNUTAR{} produces the standard
9682 checksum, however, raising incompatibilities with Sun. That is to
9683 say, @GNUTAR{} has not been modified to
9684 @emph{produce} incorrect archives to be read by buggy @command{tar}'s.
9685 I've been told that more recent Sun @command{tar} now read standard
9686 archives, so maybe Sun did a similar patch, after all?
9688 The story seems to be that when Sun first imported @command{tar}
9689 sources on their system, they recompiled it without realizing that
9690 the checksums were computed differently, because of a change in
9691 the default signing of @code{char}'s in their compiler. So they
9692 started computing checksums wrongly. When they later realized their
9693 mistake, they merely decided to stay compatible with it, and with
9694 themselves afterwards. Presumably, but I do not really know, HP-UX
9695 has chosen that their @command{tar} archives to be compatible with Sun's.
9696 The current standards do not favor Sun @command{tar} format. In any
9697 case, it now falls on the shoulders of SunOS and HP-UX users to get
9698 a @command{tar} able to read the good archives they receive.
9700 @node Large or Negative Values
9701 @subsection Large or Negative Values
9702 @cindex large values
9703 @cindex future time stamps
9704 @cindex negative time stamps
9707 The above sections suggest to use @samp{oldest possible} archive
9708 format if in doubt. However, sometimes it is not possible. If you
9709 attempt to archive a file whose metadata cannot be represented using
9710 required format, @GNUTAR{} will print error message and ignore such a
9711 file. You will than have to switch to a format that is able to
9712 handle such values. The format summary table (@pxref{Formats}) will
9715 In particular, when trying to archive files larger than 8GB or with
9716 timestamps not in the range 1970-01-01 00:00:00 through 2242-03-16
9717 12:56:31 @sc{utc}, you will have to chose between @acronym{GNU} and
9718 @acronym{POSIX} archive formats. When considering which format to
9719 choose, bear in mind that the @acronym{GNU} format uses
9720 two's-complement base-256 notation to store values that do not fit
9721 into standard @acronym{ustar} range. Such archives can generally be
9722 read only by a @GNUTAR{} implementation. Moreover, they sometimes
9723 cannot be correctly restored on another hosts even by @GNUTAR{}. For
9724 example, using two's complement representation for negative time
9725 stamps that assumes a signed 32-bit @code{time_t} generates archives
9726 that are not portable to hosts with differing @code{time_t}
9729 On the other hand, @acronym{POSIX} archives, generally speaking, can
9730 be extracted by any tar implementation that understands older
9731 @acronym{ustar} format. The only exception are files larger than 8GB.
9733 @FIXME{Describe how @acronym{POSIX} archives are extracted by non
9737 @subsection How to Extract GNU-Specific Data Using Other @command{tar} Implementations
9739 In previous sections you became acquainted with various quirks
9740 necessary to make your archives portable. Sometimes you may need to
9741 extract archives containing GNU-specific members using some
9742 third-party @command{tar} implementation or an older version of
9743 @GNUTAR{}. Of course your best bet is to have @GNUTAR{} installed,
9744 but if it is for some reason impossible, this section will explain
9745 how to cope without it.
9747 When we speak about @dfn{GNU-specific} members we mean two classes of
9748 them: members split between the volumes of a multi-volume archive and
9749 sparse members. You will be able to always recover such members if
9750 the archive is in PAX format. In addition split members can be
9751 recovered from archives in old GNU format. The following subsections
9752 describe the required procedures in detail.
9755 * Split Recovery:: Members Split Between Volumes
9756 * Sparse Recovery:: Sparse Members
9759 @node Split Recovery
9760 @subsubsection Extracting Members Split Between Volumes
9762 @cindex Mutli-volume archives, extracting using non-GNU tars
9763 If a member is split between several volumes of an old GNU format archive
9764 most third party @command{tar} implementation will fail to extract
9765 it. To extract it, use @command{tarcat} program (@pxref{Tarcat}).
9766 This program is available from
9767 @uref{http://www.gnu.org/@/software/@/tar/@/utils/@/tarcat.html, @GNUTAR{}
9768 home page}. It concatenates several archive volumes into a single
9769 valid archive. For example, if you have three volumes named from
9770 @file{vol-1.tar} to @file{vol-3.tar}, you can do the following to
9771 extract them using a third-party @command{tar}:
9774 $ @kbd{tarcat vol-1.tar vol-2.tar vol-3.tar | tar xf -}
9777 @cindex Mutli-volume archives in PAX format, extracting using non-GNU tars
9778 You could use this approach for most (although not all) PAX
9779 format archives as well. However, extracting split members from a PAX
9780 archive is a much easier task, because PAX volumes are constructed in
9781 such a way that each part of a split member is extracted to a
9782 different file by @command{tar} implementations that are not aware of
9783 GNU extensions. More specifically, the very first part retains its
9784 original name, and all subsequent parts are named using the pattern:
9787 %d/GNUFileParts.%p/%f.%n
9791 where symbols preceeded by @samp{%} are @dfn{macro characters} that
9792 have the following meaning:
9794 @multitable @columnfractions .25 .55
9795 @headitem Meta-character @tab Replaced By
9796 @item %d @tab The directory name of the file, equivalent to the
9797 result of the @command{dirname} utility on its full name.
9798 @item %f @tab The file name of the file, equivalent to the result
9799 of the @command{basename} utility on its full name.
9800 @item %p @tab The process @acronym{ID} of the @command{tar} process that
9801 created the archive.
9802 @item %n @tab Ordinal number of this particular part.
9805 For example, if the file @file{var/longfile} was split during archive
9806 creation between three volumes, and the creator @command{tar} process
9807 had process @acronym{ID} @samp{27962}, then the member names will be:
9811 var/GNUFileParts.27962/longfile.1
9812 var/GNUFileParts.27962/longfile.2
9815 When you extract your archive using a third-party @command{tar}, these
9816 files will be created on your disk, and the only thing you will need
9817 to do to restore your file in its original form is concatenate them in
9818 the proper order, for example:
9823 $ @kbd{cat GNUFileParts.27962/longfile.1 \
9824 GNUFileParts.27962/longfile.2 >> longfile}
9825 $ rm -f GNUFileParts.27962
9829 Notice, that if the @command{tar} implementation you use supports PAX
9830 format archives, it will probably emit warnings about unknown keywords
9831 during extraction. They will look like this:
9836 Unknown extended header keyword 'GNU.volume.filename' ignored.
9837 Unknown extended header keyword 'GNU.volume.size' ignored.
9838 Unknown extended header keyword 'GNU.volume.offset' ignored.
9843 You can safely ignore these warnings.
9845 If your @command{tar} implementation is not PAX-aware, you will get
9846 more warnings and more files generated on your disk, e.g.:
9850 $ @kbd{tar xf vol-1.tar}
9851 var/PaxHeaders.27962/longfile: Unknown file type 'x', extracted as
9853 Unexpected EOF in archive
9854 $ @kbd{tar xf vol-2.tar}
9855 tmp/GlobalHead.27962.1: Unknown file type 'g', extracted as normal file
9856 GNUFileParts.27962/PaxHeaders.27962/sparsefile.1: Unknown file type
9857 'x', extracted as normal file
9861 Ignore these warnings. The @file{PaxHeaders.*} directories created
9862 will contain files with @dfn{extended header keywords} describing the
9863 extracted files. You can delete them, unless they describe sparse
9864 members. Read further to learn more about them.
9866 @node Sparse Recovery
9867 @subsubsection Extracting Sparse Members
9869 @cindex sparse files, extracting with non-GNU tars
9870 Any @command{tar} implementation will be able to extract sparse members from a
9871 PAX archive. However, the extracted files will be @dfn{condensed},
9872 i.e., any zero blocks will be removed from them. When we restore such
9873 a condensed file to its original form, by adding zero blocks (or
9874 @dfn{holes}) back to their original locations, we call this process
9875 @dfn{expanding} a compressed sparse file.
9878 To expand a file, you will need a simple auxiliary program called
9879 @command{xsparse}. It is available in source form from
9880 @uref{http://www.gnu.org/@/software/@/tar/@/utils/@/xsparse.html, @GNUTAR{}
9883 @cindex sparse files v.1.0, extracting with non-GNU tars
9884 Let's begin with archive members in @dfn{sparse format
9885 version 1.0}@footnote{@xref{PAX 1}.}, which are the easiest to expand.
9886 The condensed file will contain both file map and file data, so no
9887 additional data will be needed to restore it. If the original file
9888 name was @file{@var{dir}/@var{name}}, then the condensed file will be
9889 named @file{@var{dir}/@/GNUSparseFile.@var{n}/@/@var{name}}, where
9890 @var{n} is a decimal number@footnote{Technically speaking, @var{n} is a
9891 @dfn{process @acronym{ID}} of the @command{tar} process which created the
9892 archive (@pxref{PAX keywords}).}.
9894 To expand a version 1.0 file, run @command{xsparse} as follows:
9897 $ @kbd{xsparse @file{cond-file}}
9901 where @file{cond-file} is the name of the condensed file. The utility
9902 will deduce the name for the resulting expanded file using the
9903 following algorithm:
9906 @item If @file{cond-file} does not contain any directories,
9907 @file{../cond-file} will be used;
9909 @item If @file{cond-file} has the form
9910 @file{@var{dir}/@var{t}/@var{name}}, where both @var{t} and @var{name}
9911 are simple names, with no @samp{/} characters in them, the output file
9912 name will be @file{@var{dir}/@var{name}}.
9914 @item Otherwise, if @file{cond-file} has the form
9915 @file{@var{dir}/@var{name}}, the output file name will be
9919 In the unlikely case when this algorithm does not suit your needs,
9920 you can explicitly specify output file name as a second argument to
9924 $ @kbd{xsparse @file{cond-file} @file{out-file}}
9927 It is often a good idea to run @command{xsparse} in @dfn{dry run} mode
9928 first. In this mode, the command does not actually expand the file,
9929 but verbosely lists all actions it would be taking to do so. The dry
9930 run mode is enabled by @option{-n} command line argument:
9934 $ @kbd{xsparse -n /home/gray/GNUSparseFile.6058/sparsefile}
9935 Reading v.1.0 sparse map
9936 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
9937 `/home/gray/sparsefile'
9942 To actually expand the file, you would run:
9945 $ @kbd{xsparse /home/gray/GNUSparseFile.6058/sparsefile}
9949 The program behaves the same way all UNIX utilities do: it will keep
9950 quiet unless it has simething important to tell you (e.g. an error
9951 condition or something). If you wish it to produce verbose output,
9952 similar to that from the dry run mode, use @option{-v} option:
9956 $ @kbd{xsparse -v /home/gray/GNUSparseFile.6058/sparsefile}
9957 Reading v.1.0 sparse map
9958 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
9959 `/home/gray/sparsefile'
9964 Additionally, if your @command{tar} implementation has extracted the
9965 @dfn{extended headers} for this file, you can instruct @command{xstar}
9966 to use them in order to verify the integrity of the expanded file.
