1 \input texinfo @c -*-texinfo-*-
2 @comment %**start of header
5 @settitle GNU tar @value{VERSION}
13 @include rendition.texi
16 @c Put everything in one index (arbitrarily chosen to be the concept index).
26 This manual is for @acronym{GNU} @command{tar} (version
27 @value{VERSION}, @value{UPDATED}), which creates and extracts files
30 Copyright @copyright{} 1992, 1994, 1995, 1996, 1997, 1999, 2000, 2001,
31 2003, 2004, 2005, 2006 Free Software Foundation, Inc.
34 Permission is granted to copy, distribute and/or modify this document
35 under the terms of the GNU Free Documentation License, Version 1.1 or
36 any later version published by the Free Software Foundation; with no
37 Invariant Sections, with the Front-Cover Texts being ``A GNU Manual,''
38 and with the Back-Cover Texts as in (a) below. A copy of the license
39 is included in the section entitled "GNU Free Documentation License".
41 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
42 this GNU Manual. Buying copies from GNU Press supports the FSF in
43 developing GNU and promoting software freedom.''
47 @dircategory Archiving
49 * Tar: (tar). Making tape (or disk) archives.
52 @dircategory Individual utilities
54 * tar: (tar)tar invocation. Invoking @GNUTAR{}.
57 @shorttitlepage @acronym{GNU} @command{tar}
60 @title @acronym{GNU} tar: an archiver tool
61 @subtitle @value{RENDITION} @value{VERSION}, @value{UPDATED}
62 @author John Gilmore, Jay Fenlason et al.
65 @vskip 0pt plus 1filll
71 @top @acronym{GNU} tar: an archiver tool
76 @cindex archiving files
78 The first part of this master menu lists the major nodes in this Info
79 document. The rest of the menu lists all the lower level nodes.
82 @c The master menu, created with texinfo-master-menu, goes here.
83 @c (However, getdate.texi's menu is interpolated by hand.)
92 * Date input formats::
99 * Configuring Help Summary::
103 * Free Software Needs Free Documentation::
104 * Copying This Manual::
105 * Index of Command Line Options::
109 --- The Detailed Node Listing ---
113 * Book Contents:: What this Book Contains
114 * Definitions:: Some Definitions
115 * What tar Does:: What @command{tar} Does
116 * Naming tar Archives:: How @command{tar} Archives are Named
117 * Authors:: @GNUTAR{} Authors
118 * Reports:: Reporting bugs or suggestions
120 Tutorial Introduction to @command{tar}
123 * stylistic conventions::
124 * basic tar options:: Basic @command{tar} Operations and Options
125 * frequent operations::
126 * Two Frequent Options::
127 * create:: How to Create Archives
128 * list:: How to List Archives
129 * extract:: How to Extract Members from an Archive
132 Two Frequently Used Options
138 How to Create Archives
140 * prepare for examples::
141 * Creating the archive::
150 How to Extract Members from an Archive
152 * extracting archives::
160 * using tar options::
168 The Three Option Styles
170 * Mnemonic Options:: Mnemonic Option Style
171 * Short Options:: Short Option Style
172 * Old Options:: Old Option Style
173 * Mixing:: Mixing Option Styles
175 All @command{tar} Options
177 * Operation Summary::
179 * Short Option Summary::
191 Advanced @GNUTAR{} Operations
200 How to Add Files to Existing Archives: @option{--append}
202 * appending files:: Appending Files to an Archive
209 Options Used by @option{--create}
211 * Ignore Failed Read::
213 Options Used by @option{--extract}
215 * Reading:: Options to Help Read Archives
216 * Writing:: Changing How @command{tar} Writes Files
217 * Scarce:: Coping with Scarce Resources
219 Options to Help Read Archives
221 * read full records::
224 Changing How @command{tar} Writes Files
226 * Dealing with Old Files::
227 * Overwrite Old Files::
232 * Data Modification Times::
233 * Setting Access Permissions::
234 * Writing to Standard Output::
237 Coping with Scarce Resources
242 Performing Backups and Restoring Files
244 * Full Dumps:: Using @command{tar} to Perform Full Dumps
245 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
246 * Backup Levels:: Levels of Backups
247 * Backup Parameters:: Setting Parameters for Backups and Restoration
248 * Scripted Backups:: Using the Backup Scripts
249 * Scripted Restoration:: Using the Restore Script
251 Setting Parameters for Backups and Restoration
253 * General-Purpose Variables::
254 * Magnetic Tape Control::
256 * backup-specs example:: An Example Text of @file{Backup-specs}
258 Choosing Files and Names for @command{tar}
260 * file:: Choosing the Archive's Name
261 * Selecting Archive Members::
262 * files:: Reading Names from a File
263 * exclude:: Excluding Some Files
264 * wildcards:: Wildcards Patterns and Matching
265 * quoting styles:: Ways of Quoting Special Characters in Names
266 * transform:: Modifying File and Member Names
267 * after:: Operating Only on New Files
268 * recurse:: Descending into Directories
269 * one:: Crossing File System Boundaries
271 Reading Names from a File
277 * problems with exclude::
279 Crossing File System Boundaries
281 * directory:: Changing Directory
282 * absolute:: Absolute File Names
286 * General date syntax:: Common rules.
287 * Calendar date items:: 19 Dec 1994.
288 * Time of day items:: 9:20pm.
289 * Time zone items:: @sc{est}, @sc{pdt}, @sc{gmt}, ...
290 * Day of week items:: Monday and others.
291 * Relative items in date strings:: next tuesday, 2 years ago.
292 * Pure numbers in date strings:: 19931219, 1440.
293 * Seconds since the Epoch:: @@1078100502.
294 * Authors of get_date:: Bellovin, Eggert, Salz, Berets, et al.
296 Controlling the Archive Format
298 * Portability:: Making @command{tar} Archives More Portable
299 * Compression:: Using Less Space through Compression
300 * Attributes:: Handling File Attributes
301 * Standard:: The Standard Format
302 * Extensions:: @acronym{GNU} Extensions to the Archive Format
303 * cpio:: Comparison of @command{tar} and @command{cpio}
305 Making @command{tar} Archives More Portable
307 * Portable Names:: Portable Names
308 * dereference:: Symbolic Links
309 * old:: Old V7 Archives
310 * posix:: @acronym{POSIX} archives
311 * Checksumming:: Checksumming Problems
312 * Large or Negative Values:: Large files, negative time stamps, etc.
314 Using Less Space through Compression
316 * gzip:: Creating and Reading Compressed Archives
317 * sparse:: Archiving Sparse Files
319 Tapes and Other Archive Media
321 * Device:: Device selection and switching
322 * Remote Tape Server::
323 * Common Problems and Solutions::
324 * Blocking:: Blocking
325 * Many:: Many archives on one tape
326 * Using Multiple Tapes:: Using Multiple Tapes
327 * label:: Including a Label in the Archive
333 * Format Variations:: Format Variations
334 * Blocking Factor:: The Blocking Factor of an Archive
336 Many Archives on One Tape
338 * Tape Positioning:: Tape Positions and Tape Marks
339 * mt:: The @command{mt} Utility
343 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
344 * Tape Files:: Tape Files
345 * Tarcat:: Concatenate Volumes into a Single Archive
347 GNU tar internals and development
355 * Free Software Needs Free Documentation::
356 * GNU Free Documentation License:: License for copying this manual
362 @chapter Introduction
365 and manipulates @dfn{archives} which are actually collections of
366 many other files; the program provides users with an organized and
367 systematic method for controlling a large amount of data.
368 The name ``tar'' originally came from the phrase ``Tape ARchive'', but
369 archives need not (and these days, typically do not) reside on tapes.
372 * Book Contents:: What this Book Contains
373 * Definitions:: Some Definitions
374 * What tar Does:: What @command{tar} Does
375 * Naming tar Archives:: How @command{tar} Archives are Named
376 * Authors:: @GNUTAR{} Authors
377 * Reports:: Reporting bugs or suggestions
381 @section What this Book Contains
383 The first part of this chapter introduces you to various terms that will
384 recur throughout the book. It also tells you who has worked on @GNUTAR{}
385 and its documentation, and where you should send bug reports
388 The second chapter is a tutorial (@pxref{Tutorial}) which provides a
389 gentle introduction for people who are new to using @command{tar}. It is
390 meant to be self contained, not requiring any reading from subsequent
391 chapters to make sense. It moves from topic to topic in a logical,
392 progressive order, building on information already explained.
394 Although the tutorial is paced and structured to allow beginners to
395 learn how to use @command{tar}, it is not intended solely for beginners.
396 The tutorial explains how to use the three most frequently used
397 operations (@samp{create}, @samp{list}, and @samp{extract}) as well as
398 two frequently used options (@samp{file} and @samp{verbose}). The other
399 chapters do not refer to the tutorial frequently; however, if a section
400 discusses something which is a complex variant of a basic concept, there
401 may be a cross reference to that basic concept. (The entire book,
402 including the tutorial, assumes that the reader understands some basic
403 concepts of using a Unix-type operating system; @pxref{Tutorial}.)
405 The third chapter presents the remaining five operations, and
406 information about using @command{tar} options and option syntax.
408 @FIXME{this sounds more like a @acronym{GNU} Project Manuals Concept [tm] more
409 than the reality. should think about whether this makes sense to say
410 here, or not.} The other chapters are meant to be used as a
411 reference. Each chapter presents everything that needs to be said
412 about a specific topic.
414 One of the chapters (@pxref{Date input formats}) exists in its
415 entirety in other @acronym{GNU} manuals, and is mostly self-contained.
416 In addition, one section of this manual (@pxref{Standard}) contains a
417 big quote which is taken directly from @command{tar} sources.
419 In general, we give both long and short (abbreviated) option names
420 at least once in each section where the relevant option is covered, so
421 that novice readers will become familiar with both styles. (A few
422 options have no short versions, and the relevant sections will
426 @section Some Definitions
430 The @command{tar} program is used to create and manipulate @command{tar}
431 archives. An @dfn{archive} is a single file which contains the contents
432 of many files, while still identifying the names of the files, their
433 owner(s), and so forth. (In addition, archives record access
434 permissions, user and group, size in bytes, and data modification time.
435 Some archives also record the file names in each archived directory, as
436 well as other file and directory information.) You can use @command{tar}
437 to @dfn{create} a new archive in a specified directory.
440 @cindex archive member
443 The files inside an archive are called @dfn{members}. Within this
444 manual, we use the term @dfn{file} to refer only to files accessible in
445 the normal ways (by @command{ls}, @command{cat}, and so forth), and the term
446 @dfn{member} to refer only to the members of an archive. Similarly, a
447 @dfn{file name} is the name of a file, as it resides in the file system,
448 and a @dfn{member name} is the name of an archive member within the
453 The term @dfn{extraction} refers to the process of copying an archive
454 member (or multiple members) into a file in the file system. Extracting
455 all the members of an archive is often called @dfn{extracting the
456 archive}. The term @dfn{unpack} can also be used to refer to the
457 extraction of many or all the members of an archive. Extracting an
458 archive does not destroy the archive's structure, just as creating an
459 archive does not destroy the copies of the files that exist outside of
460 the archive. You may also @dfn{list} the members in a given archive
461 (this is often thought of as ``printing'' them to the standard output,
462 or the command line), or @dfn{append} members to a pre-existing archive.
463 All of these operations can be performed using @command{tar}.
466 @section What @command{tar} Does
469 The @command{tar} program provides the ability to create @command{tar}
470 archives, as well as various other kinds of manipulation. For example,
471 you can use @command{tar} on previously created archives to extract files,
472 to store additional files, or to update or list files which were already
475 Initially, @command{tar} archives were used to store files conveniently on
476 magnetic tape. The name @command{tar} comes from this use; it stands for
477 @code{t}ape @code{ar}chiver. Despite the utility's name, @command{tar} can
478 direct its output to available devices, files, or other programs (using
479 pipes). @command{tar} may even access remote devices or files (as archives).
481 You can use @command{tar} archives in many ways. We want to stress a few
482 of them: storage, backup, and transportation.
484 @FIXME{the following table entries need a bit of work..}
487 Often, @command{tar} archives are used to store related files for
488 convenient file transfer over a network. For example, the
489 @acronym{GNU} Project distributes its software bundled into
490 @command{tar} archives, so that all the files relating to a particular
491 program (or set of related programs) can be transferred as a single
494 A magnetic tape can store several files in sequence. However, the tape
495 has no names for these files; it only knows their relative position on
496 the tape. One way to store several files on one tape and retain their
497 names is by creating a @command{tar} archive. Even when the basic transfer
498 mechanism can keep track of names, as FTP can, the nuisance of handling
499 multiple files, directories, and multiple links makes @command{tar}
502 Archive files are also used for long-term storage. You can think of
503 this as transportation from the present into the future. (It is a
504 science-fiction idiom that you can move through time as well as in
505 space; the idea here is that @command{tar} can be used to move archives in
506 all dimensions, even time!)
509 Because the archive created by @command{tar} is capable of preserving
510 file information and directory structure, @command{tar} is commonly
511 used for performing full and incremental backups of disks. A backup
512 puts a collection of files (possibly pertaining to many users and
513 projects) together on a disk or a tape. This guards against
514 accidental destruction of the information in those files.
515 @GNUTAR{} has special features that allow it to be
516 used to make incremental and full dumps of all the files in a
520 You can create an archive on one system, transfer it to another system,
521 and extract the contents there. This allows you to transport a group of
522 files from one system to another.
525 @node Naming tar Archives
526 @section How @command{tar} Archives are Named
528 Conventionally, @command{tar} archives are given names ending with
529 @samp{.tar}. This is not necessary for @command{tar} to operate properly,
530 but this manual follows that convention in order to accustom readers to
531 it and to make examples more clear.
536 Often, people refer to @command{tar} archives as ``@command{tar} files,'' and
537 archive members as ``files'' or ``entries''. For people familiar with
538 the operation of @command{tar}, this causes no difficulty. However, in
539 this manual, we consistently refer to ``archives'' and ``archive
540 members'' to make learning to use @command{tar} easier for novice users.
543 @section @GNUTAR{} Authors
545 @GNUTAR{} was originally written by John Gilmore,
546 and modified by many people. The @acronym{GNU} enhancements were
547 written by Jay Fenlason, then Joy Kendall, and the whole package has
548 been further maintained by Thomas Bushnell, n/BSG, Fran@,{c}ois
549 Pinard, Paul Eggert, and finally Sergey Poznyakoff with the help of
550 numerous and kind users.
552 We wish to stress that @command{tar} is a collective work, and owes much to
553 all those people who reported problems, offered solutions and other
554 insights, or shared their thoughts and suggestions. An impressive, yet
555 partial list of those contributors can be found in the @file{THANKS}
556 file from the @GNUTAR{} distribution.
558 @FIXME{i want all of these names mentioned, Absolutely. BUT, i'm not
559 sure i want to spell out the history in this detail, at least not for
560 the printed book. i'm just not sure it needs to be said this way.
561 i'll think about it.}
563 @FIXME{History is more important, and surely more interesting, than
564 actual names. Quoting names without history would be meaningless. FP}
566 Jay Fenlason put together a draft of a @GNUTAR{}
567 manual, borrowing notes from the original man page from John Gilmore.
568 This was withdrawn in version 1.11. Thomas Bushnell, n/BSG and Amy
569 Gorin worked on a tutorial and manual for @GNUTAR{}.
570 Fran@,{c}ois Pinard put version 1.11.8 of the manual together by
571 taking information from all these sources and merging them. Melissa
572 Weisshaus finally edited and redesigned the book to create version
573 1.12. @FIXME{update version number as necessary; i'm being
574 optimistic!} @FIXME{Someone [maybe karl berry? maybe bob chassell?
575 maybe melissa? maybe julie sussman?] needs to properly index the
578 For version 1.12, Daniel Hagerty contributed a great deal of technical
579 consulting. In particular, he is the primary author of @ref{Backups}.
581 In July, 2003 @GNUTAR{} was put on CVS at savannah.gnu.org
582 (see @url{http://savannah.gnu.org/projects/tar}), and
583 active development and maintenance work has started
584 again. Currently @GNUTAR{} is being maintained by Paul Eggert, Sergey
585 Poznyakoff and Jeff Bailey.
587 Support for @acronym{POSIX} archives was added by Sergey Poznyakoff.
590 @section Reporting bugs or suggestions
593 @cindex reporting bugs
594 If you find problems or have suggestions about this program or manual,
595 please report them to @file{bug-tar@@gnu.org}.
597 When reporting a bug, please be sure to include as much detail as
598 possible, in order to reproduce it. @FIXME{Be more specific, I'd
599 like to make this node as detailed as 'Bug reporting' node in Emacs
603 @chapter Tutorial Introduction to @command{tar}
605 This chapter guides you through some basic examples of three @command{tar}
606 operations: @option{--create}, @option{--list}, and @option{--extract}. If
607 you already know how to use some other version of @command{tar}, then you
608 may not need to read this chapter. This chapter omits most complicated
609 details about how @command{tar} works.
613 * stylistic conventions::
614 * basic tar options:: Basic @command{tar} Operations and Options
615 * frequent operations::
616 * Two Frequent Options::
617 * create:: How to Create Archives
618 * list:: How to List Archives
619 * extract:: How to Extract Members from an Archive
624 @section Assumptions this Tutorial Makes
626 This chapter is paced to allow beginners to learn about @command{tar}
627 slowly. At the same time, we will try to cover all the basic aspects of
628 these three operations. In order to accomplish both of these tasks, we
629 have made certain assumptions about your knowledge before reading this
630 manual, and the hardware you will be using:
634 Before you start to work through this tutorial, you should understand
635 what the terms ``archive'' and ``archive member'' mean
636 (@pxref{Definitions}). In addition, you should understand something
637 about how Unix-type operating systems work, and you should know how to
638 use some basic utilities. For example, you should know how to create,
639 list, copy, rename, edit, and delete files and directories; how to
640 change between directories; and how to figure out where you are in the
641 file system. You should have some basic understanding of directory
642 structure and how files are named according to which directory they are
643 in. You should understand concepts such as standard output and standard
644 input, what various definitions of the term ``argument'' mean, and the
645 differences between relative and absolute path names. @FIXME{and what
649 This manual assumes that you are working from your own home directory
650 (unless we state otherwise). In this tutorial, you will create a
651 directory to practice @command{tar} commands in. When we show path names,
652 we will assume that those paths are relative to your home directory.
653 For example, my home directory path is @file{/home/fsf/melissa}. All of
654 my examples are in a subdirectory of the directory named by that path
655 name; the subdirectory is called @file{practice}.
658 In general, we show examples of archives which exist on (or can be
659 written to, or worked with from) a directory on a hard disk. In most
660 cases, you could write those archives to, or work with them on any other
661 device, such as a tape drive. However, some of the later examples in
662 the tutorial and next chapter will not work on tape drives.
663 Additionally, working with tapes is much more complicated than working
664 with hard disks. For these reasons, the tutorial does not cover working
665 with tape drives. @xref{Media}, for complete information on using
666 @command{tar} archives with tape drives.
668 @FIXME{this is a cop out. need to add some simple tape drive info.}
671 @node stylistic conventions
672 @section Stylistic Conventions
674 In the examples, @samp{$} represents a typical shell prompt. It
675 precedes lines you should type; to make this more clear, those lines are
676 shown in @kbd{this font}, as opposed to lines which represent the
677 computer's response; those lines are shown in @code{this font}, or
678 sometimes @samp{like this}.
680 @c When we have lines which are too long to be
681 @c displayed in any other way, we will show them like this:
683 @node basic tar options
684 @section Basic @command{tar} Operations and Options
686 @command{tar} can take a wide variety of arguments which specify and define
687 the actions it will have on the particular set of files or the archive.
688 The main types of arguments to @command{tar} fall into one of two classes:
689 operations, and options.
691 Some arguments fall into a class called @dfn{operations}; exactly one of
692 these is both allowed and required for any instance of using @command{tar};
693 you may @emph{not} specify more than one. People sometimes speak of
694 @dfn{operating modes}. You are in a particular operating mode when you
695 have specified the operation which specifies it; there are eight
696 operations in total, and thus there are eight operating modes.
698 The other arguments fall into the class known as @dfn{options}. You are
699 not required to specify any options, and you are allowed to specify more
700 than one at a time (depending on the way you are using @command{tar} at
701 that time). Some options are used so frequently, and are so useful for
702 helping you type commands more carefully that they are effectively
703 ``required''. We will discuss them in this chapter.
705 You can write most of the @command{tar} operations and options in any
706 of three forms: long (mnemonic) form, short form, and old style. Some
707 of the operations and options have no short or ``old'' forms; however,
708 the operations and options which we will cover in this tutorial have
709 corresponding abbreviations. @FIXME{make sure this is still the case,
710 at the end}We will indicate those abbreviations appropriately to get
711 you used to seeing them. (Note that the ``old style'' option forms
712 exist in @GNUTAR{} for compatibility with Unix
713 @command{tar}. In this book we present a full discussion of this way
714 of writing options and operations (@pxref{Old Options}), and we discuss
715 the other two styles of writing options (@xref{Mnemonic Options}, and
716 @pxref{Short Options}).
718 In the examples and in the text of this tutorial, we usually use the
719 long forms of operations and options; but the ``short'' forms produce
720 the same result and can make typing long @command{tar} commands easier.
721 For example, instead of typing
724 @kbd{tar --create --verbose --file=afiles.tar apple angst aspic}
730 @kbd{tar -c -v -f afiles.tar apple angst aspic}
736 @kbd{tar -cvf afiles.tar apple angst aspic}
740 For more information on option syntax, see @ref{Advanced tar}. In
741 discussions in the text, when we name an option by its long form, we
742 also give the corresponding short option in parentheses.
744 The term, ``option'', can be confusing at times, since ``operations''
745 are often lumped in with the actual, @emph{optional} ``options'' in certain
746 general class statements. For example, we just talked about ``short and
747 long forms of options and operations''. However, experienced @command{tar}
748 users often refer to these by shorthand terms such as, ``short and long
749 options''. This term assumes that the ``operations'' are included, also.
750 Context will help you determine which definition of ``options'' to use.
752 Similarly, the term ``command'' can be confusing, as it is often used in
753 two different ways. People sometimes refer to @command{tar} ``commands''.
754 A @command{tar} @dfn{command} is the entire command line of user input
755 which tells @command{tar} what to do --- including the operation, options,
756 and any arguments (file names, pipes, other commands, etc). However,
757 you will also sometimes hear the term ``the @command{tar} command''. When
758 the word ``command'' is used specifically like this, a person is usually
759 referring to the @command{tar} @emph{operation}, not the whole line.
760 Again, use context to figure out which of the meanings the speaker
763 @node frequent operations
764 @section The Three Most Frequently Used Operations
766 Here are the three most frequently used operations (both short and long
767 forms), as well as a brief description of their meanings. The rest of
768 this chapter will cover how to use these operations in detail. We will
769 present the rest of the operations in the next chapter.
774 Create a new @command{tar} archive.
777 List the contents of an archive.
780 Extract one or more members from an archive.
783 @node Two Frequent Options
784 @section Two Frequently Used Options
786 To understand how to run @command{tar} in the three operating modes listed
787 previously, you also need to understand how to use two of the options to
788 @command{tar}: @option{--file} (which takes an archive file as an argument)
789 and @option{--verbose}. (You are usually not @emph{required} to specify
790 either of these options when you run @command{tar}, but they can be very
791 useful in making things more clear and helping you avoid errors.)
800 @unnumberedsubsec The @option{--file} Option
803 @opindex file, tutorial
804 @item --file=@var{archive-name}
805 @itemx -f @var{archive-name}
806 Specify the name of an archive file.
809 You can specify an argument for the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option whenever you
810 use @command{tar}; this option determines the name of the archive file
811 that @command{tar} will work on.
814 If you don't specify this argument, then @command{tar} will examine
815 the environment variable @env{TAPE}. If it is set, its value will be
816 used as the archive name. Otherwise, @command{tar} will use the
817 default archive, determined at the compile time. Usually it is
818 standard output or some physical tape drive attached to your machine
819 (you can verify what the default is by running @kbd{tar
820 --show-defaults}, @pxref{defaults}). If there is no tape drive
821 attached, or the default is not meaningful, then @command{tar} will
822 print an error message. The error message might look roughly like one
826 tar: can't open /dev/rmt8 : No such device or address
827 tar: can't open /dev/rsmt0 : I/O error
831 To avoid confusion, we recommend that you always specify an archive file
832 name by using @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) when writing your @command{tar} commands.
833 For more information on using the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option, see
836 @node verbose tutorial
837 @unnumberedsubsec The @option{--verbose} Option
840 @opindex verbose, introduced
843 Show the files being worked on as @command{tar} is running.
846 @option{--verbose} (@option{-v}) shows details about the results of running
847 @command{tar}. This can be especially useful when the results might not be
848 obvious. For example, if you want to see the progress of @command{tar} as
849 it writes files into the archive, you can use the @option{--verbose}
850 option. In the beginning, you may find it useful to use
851 @option{--verbose} at all times; when you are more accustomed to
852 @command{tar}, you will likely want to use it at certain times but not at
853 others. We will use @option{--verbose} at times to help make something
854 clear, and we will give many examples both using and not using
855 @option{--verbose} to show the differences.
857 Sometimes, a single instance of @option{--verbose} on the command line
858 will show a full, @samp{ls} style listing of an archive or files,
859 giving sizes, owners, and similar information. @FIXME{Describe the
860 exact output format, e.g., how hard links are displayed.}
861 Other times, @option{--verbose} will only show files or members that the particular
862 operation is operating on at the time. In the latter case, you can
863 use @option{--verbose} twice in a command to get a listing such as that
864 in the former case. For example, instead of saying
867 @kbd{tar -cvf afiles.tar apple angst aspic}
874 @kbd{tar -cvvf afiles.tar apple angst aspic}
878 This works equally well using short or long forms of options. Using
879 long forms, you would simply write out the mnemonic form of the option
883 $ @kbd{tar --create --verbose --verbose @dots{}}
887 Note that you must double the hyphens properly each time.
889 Later in the tutorial, we will give examples using @w{@option{--verbose
893 @unnumberedsubsec Getting Help: Using the @option{--help} Option
899 The @option{--help} option to @command{tar} prints out a very brief list of
900 all operations and option available for the current version of
901 @command{tar} available on your system.
905 @section How to Create Archives
908 @cindex Creation of the archive
909 @cindex Archive, creation of
910 One of the basic operations of @command{tar} is @option{--create} (@option{-c}), which
911 you use to create a @command{tar} archive. We will explain
912 @option{--create} first because, in order to learn about the other
913 operations, you will find it useful to have an archive available to
916 To make this easier, in this section you will first create a directory
917 containing three files. Then, we will show you how to create an
918 @emph{archive} (inside the new directory). Both the directory, and
919 the archive are specifically for you to practice on. The rest of this
920 chapter and the next chapter will show many examples using this
921 directory and the files you will create: some of those files may be
922 other directories and other archives.
924 The three files you will archive in this example are called
925 @file{blues}, @file{folk}, and @file{jazz}. The archive is called
926 @file{collection.tar}.
928 This section will proceed slowly, detailing how to use @option{--create}
929 in @code{verbose} mode, and showing examples using both short and long
930 forms. In the rest of the tutorial, and in the examples in the next
931 chapter, we will proceed at a slightly quicker pace. This section
932 moves more slowly to allow beginning users to understand how
936 * prepare for examples::
937 * Creating the archive::
943 @node prepare for examples
944 @subsection Preparing a Practice Directory for Examples
946 To follow along with this and future examples, create a new directory
947 called @file{practice} containing files called @file{blues}, @file{folk}
948 and @file{jazz}. The files can contain any information you like:
949 ideally, they should contain information which relates to their names,
950 and be of different lengths. Our examples assume that @file{practice}
951 is a subdirectory of your home directory.
953 Now @command{cd} to the directory named @file{practice}; @file{practice}
954 is now your @dfn{working directory}. (@emph{Please note}: Although
955 the full path name of this directory is
956 @file{/@var{homedir}/practice}, in our examples we will refer to
957 this directory as @file{practice}; the @var{homedir} is presumed.
959 In general, you should check that the files to be archived exist where
960 you think they do (in the working directory) by running @command{ls}.
961 Because you just created the directory and the files and have changed to
962 that directory, you probably don't need to do that this time.
964 It is very important to make sure there isn't already a file in the
965 working directory with the archive name you intend to use (in this case,
966 @samp{collection.tar}), or that you don't care about its contents.
967 Whenever you use @samp{create}, @command{tar} will erase the current
968 contents of the file named by @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) if it exists. @command{tar}
969 will not tell you if you are about to overwrite an archive unless you
970 specify an option which does this (@pxref{backup}, for the
971 information on how to do so). To add files to an existing archive,
972 you need to use a different option, such as @option{--append} (@option{-r}); see
973 @ref{append} for information on how to do this.
975 @node Creating the archive
976 @subsection Creating the Archive
978 @opindex create, introduced
979 To place the files @file{blues}, @file{folk}, and @file{jazz} into an
980 archive named @file{collection.tar}, use the following command:
983 $ @kbd{tar --create --file=collection.tar blues folk jazz}
986 The order of the arguments is not very important, @emph{when using long
987 option forms}. You could also say:
990 $ @kbd{tar blues --create folk --file=collection.tar jazz}
994 However, you can see that this order is harder to understand; this is
995 why we will list the arguments in the order that makes the commands
996 easiest to understand (and we encourage you to do the same when you use
997 @command{tar}, to avoid errors).
999 Note that the part of the command which says,
1000 @w{@option{--file=collection.tar}} is considered to be @emph{one} argument.
1001 If you substituted any other string of characters for
1002 @kbd{collection.tar}, then that string would become the name of the
1003 archive file you create.
1005 The order of the options becomes more important when you begin to use
1006 short forms. With short forms, if you type commands in the wrong order
1007 (even if you type them correctly in all other ways), you may end up with
1008 results you don't expect. For this reason, it is a good idea to get
1009 into the habit of typing options in the order that makes inherent sense.
1010 @xref{short create}, for more information on this.
1012 In this example, you type the command as shown above: @option{--create}
1013 is the operation which creates the new archive
1014 (@file{collection.tar}), and @option{--file} is the option which lets
1015 you give it the name you chose. The files, @file{blues}, @file{folk},
1016 and @file{jazz}, are now members of the archive, @file{collection.tar}
1017 (they are @dfn{file name arguments} to the @option{--create} operation).
1018 @FIXME-xref{to the discussion of file name args?}Now that they are
1019 in the archive, they are called @emph{archive members}, not files.
1020 (@pxref{Definitions,members}).
1022 When you create an archive, you @emph{must} specify which files you
1023 want placed in the archive. If you do not specify any archive
1024 members, @GNUTAR{} will complain.
1026 If you now list the contents of the working directory (@command{ls}), you will
1027 find the archive file listed as well as the files you saw previously:
1030 blues folk jazz collection.tar
1034 Creating the archive @samp{collection.tar} did not destroy the copies of
1035 the files in the directory.
1037 Keep in mind that if you don't indicate an operation, @command{tar} will not
1038 run and will prompt you for one. If you don't name any files, @command{tar}
1039 will complain. You must have write access to the working directory,
1040 or else you will not be able to create an archive in that directory.
1042 @emph{Caution}: Do not attempt to use @option{--create} (@option{-c}) to add files to
1043 an existing archive; it will delete the archive and write a new one.
1044 Use @option{--append} (@option{-r}) instead. @xref{append}.
1046 @node create verbose
1047 @subsection Running @option{--create} with @option{--verbose}
1049 @opindex create, using with @option{--verbose}
1050 @opindex verbose, using with @option{--create}
1051 If you include the @option{--verbose} (@option{-v}) option on the command line,
1052 @command{tar} will list the files it is acting on as it is working. In
1053 verbose mode, the @code{create} example above would appear as:
1056 $ @kbd{tar --create --verbose --file=collection.tar blues folk jazz}
1062 This example is just like the example we showed which did not use
1063 @option{--verbose}, except that @command{tar} generated the remaining lines
1065 (note the different font styles).
1071 In the rest of the examples in this chapter, we will frequently use
1072 @code{verbose} mode so we can show actions or @command{tar} responses that
1073 you would otherwise not see, and which are important for you to
1077 @subsection Short Forms with @samp{create}
1079 As we said before, the @option{--create} (@option{-c}) operation is one of the most
1080 basic uses of @command{tar}, and you will use it countless times.
1081 Eventually, you will probably want to use abbreviated (or ``short'')
1082 forms of options. A full discussion of the three different forms that
1083 options can take appears in @ref{Styles}; for now, here is what the
1084 previous example (including the @option{--verbose} (@option{-v}) option) looks like
1085 using short option forms:
1088 $ @kbd{tar -cvf collection.tar blues folk jazz}
1095 As you can see, the system responds the same no matter whether you use
1096 long or short option forms.
1098 @FIXME{i don't like how this is worded:} One difference between using
1099 short and long option forms is that, although the exact placement of
1100 arguments following options is no more specific when using short forms,
1101 it is easier to become confused and make a mistake when using short
1102 forms. For example, suppose you attempted the above example in the
1106 $ @kbd{tar -cfv collection.tar blues folk jazz}
1110 In this case, @command{tar} will make an archive file called @file{v},
1111 containing the files @file{blues}, @file{folk}, and @file{jazz}, because
1112 the @samp{v} is the closest ``file name'' to the @option{-f} option, and
1113 is thus taken to be the chosen archive file name. @command{tar} will try
1114 to add a file called @file{collection.tar} to the @file{v} archive file;
1115 if the file @file{collection.tar} did not already exist, @command{tar} will
1116 report an error indicating that this file does not exist. If the file
1117 @file{collection.tar} does already exist (e.g., from a previous command
1118 you may have run), then @command{tar} will add this file to the archive.
1119 Because the @option{-v} option did not get registered, @command{tar} will not
1120 run under @samp{verbose} mode, and will not report its progress.
1122 The end result is that you may be quite confused about what happened,
1123 and possibly overwrite a file. To illustrate this further, we will show
1124 you how an example we showed previously would look using short forms.
1129 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1133 is confusing as it is. When shown using short forms, however, it
1134 becomes much more so:
1137 $ @kbd{tar blues -c folk -f collection.tar jazz}
1141 It would be very easy to put the wrong string of characters
1142 immediately following the @option{-f}, but doing that could sacrifice
1145 For this reason, we recommend that you pay very careful attention to
1146 the order of options and placement of file and archive names,
1147 especially when using short option forms. Not having the option name
1148 written out mnemonically can affect how well you remember which option
1149 does what, and therefore where different names have to be placed.
1152 @subsection Archiving Directories
1154 @cindex Archiving Directories
1155 @cindex Directories, Archiving
1156 You can archive a directory by specifying its directory name as a
1157 file name argument to @command{tar}. The files in the directory will be
1158 archived relative to the working directory, and the directory will be
1159 re-created along with its contents when the archive is extracted.
1161 To archive a directory, first move to its superior directory. If you
1162 have followed the previous instructions in this tutorial, you should
1171 This will put you into the directory which contains @file{practice},
1172 i.e., your home directory. Once in the superior directory, you can
1173 specify the subdirectory, @file{practice}, as a file name argument. To
1174 store @file{practice} in the new archive file @file{music.tar}, type:
1177 $ @kbd{tar --create --verbose --file=music.tar practice}
1181 @command{tar} should output:
1188 practice/collection.tar
1191 Note that the archive thus created is not in the subdirectory
1192 @file{practice}, but rather in the current working directory---the
1193 directory from which @command{tar} was invoked. Before trying to archive a
1194 directory from its superior directory, you should make sure you have
1195 write access to the superior directory itself, not only the directory
1196 you are trying archive with @command{tar}. For example, you will probably
1197 not be able to store your home directory in an archive by invoking
1198 @command{tar} from the root directory; @xref{absolute}. (Note
1199 also that @file{collection.tar}, the original archive file, has itself
1200 been archived. @command{tar} will accept any file as a file to be
1201 archived, regardless of its content. When @file{music.tar} is
1202 extracted, the archive file @file{collection.tar} will be re-written
1203 into the file system).
1205 If you give @command{tar} a command such as
1208 $ @kbd{tar --create --file=foo.tar .}
1212 @command{tar} will report @samp{tar: ./foo.tar is the archive; not
1213 dumped}. This happens because @command{tar} creates the archive
1214 @file{foo.tar} in the current directory before putting any files into
1215 it. Then, when @command{tar} attempts to add all the files in the
1216 directory @file{.} to the archive, it notices that the file
1217 @file{./foo.tar} is the same as the archive @file{foo.tar}, and skips
1218 it. (It makes no sense to put an archive into itself.) @GNUTAR{}
1219 will continue in this case, and create the archive
1220 normally, except for the exclusion of that one file. (@emph{Please
1221 note:} Other implementations of @command{tar} may not be so clever;
1222 they will enter an infinite loop when this happens, so you should not
1223 depend on this behavior unless you are certain you are running
1224 @GNUTAR{}. In general, it is wise to always place the archive outside
1225 of the directory being dumped.
1228 @section How to List Archives
1231 Frequently, you will find yourself wanting to determine exactly what a
1232 particular archive contains. You can use the @option{--list} (@option{-t}) operation
1233 to get the member names as they currently appear in the archive, as well
1234 as various attributes of the files at the time they were archived. For
1235 example, you can examine the archive @file{collection.tar} that you
1236 created in the last section with the command,
1239 $ @kbd{tar --list --file=collection.tar}
1243 The output of @command{tar} would then be:
1252 The archive @file{bfiles.tar} would list as follows:
1261 Be sure to use a @option{--file=@var{archive-name}} (@option{-f
1262 @var{archive-name}}) option just as with @option{--create}
1263 (@option{-c}) to specify the name of the archive.
1265 @opindex list, using with @option{--verbose}
1266 @opindex verbose, using with @option{--list}
1267 If you use the @option{--verbose} (@option{-v}) option with
1268 @option{--list}, then @command{tar} will print out a listing
1269 reminiscent of @w{@samp{ls -l}}, showing owner, file size, and so forth.
1271 If you had used @option{--verbose} (@option{-v}) mode, the example
1272 above would look like:
1275 $ @kbd{tar --list --verbose --file=collection.tar folk}
1276 -rw-r--r-- myself user 62 1990-05-23 10:55 folk
1279 @cindex listing member and file names
1280 @anchor{listing member and file names}
1281 It is important to notice that the output of @kbd{tar --list
1282 --verbose} does not necessarily match that produced by @kbd{tar
1283 --create --verbose} while creating the archive. It is because
1284 @GNUTAR{}, unless told explicitly not to do so, removes some directory
1285 prefixes from file names before storing them in the archive
1286 (@xref{absolute}, for more information). In other
1287 words, in verbose mode @GNUTAR{} shows @dfn{file names} when creating
1288 an archive and @dfn{member names} when listing it. Consider this
1293 $ @kbd{tar cfv archive /etc/mail}
1294 tar: Removing leading `/' from member names
1296 /etc/mail/sendmail.cf
1298 $ @kbd{tar tf archive}
1300 etc/mail/sendmail.cf
1305 @opindex show-stored-names
1306 This default behavior can sometimes be inconvenient. You can force
1307 @GNUTAR{} show member names when creating archive by supplying
1308 @option{--show-stored-names} option.
1311 @item --show-stored-names
1312 Print member (as opposed to @emph{file}) names when creating the archive.
1315 @cindex File name arguments, using @option{--list} with
1316 @opindex list, using with file name arguments
1317 You can specify one or more individual member names as arguments when
1318 using @samp{list}. In this case, @command{tar} will only list the
1319 names of members you identify. For example, @w{@kbd{tar --list
1320 --file=afiles.tar apple}} would only print @file{apple}.
1322 Because @command{tar} preserves paths, file names must be specified as
1323 they appear in the archive (ie., relative to the directory from which
1324 the archive was created). Therefore, it is essential when specifying
1325 member names to @command{tar} that you give the exact member names.
1326 For example, @w{@kbd{tar --list --file=bfiles.tar birds}} would produce an
1327 error message something like @samp{tar: birds: Not found in archive},
1328 because there is no member named @file{birds}, only one named
1329 @file{./birds}. While the names @file{birds} and @file{./birds} name
1330 the same file, @emph{member} names by default are compared verbatim.
1332 However, @w{@kbd{tar --list --file=bfiles.tar baboon}} would respond
1333 with @file{baboon}, because this exact member name is in the archive file
1334 @file{bfiles.tar}. If you are not sure of the exact file name,
1335 use @dfn{globbing patterns}, for example:
1338 $ @kbd{tar --list --file=bfiles.tar --wildcards '*b*'}
1342 will list all members whose name contains @samp{b}. @xref{wildcards},
1343 for a detailed discussion of globbing patterns and related
1344 @command{tar} command line options.
1351 @unnumberedsubsec Listing the Contents of a Stored Directory
1353 To get information about the contents of an archived directory,
1354 use the directory name as a file name argument in conjunction with
1355 @option{--list} (@option{-t}). To find out file attributes, include the
1356 @option{--verbose} (@option{-v}) option.