9967 The option @option{-x} sets the name of the extended header file to
9968 use. Continuing our example:
9972 $ @kbd{xsparse -v -x /home/gray/PaxHeaders.6058/sparsefile \
9973 /home/gray/GNUSparseFile.6058/sparsefile}
9974 Reading extended header file
9975 Found variable GNU.sparse.major = 1
9976 Found variable GNU.sparse.minor = 0
9977 Found variable GNU.sparse.name = sparsefile
9978 Found variable GNU.sparse.realsize = 217481216
9979 Reading v.1.0 sparse map
9980 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
9981 `/home/gray/sparsefile'
9986 @anchor{extracting sparse v.0.x}
9987 @cindex sparse files v.0.1, extracting with non-GNU tars
9988 @cindex sparse files v.0.0, extracting with non-GNU tars
9989 An @dfn{extended header} is a special @command{tar} archive header
9990 that precedes an archive member and contains a set of
9991 @dfn{variables}, describing the member properties that cannot be
9992 stored in the standard @code{ustar} header. While optional for
9993 expanding sparse version 1.0 members, the use of extended headers is
9994 mandatory when expanding sparse members in older sparse formats: v.0.0
9995 and v.0.1 (The sparse formats are described in detail in @ref{Sparse
9996 Formats}.) So, for these formats, the question is: how to obtain
9997 extended headers from the archive?
9999 If you use a @command{tar} implementation that does not support PAX
10000 format, extended headers for each member will be extracted as a
10001 separate file. If we represent the member name as
10002 @file{@var{dir}/@var{name}}, then the extended header file will be
10003 named @file{@var{dir}/@/PaxHeaders.@var{n}/@/@var{name}}, where
10004 @var{n} is an integer number.
10006 Things become more difficult if your @command{tar} implementation
10007 does support PAX headers, because in this case you will have to
10008 manually extract the headers. We recommend the following algorithm:
10012 Consult the documentation of your @command{tar} implementation for an
10013 option that prints @dfn{block numbers} along with the archive
10014 listing (analogous to @GNUTAR{}'s @option{-R} option). For example,
10015 @command{star} has @option{-block-number}.
10018 Obtain verbose listing using the @samp{block number} option, and
10019 find block numbers of the sparse member in question and the member
10020 immediately following it. For example, running @command{star} on our
10025 $ @kbd{star -t -v -block-number -f arc.tar}
10027 star: Unknown extended header keyword 'GNU.sparse.size' ignored.
10028 star: Unknown extended header keyword 'GNU.sparse.numblocks' ignored.
10029 star: Unknown extended header keyword 'GNU.sparse.name' ignored.
10030 star: Unknown extended header keyword 'GNU.sparse.map' ignored.
10031 block 56: 425984 -rw-r--r-- gray/users Jun 25 14:46 2006 GNUSparseFile.28124/sparsefile
10032 block 897: 65391 -rw-r--r-- gray/users Jun 24 20:06 2006 README
10038 (as usual, ignore the warnings about unknown keywords.)
10041 Let @var{size} be the size of the sparse member, @var{Bs} be its block number
10042 and @var{Bn} be the block number of the next member.
10046 @var{N} = @var{Bs} - @var{Bn} - @var{size}/512 - 2
10050 This number gives the size of the extended header part in tar @dfn{blocks}.
10051 In our example, this formula gives: @code{897 - 56 - 425984 / 512 - 2
10055 Use @command{dd} to extract the headers:
10058 @kbd{dd if=@var{archive} of=@var{hname} bs=512 skip=@var{Bs} count=@var{N}}
10062 where @var{archive} is the archive name, @var{hname} is a name of the
10063 file to store the extended header in, @var{Bs} and @var{N} are
10064 computed in previous steps.
10066 In our example, this command will be
10069 $ @kbd{dd if=arc.tar of=xhdr bs=512 skip=56 count=7}
10073 Finally, you can expand the condensed file, using the obtained header:
10077 $ @kbd{xsparse -v -x xhdr GNUSparseFile.6058/sparsefile}
10078 Reading extended header file
10079 Found variable GNU.sparse.size = 217481216
10080 Found variable GNU.sparse.numblocks = 208
10081 Found variable GNU.sparse.name = sparsefile
10082 Found variable GNU.sparse.map = 0,2048,1050624,2048,@dots{}
10083 Expanding file `GNUSparseFile.28124/sparsefile' to `sparsefile'
10089 @section Comparison of @command{tar} and @command{cpio}
10092 @FIXME{Reorganize the following material}
10094 The @command{cpio} archive formats, like @command{tar}, do have maximum
10095 file name lengths. The binary and old @acronym{ASCII} formats have a maximum file
10096 length of 256, and the new @acronym{ASCII} and @acronym{CRC ASCII} formats have a max
10097 file length of 1024. @acronym{GNU} @command{cpio} can read and write archives
10098 with arbitrary file name lengths, but other @command{cpio} implementations
10099 may crash unexplainedly trying to read them.
10101 @command{tar} handles symbolic links in the form in which it comes in @acronym{BSD};
10102 @command{cpio} doesn't handle symbolic links in the form in which it comes
10103 in System V prior to SVR4, and some vendors may have added symlinks
10104 to their system without enhancing @command{cpio} to know about them.
10105 Others may have enhanced it in a way other than the way I did it
10106 at Sun, and which was adopted by AT&T (and which is, I think, also
10107 present in the @command{cpio} that Berkeley picked up from AT&T and put
10108 into a later @acronym{BSD} release---I think I gave them my changes).
10110 (SVR4 does some funny stuff with @command{tar}; basically, its @command{cpio}
10111 can handle @command{tar} format input, and write it on output, and it
10112 probably handles symbolic links. They may not have bothered doing
10113 anything to enhance @command{tar} as a result.)
10115 @command{cpio} handles special files; traditional @command{tar} doesn't.
10117 @command{tar} comes with V7, System III, System V, and @acronym{BSD} source;
10118 @command{cpio} comes only with System III, System V, and later @acronym{BSD}
10119 (4.3-tahoe and later).
10121 @command{tar}'s way of handling multiple hard links to a file can handle
10122 file systems that support 32-bit i-numbers (e.g., the @acronym{BSD} file system);
10123 @command{cpio}s way requires you to play some games (in its ``binary''
10124 format, i-numbers are only 16 bits, and in its ``portable @acronym{ASCII}'' format,
10125 they're 18 bits---it would have to play games with the "file system @acronym{ID}"
10126 field of the header to make sure that the file system @acronym{ID}/i-number pairs
10127 of different files were always different), and I don't know which
10128 @command{cpio}s, if any, play those games. Those that don't might get
10129 confused and think two files are the same file when they're not, and
10130 make hard links between them.
10132 @command{tar}s way of handling multiple hard links to a file places only
10133 one copy of the link on the tape, but the name attached to that copy
10134 is the @emph{only} one you can use to retrieve the file; @command{cpio}s
10135 way puts one copy for every link, but you can retrieve it using any
10139 What type of check sum (if any) is used, and how is this calculated.
10142 See the attached manual pages for @command{tar} and @command{cpio} format.
10143 @command{tar} uses a checksum which is the sum of all the bytes in the
10144 @command{tar} header for a file; @command{cpio} uses no checksum.
10147 If anyone knows why @command{cpio} was made when @command{tar} was present
10151 It wasn't. @command{cpio} first showed up in PWB/UNIX 1.0; no
10152 generally-available version of UNIX had @command{tar} at the time. I don't
10153 know whether any version that was generally available @emph{within AT&T}
10154 had @command{tar}, or, if so, whether the people within AT&T who did
10155 @command{cpio} knew about it.
10157 On restore, if there is a corruption on a tape @command{tar} will stop at
10158 that point, while @command{cpio} will skip over it and try to restore the
10161 The main difference is just in the command syntax and header format.
10163 @command{tar} is a little more tape-oriented in that everything is blocked
10164 to start on a record boundary.
10167 Is there any differences between the ability to recover crashed
10168 archives between the two of them. (Is there any chance of recovering
10169 crashed archives at all.)
10172 Theoretically it should be easier under @command{tar} since the blocking
10173 lets you find a header with some variation of @samp{dd skip=@var{nn}}.
10174 However, modern @command{cpio}'s and variations have an option to just
10175 search for the next file header after an error with a reasonable chance
10176 of resyncing. However, lots of tape driver software won't allow you to
10177 continue past a media error which should be the only reason for getting
10178 out of sync unless a file changed sizes while you were writing the
10182 If anyone knows why @command{cpio} was made when @command{tar} was present
10183 at the unix scene, please tell me about this too.
10186 Probably because it is more media efficient (by not blocking everything
10187 and using only the space needed for the headers where @command{tar}
10188 always uses 512 bytes per file header) and it knows how to archive
10191 You might want to look at the freely available alternatives. The
10192 major ones are @command{afio}, @GNUTAR{}, and
10193 @command{pax}, each of which have their own extensions with some
10194 backwards compatibility.
10196 Sparse files were @command{tar}red as sparse files (which you can
10197 easily test, because the resulting archive gets smaller, and
10198 @acronym{GNU} @command{cpio} can no longer read it).
10201 @chapter Tapes and Other Archive Media
10204 A few special cases about tape handling warrant more detailed
10205 description. These special cases are discussed below.
10207 Many complexities surround the use of @command{tar} on tape drives. Since
10208 the creation and manipulation of archives located on magnetic tape was
10209 the original purpose of @command{tar}, it contains many features making
10210 such manipulation easier.
10212 Archives are usually written on dismountable media---tape cartridges,
10213 mag tapes, or floppy disks.
10215 The amount of data a tape or disk holds depends not only on its size,
10216 but also on how it is formatted. A 2400 foot long reel of mag tape
10217 holds 40 megabytes of data when formatted at 1600 bits per inch. The
10218 physically smaller EXABYTE tape cartridge holds 2.3 gigabytes.
10220 Magnetic media are re-usable---once the archive on a tape is no longer
10221 needed, the archive can be erased and the tape or disk used over.
10222 Media quality does deteriorate with use, however. Most tapes or disks
10223 should be discarded when they begin to produce data errors. EXABYTE
10224 tape cartridges should be discarded when they generate an @dfn{error
10225 count} (number of non-usable bits) of more than 10k.
10227 Magnetic media are written and erased using magnetic fields, and
10228 should be protected from such fields to avoid damage to stored data.
10229 Sticking a floppy disk to a filing cabinet using a magnet is probably
10233 * Device:: Device selection and switching
10234 * Remote Tape Server::
10235 * Common Problems and Solutions::
10236 * Blocking:: Blocking
10237 * Many:: Many archives on one tape
10238 * Using Multiple Tapes:: Using Multiple Tapes
10239 * label:: Including a Label in the Archive
10241 * Write Protection::
10245 @section Device Selection and Switching
10249 @item -f [@var{hostname}:]@var{file}
10250 @itemx --file=[@var{hostname}:]@var{file}
10251 Use archive file or device @var{file} on @var{hostname}.
10254 This option is used to specify the file name of the archive @command{tar}
10257 If the file name is @samp{-}, @command{tar} reads the archive from standard
10258 input (when listing or extracting), or writes it to standard output
10259 (when creating). If the @samp{-} file name is given when updating an
10260 archive, @command{tar} will read the original archive from its standard
10261 input, and will write the entire new archive to its standard output.