1358 For example, to find out about files in the directory @file{practice}, in
1359 the archive file @file{music.tar}, type:
1362 $ @kbd{tar --list --verbose --file=music.tar practice}
1365 @command{tar} responds:
1368 drwxrwxrwx myself user 0 1990-05-31 21:49 practice/
1369 -rw-r--r-- myself user 42 1990-05-21 13:29 practice/blues
1370 -rw-r--r-- myself user 62 1990-05-23 10:55 practice/folk
1371 -rw-r--r-- myself user 40 1990-05-21 13:30 practice/jazz
1372 -rw-r--r-- myself user 10240 1990-05-31 21:49 practice/collection.tar
1375 When you use a directory name as a file name argument, @command{tar} acts on
1376 all the files (including sub-directories) in that directory.
1379 @section How to Extract Members from an Archive
1382 @cindex Retrieving files from an archive
1383 @cindex Resurrecting files from an archive
1386 Creating an archive is only half the job---there is no point in storing
1387 files in an archive if you can't retrieve them. The act of retrieving
1388 members from an archive so they can be used and manipulated as
1389 unarchived files again is called @dfn{extraction}. To extract files
1390 from an archive, use the @option{--extract} (@option{--get} or
1391 @option{-x}) operation. As with @option{--create}, specify the name
1392 of the archive with @option{--file} (@option{-f}) option. Extracting
1393 an archive does not modify the archive in any way; you can extract it
1394 multiple times if you want or need to.
1396 Using @option{--extract}, you can extract an entire archive, or specific
1397 files. The files can be directories containing other files, or not. As
1398 with @option{--create} (@option{-c}) and @option{--list} (@option{-t}), you may use the short or the
1399 long form of the operation without affecting the performance.
1402 * extracting archives::
1403 * extracting files::
1405 * extracting untrusted archives::
1406 * failing commands::
1409 @node extracting archives
1410 @subsection Extracting an Entire Archive
1412 To extract an entire archive, specify the archive file name only, with
1413 no individual file names as arguments. For example,
1416 $ @kbd{tar -xvf collection.tar}
1423 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
1424 -rw-r--r-- me user 21 1996-09-23 16:44 blues
1425 -rw-r--r-- me user 20 1996-09-23 16:44 folk
1428 @node extracting files
1429 @subsection Extracting Specific Files
1431 To extract specific archive members, give their exact member names as
1432 arguments, as printed by @option{--list} (@option{-t}). If you had
1433 mistakenly deleted one of the files you had placed in the archive
1434 @file{collection.tar} earlier (say, @file{blues}), you can extract it
1435 from the archive without changing the archive's structure. Its
1436 contents will be identical to the original file @file{blues} that you
1439 First, make sure you are in the @file{practice} directory, and list the
1440 files in the directory. Now, delete the file, @samp{blues}, and list
1441 the files in the directory again.
1443 You can now extract the member @file{blues} from the archive file
1444 @file{collection.tar} like this:
1447 $ @kbd{tar --extract --file=collection.tar blues}
1451 If you list the files in the directory again, you will see that the file
1452 @file{blues} has been restored, with its original permissions, data
1453 modification times, and owner.@footnote{This is only accidentally
1454 true, but not in general. Whereas modification times are always
1455 restored, in most cases, one has to be root for restoring the owner,
1456 and use a special option for restoring permissions. Here, it just
1457 happens that the restoring user is also the owner of the archived
1458 members, and that the current @code{umask} is compatible with original
1459 permissions.} (These parameters will be identical to those which
1460 the file had when you originally placed it in the archive; any changes
1461 you may have made before deleting the file from the file system,
1462 however, will @emph{not} have been made to the archive member.) The
1463 archive file, @samp{collection.tar}, is the same as it was before you
1464 extracted @samp{blues}. You can confirm this by running @command{tar} with
1465 @option{--list} (@option{-t}).
1467 Remember that as with other operations, specifying the exact member
1468 name is important. @w{@kbd{tar --extract --file=bfiles.tar birds}}
1469 will fail, because there is no member named @file{birds}. To extract
1470 the member named @file{./birds}, you must specify @w{@kbd{tar
1471 --extract --file=bfiles.tar ./birds}}. If you don't remember the
1472 exact member names, use @option{--list} (@option{-t}) option
1473 (@pxref{list}). You can also extract those members that match a
1474 specific @dfn{globbing pattern}. For example, to extract from
1475 @file{bfiles.tar} all files that begin with @samp{b}, no matter their
1476 directory prefix, you could type:
1479 $ @kbd{tar -x -f bfiles.tar --wildcards --no-anchored 'b*'}
1483 Here, @option{--wildcards} instructs @command{tar} to treat
1484 command line arguments as globbing patterns and @option{--no-anchored}
1485 informs it that the patterns apply to member names after any @samp{/}
1486 delimiter. The use of globbing patterns is discussed in detail in
1489 You can extract a file to standard output by combining the above options
1490 with the @option{--to-stdout} (@option{-O}) option (@pxref{Writing to Standard
1493 If you give the @option{--verbose} option, then @option{--extract}
1494 will print the names of the archive members as it extracts them.
1497 @subsection Extracting Files that are Directories
1499 Extracting directories which are members of an archive is similar to
1500 extracting other files. The main difference to be aware of is that if
1501 the extracted directory has the same name as any directory already in
1502 the working directory, then files in the extracted directory will be
1503 placed into the directory of the same name. Likewise, if there are
1504 files in the pre-existing directory with the same names as the members
1505 which you extract, the files from the extracted archive will replace
1506 the files already in the working directory (and possible
1507 subdirectories). This will happen regardless of whether or not the
1508 files in the working directory were more recent than those extracted
1509 (there exist, however, special options that alter this behavior
1512 However, if a file was stored with a directory name as part of its file
1513 name, and that directory does not exist under the working directory when
1514 the file is extracted, @command{tar} will create the directory.
1516 We can demonstrate how to use @option{--extract} to extract a directory
1517 file with an example. Change to the @file{practice} directory if you
1518 weren't there, and remove the files @file{folk} and @file{jazz}. Then,
1519 go back to the parent directory and extract the archive
1520 @file{music.tar}. You may either extract the entire archive, or you may
1521 extract only the files you just deleted. To extract the entire archive,
1522 don't give any file names as arguments after the archive name
1523 @file{music.tar}. To extract only the files you deleted, use the
1527 $ @kbd{tar -xvf music.tar practice/folk practice/jazz}
1533 If you were to specify two @option{--verbose} (@option{-v}) options, @command{tar}
1534 would have displayed more detail about the extracted files, as shown
1535 in the example below:
1538 $ @kbd{tar -xvvf music.tar practice/folk practice/jazz}
1539 -rw-r--r-- me user 28 1996-10-18 16:31 practice/jazz
1540 -rw-r--r-- me user 20 1996-09-23 16:44 practice/folk
1544 Because you created the directory with @file{practice} as part of the
1545 file names of each of the files by archiving the @file{practice}
1546 directory as @file{practice}, you must give @file{practice} as part
1547 of the file names when you extract those files from the archive.
1549 @node extracting untrusted archives
1550 @subsection Extracting Archives from Untrusted Sources
1552 Extracting files from archives can overwrite files that already exist.
1553 If you receive an archive from an untrusted source, you should make a
1554 new directory and extract into that directory, so that you don't have
1555 to worry about the extraction overwriting one of your existing files.
1556 For example, if @file{untrusted.tar} came from somewhere else on the
1557 Internet, and you don't necessarily trust its contents, you can
1558 extract it as follows:
1561 $ @kbd{mkdir newdir}
1563 $ @kbd{tar -xvf ../untrusted.tar}
1566 It is also a good practice to examine contents of the archive
1567 before extracting it, using @option{--list} (@option{-t}) option, possibly combined
1568 with @option{--verbose} (@option{-v}).
1570 @node failing commands
1571 @subsection Commands That Will Fail
1573 Here are some sample commands you might try which will not work, and why
1576 If you try to use this command,
1579 $ @kbd{tar -xvf music.tar folk jazz}
1583 you will get the following response:
1586 tar: folk: Not found in archive
1587 tar: jazz: Not found in archive
1592 This is because these files were not originally @emph{in} the parent
1593 directory @file{..}, where the archive is located; they were in the
1594 @file{practice} directory, and their file names reflect this:
1597 $ @kbd{tar -tvf music.tar}
1603 @FIXME{make sure the above works when going through the examples in
1607 Likewise, if you try to use this command,
1610 $ @kbd{tar -tvf music.tar folk jazz}
1614 you would get a similar response. Members with those names are not in the
1615 archive. You must use the correct member names, or wildcards, in order
1616 to extract the files from the archive.
1618 If you have forgotten the correct names of the files in the archive,
1619 use @w{@kbd{tar --list --verbose}} to list them correctly.
1621 @FIXME{more examples, here? hag thinks it's a good idea.}
1624 @section Going Further Ahead in this Manual
1626 @FIXME{need to write up a node here about the things that are going to
1627 be in the rest of the manual.}
1629 @node tar invocation
1630 @chapter Invoking @GNUTAR{}
1633 This chapter is about how one invokes the @GNUTAR{}
1634 command, from the command synopsis (@pxref{Synopsis}). There are
1635 numerous options, and many styles for writing them. One mandatory
1636 option specifies the operation @command{tar} should perform
1637 (@pxref{Operation Summary}), other options are meant to detail how
1638 this operation should be performed (@pxref{Option Summary}).
1639 Non-option arguments are not always interpreted the same way,
1640 depending on what the operation is.
1642 You will find in this chapter everything about option styles and rules for
1643 writing them (@pxref{Styles}). On the other hand, operations and options
1644 are fully described elsewhere, in other chapters. Here, you will find
1645 only synthetic descriptions for operations and options, together with
1646 pointers to other parts of the @command{tar} manual.
1648 Some options are so special they are fully described right in this
1649 chapter. They have the effect of inhibiting the normal operation of
1650 @command{tar} or else, they globally alter the amount of feedback the user
1651 receives about what is going on. These are the @option{--help} and
1652 @option{--version} (@pxref{help}), @option{--verbose} (@pxref{verbose})
1653 and @option{--interactive} options (@pxref{interactive}).
1657 * using tar options::
1667 @section General Synopsis of @command{tar}
1669 The @GNUTAR{} program is invoked as either one of:
1672 @kbd{tar @var{option}@dots{} [@var{name}]@dots{}}
1673 @kbd{tar @var{letter}@dots{} [@var{argument}]@dots{} [@var{option}]@dots{} [@var{name}]@dots{}}
1676 The second form is for when old options are being used.
1678 You can use @command{tar} to store files in an archive, to extract them from
1679 an archive, and to do other types of archive manipulation. The primary
1680 argument to @command{tar}, which is called the @dfn{operation}, specifies
1681 which action to take. The other arguments to @command{tar} are either
1682 @dfn{options}, which change the way @command{tar} performs an operation,
1683 or file names or archive members, which specify the files or members
1684 @command{tar} is to act on.
1686 You can actually type in arguments in any order, even if in this manual
1687 the options always precede the other arguments, to make examples easier
1688 to understand. Further, the option stating the main operation mode
1689 (the @command{tar} main command) is usually given first.
1691 Each @var{name} in the synopsis above is interpreted as an archive member
1692 name when the main command is one of @option{--compare}
1693 (@option{--diff}, @option{-d}), @option{--delete}, @option{--extract}
1694 (@option{--get}, @option{-x}), @option{--list} (@option{-t}) or
1695 @option{--update} (@option{-u}). When naming archive members, you
1696 must give the exact name of the member in the archive, as it is
1697 printed by @option{--list}. For @option{--append} (@option{-r}) and
1698 @option{--create} (@option{-c}), these @var{name} arguments specify
1699 the names of either files or directory hierarchies to place in the archive.
1700 These files or hierarchies should already exist in the file system,
1701 prior to the execution of the @command{tar} command.
1703 @command{tar} interprets relative file names as being relative to the
1704 working directory. @command{tar} will make all file names relative
1705 (by removing leading slashes when archiving or restoring files),
1706 unless you specify otherwise (using the @option{--absolute-names}
1707 option). @xref{absolute}, for more information about
1708 @option{--absolute-names}.
1710 If you give the name of a directory as either a file name or a member
1711 name, then @command{tar} acts recursively on all the files and directories
1712 beneath that directory. For example, the name @file{/} identifies all
1713 the files in the file system to @command{tar}.
1715 The distinction between file names and archive member names is especially
1716 important when shell globbing is used, and sometimes a source of confusion
1717 for newcomers. @xref{wildcards}, for more information about globbing.
1718 The problem is that shells may only glob using existing files in the
1719 file system. Only @command{tar} itself may glob on archive members, so when
1720 needed, you must ensure that wildcard characters reach @command{tar} without
1721 being interpreted by the shell first. Using a backslash before @samp{*}
1722 or @samp{?}, or putting the whole argument between quotes, is usually
1723 sufficient for this.
1725 Even if @var{name}s are often specified on the command line, they
1726 can also be read from a text file in the file system, using the
1727 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}}) option.
1729 If you don't use any file name arguments, @option{--append} (@option{-r}),
1730 @option{--delete} and @option{--concatenate} (@option{--catenate},
1731 @option{-A}) will do nothing, while @option{--create} (@option{-c})
1732 will usually yield a diagnostic and inhibit @command{tar} execution.
1733 The other operations of @command{tar} (@option{--list},
1734 @option{--extract}, @option{--compare}, and @option{--update})
1735 will act on the entire contents of the archive.
1738 @cindex return status
1739 Besides successful exits, @GNUTAR{} may fail for
1740 many reasons. Some reasons correspond to bad usage, that is, when the
1741 @command{tar} command is improperly written. Errors may be
1742 encountered later, while encountering an error processing the archive
1743 or the files. Some errors are recoverable, in which case the failure
1744 is delayed until @command{tar} has completed all its work. Some
1745 errors are such that it would not meaningful, or at least risky, to
1746 continue processing: @command{tar} then aborts processing immediately.
1747 All abnormal exits, whether immediate or delayed, should always be
1748 clearly diagnosed on @code{stderr}, after a line stating the nature of
1751 @GNUTAR{} returns only a few exit statuses. I'm really
1752 aiming simplicity in that area, for now. If you are not using the
1753 @option{--compare} @option{--diff}, @option{-d}) option, zero means
1754 that everything went well, besides maybe innocuous warnings. Nonzero
1755 means that something went wrong. Right now, as of today, ``nonzero''
1756 is almost always 2, except for remote operations, where it may be
1759 @node using tar options
1760 @section Using @command{tar} Options
1762 @GNUTAR{} has a total of eight operating modes which
1763 allow you to perform a variety of tasks. You are required to choose
1764 one operating mode each time you employ the @command{tar} program by
1765 specifying one, and only one operation as an argument to the
1766 @command{tar} command (two lists of four operations each may be found
1767 at @ref{frequent operations} and @ref{Operations}). Depending on
1768 circumstances, you may also wish to customize how the chosen operating
1769 mode behaves. For example, you may wish to change the way the output
1770 looks, or the format of the files that you wish to archive may require
1771 you to do something special in order to make the archive look right.
1773 You can customize and control @command{tar}'s performance by running
1774 @command{tar} with one or more options (such as @option{--verbose}
1775 (@option{-v}), which we used in the tutorial). As we said in the
1776 tutorial, @dfn{options} are arguments to @command{tar} which are (as
1777 their name suggests) optional. Depending on the operating mode, you
1778 may specify one or more options. Different options will have different
1779 effects, but in general they all change details of the operation, such
1780 as archive format, archive name, or level of user interaction. Some
1781 options make sense with all operating modes, while others are
1782 meaningful only with particular modes. You will likely use some
1783 options frequently, while you will only use others infrequently, or
1784 not at all. (A full list of options is available in @pxref{All Options}.)
1786 @vrindex TAR_OPTIONS, environment variable
1787 @anchor{TAR_OPTIONS}
1788 The @env{TAR_OPTIONS} environment variable specifies default options to
1789 be placed in front of any explicit options. For example, if
1790 @code{TAR_OPTIONS} is @samp{-v --unlink-first}, @command{tar} behaves as
1791 if the two options @option{-v} and @option{--unlink-first} had been
1792 specified before any explicit options. Option specifications are
1793 separated by whitespace. A backslash escapes the next character, so it
1794 can be used to specify an option containing whitespace or a backslash.
1796 Note that @command{tar} options are case sensitive. For example, the
1797 options @option{-T} and @option{-t} are different; the first requires an
1798 argument for stating the name of a file providing a list of @var{name}s,
1799 while the second does not require an argument and is another way to
1800 write @option{--list} (@option{-t}).
1802 In addition to the eight operations, there are many options to
1803 @command{tar}, and three different styles for writing both: long (mnemonic)
1804 form, short form, and old style. These styles are discussed below.
1805 Both the options and the operations can be written in any of these three
1808 @FIXME{menu at end of this node. need to think of an actual outline
1809 for this chapter; probably do that after stuff from chapter 4 is
1813 @section The Three Option Styles
1815 There are three styles for writing operations and options to the command
1816 line invoking @command{tar}. The different styles were developed at
1817 different times during the history of @command{tar}. These styles will be
1818 presented below, from the most recent to the oldest.
1820 Some options must take an argument. (For example, @option{--file}
1821 (@option{-f})) takes the name of an archive file as an argument. If
1822 you do not supply an archive file name, @command{tar} will use a
1823 default, but this can be confusing; thus, we recommend that you always
1824 supply a specific archive file name.) Where you @emph{place} the
1825 arguments generally depends on which style of options you choose. We
1826 will detail specific information relevant to each option style in the
1827 sections on the different option styles, below. The differences are
1828 subtle, yet can often be very important; incorrect option placement
1829 can cause you to overwrite a number of important files. We urge you
1830 to note these differences, and only use the option style(s) which
1831 makes the most sense to you until you feel comfortable with the others.
1833 Some options @emph{may} take an argument (currently, there are
1834 two such options: @option{--backup} and @option{--occurrence}). Such
1835 options may have at most long and short forms, they do not have old style
1836 equivalent. The rules for specifying an argument for such options
1837 are stricter than those for specifying mandatory arguments. Please,
1838 pay special attention to them.
1841 * Mnemonic Options:: Mnemonic Option Style
1842 * Short Options:: Short Option Style
1843 * Old Options:: Old Option Style
1844 * Mixing:: Mixing Option Styles
1847 @node Mnemonic Options
1848 @subsection Mnemonic Option Style
1850 @FIXME{have to decide whether or not to replace other occurrences of
1851 "mnemonic" with "long", or *ugh* vice versa.}
1853 Each option has at least one long (or mnemonic) name starting with two
1854 dashes in a row, e.g., @option{--list}. The long names are more clear than
1855 their corresponding short or old names. It sometimes happens that a
1856 single mnemonic option has many different different names which are
1857 synonymous, such as @option{--compare} and @option{--diff}. In addition,
1858 long option names can be given unique abbreviations. For example,
1859 @option{--cre} can be used in place of @option{--create} because there is no
1860 other mnemonic option which begins with @samp{cre}. (One way to find
1861 this out is by trying it and seeing what happens; if a particular
1862 abbreviation could represent more than one option, @command{tar} will tell
1863 you that that abbreviation is ambiguous and you'll know that that
1864 abbreviation won't work. You may also choose to run @samp{tar --help}
1865 to see a list of options. Be aware that if you run @command{tar} with a
1866 unique abbreviation for the long name of an option you didn't want to
1867 use, you are stuck; @command{tar} will perform the command as ordered.)
1869 Mnemonic options are meant to be obvious and easy to remember, and their
1870 meanings are generally easier to discern than those of their
1871 corresponding short options (see below). For example:
1874 $ @kbd{tar --create --verbose --blocking-factor=20 --file=/dev/rmt0}
1878 gives a fairly good set of hints about what the command does, even
1879 for those not fully acquainted with @command{tar}.
1881 Mnemonic options which require arguments take those arguments
1882 immediately following the option name. There are two ways of
1883 specifying a mandatory argument. It can be separated from the
1884 option name either by an equal sign, or by any amount of
1885 white space characters. For example, the @option{--file} option (which
1886 tells the name of the @command{tar} archive) is given a file such as
1887 @file{archive.tar} as argument by using any of the following notations:
1888 @option{--file=archive.tar} or @option{--file archive.tar}.
1890 In contrast, optional arguments must always be introduced using
1891 an equal sign. For example, the @option{--backup} option takes
1892 an optional argument specifying backup type. It must be used
1893 as @option{--backup=@var{backup-type}}.
1896 @subsection Short Option Style
1898 Most options also have a short option name. Short options start with
1899 a single dash, and are followed by a single character, e.g., @option{-t}
1900 (which is equivalent to @option{--list}). The forms are absolutely
1901 identical in function; they are interchangeable.
1903 The short option names are faster to type than long option names.
1905 Short options which require arguments take their arguments immediately
1906 following the option, usually separated by white space. It is also
1907 possible to stick the argument right after the short option name, using
1908 no intervening space. For example, you might write @w{@option{-f
1909 archive.tar}} or @option{-farchive.tar} instead of using
1910 @option{--file=archive.tar}. Both @option{--file=@var{archive-name}} and
1911 @w{@option{-f @var{archive-name}}} denote the option which indicates a
1912 specific archive, here named @file{archive.tar}.
1914 Short options which take optional arguments take their arguments
1915 immediately following the option letter, @emph{without any intervening
1916 white space characters}.
1918 Short options' letters may be clumped together, but you are not
1919 required to do this (as compared to old options; see below). When
1920 short options are clumped as a set, use one (single) dash for them
1921 all, e.g., @w{@samp{@command{tar} -cvf}}. Only the last option in
1922 such a set is allowed to have an argument@footnote{Clustering many
1923 options, the last of which has an argument, is a rather opaque way to
1924 write options. Some wonder if @acronym{GNU} @code{getopt} should not
1925 even be made helpful enough for considering such usages as invalid.}.
1927 When the options are separated, the argument for each option which requires
1928 an argument directly follows that option, as is usual for Unix programs.
1932 $ @kbd{tar -c -v -b 20 -f /dev/rmt0}
1935 If you reorder short options' locations, be sure to move any arguments
1936 that belong to them. If you do not move the arguments properly, you may
1937 end up overwriting files.
1940 @subsection Old Option Style
1943 Like short options, old options are single letters. However, old options
1944 must be written together as a single clumped set, without spaces separating
1945 them or dashes preceding them@footnote{Beware that if you precede options
1946 with a dash, you are announcing the short option style instead of the
1947 old option style; short options are decoded differently.}. This set
1948 of letters must be the first to appear on the command line, after the
1949 @command{tar} program name and some white space; old options cannot appear
1950 anywhere else. The letter of an old option is exactly the same letter as
1951 the corresponding short option. For example, the old option @samp{t} is
1952 the same as the short option @option{-t}, and consequently, the same as the
1953 mnemonic option @option{--list}. So for example, the command @w{@samp{tar
1954 cv}} specifies the option @option{-v} in addition to the operation @option{-c}.
1956 When options that need arguments are given together with the command,
1957 all the associated arguments follow, in the same order as the options.
1958 Thus, the example given previously could also be written in the old
1962 $ @kbd{tar cvbf 20 /dev/rmt0}
1966 Here, @samp{20} is the argument of @option{-b} and @samp{/dev/rmt0} is
1967 the argument of @option{-f}.
1969 On the other hand, this old style syntax makes it difficult to match
1970 option letters with their corresponding arguments, and is often
1971 confusing. In the command @w{@samp{tar cvbf 20 /dev/rmt0}}, for example,
1972 @samp{20} is the argument for @option{-b}, @samp{/dev/rmt0} is the
1973 argument for @option{-f}, and @option{-v} does not have a corresponding
1974 argument. Even using short options like in @w{@samp{tar -c -v -b 20 -f
1975 /dev/rmt0}} is clearer, putting all arguments next to the option they
1978 If you want to reorder the letters in the old option argument, be
1979 sure to reorder any corresponding argument appropriately.
1981 This old way of writing @command{tar} options can surprise even experienced
1982 users. For example, the two commands:
1985 @kbd{tar cfz archive.tar.gz file}
1986 @kbd{tar -cfz archive.tar.gz file}
1990 are quite different. The first example uses @file{archive.tar.gz} as
1991 the value for option @samp{f} and recognizes the option @samp{z}. The
1992 second example, however, uses @file{z} as the value for option
1993 @samp{f} --- probably not what was intended.
1995 Old options are kept for compatibility with old versions of @command{tar}.
1997 This second example could be corrected in many ways, among which the
1998 following are equivalent:
2001 @kbd{tar -czf archive.tar.gz file}
2002 @kbd{tar -cf archive.tar.gz -z file}
2003 @kbd{tar cf archive.tar.gz -z file}
2006 @cindex option syntax, traditional
2007 As far as we know, all @command{tar} programs, @acronym{GNU} and
2008 non-@acronym{GNU}, support old options. @GNUTAR{}
2009 supports them not only for historical reasons, but also because many
2010 people are used to them. For compatibility with Unix @command{tar},
2011 the first argument is always treated as containing command and option
2012 letters even if it doesn't start with @samp{-}. Thus, @samp{tar c} is
2013 equivalent to @w{@samp{tar -c}:} both of them specify the
2014 @option{--create} (@option{-c}) command to create an archive.
2017 @subsection Mixing Option Styles
2019 All three styles may be intermixed in a single @command{tar} command,
2020 so long as the rules for each style are fully
2021 respected@footnote{Before @GNUTAR{} version 1.11.6,
2022 a bug prevented intermixing old style options with mnemonic options in
2023 some cases.}. Old style options and either of the modern styles of
2024 options may be mixed within a single @command{tar} command. However,
2025 old style options must be introduced as the first arguments only,
2026 following the rule for old options (old options must appear directly
2027 after the @command{tar} command and some white space). Modern options
2028 may be given only after all arguments to the old options have been
2029 collected. If this rule is not respected, a modern option might be
2030 falsely interpreted as the value of the argument to one of the old
2033 For example, all the following commands are wholly equivalent, and
2034 illustrate the many combinations and orderings of option styles.
2037 @kbd{tar --create --file=archive.tar}
2038 @kbd{tar --create -f archive.tar}
2039 @kbd{tar --create -farchive.tar}
2040 @kbd{tar --file=archive.tar --create}
2041 @kbd{tar --file=archive.tar -c}
2042 @kbd{tar -c --file=archive.tar}
2043 @kbd{tar -c -f archive.tar}
2044 @kbd{tar -c -farchive.tar}
2045 @kbd{tar -cf archive.tar}
2046 @kbd{tar -cfarchive.tar}
2047 @kbd{tar -f archive.tar --create}
2048 @kbd{tar -f archive.tar -c}
2049 @kbd{tar -farchive.tar --create}
2050 @kbd{tar -farchive.tar -c}
2051 @kbd{tar c --file=archive.tar}
2052 @kbd{tar c -f archive.tar}
2053 @kbd{tar c -farchive.tar}
2054 @kbd{tar cf archive.tar}
2055 @kbd{tar f archive.tar --create}
2056 @kbd{tar f archive.tar -c}
2057 @kbd{tar fc archive.tar}
2060 On the other hand, the following commands are @emph{not} equivalent to
2064 @kbd{tar -f -c archive.tar}
2065 @kbd{tar -fc archive.tar}
2066 @kbd{tar -fcarchive.tar}
2067 @kbd{tar -farchive.tarc}
2068 @kbd{tar cfarchive.tar}
2072 These last examples mean something completely different from what the
2073 user intended (judging based on the example in the previous set which
2074 uses long options, whose intent is therefore very clear). The first
2075 four specify that the @command{tar} archive would be a file named
2076 @option{-c}, @samp{c}, @samp{carchive.tar} or @samp{archive.tarc},
2077 respectively. The first two examples also specify a single non-option,
2078 @var{name} argument having the value @samp{archive.tar}. The last
2079 example contains only old style option letters (repeating option
2080 @samp{c} twice), not all of which are meaningful (eg., @samp{.},
2081 @samp{h}, or @samp{i}), with no argument value. @FIXME{not sure i liked
2082 the first sentence of this paragraph..}
2085 @section All @command{tar} Options
2087 The coming manual sections contain an alphabetical listing of all
2088 @command{tar} operations and options, with brief descriptions and cross
2089 references to more in-depth explanations in the body of the manual.
2090 They also contain an alphabetically arranged table of the short option
2091 forms with their corresponding long option. You can use this table as
2092 a reference for deciphering @command{tar} commands in scripts.
2095 * Operation Summary::
2097 * Short Option Summary::
2100 @node Operation Summary
2101 @subsection Operations
2105 @opindex append, summary
2109 Appends files to the end of the archive. @xref{append}.
2111 @opindex catenate, summary
2115 Same as @option{--concatenate}. @xref{concatenate}.
2117 @opindex compare, summary
2121 Compares archive members with their counterparts in the file
2122 system, and reports differences in file size, mode, owner,
2123 modification date and contents. @xref{compare}.
2125 @opindex concatenate, summary
2129 Appends other @command{tar} archives to the end of the archive.
2132 @opindex create, summary
2136 Creates a new @command{tar} archive. @xref{create}.
2138 @opindex delete, summary
2141 Deletes members from the archive. Don't try this on a archive on a
2142 tape! @xref{delete}.
2144 @opindex diff, summary
2148 Same @option{--compare}. @xref{compare}.
2150 @opindex extract, summary
2154 Extracts members from the archive into the file system. @xref{extract}.
2156 @opindex get, summary
2160 Same as @option{--extract}. @xref{extract}.
2162 @opindex list, summary
2166 Lists the members in an archive. @xref{list}.
2168 @opindex update, summary
2172 Adds files to the end of the archive, but only if they are newer than
2173 their counterparts already in the archive, or if they do not already
2174 exist in the archive. @xref{update}.
2178 @node Option Summary
2179 @subsection @command{tar} Options
2183 @opindex absolute-names, summary
2184 @item --absolute-names
2187 Normally when creating an archive, @command{tar} strips an initial
2188 @samp{/} from member names. This option disables that behavior.
2191 @opindex after-date, summary
2194 (See @option{--newer}, @pxref{after})
2196 @opindex anchored, summary
2198 A pattern must match an initial subsequence of the name's components.
2199 @xref{controlling pattern-matching}.
2201 @opindex atime-preserve, summary
2202 @item --atime-preserve
2203 @itemx --atime-preserve=replace
2204 @itemx --atime-preserve=system
2206 Attempt to preserve the access time of files when reading them. This
2207 option currently is effective only on files that you own, unless you
2208 have superuser privileges.
2210 @option{--atime-preserve=replace} remembers the access time of a file
2211 before reading it, and then restores the access time afterwards. This
2212 may cause problems if other programs are reading the file at the same
2213 time, as the times of their accesses will be lost. On most platforms
2214 restoring the access time also requires @command{tar} to restore the
2215 data modification time too, so this option may also cause problems if
2216 other programs are writing the file at the same time. (Tar attempts
2217 to detect this situation, but cannot do so reliably due to race
2218 conditions.) Worse, on most platforms restoring the access time also
2219 updates the status change time, which means that this option is
2220 incompatible with incremental backups.
2222 @option{--atime-preserve=system} avoids changing time stamps on files,
2223 without interfering with time stamp updates
2224 caused by other programs, so it works better with incremental backups.
2225 However, it requires a special @code{O_NOATIME} option from the
2226 underlying operating and file system implementation, and it also requires
2227 that searching directories does not update their access times. As of
2228 this writing (November 2005) this works only with Linux, and only with
2229 Linux kernels 2.6.8 and later. Worse, there is currently no reliable
2230 way to know whether this feature actually works. Sometimes
2231 @command{tar} knows that it does not work, and if you use
2232 @option{--atime-preserve=system} then @command{tar} complains and
2233 exits right away. But other times @command{tar} might think that the
2234 option works when it actually does not.
2236 Currently @option{--atime-preserve} with no operand defaults to
2237 @option{--atime-preserve=replace}, but this may change in the future
2238 as support for @option{--atime-preserve=system} improves.
2240 If your operating system does not support
2241 @option{--atime-preserve=system}, you might be able to preserve access
2242 times reliably by by using the @command{mount} command. For example,
2243 you can mount the file system read-only, or access the file system via
2244 a read-only loopback mount, or use the @samp{noatime} mount option
2245 available on some systems. However, mounting typically requires
2246 superuser privileges and can be a pain to manage.
2248 @opindex backup, summary
2249 @item --backup=@var{backup-type}
2251 Rather than deleting files from the file system, @command{tar} will
2252 back them up using simple or numbered backups, depending upon
2253 @var{backup-type}. @xref{backup}.
2255 @opindex block-number, summary
2256 @item --block-number
2259 With this option present, @command{tar} prints error messages for read errors
2260 with the block number in the archive file. @xref{block-number}.
2262 @opindex blocking-factor, summary
2263 @item --blocking-factor=@var{blocking}
2264 @itemx -b @var{blocking}
2266 Sets the blocking factor @command{tar} uses to @var{blocking} x 512 bytes per
2267 record. @xref{Blocking Factor}.
2269 @opindex bzip2, summary
2273 This option tells @command{tar} to read or write archives through
2274 @code{bzip2}. @xref{gzip}.
2276 @opindex checkpoint, summary
2279 This option directs @command{tar} to print periodic checkpoint messages as it
2280 reads through the archive. It is intended for when you want a visual
2281 indication that @command{tar} is still running, but don't want to see
2282 @option{--verbose} output. @FIXME-xref{}
2284 @opindex check-links, summary
2287 If this option was given, @command{tar} will check the number of links
2288 dumped for each processed file. If this number does not match the
2289 total number of hard links for the file, a warning message will be
2290 output @footnote{Earlier versions of @GNUTAR{} understood @option{-l} as a
2291 synonym for @option{--one-file-system}. The current semantics, wich
2292 complies to UNIX98, was introduced with version
2293 1.15.91. @xref{Changes}, for more information.}.
2295 @opindex compress, summary
2296 @opindex uncompress, summary
2301 @command{tar} will use the @command{compress} program when reading or
2302 writing the archive. This allows you to directly act on archives
2303 while saving space. @xref{gzip}.
2305 @opindex confirmation, summary
2306 @item --confirmation
2308 (See @option{--interactive}.) @xref{interactive}.
2310 @opindex delay-directory-restore, summary
2311 @item --delay-directory-restore
2313 Delay setting modification times and permissions of extracted
2314 directories until the end of extraction. @xref{Directory Modification Times and Permissions}.
2316 @opindex dereference, summary
2320 When creating a @command{tar} archive, @command{tar} will archive the
2321 file that a symbolic link points to, rather than archiving the
2322 symlink. @xref{dereference}.
2324 @opindex directory, summary
2325 @item --directory=@var{dir}
2328 When this option is specified, @command{tar} will change its current directory
2329 to @var{dir} before performing any operations. When this option is used
2330 during archive creation, it is order sensitive. @xref{directory}.
2332 @opindex exclude, summary
2333 @item --exclude=@var{pattern}
2335 When performing operations, @command{tar} will skip files that match
2336 @var{pattern}. @xref{exclude}.
2338 @opindex exclude-from, summary
2339 @item --exclude-from=@var{file}
2340 @itemx -X @var{file}
2342 Similar to @option{--exclude}, except @command{tar} will use the list of
2343 patterns in the file @var{file}. @xref{exclude}.
2345 @opindex exclude-caches, summary
2346 @item --exclude-caches
2348 Automatically excludes all directories
2349 containing a cache directory tag. @xref{exclude}.
2351 @opindex file, summary
2352 @item --file=@var{archive}
2353 @itemx -f @var{archive}
2355 @command{tar} will use the file @var{archive} as the @command{tar} archive it
2356 performs operations on, rather than @command{tar}'s compilation dependent
2357 default. @xref{file tutorial}.
2359 @opindex files-from, summary
2360 @item --files-from=@var{file}
2361 @itemx -T @var{file}
2363 @command{tar} will use the contents of @var{file} as a list of archive members
2364 or files to operate on, in addition to those specified on the
2365 command-line. @xref{files}.
2367 @opindex force-local, summary
2370 Forces @command{tar} to interpret the filename given to @option{--file}
2371 as a local file, even if it looks like a remote tape drive name.
2372 @xref{local and remote archives}.
2374 @opindex format, summary
2375 @item --format=@var{format}
2377 Selects output archive format. @var{Format} may be one of the
2382 Creates an archive that is compatible with Unix V7 @command{tar}.
2385 Creates an archive that is compatible with GNU @command{tar} version
2389 Creates archive in GNU tar 1.13 format. Basically it is the same as
2390 @samp{oldgnu} with the only difference in the way it handles long
2394 Creates a @acronym{POSIX.1-1988} compatible archive.
2397 Creates a @acronym{POSIX.1-2001 archive}.
2401 @xref{Formats}, for a detailed discussion of these formats.
2403 @opindex group, summary
2404 @item --group=@var{group}
2406 Files added to the @command{tar} archive will have a group id of @var{group},
2407 rather than the group from the source file. @var{group} is first decoded
2408 as a group symbolic name, but if this interpretation fails, it has to be
2409 a decimal numeric group ID. @FIXME-xref{}
2411 Also see the comments for the @option{--owner=@var{user}} option.
2413 @opindex gzip, summary
2414 @opindex gunzip, summary
2415 @opindex ungzip, summary
2421 This option tells @command{tar} to read or write archives through
2422 @command{gzip}, allowing @command{tar} to directly operate on several
2423 kinds of compressed archives transparently. @xref{gzip}.
2425 @opindex help, summary
2428 @command{tar} will print out a short message summarizing the operations and
2429 options to @command{tar} and exit. @xref{help}.
2431 @opindex ignore-case, summary
2433 Ignore case when matching member or file names with
2434 patterns. @xref{controlling pattern-matching}.
2436 @opindex ignore-command-error, summary
2437 @item --ignore-command-error
2438 Ignore exit codes of subprocesses. @xref{Writing to an External Program}.
2440 @opindex ignore-failed-read, summary
2441 @item --ignore-failed-read
2443 Do not exit unsuccessfully merely because an unreadable file was encountered.
2446 @opindex ignore-zeros, summary
2447 @item --ignore-zeros
2450 With this option, @command{tar} will ignore zeroed blocks in the
2451 archive, which normally signals EOF. @xref{Reading}.
2453 @opindex incremental, summary
2457 Used to inform @command{tar} that it is working with an old
2458 @acronym{GNU}-format incremental backup archive. It is intended
2459 primarily for backwards compatibility only. @xref{Incremental Dumps},
2460 for a detailed discussion of incremental archives.
2462 @opindex index-file, summary
2463 @item --index-file=@var{file}
2465 Send verbose output to @var{file} instead of to standard output.
2467 @opindex info-script, summary
2468 @opindex new-volume-script, summary
2469 @item --info-script=@var{script-file}
2470 @itemx --new-volume-script=@var{script-file}
2471 @itemx -F @var{script-file}
2473 When @command{tar} is performing multi-tape backups, @var{script-file} is run
2474 at the end of each tape. If @var{script-file} exits with nonzero status,
2475 @command{tar} fails immediately. @xref{info-script}, for a detailed
2476 discussion of @var{script-file}.
2478 @opindex interactive, summary
2480 @itemx --confirmation
2483 Specifies that @command{tar} should ask the user for confirmation before
2484 performing potentially destructive options, such as overwriting files.
2487 @opindex keep-newer-files, summary
2488 @item --keep-newer-files
2490 Do not replace existing files that are newer than their archive copies
2491 when extracting files from an archive.
2493 @opindex keep-old-files, summary
2494 @item --keep-old-files
2497 Do not overwrite existing files when extracting files from an archive.
2498 @xref{Keep Old Files}.
2500 @opindex label, summary
2501 @item --label=@var{name}
2502 @itemx -V @var{name}
2504 When creating an archive, instructs @command{tar} to write @var{name}
2505 as a name record in the archive. When extracting or listing archives,
2506 @command{tar} will only operate on archives that have a label matching
2507 the pattern specified in @var{name}. @xref{Tape Files}.
2509 @opindex listed-incremental, summary
2510 @item --listed-incremental=@var{snapshot-file}
2511 @itemx -g @var{snapshot-file}
2513 During a @option{--create} operation, specifies that the archive that
2514 @command{tar} creates is a new @acronym{GNU}-format incremental
2515 backup, using @var{snapshot-file} to determine which files to backup.
2516 With other operations, informs @command{tar} that the archive is in
2517 incremental format. @xref{Incremental Dumps}.
2519 @opindex mode, summary
2520 @item --mode=@var{permissions}
2522 When adding files to an archive, @command{tar} will use
2523 @var{permissions} for the archive members, rather than the permissions
2524 from the files. The program @command{chmod} and this @command{tar}
2525 option share the same syntax for what @var{permissions} might be.
2526 @xref{File permissions, Permissions, File permissions, fileutils,
2527 @acronym{GNU} file utilities}. This reference also has useful
2528 information for those not being overly familiar with the Unix
2531 Of course, @var{permissions} might be plainly specified as an octal number.
2532 However, by using generic symbolic modifications to mode bits, this allows
2533 more flexibility. For example, the value @samp{a+rw} adds read and write
2534 permissions for everybody, while retaining executable bits on directories
2535 or on any other file already marked as executable.
2537 @opindex multi-volume, summary
2538 @item --multi-volume
2541 Informs @command{tar} that it should create or otherwise operate on a
2542 multi-volume @command{tar} archive. @xref{Using Multiple Tapes}.
2544 @opindex new-volume-script, summary
2545 @item --new-volume-script
2549 @opindex seek, summary
2553 Assume that the archive media supports seeks to arbitrary
2554 locations. Usually @command{tar} determines automatically whether
2555 the archive can be seeked or not. This option is intended for use
2556 in cases when such recognition fails.