10263 If the file name contains a @samp{:}, it is interpreted as
10264 @samp{hostname:file name}. If the @var{hostname} contains an @dfn{at}
10265 sign (@samp{@@}), it is treated as @samp{user@@hostname:file name}. In
10266 either case, @command{tar} will invoke the command @command{rsh} (or
10267 @command{remsh}) to start up an @command{/usr/libexec/rmt} on the remote
10268 machine. If you give an alternate login name, it will be given to the
10270 Naturally, the remote machine must have an executable
10271 @command{/usr/libexec/rmt}. This program is free software from the
10272 University of California, and a copy of the source code can be found
10273 with the sources for @command{tar}; it's compiled and installed by default.
10274 The exact path to this utility is determined when configuring the package.
10275 It is @file{@var{prefix}/libexec/rmt}, where @var{prefix} stands for
10276 your installation prefix. This location may also be overridden at
10277 runtime by using the @option{--rmt-command=@var{command}} option (@xref{Option Summary,
10278 ---rmt-command}, for detailed description of this option. @xref{Remote
10279 Tape Server}, for the description of @command{rmt} command).
10281 If this option is not given, but the environment variable @env{TAPE}
10282 is set, its value is used; otherwise, old versions of @command{tar}
10283 used a default archive name (which was picked when @command{tar} was
10284 compiled). The default is normally set up to be the @dfn{first} tape
10285 drive or other transportable I/O medium on the system.
10287 Starting with version 1.11.5, @GNUTAR{} uses
10288 standard input and standard output as the default device, and I will
10289 not try anymore supporting automatic device detection at installation
10290 time. This was failing really in too many cases, it was hopeless.
10291 This is now completely left to the installer to override standard
10292 input and standard output for default device, if this seems
10293 preferable. Further, I think @emph{most} actual usages of
10294 @command{tar} are done with pipes or disks, not really tapes,
10295 cartridges or diskettes.
10297 Some users think that using standard input and output is running
10298 after trouble. This could lead to a nasty surprise on your screen if
10299 you forget to specify an output file name---especially if you are going
10300 through a network or terminal server capable of buffering large amounts
10301 of output. We had so many bug reports in that area of configuring
10302 default tapes automatically, and so many contradicting requests, that
10303 we finally consider the problem to be portably intractable. We could
10304 of course use something like @samp{/dev/tape} as a default, but this
10305 is @emph{also} running after various kind of trouble, going from hung
10306 processes to accidental destruction of real tapes. After having seen
10307 all this mess, using standard input and output as a default really
10308 sounds like the only clean choice left, and a very useful one too.
10310 @GNUTAR{} reads and writes archive in records, I
10311 suspect this is the main reason why block devices are preferred over
10312 character devices. Most probably, block devices are more efficient
10313 too. The installer could also check for @samp{DEFTAPE} in
10314 @file{<sys/mtio.h>}.
10317 @xopindex{force-local, short description}
10318 @item --force-local
10319 Archive file is local even if it contains a colon.
10321 @opindex rsh-command
10322 @item --rsh-command=@var{command}
10323 Use remote @var{command} instead of @command{rsh}. This option exists
10324 so that people who use something other than the standard @command{rsh}
10325 (e.g., a Kerberized @command{rsh}) can access a remote device.
10327 When this command is not used, the shell command found when
10328 the @command{tar} program was installed is used instead. This is
10329 the first found of @file{/usr/ucb/rsh}, @file{/usr/bin/remsh},
10330 @file{/usr/bin/rsh}, @file{/usr/bsd/rsh} or @file{/usr/bin/nsh}.
10331 The installer may have overridden this by defining the environment
10332 variable @env{RSH} @emph{at installation time}.
10335 Specify drive and density.
10337 @xopindex{multi-volume, short description}
10339 @itemx --multi-volume
10340 Create/list/extract multi-volume archive.
10342 This option causes @command{tar} to write a @dfn{multi-volume} archive---one
10343 that may be larger than will fit on the medium used to hold it.
10344 @xref{Multi-Volume Archives}.
10346 @xopindex{tape-length, short description}
10348 @itemx --tape-length=@var{num}
10349 Change tape after writing @var{num} x 1024 bytes.
10351 This option might be useful when your tape drivers do not properly
10352 detect end of physical tapes. By being slightly conservative on the
10353 maximum tape length, you might avoid the problem entirely.
10355 @xopindex{info-script, short description}
10356 @xopindex{new-volume-script, short description}
10357 @item -F @var{file}
10358 @itemx --info-script=@var{file}
10359 @itemx --new-volume-script=@var{file}
10360 Execute @file{file} at end of each tape. This implies
10361 @option{--multi-volume} (@option{-M}). @xref{info-script}, for a detailed
10362 description of this option.
10365 @node Remote Tape Server
10366 @section Remote Tape Server
10368 @cindex remote tape drive
10370 In order to access the tape drive on a remote machine, @command{tar}
10371 uses the remote tape server written at the University of California at
10372 Berkeley. The remote tape server must be installed as
10373 @file{@var{prefix}/libexec/rmt} on any machine whose tape drive you
10374 want to use. @command{tar} calls @command{rmt} by running an
10375 @command{rsh} or @command{remsh} to the remote machine, optionally
10376 using a different login name if one is supplied.
10378 A copy of the source for the remote tape server is provided. It is
10379 Copyright @copyright{} 1983 by the Regents of the University of
10380 California, but can be freely distributed. It is compiled and
10381 installed by default.
10383 @cindex absolute file names
10384 Unless you use the @option{--absolute-names} (@option{-P}) option,
10385 @GNUTAR{} will not allow you to create an archive that contains
10386 absolute file names (a file name beginning with @samp{/}.) If you try,
10387 @command{tar} will automatically remove the leading @samp{/} from the
10388 file names it stores in the archive. It will also type a warning
10389 message telling you what it is doing.
10391 When reading an archive that was created with a different
10392 @command{tar} program, @GNUTAR{} automatically
10393 extracts entries in the archive which have absolute file names as if
10394 the file names were not absolute. This is an important feature. A
10395 visitor here once gave a @command{tar} tape to an operator to restore;
10396 the operator used Sun @command{tar} instead of @GNUTAR{},
10397 and the result was that it replaced large portions of
10398 our @file{/bin} and friends with versions from the tape; needless to
10399 say, we were unhappy about having to recover the file system from
10402 For example, if the archive contained a file @file{/usr/bin/computoy},
10403 @GNUTAR{} would extract the file to @file{usr/bin/computoy},
10404 relative to the current directory. If you want to extract the files in
10405 an archive to the same absolute names that they had when the archive
10406 was created, you should do a @samp{cd /} before extracting the files
10407 from the archive, or you should either use the @option{--absolute-names}
10408 option, or use the command @samp{tar -C / @dots{}}.
10410 @cindex Ultrix 3.1 and write failure
10411 Some versions of Unix (Ultrix 3.1 is known to have this problem),
10412 can claim that a short write near the end of a tape succeeded,
10413 when it actually failed. This will result in the -M option not
10414 working correctly. The best workaround at the moment is to use a
10415 significantly larger blocking factor than the default 20.
10417 In order to update an archive, @command{tar} must be able to backspace the
10418 archive in order to reread or rewrite a record that was just read (or
10419 written). This is currently possible only on two kinds of files: normal
10420 disk files (or any other file that can be backspaced with @samp{lseek}),
10421 and industry-standard 9-track magnetic tape (or any other kind of tape
10422 that can be backspaced with the @code{MTIOCTOP} @code{ioctl}).
10424 This means that the @option{--append}, @option{--concatenate}, and
10425 @option{--delete} commands will not work on any other kind of file.
10426 Some media simply cannot be backspaced, which means these commands and
10427 options will never be able to work on them. These non-backspacing
10428 media include pipes and cartridge tape drives.
10430 Some other media can be backspaced, and @command{tar} will work on them
10431 once @command{tar} is modified to do so.
10433 Archives created with the @option{--multi-volume}, @option{--label}, and
10434 @option{--incremental} (@option{-G}) options may not be readable by other version
10435 of @command{tar}. In particular, restoring a file that was split over
10436 a volume boundary will require some careful work with @command{dd}, if
10437 it can be done at all. Other versions of @command{tar} may also create
10438 an empty file whose name is that of the volume header. Some versions
10439 of @command{tar} may create normal files instead of directories archived
10440 with the @option{--incremental} (@option{-G}) option.
10442 @node Common Problems and Solutions
10443 @section Some Common Problems and their Solutions
10448 errors from system:
10450 no such file or directory
10453 errors from @command{tar}:
10454 directory checksum error
10455 header format error
10457 errors from media/system:
10469 @dfn{Block} and @dfn{record} terminology is rather confused, and it
10470 is also confusing to the expert reader. On the other hand, readers
10471 who are new to the field have a fresh mind, and they may safely skip
10472 the next two paragraphs, as the remainder of this manual uses those
10473 two terms in a quite consistent way.
10475 John Gilmore, the writer of the public domain @command{tar} from which
10476 @GNUTAR{} was originally derived, wrote (June 1995):
10479 The nomenclature of tape drives comes from IBM, where I believe
10480 they were invented for the IBM 650 or so. On IBM mainframes, what
10481 is recorded on tape are tape blocks. The logical organization of
10482 data is into records. There are various ways of putting records into
10483 blocks, including @code{F} (fixed sized records), @code{V} (variable
10484 sized records), @code{FB} (fixed blocked: fixed size records, @var{n}
10485 to a block), @code{VB} (variable size records, @var{n} to a block),
10486 @code{VSB} (variable spanned blocked: variable sized records that can
10487 occupy more than one block), etc. The @code{JCL} @samp{DD RECFORM=}
10488 parameter specified this to the operating system.
10490 The Unix man page on @command{tar} was totally confused about this.
10491 When I wrote @code{PD TAR}, I used the historically correct terminology
10492 (@command{tar} writes data records, which are grouped into blocks).
10493 It appears that the bogus terminology made it into @acronym{POSIX} (no surprise
10494 here), and now Fran@,{c}ois has migrated that terminology back
10495 into the source code too.
10498 The term @dfn{physical block} means the basic transfer chunk from or
10499 to a device, after which reading or writing may stop without anything
10500 being lost. In this manual, the term @dfn{block} usually refers to
10501 a disk physical block, @emph{assuming} that each disk block is 512
10502 bytes in length. It is true that some disk devices have different
10503 physical blocks, but @command{tar} ignore these differences in its own
10504 format, which is meant to be portable, so a @command{tar} block is always
10505 512 bytes in length, and @dfn{block} always mean a @command{tar} block.
10506 The term @dfn{logical block} often represents the basic chunk of
10507 allocation of many disk blocks as a single entity, which the operating
10508 system treats somewhat atomically; this concept is only barely used
10511 The term @dfn{physical record} is another way to speak of a physical
10512 block, those two terms are somewhat interchangeable. In this manual,
10513 the term @dfn{record} usually refers to a tape physical block,
10514 @emph{assuming} that the @command{tar} archive is kept on magnetic tape.