2558 @opindex newer, summary
2559 @item --newer=@var{date}
2560 @itemx --after-date=@var{date}
2563 When creating an archive, @command{tar} will only add files that have changed
2564 since @var{date}. If @var{date} begins with @samp{/} or @samp{.}, it
2565 is taken to be the name of a file whose data modification time specifies
2566 the date. @xref{after}.
2568 @opindex newer-mtime, summary
2569 @item --newer-mtime=@var{date}
2571 Like @option{--newer}, but add only files whose
2572 contents have changed (as opposed to just @option{--newer}, which will
2573 also back up files for which any status information has changed).
2575 @opindex no-anchored, summary
2577 An exclude pattern can match any subsequence of the name's components.
2578 @xref{controlling pattern-matching}.
2580 @opindex no-delay-directory-restore, summary
2581 @item --no-delay-directory-restore
2583 Setting modification times and permissions of extracted
2584 directories when all files from this directory has been
2585 extracted. This is the default. @xref{Directory Modification Times and Permissions}.
2587 @opindex no-ignore-case, summary
2588 @item --no-ignore-case
2589 Use case-sensitive matching.
2590 @xref{controlling pattern-matching}.
2592 @opindex no-ignore-command-error, summary
2593 @item --no-ignore-command-error
2594 Print warnings about subprocesses terminated with a non-zero exit
2595 code. @xref{Writing to an External Program}.
2597 @opindex no-quote-chars, summary
2598 @item --no-quote-chars=@var{string}
2599 Remove characters listed in @var{string} from the list of quoted
2600 characters set by the previous @option{--quote-chars} option
2601 (@pxref{quoting styles}).
2603 @opindex no-recursion, summary
2604 @item --no-recursion
2606 With this option, @command{tar} will not recurse into directories.
2609 @opindex no-same-owner, summary
2610 @item --no-same-owner
2613 When extracting an archive, do not attempt to preserve the owner
2614 specified in the @command{tar} archive. This the default behavior
2617 @opindex no-same-permissions, summary
2618 @item --no-same-permissions
2620 When extracting an archive, subtract the user's umask from files from
2621 the permissions specified in the archive. This is the default behavior
2624 @opindex no-wildcards, summary
2625 @item --no-wildcards
2626 Do not use wildcards.
2627 @xref{controlling pattern-matching}.
2629 @opindex no-wildcards-match-slash, summary
2630 @item --no-wildcards-match-slash
2631 Wildcards do not match @samp{/}.
2632 @xref{controlling pattern-matching}.
2634 @opindex null, summary
2637 When @command{tar} is using the @option{--files-from} option, this option
2638 instructs @command{tar} to expect filenames terminated with @option{NUL}, so
2639 @command{tar} can correctly work with file names that contain newlines.
2642 @opindex numeric-owner, summary
2643 @item --numeric-owner
2645 This option will notify @command{tar} that it should use numeric user
2646 and group IDs when creating a @command{tar} file, rather than names.
2650 When extracting files, this option is a synonym for
2651 @option{--no-same-owner}, i.e. it prevents @command{tar} from
2652 restoring ownership of files being extracted.
2654 When creating an archive, @option{-o} is a synonym for
2655 @option{--old-archive}. This behavior is for compatibility
2656 with previous versions of @GNUTAR{}, and will be
2657 removed in the future releases.
2659 @xref{Changes}, for more information.
2661 @opindex occurrence, summary
2662 @item --occurrence[=@var{number}]
2664 This option can be used in conjunction with one of the subcommands
2665 @option{--delete}, @option{--diff}, @option{--extract} or
2666 @option{--list} when a list of files is given either on the command
2667 line or via @option{-T} option.
2669 This option instructs @command{tar} to process only the @var{number}th
2670 occurrence of each named file. @var{Number} defaults to 1, so
2673 tar -x -f archive.tar --occurrence filename
2677 will extract the first occurrence of @file{filename} from @file{archive.tar}
2678 and will terminate without scanning to the end of the archive.
2680 @opindex old-archive, summary
2682 Synonym for @option{--format=v7}.
2684 @opindex one-file-system, summary
2685 @item --one-file-system
2686 Used when creating an archive. Prevents @command{tar} from recursing into
2687 directories that are on different file systems from the current
2688 directory @footnote{Earlier versions of @GNUTAR{} understood @option{-l} as a
2689 synonym for @option{--one-file-system}. This has changed in version
2690 1.15.91. @xref{Changes}, for more information.}.
2692 @opindex overwrite, summary
2695 Overwrite existing files and directory metadata when extracting files
2696 from an archive. @xref{Overwrite Old Files}.
2698 @opindex overwrite-dir, summary
2699 @item --overwrite-dir
2701 Overwrite the metadata of existing directories when extracting files
2702 from an archive. @xref{Overwrite Old Files}.
2704 @opindex owner, summary
2705 @item --owner=@var{user}
2707 Specifies that @command{tar} should use @var{user} as the owner of members
2708 when creating archives, instead of the user associated with the source
2709 file. @var{user} is first decoded as a user symbolic name, but if
2710 this interpretation fails, it has to be a decimal numeric user ID.
2713 There is no value indicating a missing number, and @samp{0} usually means
2714 @code{root}. Some people like to force @samp{0} as the value to offer in
2715 their distributions for the owner of files, because the @code{root} user is
2716 anonymous anyway, so that might as well be the owner of anonymous archives.
2718 This option does not affect extraction from archives.
2720 @opindex transform, summary
2721 @item --transform=@var{sed-expr}
2723 Transform file or member names using @command{sed} replacement expression
2724 @var{sed-expr}. For example,
2727 $ @kbd{tar cf archive.tar --transform 's,^\./,usr/,' .}
2731 will add to @file{archive} files from the current working directory,
2732 replacing initial @samp{./} prefix with @samp{usr/}. For the detailed
2733 discussion, @xref{transform}.
2735 To see transformed member names in verbose listings, use
2736 @option{--show-transformed-names} option
2737 (@pxref{show-transformed-names}).
2739 @opindex quote-chars, summary
2740 @item --quote-chars=@var{string}
2741 Always quote characters from @var{string}, even if the selected
2742 quoting style would not quote them (@pxref{quoting styles}).
2744 @opindex quoting-style, summary
2745 @item --quoting-style=@var{style}
2746 Set quoting style to use when printing member and file names
2747 (@pxref{quoting styles}). Valid @var{style} values are:
2748 @code{literal}, @code{shell}, @code{shell-always}, @code{c},
2749 @code{escape}, @code{locale}, and @code{clocale}. Default quoting
2750 style is @code{escape}, unless overridden while configuring the
2753 @opindex pax-option, summary
2754 @item --pax-option=@var{keyword-list}
2755 @FIXME{Such a detailed description does not belong there, move it elsewhere.}
2756 This option is meaningful only with @acronym{POSIX.1-2001} archives
2757 (@pxref{posix}). It modifies the way @command{tar} handles the
2758 extended header keywords. @var{Keyword-list} is a comma-separated
2759 list of keyword options, each keyword option taking one of
2760 the following forms:
2763 @item delete=@var{pattern}
2764 When used with one of archive-creation commands,
2765 this option instructs @command{tar} to omit from extended header records
2766 that it produces any keywords matching the string @var{pattern}.
2768 When used in extract or list mode, this option instructs tar
2769 to ignore any keywords matching the given @var{pattern} in the extended
2770 header records. In both cases, matching is performed using the pattern
2771 matching notation described in @acronym{POSIX 1003.2}, 3.13
2772 (See @cite{glob(7)}). For example:
2775 --pax-option delete=security.*
2778 would suppress security-related information.
2780 @item exthdr.name=@var{string}
2782 This keyword allows user control over the name that is written into the
2783 ustar header blocks for the extended headers. The name is obtained
2784 from @var{string} after making the following substitutions:
2786 @multitable @columnfractions .30 .70
2787 @headitem Meta-character @tab Replaced By
2788 @item %d @tab The directory name of the file, equivalent to the
2789 result of the @command{dirname} utility on the translated pathname.
2790 @item %f @tab The filename of the file, equivalent to the result
2791 of the @command{basename} utility on the translated pathname.
2792 @item %p @tab The process ID of the @command{tar} process.
2793 @item %% @tab A @samp{%} character.
2796 Any other @samp{%} characters in @var{string} produce undefined
2799 If no option @samp{exthdr.name=string} is specified, @command{tar}
2800 will use the following default value:
2806 @item globexthdr.name=@var{string}
2807 This keyword allows user control over the name that is written into
2808 the ustar header blocks for global extended header records. The name
2809 is obtained from the contents of @var{string}, after making
2810 the following substitutions:
2812 @multitable @columnfractions .30 .70
2813 @headitem Meta-character @tab Replaced By
2814 @item %n @tab An integer that represents the
2815 sequence number of the global extended header record in the archive,
2817 @item %p @tab The process ID of the @command{tar} process.
2818 @item %% @tab A @samp{%} character.
2821 Any other @samp{%} characters in @var{string} produce undefined results.
2823 If no option @samp{globexthdr.name=string} is specified, @command{tar}
2824 will use the following default value:
2827 $TMPDIR/GlobalHead.%p.%n
2831 where @samp{$TMPDIR} represents the value of the @var{TMPDIR}
2832 environment variable. If @var{TMPDIR} is not set, @command{tar}
2835 @item @var{keyword}=@var{value}
2836 When used with one of archive-creation commands, these keyword/value pairs
2837 will be included at the beginning of the archive in a global extended
2838 header record. When used with one of archive-reading commands,
2839 @command{tar} will behave as if it has encountered these keyword/value
2840 pairs at the beginning of the archive in a global extended header
2843 @item @var{keyword}:=@var{value}
2844 When used with one of archive-creation commands, these keyword/value pairs
2845 will be included as records at the beginning of an extended header for
2846 each file. This is effectively equivalent to @var{keyword}=@var{value}
2847 form except that it creates no global extended header records.
2849 When used with one of archive-reading commands, @command{tar} will
2850 behave as if these keyword/value pairs were included as records at the
2851 end of each extended header; thus, they will override any global or
2852 file-specific extended header record keywords of the same names.
2853 For example, in the command:
2856 tar --format=posix --create \
2857 --file archive --pax-option gname:=user .
2860 the group name will be forced to a new value for all files
2861 stored in the archive.
2864 @opindex portability, summary
2866 @itemx --old-archive
2867 Synonym for @option{--format=v7}.
2869 @opindex posix, summary
2871 Same as @option{--format=posix}.
2873 @opindex preserve, summary
2876 Synonymous with specifying both @option{--preserve-permissions} and
2877 @option{--same-order}. @xref{Setting Access Permissions}.
2879 @opindex preserve-order, summary
2880 @item --preserve-order
2882 (See @option{--same-order}; @pxref{Reading}.)
2884 @opindex preserve-permissions, summary
2885 @opindex same-permissions, summary
2886 @item --preserve-permissions
2887 @itemx --same-permissions
2890 When @command{tar} is extracting an archive, it normally subtracts the
2891 users' umask from the permissions specified in the archive and uses
2892 that number as the permissions to create the destination file.
2893 Specifying this option instructs @command{tar} that it should use the
2894 permissions directly from the archive. @xref{Setting Access Permissions}.
2896 @opindex read-full-records, summary
2897 @item --read-full-records
2900 Specifies that @command{tar} should reblock its input, for reading
2901 from pipes on systems with buggy implementations. @xref{Reading}.
2903 @opindex record-size, summary
2904 @item --record-size=@var{size}
2906 Instructs @command{tar} to use @var{size} bytes per record when accessing the
2907 archive. @xref{Blocking Factor}.
2909 @opindex recursion, summary
2912 With this option, @command{tar} recurses into directories.
2915 @opindex recursive-unlink, summary
2916 @item --recursive-unlink
2919 directory hierarchies before extracting directories of the same name
2920 from the archive. @xref{Recursive Unlink}.
2922 @opindex remove-files, summary
2923 @item --remove-files
2925 Directs @command{tar} to remove the source file from the file system after
2926 appending it to an archive. @xref{remove files}.
2928 @opindex restrict, summary
2931 Disable use of some potentially harmful @command{tar} options.
2932 Currently this option disables shell invocaton from multi-volume menu
2933 (@pxref{Using Multiple Tapes}).
2935 @opindex rmt-command, summary
2936 @item --rmt-command=@var{cmd}
2938 Notifies @command{tar} that it should use @var{cmd} instead of
2939 the default @file{/usr/libexec/rmt} (@pxref{Remote Tape Server}).
2941 @opindex rsh-command, summary
2942 @item --rsh-command=@var{cmd}
2944 Notifies @command{tar} that is should use @var{cmd} to communicate with remote
2945 devices. @xref{Device}.
2947 @opindex same-order, summary
2949 @itemx --preserve-order
2952 This option is an optimization for @command{tar} when running on machines with
2953 small amounts of memory. It informs @command{tar} that the list of file
2954 arguments has already been sorted to match the order of files in the
2955 archive. @xref{Reading}.
2957 @opindex same-owner, summary
2960 When extracting an archive, @command{tar} will attempt to preserve the owner
2961 specified in the @command{tar} archive with this option present.
2962 This is the default behavior for the superuser; this option has an
2963 effect only for ordinary users. @xref{Attributes}.
2965 @opindex same-permissions, summary
2966 @item --same-permissions
2968 (See @option{--preserve-permissions}; @pxref{Setting Access Permissions}.)
2970 @opindex show-defaults, summary
2971 @item --show-defaults
2973 Displays the default options used by @command{tar} and exits
2974 successfully. This option is intended for use in shell scripts.
2975 Here is an example of what you can see using this option:
2978 $ tar --show-defaults
2979 --format=gnu -f- -b20 --quoting-style=escape \
2980 --rmt-command=/usr/libexec/rmt --rsh-command=/usr/bin/rsh
2983 @opindex show-omitted-dirs, summary
2984 @item --show-omitted-dirs
2986 Instructs @command{tar} to mention directories its skipping over when
2987 operating on a @command{tar} archive. @xref{show-omitted-dirs}.
2989 @opindex show-transformed-names, summary
2990 @opindex show-stored-names, summary
2991 @item --show-transformed-names
2992 @itemx --show-stored-names
2994 Display file or member names after applying any transformations
2995 (@pxref{transform}). In particular, when used in conjunction with one of
2996 archive creation operations it instructs tar to list the member names
2997 stored in the archive, as opposed to the actual file
2998 names. @xref{listing member and file names}.
3000 @opindex sparse, summary
3004 Invokes a @acronym{GNU} extension when adding files to an archive that handles
3005 sparse files efficiently. @xref{sparse}.
3007 @opindex starting-file, summary
3008 @item --starting-file=@var{name}
3009 @itemx -K @var{name}
3011 This option affects extraction only; @command{tar} will skip extracting
3012 files in the archive until it finds one that matches @var{name}.
3015 @opindex strip-components, summary
3016 @item --strip-components=@var{number}
3017 Strip given @var{number} of leading components from file names before
3018 extraction.@footnote{This option was called @option{--strip-path} in
3019 version 1.14.} For example, if archive @file{archive.tar} contained
3020 @file{/some/file/name}, then running
3023 tar --extract --file archive.tar --strip-components=2
3027 would extract this file to file @file{name}.
3029 @opindex suffix, summary
3030 @item --suffix=@var{suffix}
3032 Alters the suffix @command{tar} uses when backing up files from the default
3033 @samp{~}. @xref{backup}.
3035 @opindex tape-length, summary
3036 @item --tape-length=@var{num}
3039 Specifies the length of tapes that @command{tar} is writing as being
3040 @w{@var{num} x 1024} bytes long. @xref{Using Multiple Tapes}.
3042 @opindex test-label, summary
3045 Reads the volume label. If an argument is specified, test whether it
3046 matches the volume label. @xref{--test-label option}.
3048 @opindex to-command, summary
3049 @item --to-command=@var{command}
3051 During extraction @command{tar} will pipe extracted files to the
3052 standard input of @var{command}. @xref{Writing to an External Program}.
3054 @opindex to-stdout, summary
3058 During extraction, @command{tar} will extract files to stdout rather
3059 than to the file system. @xref{Writing to Standard Output}.
3061 @opindex totals, summary
3064 Displays the total number of bytes written after creating an archive.
3067 @opindex touch, summary
3071 Sets the data modification time of extracted files to the extraction time,
3072 rather than the data modification time stored in the archive.
3073 @xref{Data Modification Times}.
3075 @opindex uncompress, summary
3078 (See @option{--compress}. @pxref{gzip})
3080 @opindex ungzip, summary
3083 (See @option{--gzip}. @pxref{gzip})
3085 @opindex unlink-first, summary
3086 @item --unlink-first
3089 Directs @command{tar} to remove the corresponding file from the file
3090 system before extracting it from the archive. @xref{Unlink First}.
3092 @opindex use-compress-program, summary
3093 @item --use-compress-program=@var{prog}
3095 Instructs @command{tar} to access the archive through @var{prog}, which is
3096 presumed to be a compression program of some sort. @xref{gzip}.
3098 @opindex utc, summary
3101 Display file modification dates in @acronym{UTC}. This option implies
3104 @opindex verbose, summary
3108 Specifies that @command{tar} should be more verbose about the operations its
3109 performing. This option can be specified multiple times for some
3110 operations to increase the amount of information displayed.
3113 @opindex verify, summary
3117 Verifies that the archive was correctly written when creating an
3118 archive. @xref{verify}.
3120 @opindex version, summary
3123 Print information about the program's name, version, origin and legal
3124 status, all on standard output, and then exit successfully.
3127 @opindex volno-file, summary
3128 @item --volno-file=@var{file}
3130 Used in conjunction with @option{--multi-volume}. @command{tar} will keep track
3131 of which volume of a multi-volume archive its working in @var{file}.
3134 @opindex wildcards, summary
3136 Use wildcards when matching member names with patterns.
3137 @xref{controlling pattern-matching}.
3139 @opindex wildcards-match-slash, summary
3140 @item --wildcards-match-slash
3141 Wildcards match @samp{/}.
3142 @xref{controlling pattern-matching}.
3145 @node Short Option Summary
3146 @subsection Short Options Cross Reference
3148 Here is an alphabetized list of all of the short option forms, matching
3149 them with the equivalent long option.
3155 @option{--concatenate}
3159 @option{--read-full-records}
3163 @option{--directory}
3167 @option{--info-script}
3171 @option{--incremental}
3175 @option{--starting-file}
3179 @option{--tape-length}
3183 @option{--multi-volume}
3191 @option{--to-stdout}
3195 @option{--absolute-names}
3199 @option{--block-number}
3207 @option{--files-from}
3211 @option{--unlink-first}
3223 @option{--exclude-from}
3231 @option{--blocking-factor}
3247 @option{--listed-incremental}
3251 @option{--dereference}
3255 @option{--ignore-zeros}
3263 @option{--keep-old-files}
3267 @option{--one-file-system}. Use of this short option is deprecated. It
3268 is retained for compatibility with the earlier versions of GNU
3269 @command{tar}, and will be changed in future releases.
3271 @xref{Changes}, for more information.
3279 When creating --- @option{--no-same-owner}, when extracting ---
3280 @option{--portability}.
3282 The later usage is deprecated. It is retained for compatibility with
3283 the earlier versions of @GNUTAR{}. In the future releases
3284 @option{-o} will be equivalent to @option{--no-same-owner} only.
3288 @option{--preserve-permissions}
3296 @option{--same-order}
3312 @option{--interactive}
3325 @section @GNUTAR{} documentation
3327 @cindex Getting program version number
3329 @cindex Version of the @command{tar} program
3330 Being careful, the first thing is really checking that you are using
3331 @GNUTAR{}, indeed. The @option{--version} option
3332 causes @command{tar} to print information about its name, version,
3333 origin and legal status, all on standard output, and then exit
3334 successfully. For example, @w{@samp{tar --version}} might print:
3337 tar (GNU tar) 1.15.2
3338 Copyright (C) 2006 Free Software Foundation, Inc.
3339 This is free software. You may redistribute copies of it under the terms of
3340 the GNU General Public License <http://www.gnu.org/licenses/gpl.html>.
3341 There is NO WARRANTY, to the extent permitted by law.
3343 Written by John Gilmore and Jay Fenlason.
3347 The first occurrence of @samp{tar} in the result above is the program
3348 name in the package (for example, @command{rmt} is another program),
3349 while the second occurrence of @samp{tar} is the name of the package
3350 itself, containing possibly many programs. The package is currently
3351 named @samp{tar}, after the name of the main program it
3352 contains@footnote{There are plans to merge the @command{cpio} and
3353 @command{tar} packages into a single one which would be called
3354 @code{paxutils}. So, who knows if, one of this days, the
3355 @option{--version} would not output @w{@samp{tar (@acronym{GNU}
3358 @cindex Obtaining help
3359 @cindex Listing all @command{tar} options
3360 @opindex help, introduction
3361 Another thing you might want to do is checking the spelling or meaning
3362 of some particular @command{tar} option, without resorting to this
3363 manual, for once you have carefully read it. @GNUTAR{}
3364 has a short help feature, triggerable through the
3365 @option{--help} option. By using this option, @command{tar} will
3366 print a usage message listing all available options on standard
3367 output, then exit successfully, without doing anything else and
3368 ignoring all other options. Even if this is only a brief summary, it
3369 may be several screens long. So, if you are not using some kind of
3370 scrollable window, you might prefer to use something like:
3373 $ @kbd{tar --help | less}
3377 presuming, here, that you like using @command{less} for a pager. Other
3378 popular pagers are @command{more} and @command{pg}. If you know about some
3379 @var{keyword} which interests you and do not want to read all the
3380 @option{--help} output, another common idiom is doing:
3383 tar --help | grep @var{keyword}
3387 for getting only the pertinent lines. Notice, however, that some
3388 @command{tar} options have long description lines and the above
3389 command will list only the first of them.
3391 The exact look of the option summary displayed by @kbd{tar --help} is
3392 configurable. @xref{Configuring Help Summary}, for a detailed description.
3395 If you only wish to check the spelling of an option, running @kbd{tar
3396 --usage} may be a better choice. This will display a terse list of
3397 @command{tar} option without accompanying explanations.
3399 The short help output is quite succinct, and you might have to get
3400 back to the full documentation for precise points. If you are reading
3401 this paragraph, you already have the @command{tar} manual in some
3402 form. This manual is available in a variety of forms from
3403 @url{http://www.gnu.org/software/tar/manual}. It may printed out of the @GNUTAR{}
3404 distribution, provided you have @TeX{} already installed somewhere,
3405 and a laser printer around. Just configure the distribution, execute
3406 the command @w{@samp{make dvi}}, then print @file{doc/tar.dvi} the
3407 usual way (contact your local guru to know how). If @GNUTAR{}
3408 has been conveniently installed at your place, this
3409 manual is also available in interactive, hypertextual form as an Info
3410 file. Just call @w{@samp{info tar}} or, if you do not have the
3411 @command{info} program handy, use the Info reader provided within
3412 @acronym{GNU} Emacs, calling @samp{tar} from the main Info menu.
3414 There is currently no @code{man} page for @GNUTAR{}.
3415 If you observe such a @code{man} page on the system you are running,
3416 either it does not belong to @GNUTAR{}, or it has not
3417 been produced by @acronym{GNU}. Some package maintainers convert
3418 @kbd{tar --help} output to a man page, using @command{help2man}. In
3419 any case, please bear in mind that the authoritative source of
3420 information about @GNUTAR{} is this Texinfo documentation.
3423 @section Obtaining @GNUTAR{} default values
3425 @opindex show-defaults
3426 @GNUTAR{} has some predefined defaults that are used when you do not
3427 explicitely specify another values. To obtain a list of such
3428 defaults, use @option{--show-defaults} option. This will output the
3429 values in the form of @command{tar} command line options:
3433 @kbd{tar --show-defaults}
3434 --format=gnu -f- -b20 --rmt-command=/etc/rmt --rsh-command=/usr/bin/rsh
3439 The above output shows that this version of @GNUTAR{} defaults to
3440 using @samp{gnu} archive format (@pxref{Formats}), it uses standard
3441 output as the archive, if no @option{--file} option has been given
3442 (@pxref{file tutorial}), the default blocking factor is 20
3443 (@pxref{Blocking Factor}). It also shows the default locations where
3444 @command{tar} will look for @command{rmt} and @command{rsh} binaries.
3447 @section Checking @command{tar} progress
3449 Typically, @command{tar} performs most operations without reporting any
3450 information to the user except error messages. When using @command{tar}
3451 with many options, particularly ones with complicated or
3452 difficult-to-predict behavior, it is possible to make serious mistakes.
3453 @command{tar} provides several options that make observing @command{tar}
3454 easier. These options cause @command{tar} to print information as it
3455 progresses in its job, and you might want to use them just for being
3456 more careful about what is going on, or merely for entertaining
3457 yourself. If you have encountered a problem when operating on an
3458 archive, however, you may need more information than just an error
3459 message in order to solve the problem. The following options can be
3460 helpful diagnostic tools.
3462 @cindex Verbose operation
3464 Normally, the @option{--list} (@option{-t}) command to list an archive
3465 prints just the file names (one per line) and the other commands are
3466 silent. When used with most operations, the @option{--verbose}
3467 (@option{-v}) option causes @command{tar} to print the name of each
3468 file or archive member as it is processed. This and the other options
3469 which make @command{tar} print status information can be useful in
3470 monitoring @command{tar}.
3472 With @option{--create} or @option{--extract}, @option{--verbose} used
3473 once just prints the names of the files or members as they are processed.
3474 Using it twice causes @command{tar} to print a longer listing
3475 (reminiscent of @samp{ls -l}) for each member. Since @option{--list}
3476 already prints the names of the members, @option{--verbose} used once
3477 with @option{--list} causes @command{tar} to print an @samp{ls -l}
3478 type listing of the files in the archive. The following examples both
3479 extract members with long list output:
3482 $ @kbd{tar --extract --file=archive.tar --verbose --verbose}
3483 $ @kbd{tar xvvf archive.tar}
3486 Verbose output appears on the standard output except when an archive is
3487 being written to the standard output, as with @samp{tar --create
3488 --file=- --verbose} (@samp{tar cfv -}, or even @samp{tar cv}---if the
3489 installer let standard output be the default archive). In that case
3490 @command{tar} writes verbose output to the standard error stream.
3492 If @option{--index-file=@var{file}} is specified, @command{tar} sends
3493 verbose output to @var{file} rather than to standard output or standard
3496 @cindex Obtaining total status information
3498 The @option{--totals} option---which is only meaningful when used with
3499 @option{--create} (@option{-c})---causes @command{tar} to print the total
3500 amount written to the archive, after it has been fully created.
3502 @cindex Progress information
3504 The @option{--checkpoint} option prints an occasional message
3505 as @command{tar} reads or writes the archive. In fact, it prints
3506 a message each 10 records read or written. It is designed for
3507 those who don't need the more detailed (and voluminous) output of
3508 @option{--block-number} (@option{-R}), but do want visual confirmation
3509 that @command{tar} is actually making forward progress.
3511 @FIXME{There is some confusion here. It seems that -R once wrote a
3512 message at @samp{every} record read or written.}
3514 @opindex show-omitted-dirs
3515 @anchor{show-omitted-dirs}
3516 The @option{--show-omitted-dirs} option, when reading an archive---with
3517 @option{--list} or @option{--extract}, for example---causes a message
3518 to be printed for each directory in the archive which is skipped.
3519 This happens regardless of the reason for skipping: the directory might
3520 not have been named on the command line (implicitly or explicitly),
3521 it might be excluded by the use of the @option{--exclude=@var{pattern}} option, or
3524 @opindex block-number
3525 @cindex Block number where error occurred
3526 @anchor{block-number}
3527 If @option{--block-number} (@option{-R}) is used, @command{tar} prints, along with
3528 every message it would normally produce, the block number within the
3529 archive where the message was triggered. Also, supplementary messages
3530 are triggered when reading blocks full of NULs, or when hitting end of
3531 file on the archive. As of now, if the archive if properly terminated
3532 with a NUL block, the reading of the file may stop before end of file
3533 is met, so the position of end of file will not usually show when
3534 @option{--block-number} (@option{-R}) is used. Note that @GNUTAR{}
3535 drains the archive before exiting when reading the
3536 archive from a pipe.
3538 @cindex Error message, block number of
3539 This option is especially useful when reading damaged archives, since
3540 it helps pinpoint the damaged sections. It can also be used with
3541 @option{--list} (@option{-t}) when listing a file-system backup tape, allowing you to
3542 choose among several backup tapes when retrieving a file later, in
3543 favor of the tape where the file appears earliest (closest to the
3544 front of the tape). @xref{backup}.
3547 @section Asking for Confirmation During Operations
3548 @cindex Interactive operation
3550 Typically, @command{tar} carries out a command without stopping for
3551 further instructions. In some situations however, you may want to
3552 exclude some files and archive members from the operation (for instance
3553 if disk or storage space is tight). You can do this by excluding
3554 certain files automatically (@pxref{Choosing}), or by performing
3555 an operation interactively, using the @option{--interactive} (@option{-w}) option.
3556 @command{tar} also accepts @option{--confirmation} for this option.
3558 @opindex interactive
3559 When the @option{--interactive} (@option{-w}) option is specified, before
3560 reading, writing, or deleting files, @command{tar} first prints a message
3561 for each such file, telling what operation it intends to take, then asks
3562 for confirmation on the terminal. The actions which require
3563 confirmation include adding a file to the archive, extracting a file
3564 from the archive, deleting a file from the archive, and deleting a file
3565 from disk. To confirm the action, you must type a line of input
3566 beginning with @samp{y}. If your input line begins with anything other
3567 than @samp{y}, @command{tar} skips that file.
3569 If @command{tar} is reading the archive from the standard input,
3570 @command{tar} opens the file @file{/dev/tty} to support the interactive
3573 Verbose output is normally sent to standard output, separate from
3574 other error messages. However, if the archive is produced directly
3575 on standard output, then verbose output is mixed with errors on
3576 @code{stderr}. Producing the archive on standard output may be used
3577 as a way to avoid using disk space, when the archive is soon to be
3578 consumed by another process reading it, say. Some people felt the need
3579 of producing an archive on stdout, still willing to segregate between
3580 verbose output and error output. A possible approach would be using a
3581 named pipe to receive the archive, and having the consumer process to
3582 read from that named pipe. This has the advantage of letting standard
3583 output free to receive verbose output, all separate from errors.
3586 @chapter @GNUTAR{} Operations
3599 @section Basic @GNUTAR{} Operations
3601 The basic @command{tar} operations, @option{--create} (@option{-c}),
3602 @option{--list} (@option{-t}) and @option{--extract} (@option{--get},
3603 @option{-x}), are currently presented and described in the tutorial
3604 chapter of this manual. This section provides some complementary notes
3605 for these operations.
3608 @opindex create, complementary notes
3612 Creating an empty archive would have some kind of elegance. One can
3613 initialize an empty archive and later use @option{--append}
3614 (@option{-r}) for adding all members. Some applications would not
3615 welcome making an exception in the way of adding the first archive
3616 member. On the other hand, many people reported that it is
3617 dangerously too easy for @command{tar} to destroy a magnetic tape with
3618 an empty archive@footnote{This is well described in @cite{Unix-haters
3619 Handbook}, by Simson Garfinkel, Daniel Weise & Steven Strassmann, IDG
3620 Books, ISBN 1-56884-203-1.}. The two most common errors are:
3624 Mistakingly using @code{create} instead of @code{extract}, when the
3625 intent was to extract the full contents of an archive. This error
3626 is likely: keys @kbd{c} and @kbd{x} are right next to each other on
3627 the QWERTY keyboard. Instead of being unpacked, the archive then
3628 gets wholly destroyed. When users speak about @dfn{exploding} an
3629 archive, they usually mean something else :-).
3632 Forgetting the argument to @code{file}, when the intent was to create
3633 an archive with a single file in it. This error is likely because a
3634 tired user can easily add the @kbd{f} key to the cluster of option
3635 letters, by the mere force of habit, without realizing the full
3636 consequence of doing so. The usual consequence is that the single
3637 file, which was meant to be saved, is rather destroyed.
3640 So, recognizing the likelihood and the catastrophical nature of these
3641 errors, @GNUTAR{} now takes some distance from elegance, and
3642 cowardly refuses to create an archive when @option{--create} option is
3643 given, there are no arguments besides options, and
3644 @option{--files-from} (@option{-T}) option is @emph{not} used. To get
3645 around the cautiousness of @GNUTAR{} and nevertheless create an
3646 archive with nothing in it, one may still use, as the value for the
3647 @option{--files-from} option, a file with no names in it, as shown in
3648 the following commands:
3651 @kbd{tar --create --file=empty-archive.tar --files-from=/dev/null}
3652 @kbd{tar cfT empty-archive.tar /dev/null}
3655 @opindex extract, complementary notes
3660 A socket is stored, within a @GNUTAR{} archive, as a pipe.
3662 @item @option{--list} (@option{-t})
3664 @GNUTAR{} now shows dates as @samp{1996-08-30},
3665 while it used to show them as @samp{Aug 30 1996}. Preferably,
3666 people should get used to ISO 8601 dates. Local American dates should
3667 be made available again with full date localization support, once
3668 ready. In the meantime, programs not being localizable for dates
3669 should prefer international dates, that's really the way to go.
3671 Look up @url{http://www.ft.uni-erlangen.de/~mskuhn/iso-time.html} if you
3672 are curious, it contains a detailed explanation of the ISO 8601 standard.
3677 @section Advanced @GNUTAR{} Operations
3679 Now that you have learned the basics of using @GNUTAR{}, you may want
3680 to learn about further ways in which @command{tar} can help you.
3682 This chapter presents five, more advanced operations which you probably
3683 won't use on a daily basis, but which serve more specialized functions.
3684 We also explain the different styles of options and why you might want
3685 to use one or another, or a combination of them in your @command{tar}
3686 commands. Additionally, this chapter includes options which allow you to
3687 define the output from @command{tar} more carefully, and provide help and
3688 error correction in special circumstances.
3690 @FIXME{check this after the chapter is actually revised to make sure
3691 it still introduces the info in the chapter correctly : ).}
3703 @subsection The Five Advanced @command{tar} Operations
3706 In the last chapter, you learned about the first three operations to
3707 @command{tar}. This chapter presents the remaining five operations to
3708 @command{tar}: @option{--append}, @option{--update}, @option{--concatenate},
3709 @option{--delete}, and @option{--compare}.
3711 You are not likely to use these operations as frequently as those
3712 covered in the last chapter; however, since they perform specialized
3713 functions, they are quite useful when you do need to use them. We
3714 will give examples using the same directory and files that you created
3715 in the last chapter. As you may recall, the directory is called
3716 @file{practice}, the files are @samp{jazz}, @samp{blues}, @samp{folk},
3717 @samp{rock}, and the two archive files you created are
3718 @samp{collection.tar} and @samp{music.tar}.
3720 We will also use the archive files @samp{afiles.tar} and
3721 @samp{bfiles.tar}. @samp{afiles.tar} contains the members @samp{apple},
3722 @samp{angst}, and @samp{aspic}. @samp{bfiles.tar} contains the members
3723 @samp{./birds}, @samp{baboon}, and @samp{./box}.
3725 Unless we state otherwise, all practicing you do and examples you follow
3726 in this chapter will take place in the @file{practice} directory that
3727 you created in the previous chapter; see @ref{prepare for examples}.
3728 (Below in this section, we will remind you of the state of the examples
3729 where the last chapter left them.)
3731 The five operations that we will cover in this chapter are:
3736 Add new entries to an archive that already exists.
3739 Add more recent copies of archive members to the end of an archive, if
3744 Add one or more pre-existing archives to the end of another archive.
3746 Delete items from an archive (does not work on tapes).
3750 Compare archive members to their counterparts in the file system.
3754 @subsection How to Add Files to Existing Archives: @option{--append}
3758 If you want to add files to an existing archive, you don't need to
3759 create a new archive; you can use @option{--append} (@option{-r}).
3760 The archive must already exist in order to use @option{--append}. (A
3761 related operation is the @option{--update} operation; you can use this
3762 to add newer versions of archive members to an existing archive. To learn how to
3763 do this with @option{--update}, @pxref{update}.)
3765 If you use @option{--append} to add a file that has the same name as an
3766 archive member to an archive containing that archive member, then the
3767 old member is not deleted. What does happen, however, is somewhat
3768 complex. @command{tar} @emph{allows} you to have infinite number of files
3769 with the same name. Some operations treat these same-named members no
3770 differently than any other set of archive members: for example, if you
3771 view an archive with @option{--list} (@option{-t}), you will see all
3772 of those members listed, with their data modification times, owners, etc.
3774 Other operations don't deal with these members as perfectly as you might
3775 prefer; if you were to use @option{--extract} to extract the archive,
3776 only the most recently added copy of a member with the same name as four
3777 other members would end up in the working directory. This is because
3778 @option{--extract} extracts an archive in the order the members appeared
3779 in the archive; the most recently archived members will be extracted
3780 last. Additionally, an extracted member will @emph{replace} a file of
3781 the same name which existed in the directory already, and @command{tar}
3782 will not prompt you about this@footnote{Unless you give it
3783 @option{--keep-old-files} option, or the disk copy is newer than the
3784 the one in the archive and you invoke @command{tar} with
3785 @option{--keep-newer-files} option}. Thus, only the most recently archived
3786 member will end up being extracted, as it will replace the one
3787 extracted before it, and so on.
3789 There exists a special option that allows you to get around this
3790 behavior and extract (or list) only a particular copy of the file.
3791 This is @option{--occurrence} option. If you run @command{tar} with
3792 this option, it will extract only the first copy of the file. You
3793 may also give this option an argument specifying the number of
3794 copy to be extracted. Thus, for example if the archive
3795 @file{archive.tar} contained three copies of file @file{myfile}, then
3799 tar --extract --file archive.tar --occurrence=2 myfile
3803 would extract only the second copy. @xref{Option
3804 Summary,---occurrence}, for the description of @option{--occurrence}
3807 @FIXME{ hag -- you might want to incorporate some of the above into the
3808 MMwtSN node; not sure. i didn't know how to make it simpler...
3810 There are a few ways to get around this. (maybe xref Multiple Members
3811 with the Same Name.}
3813 @cindex Members, replacing with other members
3814 @cindex Replacing members with other members
3815 If you want to replace an archive member, use @option{--delete} to
3816 delete the member you want to remove from the archive, , and then use
3817 @option{--append} to add the member you want to be in the archive. Note
3818 that you can not change the order of the archive; the most recently
3819 added member will still appear last. In this sense, you cannot truly
3820 ``replace'' one member with another. (Replacing one member with another
3821 will not work on certain types of media, such as tapes; see @ref{delete}
3822 and @ref{Media}, for more information.)
3825 * appending files:: Appending Files to an Archive
3829 @node appending files
3830 @subsubsection Appending Files to an Archive
3832 @cindex Adding files to an Archive
3833 @cindex Appending files to an Archive
3834 @cindex Archives, Appending files to
3836 The simplest way to add a file to an already existing archive is the
3837 @option{--append} (@option{-r}) operation, which writes specified
3838 files into the archive whether or not they are already among the
3841 When you use @option{--append}, you @emph{must} specify file name
3842 arguments, as there is no default. If you specify a file that already
3843 exists in the archive, another copy of the file will be added to the
3844 end of the archive. As with other operations, the member names of the
3845 newly added files will be exactly the same as their names given on the
3846 command line. The @option{--verbose} (@option{-v}) option will print
3847 out the names of the files as they are written into the archive.
3849 @option{--append} cannot be performed on some tape drives, unfortunately,
3850 due to deficiencies in the formats those tape drives use. The archive
3851 must be a valid @command{tar} archive, or else the results of using this
3852 operation will be unpredictable. @xref{Media}.
3854 To demonstrate using @option{--append} to add a file to an archive,
3855 create a file called @file{rock} in the @file{practice} directory.
3856 Make sure you are in the @file{practice} directory. Then, run the
3857 following @command{tar} command to add @file{rock} to
3858 @file{collection.tar}:
3861 $ @kbd{tar --append --file=collection.tar rock}
3865 If you now use the @option{--list} (@option{-t}) operation, you will see that
3866 @file{rock} has been added to the archive:
3869 $ @kbd{tar --list --file=collection.tar}
3870 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
3871 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3872 -rw-r--r-- me user 20 1996-09-23 16:44 folk
3873 -rw-r--r-- me user 20 1996-09-23 16:44 rock
3877 @subsubsection Multiple Files with the Same Name
3879 You can use @option{--append} (@option{-r}) to add copies of files
3880 which have been updated since the archive was created. (However, we
3881 do not recommend doing this since there is another @command{tar}
3882 option called @option{--update}; @xref{update}, for more information.
3883 We describe this use of @option{--append} here for the sake of
3884 completeness.) When you extract the archive, the older version will
3885 be effectively lost. This works because files are extracted from an
3886 archive in the order in which they were archived. Thus, when the
3887 archive is extracted, a file archived later in time will replace a
3888 file of the same name which was archived earlier, even though the
3889 older version of the file will remain in the archive unless you delete
3890 all versions of the file.
3892 Supposing you change the file @file{blues} and then append the changed
3893 version to @file{collection.tar}. As you saw above, the original
3894 @file{blues} is in the archive @file{collection.tar}. If you change the
3895 file and append the new version of the file to the archive, there will
3896 be two copies in the archive. When you extract the archive, the older
3897 version of the file will be extracted first, and then replaced by the
3898 newer version when it is extracted.