10515 It is true that archives may be put on disk or used with pipes,
10516 but nevertheless, @command{tar} tries to read and write the archive one
10517 @dfn{record} at a time, whatever the medium in use. One record is made
10518 up of an integral number of blocks, and this operation of putting many
10519 disk blocks into a single tape block is called @dfn{reblocking}, or
10520 more simply, @dfn{blocking}. The term @dfn{logical record} refers to
10521 the logical organization of many characters into something meaningful
10522 to the application. The term @dfn{unit record} describes a small set
10523 of characters which are transmitted whole to or by the application,
10524 and often refers to a line of text. Those two last terms are unrelated
10525 to what we call a @dfn{record} in @GNUTAR{}.
10527 When writing to tapes, @command{tar} writes the contents of the archive
10528 in chunks known as @dfn{records}. To change the default blocking
10529 factor, use the @option{--blocking-factor=@var{512-size}} (@option{-b
10530 @var{512-size}}) option. Each record will then be composed of
10531 @var{512-size} blocks. (Each @command{tar} block is 512 bytes.
10532 @xref{Standard}.) Each file written to the archive uses at least one
10533 full record. As a result, using a larger record size can result in
10534 more wasted space for small files. On the other hand, a larger record
10535 size can often be read and written much more efficiently.
10537 Further complicating the problem is that some tape drives ignore the
10538 blocking entirely. For these, a larger record size can still improve
10539 performance (because the software layers above the tape drive still
10540 honor the blocking), but not as dramatically as on tape drives that
10543 When reading an archive, @command{tar} can usually figure out the
10544 record size on itself. When this is the case, and a non-standard
10545 record size was used when the archive was created, @command{tar} will
10546 print a message about a non-standard blocking factor, and then operate
10547 normally. On some tape devices, however, @command{tar} cannot figure
10548 out the record size itself. On most of those, you can specify a
10549 blocking factor (with @option{--blocking-factor}) larger than the
10550 actual blocking factor, and then use the @option{--read-full-records}
10551 (@option{-B}) option. (If you specify a blocking factor with
10552 @option{--blocking-factor} and don't use the
10553 @option{--read-full-records} option, then @command{tar} will not
10554 attempt to figure out the recording size itself.) On some devices,
10555 you must always specify the record size exactly with
10556 @option{--blocking-factor} when reading, because @command{tar} cannot
10557 figure it out. In any case, use @option{--list} (@option{-t}) before
10558 doing any extractions to see whether @command{tar} is reading the archive
10561 @command{tar} blocks are all fixed size (512 bytes), and its scheme for
10562 putting them into records is to put a whole number of them (one or
10563 more) into each record. @command{tar} records are all the same size;
10564 at the end of the file there's a block containing all zeros, which
10565 is how you tell that the remainder of the last record(s) are garbage.
10567 In a standard @command{tar} file (no options), the block size is 512
10568 and the record size is 10240, for a blocking factor of 20. What the
10569 @option{--blocking-factor} option does is sets the blocking factor,
10570 changing the record size while leaving the block size at 512 bytes.
10571 20 was fine for ancient 800 or 1600 bpi reel-to-reel tape drives;
10572 most tape drives these days prefer much bigger records in order to
10573 stream and not waste tape. When writing tapes for myself, some tend
10574 to use a factor of the order of 2048, say, giving a record size of
10575 around one megabyte.
10577 If you use a blocking factor larger than 20, older @command{tar}
10578 programs might not be able to read the archive, so we recommend this
10579 as a limit to use in practice. @GNUTAR{}, however,
10580 will support arbitrarily large record sizes, limited only by the
10581 amount of virtual memory or the physical characteristics of the tape
10585 * Format Variations:: Format Variations
10586 * Blocking Factor:: The Blocking Factor of an Archive
10589 @node Format Variations
10590 @subsection Format Variations
10591 @cindex Format Parameters
10592 @cindex Format Options
10593 @cindex Options, archive format specifying
10594 @cindex Options, format specifying
10597 Format parameters specify how an archive is written on the archive
10598 media. The best choice of format parameters will vary depending on
10599 the type and number of files being archived, and on the media used to
10602 To specify format parameters when accessing or creating an archive,
10603 you can use the options described in the following sections.
10604 If you do not specify any format parameters, @command{tar} uses
10605 default parameters. You cannot modify a compressed archive.
10606 If you create an archive with the @option{--blocking-factor} option
10607 specified (@pxref{Blocking Factor}), you must specify that
10608 blocking-factor when operating on the archive. @xref{Formats}, for other
10609 examples of format parameter considerations.
10611 @node Blocking Factor
10612 @subsection The Blocking Factor of an Archive
10613 @cindex Blocking Factor
10614 @cindex Record Size
10615 @cindex Number of blocks per record
10616 @cindex Number of bytes per record
10617 @cindex Bytes per record
10618 @cindex Blocks per record
10621 @opindex blocking-factor
10622 The data in an archive is grouped into blocks, which are 512 bytes.
10623 Blocks are read and written in whole number multiples called
10624 @dfn{records}. The number of blocks in a record (i.e., the size of a
10625 record in units of 512 bytes) is called the @dfn{blocking factor}.
10626 The @option{--blocking-factor=@var{512-size}} (@option{-b
10627 @var{512-size}}) option specifies the blocking factor of an archive.
10628 The default blocking factor is typically 20 (i.e., 10240 bytes), but
10629 can be specified at installation. To find out the blocking factor of
10630 an existing archive, use @samp{tar --list --file=@var{archive-name}}.
10631 This may not work on some devices.
10633 Records are separated by gaps, which waste space on the archive media.
10634 If you are archiving on magnetic tape, using a larger blocking factor
10635 (and therefore larger records) provides faster throughput and allows you
10636 to fit more data on a tape (because there are fewer gaps). If you are
10637 archiving on cartridge, a very large blocking factor (say 126 or more)
10638 greatly increases performance. A smaller blocking factor, on the other
10639 hand, may be useful when archiving small files, to avoid archiving lots
10640 of nulls as @command{tar} fills out the archive to the end of the record.
10641 In general, the ideal record size depends on the size of the
10642 inter-record gaps on the tape you are using, and the average size of the
10643 files you are archiving. @xref{create}, for information on
10646 @FIXME{Need example of using a cartridge with blocking factor=126 or more.}
10648 Archives with blocking factors larger than 20 cannot be read
10649 by very old versions of @command{tar}, or by some newer versions
10650 of @command{tar} running on old machines with small address spaces.
10651 With @GNUTAR{}, the blocking factor of an archive is limited
10652 only by the maximum record size of the device containing the archive,
10653 or by the amount of available virtual memory.
10655 Also, on some systems, not using adequate blocking factors, as sometimes
10656 imposed by the device drivers, may yield unexpected diagnostics. For
10657 example, this has been reported:
10660 Cannot write to /dev/dlt: Invalid argument
10664 In such cases, it sometimes happen that the @command{tar} bundled by
10665 the system is aware of block size idiosyncrasies, while @GNUTAR{}
10666 requires an explicit specification for the block size,
10667 which it cannot guess. This yields some people to consider
10668 @GNUTAR{} is misbehaving, because by comparison,
10669 @cite{the bundle @command{tar} works OK}. Adding @w{@kbd{-b 256}},
10670 for example, might resolve the problem.
10672 If you use a non-default blocking factor when you create an archive, you
10673 must specify the same blocking factor when you modify that archive. Some
10674 archive devices will also require you to specify the blocking factor when
10675 reading that archive, however this is not typically the case. Usually, you
10676 can use @option{--list} (@option{-t}) without specifying a blocking factor---@command{tar}
10677 reports a non-default record size and then lists the archive members as
10678 it would normally. To extract files from an archive with a non-standard
10679 blocking factor (particularly if you're not sure what the blocking factor
10680 is), you can usually use the @option{--read-full-records} (@option{-B}) option while
10681 specifying a blocking factor larger then the blocking factor of the archive
10682 (i.e., @samp{tar --extract --read-full-records --blocking-factor=300}).
10683 @xref{list}, for more information on the @option{--list} (@option{-t})
10684 operation. @xref{Reading}, for a more detailed explanation of that option.
10687 @item --blocking-factor=@var{number}
10688 @itemx -b @var{number}
10689 Specifies the blocking factor of an archive. Can be used with any
10690 operation, but is usually not necessary with @option{--list} (@option{-t}).
10696 @item -b @var{blocks}
10697 @itemx --blocking-factor=@var{blocks}
10698 Set record size to @math{@var{blocks}*512} bytes.
10700 This option is used to specify a @dfn{blocking factor} for the archive.
10701 When reading or writing the archive, @command{tar}, will do reads and writes
10702 of the archive in records of @math{@var{block}*512} bytes. This is true
10703 even when the archive is compressed. Some devices requires that all
10704 write operations be a multiple of a certain size, and so, @command{tar}
10705 pads the archive out to the next record boundary.
10707 The default blocking factor is set when @command{tar} is compiled, and is
10708 typically 20. Blocking factors larger than 20 cannot be read by very
10709 old versions of @command{tar}, or by some newer versions of @command{tar}
10710 running on old machines with small address spaces.
10712 With a magnetic tape, larger records give faster throughput and fit
10713 more data on a tape (because there are fewer inter-record gaps).
10714 If the archive is in a disk file or a pipe, you may want to specify
10715 a smaller blocking factor, since a large one will result in a large
10716 number of null bytes at the end of the archive.
10718 When writing cartridge or other streaming tapes, a much larger
10719 blocking factor (say 126 or more) will greatly increase performance.
10720 However, you must specify the same blocking factor when reading or
10721 updating the archive.
10723 Apparently, Exabyte drives have a physical block size of 8K bytes.
10724 If we choose our blocksize as a multiple of 8k bytes, then the problem
10725 seems to disappear. Id est, we are using block size of 112 right
10726 now, and we haven't had the problem since we switched@dots{}
10728 With @GNUTAR{} the blocking factor is limited only
10729 by the maximum record size of the device containing the archive, or by
10730 the amount of available virtual memory.
10732 However, deblocking or reblocking is virtually avoided in a special
10733 case which often occurs in practice, but which requires all the
10734 following conditions to be simultaneously true:
10737 the archive is subject to a compression option,
10739 the archive is not handled through standard input or output, nor
10740 redirected nor piped,
10742 the archive is directly handled to a local disk, instead of any special
10745 @option{--blocking-factor} is not explicitly specified on the @command{tar}
10749 If the output goes directly to a local disk, and not through
10750 stdout, then the last write is not extended to a full record size.
10751 Otherwise, reblocking occurs. Here are a few other remarks on this
10757 @command{gzip} will complain about trailing garbage if asked to
10758 uncompress a compressed archive on tape, there is an option to turn
10759 the message off, but it breaks the regularity of simply having to use
10760 @samp{@var{prog} -d} for decompression. It would be nice if gzip was
10761 silently ignoring any number of trailing zeros. I'll ask Jean-loup
10762 Gailly, by sending a copy of this message to him.
10765 @command{compress} does not show this problem, but as Jean-loup pointed
10766 out to Michael, @samp{compress -d} silently adds garbage after
10767 the result of decompression, which tar ignores because it already
10768 recognized its end-of-file indicator. So this bug may be safely
10772 @samp{gzip -d -q} will be silent about the trailing zeros indeed,
10773 but will still return an exit status of 2 which tar reports in turn.