3900 You can append the new, changed copy of the file @file{blues} to the
3901 archive in this way:
3904 $ @kbd{tar --append --verbose --file=collection.tar blues}
3909 Because you specified the @option{--verbose} option, @command{tar} has
3910 printed the name of the file being appended as it was acted on. Now
3911 list the contents of the archive:
3914 $ @kbd{tar --list --verbose --file=collection.tar}
3915 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
3916 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3917 -rw-r--r-- me user 20 1996-09-23 16:44 folk
3918 -rw-r--r-- me user 20 1996-09-23 16:44 rock
3919 -rw-r--r-- me user 58 1996-10-24 18:30 blues
3923 The newest version of @file{blues} is now at the end of the archive
3924 (note the different creation dates and file sizes). If you extract
3925 the archive, the older version of the file @file{blues} will be
3926 replaced by the newer version. You can confirm this by extracting
3927 the archive and running @samp{ls} on the directory.
3929 If you wish to extract the first occurrence of the file @file{blues}
3930 from the archive, use @option{--occurrence} option, as shown in
3931 the following example:
3934 $ @kbd{tar --extract -vv --occurrence --file=collection.tar blues}
3935 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3938 @xref{Writing}, for more information on @option{--extract} and
3939 @xref{Option Summary, --occurrence}, for the description of
3940 @option{--occurrence} option.
3943 @subsection Updating an Archive
3945 @cindex Updating an archive
3948 In the previous section, you learned how to use @option{--append} to
3949 add a file to an existing archive. A related operation is
3950 @option{--update} (@option{-u}). The @option{--update} operation
3951 updates a @command{tar} archive by comparing the date of the specified
3952 archive members against the date of the file with the same name. If
3953 the file has been modified more recently than the archive member, then
3954 the newer version of the file is added to the archive (as with
3957 Unfortunately, you cannot use @option{--update} with magnetic tape drives.
3958 The operation will fail.
3960 @FIXME{other examples of media on which --update will fail? need to ask
3961 charles and/or mib/thomas/dave shevett..}
3963 Both @option{--update} and @option{--append} work by adding to the end
3964 of the archive. When you extract a file from the archive, only the
3965 version stored last will wind up in the file system, unless you use
3966 the @option{--backup} option. @FIXME-ref{Multiple Members with the
3974 @subsubsection How to Update an Archive Using @option{--update}
3976 You must use file name arguments with the @option{--update}
3977 (@option{-u}) operation. If you don't specify any files,
3978 @command{tar} won't act on any files and won't tell you that it didn't
3979 do anything (which may end up confusing you).
3981 @c note: the above parenthetical added because in fact, this
3982 @c behavior just confused the author. :-)
3984 To see the @option{--update} option at work, create a new file,
3985 @file{classical}, in your practice directory, and some extra text to the
3986 file @file{blues}, using any text editor. Then invoke @command{tar} with
3987 the @samp{update} operation and the @option{--verbose} (@option{-v})
3988 option specified, using the names of all the files in the practice
3989 directory as file name arguments:
3992 $ @kbd{tar --update -v -f collection.tar blues folk rock classical}
3999 Because we have specified verbose mode, @command{tar} prints out the names
4000 of the files it is working on, which in this case are the names of the
4001 files that needed to be updated. If you run @samp{tar --list} and look
4002 at the archive, you will see @file{blues} and @file{classical} at its
4003 end. There will be a total of two versions of the member @samp{blues};
4004 the one at the end will be newer and larger, since you added text before
4007 (The reason @command{tar} does not overwrite the older file when updating
4008 it is because writing to the middle of a section of tape is a difficult
4009 process. Tapes are not designed to go backward. @xref{Media}, for more
4010 information about tapes.
4012 @option{--update} (@option{-u}) is not suitable for performing backups for two
4013 reasons: it does not change directory content entries, and it
4014 lengthens the archive every time it is used. The @GNUTAR{}
4015 options intended specifically for backups are more
4016 efficient. If you need to run backups, please consult @ref{Backups}.
4019 @subsection Combining Archives with @option{--concatenate}
4021 @cindex Adding archives to an archive
4022 @cindex Concatenating Archives
4023 @opindex concatenate
4025 @c @cindex @option{-A} described
4026 Sometimes it may be convenient to add a second archive onto the end of
4027 an archive rather than adding individual files to the archive. To add
4028 one or more archives to the end of another archive, you should use the
4029 @option{--concatenate} (@option{--catenate}, @option{-A}) operation.
4031 To use @option{--concatenate}, give the first archive with
4032 @option{--file} option and name the rest of archives to be
4033 concatenated on the command line. The members, and their member
4034 names, will be copied verbatim from those archives to the first one.
4035 @FIXME-ref{This can cause multiple members to have the same name, for
4036 information on how this affects reading the archive, Multiple
4037 Members with the Same Name.}
4038 The new, concatenated archive will be called by the same name as the
4039 one given with the @option{--file} option. As usual, if you omit
4040 @option{--file}, @command{tar} will use the value of the environment
4041 variable @env{TAPE}, or, if this has not been set, the default archive name.
4043 @FIXME{There is no way to specify a new name...}
4045 To demonstrate how @option{--concatenate} works, create two small archives
4046 called @file{bluesrock.tar} and @file{folkjazz.tar}, using the relevant
4047 files from @file{practice}:
4050 $ @kbd{tar -cvf bluesrock.tar blues rock}
4053 $ @kbd{tar -cvf folkjazz.tar folk jazz}
4059 If you like, You can run @samp{tar --list} to make sure the archives
4060 contain what they are supposed to:
4063 $ @kbd{tar -tvf bluesrock.tar}
4064 -rw-r--r-- melissa user 105 1997-01-21 19:42 blues
4065 -rw-r--r-- melissa user 33 1997-01-20 15:34 rock
4066 $ @kbd{tar -tvf jazzfolk.tar}
4067 -rw-r--r-- melissa user 20 1996-09-23 16:44 folk
4068 -rw-r--r-- melissa user 65 1997-01-30 14:15 jazz
4071 We can concatenate these two archives with @command{tar}:
4075 $ @kbd{tar --concatenate --file=bluesrock.tar jazzfolk.tar}
4078 If you now list the contents of the @file{bluesrock.tar}, you will see
4079 that now it also contains the archive members of @file{jazzfolk.tar}:
4082 $ @kbd{tar --list --file=bluesrock.tar}
4089 When you use @option{--concatenate}, the source and target archives must
4090 already exist and must have been created using compatible format
4091 parameters. Notice, that @command{tar} does not check whether the
4092 archives it concatenates have compatible formats, it does not
4093 even check if the files are really tar archives.
4095 Like @option{--append} (@option{-r}), this operation cannot be performed on some
4096 tape drives, due to deficiencies in the formats those tape drives use.
4098 @cindex @code{concatenate} vs @command{cat}
4099 @cindex @command{cat} vs @code{concatenate}
4100 It may seem more intuitive to you to want or try to use @command{cat} to
4101 concatenate two archives instead of using the @option{--concatenate}
4102 operation; after all, @command{cat} is the utility for combining files.
4104 However, @command{tar} archives incorporate an end-of-file marker which
4105 must be removed if the concatenated archives are to be read properly as
4106 one archive. @option{--concatenate} removes the end-of-archive marker
4107 from the target archive before each new archive is appended. If you use
4108 @command{cat} to combine the archives, the result will not be a valid
4109 @command{tar} format archive. If you need to retrieve files from an
4110 archive that was added to using the @command{cat} utility, use the
4111 @option{--ignore-zeros} (@option{-i}) option. @xref{Ignore Zeros}, for further
4112 information on dealing with archives improperly combined using the
4113 @command{cat} shell utility.
4116 @subsection Removing Archive Members Using @option{--delete}
4118 @cindex Deleting files from an archive
4119 @cindex Removing files from an archive
4122 You can remove members from an archive by using the @option{--delete}
4123 option. Specify the name of the archive with @option{--file}
4124 (@option{-f}) and then specify the names of the members to be deleted;
4125 if you list no member names, nothing will be deleted. The
4126 @option{--verbose} option will cause @command{tar} to print the names
4127 of the members as they are deleted. As with @option{--extract}, you
4128 must give the exact member names when using @samp{tar --delete}.
4129 @option{--delete} will remove all versions of the named file from the
4130 archive. The @option{--delete} operation can run very slowly.
4132 Unlike other operations, @option{--delete} has no short form.
4134 @cindex Tapes, using @option{--delete} and
4135 @cindex Deleting from tape archives
4136 This operation will rewrite the archive. You can only use
4137 @option{--delete} on an archive if the archive device allows you to
4138 write to any point on the media, such as a disk; because of this, it
4139 does not work on magnetic tapes. Do not try to delete an archive member
4140 from a magnetic tape; the action will not succeed, and you will be
4141 likely to scramble the archive and damage your tape. There is no safe
4142 way (except by completely re-writing the archive) to delete files from
4143 most kinds of magnetic tape. @xref{Media}.
4145 To delete all versions of the file @file{blues} from the archive
4146 @file{collection.tar} in the @file{practice} directory, make sure you
4147 are in that directory, and then,
4150 $ @kbd{tar --list --file=collection.tar}
4155 $ @kbd{tar --delete --file=collection.tar blues}
4156 $ @kbd{tar --list --file=collection.tar}
4163 @FIXME{Check if the above listing is actually produced after running
4164 all the examples on collection.tar.}
4166 The @option{--delete} option has been reported to work properly when
4167 @command{tar} acts as a filter from @code{stdin} to @code{stdout}.
4170 @subsection Comparing Archive Members with the File System
4171 @cindex Verifying the currency of an archive
4175 The @option{--compare} (@option{-d}), or @option{--diff} operation compares
4176 specified archive members against files with the same names, and then
4177 reports differences in file size, mode, owner, modification date and
4178 contents. You should @emph{only} specify archive member names, not file
4179 names. If you do not name any members, then @command{tar} will compare the
4180 entire archive. If a file is represented in the archive but does not
4181 exist in the file system, @command{tar} reports a difference.
4183 You have to specify the record size of the archive when modifying an
4184 archive with a non-default record size.
4186 @command{tar} ignores files in the file system that do not have
4187 corresponding members in the archive.
4189 The following example compares the archive members @file{rock},
4190 @file{blues} and @file{funk} in the archive @file{bluesrock.tar} with
4191 files of the same name in the file system. (Note that there is no file,
4192 @file{funk}; @command{tar} will report an error message.)
4195 $ @kbd{tar --compare --file=bluesrock.tar rock blues funk}
4198 tar: funk not found in archive
4201 The spirit behind the @option{--compare} (@option{--diff},
4202 @option{-d}) option is to check whether the archive represents the
4203 current state of files on disk, more than validating the integrity of
4204 the archive media. For this later goal, @xref{verify}.
4206 @node create options
4207 @section Options Used by @option{--create}
4209 @opindex create, additional options
4210 The previous chapter described the basics of how to use
4211 @option{--create} (@option{-c}) to create an archive from a set of files.
4212 @xref{create}. This section described advanced options to be used with
4216 * Ignore Failed Read::
4219 @node Ignore Failed Read
4220 @subsection Ignore Fail Read
4223 @item --ignore-failed-read
4224 Do not exit with nonzero on unreadable files or directories.
4227 @node extract options
4228 @section Options Used by @option{--extract}
4231 @opindex extract, additional options
4232 The previous chapter showed how to use @option{--extract} to extract
4233 an archive into the file system. Various options cause @command{tar} to
4234 extract more information than just file contents, such as the owner,
4235 the permissions, the modification date, and so forth. This section
4236 presents options to be used with @option{--extract} when certain special
4237 considerations arise. You may review the information presented in
4238 @ref{extract} for more basic information about the
4239 @option{--extract} operation.
4242 * Reading:: Options to Help Read Archives
4243 * Writing:: Changing How @command{tar} Writes Files
4244 * Scarce:: Coping with Scarce Resources
4248 @subsection Options to Help Read Archives
4249 @cindex Options when reading archives
4252 @cindex Reading incomplete records
4253 @cindex Records, incomplete
4254 @opindex read-full-records
4255 Normally, @command{tar} will request data in full record increments from
4256 an archive storage device. If the device cannot return a full record,
4257 @command{tar} will report an error. However, some devices do not always
4258 return full records, or do not require the last record of an archive to
4259 be padded out to the next record boundary. To keep reading until you
4260 obtain a full record, or to accept an incomplete record if it contains
4261 an end-of-archive marker, specify the @option{--read-full-records} (@option{-B}) option
4262 in conjunction with the @option{--extract} or @option{--list} operations.
4265 The @option{--read-full-records} (@option{-B}) option is turned on by default when
4266 @command{tar} reads an archive from standard input, or from a remote
4267 machine. This is because on BSD Unix systems, attempting to read a
4268 pipe returns however much happens to be in the pipe, even if it is
4269 less than was requested. If this option were not enabled, @command{tar}
4270 would fail as soon as it read an incomplete record from the pipe.
4272 If you're not sure of the blocking factor of an archive, you can
4273 read the archive by specifying @option{--read-full-records} (@option{-B}) and
4274 @option{--blocking-factor=@var{512-size}} (@option{-b
4275 @var{512-size}}), using a blocking factor larger than what the archive
4276 uses. This lets you avoid having to determine the blocking factor
4277 of an archive. @xref{Blocking Factor}.
4280 * read full records::
4284 @node read full records
4285 @unnumberedsubsubsec Reading Full Records
4287 @FIXME{need sentence or so of intro here}
4290 @opindex read-full-records
4291 @item --read-full-records
4293 Use in conjunction with @option{--extract} (@option{--get},
4294 @option{-x}) to read an archive which contains incomplete records, or
4295 one which has a blocking factor less than the one specified.
4299 @unnumberedsubsubsec Ignoring Blocks of Zeros
4301 @cindex End-of-archive blocks, ignoring
4302 @cindex Ignoring end-of-archive blocks
4303 @opindex ignore-zeros
4304 Normally, @command{tar} stops reading when it encounters a block of zeros
4305 between file entries (which usually indicates the end of the archive).
4306 @option{--ignore-zeros} (@option{-i}) allows @command{tar} to
4307 completely read an archive which contains a block of zeros before the
4308 end (i.e., a damaged archive, or one that was created by concatenating
4309 several archives together).
4311 The @option{--ignore-zeros} (@option{-i}) option is turned off by default because many
4312 versions of @command{tar} write garbage after the end-of-archive entry,
4313 since that part of the media is never supposed to be read. @GNUTAR{}
4314 does not write after the end of an archive, but seeks to
4315 maintain compatiblity among archiving utilities.
4318 @item --ignore-zeros
4320 To ignore blocks of zeros (i.e., end-of-archive entries) which may be
4321 encountered while reading an archive. Use in conjunction with
4322 @option{--extract} or @option{--list}.
4326 @subsection Changing How @command{tar} Writes Files
4329 @FIXME{Introductory paragraph}
4332 * Dealing with Old Files::
4333 * Overwrite Old Files::
4335 * Keep Newer Files::
4337 * Recursive Unlink::
4338 * Data Modification Times::
4339 * Setting Access Permissions::
4340 * Directory Modification Times and Permissions::
4341 * Writing to Standard Output::
4342 * Writing to an External Program::
4346 @node Dealing with Old Files
4347 @unnumberedsubsubsec Options Controlling the Overwriting of Existing Files
4349 @opindex overwrite-dir, introduced
4350 When extracting files, if @command{tar} discovers that the extracted
4351 file already exists, it normally replaces the file by removing it before
4352 extracting it, to prevent confusion in the presence of hard or symbolic
4353 links. (If the existing file is a symbolic link, it is removed, not
4354 followed.) However, if a directory cannot be removed because it is
4355 nonempty, @command{tar} normally overwrites its metadata (ownership,
4356 permission, etc.). The @option{--overwrite-dir} option enables this
4357 default behavior. To be more cautious and preserve the metadata of
4358 such a directory, use the @option{--no-overwrite-dir} option.
4360 @cindex Overwriting old files, prevention
4361 @opindex keep-old-files, introduced
4362 To be even more cautious and prevent existing files from being replaced, use
4363 the @option{--keep-old-files} (@option{-k}) option. It causes @command{tar} to refuse
4364 to replace or update a file that already exists, i.e., a file with the
4365 same name as an archive member prevents extraction of that archive
4366 member. Instead, it reports an error.
4368 @opindex overwrite, introduced
4369 To be more aggressive about altering existing files, use the
4370 @option{--overwrite} option. It causes @command{tar} to overwrite
4371 existing files and to follow existing symbolic links when extracting.
4373 @cindex Protecting old files
4374 Some people argue that @GNUTAR{} should not hesitate
4375 to overwrite files with other files when extracting. When extracting
4376 a @command{tar} archive, they expect to see a faithful copy of the
4377 state of the file system when the archive was created. It is debatable
4378 that this would always be a proper behavior. For example, suppose one
4379 has an archive in which @file{usr/local} is a link to
4380 @file{usr/local2}. Since then, maybe the site removed the link and
4381 renamed the whole hierarchy from @file{/usr/local2} to
4382 @file{/usr/local}. Such things happen all the time. I guess it would
4383 not be welcome at all that @GNUTAR{} removes the
4384 whole hierarchy just to make room for the link to be reinstated
4385 (unless it @emph{also} simultaneously restores the full
4386 @file{/usr/local2}, of course!) @GNUTAR{} is indeed
4387 able to remove a whole hierarchy to reestablish a symbolic link, for
4388 example, but @emph{only if} @option{--recursive-unlink} is specified
4389 to allow this behavior. In any case, single files are silently
4392 @opindex unlink-first, introduced
4393 Finally, the @option{--unlink-first} (@option{-U}) option can improve performance in
4394 some cases by causing @command{tar} to remove files unconditionally
4395 before extracting them.
4397 @node Overwrite Old Files
4398 @unnumberedsubsubsec Overwrite Old Files
4403 Overwrite existing files and directory metadata when extracting files
4406 This causes @command{tar} to write extracted files into the file system without
4407 regard to the files already on the system; i.e., files with the same
4408 names as archive members are overwritten when the archive is extracted.
4409 It also causes @command{tar} to extract the ownership, permissions,
4410 and time stamps onto any preexisting files or directories.
4411 If the name of a corresponding file name is a symbolic link, the file
4412 pointed to by the symbolic link will be overwritten instead of the
4413 symbolic link itself (if this is possible). Moreover, special devices,
4414 empty directories and even symbolic links are automatically removed if
4415 they are in the way of extraction.
4417 Be careful when using the @option{--overwrite} option, particularly when
4418 combined with the @option{--absolute-names} (@option{-P}) option, as this combination
4419 can change the contents, ownership or permissions of any file on your
4420 system. Also, many systems do not take kindly to overwriting files that
4421 are currently being executed.
4423 @opindex overwrite-dir
4424 @item --overwrite-dir
4425 Overwrite the metadata of directories when extracting files from an
4426 archive, but remove other files before extracting.
4429 @node Keep Old Files
4430 @unnumberedsubsubsec Keep Old Files
4433 @opindex keep-old-files
4434 @item --keep-old-files
4436 Do not replace existing files from archive. The
4437 @option{--keep-old-files} (@option{-k}) option prevents @command{tar}
4438 from replacing existing files with files with the same name from the
4439 archive. The @option{--keep-old-files} option is meaningless with
4440 @option{--list} (@option{-t}). Prevents @command{tar} from replacing
4441 files in the file system during extraction.
4444 @node Keep Newer Files
4445 @unnumberedsubsubsec Keep Newer Files
4448 @opindex keep-newer-files
4449 @item --keep-newer-files
4450 Do not replace existing files that are newer than their archive
4451 copies. This option is meaningless with @option{--list} (@option{-t}).
4455 @unnumberedsubsubsec Unlink First
4458 @opindex unlink-first
4459 @item --unlink-first
4461 Remove files before extracting over them.
4462 This can make @command{tar} run a bit faster if you know in advance
4463 that the extracted files all need to be removed. Normally this option
4464 slows @command{tar} down slightly, so it is disabled by default.
4467 @node Recursive Unlink
4468 @unnumberedsubsubsec Recursive Unlink
4471 @opindex recursive-unlink
4472 @item --recursive-unlink
4473 When this option is specified, try removing files and directory hierarchies
4474 before extracting over them. @emph{This is a dangerous option!}
4477 If you specify the @option{--recursive-unlink} option,
4478 @command{tar} removes @emph{anything} that keeps you from extracting a file
4479 as far as current permissions will allow it. This could include removal
4480 of the contents of a full directory hierarchy.
4482 @node Data Modification Times
4483 @unnumberedsubsubsec Setting Data Modification Times
4485 @cindex Data modification times of extracted files
4486 @cindex Modification times of extracted files
4487 Normally, @command{tar} sets the data modification times of extracted
4488 files to the corresponding times recorded for the files in the archive, but
4489 limits the permissions of extracted files by the current @code{umask}
4492 To set the data modification times of extracted files to the time when
4493 the files were extracted, use the @option{--touch} (@option{-m}) option in
4494 conjunction with @option{--extract} (@option{--get}, @option{-x}).
4500 Sets the data modification time of extracted archive members to the time
4501 they were extracted, not the time recorded for them in the archive.
4502 Use in conjunction with @option{--extract} (@option{--get}, @option{-x}).
4505 @node Setting Access Permissions
4506 @unnumberedsubsubsec Setting Access Permissions
4508 @cindex Permissions of extracted files
4509 @cindex Modes of extracted files
4510 To set the modes (access permissions) of extracted files to those
4511 recorded for those files in the archive, use @option{--same-permissions}
4512 in conjunction with the @option{--extract} (@option{--get},
4513 @option{-x}) operation.
4516 @opindex preserve-permission
4517 @opindex same-permission
4518 @item --preserve-permission
4519 @itemx --same-permission
4520 @c @itemx --ignore-umask
4522 Set modes of extracted archive members to those recorded in the
4523 archive, instead of current umask settings. Use in conjunction with
4524 @option{--extract} (@option{--get}, @option{-x}).
4527 @node Directory Modification Times and Permissions
4528 @unnumberedsubsubsec Directory Modification Times and Permissions
4530 After sucessfully extracting a file member, @GNUTAR{} normally
4531 restores its permissions and modification times, as described in the
4532 previous sections. This cannot be done for directories, because
4533 after extracting a directory @command{tar} will almost certainly
4534 extract files into that directory and this will cause the directory
4535 modification time to be updated. Moreover, restoring that directory
4536 permissions may not permit file creation within it. Thus, restoring
4537 directory permissions and modification times must be delayed at least
4538 until all files have been extracted into that directory. @GNUTAR{}
4539 restores directories using the following approach.
4541 The extracted directories are created with the mode specified in the
4542 archive, as modified by the umask of the user, which gives sufficient
4543 permissions to allow file creation. The meta-information about the
4544 directory is recorded in the temporary list of directories. When
4545 preparing to extract next archive member, @GNUTAR{} checks if the
4546 directory prefix of this file contains the remembered directory. If
4547 it does not, the program assumes that all files have been extracted
4548 into that directory, restores its modification time and permissions
4549 and removes its entry from the internal list. This approach allows
4550 to correctly restore directory meta-information in the majority of
4551 cases, while keeping memory requirements sufficiently small. It is
4552 based on the fact, that most @command{tar} archives use the predefined
4553 order of members: first the directory, then all the files and
4554 subdirectories in that directory.
4556 However, this is not always true. The most important exception are
4557 incremental archives (@pxref{Incremental Dumps}). The member order in
4558 an incremental archive is reversed: first all directory members are
4559 stored, followed by other (non-directory) members. So, when extracting
4560 from incremental archives, @GNUTAR{} alters the above procedure. It
4561 remebers all restored directories, and restores their meta-data
4562 only after the entire archive has been processed. Notice, that you do
4563 not need to specity any special options for that, as @GNUTAR{}
4564 automatically detects archives in incremental format.
4566 There may be cases, when such processing is required for normal archives
4567 too. Consider the following example:
4571 $ @kbd{tar --no-recursion -cvf archive \
4572 foo foo/file1 bar bar/file foo/file2}
4581 During the normal operation, after encountering @file{bar}
4582 @GNUTAR{} will assume that all files from the directory @file{foo}
4583 were already extracted and will therefore restore its timestamp and
4584 permission bits. However, after extracting @file{foo/file2} the
4585 directory timestamp will be offset again.
4587 To correctly restore directory meta-information in such cases, use
4588 @option{delay-directory-restore} command line option:
4591 @opindex delay-directory-restore
4592 @item --delay-directory-restore
4593 Delays restoring of the modification times and permissions of extracted
4594 directories until the end of extraction. This way, correct
4595 meta-information is restored even if the archive has unusual member
4598 @opindex no-delay-directory-restore
4599 @item --no-delay-directory-restore
4600 Cancel the effect of the previous @option{--delay-directory-restore}.
4601 Use this option if you have used @option{--delay-directory-restore} in
4602 @env{TAR_OPTIONS} variable (@pxref{TAR_OPTIONS}) and wish to
4603 temporarily disable it.
4606 @node Writing to Standard Output
4607 @unnumberedsubsubsec Writing to Standard Output
4609 @cindex Writing extracted files to standard output
4610 @cindex Standard output, writing extracted files to
4611 To write the extracted files to the standard output, instead of
4612 creating the files on the file system, use @option{--to-stdout} (@option{-O}) in
4613 conjunction with @option{--extract} (@option{--get}, @option{-x}). This option is useful if you are
4614 extracting files to send them through a pipe, and do not need to
4615 preserve them in the file system. If you extract multiple members,
4616 they appear on standard output concatenated, in the order they are
4617 found in the archive.
4623 Writes files to the standard output. Use only in conjunction with
4624 @option{--extract} (@option{--get}, @option{-x}). When this option is
4625 used, instead of creating the files specified, @command{tar} writes
4626 the contents of the files extracted to its standard output. This may
4627 be useful if you are only extracting the files in order to send them
4628 through a pipe. This option is meaningless with @option{--list}
4632 This can be useful, for example, if you have a tar archive containing
4633 a big file and don't want to store the file on disk before processing
4634 it. You can use a command like this:
4637 tar -xOzf foo.tgz bigfile | process
4640 or even like this if you want to process the concatenation of the files:
4643 tar -xOzf foo.tgz bigfile1 bigfile2 | process
4646 Hovewer, @option{--to-command} may be more convenient for use with
4647 multiple files. See the next section.
4649 @node Writing to an External Program
4650 @unnumberedsubsubsec Writing to an External Program
4652 You can instruct @command{tar} to send the contents of each extracted
4653 file to the standard input of an external program:
4657 @item --to-command=@var{command}
4658 Extract files and pipe their contents to the standard input of
4659 @var{command}. When this option is used, instead of creating the
4660 files specified, @command{tar} invokes @var{command} and pipes the
4661 contents of the files to its standard output. @var{Command} may
4662 contain command line arguments. The program is executed via
4663 @code{sh -c}. Notice, that @var{command} is executed once for each regular file
4664 extracted. Non-regular files (directories, etc.) are ignored when this
4668 The command can obtain the information about the file it processes
4669 from the following environment variables:
4672 @vrindex TAR_FILETYPE, to-command environment
4674 Type of the file. It is a single letter with the following meaning:
4676 @multitable @columnfractions 0.10 0.90
4677 @item f @tab Regular file
4678 @item d @tab Directory
4679 @item l @tab Symbolic link
4680 @item h @tab Hard link
4681 @item b @tab Block device
4682 @item c @tab Character device
4685 Currently only regular files are supported.
4687 @vrindex TAR_MODE, to-command environment
4689 File mode, an octal number.
4691 @vrindex TAR_FILENAME, to-command environment
4693 The name of the file.
4695 @vrindex TAR_REALNAME, to-command environment
4697 Name of the file as stored in the archive.
4699 @vrindex TAR_UNAME, to-command environment
4701 Name of the file owner.
4703 @vrindex TAR_GNAME, to-command environment
4705 Name of the file owner group.
4707 @vrindex TAR_ATIME, to-command environment
4709 Time of last access. It is a decimal number, representing seconds
4710 since the epoch. If the archive provides times with nanosecond
4711 precision, the nanoseconds are appended to the timestamp after a
4714 @vrindex TAR_MTIME, to-command environment
4716 Time of last modification.
4718 @vrindex TAR_CTIME, to-command environment
4720 Time of last status change.
4722 @vrindex TAR_SIZE, to-command environment
4726 @vrindex TAR_UID, to-command environment
4728 UID of the file owner.
4730 @vrindex TAR_GID, to-command environment
4732 GID of the file owner.
4735 In addition to these variables, @env{TAR_VERSION} contains the
4736 @GNUTAR{} version number.
4738 If @var{command} exits with a non-0 status, @command{tar} will print
4739 an error message similar to the following:
4742 tar: 2345: Child returned status 1
4745 Here, @samp{2345} is the PID of the finished process.
4747 If this behavior is not wanted, use @option{--ignore-command-error}:
4750 @opindex ignore-command-error
4751 @item --ignore-command-error
4752 Ignore exit codes of subprocesses. Notice that if the program
4753 exits on signal or otherwise terminates abnormally, the error message
4754 will be printed even if this option is used.
4756 @opindex no-ignore-command-error
4757 @item --no-ignore-command-error
4758 Cancel the effect of any previous @option{--ignore-command-error}
4759 option. This option is useful if you have set
4760 @option{--ignore-command-error} in @env{TAR_OPTIONS}
4761 (@pxref{TAR_OPTIONS}) and wish to temporarily cancel it.
4765 @unnumberedsubsubsec Removing Files
4767 @FIXME{The section is too terse. Something more to add? An example,
4771 @opindex remove-files
4772 @item --remove-files
4773 Remove files after adding them to the archive.
4777 @subsection Coping with Scarce Resources
4780 @cindex Small memory
4781 @cindex Running out of space
4789 @unnumberedsubsubsec Starting File
4792 @opindex starting-file
4793 @item --starting-file=@var{name}
4794 @itemx -K @var{name}
4795 Starts an operation in the middle of an archive. Use in conjunction
4796 with @option{--extract} (@option{--get}, @option{-x}) or @option{--list} (@option{-t}).
4799 @cindex Middle of the archive, starting in the
4800 If a previous attempt to extract files failed due to lack of disk
4801 space, you can use @option{--starting-file=@var{name}} (@option{-K
4802 @var{name}}) to start extracting only after member @var{name} of the
4803 archive. This assumes, of course, that there is now free space, or
4804 that you are now extracting into a different file system. (You could
4805 also choose to suspend @command{tar}, remove unnecessary files from
4806 the file system, and then restart the same @command{tar} operation.
4807 In this case, @option{--starting-file} is not necessary.
4808 @xref{Incremental Dumps}, @xref{interactive}, and @ref{exclude}.)
4811 @unnumberedsubsubsec Same Order
4814 @cindex Large lists of file names on small machines
4816 @opindex preserve-order
4818 @itemx --preserve-order
4820 To process large lists of file names on machines with small amounts of
4821 memory. Use in conjunction with @option{--compare} (@option{--diff},
4822 @option{-d}), @option{--list} (@option{-t}) or @option{--extract}
4823 (@option{--get}, @option{-x}).
4826 The @option{--same-order} (@option{--preserve-order}, @option{-s}) option tells @command{tar} that the list of file
4827 names to be listed or extracted is sorted in the same order as the
4828 files in the archive. This allows a large list of names to be used,
4829 even on a small machine that would not otherwise be able to hold all
4830 the names in memory at the same time. Such a sorted list can easily be
4831 created by running @samp{tar -t} on the archive and editing its output.
4833 This option is probably never needed on modern computer systems.
4836 @section Backup options
4838 @cindex backup options
4840 @GNUTAR{} offers options for making backups of files
4841 before writing new versions. These options control the details of
4842 these backups. They may apply to the archive itself before it is
4843 created or rewritten, as well as individual extracted members. Other
4844 @acronym{GNU} programs (@command{cp}, @command{install}, @command{ln},
4845 and @command{mv}, for example) offer similar options.
4847 Backup options may prove unexpectedly useful when extracting archives
4848 containing many members having identical name, or when extracting archives
4849 on systems having file name limitations, making different members appear
4850 has having similar names through the side-effect of name truncation.
4851 (This is true only if we have a good scheme for truncated backup names,
4852 which I'm not sure at all: I suspect work is needed in this area.)
4853 When any existing file is backed up before being overwritten by extraction,
4854 then clashing files are automatically be renamed to be unique, and the
4855 true name is kept for only the last file of a series of clashing files.
4856 By using verbose mode, users may track exactly what happens.
4858 At the detail level, some decisions are still experimental, and may
4859 change in the future, we are waiting comments from our users. So, please
4860 do not learn to depend blindly on the details of the backup features.
4861 For example, currently, directories themselves are never renamed through
4862 using these options, so, extracting a file over a directory still has
4863 good chances to fail. Also, backup options apply to created archives,
4864 not only to extracted members. For created archives, backups will not
4865 be attempted when the archive is a block or character device, or when it
4866 refers to a remote file.
4868 For the sake of simplicity and efficiency, backups are made by renaming old
4869 files prior to creation or extraction, and not by copying. The original
4870 name is restored if the file creation fails. If a failure occurs after a
4871 partial extraction of a file, both the backup and the partially extracted
4875 @item --backup[=@var{method}]
4877 @vindex VERSION_CONTROL
4879 Back up files that are about to be overwritten or removed.
4880 Without this option, the original versions are destroyed.
4882 Use @var{method} to determine the type of backups made.
4883 If @var{method} is not specified, use the value of the @env{VERSION_CONTROL}
4884 environment variable. And if @env{VERSION_CONTROL} is not set,
4885 use the @samp{existing} method.
4887 @vindex version-control @r{Emacs variable}
4888 This option corresponds to the Emacs variable @samp{version-control};
4889 the same values for @var{method} are accepted as in Emacs. This option
4890 also allows more descriptive names. The valid @var{method}s are:
4895 @cindex numbered @r{backup method}
4896 Always make numbered backups.
4900 @cindex existing @r{backup method}
4901 Make numbered backups of files that already have them, simple backups
4906 @cindex simple @r{backup method}
4907 Always make simple backups.
4911 @item --suffix=@var{suffix}
4913 @cindex backup suffix
4914 @vindex SIMPLE_BACKUP_SUFFIX
4915 Append @var{suffix} to each backup file made with @option{--backup}. If this
4916 option is not specified, the value of the @env{SIMPLE_BACKUP_SUFFIX}
4917 environment variable is used. And if @env{SIMPLE_BACKUP_SUFFIX} is not
4918 set, the default is @samp{~}, just as in Emacs.
4922 Some people express the desire to @emph{always} use the @option{--backup}
4923 option, by defining some kind of alias or script. This is not as easy
4924 as one may think, due to the fact that old style options should appear first
4925 and consume arguments a bit unpredictably for an alias or script. But,
4926 if you are ready to give up using old style options, you may resort to
4927 using something like (a Bourne shell function here):
4930 tar () @{ /usr/local/bin/tar --backup $*; @}
4934 @section Notable @command{tar} Usages
4937 @FIXME{Using Unix file linking capability to recreate directory
4938 structures---linking files into one subdirectory and then
4939 @command{tar}ring that directory.}
4941 @FIXME{Nice hairy example using absolute-names, newer, etc.}
4944 You can easily use archive files to transport a group of files from
4945 one system to another: put all relevant files into an archive on one
4946 computer system, transfer the archive to another system, and extract
4947 the contents there. The basic transfer medium might be magnetic tape,
4948 Internet FTP, or even electronic mail (though you must encode the
4949 archive with @command{uuencode} in order to transport it properly by
4950 mail). Both machines do not have to use the same operating system, as
4951 long as they both support the @command{tar} program.
4953 For example, here is how you might copy a directory's contents from
4954 one disk to another, while preserving the dates, modes, owners and
4955 link-structure of all the files therein. In this case, the transfer
4956 medium is a @dfn{pipe}, which is one a Unix redirection mechanism:
4959 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xf -)}
4963 You can avoid subshells by using @option{-C} option:
4966 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xf -}
4970 The command also works using short option forms:
4973 $ @kbd{(cd sourcedir; tar --create --file=- . ) \
4974 | (cd targetdir; tar --extract --file=-)}
4976 $ @kbd{tar --directory sourcedir --create --file=- . ) \
4977 | tar --directory targetdir --extract --file=-}
4981 This is one of the easiest methods to transfer a @command{tar} archive.
4984 @section Looking Ahead: The Rest of this Manual
4986 You have now seen how to use all eight of the operations available to
4987 @command{tar}, and a number of the possible options. The next chapter
4988 explains how to choose and change file and archive names, how to use
4989 files to store names of other files which you can then call as
4990 arguments to @command{tar} (this can help you save time if you expect to
4991 archive the same list of files a number of times), and so forth.
4992 @FIXME{in case it's not obvious, i'm making this up in some sense
4993 based on my limited memory of what the next chapter *really* does. i
4994 just wanted to flesh out this final section a little bit so i'd
4995 remember to stick it in here. :-)}
4997 If there are too many files to conveniently list on the command line,
4998 you can list the names in a file, and @command{tar} will read that file.
5001 There are various ways of causing @command{tar} to skip over some files,
5002 and not archive them. @xref{Choosing}.
5005 @chapter Performing Backups and Restoring Files
5008 @GNUTAR{} is distributed along with the scripts
5009 which the Free Software Foundation uses for performing backups. There
5010 is no corresponding scripts available yet for doing restoration of
5011 files. Even if there is a good chance those scripts may be satisfying
5012 to you, they are not the only scripts or methods available for doing
5013 backups and restore. You may well create your own, or use more
5014 sophisticated packages dedicated to that purpose.
5016 Some users are enthusiastic about @code{Amanda} (The Advanced Maryland
5017 Automatic Network Disk Archiver), a backup system developed by James
5018 da Silva @file{jds@@cs.umd.edu} and available on many Unix systems.
5019 This is free software, and it is available at these places:
5022 http://www.cs.umd.edu/projects/amanda/amanda.html
5023 ftp://ftp.cs.umd.edu/pub/amanda
5028 Here is a possible plan for a future documentation about the backuping
5029 scripts which are provided within the @GNUTAR{}
5035 @item what are dumps
5036 @item different levels of dumps
5038 @item full dump = dump everything
5039 @item level 1, level 2 dumps etc
5040 A level @var{n} dump dumps everything changed since the last level
5043 @item how to use scripts for dumps (ie, the concept)
5045 @item scripts to run after editing backup specs (details)
5047 @item Backup Specs, what is it.
5049 @item how to customize
5050 @item actual text of script [/sp/dump/backup-specs]
5054 @item rsh doesn't work
5055 @item rtape isn't installed
5058 @item the @option{--incremental} option of tar
5061 @item write protection
5062 @item types of media, different sizes and types, useful for different things
5063 @item files and tape marks
5064 one tape mark between files, two at end.
5065 @item positioning the tape
5066 MT writes two at end of write,
5067 backspaces over one when writing again.
5073 This chapter documents both the provided shell scripts and @command{tar}
5074 options which are more specific to usage as a backup tool.
5076 To @dfn{back up} a file system means to create archives that contain
5077 all the files in that file system. Those archives can then be used to
5078 restore any or all of those files (for instance if a disk crashes or a
5079 file is accidentally deleted). File system @dfn{backups} are also
5083 * Full Dumps:: Using @command{tar} to Perform Full Dumps
5084 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
5085 * Backup Levels:: Levels of Backups
5086 * Backup Parameters:: Setting Parameters for Backups and Restoration
5087 * Scripted Backups:: Using the Backup Scripts
5088 * Scripted Restoration:: Using the Restore Script
5092 @section Using @command{tar} to Perform Full Dumps
5098 @cindex corrupted archives
5099 Full dumps should only be made when no other people or programs
5100 are modifying files in the file system. If files are modified while
5101 @command{tar} is making the backup, they may not be stored properly in
5102 the archive, in which case you won't be able to restore them if you
5103 have to. (Files not being modified are written with no trouble, and do
5104 not corrupt the entire archive.)
5106 You will want to use the @option{--label=@var{archive-label}}
5107 (@option{-V @var{archive-label}}) option to give the archive a
5108 volume label, so you can tell what this archive is even if the label
5109 falls off the tape, or anything like that.
5111 Unless the file system you are dumping is guaranteed to fit on
5112 one volume, you will need to use the @option{--multi-volume} (@option{-M}) option.
5113 Make sure you have enough tapes on hand to complete the backup.
5115 If you want to dump each file system separately you will need to use
5116 the @option{--one-file-system} option to prevent
5117 @command{tar} from crossing file system boundaries when storing
5120 The @option{--incremental} (@option{-G}) (@pxref{Incremental Dumps})
5121 option is not needed, since this is a complete copy of everything in
5122 the file system, and a full restore from this backup would only be
5123 done onto a completely
5126 Unless you are in a hurry, and trust the @command{tar} program (and your
5127 tapes), it is a good idea to use the @option{--verify} (@option{-W})
5128 option, to make sure your files really made it onto the dump properly.
5129 This will also detect cases where the file was modified while (or just
5130 after) it was being archived. Not all media (notably cartridge tapes)
5131 are capable of being verified, unfortunately.
5133 @node Incremental Dumps
5134 @section Using @command{tar} to Perform Incremental Dumps
5136 @dfn{Incremental backup} is a special form of @GNUTAR{} archive that
5137 stores additional metadata so that exact state of the file system
5138 can be restored when extracting the archive.