10774 @command{tar} might ignore the exit status returned, but I hate doing
10775 that, as it weakens the protection @command{tar} offers users against
10776 other possible problems at decompression time. If @command{gzip} was
10777 silently skipping trailing zeros @emph{and} also avoiding setting the
10778 exit status in this innocuous case, that would solve this situation.
10781 @command{tar} should become more solid at not stopping to read a pipe at
10782 the first null block encountered. This inelegantly breaks the pipe.
10783 @command{tar} should rather drain the pipe out before exiting itself.
10786 @xopindex{ignore-zeros, short description}
10788 @itemx --ignore-zeros
10789 Ignore blocks of zeros in archive (means EOF).
10791 The @option{--ignore-zeros} (@option{-i}) option causes @command{tar} to ignore blocks
10792 of zeros in the archive. Normally a block of zeros indicates the
10793 end of the archive, but when reading a damaged archive, or one which
10794 was created by concatenating several archives together, this option
10795 allows @command{tar} to read the entire archive. This option is not on
10796 by default because many versions of @command{tar} write garbage after
10799 Note that this option causes @command{tar} to read to the end of the
10800 archive file, which may sometimes avoid problems when multiple files
10801 are stored on a single physical tape.
10803 @xopindex{read-full-records, short description}
10805 @itemx --read-full-records
10806 Reblock as we read (for reading 4.2@acronym{BSD} pipes).
10808 If @option{--read-full-records} is used, @command{tar}
10809 will not panic if an attempt to read a record from the archive does
10810 not return a full record. Instead, @command{tar} will keep reading
10811 until it has obtained a full
10814 This option is turned on by default when @command{tar} is reading
10815 an archive from standard input, or from a remote machine. This is
10816 because on @acronym{BSD} Unix systems, a read of a pipe will return however
10817 much happens to be in the pipe, even if it is less than @command{tar}
10818 requested. If this option was not used, @command{tar} would fail as
10819 soon as it read an incomplete record from the pipe.
10821 This option is also useful with the commands for updating an archive.
10827 @FIXME{Appropriate options should be moved here from elsewhere.}
10829 @cindex blocking factor
10830 @cindex tape blocking
10832 When handling various tapes or cartridges, you have to take care of
10833 selecting a proper blocking, that is, the number of disk blocks you
10834 put together as a single tape block on the tape, without intervening
10835 tape gaps. A @dfn{tape gap} is a small landing area on the tape
10836 with no information on it, used for decelerating the tape to a
10837 full stop, and for later regaining the reading or writing speed.
10838 When the tape driver starts reading a record, the record has to
10839 be read whole without stopping, as a tape gap is needed to stop the
10840 tape motion without losing information.
10842 @cindex Exabyte blocking
10843 @cindex DAT blocking
10844 Using higher blocking (putting more disk blocks per tape block) will use
10845 the tape more efficiently as there will be less tape gaps. But reading
10846 such tapes may be more difficult for the system, as more memory will be
10847 required to receive at once the whole record. Further, if there is a
10848 reading error on a huge record, this is less likely that the system will
10849 succeed in recovering the information. So, blocking should not be too
10850 low, nor it should be too high. @command{tar} uses by default a blocking of
10851 20 for historical reasons, and it does not really matter when reading or
10852 writing to disk. Current tape technology would easily accommodate higher
10853 blockings. Sun recommends a blocking of 126 for Exabytes and 96 for DATs.
10854 We were told that for some DLT drives, the blocking should be a multiple
10855 of 4Kb, preferably 64Kb (@w{@kbd{-b 128}}) or 256 for decent performance.
10856 Other manufacturers may use different recommendations for the same tapes.
10857 This might also depends of the buffering techniques used inside modern
10858 tape controllers. Some imposes a minimum blocking, or a maximum blocking.
10859 Others request blocking to be some exponent of two.
10861 So, there is no fixed rule for blocking. But blocking at read time
10862 should ideally be the same as blocking used at write time. At one place
10863 I know, with a wide variety of equipment, they found it best to use a
10864 blocking of 32 to guarantee that their tapes are fully interchangeable.
10866 I was also told that, for recycled tapes, prior erasure (by the same
10867 drive unit that will be used to create the archives) sometimes lowers
10868 the error rates observed at rewriting time.
10870 I might also use @option{--number-blocks} instead of
10871 @option{--block-number}, so @option{--block} will then expand to
10872 @option{--blocking-factor} unambiguously.
10875 @section Many Archives on One Tape
10877 @FIXME{Appropriate options should be moved here from elsewhere.}
10879 @findex ntape @r{device}
10880 Most tape devices have two entries in the @file{/dev} directory, or
10881 entries that come in pairs, which differ only in the minor number for
10882 this device. Let's take for example @file{/dev/tape}, which often
10883 points to the only or usual tape device of a given system. There might
10884 be a corresponding @file{/dev/nrtape} or @file{/dev/ntape}. The simpler
10885 name is the @emph{rewinding} version of the device, while the name
10886 having @samp{nr} in it is the @emph{no rewinding} version of the same
10889 A rewinding tape device will bring back the tape to its beginning point
10890 automatically when this device is opened or closed. Since @command{tar}
10891 opens the archive file before using it and closes it afterwards, this
10892 means that a simple:
10895 $ @kbd{tar cf /dev/tape @var{directory}}
10899 will reposition the tape to its beginning both prior and after saving
10900 @var{directory} contents to it, thus erasing prior tape contents and
10901 making it so that any subsequent write operation will destroy what has
10904 @cindex tape positioning
10905 So, a rewinding device is normally meant to hold one and only one file.
10906 If you want to put more than one @command{tar} archive on a given tape, you
10907 will need to avoid using the rewinding version of the tape device. You
10908 will also have to pay special attention to tape positioning. Errors in
10909 positioning may overwrite the valuable data already on your tape. Many
10910 people, burnt by past experiences, will only use rewinding devices and
10911 limit themselves to one file per tape, precisely to avoid the risk of
10912 such errors. Be fully aware that writing at the wrong position on a
10913 tape loses all information past this point and most probably until the
10914 end of the tape, and this destroyed information @emph{cannot} be
10917 To save @var{directory-1} as a first archive at the beginning of a
10918 tape, and leave that tape ready for a second archive, you should use:
10921 $ @kbd{mt -f /dev/nrtape rewind}
10922 $ @kbd{tar cf /dev/nrtape @var{directory-1}}
10926 @dfn{Tape marks} are special magnetic patterns written on the tape
10927 media, which are later recognizable by the reading hardware. These
10928 marks are used after each file, when there are many on a single tape.
10929 An empty file (that is to say, two tape marks in a row) signal the
10930 logical end of the tape, after which no file exist. Usually,
10931 non-rewinding tape device drivers will react to the close request issued
10932 by @command{tar} by first writing two tape marks after your archive, and by
10933 backspacing over one of these. So, if you remove the tape at that time
10934 from the tape drive, it is properly terminated. But if you write
10935 another file at the current position, the second tape mark will be
10936 erased by the new information, leaving only one tape mark between files.
10938 So, you may now save @var{directory-2} as a second archive after the
10939 first on the same tape by issuing the command:
10942 $ @kbd{tar cf /dev/nrtape @var{directory-2}}
10946 and so on for all the archives you want to put on the same tape.
10948 Another usual case is that you do not write all the archives the same
10949 day, and you need to remove and store the tape between two archive
10950 sessions. In general, you must remember how many files are already
10951 saved on your tape. Suppose your tape already has 16 files on it, and
10952 that you are ready to write the 17th. You have to take care of skipping
10953 the first 16 tape marks before saving @var{directory-17}, say, by using
10957 $ @kbd{mt -f /dev/nrtape rewind}
10958 $ @kbd{mt -f /dev/nrtape fsf 16}
10959 $ @kbd{tar cf /dev/nrtape @var{directory-17}}
10962 In all the previous examples, we put aside blocking considerations, but
10963 you should do the proper things for that as well. @xref{Blocking}.
10966 * Tape Positioning:: Tape Positions and Tape Marks
10967 * mt:: The @command{mt} Utility
10970 @node Tape Positioning
10971 @subsection Tape Positions and Tape Marks
10974 Just as archives can store more than one file from the file system,
10975 tapes can store more than one archive file. To keep track of where
10976 archive files (or any other type of file stored on tape) begin and
10977 end, tape archive devices write magnetic @dfn{tape marks} on the
10978 archive media. Tape drives write one tape mark between files,
10979 two at the end of all the file entries.
10981 If you think of data as a series of records "rrrr"'s, and tape marks as
10982 "*"'s, a tape might look like the following:
10985 rrrr*rrrrrr*rrrrr*rr*rrrrr**-------------------------
10988 Tape devices read and write tapes using a read/write @dfn{tape
10989 head}---a physical part of the device which can only access one
10990 point on the tape at a time. When you use @command{tar} to read or
10991 write archive data from a tape device, the device will begin reading
10992 or writing from wherever on the tape the tape head happens to be,
10993 regardless of which archive or what part of the archive the tape
10994 head is on. Before writing an archive, you should make sure that no
10995 data on the tape will be overwritten (unless it is no longer needed).
10996 Before reading an archive, you should make sure the tape head is at
10997 the beginning of the archive you want to read. You can do it manually
10998 via @code{mt} utility (@pxref{mt}). The @code{restore} script does
10999 that automatically (@pxref{Scripted Restoration}).
11001 If you want to add new archive file entries to a tape, you should
11002 advance the tape to the end of the existing file entries, backspace
11003 over the last tape mark, and write the new archive file. If you were
11004 to add two archives to the example above, the tape might look like the
11008 rrrr*rrrrrr*rrrrr*rr*rrrrr*rrr*rrrr**----------------
11012 @subsection The @command{mt} Utility
11015 @FIXME{Is it true that this only works on non-block devices?
11016 should explain the difference, (fixed or variable).}
11017 @xref{Blocking Factor}.
11019 You can use the @command{mt} utility to advance or rewind a tape past a
11020 specified number of archive files on the tape. This will allow you
11021 to move to the beginning of an archive before extracting or reading
11022 it, or to the end of all the archives before writing a new one.
11023 @FIXME{Why isn't there an "advance 'til you find two tape marks
11026 The syntax of the @command{mt} command is:
11029 @kbd{mt [-f @var{tapename}] @var{operation} [@var{number}]}
11032 where @var{tapename} is the name of the tape device, @var{number} is
11033 the number of times an operation is performed (with a default of one),
11034 and @var{operation} is one of the following:
11036 @FIXME{is there any use for record operations?}
11041 Writes @var{number} tape marks at the current position on the tape.
11044 Moves tape position forward @var{number} files.
11047 Moves tape position back @var{number} files.
11050 Rewinds the tape. (Ignores @var{number}.)
11054 Rewinds the tape and takes the tape device off-line. (Ignores @var{number}.)
11057 Prints status information about the tape unit.
11061 If you don't specify a @var{tapename}, @command{mt} uses the environment
11062 variable @env{TAPE}; if @env{TAPE} is not set, @command{mt} will use
11063 the default device specified in your @file{sys/mtio.h} file
11064 (@code{DEFTAPE} variable). If this is not defined, the program will
11065 display a descriptive error message and exit with code 1.