5140 @GNUTAR{} currently offers two options for handling incremental
5141 backups: @option{--listed-incremental=@var{snapshot-file}} (@option{-g
5142 @var{snapshot-file}}) and @option{--incremental} (@option{-G}).
5144 @opindex listed-incremental
5145 The option @option{--listed-incremental} instructs tar to operate on
5146 an incremental archive with additional metadata stored in a standalone
5147 file, called a @dfn{snapshot file}. The purpose of this file is to help
5148 determine which files have been changed, added or deleted since the
5149 last backup, so that the next incremental backup will contain only
5150 modified files. The name of the snapshot file is given as an argument
5154 @item --listed-incremental=@var{file}
5155 @itemx -g @var{file}
5156 Handle incremental backups with snapshot data in @var{file}.
5159 To create an incremental backup, you would use
5160 @option{--listed-incremental} together with @option{--create}
5161 (@pxref{create}). For example:
5164 $ @kbd{tar --create \
5165 --file=archive.1.tar \
5166 --listed-incremental=/var/log/usr.snar \
5170 This will create in @file{archive.1.tar} an incremental backup of
5171 the @file{/usr} file system, storing additional metadata in the file
5172 @file{/var/log/usr.snar}. If this file does not exist, it will be
5173 created. The created archive will then be a @dfn{level 0 backup};
5174 please see the next section for more on backup levels.
5176 Otherwise, if the file @file{/var/log/usr.snar} exists, it
5177 determines which files are modified. In this case only these files will be
5178 stored in the archive. Suppose, for example, that after running the
5179 above command, you delete file @file{/usr/doc/old} and create
5180 directory @file{/usr/local/db} with the following contents:
5183 $ @kbd{ls /usr/local/db}
5188 Some time later you create another incremental backup. You will
5192 $ @kbd{tar --create \
5193 --file=archive.2.tar \
5194 --listed-incremental=/var/log/usr.snar \
5196 tar: usr/local/db: Directory is new
5203 The created archive @file{archive.2.tar} will contain only these
5204 three members. This archive is called a @dfn{level 1 backup}. Notice
5205 that @file{/var/log/usr.snar} will be updated with the new data, so if
5206 you plan to create more @samp{level 1} backups, it is necessary to
5207 create a working copy of the snapshot file before running
5208 @command{tar}. The above example will then be modified as follows:
5211 $ @kbd{cp /var/log/usr.snar /var/log/usr.snar-1}
5212 $ @kbd{tar --create \
5213 --file=archive.2.tar \
5214 --listed-incremental=/var/log/usr.snar-1 \
5218 Incremental dumps depend crucially on time stamps, so the results are
5219 unreliable if you modify a file's time stamps during dumping (e.g.,
5220 with the @option{--atime-preserve=replace} option), or if you set the clock
5223 Metadata stored in snapshot files include device numbers, which,
5224 obviously is supposed to be a non-volatile value. However, it turns
5225 out that NFS devices have undependable values when an automounter
5226 gets in the picture. This can lead to a great deal of spurious
5227 redumping in incremental dumps, so it is somewhat useless to compare
5228 two NFS devices numbers over time. The solution implemented currently
5229 is to considers all NFS devices as being equal when it comes to
5230 comparing directories; this is fairly gross, but there does not seem
5231 to be a better way to go.
5233 Note that incremental archives use @command{tar} extensions and may
5234 not be readable by non-@acronym{GNU} versions of the @command{tar} program.
5236 @opindex listed-incremental, using with @option{--extract}
5237 @opindex extract, using with @option{--listed-incremental}
5238 To extract from the incremental dumps, use
5239 @option{--listed-incremental} together with @option{--extract}
5240 option (@pxref{extracting files}). In this case, @command{tar} does
5241 not need to access snapshot file, since all the data necessary for
5242 extraction are stored in the archive itself. So, when extracting, you
5243 can give whatever argument to @option{--listed-incremental}, the usual
5244 practice is to use @option{--listed-incremental=/dev/null}.
5245 Alternatively, you can use @option{--incremental}, which needs no
5246 arguments. In general, @option{--incremental} (@option{-G}) can be
5247 used as a shortcut for @option{--listed-incremental} when listing or
5248 extracting incremental backups (for more information, regarding this
5249 option, @pxref{incremental-op}).
5251 When extracting from the incremental backup @GNUTAR{} attempts to
5252 restore the exact state the file system had when the archive was
5253 created. In particular, it will @emph{delete} those files in the file
5254 system that did not exist in their directories when the archive was
5255 created. If you have created several levels of incremental files,
5256 then in order to restore the exact contents the file system had when
5257 the last level was created, you will need to restore from all backups
5258 in turn. Continuing our example, to restore the state of @file{/usr}
5259 file system, one would do@footnote{Notice, that since both archives
5260 were created withouth @option{-P} option (@pxref{absolute}), these
5261 commands should be run from the root file system.}:
5264 $ @kbd{tar --extract \
5265 --listed-incremental=/dev/null \
5266 --file archive.1.tar}
5267 $ @kbd{tar --extract \
5268 --listed-incremental=/dev/null \
5269 --file archive.2.tar}
5272 To list the contents of an incremental archive, use @option{--list}
5273 (@pxref{list}), as usual. To obtain more information about the
5274 archive, use @option{--listed-incremental} or @option{--incremental}
5275 combined with two @option{--verbose} options@footnote{Two
5276 @option{--verbose} options were selected to avoid breaking usual
5277 verbose listing output (@option{--list --verbose}) when using in
5280 @opindex incremental, using with @option{--list}
5281 @opindex listed-incremental, using with @option{--list}
5282 @opindex list, using with @option{--incremental}
5283 @opindex list, using with @option{--listed-incremental}
5284 Versions of @GNUTAR{} up to 1.15.1 used to dump verbatim binary
5285 contents of the DUMPDIR header (with terminating nulls) when
5286 @option{--incremental} or @option{--listed-incremental} option was
5287 given, no matter what the verbosity level. This behavior, and,
5288 especially, the binary output it produced were considered incovenient
5289 and were changed in version 1.16}:
5292 @kbd{tar --list --incremental --verbose --verbose archive.tar}
5295 This command will print, for each directory in the archive, the list
5296 of files in that directory at the time the archive was created. This
5297 information is put out in a format which is both human-readable and
5298 unambiguous for a program: each file name is printed as
5305 where @var{x} is a letter describing the status of the file: @samp{Y}
5306 if the file is present in the archive, @samp{N} if the file is not
5307 included in the archive, or a @samp{D} if the file is a directory (and
5308 is included in the archive). @xref{Dumpdir}, for the detailed
5309 description of dumpdirs and status codes. Each such
5310 line is terminated by a newline character. The last line is followed
5311 by an additional newline to indicate the end of the data.
5313 @anchor{incremental-op}The option @option{--incremental} (@option{-G})
5314 gives the same behavior as @option{--listed-incremental} when used
5315 with @option{--list} and @option{--extract} options. When used with
5316 @option{--create} option, it creates an incremental archive without
5317 creating snapshot file. Thus, it is impossible to create several
5318 levels of incremental backups with @option{--incremental} option.
5321 @section Levels of Backups
5323 An archive containing all the files in the file system is called a
5324 @dfn{full backup} or @dfn{full dump}. You could insure your data by
5325 creating a full dump every day. This strategy, however, would waste a
5326 substantial amount of archive media and user time, as unchanged files
5327 are daily re-archived.
5329 It is more efficient to do a full dump only occasionally. To back up
5330 files between full dumps, you can use @dfn{incremental dumps}. A @dfn{level
5331 one} dump archives all the files that have changed since the last full
5334 A typical dump strategy would be to perform a full dump once a week,
5335 and a level one dump once a day. This means some versions of files
5336 will in fact be archived more than once, but this dump strategy makes
5337 it possible to restore a file system to within one day of accuracy by
5338 only extracting two archives---the last weekly (full) dump and the
5339 last daily (level one) dump. The only information lost would be in
5340 files changed or created since the last daily backup. (Doing dumps
5341 more than once a day is usually not worth the trouble).
5343 @GNUTAR{} comes with scripts you can use to do full
5344 and level-one (actually, even level-two and so on) dumps. Using
5345 scripts (shell programs) to perform backups and restoration is a
5346 convenient and reliable alternative to typing out file name lists
5347 and @command{tar} commands by hand.
5349 Before you use these scripts, you need to edit the file
5350 @file{backup-specs}, which specifies parameters used by the backup
5351 scripts and by the restore script. This file is usually located
5352 in @file{/etc/backup} directory. @xref{Backup Parameters}, for its
5353 detailed description. Once the backup parameters are set, you can
5354 perform backups or restoration by running the appropriate script.
5356 The name of the backup script is @code{backup}. The name of the
5357 restore script is @code{restore}. The following sections describe
5358 their use in detail.
5360 @emph{Please Note:} The backup and restoration scripts are
5361 designed to be used together. While it is possible to restore files by
5362 hand from an archive which was created using a backup script, and to create
5363 an archive by hand which could then be extracted using the restore script,
5364 it is easier to use the scripts. @xref{Incremental Dumps}, before
5365 making such an attempt.
5367 @node Backup Parameters
5368 @section Setting Parameters for Backups and Restoration
5370 The file @file{backup-specs} specifies backup parameters for the
5371 backup and restoration scripts provided with @command{tar}. You must
5372 edit @file{backup-specs} to fit your system configuration and schedule
5373 before using these scripts.
5375 Syntactically, @file{backup-specs} is a shell script, containing
5376 mainly variable assignments. However, any valid shell construct
5377 is allowed in this file. Particularly, you may wish to define
5378 functions within that script (e.g., see @code{RESTORE_BEGIN} below).
5379 For more information about shell script syntax, please refer to
5380 @url{http://www.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html#ta
5381 g_02, the definition of the Shell Command Language}. See also
5382 @ref{Top,,Bash Features,bashref,Bash Reference Manual}.
5384 The shell variables controlling behavior of @code{backup} and
5385 @code{restore} are described in the following subsections.
5388 * General-Purpose Variables::
5389 * Magnetic Tape Control::
5391 * backup-specs example:: An Example Text of @file{Backup-specs}
5394 @node General-Purpose Variables
5395 @subsection General-Purpose Variables
5397 @defvr {Backup variable} ADMINISTRATOR
5398 The user name of the backup administrator. @code{Backup} scripts
5399 sends a backup report to this address.
5402 @defvr {Backup variable} BACKUP_HOUR
5403 The hour at which the backups are done. This can be a number from 0
5404 to 23, or the time specification in form @var{hours}:@var{minutes},
5405 or the string @samp{now}.
5407 This variable is used by @code{backup}. Its value may be overridden
5408 using @option{--time} option (@pxref{Scripted Backups}).
5411 @defvr {Backup variable} TAPE_FILE
5413 The device @command{tar} writes the archive to. If @var{TAPE_FILE}
5414 is a remote archive (@pxref{remote-dev}), backup script will suppose
5415 that your @command{mt} is able to access remote devices. If @var{RSH}
5416 (@pxref{RSH}) is set, @option{--rsh-command} option will be added to
5417 invocations of @command{mt}.
5420 @defvr {Backup variable} BLOCKING
5422 The blocking factor @command{tar} will use when writing the dump archive.
5423 @xref{Blocking Factor}.
5426 @defvr {Backup variable} BACKUP_DIRS
5428 A list of file systems to be dumped (for @code{backup}), or restored
5429 (for @code{restore}). You can include any directory
5430 name in the list --- subdirectories on that file system will be
5431 included, regardless of how they may look to other networked machines.
5432 Subdirectories on other file systems will be ignored.
5434 The host name specifies which host to run @command{tar} on, and should
5435 normally be the host that actually contains the file system. However,
5436 the host machine must have @GNUTAR{} installed, and
5437 must be able to access the directory containing the backup scripts and
5438 their support files using the same file name that is used on the
5439 machine where the scripts are run (ie. what @command{pwd} will print
5440 when in that directory on that machine). If the host that contains
5441 the file system does not have this capability, you can specify another
5442 host as long as it can access the file system through NFS.
5444 If the list of file systems is very long you may wish to put it
5445 in a separate file. This file is usually named
5446 @file{/etc/backup/dirs}, but this name may be overridden in
5447 @file{backup-specs} using @code{DIRLIST} variable.
5450 @defvr {Backup variable} DIRLIST
5452 A path to the file containing the list of the file systems to backup
5453 or restore. By default it is @file{/etc/backup/dirs}.
5456 @defvr {Backup variable} BACKUP_FILES
5458 A list of individual files to be dumped (for @code{backup}), or restored
5459 (for @code{restore}). These should be accessible from the machine on
5460 which the backup script is run.
5462 If the list of file systems is very long you may wish to store it
5463 in a separate file. This file is usually named
5464 @file{/etc/backup/files}, but this name may be overridden in
5465 @file{backup-specs} using @code{FILELIST} variable.
5468 @defvr {Backup variable} FILELIST
5470 A path to the file containing the list of the individual files to backup
5471 or restore. By default it is @file{/etc/backup/files}.
5474 @defvr {Backup variable} MT
5476 Full file name of @command{mt} binary.
5479 @defvr {Backup variable} RSH
5481 Full file name of @command{rsh} binary or its equivalent. You may wish to
5482 set it to @code{ssh}, to improve security. In this case you will have
5483 to use public key authentication.
5486 @defvr {Backup variable} RSH_COMMAND
5488 Full file name of @command{rsh} binary on remote mashines. This will
5489 be passed via @option{--rsh-command} option to the remote invocation
5493 @defvr {Backup variable} VOLNO_FILE
5495 Name of temporary file to hold volume numbers. This needs to be accessible
5496 by all the machines which have file systems to be dumped.
5499 @defvr {Backup variable} XLIST
5501 Name of @dfn{exclude file list}. An @dfn{exclude file list} is a file
5502 located on the remote machine and containing the list of files to
5503 be excluded from the backup. Exclude file lists are searched in
5504 /etc/tar-backup directory. A common use for exclude file lists
5505 is to exclude files containing security-sensitive information
5506 (e.g., @file{/etc/shadow} from backups).
5508 This variable affects only @code{backup}.
5511 @defvr {Backup variable} SLEEP_TIME
5513 Time to sleep between dumps of any two successive file systems
5515 This variable affects only @code{backup}.
5518 @defvr {Backup variable} DUMP_REMIND_SCRIPT
5520 Script to be run when it's time to insert a new tape in for the next
5521 volume. Administrators may want to tailor this script for their site.
5522 If this variable isn't set, @GNUTAR{} will display its built-in prompt
5523 @FIXME-xref{describe it somewhere!}, and will expect confirmation from
5527 @defvr {Backup variable} SLEEP_MESSAGE
5529 Message to display on the terminal while waiting for dump time. Usually
5530 this will just be some literal text.
5533 @defvr {Backup variable} TAR
5535 Full file name of the @GNUTAR{} executable. If this is not set, backup
5536 scripts will search @command{tar} in the current shell path.
5539 @node Magnetic Tape Control
5540 @subsection Magnetic Tape Control
5542 Backup scripts access tape device using special @dfn{hook functions}.
5543 These functions take a single argument -- the name of the tape
5544 device. Their names are kept in the following variables:
5546 @defvr {Backup variable} MT_BEGIN
5547 The name of @dfn{begin} function. This function is called before
5548 accessing the drive. By default it retensions the tape:
5554 mt -f "$1" retension
5559 @defvr {Backup variable} MT_REWIND
5560 The name of @dfn{rewind} function. The default definition is as
5573 @defvr {Backup variable} MT_OFFLINE
5574 The name of the function switching the tape off line. By default
5575 it is defined as follows:
5578 MT_OFFLINE=mt_offline
5586 @defvr {Backup variable} MT_STATUS
5587 The name of the function used to obtain the status of the archive device,
5588 including error count. Default definition:
5600 @subsection User Hooks
5602 @dfn{User hooks} are shell functions executed before and after
5603 each @command{tar} invocation. Thus, there are @dfn{backup
5604 hooks}, which are executed before and after dumping each file
5605 system, and @dfn{restore hooks}, executed before and
5606 after restoring a file system. Each user hook is a shell function
5607 taking four arguments:
5609 @deffn {User Hook Function} hook @var{level} @var{host} @var{fs} @var{fsname}
5614 Current backup or restore level.
5617 Name or IP address of the host machine being dumped or restored.
5620 Full path name to the file system being dumped or restored.
5623 File system name with directory separators replaced with colons. This
5624 is useful, e.g., for creating unique files.
5628 Following variables keep the names of user hook functions
5630 @defvr {Backup variable} DUMP_BEGIN
5631 Dump begin function. It is executed before dumping the file system.
5634 @defvr {Backup variable} DUMP_END
5635 Executed after dumping the file system.
5638 @defvr {Backup variable} RESTORE_BEGIN
5639 Executed before restoring the file system.
5642 @defvr {Backup variable} RESTORE_END
5643 Executed after restoring the file system.
5646 @node backup-specs example
5647 @subsection An Example Text of @file{Backup-specs}
5649 The following is an example of @file{backup-specs}:
5652 # site-specific parameters for file system backup.
5654 ADMINISTRATOR=friedman
5656 TAPE_FILE=/dev/nrsmt0
5658 # Use @code{ssh} instead of the less secure @code{rsh}
5660 RSH_COMMAND=/usr/bin/ssh
5662 # Override MT_STATUS function:
5668 # Disable MT_OFFLINE function
5685 apple-gunkies:/com/mailer/gnu
5686 apple-gunkies:/com/archive/gnu"
5688 BACKUP_FILES="/com/mailer/aliases /com/mailer/league*[a-z]"
5692 @node Scripted Backups
5693 @section Using the Backup Scripts
5695 The syntax for running a backup script is:
5698 backup --level=@var{level} --time=@var{time}
5701 The @option{level} option requests the dump level. Thus, to produce
5702 a full dump, specify @code{--level=0} (this is the default, so
5703 @option{--level} may be omitted if its value is @code{0}).
5704 @footnote{For backward compatibility, the @code{backup} will also
5705 try to deduce the requested dump level from the name of the
5706 script itself. If the name consists of a string @samp{level-}
5707 followed by a single decimal digit, that digit is taken as
5708 the dump level number. Thus, you may create a link from @code{backup}
5709 to @code{level-1} and then run @code{level-1} whenever you need to
5710 create a level one dump.}
5712 The @option{--time} option determines when should the backup be
5713 run. @var{Time} may take three forms:
5716 @item @var{hh}:@var{mm}
5718 The dump must be run at @var{hh} hours @var{mm} minutes.
5722 The dump must be run at @var{hh} hours
5726 The dump must be run immediately.
5729 You should start a script with a tape or disk mounted. Once you
5730 start a script, it prompts you for new tapes or disks as it
5731 needs them. Media volumes don't have to correspond to archive
5732 files --- a multi-volume archive can be started in the middle of a
5733 tape that already contains the end of another multi-volume archive.
5734 The @code{restore} script prompts for media by its archive volume,
5735 so to avoid an error message you should keep track of which tape
5736 (or disk) contains which volume of the archive (@pxref{Scripted
5739 The backup scripts write two files on the file system. The first is a
5740 record file in @file{/etc/tar-backup/}, which is used by the scripts
5741 to store and retrieve information about which files were dumped. This
5742 file is not meant to be read by humans, and should not be deleted by
5743 them. @xref{Snapshot Files}, for a more detailed explanation of this
5746 The second file is a log file containing the names of the file systems
5747 and files dumped, what time the backup was made, and any error
5748 messages that were generated, as well as how much space was left in
5749 the media volume after the last volume of the archive was written.
5750 You should check this log file after every backup. The file name is
5751 @file{log-@var{mm-dd-yyyy}-level-@var{n}}, where @var{mm-dd-yyyy}
5752 represents current date, and @var{n} represents current dump level number.
5754 The script also prints the name of each system being dumped to the
5757 Following is the full list of options accepted by @code{backup}
5761 @item -l @var{level}
5762 @itemx --level=@var{level}
5763 Do backup level @var{level} (default 0).
5767 Force backup even if today's log file already exists.
5769 @item -v[@var{level}]
5770 @itemx --verbose[=@var{level}]
5771 Set verbosity level. The higher the level is, the more debugging
5772 information will be output during execution. Devault @var{level}
5773 is 100, which means the highest debugging level.
5775 @item -t @var{start-time}
5776 @itemx --time=@var{start-time}
5777 Wait till @var{time}, then do backup.
5781 Display short help message and exit.
5785 Display information about the program's name, version, origin and legal
5786 status, all on standard output, and then exit successfully.
5790 @node Scripted Restoration
5791 @section Using the Restore Script
5793 To restore files that were archived using a scripted backup, use the
5794 @code{restore} script. Its usage is quite straightforward. In the
5795 simplest form, invoke @code{restore --all}, it will
5796 then restore all the file systems and files specified in
5797 @file{backup-specs} (@pxref{General-Purpose Variables,BACKUP_DIRS}).
5799 You may select the file systems (and/or files) to restore by
5800 giving @code{restore} list of @dfn{patterns} in its command
5801 line. For example, running
5808 will restore all file systems on the machine @samp{albert}. A more
5809 complicated example:
5812 restore 'albert:*' '*:/var'
5816 This command will restore all file systems on the machine @samp{albert}
5817 as well as @file{/var} file system on all machines.
5819 By default @code{restore} will start restoring files from the lowest
5820 available dump level (usually zero) and will continue through
5821 all available dump levels. There may be situations where such a
5822 thorough restore is not necessary. For example, you may wish to
5823 restore only files from the recent level one backup. To do so,
5824 use @option{--level} option, as shown in the example below:
5830 The full list of options accepted by @code{restore} follows:
5835 Restore all file systems and files specified in @file{backup-specs}
5837 @item -l @var{level}
5838 @itemx --level=@var{level}
5839 Start restoring from the given backup level, instead of the default 0.
5841 @item -v[@var{level}]
5842 @itemx --verbose[=@var{level}]
5843 Set verbosity level. The higher the level is, the more debugging
5844 information will be output during execution. Devault @var{level}
5845 is 100, which means the highest debugging level.
5849 Display short help message and exit.
5853 Display information about the program's name, version, origin and legal
5854 status, all on standard output, and then exit successfully.
5857 You should start the restore script with the media containing the
5858 first volume of the archive mounted. The script will prompt for other
5859 volumes as they are needed. If the archive is on tape, you don't need
5860 to rewind the tape to to its beginning---if the tape head is
5861 positioned past the beginning of the archive, the script will rewind
5862 the tape as needed. @FIXME-xref{Media, for a discussion of tape
5866 @strong{Warning:} The script will delete files from the active file
5867 system if they were not in the file system when the archive was made.
5870 @xref{Incremental Dumps}, for an explanation of how the script makes
5874 @chapter Choosing Files and Names for @command{tar}
5877 Certain options to @command{tar} enable you to specify a name for your
5878 archive. Other options let you decide which files to include or exclude
5879 from the archive, based on when or whether files were modified, whether
5880 the file names do or don't match specified patterns, or whether files
5881 are in specified directories.
5883 This chapter discusses these options in detail.
5886 * file:: Choosing the Archive's Name
5887 * Selecting Archive Members::
5888 * files:: Reading Names from a File
5889 * exclude:: Excluding Some Files
5890 * wildcards:: Wildcards Patterns and Matching
5891 * quoting styles:: Ways of Quoting Special Characters in Names
5892 * transform:: Modifying File and Member Names
5893 * after:: Operating Only on New Files
5894 * recurse:: Descending into Directories
5895 * one:: Crossing File System Boundaries
5899 @section Choosing and Naming Archive Files
5902 @cindex Naming an archive
5903 @cindex Archive Name
5904 @cindex Choosing an archive file
5905 @cindex Where is the archive?
5906 By default, @command{tar} uses an archive file name that was compiled when
5907 it was built on the system; usually this name refers to some physical
5908 tape drive on the machine. However, the person who installed @command{tar}
5909 on the system may not have set the default to a meaningful value as far as
5910 most users are concerned. As a result, you will usually want to tell
5911 @command{tar} where to find (or create) the archive. The
5912 @option{--file=@var{archive-name}} (@option{-f @var{archive-name}})
5913 option allows you to either specify or name a file to use as the archive
5914 instead of the default archive file location.
5917 @opindex file, short description
5918 @item --file=@var{archive-name}
5919 @itemx -f @var{archive-name}
5920 Name the archive to create or operate on. Use in conjunction with
5924 For example, in this @command{tar} command,
5927 $ @kbd{tar -cvf collection.tar blues folk jazz}
5931 @file{collection.tar} is the name of the archive. It must directly
5932 follow the @option{-f} option, since whatever directly follows @option{-f}
5933 @emph{will} end up naming the archive. If you neglect to specify an
5934 archive name, you may end up overwriting a file in the working directory
5935 with the archive you create since @command{tar} will use this file's name
5936 for the archive name.
5938 An archive can be saved as a file in the file system, sent through a
5939 pipe or over a network, or written to an I/O device such as a tape,
5940 floppy disk, or CD write drive.
5942 @cindex Writing new archives
5943 @cindex Archive creation
5944 If you do not name the archive, @command{tar} uses the value of the
5945 environment variable @env{TAPE} as the file name for the archive. If
5946 that is not available, @command{tar} uses a default, compiled-in archive
5947 name, usually that for tape unit zero (ie. @file{/dev/tu00}).
5949 @cindex Standard input and output
5950 @cindex tar to standard input and output
5951 If you use @file{-} as an @var{archive-name}, @command{tar} reads the
5952 archive from standard input (when listing or extracting files), or
5953 writes it to standard output (when creating an archive). If you use
5954 @file{-} as an @var{archive-name} when modifying an archive,
5955 @command{tar} reads the original archive from its standard input and
5956 writes the entire new archive to its standard output.
5958 The following example is a convenient way of copying directory
5959 hierarchy from @file{sourcedir} to @file{targetdir}.
5962 $ @kbd{(cd sourcedir; tar -cf - .) | (cd targetdir; tar -xpf -)}
5965 The @option{-C} option allows to avoid using subshells:
5968 $ @kbd{tar -C sourcedir -cf - . | tar -C targetdir -xpf -}
5971 In both examples above, the leftmost @command{tar} invocation archives
5972 the contents of @file{sourcedir} to the standard output, while the
5973 rightmost one reads this archive from its standard input and
5974 extracts it. The @option{-p} option tells it to restore permissions
5975 of the extracted files.
5977 @cindex Remote devices
5978 @cindex tar to a remote device
5980 To specify an archive file on a device attached to a remote machine,
5984 @kbd{--file=@var{hostname}:/@var{dev}/@var{file-name}}
5988 @command{tar} will complete the remote connection, if possible, and
5989 prompt you for a username and password. If you use
5990 @option{--file=@@@var{hostname}:/@var{dev}/@var{file-name}}, @command{tar}
5991 will complete the remote connection, if possible, using your username
5992 as the username on the remote machine.
5994 @cindex Local and remote archives
5995 @anchor{local and remote archives}
5996 If the archive file name includes a colon (@samp{:}), then it is assumed
5997 to be a file on another machine. If the archive file is
5998 @samp{@var{user}@@@var{host}:@var{file}}, then @var{file} is used on the
5999 host @var{host}. The remote host is accessed using the @command{rsh}
6000 program, with a username of @var{user}. If the username is omitted
6001 (along with the @samp{@@} sign), then your user name will be used.
6002 (This is the normal @command{rsh} behavior.) It is necessary for the
6003 remote machine, in addition to permitting your @command{rsh} access, to
6004 have the @file{rmt} program installed (This command is included in
6005 the @GNUTAR{} distribution and by default is installed under
6006 @file{@var{prefix}/libexec/rmt}, were @var{prefix} means your
6007 installation prefix). If you need to use a file whose name includes a
6008 colon, then the remote tape drive behavior
6009 can be inhibited by using the @option{--force-local} option.
6011 When the archive is being created to @file{/dev/null}, @GNUTAR{}
6012 tries to minimize input and output operations. The Amanda backup
6013 system, when used with @GNUTAR{}, has an initial sizing pass which
6016 @node Selecting Archive Members
6017 @section Selecting Archive Members
6018 @cindex Specifying files to act on
6019 @cindex Specifying archive members
6021 @dfn{File Name arguments} specify which files in the file system
6022 @command{tar} operates on, when creating or adding to an archive, or which
6023 archive members @command{tar} operates on, when reading or deleting from
6024 an archive. @xref{Operations}.
6026 To specify file names, you can include them as the last arguments on
6027 the command line, as follows:
6029 @kbd{tar} @var{operation} [@var{option1} @var{option2} @dots{}] [@var{file name-1} @var{file name-2} @dots{}]
6032 If a file name begins with dash (@samp{-}), precede it with
6033 @option{--add-file} option to prevent it from being treated as an
6036 If you specify a directory name as a file name argument, all the files
6037 in that directory are operated on by @command{tar}.
6039 If you do not specify files, @command{tar} behavior differs depending
6040 on the operation mode as described below:
6042 When @command{tar} is invoked with @option{--create} (@option{-c}),
6043 @command{tar} will stop immediately, reporting the following:
6047 $ @kbd{tar cf a.tar}
6048 tar: Cowardly refusing to create an empty archive
6049 Try `tar --help' or `tar --usage' for more information.
6053 If you specify either @option{--list} (@option{-t}) or
6054 @option{--extract} (@option{--get}, @option{-x}), @command{tar}
6055 operates on all the archive members in the archive.
6057 If run with @option{--diff} option, tar will compare the archive with
6058 the contents of the current working directory.
6060 If you specify any other operation, @command{tar} does nothing.
6062 By default, @command{tar} takes file names from the command line. However,
6063 there are other ways to specify file or member names, or to modify the
6064 manner in which @command{tar} selects the files or members upon which to
6065 operate. In general, these methods work both for specifying the names
6066 of files and archive members.
6069 @section Reading Names from a File
6071 @cindex Reading file names from a file
6072 @cindex Lists of file names
6073 @cindex File Name arguments, alternatives
6074 Instead of giving the names of files or archive members on the command
6075 line, you can put the names into a file, and then use the
6076 @option{--files-from=@var{file-of-names}} (@option{-T
6077 @var{file-of-names}}) option to @command{tar}. Give the name of the
6078 file which contains the list of files to include as the argument to
6079 @option{--files-from}. In the list, the file names should be separated by
6080 newlines. You will frequently use this option when you have generated
6081 the list of files to archive with the @command{find} utility.
6085 @item --files-from=@var{file-name}
6086 @itemx -T @var{file-name}
6087 Get names to extract or create from file @var{file-name}.
6090 If you give a single dash as a file name for @option{--files-from}, (i.e.,
6091 you specify either @code{--files-from=-} or @code{-T -}), then the file
6092 names are read from standard input.
6094 Unless you are running @command{tar} with @option{--create}, you can not use
6095 both @code{--files-from=-} and @code{--file=-} (@code{-f -}) in the same
6098 Any number of @option{-T} options can be given in the command line.
6100 The following example shows how to use @command{find} to generate a list of
6101 files smaller than 400K in length and put that list into a file
6102 called @file{small-files}. You can then use the @option{-T} option to
6103 @command{tar} to specify the files from that file, @file{small-files}, to
6104 create the archive @file{little.tgz}. (The @option{-z} option to
6105 @command{tar} compresses the archive with @command{gzip}; @pxref{gzip} for
6109 $ @kbd{find . -size -400 -print > small-files}
6110 $ @kbd{tar -c -v -z -T small-files -f little.tgz}
6114 In the file list given by @option{-T} option, any file name beginning
6115 with @samp{-} character is considered a @command{tar} option and is
6116 processed accordingly.@footnote{Versions of @GNUTAR{} up to 1.15.1
6117 recognized only @option{-C} option in file lists, and only if the
6118 option and its argument occupied two consecutive lines.} For example,
6119 the common use of this feature is to change to another directory by
6120 specifying @option{-C} option:
6130 $ @kbd{tar -c -f foo.tar --files-from list}
6135 In this example, @command{tar} will first switch to @file{/etc}
6136 directory and add files @file{passwd} and @file{hosts} to the
6137 archive. Then it will change to @file{/lib} directory and will archive
6138 the file @file{libc.a}. Thus, the resulting archive @file{foo.tar} will
6143 $ @kbd{tar tf foo.tar}
6151 @opindex directory, using in @option{--files-from} argument
6152 Notice that the option parsing algorithm used with @option{-T} is
6153 stricter than the one used by shell. Namely, when specifying option
6154 arguments, you should observe the following rules:
6158 When using short (single-letter) option form, its argument must
6159 immediately follow the option letter, without any intervening
6160 whitespace. For example: @code{-Cdir}.
6163 When using long option form, the option argument must be separated
6164 from the option by a single equal sign. No whitespace is allowed on
6165 any side of the equal sign. For example: @code{--directory=dir}.
6168 For both short and long option forms, the option argument can be given
6169 on the next line after the option name, e.g.:
6190 If you happen to have a file whose name starts with @samp{-},
6191 precede it with @option{--add-file} option to prevent it from
6192 being recognized as an option. For example: @code{--add-file=--my-file}.
6199 @subsection @code{NUL} Terminated File Names
6201 @cindex File names, terminated by @code{NUL}
6202 @cindex @code{NUL} terminated file names
6203 The @option{--null} option causes
6204 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}})
6205 to read file names terminated by a @code{NUL} instead of a newline, so
6206 files whose names contain newlines can be archived using
6207 @option{--files-from}.
6212 Only consider @code{NUL} terminated file names, instead of files that
6213 terminate in a newline.
6216 The @option{--null} option is just like the one in @acronym{GNU}
6217 @command{xargs} and @command{cpio}, and is useful with the
6218 @option{-print0} predicate of @acronym{GNU} @command{find}. In
6219 @command{tar}, @option{--null} also disables special handling for
6220 file names that begin with dash.
6222 This example shows how to use @command{find} to generate a list of files
6223 larger than 800K in length and put that list into a file called
6224 @file{long-files}. The @option{-print0} option to @command{find} is just
6225 like @option{-print}, except that it separates files with a @code{NUL}
6226 rather than with a newline. You can then run @command{tar} with both the
6227 @option{--null} and @option{-T} options to specify that @command{tar} get the
6228 files from that file, @file{long-files}, to create the archive
6229 @file{big.tgz}. The @option{--null} option to @command{tar} will cause
6230 @command{tar} to recognize the @code{NUL} separator between files.
6233 $ @kbd{find . -size +800 -print0 > long-files}
6234 $ @kbd{tar -c -v --null --files-from=long-files --file=big.tar}
6237 @FIXME{say anything else here to conclude the section?}
6240 @section Excluding Some Files
6243 @cindex File names, excluding files by
6244 @cindex Excluding files by name and pattern
6245 @cindex Excluding files by file system
6246 To avoid operating on files whose names match a particular pattern,
6247 use the @option{--exclude} or @option{--exclude-from} options.
6251 @item --exclude=@var{pattern}
6252 Causes @command{tar} to ignore files that match the @var{pattern}.
6256 The @option{--exclude=@var{pattern}} option prevents any file or
6257 member whose name matches the shell wildcard (@var{pattern}) from
6259 For example, to create an archive with all the contents of the directory
6260 @file{src} except for files whose names end in @file{.o}, use the
6261 command @samp{tar -cf src.tar --exclude='*.o' src}.
6263 You may give multiple @option{--exclude} options.
6266 @opindex exclude-from
6267 @item --exclude-from=@var{file}
6268 @itemx -X @var{file}
6269 Causes @command{tar} to ignore files that match the patterns listed in
6273 @findex exclude-from
6274 Use the @option{--exclude-from} option to read a
6275 list of patterns, one per line, from @var{file}; @command{tar} will
6276 ignore files matching those patterns. Thus if @command{tar} is
6277 called as @w{@samp{tar -c -X foo .}} and the file @file{foo} contains a
6278 single line @file{*.o}, no files whose names end in @file{.o} will be
6279 added to the archive.
6282 @opindex exclude-caches
6283 @item --exclude-caches
6284 Causes @command{tar} to ignore directories containing a cache directory tag.
6287 @findex exclude-caches
6288 When creating an archive, the @option{--exclude-caches} option causes
6289 @command{tar} to exclude all directories that contain a @dfn{cache
6290 directory tag}. A cache directory tag is a short file with the
6291 well-known name @file{CACHEDIR.TAG} and having a standard header
6292 specified in @url{http://www.brynosaurus.com/cachedir/spec.html}.
6293 Various applications write cache directory tags into directories they
6294 use to hold regenerable, non-precious data, so that such data can be
6295 more easily excluded from backups.
6298 * problems with exclude::
6301 @node problems with exclude
6302 @unnumberedsubsec Problems with Using the @code{exclude} Options
6304 @opindex exclude, potential problems with
6305 Some users find @samp{exclude} options confusing. Here are some common
6310 The main operating mode of @command{tar} does not act on a path name
6311 explicitly listed on the command line if one of its file name
6312 components is excluded. In the example above, if
6313 you create an archive and exclude files that end with @samp{*.o}, but
6314 explicitly name the file @samp{dir.o/foo} after all the options have been
6315 listed, @samp{dir.o/foo} will be excluded from the archive.
6318 You can sometimes confuse the meanings of @option{--exclude} and
6319 @option{--exclude-from}. Be careful: use @option{--exclude} when files
6320 to be excluded are given as a pattern on the command line. Use
6321 @option{--exclude-from} to introduce the name of a file which contains
6322 a list of patterns, one per line; each of these patterns can exclude
6323 zero, one, or many files.
6326 When you use @option{--exclude=@var{pattern}}, be sure to quote the
6327 @var{pattern} parameter, so @GNUTAR{} sees wildcard characters
6328 like @samp{*}. If you do not do this, the shell might expand the
6329 @samp{*} itself using files at hand, so @command{tar} might receive a
6330 list of files instead of one pattern, or none at all, making the
6331 command somewhat illegal. This might not correspond to what you want.
6336 $ @kbd{tar -c -f @var{archive.tar} --exclude '*.o' @var{directory}}
6344 $ @kbd{tar -c -f @var{archive.tar} --exclude *.o @var{directory}}
6348 You must use use shell syntax, or globbing, rather than @code{regexp}
6349 syntax, when using exclude options in @command{tar}. If you try to use
6350 @code{regexp} syntax to describe files to be excluded, your command
6354 @FIXME{The change in semantics must have occurred before 1.11,
6355 so I doubt if it is worth mentioning at all. Anyway, should at
6356 least specify in which version the semantics changed.}
6357 In earlier versions of @command{tar}, what is now the
6358 @option{--exclude-from} option was called @option{--exclude} instead.
6359 Now, @option{--exclude} applies to patterns listed on the command
6360 line and @option{--exclude-from} applies to patterns listed in a
6366 @section Wildcards Patterns and Matching
6368 @dfn{Globbing} is the operation by which @dfn{wildcard} characters,
6369 @samp{*} or @samp{?} for example, are replaced and expanded into all
6370 existing files matching the given pattern. @GNUTAR{} can use wildcard
6371 patterns for matching (or globbing) archive members when extracting
6372 from or listing an archive. Wildcard patterns are also used for
6373 verifying volume labels of @command{tar} archives. This section has the
6374 purpose of explaining wildcard syntax for @command{tar}.
6376 @FIXME{the next few paragraphs need work.}
6378 A @var{pattern} should be written according to shell syntax, using wildcard
6379 characters to effect globbing. Most characters in the pattern stand
6380 for themselves in the matched string, and case is significant: @samp{a}
6381 will match only @samp{a}, and not @samp{A}. The character @samp{?} in the
6382 pattern matches any single character in the matched string. The character
6383 @samp{*} in the pattern matches zero, one, or more single characters in
6384 the matched string. The character @samp{\} says to take the following
6385 character of the pattern @emph{literally}; it is useful when one needs to
6386 match the @samp{?}, @samp{*}, @samp{[} or @samp{\} characters, themselves.
6388 The character @samp{[}, up to the matching @samp{]}, introduces a character
6389 class. A @dfn{character class} is a list of acceptable characters
6390 for the next single character of the matched string. For example,
6391 @samp{[abcde]} would match any of the first five letters of the alphabet.
6392 Note that within a character class, all of the ``special characters''
6393 listed above other than @samp{\} lose their special meaning; for example,
6394 @samp{[-\\[*?]]} would match any of the characters, @samp{-}, @samp{\},
6395 @samp{[}, @samp{*}, @samp{?}, or @samp{]}. (Due to parsing constraints,
6396 the characters @samp{-} and @samp{]} must either come @emph{first} or
6397 @emph{last} in a character class.)
6399 @cindex Excluding characters from a character class
6400 @cindex Character class, excluding characters from
6401 If the first character of the class after the opening @samp{[}
6402 is @samp{!} or @samp{^}, then the meaning of the class is reversed.
6403 Rather than listing character to match, it lists those characters which
6404 are @emph{forbidden} as the next single character of the matched string.
6406 Other characters of the class stand for themselves. The special
6407 construction @samp{[@var{a}-@var{e}]}, using an hyphen between two
6408 letters, is meant to represent all characters between @var{a} and
6411 @FIXME{need to add a sentence or so here to make this clear for those
6412 who don't have dan around.}
6414 Periods (@samp{.}) or forward slashes (@samp{/}) are not considered
6415 special for wildcard matches. However, if a pattern completely matches
6416 a directory prefix of a matched string, then it matches the full matched
6417 string: thus, excluding a directory also excludes all the files beneath it.
6420 * controlling pattern-matching::
6423 @node controlling pattern-matching
6424 @unnumberedsubsec Controlling Pattern-Matching
6426 For the purposes of this section, we call @dfn{exclusion members} all
6427 member names obtained while processing @option{--exclude} and
6428 @option{--exclude-from} options, and @dfn{inclusion members} those
6429 member names that were given in the command line or read from the file
6430 specified with @option{--files-from} option.
6432 These two pairs of member lists are used in the following operations:
6433 @option{--diff}, @option{--extract}, @option{--list},
6436 There are no inclusion members in create mode (@option{--create} and
6437 @option{--append}), since in this mode the names obtained from the
6438 command line refer to @emph{files}, not archive members.
6440 By default, inclusion members are compared with archive members
6441 literally @footnote{Notice that earlier @GNUTAR{} versions used
6442 globbing for inclusion members, which contradicted to UNIX98
6443 specification and was not documented. @xref{Changes}, for more
6444 information on this and other changes} and exclusion members are
6445 treated as globbing patterns. For example:
6449 $ @kbd{tar tf foo.tar}
6454 # @i{Member names are used verbatim:}
6455 $ @kbd{tar -xf foo.tar -v '[remarks]'}
6457 # @i{Exclude member names are globbed:}
6458 $ @kbd{tar -xf foo.tar -v --exclude '*.c'}
6464 This behavior can be altered by using the following options:
6469 Treat all member names as wildcards.
6471 @opindex no-wildcards
6472 @item --no-wildcards
6473 Treat all member names as literal strings.
6476 Thus, to extract files whose names end in @samp{.c}, you can use:
6479 $ @kbd{tar -xf foo.tar -v --wildcards '*.c'}
6485 Notice quoting of the pattern to prevent the shell from interpreting
6488 The effect of @option{--wildcards} option is cancelled by
6489 @option{--no-wildcards}. This can be used to pass part of
6490 the command line arguments verbatim and other part as globbing
6491 patterns. For example, the following invocation:
6494 $ @kbd{tar -xf foo.tar --wildcards '*.txt' --no-wildcards '[remarks]'}
6498 instructs @command{tar} to extract from @file{foo.tar} all files whose
6499 names end in @samp{.txt} and the file named @file{[remarks]}.
6501 Normally, a pattern matches a name if an initial subsequence of the
6502 name's components matches the pattern, where @samp{*}, @samp{?}, and
6503 @samp{[...]} are the usual shell wildcards, @samp{\} escapes wildcards,
6504 and wildcards can match @samp{/}.
6506 Other than optionally stripping leading @samp{/} from names
6507 (@pxref{absolute}), patterns and names are used as-is. For
6508 example, trailing @samp{/} is not trimmed from a user-specified name
6509 before deciding whether to exclude it.
6511 However, this matching procedure can be altered by the options listed
6512 below. These options accumulate. For example:
6515 --ignore-case --exclude='makefile' --no-ignore-case ---exclude='readme'
6518 ignores case when excluding @samp{makefile}, but not when excluding
6523 @opindex no-anchored
6525 @itemx --no-anchored
6526 If anchored, a pattern must match an initial subsequence
6527 of the name's components. Otherwise, the pattern can match any
6528 subsequence. Default is @option{--no-anchored} for exclusion members
6529 and @option{--anchored} inclusion members.
6531 @opindex ignore-case
6532 @opindex no-ignore-case
6534 @itemx --no-ignore-case
6535 When ignoring case, upper-case patterns match lower-case names and vice versa.
6536 When not ignoring case (the default), matching is case-sensitive.
6538 @opindex wildcards-match-slash
6539 @opindex no-wildcards-match-slash
6540 @item --wildcards-match-slash
6541 @itemx --no-wildcards-match-slash
6542 When wildcards match slash (the default for exclusion members), a
6543 wildcard like @samp{*} in the pattern can match a @samp{/} in the
6544 name. Otherwise, @samp{/} is matched only by @samp{/}.
6548 The @option{--recursion} and @option{--no-recursion} options
6549 (@pxref{recurse}) also affect how member patterns are interpreted. If
6550 recursion is in effect, a pattern matches a name if it matches any of
6551 the name's parent directories.
6553 The following table summarizes pattern-matching default values:
6555 @multitable @columnfractions .3 .7
6556 @headitem Members @tab Default settings
6557 @item Inclusion @tab @option{--no-wildcards --anchored --no-wildcards-match-slash}
6558 @item Exclusion @tab @option{--wildcards --no-anchored --wildcards-match-slash}
6561 @node quoting styles
6562 @section Quoting Member Names
6564 When displaying member names, @command{tar} takes care to avoid
6565 ambiguities caused by certain characters. This is called @dfn{name
6566 quoting}. The characters in question are:
6569 @item Non-printable control characters:
6571 @multitable @columnfractions 0.20 0.10 0.60
6572 @headitem Character @tab ASCII @tab Character name
6573 @item \a @tab 7 @tab Audible bell
6574 @item \b @tab 8 @tab Backspace
6575 @item \f @tab 12 @tab Form feed
6576 @item \n @tab 10 @tab New line
6577 @item \r @tab 13 @tab Carriage return
6578 @item \t @tab 9 @tab Horizontal tabulation
6579 @item \v @tab 11 @tab Vertical tabulation
6582 @item Space (ASCII 32)
6584 @item Single and double quotes (@samp{'} and @samp{"})
6586 @item Backslash (@samp{\})
6589 The exact way @command{tar} uses to quote these characters depends on
6590 the @dfn{quoting style}. The default quoting style, called
6591 @dfn{escape} (see below), uses backslash notation to represent control
6592 characters, space and backslash. Using this quoting style, control
6593 characters are represented as listed in column @samp{Character} in the
6594 above table, a space is printed as @samp{\ } and a backslash as @samp{\\}.
6596 @GNUTAR{} offers seven distinct quoting styles, which can be selected
6597 using @option{--quoting-style} option:
6600 @item --quoting-style=@var{style}
6601 @opindex quoting-style
6603 Sets quoting style. Valid values for @var{style} argument are:
6604 literal, shell, shell-always, c, escape, locale, clocale.
6607 These styles are described in detail below. To illustrate their
6608 effect, we will use an imaginary tar archive @file{arch.tar}
6609 containing the following members:
6613 # 1. Contains horizontal tabulation character.
6615 # 2. Contains newline character
6618 # 3. Contains a space
6620 # 4. Contains double quotes
6622 # 5. Contains single quotes
6624 # 6. Contains a backslash character:
6629 Here is how usual @command{ls} command would have listed them, if they
6630 had existed in the current working directory:
6648 No quoting, display each character as is:
6652 $ @kbd{tar tf arch.tar --quoting-style=literal}
6665 Display characters the same way Bourne shell does:
6666 control characters, except @samp{\t} and @samp{\n}, are printed using
6667 backslash escapes, @samp{\t} and @samp{\n} are printed as is, and a
6668 single quote is printed as @samp{\'}. If a name contains any quoted
6669 characters, it is enclosed in single quotes. In particular, if a name
6670 contains single quotes, it is printed as several single-quoted strings:
6674 $ @kbd{tar tf arch.tar --quoting-style=shell}
6677 './a'\''single'\''quote'
6687 Same as @samp{shell}, but the names are always enclosed in single
6692 $ @kbd{tar tf arch.tar --quoting-style=shell-always}
6695 './a'\''single'\''quote'
6705 Use the notation of the C programming language. All names are
6706 enclosed in double quotes. Control characters are quoted using
6707 backslash notations, double quotes are represented as @samp{\"},
6708 backslash characters are represented as @samp{\\}. Single quotes and
6709 spaces are not quoted:
6713 $ @kbd{tar tf arch.tar --quoting-style=c}
6717 "./a\"double\"quote"
6725 Control characters are printed using backslash notation, a space is
6726 printed as @samp{\ } and a backslash as @samp{\\}. This is the
6727 default quoting style, unless it was changed when configured the
6732 $ @kbd{tar tf arch.tar --quoting-style=escape}
6744 Control characters, single quote and backslash are printed using
6745 backslash notation. All names are quoted using left and right
6746 quotation marks, appropriate to the current locale. If it does not
6747 define quotation marks, use @samp{`} as left and @samp{'} as right
6748 quotation marks. Any occurrences of the right quotation mark in a
6749 name are escaped with @samp{\}, for example:
6755 $ @kbd{tar tf arch.tar --quoting-style=locale}
6758 `./a\'single\'quote'
6767 Same as @samp{locale}, but @samp{"} is used for both left and right
6768 quotation marks, if not provided by the currently selected locale:
6772 $ @kbd{tar tf arch.tar --quoting-style=clocale}
6776 "./a\"double\"quote"
6784 You can specify which characters should be quoted in addition to those
6785 implied by the current quoting style:
6788 @item --quote-chars=@var{string}
6789 Always quote characters from @var{string}, even if the selected
6790 quoting style would not quote them.
6793 For example, using @samp{escape} quoting (compare with the usual
6794 escape listing above):
6798 $ @kbd{tar tf arch.tar --quoting-style=escape --quote-chars=' "'}
6809 To disable quoting of such additional characters, use the following
6813 @item --no-quote-chars=@var{string}
6814 Remove characters listed in @var{string} from the list of quoted
6815 characters set by the previous @option{--quote-chars} option.
6818 This option is particularly useful if you have added
6819 @option{--quote-chars} to your @env{TAR_OPTIONS} (@pxref{TAR_OPTIONS})
6820 and wish to disable it for the current invocation.
6822 Note, that @option{--no-quote-chars} does @emph{not} disable those
6823 characters that are quoted by default in the selected quoting style.
6826 @section Modifying File and Member Names
6828 @command{Tar} archives contain detailed information about files stored
6829 in them and full file names are part of that information. When
6830 storing file to an archive, its file name is recorded in the archive
6831 along with the actual file contents. When restoring from an archive,
6832 a file is created on disk with exactly the same name as that stored
6833 in the archive. In the majority of cases this is the desired behavior
6834 of a file archiver. However, there are some cases when it is not.
6836 First of all, it is often unsafe to extract archive members with
6837 absolute file names or those that begin with a @file{../}. @GNUTAR{}
6838 takes special precautions when extracting such names and provides a
6839 special option for handling them, which is described in
6842 Secondly, you may wish to extract file names without some leading
6843 directory components, or with otherwise modified names. In other
6844 cases it is desirable to store files under differing names in the
6847 @GNUTAR{} provides two options for these needs.
6850 @opindex strip-components
6851 @item --strip-components=@var{number}
6852 Strip given @var{number} of leading components from file names before
6856 For example, suppose you have archived whole @file{/usr} hierarchy to
6857 a tar archive named @file{usr.tar}. Among other files, this archive
6858 contains @file{usr/include/stdlib.h}, which you wish to extract to
6859 the current working directory. To do so, you type:
6862 $ @kbd{tar -xf usr.tar --strip=2 usr/include/stdlib.h}
6865 The option @option{--strip=2} instructs @command{tar} to strip the
6866 two leading components (@file{usr/} and @file{include/}) off the file
6869 If you add to the above invocation @option{--verbose} (@option{-v})
6870 option, you will note that the verbose listing still contains the
6871 full file name, with the two removed components still in place. This
6872 can be inconvenient, so @command{tar} provides a special option for
6873 altering this behavior:
6875 @anchor{show-transformed-names}
6877 @opindex --show-transformed-names
6878 @item --show-transformed-names
6879 Display file or member names with all requested transformations
6887 $ @kbd{tar -xf usr.tar -v --strip=2 usr/include/stdlib.h}
6888 usr/include/stdlib.h
6889 $ @kbd{tar -xf usr.tar -v --strip=2 --show-transformed usr/include/stdlib.h}
6894 Notice that in both cases the file is @file{stdlib.h} extracted to the
6895 current working directory, @option{--show-transformed-names} affects
6896 only the way its name is displayed.
6898 This option is especially useful for verifying whether the invocation
6899 will have the desired effect. Thus, before running
6902 $ @kbd{tar -x --strip=@var{n}}
6906 it is often advisable to run
6909 $ @kbd{tar -t -v --show-transformed --strip=@var{n}}
6913 to make sure the command will produce the intended results.
6915 In case you need to apply more complex modifications to the file name,
6916 @GNUTAR{} provides a general-purpose transformation option:
6919 @opindex --transform
6920 @item --transform=@var{expression}
6921 Modify file names using supplied @var{expression}.
6925 The @var{expression} is a @command{sed}-like replace expression of the
6929 s/@var{regexp}/@var{replace}/[@var{flags}]
6933 where @var{regexp} is a @dfn{regular expression}, @var{replace} is a
6934 replacement for each file name part that matches @var{regexp}. Both
6935 @var{regexp} and @var{replace} are described in detail in
6936 @ref{The "s" Command, The "s" Command, The `s' Command, sed, GNU sed}.
6938 Supported @var{flags} are:
6942 Apply the replacement to @emph{all} matches to the @var{regexp}, not
6946 Use case-insensitive matching
6949 @var{regexp} is an @dfn{extended regular expression} (@pxref{Extended
6950 regexps, Extended regular expressions, Extended regular expressions,
6954 Only replace the @var{number}th match of the @var{regexp}.
6956 Note: the @var{posix} standard does not specify what should happen
6957 when you mix the @samp{g} and @var{number} modifiers. @GNUTAR{}
6958 follows the GNU @command{sed} implementation in this regard, so
6959 the the interaction is defined to be: ignore matches before the
6960 @var{number}th, and then match and replace all matches from the
6965 Any delimiter can be used in lieue of @samp{/}, the only requirement being
6966 that it be used consistently throughout the expression. For example,
6967 the following two expressions are equivalent:
6976 Changing of delimiter is often useful when the @var{regex} contains
6977 slashes. For example, it is more convenient to write:
6990 Here are several examples of @option{--transform} usage:
6993 @item Extract @file{usr/} hierarchy into @file{usr/local/}:
6996 $ @kbd{tar --transform='s,usr/,usr/local/,' -x -f arch.tar}
6999 @item Strip two leading directory components (equivalent to
7000 @option{--strip-components=2}):
7003 $ @kbd{tar --transform='s,/*[^/]*/[^/]*/,,' -x -f arch.tar}
7006 @item Prepend @file{/prefix/} to each file name:
7009 $ @kbd{tar --transform 's,^,/prefix/,' -x -f arch.tar}
7012 @item Convert each file name to lower case:
7015 $ @kbd{tar --transform 's/.*/\L&/' -x -f arch.tar}
7020 Unlike @option{--strip-components}, @option{--transform} can be used
7021 in any @GNUTAR{} operation mode. For example, the following command
7022 adds files to the archive while replacing the leading @file{usr/}
7023 component with @file{var/}:
7026 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' /}
7029 To test @option{--transform} effect we suggest to use
7030 @option{--show-transformed-names}:
7033 $ @kbd{tar -cf arch.tar --transform='s,^usr/,var/,' \
7034 --verbose --show-transformed-names /}
7037 If both @option{--strip-components} and @option{--transform} are used
7038 together, then @option{--transform} is applied first, and the required
7039 number of components is then stripped from its result.
7042 @section Operating Only on New Files
7045 @cindex Excluding file by age
7046 @cindex Data Modification time, excluding files by
7047 @cindex Modification time, excluding files by
7048 @cindex Age, excluding files by
7049 The @option{--after-date=@var{date}} (@option{--newer=@var{date}},
7050 @option{-N @var{date}}) option causes @command{tar} to only work on
7051 files whose data modification or status change times are newer than
7052 the @var{date} given. If @var{date} starts with @samp{/} or @samp{.},
7053 it is taken to be a file name; the data modification time of that file
7054 is used as the date. If you use this option when creating or appending
7055 to an archive, the archive will only include new files. If you use
7056 @option{--after-date} when extracting an archive, @command{tar} will
7057 only extract files newer than the @var{date} you specify.
7059 If you only want @command{tar} to make the date comparison based on
7060 modification of the file's data (rather than status
7061 changes), then use the @option{--newer-mtime=@var{date}} option.
7063 You may use these options with any operation. Note that these options
7064 differ from the @option{--update} (@option{-u}) operation in that they
7065 allow you to specify a particular date against which @command{tar} can
7066 compare when deciding whether or not to archive the files.
7071 @item --after-date=@var{date}
7072 @itemx --newer=@var{date}
7073 @itemx -N @var{date}
7074 Only store files newer than @var{date}.
7076 Acts on files only if their data modification or status change times are
7077 later than @var{date}. Use in conjunction with any operation.
7079 If @var{date} starts with @samp{/} or @samp{.}, it is taken to be a file
7080 name; the data modification time of that file is used as the date.
7082 @opindex newer-mtime
7083 @item --newer-mtime=@var{date}
7084 Acts like @option{--after-date}, but only looks at data modification times.
7087 These options limit @command{tar} to operate only on files which have
7088 been modified after the date specified. A file's status is considered to have
7089 changed if its contents have been modified, or if its owner,
7090 permissions, and so forth, have been changed. (For more information on
7091 how to specify a date, see @ref{Date input formats}; remember that the
7092 entire date argument must be quoted if it contains any spaces.)
7094 Gurus would say that @option{--after-date} tests both the data
7095 modification time (@code{mtime}, the time the contents of the file
7096 were last modified) and the status change time (@code{ctime}, the time
7097 the file's status was last changed: owner, permissions, etc.@:)
7098 fields, while @option{--newer-mtime} tests only the @code{mtime}
7101 To be precise, @option{--after-date} checks @emph{both} @code{mtime} and
7102 @code{ctime} and processes the file if either one is more recent than
7103 @var{date}, while @option{--newer-mtime} only checks @code{mtime} and
7104 disregards @code{ctime}. Neither does it use @code{atime} (the last time the
7105 contents of the file were looked at).
7107 Date specifiers can have embedded spaces. Because of this, you may need
7108 to quote date arguments to keep the shell from parsing them as separate
7109 arguments. For example, the following command will add to the archive
7110 all the files modified less than two days ago:
7113 $ @kbd{tar -cf foo.tar --newer-mtime '2 days ago'}
7117 @strong{Please Note:} @option{--after-date} and @option{--newer-mtime}
7118 should not be used for incremental backups. @xref{Incremental Dumps},
7119 for proper way of creating incremental backups.
7123 @section Descending into Directories
7125 @cindex Avoiding recursion in directories
7126 @cindex Descending directories, avoiding
7127 @cindex Directories, avoiding recursion
7128 @cindex Recursion in directories, avoiding
7130 @FIXME{arrggh! this is still somewhat confusing to me. :-< }
7132 Usually, @command{tar} will recursively explore all directories (either
7133 those given on the command line or through the @option{--files-from}
7134 option) for the various files they contain. However, you may not always
7135 want @command{tar} to act this way.
7137 @opindex no-recursion
7138 The @option{--no-recursion} option inhibits @command{tar}'s recursive descent
7139 into specified directories. If you specify @option{--no-recursion}, you can
7140 use the @command{find} utility for hunting through levels of directories to
7141 construct a list of file names which you could then pass to @command{tar}.
7142 @command{find} allows you to be more selective when choosing which files to
7143 archive; see @ref{files}, for more information on using @command{find} with
7144 @command{tar}, or look.
7147 @item --no-recursion
7148 Prevents @command{tar} from recursively descending directories.
7152 Requires @command{tar} to recursively descend directories.
7153 This is the default.
7156 When you use @option{--no-recursion}, @GNUTAR{} grabs
7157 directory entries themselves, but does not descend on them
7158 recursively. Many people use @command{find} for locating files they
7159 want to back up, and since @command{tar} @emph{usually} recursively
7160 descends on directories, they have to use the @samp{@w{-not -type d}}
7161 test in their @command{find} invocation (@pxref{Type, Type, Type test,
7162 find, Finding Files}), as they usually do not want all the files in a
7163 directory. They then use the @option{--files-from} option to archive
7164 the files located via @command{find}.
7166 The problem when restoring files archived in this manner is that the
7167 directories themselves are not in the archive; so the
7168 @option{--same-permissions} (@option{--preserve-permissions},
7169 @option{-p}) option does not affect them---while users might really
7170 like it to. Specifying @option{--no-recursion} is a way to tell
7171 @command{tar} to grab only the directory entries given to it, adding
7172 no new files on its own. To summarize, if you use @command{find} to
7173 create a list of files to be stored in an archive, use it as follows:
7177 $ @kbd{find @var{dir} @var{tests} | \
7178 tar -cf @var{archive} -T - --no-recursion}
7182 The @option{--no-recursion} option also applies when extracting: it
7183 causes @command{tar} to extract only the matched directory entries, not
7184 the files under those directories.
7186 The @option{--no-recursion} option also affects how globbing patterns
7187 are interpreted (@pxref{controlling pattern-matching}).
7189 The @option{--no-recursion} and @option{--recursion} options apply to
7190 later options and operands, and can be overridden by later occurrences
7191 of @option{--no-recursion} and @option{--recursion}. For example:
7194 $ @kbd{tar -cf jams.tar --no-recursion grape --recursion grape/concord}
7198 creates an archive with one entry for @file{grape}, and the recursive
7199 contents of @file{grape/concord}, but no entries under @file{grape}
7200 other than @file{grape/concord}.
7203 @section Crossing File System Boundaries
7204 @cindex File system boundaries, not crossing
7207 @command{tar} will normally automatically cross file system boundaries in
7208 order to archive files which are part of a directory tree. You can
7209 change this behavior by running @command{tar} and specifying
7210 @option{--one-file-system}. This option only affects files that are
7211 archived because they are in a directory that is being archived;
7212 @command{tar} will still archive files explicitly named on the command line
7213 or through @option{--files-from}, regardless of where they reside.
7216 @opindex one-file-system
7217 @item --one-file-system
7218 Prevents @command{tar} from crossing file system boundaries when
7219 archiving. Use in conjunction with any write operation.
7222 The @option{--one-file-system} option causes @command{tar} to modify its
7223 normal behavior in archiving the contents of directories. If a file in
7224 a directory is not on the same file system as the directory itself, then
7225 @command{tar} will not archive that file. If the file is a directory
7226 itself, @command{tar} will not archive anything beneath it; in other words,
7227 @command{tar} will not cross mount points.
7229 This option is useful for making full or incremental archival backups of
7230 a file system. If this option is used in conjunction with
7231 @option{--verbose} (@option{-v}), files that are excluded are
7232 mentioned by name on the standard error.
7235 * directory:: Changing Directory
7236 * absolute:: Absolute File Names
7240 @subsection Changing the Working Directory
7243 @FIXME{need to read over this node now for continuity; i've switched
7244 things around some.}
7246 @cindex Changing directory mid-stream
7247 @cindex Directory, changing mid-stream
7248 @cindex Working directory, specifying
7249 To change the working directory in the middle of a list of file names,
7250 either on the command line or in a file specified using
7251 @option{--files-from} (@option{-T}), use @option{--directory} (@option{-C}).
7252 This will change the working directory to the specified directory
7253 after that point in the list.
7257 @item --directory=@var{directory}
7258 @itemx -C @var{directory}
7259 Changes the working directory in the middle of a command line.
7265 $ @kbd{tar -c -f jams.tar grape prune -C food cherry}
7269 will place the files @file{grape} and @file{prune} from the current
7270 directory into the archive @file{jams.tar}, followed by the file
7271 @file{cherry} from the directory @file{food}. This option is especially
7272 useful when you have several widely separated files that you want to
7273 store in the same archive.
7275 Note that the file @file{cherry} is recorded in the archive under the
7276 precise name @file{cherry}, @emph{not} @file{food/cherry}. Thus, the
7277 archive will contain three files that all appear to have come from the
7278 same directory; if the archive is extracted with plain @samp{tar
7279 --extract}, all three files will be written in the current directory.
7281 Contrast this with the command,
7284 $ @kbd{tar -c -f jams.tar grape prune -C food red/cherry}
7288 which records the third file in the archive under the name
7289 @file{red/cherry} so that, if the archive is extracted using
7290 @samp{tar --extract}, the third file will be written in a subdirectory
7291 named @file{orange-colored}.
7293 You can use the @option{--directory} option to make the archive
7294 independent of the original name of the directory holding the files.
7295 The following command places the files @file{/etc/passwd},
7296 @file{/etc/hosts}, and @file{/lib/libc.a} into the archive
7300 $ @kbd{tar -c -f foo.tar -C /etc passwd hosts -C /lib libc.a}
7304 However, the names of the archive members will be exactly what they were
7305 on the command line: @file{passwd}, @file{hosts}, and @file{libc.a}.
7306 They will not appear to be related by file name to the original
7307 directories where those files were located.
7309 Note that @option{--directory} options are interpreted consecutively. If
7310 @option{--directory} specifies a relative file name, it is interpreted
7311 relative to the then current directory, which might not be the same as
7312 the original current working directory of @command{tar}, due to a previous
7313 @option{--directory} option.
7315 When using @option{--files-from} (@pxref{files}), you can put various
7316 @command{tar} options (including @option{-C}) in the file list. Notice,
7317 however, that in this case the option and its argument may not be
7318 separated by whitespace. If you use short option, its argument must
7319 either follow the option letter immediately, without any intervening
7320 whitespace, or occupy the next line. Otherwise, if you use long
7321 option, separate its argument by an equal sign.
7323 For instance, the file list for the above example will be:
7338 To use it, you would invoke @command{tar} as follows:
7341 $ @kbd{tar -c -f foo.tar --files-from list}
7344 Notice also that you can only use the short option variant in the file
7345 list, i.e., always use @option{-C}, not @option{--directory}.
7347 The interpretation of @option{--directory} is disabled by
7348 @option{--null} option.
7351 @subsection Absolute File Names
7355 @opindex absolute-names
7356 @item --absolute-names
7358 Do not strip leading slashes from file names, and permit file names
7359 containing a @file{..} file name component.
7362 By default, @GNUTAR{} drops a leading @samp{/} on
7363 input or output, and complains about file names containing a @file{..}
7364 component. This option turns off this behavior.
7366 When @command{tar} extracts archive members from an archive, it strips any
7367 leading slashes (@samp{/}) from the member name. This causes absolute
7368 member names in the archive to be treated as relative file names. This
7369 allows you to have such members extracted wherever you want, instead of
7370 being restricted to extracting the member in the exact directory named
7371 in the archive. For example, if the archive member has the name
7372 @file{/etc/passwd}, @command{tar} will extract it as if the name were
7373 really @file{etc/passwd}.
7375 File names containing @file{..} can cause problems when extracting, so
7376 @command{tar} normally warns you about such files when creating an
7377 archive, and rejects attempts to extracts such files.
7379 Other @command{tar} programs do not do this. As a result, if you
7380 create an archive whose member names start with a slash, they will be
7381 difficult for other people with a non-@GNUTAR{}
7382 program to use. Therefore, @GNUTAR{} also strips
7383 leading slashes from member names when putting members into the
7384 archive. For example, if you ask @command{tar} to add the file
7385 @file{/bin/ls} to an archive, it will do so, but the member name will
7386 be @file{bin/ls}.@footnote{A side effect of this is that when
7387 @option{--create} is used with @option{--verbose} the resulting output
7388 is not, generally speaking, the same as the one you'd get running
7389 @kbd{tar --list} command. This may be important if you use some
7390 scripts for comparing both outputs. @xref{listing member and file names},
7391 for the information on how to handle this case.}
7393 If you use the @option{--absolute-names} (@option{-P}) option,
7394 @command{tar} will do none of these transformations.
7396 To archive or extract files relative to the root directory, specify
7397 the @option{--absolute-names} (@option{-P}) option.
7399 Normally, @command{tar} acts on files relative to the working
7400 directory---ignoring superior directory names when archiving, and
7401 ignoring leading slashes when extracting.
7403 When you specify @option{--absolute-names} (@option{-P}),
7404 @command{tar} stores file names including all superior directory
7405 names, and preserves leading slashes. If you only invoked
7406 @command{tar} from the root directory you would never need the
7407 @option{--absolute-names} option, but using this option
7408 may be more convenient than switching to root.
7410 @FIXME{Should be an example in the tutorial/wizardry section using this
7411 to transfer files between systems.}
7413 @FIXME{Is write access an issue?}
7416 @item --absolute-names
7417 Preserves full file names (including superior directory names) when
7418 archiving files. Preserves leading slash when extracting files.
7422 @FIXME{this is still horrible; need to talk with dan on monday.}
7424 @command{tar} prints out a message about removing the @samp{/} from
7425 file names. This message appears once per @GNUTAR{}
7426 invocation. It represents something which ought to be told; ignoring
7427 what it means can cause very serious surprises, later.
7429 Some people, nevertheless, do not want to see this message. Wanting to
7430 play really dangerously, one may of course redirect @command{tar} standard
7431 error to the sink. For example, under @command{sh}:
7434 $ @kbd{tar -c -f archive.tar /home 2> /dev/null}
7438 Another solution, both nicer and simpler, would be to change to
7439 the @file{/} directory first, and then avoid absolute notation.
7443 $ @kbd{(cd / && tar -c -f archive.tar home)}
7445 $ @kbd{tar -c -f archive.tar -C / home}
7448 @include getdate.texi
7451 @chapter Controlling the Archive Format
7453 @cindex Tar archive formats
7454 Due to historical reasons, there are several formats of tar archives.
7455 All of them are based on the same principles, but have some subtle
7456 differences that often make them incompatible with each other.
7458 GNU tar is able to create and handle archives in a variety of formats.
7459 The most frequently used formats are (in alphabetical order):
7463 Format used by @GNUTAR{} versions up to 1.13.25. This format derived
7464 from an early @acronym{POSIX} standard, adding some improvements such as
7465 sparse file handling and incremental archives. Unfortunately these
7466 features were implemented in a way incompatible with other archive
7469 Archives in @samp{gnu} format are able to hold pathnames of unlimited
7473 Format used by @GNUTAR{} of versions prior to 1.12.
7476 Archive format, compatible with the V7 implementation of tar. This
7477 format imposes a number of limitations. The most important of them
7481 @item The maximum length of a file name is limited to 99 characters.
7482 @item The maximum length of a symbolic link is limited to 99 characters.
7483 @item It is impossible to store special files (block and character
7484 devices, fifos etc.)
7485 @item Maximum value of user or group ID is limited to 2097151 (7777777
7487 @item V7 archives do not contain symbolic ownership information (user
7488 and group name of the file owner).
7491 This format has traditionally been used by Automake when producing
7492 Makefiles. This practice will change in the future, in the meantime,
7493 however this means that projects containing filenames more than 99
7494 characters long will not be able to use @GNUTAR{} @value{VERSION} and
7495 Automake prior to 1.9.
7498 Archive format defined by @acronym{POSIX.1-1988} specification. It stores
7499 symbolic ownership information. It is also able to store
7500 special files. However, it imposes several restrictions as well:
7503 @item The maximum length of a file name is limited to 256 characters,
7504 provided that the filename can be split at directory separator in
7505 two parts, first of them being at most 155 bytes long. So, in most
7506 cases the maximum file name length will be shorter than 256
7508 @item The maximum length of a symbolic link name is limited to
7510 @item Maximum size of a file the archive is able to accomodate
7512 @item Maximum value of UID/GID is 2097151.
7513 @item Maximum number of bits in device major and minor numbers is 21.
7517 Format used by J@"org Schilling @command{star}
7518 implementation. @GNUTAR{} is able to read @samp{star} archives but
7519 currently does not produce them.
7522 Archive format defined by @acronym{POSIX.1-2001} specification. This is the
7523 most flexible and feature-rich format. It does not impose any
7524 restrictions on file sizes or filename lengths. This format is quite
7525 recent, so not all tar implementations are able to handle it properly.
7526 However, this format is designed in such a way that any tar
7527 implementation able to read @samp{ustar} archives will be able to read
7528 most @samp{posix} archives as well, with the only exception that any
7529 additional information (such as long file names etc.) will in such
7530 case be extracted as plain text files along with the files it refers to.
7532 This archive format will be the default format for future versions
7537 The following table summarizes the limitations of each of these
7540 @multitable @columnfractions .10 .20 .20 .20 .20
7541 @headitem Format @tab UID @tab File Size @tab Path Name @tab Devn
7542 @item gnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
7543 @item oldgnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
7544 @item v7 @tab 2097151 @tab 8GB @tab 99 @tab n/a
7545 @item ustar @tab 2097151 @tab 8GB @tab 256 @tab 21
7546 @item posix @tab Unlimited @tab Unlimited @tab Unlimited @tab Unlimited
7549 The default format for @GNUTAR{} is defined at compilation
7550 time. You may check it by running @command{tar --help}, and examining
7551 the last lines of its output. Usually, @GNUTAR{} is configured
7552 to create archives in @samp{gnu} format, however, future version will
7553 switch to @samp{posix}.
7556 * Portability:: Making @command{tar} Archives More Portable
7557 * Compression:: Using Less Space through Compression
7558 * Attributes:: Handling File Attributes
7559 * Standard:: The Standard Format
7560 * Extensions:: @acronym{GNU} Extensions to the Archive Format
7561 * cpio:: Comparison of @command{tar} and @command{cpio}
7565 @section Making @command{tar} Archives More Portable
7567 Creating a @command{tar} archive on a particular system that is meant to be
7568 useful later on many other machines and with other versions of @command{tar}
7569 is more challenging than you might think. @command{tar} archive formats
7570 have been evolving since the first versions of Unix. Many such formats
7571 are around, and are not always compatible with each other. This section
7572 discusses a few problems, and gives some advice about making @command{tar}
7573 archives more portable.
7575 One golden rule is simplicity. For example, limit your @command{tar}
7576 archives to contain only regular files and directories, avoiding
7577 other kind of special files. Do not attempt to save sparse files or
7578 contiguous files as such. Let's discuss a few more problems, in turn.
7580 @FIXME{Discuss GNU extensions (incremental backups, multi-volume
7581 archives and archive labels) in GNU and PAX formats.}
7584 * Portable Names:: Portable Names
7585 * dereference:: Symbolic Links
7586 * old:: Old V7 Archives
7587 * ustar:: Ustar Archives
7588 * gnu:: GNU and old GNU format archives.
7589 * posix:: @acronym{POSIX} archives
7590 * Checksumming:: Checksumming Problems
7591 * Large or Negative Values:: Large files, negative time stamps, etc.
7594 @node Portable Names
7595 @subsection Portable Names
7597 Use portable file and member names. A name is portable if it contains
7598 only ASCII letters and digits, @samp{/}, @samp{.}, @samp{_}, and
7599 @samp{-}; it cannot be empty, start with @samp{-} or @samp{//}, or
7600 contain @samp{/-}. Avoid deep directory nesting. For portability to
7601 old Unix hosts, limit your file name components to 14 characters or
7604 If you intend to have your @command{tar} archives to be read under
7605 MSDOS, you should not rely on case distinction for file names, and you
7606 might use the @acronym{GNU} @command{doschk} program for helping you
7607 further diagnosing illegal MSDOS names, which are even more limited
7611 @subsection Symbolic Links
7612 @cindex File names, using symbolic links
7613 @cindex Symbolic link as file name
7615 @opindex dereference
7616 Normally, when @command{tar} archives a symbolic link, it writes a
7617 block to the archive naming the target of the link. In that way, the
7618 @command{tar} archive is a faithful record of the file system contents.
7619 @option{--dereference} (@option{-h}) is used with @option{--create} (@option{-c}), and causes
7620 @command{tar} to archive the files symbolic links point to, instead of
7621 the links themselves. When this option is used, when @command{tar}
7622 encounters a symbolic link, it will archive the linked-to file,
7623 instead of simply recording the presence of a symbolic link.
7625 The name under which the file is stored in the file system is not
7626 recorded in the archive. To record both the symbolic link name and
7627 the file name in the system, archive the file under both names. If
7628 all links were recorded automatically by @command{tar}, an extracted file
7629 might be linked to a file name that no longer exists in the file
7632 If a linked-to file is encountered again by @command{tar} while creating
7633 the same archive, an entire second copy of it will be stored. (This
7634 @emph{might} be considered a bug.)
7636 So, for portable archives, do not archive symbolic links as such,
7637 and use @option{--dereference} (@option{-h}): many systems do not support
7638 symbolic links, and moreover, your distribution might be unusable if
7639 it contains unresolved symbolic links.
7642 @subsection Old V7 Archives
7643 @cindex Format, old style
7644 @cindex Old style format
7645 @cindex Old style archives
7646 @cindex v7 archive format
7648 Certain old versions of @command{tar} cannot handle additional
7649 information recorded by newer @command{tar} programs. To create an
7650 archive in V7 format (not ANSI), which can be read by these old
7651 versions, specify the @option{--format=v7} option in
7652 conjunction with the @option{--create} (@option{-c}) (@command{tar} also
7653 accepts @option{--portability} or @samp{op-old-archive} for this
7654 option). When you specify it,
7655 @command{tar} leaves out information about directories, pipes, fifos,
7656 contiguous files, and device files, and specifies file ownership by
7657 group and user IDs instead of group and user names.
7659 When updating an archive, do not use @option{--format=v7}
7660 unless the archive was created using this option.
7662 In most cases, a @emph{new} format archive can be read by an @emph{old}
7663 @command{tar} program without serious trouble, so this option should
7664 seldom be needed. On the other hand, most modern @command{tar}s are
7665 able to read old format archives, so it might be safer for you to
7666 always use @option{--format=v7} for your distributions.
7669 @subsection Ustar Archive Format
7671 @cindex ustar archive format
7672 Archive format defined by @acronym{POSIX}.1-1988 specification is called
7673 @code{ustar}. Although it is more flexible than the V7 format, it
7674 still has many restrictions (@xref{Formats,ustar}, for the detailed
7675 description of @code{ustar} format). Along with V7 format,
7676 @code{ustar} format is a good choice for archives intended to be read
7677 with other implementations of @command{tar}.
7679 To create archive in @code{ustar} format, use @option{--format=ustar}
7680 option in conjunction with the @option{--create} (@option{-c}).
7683 @subsection @acronym{GNU} and old @GNUTAR{} format
7685 @cindex GNU archive format
7686 @cindex Old GNU archive format
7687 @GNUTAR{} was based on an early draft of the
7688 @acronym{POSIX} 1003.1 @code{ustar} standard. @acronym{GNU} extensions to
7689 @command{tar}, such as the support for file names longer than 100
7690 characters, use portions of the @command{tar} header record which were
7691 specified in that @acronym{POSIX} draft as unused. Subsequent changes in
7692 @acronym{POSIX} have allocated the same parts of the header record for
7693 other purposes. As a result, @GNUTAR{} format is
7694 incompatible with the current @acronym{POSIX} specification, and with
7695 @command{tar} programs that follow it.
7697 In the majority of cases, @command{tar} will be configured to create
7698 this format by default. This will change in the future releases, since
7699 we plan to make @samp{posix} format the default.
7701 To force creation a @GNUTAR{} archive, use option
7702 @option{--format=gnu}.
7705 @subsection @GNUTAR{} and @acronym{POSIX} @command{tar}
7707 @cindex POSIX archive format
7708 @cindex PAX archive format
7709 The version @value{VERSION} of @GNUTAR{} is able
7710 to read and create archives conforming to @acronym{POSIX.1-2001} standard.
7712 A @acronym{POSIX} conformant archive will be created if @command{tar}
7713 was given @option{--format=posix} option.
7716 @subsection Checksumming Problems
7718 SunOS and HP-UX @command{tar} fail to accept archives created using
7719 @GNUTAR{} and containing non-ASCII file names, that
7720 is, file names having characters with the eight bit set, because they
7721 use signed checksums, while @GNUTAR{} uses unsigned
7722 checksums while creating archives, as per @acronym{POSIX} standards. On
7723 reading, @GNUTAR{} computes both checksums and
7724 accept any. It is somewhat worrying that a lot of people may go
7725 around doing backup of their files using faulty (or at least
7726 non-standard) software, not learning about it until it's time to
7727 restore their missing files with an incompatible file extractor, or
7730 @GNUTAR{} compute checksums both ways, and accept
7731 any on read, so @acronym{GNU} tar can read Sun tapes even with their
7732 wrong checksums. @GNUTAR{} produces the standard
7733 checksum, however, raising incompatibilities with Sun. That is to
7734 say, @GNUTAR{} has not been modified to
7735 @emph{produce} incorrect archives to be read by buggy @command{tar}'s.
7736 I've been told that more recent Sun @command{tar} now read standard
7737 archives, so maybe Sun did a similar patch, after all?
7739 The story seems to be that when Sun first imported @command{tar}
7740 sources on their system, they recompiled it without realizing that
7741 the checksums were computed differently, because of a change in
7742 the default signing of @code{char}'s in their compiler. So they
7743 started computing checksums wrongly. When they later realized their
7744 mistake, they merely decided to stay compatible with it, and with
7745 themselves afterwards. Presumably, but I do not really know, HP-UX
7746 has chosen that their @command{tar} archives to be compatible with Sun's.
7747 The current standards do not favor Sun @command{tar} format. In any
7748 case, it now falls on the shoulders of SunOS and HP-UX users to get
7749 a @command{tar} able to read the good archives they receive.