11067 @command{mt} returns a 0 exit status when the operation(s) were
11068 successful, 1 if the command was unrecognized, and 2 if an operation
11071 @node Using Multiple Tapes
11072 @section Using Multiple Tapes
11074 Often you might want to write a large archive, one larger than will fit
11075 on the actual tape you are using. In such a case, you can run multiple
11076 @command{tar} commands, but this can be inconvenient, particularly if you
11077 are using options like @option{--exclude=@var{pattern}} or dumping entire file systems.
11078 Therefore, @command{tar} provides a special mode for creating
11079 multi-volume archives.
11081 @dfn{Multi-volume} archive is a single @command{tar} archive, stored
11082 on several media volumes of fixed size. Although in this section we will
11083 often call @samp{volume} a @dfn{tape}, there is absolutely no
11084 requirement for multi-volume archives to be stored on tapes. Instead,
11085 they can use whatever media type the user finds convenient, they can
11086 even be located on files.
11088 When creating a multi-volume archive, @GNUTAR{} continues to fill
11089 current volume until it runs out of space, then it switches to
11090 next volume (usually the operator is queried to replace the tape on
11091 this point), and continues working on the new volume. This operation
11092 continues until all requested files are dumped. If @GNUTAR{} detects
11093 end of media while dumping a file, such a file is archived in split
11094 form. Some very big files can even be split across several volumes.
11096 Each volume is itself a valid @GNUTAR{} archive, so it can be read
11097 without any special options. Consequently any file member residing
11098 entirely on one volume can be extracted or otherwise operated upon
11099 without needing the other volume. Sure enough, to extract a split
11100 member you would need all volumes its parts reside on.
11102 Multi-volume archives suffer from several limitations. In particular,
11103 they cannot be compressed.
11105 @GNUTAR{} is able to create multi-volume archives of two formats
11106 (@pxref{Formats}): @samp{GNU} and @samp{POSIX}.
11109 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
11110 * Tape Files:: Tape Files
11111 * Tarcat:: Concatenate Volumes into a Single Archive
11115 @node Multi-Volume Archives
11116 @subsection Archives Longer than One Tape or Disk
11117 @cindex Multi-volume archives
11119 @opindex multi-volume
11120 To create an archive that is larger than will fit on a single unit of
11121 the media, use the @option{--multi-volume} (@option{-M}) option in conjunction with
11122 the @option{--create} option (@pxref{create}). A @dfn{multi-volume}
11123 archive can be manipulated like any other archive (provided the
11124 @option{--multi-volume} option is specified), but is stored on more
11125 than one tape or file.
11127 When you specify @option{--multi-volume}, @command{tar} does not report an
11128 error when it comes to the end of an archive volume (when reading), or
11129 the end of the media (when writing). Instead, it prompts you to load
11130 a new storage volume. If the archive is on a magnetic tape, you
11131 should change tapes when you see the prompt; if the archive is on a
11132 floppy disk, you should change disks; etc.
11135 @item --multi-volume
11137 Creates a multi-volume archive, when used in conjunction with
11138 @option{--create} (@option{-c}). To perform any other operation on a multi-volume
11139 archive, specify @option{--multi-volume} in conjunction with that
11144 $ @kbd{tar --create --multi-volume --file=/dev/tape @var{files}}
11148 The method @command{tar} uses to detect end of tape is not perfect, and
11149 fails on some operating systems or on some devices. If @command{tar}
11150 cannot detect the end of the tape itself, you can use
11151 @option{--tape-length} option to inform it about the capacity of the
11154 @anchor{tape-length}
11156 @opindex tape-length
11157 @item --tape-length=@var{size}
11158 @itemx -L @var{size}
11159 Set maximum length of a volume. The @var{size} argument should then
11160 be the usable size of the tape in units of 1024 bytes. This option
11161 selects @option{--multi-volume} automatically. For example:
11164 $ @kbd{tar --create --tape-length=41943040 --file=/dev/tape @var{files}}
11168 @anchor{change volume prompt}
11169 When @GNUTAR{} comes to the end of a storage media, it asks you to
11170 change the volume. The built-in prompt for POSIX locale
11171 is@footnote{If you run @GNUTAR{} under a different locale, the
11172 translation to the locale's language will be used.}:
11175 Prepare volume #@var{n} for `@var{archive}' and hit return:
11179 where @var{n} is the ordinal number of the volume to be created and
11180 @var{archive} is archive file or device name.
11182 When prompting for a new tape, @command{tar} accepts any of the following
11187 Request @command{tar} to explain possible responses.
11189 Request @command{tar} to exit immediately.
11190 @item n @var{file-name}
11191 Request @command{tar} to write the next volume on the file @var{file-name}.
11193 Request @command{tar} to run a subshell. This option can be disabled
11194 by giving @option{--restrict} command line option to
11195 @command{tar}@footnote{@xref{--restrict}, for more information about
11198 Request @command{tar} to begin writing the next volume.
11201 (You should only type @samp{y} after you have changed the tape;
11202 otherwise @command{tar} will write over the volume it just finished.)
11204 @cindex Volume number file
11206 @anchor{volno-file}
11207 @opindex volno-file
11208 The volume number used by @command{tar} in its tape-changing prompt
11209 can be changed; if you give the
11210 @option{--volno-file=@var{file-of-number}} option, then
11211 @var{file-of-number} should be an non-existing file to be created, or
11212 else, a file already containing a decimal number. That number will be
11213 used as the volume number of the first volume written. When
11214 @command{tar} is finished, it will rewrite the file with the
11215 now-current volume number. (This does not change the volume number
11216 written on a tape label, as per @ref{label}, it @emph{only} affects
11217 the number used in the prompt.)
11219 @cindex End-of-archive info script
11220 @cindex Info script
11221 @anchor{info-script}
11222 @opindex info-script
11223 @opindex new-volume-script
11224 If you want more elaborate behavior than this, you can write a special
11225 @dfn{new volume script}, that will be responsible for changing the
11226 volume, and instruct @command{tar} to use it instead of its normal
11227 prompting procedure:
11230 @item --info-script=@var{script-name}
11231 @itemx --new-volume-script=@var{script-name}
11232 @itemx -F @var{script-name}
11233 Specify the full name of the volume script to use. The script can be
11234 used to eject cassettes, or to broadcast messages such as
11235 @samp{Someone please come change my tape} when performing unattended
11239 The @var{script-name} is executed without any command line
11240 arguments. It inherits @command{tar}'s shell environment.
11241 Additional data is passed to it via the following
11242 environment variables:
11245 @vrindex TAR_VERSION, info script environment variable
11247 @GNUTAR{} version number.
11249 @vrindex TAR_ARCHIVE, info script environment variable
11251 The name of the archive @command{tar} is processing.
11253 @vrindex TAR_BLOCKING_FACTOR, info script environment variable
11254 @item TAR_BLOCKING_FACTOR
11255 Current blocking factor (@pxref{Blocking}).
11257 @vrindex TAR_VOLUME, info script environment variable
11259 Ordinal number of the volume @command{tar} is about to start.
11261 @vrindex TAR_SUBCOMMAND, info script environment variable
11262 @item TAR_SUBCOMMAND
11263 A short option describing the operation @command{tar} is executing.
11264 @xref{Operations}, for a complete list of subcommand options.
11266 @vrindex TAR_FORMAT, info script environment variable
11268 Format of the archive being processed. @xref{Formats}, for a complete
11269 list of archive format names.
11271 @vrindex TAR_FD, info script environment variable
11273 File descriptor which can be used to communicate the new volume
11274 name to @command{tar}.
11277 The volume script can instruct @command{tar} to use new archive name,
11278 by writing in to file descriptor @env{$TAR_FD} (see below for an example).
11280 If the info script fails, @command{tar} exits; otherwise, it begins
11281 writing the next volume.
11283 If you want @command{tar} to cycle through a series of files or tape
11284 drives, there are three approaches to choose from. First of all, you
11285 can give @command{tar} multiple @option{--file} options. In this case
11286 the specified files will be used, in sequence, as the successive
11287 volumes of the archive. Only when the first one in the sequence needs
11288 to be used again will @command{tar} prompt for a tape change (or run
11289 the info script). For example, suppose someone has two tape drives on
11290 a system named @file{/dev/tape0} and @file{/dev/tape1}. For having
11291 @GNUTAR{} to switch to the second drive when it needs to write the
11292 second tape, and then back to the first tape, etc., just do either of:
11295 $ @kbd{tar --create --multi-volume --file=/dev/tape0 --file=/dev/tape1 @var{files}}
11296 $ @kbd{tar cMff /dev/tape0 /dev/tape1 @var{files}}
11299 The second method is to use the @samp{n} response to the tape-change
11302 Finally, the most flexible approach is to use a volume script, that
11303 writes new archive name to the file descriptor @env{$TAR_FD}. For example, the
11304 following volume script will create a series of archive files, named
11305 @file{@var{archive}-@var{vol}}, where @var{archive} is the name of the
11306 archive being created (as given by @option{--file} option) and
11307 @var{vol} is the ordinal number of the archive being created:
11312 echo Preparing volume $TAR_VOLUME of $TAR_ARCHIVE.
11314 name=`expr $TAR_ARCHIVE : '\(.*\)-.*'`
11315 case $TAR_SUBCOMMAND in
11317 -d|-x|-t) test -r $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME || exit 1
11322 echo $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME >&$TAR_FD
11326 The same script can be used while listing, comparing or extracting
11327 from the created archive. For example:
11331 # @r{Create a multi-volume archive:}
11332 $ @kbd{tar -c -L1024 -f archive.tar -F new-volume .}
11333 # @r{Extract from the created archive:}
11334 $ @kbd{tar -x -f archive.tar -F new-volume .}
11339 Notice, that the first command had to use @option{-L} option, since
11340 otherwise @GNUTAR{} will end up writing everything to file
11341 @file{archive.tar}.
11343 You can read each individual volume of a multi-volume archive as if it
11344 were an archive by itself. For example, to list the contents of one
11345 volume, use @option{--list}, without @option{--multi-volume} specified.
11346 To extract an archive member from one volume (assuming it is described
11347 that volume), use @option{--extract}, again without
11348 @option{--multi-volume}.
11350 If an archive member is split across volumes (i.e., its entry begins on
11351 one volume of the media and ends on another), you need to specify
11352 @option{--multi-volume} to extract it successfully. In this case, you
11353 should load the volume where the archive member starts, and use
11354 @samp{tar --extract --multi-volume}---@command{tar} will prompt for later
11355 volumes as it needs them. @xref{extracting archives}, for more
11356 information about extracting archives.
11358 Multi-volume archives can be modified like any other archive. To add
11359 files to a multi-volume archive, you need to only mount the last
11360 volume of the archive media (and new volumes, if needed). For all
11361 other operations, you need to use the entire archive.
11363 If a multi-volume archive was labeled using
11364 @option{--label=@var{archive-label}} (@pxref{label}) when it was
11365 created, @command{tar} will not automatically label volumes which are
11366 added later. To label subsequent volumes, specify
11367 @option{--label=@var{archive-label}} again in conjunction with the
11368 @option{--append}, @option{--update} or @option{--concatenate} operation.