7751 @node Large or Negative Values
7752 @subsection Large or Negative Values
7753 @cindex large values
7754 @cindex future time stamps
7755 @cindex negative time stamps
7758 The above sections suggest to use @samp{oldest possible} archive
7759 format if in doubt. However, sometimes it is not possible. If you
7760 attempt to archive a file whose metadata cannot be represented using
7761 required format, @GNUTAR{} will print error message and ignore such a
7762 file. You will than have to switch to a format that is able to
7763 handle such values. The format summary table (@pxref{Formats}) will
7766 In particular, when trying to archive files larger than 8GB or with
7767 timestamps not in the range 1970-01-01 00:00:00 through 2242-03-16
7768 12:56:31 @sc{utc}, you will have to chose between @acronym{GNU} and
7769 @acronym{POSIX} archive formats. When considering which format to
7770 choose, bear in mind that the @acronym{GNU} format uses
7771 two's-complement base-256 notation to store values that do not fit
7772 into standard @acronym{ustar} range. Such archives can generally be
7773 read only by a @GNUTAR{} implementation. Moreover, they sometimes
7774 cannot be correctly restored on another hosts even by @GNUTAR{}. For
7775 example, using two's complement representation for negative time
7776 stamps that assumes a signed 32-bit @code{time_t} generates archives
7777 that are not portable to hosts with differing @code{time_t}
7780 On the other hand, @acronym{POSIX} archives, generally speaking, can
7781 be extracted by any tar implementation that understands older
7782 @acronym{ustar} format. The only exception are files larger than 8GB.
7784 @FIXME{Describe how @acronym{POSIX} archives are extracted by non
7788 @section Using Less Space through Compression
7791 * gzip:: Creating and Reading Compressed Archives
7792 * sparse:: Archiving Sparse Files
7796 @subsection Creating and Reading Compressed Archives
7797 @cindex Compressed archives
7798 @cindex Storing archives in compressed format
7800 @GNUTAR{} is able to create and read compressed archives. It supports
7801 @command{gzip} and @command{bzip2} compression programs. For backward
7802 compatibilty, it also supports @command{compress} command, although
7803 we strongly recommend against using it, since there is a patent
7804 covering the algorithm it uses and you could be sued for patent
7805 infringement merely by running @command{compress}! Besides, it is less
7806 effective than @command{gzip} and @command{bzip2}.
7808 Creating a compressed archive is simple: you just specify a
7809 @dfn{compression option} along with the usual archive creation
7810 commands. The compression option is @option{-z} (@option{--gzip}) to
7811 create a @command{gzip} compressed archive, @option{-j}
7812 (@option{--bzip2}) to create a @command{bzip2} compressed archive, and
7813 @option{-Z} (@option{--compress}) to use @command{compress} program.
7817 $ @kbd{tar cfz archive.tar.gz .}
7820 Reading compressed archive is even simpler: you don't need to specify
7821 any additional options as @GNUTAR{} recognizes its format
7822 automatically. Thus, the following commands will list and extract the
7823 archive created in previous example:
7826 # List the compressed archive
7827 $ @kbd{tar tf archive.tar.gz}
7828 # Extract the compressed archive
7829 $ @kbd{tar xf archive.tar.gz}
7832 The only case when you have to specify a decompression option while
7833 reading the archive is when reading from a pipe or from a tape drive
7834 that does not support random access. However, in this case @GNUTAR{}
7835 will indicate which option you should use. For example:
7838 $ @kbd{cat archive.tar.gz | tar tf -}
7839 tar: Archive is compressed. Use -z option
7840 tar: Error is not recoverable: exiting now
7843 If you see such diagnostics, just add the suggested option to the
7844 invocation of @GNUTAR{}:
7847 $ @kbd{cat archive.tar.gz | tar tfz -}
7850 Notice also, that there are several restrictions on operations on
7851 compressed archives. First of all, compressed archives cannot be
7852 modified, i.e., you cannot update (@option{--update} (@option{-u})) them or delete
7853 (@option{--delete}) members from them. Likewise, you cannot append
7854 another @command{tar} archive to a compressed archive using
7855 @option{--append} (@option{-r})). Secondly, multi-volume archives cannot be
7858 The following table summarizes compression options used by @GNUTAR{}.
7866 Filter the archive through @command{gzip}.
7868 You can use @option{--gzip} and @option{--gunzip} on physical devices
7869 (tape drives, etc.) and remote files as well as on normal files; data
7870 to or from such devices or remote files is reblocked by another copy
7871 of the @command{tar} program to enforce the specified (or default) record
7872 size. The default compression parameters are used; if you need to
7873 override them, set @env{GZIP} environment variable, e.g.:
7876 $ @kbd{GZIP=--best tar cfz archive.tar.gz subdir}
7880 Another way would be to avoid the @option{--gzip} (@option{--gunzip}, @option{--ungzip}, @option{-z}) option and run
7881 @command{gzip} explicitly:
7884 $ @kbd{tar cf - subdir | gzip --best -c - > archive.tar.gz}
7887 @cindex corrupted archives
7888 About corrupted compressed archives: @command{gzip}'ed files have no
7889 redundancy, for maximum compression. The adaptive nature of the
7890 compression scheme means that the compression tables are implicitly
7891 spread all over the archive. If you lose a few blocks, the dynamic
7892 construction of the compression tables becomes unsynchronized, and there
7893 is little chance that you could recover later in the archive.
7895 There are pending suggestions for having a per-volume or per-file
7896 compression in @GNUTAR{}. This would allow for viewing the
7897 contents without decompression, and for resynchronizing decompression at
7898 every volume or file, in case of corrupted archives. Doing so, we might
7899 lose some compressibility. But this would have make recovering easier.
7900 So, there are pros and cons. We'll see!
7905 Filter the archive through @code{bzip2}. Otherwise like @option{--gzip}.
7912 Filter the archive through @command{compress}. Otherwise like @option{--gzip}.
7914 The @acronym{GNU} Project recommends you not use
7915 @command{compress}, because there is a patent covering the algorithm it
7916 uses. You could be sued for patent infringement merely by running
7919 @opindex use-compress-program
7920 @item --use-compress-program=@var{prog}
7921 Use external compression program @var{prog}. Use this option if you
7922 have a compression program that @GNUTAR{} does not support. There
7923 are two requirements to which @var{prog} should comply:
7925 First, when called without options, it should read data from standard
7926 input, compress it and output it on standard output.
7928 Secondly, if called with @option{-d} argument, it should do exactly
7929 the opposite, i.e., read the compressed data from the standard input
7930 and produce uncompressed data on the standard output.
7933 @cindex gpg, using with tar
7934 @cindex gnupg, using with tar
7935 @cindex Using encrypted archives
7936 The @option{--use-compress-program} option, in particular, lets you
7937 implement your own filters, not necessarily dealing with
7938 compression/decomression. For example, suppose you wish to implement
7939 PGP encryption on top of compression, using @command{gpg} (@pxref{Top,
7940 gpg, gpg ---- encryption and signing tool, gpg}). The following
7947 -d) gpg --decrypt - | gzip -d -c;;
7948 '') gzip -c | gpg -s ;;
7949 *) echo "Unknown option $1">&2; exit 1;;
7954 Suppose you name it @file{gpgz} and save it somewhere in your
7955 @env{PATH}. Then the following command will create a commpressed
7956 archive signed with your private key:
7959 $ @kbd{tar -cf foo.tar.gpgz --use-compress=gpgz .}
7963 Likewise, the following command will list its contents:
7966 $ @kbd{tar -tf foo.tar.gpgz --use-compress=gpgz .}
7970 The above is based on the following discussion:
7972 I have one question, or maybe it's a suggestion if there isn't a way
7973 to do it now. I would like to use @option{--gzip}, but I'd also like
7974 the output to be fed through a program like @acronym{GNU}
7975 @command{ecc} (actually, right now that's @samp{exactly} what I'd like
7976 to use :-)), basically adding ECC protection on top of compression.
7977 It seems as if this should be quite easy to do, but I can't work out
7978 exactly how to go about it. Of course, I can pipe the standard output
7979 of @command{tar} through @command{ecc}, but then I lose (though I
7980 haven't started using it yet, I confess) the ability to have
7981 @command{tar} use @command{rmt} for it's I/O (I think).
7983 I think the most straightforward thing would be to let me specify a
7984 general set of filters outboard of compression (preferably ordered,
7985 so the order can be automatically reversed on input operations, and
7986 with the options they require specifiable), but beggars shouldn't be
7987 choosers and anything you decide on would be fine with me.
7989 By the way, I like @command{ecc} but if (as the comments say) it can't
7990 deal with loss of block sync, I'm tempted to throw some time at adding
7991 that capability. Supposing I were to actually do such a thing and
7992 get it (apparently) working, do you accept contributed changes to
7993 utilities like that? (Leigh Clayton @file{loc@@soliton.com}, May 1995).
7995 Isn't that exactly the role of the
7996 @option{--use-compress-prog=@var{program}} option?
7997 I never tried it myself, but I suspect you may want to write a
7998 @var{prog} script or program able to filter stdin to stdout to
7999 way you want. It should recognize the @option{-d} option, for when
8000 extraction is needed rather than creation.
8002 It has been reported that if one writes compressed data (through the
8003 @option{--gzip} or @option{--compress} options) to a DLT and tries to use
8004 the DLT compression mode, the data will actually get bigger and one will
8005 end up with less space on the tape.
8009 @subsection Archiving Sparse Files
8010 @cindex Sparse Files
8017 Handle sparse files efficiently.
8020 This option causes all files to be put in the archive to be tested for
8021 sparseness, and handled specially if they are. The @option{--sparse}
8022 (@option{-S}) option is useful when many @code{dbm} files, for example, are being
8023 backed up. Using this option dramatically decreases the amount of
8024 space needed to store such a file.
8026 In later versions, this option may be removed, and the testing and
8027 treatment of sparse files may be done automatically with any special
8028 @acronym{GNU} options. For now, it is an option needing to be specified on
8029 the command line with the creation or updating of an archive.
8031 Files in the file system occasionally have ``holes.'' A hole in a file
8032 is a section of the file's contents which was never written. The
8033 contents of a hole read as all zeros. On many operating systems,
8034 actual disk storage is not allocated for holes, but they are counted
8035 in the length of the file. If you archive such a file, @command{tar}
8036 could create an archive longer than the original. To have @command{tar}
8037 attempt to recognize the holes in a file, use @option{--sparse} (@option{-S}). When
8038 you use this option, then, for any file using less disk space than
8039 would be expected from its length, @command{tar} searches the file for
8040 consecutive stretches of zeros. It then records in the archive for
8041 the file where the consecutive stretches of zeros are, and only
8042 archives the ``real contents'' of the file. On extraction (using
8043 @option{--sparse} is not needed on extraction) any such
8044 files have holes created wherever the continuous stretches of zeros
8045 were found. Thus, if you use @option{--sparse}, @command{tar} archives
8046 won't take more space than the original.
8048 A file is sparse if it contains blocks of zeros whose existence is
8049 recorded, but that have no space allocated on disk. When you specify
8050 the @option{--sparse} option in conjunction with the @option{--create}
8051 (@option{-c}) operation, @command{tar} tests all files for sparseness
8052 while archiving. If @command{tar} finds a file to be sparse, it uses a
8053 sparse representation of the file in the archive. @xref{create}, for
8054 more information about creating archives.
8056 @option{--sparse} is useful when archiving files, such as dbm files,
8057 likely to contain many nulls. This option dramatically
8058 decreases the amount of space needed to store such an archive.
8061 @strong{Please Note:} Always use @option{--sparse} when performing file
8062 system backups, to avoid archiving the expanded forms of files stored
8063 sparsely in the system.
8065 Even if your system has no sparse files currently, some may be
8066 created in the future. If you use @option{--sparse} while making file
8067 system backups as a matter of course, you can be assured the archive
8068 will never take more space on the media than the files take on disk
8069 (otherwise, archiving a disk filled with sparse files might take
8070 hundreds of tapes). @xref{Incremental Dumps}.
8073 @command{tar} ignores the @option{--sparse} option when reading an archive.
8078 Files stored sparsely in the file system are represented sparsely in
8079 the archive. Use in conjunction with write operations.
8082 However, users should be well aware that at archive creation time,
8083 @GNUTAR{} still has to read whole disk file to
8084 locate the @dfn{holes}, and so, even if sparse files use little space
8085 on disk and in the archive, they may sometimes require inordinate
8086 amount of time for reading and examining all-zero blocks of a file.
8087 Although it works, it's painfully slow for a large (sparse) file, even
8088 though the resulting tar archive may be small. (One user reports that
8089 dumping a @file{core} file of over 400 megabytes, but with only about
8090 3 megabytes of actual data, took about 9 minutes on a Sun Sparcstation
8091 ELC, with full CPU utilization.)
8093 This reading is required in all cases and is not related to the fact
8094 the @option{--sparse} option is used or not, so by merely @emph{not}
8095 using the option, you are not saving time@footnote{Well! We should say
8096 the whole truth, here. When @option{--sparse} is selected while creating
8097 an archive, the current @command{tar} algorithm requires sparse files to be
8098 read twice, not once. We hope to develop a new archive format for saving
8099 sparse files in which one pass will be sufficient.}.
8101 Programs like @command{dump} do not have to read the entire file; by
8102 examining the file system directly, they can determine in advance
8103 exactly where the holes are and thus avoid reading through them. The
8104 only data it need read are the actual allocated data blocks.
8105 @GNUTAR{} uses a more portable and straightforward
8106 archiving approach, it would be fairly difficult that it does
8107 otherwise. Elizabeth Zwicky writes to @file{comp.unix.internals}, on
8111 What I did say is that you cannot tell the difference between a hole and an
8112 equivalent number of nulls without reading raw blocks. @code{st_blocks} at
8113 best tells you how many holes there are; it doesn't tell you @emph{where}.
8114 Just as programs may, conceivably, care what @code{st_blocks} is (care
8115 to name one that does?), they may also care where the holes are (I have
8116 no examples of this one either, but it's equally imaginable).
8118 I conclude from this that good archivers are not portable. One can
8119 arguably conclude that if you want a portable program, you can in good
8120 conscience restore files with as many holes as possible, since you can't
8125 @section Handling File Attributes
8128 When @command{tar} reads files, it updates their access times. To
8129 avoid this, use the @option{--atime-preserve[=METHOD]} option, which can either
8130 reset the access time retroactively or avoid changing it in the first
8133 Handling of file attributes
8136 @opindex atime-preserve
8137 @item --atime-preserve
8138 @itemx --atime-preserve=replace
8139 @itemx --atime-preserve=system
8140 Preserve the access times of files that are read. This works only for
8141 files that you own, unless you have superuser privileges.
8143 @option{--atime-preserve=replace} works on most systems, but it also
8144 restores the data modification time and updates the status change
8145 time. Hence it doesn't interact with incremental dumps nicely
8146 (@pxref{Backups}), and it can set access or data modification times
8147 incorrectly if other programs access the file while @command{tar} is
8150 @option{--atime-preserve=system} avoids changing the access time in
8151 the first place, if the operating system supports this.
8152 Unfortunately, this may or may not work on any given operating system
8153 or file system. If @command{tar} knows for sure it won't work, it
8154 complains right away.
8156 Currently @option{--atime-preserve} with no operand defaults to
8157 @option{--atime-preserve=replace}, but this is intended to change to
8158 @option{--atime-preserve=system} when the latter is better-supported.
8163 Do not extract data modification time.
8165 When this option is used, @command{tar} leaves the data modification times
8166 of the files it extracts as the times when the files were extracted,
8167 instead of setting it to the times recorded in the archive.
8169 This option is meaningless with @option{--list} (@option{-t}).
8173 Create extracted files with the same ownership they have in the
8176 This is the default behavior for the superuser,
8177 so this option is meaningful only for non-root users, when @command{tar}
8178 is executed on those systems able to give files away. This is
8179 considered as a security flaw by many people, at least because it
8180 makes quite difficult to correctly account users for the disk space
8181 they occupy. Also, the @code{suid} or @code{sgid} attributes of
8182 files are easily and silently lost when files are given away.
8184 When writing an archive, @command{tar} writes the user id and user name
8185 separately. If it can't find a user name (because the user id is not
8186 in @file{/etc/passwd}), then it does not write one. When restoring,
8187 it tries to look the name (if one was written) up in
8188 @file{/etc/passwd}. If it fails, then it uses the user id stored in
8189 the archive instead.
8191 @opindex no-same-owner
8192 @item --no-same-owner
8194 Do not attempt to restore ownership when extracting. This is the
8195 default behavior for ordinary users, so this option has an effect
8196 only for the superuser.
8198 @opindex numeric-owner
8199 @item --numeric-owner
8200 The @option{--numeric-owner} option allows (ANSI) archives to be written
8201 without user/group name information or such information to be ignored
8202 when extracting. It effectively disables the generation and/or use
8203 of user/group name information. This option forces extraction using
8204 the numeric ids from the archive, ignoring the names.
8206 This is useful in certain circumstances, when restoring a backup from
8207 an emergency floppy with different passwd/group files for example.
8208 It is otherwise impossible to extract files with the right ownerships
8209 if the password file in use during the extraction does not match the
8210 one belonging to the file system(s) being extracted. This occurs,
8211 for example, if you are restoring your files after a major crash and
8212 had booted from an emergency floppy with no password file or put your
8213 disk into another machine to do the restore.
8215 The numeric ids are @emph{always} saved into @command{tar} archives.
8216 The identifying names are added at create time when provided by the
8217 system, unless @option{--old-archive} (@option{-o}) is used. Numeric ids could be
8218 used when moving archives between a collection of machines using
8219 a centralized management for attribution of numeric ids to users
8220 and groups. This is often made through using the NIS capabilities.
8222 When making a @command{tar} file for distribution to other sites, it
8223 is sometimes cleaner to use a single owner for all files in the
8224 distribution, and nicer to specify the write permission bits of the
8225 files as stored in the archive independently of their actual value on
8226 the file system. The way to prepare a clean distribution is usually
8227 to have some Makefile rule creating a directory, copying all needed
8228 files in that directory, then setting ownership and permissions as
8229 wanted (there are a lot of possible schemes), and only then making a
8230 @command{tar} archive out of this directory, before cleaning
8231 everything out. Of course, we could add a lot of options to
8232 @GNUTAR{} for fine tuning permissions and ownership.
8233 This is not the good way, I think. @GNUTAR{} is
8234 already crowded with options and moreover, the approach just explained
8235 gives you a great deal of control already.
8237 @opindex same-permissions, short description
8238 @opindex preserve-permissions, short description
8240 @itemx --same-permissions
8241 @itemx --preserve-permissions
8242 Extract all protection information.
8244 This option causes @command{tar} to set the modes (access permissions) of
8245 extracted files exactly as recorded in the archive. If this option
8246 is not used, the current @code{umask} setting limits the permissions
8247 on extracted files. This option is by default enabled when
8248 @command{tar} is executed by a superuser.
8251 This option is meaningless with @option{--list} (@option{-t}).
8255 Same as both @option{--same-permissions} and @option{--same-order}.
8257 The @option{--preserve} option has no equivalent short option name.
8258 It is equivalent to @option{--same-permissions} plus @option{--same-order}.
8260 @FIXME{I do not see the purpose of such an option. (Neither I. FP.)
8261 Neither do I. --Sergey}
8266 @section Basic Tar Format
8269 While an archive may contain many files, the archive itself is a
8270 single ordinary file. Like any other file, an archive file can be
8271 written to a storage device such as a tape or disk, sent through a
8272 pipe or over a network, saved on the active file system, or even
8273 stored in another archive. An archive file is not easy to read or
8274 manipulate without using the @command{tar} utility or Tar mode in
8275 @acronym{GNU} Emacs.
8277 Physically, an archive consists of a series of file entries terminated
8278 by an end-of-archive entry, which consists of two 512 blocks of zero
8280 entry usually describes one of the files in the archive (an
8281 @dfn{archive member}), and consists of a file header and the contents
8282 of the file. File headers contain file names and statistics, checksum
8283 information which @command{tar} uses to detect file corruption, and
8284 information about file types.
8286 Archives are permitted to have more than one member with the same
8287 member name. One way this situation can occur is if more than one
8288 version of a file has been stored in the archive. For information
8289 about adding new versions of a file to an archive, see @ref{update}.
8290 @FIXME-xref{To learn more about having more than one archive member with the
8291 same name, see -backup node, when it's written.}
8293 In addition to entries describing archive members, an archive may
8294 contain entries which @command{tar} itself uses to store information.
8295 @xref{label}, for an example of such an archive entry.
8297 A @command{tar} archive file contains a series of blocks. Each block
8298 contains @code{BLOCKSIZE} bytes. Although this format may be thought
8299 of as being on magnetic tape, other media are often used.
8301 Each file archived is represented by a header block which describes
8302 the file, followed by zero or more blocks which give the contents
8303 of the file. At the end of the archive file there are two 512-byte blocks
8304 filled with binary zeros as an end-of-file marker. A reasonable system
8305 should write such end-of-file marker at the end of an archive, but
8306 must not assume that such a block exists when reading an archive. In
8307 particular @GNUTAR{} always issues a warning if it does not encounter it.
8309 The blocks may be @dfn{blocked} for physical I/O operations.
8310 Each record of @var{n} blocks (where @var{n} is set by the
8311 @option{--blocking-factor=@var{512-size}} (@option{-b @var{512-size}}) option to @command{tar}) is written with a single
8312 @w{@samp{write ()}} operation. On magnetic tapes, the result of
8313 such a write is a single record. When writing an archive,
8314 the last record of blocks should be written at the full size, with
8315 blocks after the zero block containing all zeros. When reading
8316 an archive, a reasonable system should properly handle an archive
8317 whose last record is shorter than the rest, or which contains garbage
8318 records after a zero block.
8320 The header block is defined in C as follows. In the @GNUTAR{}
8321 distribution, this is part of file @file{src/tar.h}:
8324 @include header.texi
8327 All characters in header blocks are represented by using 8-bit
8328 characters in the local variant of ASCII. Each field within the
8329 structure is contiguous; that is, there is no padding used within
8330 the structure. Each character on the archive medium is stored
8333 Bytes representing the contents of files (after the header block
8334 of each file) are not translated in any way and are not constrained
8335 to represent characters in any character set. The @command{tar} format
8336 does not distinguish text files from binary files, and no translation
8337 of file contents is performed.
8339 The @code{name}, @code{linkname}, @code{magic}, @code{uname}, and
8340 @code{gname} are null-terminated character strings. All other fields
8341 are zero-filled octal numbers in ASCII. Each numeric field of width
8342 @var{w} contains @var{w} minus 1 digits, and a null.
8344 The @code{name} field is the file name of the file, with directory names
8345 (if any) preceding the file name, separated by slashes.
8347 @FIXME{how big a name before field overflows?}
8349 The @code{mode} field provides nine bits specifying file permissions
8350 and three bits to specify the Set UID, Set GID, and Save Text
8351 (@dfn{sticky}) modes. Values for these bits are defined above.
8352 When special permissions are required to create a file with a given
8353 mode, and the user restoring files from the archive does not hold such
8354 permissions, the mode bit(s) specifying those special permissions
8355 are ignored. Modes which are not supported by the operating system
8356 restoring files from the archive will be ignored. Unsupported modes
8357 should be faked up when creating or updating an archive; e.g., the
8358 group permission could be copied from the @emph{other} permission.
8360 The @code{uid} and @code{gid} fields are the numeric user and group
8361 ID of the file owners, respectively. If the operating system does
8362 not support numeric user or group IDs, these fields should be ignored.
8364 The @code{size} field is the size of the file in bytes; linked files
8365 are archived with this field specified as zero. @FIXME-xref{Modifiers, in
8366 particular the @option{--incremental} (@option{-G}) option.}
8368 The @code{mtime} field is the data modification time of the file at
8369 the time it was archived. It is the ASCII representation of the octal
8370 value of the last time the file's contents were modified, represented
8371 as an integer number of
8372 seconds since January 1, 1970, 00:00 Coordinated Universal Time.
8374 The @code{chksum} field is the ASCII representation of the octal value
8375 of the simple sum of all bytes in the header block. Each 8-bit
8376 byte in the header is added to an unsigned integer, initialized to
8377 zero, the precision of which shall be no less than seventeen bits.
8378 When calculating the checksum, the @code{chksum} field is treated as
8379 if it were all blanks.
8381 The @code{typeflag} field specifies the type of file archived. If a
8382 particular implementation does not recognize or permit the specified
8383 type, the file will be extracted as if it were a regular file. As this
8384 action occurs, @command{tar} issues a warning to the standard error.
8386 The @code{atime} and @code{ctime} fields are used in making incremental
8387 backups; they store, respectively, the particular file's access and
8388 status change times.
8390 The @code{offset} is used by the @option{--multi-volume} (@option{-M}) option, when
8391 making a multi-volume archive. The offset is number of bytes into
8392 the file that we need to restart at to continue the file on the next
8393 tape, i.e., where we store the location that a continued file is
8396 The following fields were added to deal with sparse files. A file
8397 is @dfn{sparse} if it takes in unallocated blocks which end up being
8398 represented as zeros, i.e., no useful data. A test to see if a file
8399 is sparse is to look at the number blocks allocated for it versus the
8400 number of characters in the file; if there are fewer blocks allocated
8401 for the file than would normally be allocated for a file of that
8402 size, then the file is sparse. This is the method @command{tar} uses to
8403 detect a sparse file, and once such a file is detected, it is treated
8404 differently from non-sparse files.
8406 Sparse files are often @code{dbm} files, or other database-type files
8407 which have data at some points and emptiness in the greater part of
8408 the file. Such files can appear to be very large when an @samp{ls
8409 -l} is done on them, when in truth, there may be a very small amount
8410 of important data contained in the file. It is thus undesirable
8411 to have @command{tar} think that it must back up this entire file, as
8412 great quantities of room are wasted on empty blocks, which can lead
8413 to running out of room on a tape far earlier than is necessary.
8414 Thus, sparse files are dealt with so that these empty blocks are
8415 not written to the tape. Instead, what is written to the tape is a
8416 description, of sorts, of the sparse file: where the holes are, how
8417 big the holes are, and how much data is found at the end of the hole.
8418 This way, the file takes up potentially far less room on the tape,
8419 and when the file is extracted later on, it will look exactly the way
8420 it looked beforehand. The following is a description of the fields
8421 used to handle a sparse file:
8423 The @code{sp} is an array of @code{struct sparse}. Each @code{struct
8424 sparse} contains two 12-character strings which represent an offset
8425 into the file and a number of bytes to be written at that offset.
8426 The offset is absolute, and not relative to the offset in preceding
8429 The header can hold four of these @code{struct sparse} at the moment;
8430 if more are needed, they are not stored in the header.
8432 The @code{isextended} flag is set when an @code{extended_header}
8433 is needed to deal with a file. Note that this means that this flag
8434 can only be set when dealing with a sparse file, and it is only set
8435 in the event that the description of the file will not fit in the
8436 allotted room for sparse structures in the header. In other words,
8437 an extended_header is needed.
8439 The @code{extended_header} structure is used for sparse files which
8440 need more sparse structures than can fit in the header. The header can
8441 fit 4 such structures; if more are needed, the flag @code{isextended}
8442 gets set and the next block is an @code{extended_header}.
8444 Each @code{extended_header} structure contains an array of 21
8445 sparse structures, along with a similar @code{isextended} flag
8446 that the header had. There can be an indeterminate number of such
8447 @code{extended_header}s to describe a sparse file.
8451 @item @code{REGTYPE}
8452 @itemx @code{AREGTYPE}
8453 These flags represent a regular file. In order to be compatible
8454 with older versions of @command{tar}, a @code{typeflag} value of
8455 @code{AREGTYPE} should be silently recognized as a regular file.
8456 New archives should be created using @code{REGTYPE}. Also, for
8457 backward compatibility, @command{tar} treats a regular file whose name
8458 ends with a slash as a directory.
8460 @item @code{LNKTYPE}
8461 This flag represents a file linked to another file, of any type,
8462 previously archived. Such files are identified in Unix by each
8463 file having the same device and inode number. The linked-to name is
8464 specified in the @code{linkname} field with a trailing null.
8466 @item @code{SYMTYPE}
8467 This represents a symbolic link to another file. The linked-to name
8468 is specified in the @code{linkname} field with a trailing null.
8470 @item @code{CHRTYPE}
8471 @itemx @code{BLKTYPE}
8472 These represent character special files and block special files
8473 respectively. In this case the @code{devmajor} and @code{devminor}
8474 fields will contain the major and minor device numbers respectively.
8475 Operating systems may map the device specifications to their own
8476 local specification, or may ignore the entry.
8478 @item @code{DIRTYPE}
8479 This flag specifies a directory or sub-directory. The directory
8480 name in the @code{name} field should end with a slash. On systems where
8481 disk allocation is performed on a directory basis, the @code{size} field
8482 will contain the maximum number of bytes (which may be rounded to
8483 the nearest disk block allocation unit) which the directory may
8484 hold. A @code{size} field of zero indicates no such limiting. Systems
8485 which do not support limiting in this manner should ignore the
8488 @item @code{FIFOTYPE}
8489 This specifies a FIFO special file. Note that the archiving of a
8490 FIFO file archives the existence of this file and not its contents.
8492 @item @code{CONTTYPE}
8493 This specifies a contiguous file, which is the same as a normal
8494 file except that, in operating systems which support it, all its
8495 space is allocated contiguously on the disk. Operating systems
8496 which do not allow contiguous allocation should silently treat this
8497 type as a normal file.
8499 @item @code{A} @dots{} @code{Z}
8500 These are reserved for custom implementations. Some of these are
8501 used in the @acronym{GNU} modified format, as described below.
8505 Other values are reserved for specification in future revisions of
8506 the P1003 standard, and should not be used by any @command{tar} program.
8508 The @code{magic} field indicates that this archive was output in
8509 the P1003 archive format. If this field contains @code{TMAGIC},
8510 the @code{uname} and @code{gname} fields will contain the ASCII
8511 representation of the owner and group of the file respectively.
8512 If found, the user and group IDs are used rather than the values in
8513 the @code{uid} and @code{gid} fields.
8515 For references, see ISO/IEC 9945-1:1990 or IEEE Std 1003.1-1990, pages
8516 169-173 (section 10.1) for @cite{Archive/Interchange File Format}; and
8517 IEEE Std 1003.2-1992, pages 380-388 (section 4.48) and pages 936-940
8518 (section E.4.48) for @cite{pax - Portable archive interchange}.
8521 @section @acronym{GNU} Extensions to the Archive Format
8524 The @acronym{GNU} format uses additional file types to describe new types of
8525 files in an archive. These are listed below.
8528 @item GNUTYPE_DUMPDIR
8530 This represents a directory and a list of files created by the
8531 @option{--incremental} (@option{-G}) option. The @code{size} field gives the total
8532 size of the associated list of files. Each file name is preceded by
8533 either a @samp{Y} (the file should be in this archive) or an @samp{N}.
8534 (The file is a directory, or is not stored in the archive.) Each file
8535 name is terminated by a null. There is an additional null after the
8538 @item GNUTYPE_MULTIVOL
8540 This represents a file continued from another volume of a multi-volume
8541 archive created with the @option{--multi-volume} (@option{-M}) option. The original
8542 type of the file is not given here. The @code{size} field gives the
8543 maximum size of this piece of the file (assuming the volume does
8544 not end before the file is written out). The @code{offset} field
8545 gives the offset from the beginning of the file where this part of
8546 the file begins. Thus @code{size} plus @code{offset} should equal
8547 the original size of the file.
8549 @item GNUTYPE_SPARSE
8551 This flag indicates that we are dealing with a sparse file. Note
8552 that archiving a sparse file requires special operations to find
8553 holes in the file, which mark the positions of these holes, along
8554 with the number of bytes of data to be found after the hole.
8556 @item GNUTYPE_VOLHDR
8558 This file type is used to mark the volume header that was given with
8559 the @option{--label=@var{archive-label}} (@option{-V @var{archive-label}}) option when the archive was created. The @code{name}
8560 field contains the @code{name} given after the @option{--label=@var{archive-label}} (@option{-V @var{archive-label}}) option.
8561 The @code{size} field is zero. Only the first file in each volume
8562 of an archive should have this type.
8566 You may have trouble reading a @acronym{GNU} format archive on a
8567 non-@acronym{GNU} system if the options @option{--incremental} (@option{-G}),
8568 @option{--multi-volume} (@option{-M}), @option{--sparse} (@option{-S}), or @option{--label=@var{archive-label}} (@option{-V @var{archive-label}}) were
8569 used when writing the archive. In general, if @command{tar} does not
8570 use the @acronym{GNU}-added fields of the header, other versions of
8571 @command{tar} should be able to read the archive. Otherwise, the
8572 @command{tar} program will give an error, the most likely one being a
8576 @section Comparison of @command{tar} and @command{cpio}
8579 @FIXME{Reorganize the following material}
8581 The @command{cpio} archive formats, like @command{tar}, do have maximum
8582 pathname lengths. The binary and old ASCII formats have a max path
8583 length of 256, and the new ASCII and CRC ASCII formats have a max
8584 path length of 1024. @acronym{GNU} @command{cpio} can read and write archives
8585 with arbitrary pathname lengths, but other @command{cpio} implementations
8586 may crash unexplainedly trying to read them.
8588 @command{tar} handles symbolic links in the form in which it comes in BSD;
8589 @command{cpio} doesn't handle symbolic links in the form in which it comes
8590 in System V prior to SVR4, and some vendors may have added symlinks
8591 to their system without enhancing @command{cpio} to know about them.
8592 Others may have enhanced it in a way other than the way I did it
8593 at Sun, and which was adopted by AT&T (and which is, I think, also
8594 present in the @command{cpio} that Berkeley picked up from AT&T and put
8595 into a later BSD release---I think I gave them my changes).
8597 (SVR4 does some funny stuff with @command{tar}; basically, its @command{cpio}
8598 can handle @command{tar} format input, and write it on output, and it
8599 probably handles symbolic links. They may not have bothered doing
8600 anything to enhance @command{tar} as a result.)
8602 @command{cpio} handles special files; traditional @command{tar} doesn't.
8604 @command{tar} comes with V7, System III, System V, and BSD source;
8605 @command{cpio} comes only with System III, System V, and later BSD
8606 (4.3-tahoe and later).
8608 @command{tar}'s way of handling multiple hard links to a file can handle
8609 file systems that support 32-bit inumbers (e.g., the BSD file system);
8610 @command{cpio}s way requires you to play some games (in its "binary"
8611 format, i-numbers are only 16 bits, and in its "portable ASCII" format,
8612 they're 18 bits---it would have to play games with the "file system ID"
8613 field of the header to make sure that the file system ID/i-number pairs
8614 of different files were always different), and I don't know which
8615 @command{cpio}s, if any, play those games. Those that don't might get
8616 confused and think two files are the same file when they're not, and
8617 make hard links between them.
8619 @command{tar}s way of handling multiple hard links to a file places only
8620 one copy of the link on the tape, but the name attached to that copy
8621 is the @emph{only} one you can use to retrieve the file; @command{cpio}s
8622 way puts one copy for every link, but you can retrieve it using any
8626 What type of check sum (if any) is used, and how is this calculated.
8629 See the attached manual pages for @command{tar} and @command{cpio} format.
8630 @command{tar} uses a checksum which is the sum of all the bytes in the
8631 @command{tar} header for a file; @command{cpio} uses no checksum.
8634 If anyone knows why @command{cpio} was made when @command{tar} was present
8638 It wasn't. @command{cpio} first showed up in PWB/UNIX 1.0; no
8639 generally-available version of UNIX had @command{tar} at the time. I don't
8640 know whether any version that was generally available @emph{within AT&T}
8641 had @command{tar}, or, if so, whether the people within AT&T who did
8642 @command{cpio} knew about it.
8644 On restore, if there is a corruption on a tape @command{tar} will stop at
8645 that point, while @command{cpio} will skip over it and try to restore the
8648 The main difference is just in the command syntax and header format.
8650 @command{tar} is a little more tape-oriented in that everything is blocked
8651 to start on a record boundary.
8654 Is there any differences between the ability to recover crashed
8655 archives between the two of them. (Is there any chance of recovering
8656 crashed archives at all.)
8659 Theoretically it should be easier under @command{tar} since the blocking
8660 lets you find a header with some variation of @samp{dd skip=@var{nn}}.
8661 However, modern @command{cpio}'s and variations have an option to just
8662 search for the next file header after an error with a reasonable chance
8663 of resyncing. However, lots of tape driver software won't allow you to
8664 continue past a media error which should be the only reason for getting
8665 out of sync unless a file changed sizes while you were writing the
8669 If anyone knows why @command{cpio} was made when @command{tar} was present
8670 at the unix scene, please tell me about this too.
8673 Probably because it is more media efficient (by not blocking everything
8674 and using only the space needed for the headers where @command{tar}
8675 always uses 512 bytes per file header) and it knows how to archive
8678 You might want to look at the freely available alternatives. The
8679 major ones are @command{afio}, @GNUTAR{}, and
8680 @command{pax}, each of which have their own extensions with some
8681 backwards compatibility.
8683 Sparse files were @command{tar}red as sparse files (which you can
8684 easily test, because the resulting archive gets smaller, and
8685 @acronym{GNU} @command{cpio} can no longer read it).
8688 @chapter Tapes and Other Archive Media
8691 A few special cases about tape handling warrant more detailed
8692 description. These special cases are discussed below.
8694 Many complexities surround the use of @command{tar} on tape drives. Since
8695 the creation and manipulation of archives located on magnetic tape was
8696 the original purpose of @command{tar}, it contains many features making
8697 such manipulation easier.
8699 Archives are usually written on dismountable media---tape cartridges,
8700 mag tapes, or floppy disks.
8702 The amount of data a tape or disk holds depends not only on its size,
8703 but also on how it is formatted. A 2400 foot long reel of mag tape
8704 holds 40 megabytes of data when formatted at 1600 bits per inch. The
8705 physically smaller EXABYTE tape cartridge holds 2.3 gigabytes.
8707 Magnetic media are re-usable---once the archive on a tape is no longer
8708 needed, the archive can be erased and the tape or disk used over.
8709 Media quality does deteriorate with use, however. Most tapes or disks
8710 should be discarded when they begin to produce data errors. EXABYTE
8711 tape cartridges should be discarded when they generate an @dfn{error
8712 count} (number of non-usable bits) of more than 10k.
8714 Magnetic media are written and erased using magnetic fields, and
8715 should be protected from such fields to avoid damage to stored data.
8716 Sticking a floppy disk to a filing cabinet using a magnet is probably
8720 * Device:: Device selection and switching
8721 * Remote Tape Server::
8722 * Common Problems and Solutions::
8723 * Blocking:: Blocking
8724 * Many:: Many archives on one tape
8725 * Using Multiple Tapes:: Using Multiple Tapes
8726 * label:: Including a Label in the Archive
8728 * Write Protection::
8732 @section Device Selection and Switching
8736 @item -f [@var{hostname}:]@var{file}
8737 @itemx --file=[@var{hostname}:]@var{file}
8738 Use archive file or device @var{file} on @var{hostname}.
8741 This option is used to specify the file name of the archive @command{tar}
8744 If the file name is @samp{-}, @command{tar} reads the archive from standard
8745 input (when listing or extracting), or writes it to standard output
8746 (when creating). If the @samp{-} file name is given when updating an
8747 archive, @command{tar} will read the original archive from its standard
8748 input, and will write the entire new archive to its standard output.
8750 If the file name contains a @samp{:}, it is interpreted as
8751 @samp{hostname:file name}. If the @var{hostname} contains an @dfn{at}
8752 sign (@samp{@@}), it is treated as @samp{user@@hostname:file name}. In
8753 either case, @command{tar} will invoke the command @command{rsh} (or
8754 @command{remsh}) to start up an @command{/usr/libexec/rmt} on the remote
8755 machine. If you give an alternate login name, it will be given to the
8757 Naturally, the remote machine must have an executable
8758 @command{/usr/libexec/rmt}. This program is free software from the
8759 University of California, and a copy of the source code can be found
8760 with the sources for @command{tar}; it's compiled and installed by default.
8761 The exact path to this utility is determined when configuring the package.
8762 It is @file{@var{prefix}/libexec/rmt}, where @var{prefix} stands for
8763 your installation prefix. This location may also be overridden at
8764 runtime by using @option{rmt-command=@var{command}} option (@xref{Option Summary,
8765 ---rmt-command}, for detailed description of this option. @xref{Remote
8766 Tape Server}, for the description of @command{rmt} command).
8768 If this option is not given, but the environment variable @env{TAPE}
8769 is set, its value is used; otherwise, old versions of @command{tar}
8770 used a default archive name (which was picked when @command{tar} was
8771 compiled). The default is normally set up to be the @dfn{first} tape
8772 drive or other transportable I/O medium on the system.
8774 Starting with version 1.11.5, @GNUTAR{} uses
8775 standard input and standard output as the default device, and I will
8776 not try anymore supporting automatic device detection at installation
8777 time. This was failing really in too many cases, it was hopeless.
8778 This is now completely left to the installer to override standard
8779 input and standard output for default device, if this seems
8780 preferable. Further, I think @emph{most} actual usages of
8781 @command{tar} are done with pipes or disks, not really tapes,
8782 cartridges or diskettes.
8784 Some users think that using standard input and output is running
8785 after trouble. This could lead to a nasty surprise on your screen if
8786 you forget to specify an output file name---especially if you are going
8787 through a network or terminal server capable of buffering large amounts
8788 of output. We had so many bug reports in that area of configuring
8789 default tapes automatically, and so many contradicting requests, that
8790 we finally consider the problem to be portably intractable. We could
8791 of course use something like @samp{/dev/tape} as a default, but this
8792 is @emph{also} running after various kind of trouble, going from hung
8793 processes to accidental destruction of real tapes. After having seen
8794 all this mess, using standard input and output as a default really
8795 sounds like the only clean choice left, and a very useful one too.