11370 Notice that multi-volume support is a GNU extension and the archives
11371 created in this mode should be read only using @GNUTAR{}. If you
11372 absolutely have to process such archives using a third-party @command{tar}
11373 implementation, read @ref{Split Recovery}.
11376 @subsection Tape Files
11377 @cindex labeling archives
11381 To give the archive a name which will be recorded in it, use the
11382 @option{--label=@var{volume-label}} (@option{-V @var{volume-label}})
11383 option. This will write a special block identifying
11384 @var{volume-label} as the name of the archive to the front of the
11385 archive which will be displayed when the archive is listed with
11386 @option{--list}. If you are creating a multi-volume archive with
11387 @option{--multi-volume} (@pxref{Using Multiple Tapes}), then the
11388 volume label will have @samp{Volume @var{nnn}} appended to the name
11389 you give, where @var{nnn} is the number of the volume of the archive.
11390 If you use the @option{--label=@var{volume-label}} option when
11391 reading an archive, it checks to make sure the label on the tape
11392 matches the one you gave. @xref{label}.
11394 When @command{tar} writes an archive to tape, it creates a single
11395 tape file. If multiple archives are written to the same tape, one
11396 after the other, they each get written as separate tape files. When
11397 extracting, it is necessary to position the tape at the right place
11398 before running @command{tar}. To do this, use the @command{mt} command.
11399 For more information on the @command{mt} command and on the organization
11400 of tapes into a sequence of tape files, see @ref{mt}.
11402 People seem to often do:
11405 @kbd{--label="@var{some-prefix} `date +@var{some-format}`"}
11408 or such, for pushing a common date in all volumes or an archive set.
11411 @subsection Concatenate Volumes into a Single Archive
11414 Sometimes it is necessary to convert existing @GNUTAR{} multi-volume
11415 archive to a single @command{tar} archive. Simply concatenating all
11416 volumes into one will not work, since each volume carries an additional
11417 information at the beginning. @GNUTAR{} is shipped with the shell
11418 script @command{tarcat} designed for this purpose.
11420 The script takes a list of files comprising a multi-volume archive
11421 and creates the resulting archive at the standard output. For example:
11424 @kbd{tarcat vol.1 vol.2 vol.3 | tar tf -}
11427 The script implements a simple heuristics to determine the format of
11428 the first volume file and to decide how to process the rest of the
11429 files. However, it makes no attempt to verify whether the files are
11430 given in order or even if they are valid @command{tar} archives.
11431 It uses @command{dd} and does not filter its standard error, so you
11432 will usually see lots of spurious messages.
11434 @FIXME{The script is not installed. Should we install it?}
11437 @section Including a Label in the Archive
11438 @cindex Labeling an archive
11439 @cindex Labels on the archive media
11440 @cindex Labeling multi-volume archives
11443 To avoid problems caused by misplaced paper labels on the archive
11444 media, you can include a @dfn{label} entry --- an archive member which
11445 contains the name of the archive --- in the archive itself. Use the
11446 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
11447 option@footnote{Until version 1.10, that option was called
11448 @option{--volume}, but is not available under that name anymore.} in
11449 conjunction with the @option{--create} operation to include a label
11450 entry in the archive as it is being created.
11453 @item --label=@var{archive-label}
11454 @itemx -V @var{archive-label}
11455 Includes an @dfn{archive-label} at the beginning of the archive when
11456 the archive is being created, when used in conjunction with the
11457 @option{--create} operation. Checks to make sure the archive label
11458 matches the one specified (when used in conjunction with any other
11462 If you create an archive using both
11463 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
11464 and @option{--multi-volume} (@option{-M}), each volume of the archive
11465 will have an archive label of the form @samp{@var{archive-label}
11466 Volume @var{n}}, where @var{n} is 1 for the first volume, 2 for the
11467 next, and so on. @xref{Using Multiple Tapes}, for information on
11468 creating multiple volume archives.
11470 @cindex Volume label, listing
11471 @cindex Listing volume label
11472 The volume label will be displayed by @option{--list} along with
11473 the file contents. If verbose display is requested, it will also be
11474 explicitly marked as in the example below:
11478 $ @kbd{tar --verbose --list --file=iamanarchive}
11479 V--------- 0 0 0 1992-03-07 12:01 iamalabel--Volume Header--
11480 -rw-r--r-- ringo user 40 1990-05-21 13:30 iamafilename
11484 @opindex test-label
11485 @anchor{--test-label option}
11486 However, @option{--list} option will cause listing entire
11487 contents of the archive, which may be undesirable (for example, if the
11488 archive is stored on a tape). You can request checking only the volume
11489 label by specifying @option{--test-label} option. This option reads only the
11490 first block of an archive, so it can be used with slow storage
11491 devices. For example:
11495 $ @kbd{tar --test-label --file=iamanarchive}
11500 If @option{--test-label} is used with one or more command line
11501 arguments, @command{tar} compares the volume label with each
11502 argument. It exits with code 0 if a match is found, and with code 1
11503 otherwise@footnote{Note that @GNUTAR{} versions up to 1.23 indicated
11504 mismatch with an exit code 2 and printed a spurious diagnostics on
11505 stderr.}. No output is displayed, unless you also used the
11506 @option{--verbose} option. For example:
11510 $ @kbd{tar --test-label --file=iamanarchive 'iamalabel'}
11512 $ @kbd{tar --test-label --file=iamanarchive 'alabel'}
11517 When used with the @option{--verbose} option, @command{tar}
11518 prints the actual volume label (if any), and a verbose diagnostics in
11519 case of a mismatch:
11523 $ @kbd{tar --test-label --verbose --file=iamanarchive 'iamalabel'}
11526 $ @kbd{tar --test-label --verbose --file=iamanarchive 'alabel'}
11528 tar: Archive label mismatch
11533 If you request any operation, other than @option{--create}, along
11534 with using @option{--label} option, @command{tar} will first check if
11535 the archive label matches the one specified and will refuse to proceed
11536 if it does not. Use this as a safety precaution to avoid accidentally
11537 overwriting existing archives. For example, if you wish to add files
11538 to @file{archive}, presumably labeled with string @samp{My volume},
11543 $ @kbd{tar -rf archive --label 'My volume' .}
11544 tar: Archive not labeled to match `My volume'
11549 in case its label does not match. This will work even if
11550 @file{archive} is not labeled at all.
11552 Similarly, @command{tar} will refuse to list or extract the
11553 archive if its label doesn't match the @var{archive-label}
11554 specified. In those cases, @var{archive-label} argument is interpreted
11555 as a globbing-style pattern which must match the actual magnetic
11556 volume label. @xref{exclude}, for a precise description of how match
11557 is attempted@footnote{Previous versions of @command{tar} used full
11558 regular expression matching, or before that, only exact string
11559 matching, instead of wildcard matchers. We decided for the sake of
11560 simplicity to use a uniform matching device through
11561 @command{tar}.}. If the switch @option{--multi-volume} (@option{-M}) is being used,
11562 the volume label matcher will also suffix @var{archive-label} by
11563 @w{@samp{ Volume [1-9]*}} if the initial match fails, before giving
11564 up. Since the volume numbering is automatically added in labels at
11565 creation time, it sounded logical to equally help the user taking care
11566 of it when the archive is being read.
11568 You can also use @option{--label} to get a common information on
11569 all tapes of a series. For having this information different in each
11570 series created through a single script used on a regular basis, just
11571 manage to get some date string as part of the label. For example:
11575 $ @kbd{tar cfMV /dev/tape "Daily backup for `date +%Y-%m-%d`"}
11576 $ @kbd{tar --create --file=/dev/tape --multi-volume \
11577 --label="Daily backup for `date +%Y-%m-%d`"}
11581 Some more notes about volume labels:
11584 @item Each label has its own date and time, which corresponds
11585 to the time when @GNUTAR{} initially attempted to write it,
11586 often soon after the operator launches @command{tar} or types the
11587 carriage return telling that the next tape is ready.
11589 @item Comparing date labels to get an idea of tape throughput is
11590 unreliable. It gives correct results only if the delays for rewinding
11591 tapes and the operator switching them were negligible, which is
11592 usually not the case.
11596 @section Verifying Data as It is Stored
11597 @cindex Verifying a write operation
11598 @cindex Double-checking a write operation
11603 @opindex verify, short description
11604 Attempt to verify the archive after writing.
11607 This option causes @command{tar} to verify the archive after writing it.
11608 Each volume is checked after it is written, and any discrepancies
11609 are recorded on the standard error output.
11611 Verification requires that the archive be on a back-space-able medium.
11612 This means pipes, some cartridge tape drives, and some other devices
11613 cannot be verified.
11615 You can insure the accuracy of an archive by comparing files in the
11616 system with archive members. @command{tar} can compare an archive to the
11617 file system as the archive is being written, to verify a write
11618 operation, or can compare a previously written archive, to insure that
11621 @xopindex{verify, using with @option{--create}}
11622 @xopindex{create, using with @option{--verify}}
11623 To check for discrepancies in an archive immediately after it is
11624 written, use the @option{--verify} (@option{-W}) option in conjunction with
11625 the @option{--create} operation. When this option is
11626 specified, @command{tar} checks archive members against their counterparts
11627 in the file system, and reports discrepancies on the standard error.
11629 To verify an archive, you must be able to read it from before the end
11630 of the last written entry. This option is useful for detecting data
11631 errors on some tapes. Archives written to pipes, some cartridge tape
11632 drives, and some other devices cannot be verified.
11634 One can explicitly compare an already made archive with the file
11635 system by using the @option{--compare} (@option{--diff}, @option{-d})
11636 option, instead of using the more automatic @option{--verify} option.
11639 Note that these two options have a slightly different intent. The
11640 @option{--compare} option checks how identical are the logical contents of some
11641 archive with what is on your disks, while the @option{--verify} option is
11642 really for checking if the physical contents agree and if the recording
11643 media itself is of dependable quality. So, for the @option{--verify}
11644 operation, @command{tar} tries to defeat all in-memory cache pertaining to
11645 the archive, while it lets the speed optimization undisturbed for the
11646 @option{--compare} option. If you nevertheless use @option{--compare} for
11647 media verification, you may have to defeat the in-memory cache yourself,
11648 maybe by opening and reclosing the door latch of your recording unit,
11649 forcing some doubt in your operating system about the fact this is really
11650 the same volume as the one just written or read.
11652 The @option{--verify} option would not be necessary if drivers were indeed
11653 able to detect dependably all write failures. This sometimes require many
11654 magnetic heads, some able to read after the writes occurred. One would
11655 not say that drivers unable to detect all cases are necessarily flawed,
11656 as long as programming is concerned.
11658 The @option{--verify} (@option{-W}) option will not work in
11659 conjunction with the @option{--multi-volume} (@option{-M}) option or
11660 the @option{--append} (@option{-r}), @option{--update} (@option{-u})
11661 and @option{--delete} operations. @xref{Operations}, for more
11662 information on these operations.
11664 Also, since @command{tar} normally strips leading @samp{/} from file
11665 names (@pxref{absolute}), a command like @samp{tar --verify -cf
11666 /tmp/foo.tar /etc} will work as desired only if the working directory is
11667 @file{/}, as @command{tar} uses the archive's relative member names
11668 (e.g., @file{etc/motd}) when verifying the archive.
11670 @node Write Protection
11671 @section Write Protection
11673 Almost all tapes and diskettes, and in a few rare cases, even disks can
11674 be @dfn{write protected}, to protect data on them from being changed.
11675 Once an archive is written, you should write protect the media to prevent
11676 the archive from being accidentally overwritten or deleted. (This will
11677 protect the archive from being changed with a tape or floppy drive---it
11678 will not protect it from magnet fields or other physical hazards.)
11680 The write protection device itself is usually an integral part of the
11681 physical media, and can be a two position (write enabled/write
11682 disabled) switch, a notch which can be popped out or covered, a ring
11683 which can be removed from the center of a tape reel, or some other
11684 changeable feature.
11689 This appendix lists some important user-visible changes between
11690 version @GNUTAR{} @value{VERSION} and previous versions. An up-to-date
11691 version of this document is available at
11692 @uref{http://www.gnu.org/@/software/@/tar/manual/changes.html,the
11693 @GNUTAR{} documentation page}.
11696 @item Use of globbing patterns when listing and extracting.
11698 Previous versions of GNU tar assumed shell-style globbing when
11699 extracting from or listing an archive. For example:
11702 $ @kbd{tar xf foo.tar '*.c'}
11705 would extract all files whose names end in @samp{.c}. This behavior
11706 was not documented and was incompatible with traditional tar
11707 implementations. Therefore, starting from version 1.15.91, GNU tar
11708 no longer uses globbing by default. For example, the above invocation
11709 is now interpreted as a request to extract from the archive the file
11712 To facilitate transition to the new behavior for those users who got
11713 used to the previous incorrect one, @command{tar} will print a warning
11714 if it finds out that a requested member was not found in the archive
11715 and its name looks like a globbing pattern. For example:
11718 $ @kbd{tar xf foo.tar '*.c'}
11719 tar: Pattern matching characters used in file names. Please,
11720 tar: use --wildcards to enable pattern matching, or --no-wildcards to
11721 tar: suppress this warning.
11722 tar: *.c: Not found in archive
11723 tar: Error exit delayed from previous errors
11726 To treat member names as globbing patterns, use the @option{--wildcards} option.
11727 If you want to tar to mimic the behavior of versions prior to 1.15.91,
11728 add this option to your @env{TAR_OPTIONS} variable.
11730 @xref{wildcards}, for the detailed discussion of the use of globbing
11731 patterns by @GNUTAR{}.
11733 @item Use of short option @option{-o}.
11735 Earlier versions of @GNUTAR{} understood @option{-o} command line
11736 option as a synonym for @option{--old-archive}.
11738 @GNUTAR{} starting from version 1.13.90 understands this option as
11739 a synonym for @option{--no-same-owner}. This is compatible with
11740 UNIX98 @command{tar} implementations.
11742 However, to facilitate transition, @option{-o} option retains its
11743 old semantics when it is used with one of archive-creation commands.
11744 Users are encouraged to use @option{--format=oldgnu} instead.
11746 It is especially important, since versions of @acronym{GNU} Automake
11747 up to and including 1.8.4 invoke tar with this option to produce
11748 distribution tarballs. @xref{Formats,v7}, for the detailed discussion
11749 of this issue and its implications.
11751 @xref{Options, tar-formats, Changing Automake's Behavior,
11752 automake, GNU Automake}, for a description on how to use various
11753 archive formats with @command{automake}.
11755 Future versions of @GNUTAR{} will understand @option{-o} only as a
11756 synonym for @option{--no-same-owner}.
11758 @item Use of short option @option{-l}
11760 Earlier versions of @GNUTAR{} understood @option{-l} option as a
11761 synonym for @option{--one-file-system}. Since such usage contradicted
11762 to UNIX98 specification and harmed compatibility with other
11763 implementations, it was declared deprecated in version 1.14. However,
11764 to facilitate transition to its new semantics, it was supported by
11765 versions 1.15 and 1.15.90. The present use of @option{-l} as a short
11766 variant of @option{--check-links} was introduced in version 1.15.91.
11768 @item Use of options @option{--portability} and @option{--old-archive}
11770 These options are deprecated. Please use @option{--format=v7} instead.
11772 @item Use of option @option{--posix}
11774 This option is deprecated. Please use @option{--format=posix} instead.
11777 @node Configuring Help Summary
11778 @appendix Configuring Help Summary
11780 Running @kbd{tar --help} displays the short @command{tar} option
11781 summary (@pxref{help}). This summary is organized by @dfn{groups} of
11782 semantically close options. The options within each group are printed
11783 in the following order: a short option, eventually followed by a list
11784 of corresponding long option names, followed by a short description of
11785 the option. For example, here is an excerpt from the actual @kbd{tar
11789 Main operation mode:
11791 -A, --catenate, --concatenate append tar files to an archive
11792 -c, --create create a new archive
11793 -d, --diff, --compare find differences between archive and
11795 --delete delete from the archive
11798 @vrindex ARGP_HELP_FMT, environment variable
11799 The exact visual representation of the help output is configurable via
11800 @env{ARGP_HELP_FMT} environment variable. The value of this variable
11801 is a comma-separated list of @dfn{format variable} assignments. There
11802 are two kinds of format variables. An @dfn{offset variable} keeps the
11803 offset of some part of help output text from the leftmost column on
11804 the screen. A @dfn{boolean} variable is a flag that toggles some
11805 output feature on or off. Depending on the type of the corresponding
11806 variable, there are two kinds of assignments:
11809 @item Offset assignment
11811 The assignment to an offset variable has the following syntax:
11814 @var{variable}=@var{value}
11818 where @var{variable} is the variable name, and @var{value} is a
11819 numeric value to be assigned to the variable.
11821 @item Boolean assignment
11823 To assign @code{true} value to a variable, simply put this variable name. To
11824 assign @code{false} value, prefix the variable name with @samp{no-}. For
11829 # Assign @code{true} value:
11831 # Assign @code{false} value:
11837 Following variables are declared:
11839 @deftypevr {Help Output} boolean dup-args
11840 If true, arguments for an option are shown with both short and long
11841 options, even when a given option has both forms, for example:
11844 -f ARCHIVE, --file=ARCHIVE use archive file or device ARCHIVE
11847 If false, then if an option has both short and long forms, the
11848 argument is only shown with the long one, for example:
11851 -f, --file=ARCHIVE use archive file or device ARCHIVE
11855 and a message indicating that the argument is applicable to both
11856 forms is printed below the options. This message can be disabled
11857 using @code{dup-args-note} (see below).
11859 The default is false.
11862 @deftypevr {Help Output} boolean dup-args-note
11863 If this variable is true, which is the default, the following notice
11864 is displayed at the end of the help output:
11867 Mandatory or optional arguments to long options are also mandatory or
11868 optional for any corresponding short options.
11871 Setting @code{no-dup-args-note} inhibits this message. Normally, only one of
11872 variables @code{dup-args} or @code{dup-args-note} should be set.
11875 @deftypevr {Help Output} offset short-opt-col
11876 Column in which short options start. Default is 2.
11880 $ @kbd{tar --help|grep ARCHIVE}
11881 -f, --file=ARCHIVE use archive file or device ARCHIVE
11882 $ @kbd{ARGP_HELP_FMT=short-opt-col=6 tar --help|grep ARCHIVE}
11883 -f, --file=ARCHIVE use archive file or device ARCHIVE
11888 @deftypevr {Help Output} offset long-opt-col
11889 Column in which long options start. Default is 6. For example:
11893 $ @kbd{tar --help|grep ARCHIVE}
11894 -f, --file=ARCHIVE use archive file or device ARCHIVE
11895 $ @kbd{ARGP_HELP_FMT=long-opt-col=16 tar --help|grep ARCHIVE}
11896 -f, --file=ARCHIVE use archive file or device ARCHIVE
11901 @deftypevr {Help Output} offset doc-opt-col
11902 Column in which @dfn{doc options} start. A doc option isn't actually
11903 an option, but rather an arbitrary piece of documentation that is
11904 displayed in much the same manner as the options. For example, in
11905 the description of @option{--format} option:
11909 -H, --format=FORMAT create archive of the given format.
11911 FORMAT is one of the following:
11913 gnu GNU tar 1.13.x format
11914 oldgnu GNU format as per tar <= 1.12
11915 pax POSIX 1003.1-2001 (pax) format
11917 ustar POSIX 1003.1-1988 (ustar) format
11918 v7 old V7 tar format
11923 the format names are doc options. Thus, if you set
11924 @kbd{ARGP_HELP_FMT=doc-opt-col=6} the above part of the help output
11925 will look as follows:
11929 -H, --format=FORMAT create archive of the given format.
11931 FORMAT is one of the following:
11933 gnu GNU tar 1.13.x format
11934 oldgnu GNU format as per tar <= 1.12
11935 pax POSIX 1003.1-2001 (pax) format
11937 ustar POSIX 1003.1-1988 (ustar) format
11938 v7 old V7 tar format
11943 @deftypevr {Help Output} offset opt-doc-col
11944 Column in which option description starts. Default is 29.
11948 $ @kbd{tar --help|grep ARCHIVE}
11949 -f, --file=ARCHIVE use archive file or device ARCHIVE
11950 $ @kbd{ARGP_HELP_FMT=opt-doc-col=19 tar --help|grep ARCHIVE}
11951 -f, --file=ARCHIVE use archive file or device ARCHIVE
11952 $ @kbd{ARGP_HELP_FMT=opt-doc-col=9 tar --help|grep ARCHIVE}
11954 use archive file or device ARCHIVE
11959 Notice, that the description starts on a separate line if
11960 @code{opt-doc-col} value is too small.
11963 @deftypevr {Help Output} offset header-col
11964 Column in which @dfn{group headers} are printed. A group header is a
11965 descriptive text preceding an option group. For example, in the
11969 Main operation mode:
11971 -A, --catenate, --concatenate append tar files to
11973 -c, --create create a new archive
11976 @samp{Main operation mode:} is the group header.
11978 The default value is 1.
11981 @deftypevr {Help Output} offset usage-indent
11982 Indentation of wrapped usage lines. Affects @option{--usage}
11983 output. Default is 12.
11986 @deftypevr {Help Output} offset rmargin
11987 Right margin of the text output. Used for wrapping.
11990 @node Fixing Snapshot Files
11991 @appendix Fixing Snapshot Files
11992 @include tar-snapshot-edit.texi
11994 @node Tar Internals
11995 @appendix Tar Internals
11996 @include intern.texi
12000 @include genfile.texi
12002 @node Free Software Needs Free Documentation
12003 @appendix Free Software Needs Free Documentation
12004 @include freemanuals.texi
12006 @node Copying This Manual
12007 @appendix Copying This Manual
12010 * GNU Free Documentation License:: License for copying this manual
12015 @node Index of Command Line Options
12016 @appendix Index of Command Line Options
12018 This appendix contains an index of all @GNUTAR{} long command line
12019 options. The options are listed without the preceding double-dash.
12020 For a cross-reference of short command line options, see
12021 @ref{Short Option Summary}.
12034 @c Local variables:
12035 @c texinfo-column-for-description: 32
projects / chaz / tar / blob