8797 @GNUTAR{} reads and writes archive in records, I
8798 suspect this is the main reason why block devices are preferred over
8799 character devices. Most probably, block devices are more efficient
8800 too. The installer could also check for @samp{DEFTAPE} in
8801 @file{<sys/mtio.h>}.
8804 @opindex force-local, short description
8806 Archive file is local even if it contains a colon.
8808 @opindex rsh-command
8809 @item --rsh-command=@var{command}
8810 Use remote @var{command} instead of @command{rsh}. This option exists
8811 so that people who use something other than the standard @command{rsh}
8812 (e.g., a Kerberized @command{rsh}) can access a remote device.
8814 When this command is not used, the shell command found when
8815 the @command{tar} program was installed is used instead. This is
8816 the first found of @file{/usr/ucb/rsh}, @file{/usr/bin/remsh},
8817 @file{/usr/bin/rsh}, @file{/usr/bsd/rsh} or @file{/usr/bin/nsh}.
8818 The installer may have overridden this by defining the environment
8819 variable @env{RSH} @emph{at installation time}.
8822 Specify drive and density.
8824 @opindex multi-volume, short description
8826 @itemx --multi-volume
8827 Create/list/extract multi-volume archive.
8829 This option causes @command{tar} to write a @dfn{multi-volume} archive---one
8830 that may be larger than will fit on the medium used to hold it.
8831 @xref{Multi-Volume Archives}.
8833 @opindex tape-length, short description
8835 @itemx --tape-length=@var{num}
8836 Change tape after writing @var{num} x 1024 bytes.
8838 This option might be useful when your tape drivers do not properly
8839 detect end of physical tapes. By being slightly conservative on the
8840 maximum tape length, you might avoid the problem entirely.
8842 @opindex info-script, short description
8843 @opindex new-volume-script, short description
8845 @itemx --info-script=@var{file}
8846 @itemx --new-volume-script=@var{file}
8847 Execute @file{file} at end of each tape. This implies
8848 @option{--multi-volume} (@option{-M}). @xref{info-script}, for a detailed
8849 description of this option.
8852 @node Remote Tape Server
8853 @section The Remote Tape Server
8855 @cindex remote tape drive
8857 In order to access the tape drive on a remote machine, @command{tar}
8858 uses the remote tape server written at the University of California at
8859 Berkeley. The remote tape server must be installed as
8860 @file{@var{prefix}/libexec/rmt} on any machine whose tape drive you
8861 want to use. @command{tar} calls @command{rmt} by running an
8862 @command{rsh} or @command{remsh} to the remote machine, optionally
8863 using a different login name if one is supplied.
8865 A copy of the source for the remote tape server is provided. It is
8866 Copyright @copyright{} 1983 by the Regents of the University of
8867 California, but can be freely distributed. It is compiled and
8868 installed by default.
8870 @cindex absolute file names
8871 Unless you use the @option{--absolute-names} (@option{-P}) option,
8872 @GNUTAR{} will not allow you to create an archive that contains
8873 absolute file names (a file name beginning with @samp{/}.) If you try,
8874 @command{tar} will automatically remove the leading @samp{/} from the
8875 file names it stores in the archive. It will also type a warning
8876 message telling you what it is doing.
8878 When reading an archive that was created with a different
8879 @command{tar} program, @GNUTAR{} automatically
8880 extracts entries in the archive which have absolute file names as if
8881 the file names were not absolute. This is an important feature. A
8882 visitor here once gave a @command{tar} tape to an operator to restore;
8883 the operator used Sun @command{tar} instead of @GNUTAR{},
8884 and the result was that it replaced large portions of
8885 our @file{/bin} and friends with versions from the tape; needless to
8886 say, we were unhappy about having to recover the file system from
8889 For example, if the archive contained a file @file{/usr/bin/computoy},
8890 @GNUTAR{} would extract the file to @file{usr/bin/computoy},
8891 relative to the current directory. If you want to extract the files in
8892 an archive to the same absolute names that they had when the archive
8893 was created, you should do a @samp{cd /} before extracting the files
8894 from the archive, or you should either use the @option{--absolute-names}
8895 option, or use the command @samp{tar -C / @dots{}}.
8897 @cindex Ultrix 3.1 and write failure
8898 Some versions of Unix (Ultrix 3.1 is known to have this problem),
8899 can claim that a short write near the end of a tape succeeded,
8900 when it actually failed. This will result in the -M option not
8901 working correctly. The best workaround at the moment is to use a
8902 significantly larger blocking factor than the default 20.
8904 In order to update an archive, @command{tar} must be able to backspace the
8905 archive in order to reread or rewrite a record that was just read (or
8906 written). This is currently possible only on two kinds of files: normal
8907 disk files (or any other file that can be backspaced with @samp{lseek}),
8908 and industry-standard 9-track magnetic tape (or any other kind of tape
8909 that can be backspaced with the @code{MTIOCTOP} @code{ioctl}.
8911 This means that the @option{--append}, @option{--concatenate}, and
8912 @option{--delete} commands will not work on any other kind of file.
8913 Some media simply cannot be backspaced, which means these commands and
8914 options will never be able to work on them. These non-backspacing
8915 media include pipes and cartridge tape drives.
8917 Some other media can be backspaced, and @command{tar} will work on them
8918 once @command{tar} is modified to do so.
8920 Archives created with the @option{--multi-volume}, @option{--label}, and
8921 @option{--incremental} (@option{-G}) options may not be readable by other version
8922 of @command{tar}. In particular, restoring a file that was split over
8923 a volume boundary will require some careful work with @command{dd}, if
8924 it can be done at all. Other versions of @command{tar} may also create
8925 an empty file whose name is that of the volume header. Some versions
8926 of @command{tar} may create normal files instead of directories archived
8927 with the @option{--incremental} (@option{-G}) option.
8929 @node Common Problems and Solutions
8930 @section Some Common Problems and their Solutions
8937 no such file or directory
8940 errors from @command{tar}:
8941 directory checksum error
8944 errors from media/system:
8955 @dfn{Block} and @dfn{record} terminology is rather confused, and it
8956 is also confusing to the expert reader. On the other hand, readers
8957 who are new to the field have a fresh mind, and they may safely skip
8958 the next two paragraphs, as the remainder of this manual uses those
8959 two terms in a quite consistent way.
8961 John Gilmore, the writer of the public domain @command{tar} from which
8962 @GNUTAR{} was originally derived, wrote (June 1995):
8965 The nomenclature of tape drives comes from IBM, where I believe
8966 they were invented for the IBM 650 or so. On IBM mainframes, what
8967 is recorded on tape are tape blocks. The logical organization of
8968 data is into records. There are various ways of putting records into
8969 blocks, including @code{F} (fixed sized records), @code{V} (variable
8970 sized records), @code{FB} (fixed blocked: fixed size records, @var{n}
8971 to a block), @code{VB} (variable size records, @var{n} to a block),
8972 @code{VSB} (variable spanned blocked: variable sized records that can
8973 occupy more than one block), etc. The @code{JCL} @samp{DD RECFORM=}
8974 parameter specified this to the operating system.
8976 The Unix man page on @command{tar} was totally confused about this.
8977 When I wrote @code{PD TAR}, I used the historically correct terminology
8978 (@command{tar} writes data records, which are grouped into blocks).
8979 It appears that the bogus terminology made it into @acronym{POSIX} (no surprise
8980 here), and now Fran@,{c}ois has migrated that terminology back
8981 into the source code too.
8984 The term @dfn{physical block} means the basic transfer chunk from or
8985 to a device, after which reading or writing may stop without anything
8986 being lost. In this manual, the term @dfn{block} usually refers to
8987 a disk physical block, @emph{assuming} that each disk block is 512
8988 bytes in length. It is true that some disk devices have different
8989 physical blocks, but @command{tar} ignore these differences in its own
8990 format, which is meant to be portable, so a @command{tar} block is always
8991 512 bytes in length, and @dfn{block} always mean a @command{tar} block.
8992 The term @dfn{logical block} often represents the basic chunk of
8993 allocation of many disk blocks as a single entity, which the operating
8994 system treats somewhat atomically; this concept is only barely used
8997 The term @dfn{physical record} is another way to speak of a physical
8998 block, those two terms are somewhat interchangeable. In this manual,
8999 the term @dfn{record} usually refers to a tape physical block,
9000 @emph{assuming} that the @command{tar} archive is kept on magnetic tape.
9001 It is true that archives may be put on disk or used with pipes,
9002 but nevertheless, @command{tar} tries to read and write the archive one
9003 @dfn{record} at a time, whatever the medium in use. One record is made
9004 up of an integral number of blocks, and this operation of putting many
9005 disk blocks into a single tape block is called @dfn{reblocking}, or
9006 more simply, @dfn{blocking}. The term @dfn{logical record} refers to
9007 the logical organization of many characters into something meaningful
9008 to the application. The term @dfn{unit record} describes a small set
9009 of characters which are transmitted whole to or by the application,
9010 and often refers to a line of text. Those two last terms are unrelated
9011 to what we call a @dfn{record} in @GNUTAR{}.
9013 When writing to tapes, @command{tar} writes the contents of the archive
9014 in chunks known as @dfn{records}. To change the default blocking
9015 factor, use the @option{--blocking-factor=@var{512-size}} (@option{-b
9016 @var{512-size}}) option. Each record will then be composed of
9017 @var{512-size} blocks. (Each @command{tar} block is 512 bytes.
9018 @xref{Standard}.) Each file written to the archive uses at least one
9019 full record. As a result, using a larger record size can result in
9020 more wasted space for small files. On the other hand, a larger record
9021 size can often be read and written much more efficiently.
9023 Further complicating the problem is that some tape drives ignore the
9024 blocking entirely. For these, a larger record size can still improve
9025 performance (because the software layers above the tape drive still
9026 honor the blocking), but not as dramatically as on tape drives that
9029 When reading an archive, @command{tar} can usually figure out the
9030 record size on itself. When this is the case, and a non-standard
9031 record size was used when the archive was created, @command{tar} will
9032 print a message about a non-standard blocking factor, and then operate
9033 normally. On some tape devices, however, @command{tar} cannot figure
9034 out the record size itself. On most of those, you can specify a
9035 blocking factor (with @option{--blocking-factor}) larger than the
9036 actual blocking factor, and then use the @option{--read-full-records}
9037 (@option{-B}) option. (If you specify a blocking factor with
9038 @option{--blocking-factor} and don't use the
9039 @option{--read-full-records} option, then @command{tar} will not
9040 attempt to figure out the recording size itself.) On some devices,
9041 you must always specify the record size exactly with
9042 @option{--blocking-factor} when reading, because @command{tar} cannot
9043 figure it out. In any case, use @option{--list} (@option{-t}) before
9044 doing any extractions to see whether @command{tar} is reading the archive
9047 @command{tar} blocks are all fixed size (512 bytes), and its scheme for
9048 putting them into records is to put a whole number of them (one or
9049 more) into each record. @command{tar} records are all the same size;
9050 at the end of the file there's a block containing all zeros, which
9051 is how you tell that the remainder of the last record(s) are garbage.
9053 In a standard @command{tar} file (no options), the block size is 512
9054 and the record size is 10240, for a blocking factor of 20. What the
9055 @option{--blocking-factor} option does is sets the blocking factor,
9056 changing the record size while leaving the block size at 512 bytes.
9057 20 was fine for ancient 800 or 1600 bpi reel-to-reel tape drives;
9058 most tape drives these days prefer much bigger records in order to
9059 stream and not waste tape. When writing tapes for myself, some tend
9060 to use a factor of the order of 2048, say, giving a record size of
9061 around one megabyte.
9063 If you use a blocking factor larger than 20, older @command{tar}
9064 programs might not be able to read the archive, so we recommend this
9065 as a limit to use in practice. @GNUTAR{}, however,
9066 will support arbitrarily large record sizes, limited only by the
9067 amount of virtual memory or the physical characteristics of the tape
9071 * Format Variations:: Format Variations
9072 * Blocking Factor:: The Blocking Factor of an Archive
9075 @node Format Variations
9076 @subsection Format Variations
9077 @cindex Format Parameters
9078 @cindex Format Options
9079 @cindex Options, archive format specifying
9080 @cindex Options, format specifying
9083 Format parameters specify how an archive is written on the archive
9084 media. The best choice of format parameters will vary depending on
9085 the type and number of files being archived, and on the media used to
9088 To specify format parameters when accessing or creating an archive,
9089 you can use the options described in the following sections.
9090 If you do not specify any format parameters, @command{tar} uses
9091 default parameters. You cannot modify a compressed archive.
9092 If you create an archive with the @option{--blocking-factor} option
9093 specified (@pxref{Blocking Factor}), you must specify that
9094 blocking-factor when operating on the archive. @xref{Formats}, for other
9095 examples of format parameter considerations.
9097 @node Blocking Factor
9098 @subsection The Blocking Factor of an Archive
9099 @cindex Blocking Factor
9101 @cindex Number of blocks per record
9102 @cindex Number of bytes per record
9103 @cindex Bytes per record
9104 @cindex Blocks per record
9107 @opindex blocking-factor
9108 The data in an archive is grouped into blocks, which are 512 bytes.
9109 Blocks are read and written in whole number multiples called
9110 @dfn{records}. The number of blocks in a record (ie. the size of a
9111 record in units of 512 bytes) is called the @dfn{blocking factor}.
9112 The @option{--blocking-factor=@var{512-size}} (@option{-b
9113 @var{512-size}}) option specifies the blocking factor of an archive.
9114 The default blocking factor is typically 20 (i.e., 10240 bytes), but
9115 can be specified at installation. To find out the blocking factor of
9116 an existing archive, use @samp{tar --list --file=@var{archive-name}}.
9117 This may not work on some devices.
9119 Records are separated by gaps, which waste space on the archive media.
9120 If you are archiving on magnetic tape, using a larger blocking factor
9121 (and therefore larger records) provides faster throughput and allows you
9122 to fit more data on a tape (because there are fewer gaps). If you are
9123 archiving on cartridge, a very large blocking factor (say 126 or more)
9124 greatly increases performance. A smaller blocking factor, on the other
9125 hand, may be useful when archiving small files, to avoid archiving lots
9126 of nulls as @command{tar} fills out the archive to the end of the record.
9127 In general, the ideal record size depends on the size of the
9128 inter-record gaps on the tape you are using, and the average size of the
9129 files you are archiving. @xref{create}, for information on
9132 @FIXME{Need example of using a cartridge with blocking factor=126 or more.}
9134 Archives with blocking factors larger than 20 cannot be read
9135 by very old versions of @command{tar}, or by some newer versions
9136 of @command{tar} running on old machines with small address spaces.
9137 With @GNUTAR{}, the blocking factor of an archive is limited
9138 only by the maximum record size of the device containing the archive,
9139 or by the amount of available virtual memory.
9141 Also, on some systems, not using adequate blocking factors, as sometimes
9142 imposed by the device drivers, may yield unexpected diagnostics. For
9143 example, this has been reported:
9146 Cannot write to /dev/dlt: Invalid argument
9150 In such cases, it sometimes happen that the @command{tar} bundled by
9151 the system is aware of block size idiosyncrasies, while @GNUTAR{}
9152 requires an explicit specification for the block size,
9153 which it cannot guess. This yields some people to consider
9154 @GNUTAR{} is misbehaving, because by comparison,
9155 @cite{the bundle @command{tar} works OK}. Adding @w{@kbd{-b 256}},
9156 for example, might resolve the problem.
9158 If you use a non-default blocking factor when you create an archive, you
9159 must specify the same blocking factor when you modify that archive. Some
9160 archive devices will also require you to specify the blocking factor when
9161 reading that archive, however this is not typically the case. Usually, you
9162 can use @option{--list} (@option{-t}) without specifying a blocking factor---@command{tar}
9163 reports a non-default record size and then lists the archive members as
9164 it would normally. To extract files from an archive with a non-standard
9165 blocking factor (particularly if you're not sure what the blocking factor
9166 is), you can usually use the @option{--read-full-records} (@option{-B}) option while
9167 specifying a blocking factor larger then the blocking factor of the archive
9168 (ie. @samp{tar --extract --read-full-records --blocking-factor=300}.
9169 @xref{list}, for more information on the @option{--list} (@option{-t})
9170 operation. @xref{Reading}, for a more detailed explanation of that option.
9173 @item --blocking-factor=@var{number}
9174 @itemx -b @var{number}
9175 Specifies the blocking factor of an archive. Can be used with any
9176 operation, but is usually not necessary with @option{--list} (@option{-t}).
9182 @item -b @var{blocks}
9183 @itemx --blocking-factor=@var{blocks}
9184 Set record size to @math{@var{blocks} * 512} bytes.
9186 This option is used to specify a @dfn{blocking factor} for the archive.
9187 When reading or writing the archive, @command{tar}, will do reads and writes
9188 of the archive in records of @math{@var{block}*512} bytes. This is true
9189 even when the archive is compressed. Some devices requires that all
9190 write operations be a multiple of a certain size, and so, @command{tar}
9191 pads the archive out to the next record boundary.
9193 The default blocking factor is set when @command{tar} is compiled, and is
9194 typically 20. Blocking factors larger than 20 cannot be read by very
9195 old versions of @command{tar}, or by some newer versions of @command{tar}
9196 running on old machines with small address spaces.
9198 With a magnetic tape, larger records give faster throughput and fit
9199 more data on a tape (because there are fewer inter-record gaps).
9200 If the archive is in a disk file or a pipe, you may want to specify
9201 a smaller blocking factor, since a large one will result in a large
9202 number of null bytes at the end of the archive.
9204 When writing cartridge or other streaming tapes, a much larger
9205 blocking factor (say 126 or more) will greatly increase performance.
9206 However, you must specify the same blocking factor when reading or
9207 updating the archive.
9209 Apparently, Exabyte drives have a physical block size of 8K bytes.
9210 If we choose our blocksize as a multiple of 8k bytes, then the problem
9211 seems to disappear. Id est, we are using block size of 112 right
9212 now, and we haven't had the problem since we switched@dots{}
9214 With @GNUTAR{} the blocking factor is limited only
9215 by the maximum record size of the device containing the archive, or by
9216 the amount of available virtual memory.
9218 However, deblocking or reblocking is virtually avoided in a special
9219 case which often occurs in practice, but which requires all the
9220 following conditions to be simultaneously true:
9223 the archive is subject to a compression option,
9225 the archive is not handled through standard input or output, nor
9226 redirected nor piped,
9228 the archive is directly handled to a local disk, instead of any special
9231 @option{--blocking-factor} is not explicitly specified on the @command{tar}
9235 If the output goes directly to a local disk, and not through
9236 stdout, then the last write is not extended to a full record size.
9237 Otherwise, reblocking occurs. Here are a few other remarks on this
9243 @command{gzip} will complain about trailing garbage if asked to
9244 uncompress a compressed archive on tape, there is an option to turn
9245 the message off, but it breaks the regularity of simply having to use
9246 @samp{@var{prog} -d} for decompression. It would be nice if gzip was
9247 silently ignoring any number of trailing zeros. I'll ask Jean-loup
9248 Gailly, by sending a copy of this message to him.
9251 @command{compress} does not show this problem, but as Jean-loup pointed
9252 out to Michael, @samp{compress -d} silently adds garbage after
9253 the result of decompression, which tar ignores because it already
9254 recognized its end-of-file indicator. So this bug may be safely
9258 @samp{gzip -d -q} will be silent about the trailing zeros indeed,
9259 but will still return an exit status of 2 which tar reports in turn.
9260 @command{tar} might ignore the exit status returned, but I hate doing
9261 that, as it weakens the protection @command{tar} offers users against
9262 other possible problems at decompression time. If @command{gzip} was
9263 silently skipping trailing zeros @emph{and} also avoiding setting the
9264 exit status in this innocuous case, that would solve this situation.
9267 @command{tar} should become more solid at not stopping to read a pipe at
9268 the first null block encountered. This inelegantly breaks the pipe.
9269 @command{tar} should rather drain the pipe out before exiting itself.
9272 @opindex ignore-zeros, short description
9274 @itemx --ignore-zeros
9275 Ignore blocks of zeros in archive (means EOF).
9277 The @option{--ignore-zeros} (@option{-i}) option causes @command{tar} to ignore blocks
9278 of zeros in the archive. Normally a block of zeros indicates the
9279 end of the archive, but when reading a damaged archive, or one which
9280 was created by concatenating several archives together, this option
9281 allows @command{tar} to read the entire archive. This option is not on
9282 by default because many versions of @command{tar} write garbage after
9285 Note that this option causes @command{tar} to read to the end of the
9286 archive file, which may sometimes avoid problems when multiple files
9287 are stored on a single physical tape.
9289 @opindex read-full-records, short description
9291 @itemx --read-full-records
9292 Reblock as we read (for reading 4.2BSD pipes).
9294 If @option{--read-full-records} is used, @command{tar}
9295 will not panic if an attempt to read a record from the archive does
9296 not return a full record. Instead, @command{tar} will keep reading
9297 until it has obtained a full
9300 This option is turned on by default when @command{tar} is reading
9301 an archive from standard input, or from a remote machine. This is
9302 because on BSD Unix systems, a read of a pipe will return however
9303 much happens to be in the pipe, even if it is less than @command{tar}
9304 requested. If this option was not used, @command{tar} would fail as
9305 soon as it read an incomplete record from the pipe.
9307 This option is also useful with the commands for updating an archive.
9313 @FIXME{Appropriate options should be moved here from elsewhere.}
9315 @cindex blocking factor
9316 @cindex tape blocking
9318 When handling various tapes or cartridges, you have to take care of
9319 selecting a proper blocking, that is, the number of disk blocks you
9320 put together as a single tape block on the tape, without intervening
9321 tape gaps. A @dfn{tape gap} is a small landing area on the tape
9322 with no information on it, used for decelerating the tape to a
9323 full stop, and for later regaining the reading or writing speed.
9324 When the tape driver starts reading a record, the record has to
9325 be read whole without stopping, as a tape gap is needed to stop the
9326 tape motion without loosing information.
9328 @cindex Exabyte blocking
9329 @cindex DAT blocking
9330 Using higher blocking (putting more disk blocks per tape block) will use
9331 the tape more efficiently as there will be less tape gaps. But reading
9332 such tapes may be more difficult for the system, as more memory will be
9333 required to receive at once the whole record. Further, if there is a
9334 reading error on a huge record, this is less likely that the system will
9335 succeed in recovering the information. So, blocking should not be too
9336 low, nor it should be too high. @command{tar} uses by default a blocking of
9337 20 for historical reasons, and it does not really matter when reading or
9338 writing to disk. Current tape technology would easily accommodate higher
9339 blockings. Sun recommends a blocking of 126 for Exabytes and 96 for DATs.
9340 We were told that for some DLT drives, the blocking should be a multiple
9341 of 4Kb, preferably 64Kb (@w{@kbd{-b 128}}) or 256 for decent performance.
9342 Other manufacturers may use different recommendations for the same tapes.
9343 This might also depends of the buffering techniques used inside modern
9344 tape controllers. Some imposes a minimum blocking, or a maximum blocking.
9345 Others request blocking to be some exponent of two.
9347 So, there is no fixed rule for blocking. But blocking at read time
9348 should ideally be the same as blocking used at write time. At one place
9349 I know, with a wide variety of equipment, they found it best to use a
9350 blocking of 32 to guarantee that their tapes are fully interchangeable.
9352 I was also told that, for recycled tapes, prior erasure (by the same
9353 drive unit that will be used to create the archives) sometimes lowers
9354 the error rates observed at rewriting time.
9356 I might also use @option{--number-blocks} instead of
9357 @option{--block-number}, so @option{--block} will then expand to
9358 @option{--blocking-factor} unambiguously.
9361 @section Many Archives on One Tape
9363 @FIXME{Appropriate options should be moved here from elsewhere.}
9365 @findex ntape @r{device}
9366 Most tape devices have two entries in the @file{/dev} directory, or
9367 entries that come in pairs, which differ only in the minor number for
9368 this device. Let's take for example @file{/dev/tape}, which often
9369 points to the only or usual tape device of a given system. There might
9370 be a corresponding @file{/dev/nrtape} or @file{/dev/ntape}. The simpler
9371 name is the @emph{rewinding} version of the device, while the name
9372 having @samp{nr} in it is the @emph{no rewinding} version of the same
9375 A rewinding tape device will bring back the tape to its beginning point
9376 automatically when this device is opened or closed. Since @command{tar}
9377 opens the archive file before using it and closes it afterwards, this
9378 means that a simple:
9381 $ @kbd{tar cf /dev/tape @var{directory}}
9385 will reposition the tape to its beginning both prior and after saving
9386 @var{directory} contents to it, thus erasing prior tape contents and
9387 making it so that any subsequent write operation will destroy what has
9390 @cindex tape positioning
9391 So, a rewinding device is normally meant to hold one and only one file.
9392 If you want to put more than one @command{tar} archive on a given tape, you
9393 will need to avoid using the rewinding version of the tape device. You
9394 will also have to pay special attention to tape positioning. Errors in
9395 positioning may overwrite the valuable data already on your tape. Many
9396 people, burnt by past experiences, will only use rewinding devices and
9397 limit themselves to one file per tape, precisely to avoid the risk of
9398 such errors. Be fully aware that writing at the wrong position on a
9399 tape loses all information past this point and most probably until the
9400 end of the tape, and this destroyed information @emph{cannot} be
9403 To save @var{directory-1} as a first archive at the beginning of a
9404 tape, and leave that tape ready for a second archive, you should use:
9407 $ @kbd{mt -f /dev/nrtape rewind}
9408 $ @kbd{tar cf /dev/nrtape @var{directory-1}}
9412 @dfn{Tape marks} are special magnetic patterns written on the tape
9413 media, which are later recognizable by the reading hardware. These
9414 marks are used after each file, when there are many on a single tape.
9415 An empty file (that is to say, two tape marks in a row) signal the
9416 logical end of the tape, after which no file exist. Usually,
9417 non-rewinding tape device drivers will react to the close request issued
9418 by @command{tar} by first writing two tape marks after your archive, and by
9419 backspacing over one of these. So, if you remove the tape at that time
9420 from the tape drive, it is properly terminated. But if you write
9421 another file at the current position, the second tape mark will be
9422 erased by the new information, leaving only one tape mark between files.
9424 So, you may now save @var{directory-2} as a second archive after the
9425 first on the same tape by issuing the command:
9428 $ @kbd{tar cf /dev/nrtape @var{directory-2}}
9432 and so on for all the archives you want to put on the same tape.
9434 Another usual case is that you do not write all the archives the same
9435 day, and you need to remove and store the tape between two archive
9436 sessions. In general, you must remember how many files are already
9437 saved on your tape. Suppose your tape already has 16 files on it, and
9438 that you are ready to write the 17th. You have to take care of skipping
9439 the first 16 tape marks before saving @var{directory-17}, say, by using
9443 $ @kbd{mt -f /dev/nrtape rewind}
9444 $ @kbd{mt -f /dev/nrtape fsf 16}
9445 $ @kbd{tar cf /dev/nrtape @var{directory-17}}
9448 In all the previous examples, we put aside blocking considerations, but
9449 you should do the proper things for that as well. @xref{Blocking}.
9452 * Tape Positioning:: Tape Positions and Tape Marks
9453 * mt:: The @command{mt} Utility
9456 @node Tape Positioning
9457 @subsection Tape Positions and Tape Marks
9460 Just as archives can store more than one file from the file system,
9461 tapes can store more than one archive file. To keep track of where
9462 archive files (or any other type of file stored on tape) begin and
9463 end, tape archive devices write magnetic @dfn{tape marks} on the
9464 archive media. Tape drives write one tape mark between files,
9465 two at the end of all the file entries.
9467 If you think of data as a series of records "rrrr"'s, and tape marks as
9468 "*"'s, a tape might look like the following:
9471 rrrr*rrrrrr*rrrrr*rr*rrrrr**-------------------------
9474 Tape devices read and write tapes using a read/write @dfn{tape
9475 head}---a physical part of the device which can only access one
9476 point on the tape at a time. When you use @command{tar} to read or
9477 write archive data from a tape device, the device will begin reading
9478 or writing from wherever on the tape the tape head happens to be,
9479 regardless of which archive or what part of the archive the tape
9480 head is on. Before writing an archive, you should make sure that no
9481 data on the tape will be overwritten (unless it is no longer needed).
9482 Before reading an archive, you should make sure the tape head is at
9483 the beginning of the archive you want to read. You can do it manually
9484 via @code{mt} utility (@pxref{mt}). The @code{restore} script does
9485 that automatically (@pxref{Scripted Restoration}).
9487 If you want to add new archive file entries to a tape, you should
9488 advance the tape to the end of the existing file entries, backspace
9489 over the last tape mark, and write the new archive file. If you were
9490 to add two archives to the example above, the tape might look like the
9494 rrrr*rrrrrr*rrrrr*rr*rrrrr*rrr*rrrr**----------------
9498 @subsection The @command{mt} Utility
9501 @FIXME{Is it true that this only works on non-block devices?
9502 should explain the difference, (fixed or variable).}
9503 @xref{Blocking Factor}.
9505 You can use the @command{mt} utility to advance or rewind a tape past a
9506 specified number of archive files on the tape. This will allow you
9507 to move to the beginning of an archive before extracting or reading
9508 it, or to the end of all the archives before writing a new one.
9509 @FIXME{Why isn't there an "advance 'til you find two tape marks
9512 The syntax of the @command{mt} command is:
9515 @kbd{mt [-f @var{tapename}] @var{operation} [@var{number}]}
9518 where @var{tapename} is the name of the tape device, @var{number} is
9519 the number of times an operation is performed (with a default of one),
9520 and @var{operation} is one of the following:
9522 @FIXME{is there any use for record operations?}
9527 Writes @var{number} tape marks at the current position on the tape.
9530 Moves tape position forward @var{number} files.
9533 Moves tape position back @var{number} files.
9536 Rewinds the tape. (Ignores @var{number}).
9540 Rewinds the tape and takes the tape device off-line. (Ignores @var{number}).
9543 Prints status information about the tape unit.
9547 @FIXME{Is there a better way to frob the spacing on the list?}
9549 If you don't specify a @var{tapename}, @command{mt} uses the environment
9550 variable @env{TAPE}; if @env{TAPE} is not set, @command{mt} uses the device
9553 @command{mt} returns a 0 exit status when the operation(s) were
9554 successful, 1 if the command was unrecognized, and 2 if an operation
9557 @node Using Multiple Tapes
9558 @section Using Multiple Tapes
9561 Often you might want to write a large archive, one larger than will fit
9562 on the actual tape you are using. In such a case, you can run multiple
9563 @command{tar} commands, but this can be inconvenient, particularly if you
9564 are using options like @option{--exclude=@var{pattern}} or dumping entire file systems.
9565 Therefore, @command{tar} supports multiple tapes automatically.
9567 Use @option{--multi-volume} (@option{-M}) on the command line, and
9568 then @command{tar} will, when it reaches the end of the tape, prompt
9569 for another tape, and continue the archive. Each tape will have an
9570 independent archive, and can be read without needing the other. (As
9571 an exception to this, the file that @command{tar} was archiving when
9572 it ran out of tape will usually be split between the two archives; in
9573 this case you need to extract from the first archive, using
9574 @option{--multi-volume}, and then put in the second tape when
9575 prompted, so @command{tar} can restore both halves of the file.)
9577 @GNUTAR{} multi-volume archives do not use a truly portable format.
9578 You need @GNUTAR{} at both ends to process them properly.
9580 When prompting for a new tape, @command{tar} accepts any of the following
9585 Request @command{tar} to explain possible responses
9587 Request @command{tar} to exit immediately.
9588 @item n @var{file-name}
9589 Request @command{tar} to write the next volume on the file @var{file-name}.
9591 Request @command{tar} to run a subshell. This option can be disabled
9592 by giving @option{--restrict} command line option to @command{tar}.
9594 Request @command{tar} to begin writing the next volume.
9597 (You should only type @samp{y} after you have changed the tape;
9598 otherwise @command{tar} will write over the volume it just finished.)
9600 @cindex End-of-archive info script
9602 @anchor{info-script}
9603 @opindex info-script
9604 @opindex new-volume-script
9605 If you want more elaborate behavior than this, give @command{tar} the
9606 @option{--info-script=@var{script-name}}
9607 (@option{--new-volume-script=@var{script-name}}, @option{-F
9608 @var{script-name}}) option. The file @var{script-name} is expected to
9609 be a program (or shell script) to be run instead of the normal
9610 prompting procedure. It is executed without any command line
9611 arguments. Additional data is passed to it via the following
9612 environment variables:
9615 @vrindex TAR_VERSION, info script environment variable
9617 @GNUTAR{} version number.
9619 @vrindex TAR_ARCHIVE, info script environment variable
9621 The name of the archive @command{tar} is processing.
9623 @vrindex TAR_VOLUME, info script environment variable
9625 Ordinal number of the volume @command{tar} is about to start.
9627 @vrindex TAR_SUBCOMMAND, info script environment variable
9628 @item TAR_SUBCOMMAND
9629 Short option describing the operation @command{tar} is executed.
9630 @xref{Operations}, for a complete list of subcommand options.
9632 @vrindex TAR_FORMAT, info script environment variable
9634 Format of the archive being processed. @xref{Formats}, for a complete
9635 list of archive format names.
9638 The info script can instruct @command{tar} to use new archive name,
9639 by writing in to file descriptor 3 (see below for an
9642 If the info script fails, @command{tar} exits; otherwise, it begins
9643 writing the next volume.
9645 The method @command{tar} uses to detect end of tape is not perfect, and
9646 fails on some operating systems or on some devices. You can use the
9647 @option{--tape-length=@var{size}} (@option{-L @var{size}}) option if
9648 @command{tar} can't detect the end of the tape itself. This option
9649 selects @option{--multi-volume} (@option{-M}) automatically. The
9650 @var{size} argument should then be the usable size of the tape in
9651 units of 1024 bytes. But for many devices, and floppy disks in
9652 particular, this option is never required for real, as far as we know.
9654 @cindex Volume number file
9658 The volume number used by @command{tar} in its tape-change prompt
9659 can be changed; if you give the
9660 @option{--volno-file=@var{file-of-number}} option, then
9661 @var{file-of-number} should be an unexisting file to be created, or
9662 else, a file already containing a decimal number. That number will be
9663 used as the volume number of the first volume written. When
9664 @command{tar} is finished, it will rewrite the file with the
9665 now-current volume number. (This does not change the volume number
9666 written on a tape label, as per @ref{label}, it @emph{only} affects
9667 the number used in the prompt.)
9669 If you want @command{tar} to cycle through a series of files or tape
9670 drives, there are three approaches to choose from. First of all, you
9671 can give @command{tar} multiple @option{--file} options. In this case
9672 the specified files will be used, in sequence, as the successive
9673 volumes of the archive. Only when the first one in the sequence needs
9674 to be used again will @command{tar} prompt for a tape change (or run
9675 the info script). Secondly, you can use the @samp{n} response to the
9676 tape-change prompt, and, finally, you can use an info script, that
9677 writes new archive name to file descriptor. The following example
9678 illustrates this approach:
9683 echo Preparing volume $TAR_VOLUME of $TAR_ARCHIVE.
9685 name=`expr $TAR_ARCHIVE : '\(.*\)-.*'`
9686 case $TAR_SUBCOMMAND in
9688 -d|-x|-t) test -r $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME || exit 1
9693 echo $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME >&3
9697 Each volume of a multi-volume archive is an independent @command{tar}
9698 archive, complete in itself. For example, you can list or extract any
9699 volume alone; just don't specify @option{--multi-volume}
9700 (@option{-M}). However, if one file in the archive is split across
9701 volumes, the only way to extract it successfully is with a
9702 multi-volume extract command @option{--extract --multi-volume}
9703 (@option{-xM}) starting on or before the volume where the file begins.
9705 For example, let's presume someone has two tape drives on a system
9706 named @file{/dev/tape0} and @file{/dev/tape1}. For having @GNUTAR{}
9707 to switch to the second drive when it needs to write the
9708 second tape, and then back to the first tape, etc., just do either of:
9711 $ @kbd{tar --create --multi-volume --file=/dev/tape0 --file=/dev/tape1 @var{files}}
9712 $ @kbd{tar cMff /dev/tape0 /dev/tape1 @var{files}}
9716 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
9717 * Tape Files:: Tape Files
9718 * Tarcat:: Concatenate Volumes into a Single Archive
9722 @node Multi-Volume Archives
9723 @subsection Archives Longer than One Tape or Disk
9724 @cindex Multi-volume archives
9727 @opindex multi-volume
9728 To create an archive that is larger than will fit on a single unit of
9729 the media, use the @option{--multi-volume} (@option{-M}) option in conjunction with
9730 the @option{--create} option (@pxref{create}). A @dfn{multi-volume}
9731 archive can be manipulated like any other archive (provided the
9732 @option{--multi-volume} option is specified), but is stored on more
9733 than one tape or disk.
9735 When you specify @option{--multi-volume}, @command{tar} does not report an
9736 error when it comes to the end of an archive volume (when reading), or
9737 the end of the media (when writing). Instead, it prompts you to load
9738 a new storage volume. If the archive is on a magnetic tape, you
9739 should change tapes when you see the prompt; if the archive is on a
9740 floppy disk, you should change disks; etc.
9742 You can read each individual volume of a multi-volume archive as if it
9743 were an archive by itself. For example, to list the contents of one
9744 volume, use @option{--list}, without @option{--multi-volume} specified.
9745 To extract an archive member from one volume (assuming it is described
9746 that volume), use @option{--extract}, again without
9747 @option{--multi-volume}.
9749 If an archive member is split across volumes (ie. its entry begins on
9750 one volume of the media and ends on another), you need to specify
9751 @option{--multi-volume} to extract it successfully. In this case, you
9752 should load the volume where the archive member starts, and use
9753 @samp{tar --extract --multi-volume}---@command{tar} will prompt for later
9754 volumes as it needs them. @xref{extracting archives}, for more
9755 information about extracting archives.
9757 @option{--info-script=@var{script-name}}
9758 (@option{--new-volume-script=@var{script-name}}, @option{-F
9759 @var{script-name}}) (@pxref{info-script}) is like
9760 @option{--multi-volume} (@option{-M}), except that @command{tar} does
9761 not prompt you directly to change media volumes when a volume is
9762 full---instead, @command{tar} runs commands you have stored in
9763 @var{script-name}. For example, this option can be used to eject
9764 cassettes, or to broadcast messages such as @samp{Someone please come
9765 change my tape} when performing unattended backups. When
9766 @var{script-name} is done, @command{tar} will assume that the media
9769 Multi-volume archives can be modified like any other archive. To add
9770 files to a multi-volume archive, you need to only mount the last
9771 volume of the archive media (and new volumes, if needed). For all
9772 other operations, you need to use the entire archive.
9774 If a multi-volume archive was labeled using
9775 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9776 (@pxref{label}) when it was created, @command{tar} will not
9777 automatically label volumes which are added later. To label
9778 subsequent volumes, specify @option{--label=@var{archive-label}} again
9779 in conjunction with the @option{--append}, @option{--update} or
9780 @option{--concatenate} operation.
9782 @cindex Labeling multi-volume archives
9785 @FIXME{There should be a sample program here, including an exit
9786 before end. Is the exit status even checked in tar? :-(}
9789 @item --multi-volume
9791 Creates a multi-volume archive, when used in conjunction with
9792 @option{--create} (@option{-c}). To perform any other operation on a multi-volume
9793 archive, specify @option{--multi-volume} in conjunction with that
9796 @item --info-script=@var{program-file}
9797 @itemx --new-volume-script=@var{program-file}
9798 @itemx -F @var{program-file}
9799 Creates a multi-volume archive via a script. Used in conjunction with
9800 @option{--create} (@option{-c}). @xref{info-script}, dor a detailed discussion.
9803 Beware that there is @emph{no} real standard about the proper way, for
9804 a @command{tar} archive, to span volume boundaries. If you have a
9805 multi-volume created by some vendor's @command{tar}, there is almost
9806 no chance you could read all the volumes with @GNUTAR{}.
9807 The converse is also true: you may not expect
9808 multi-volume archives created by @GNUTAR{} to be
9809 fully recovered by vendor's @command{tar}. Since there is little
9810 chance that, in mixed system configurations, some vendor's
9811 @command{tar} will work on another vendor's machine, and there is a
9812 great chance that @GNUTAR{} will work on most of
9813 them, your best bet is to install @GNUTAR{} on all
9814 machines between which you know exchange of files is possible.
9817 @subsection Tape Files
9820 To give the archive a name which will be recorded in it, use the
9821 @option{--label=@var{volume-label}} (@option{-V @var{volume-label}})
9822 option. This will write a special block identifying
9823 @var{volume-label} as the name of the archive to the front of the
9824 archive which will be displayed when the archive is listed with
9825 @option{--list}. If you are creating a multi-volume archive with
9826 @option{--multi-volume} (@pxref{Using Multiple Tapes}), then the
9827 volume label will have @samp{Volume @var{nnn}} appended to the name
9828 you give, where @var{nnn} is the number of the volume of the archive.
9829 (If you use the @option{--label=@var{volume-label}}) option when
9830 reading an archive, it checks to make sure the label on the tape
9831 matches the one you give. @xref{label}.
9833 When @command{tar} writes an archive to tape, it creates a single
9834 tape file. If multiple archives are written to the same tape, one
9835 after the other, they each get written as separate tape files. When
9836 extracting, it is necessary to position the tape at the right place
9837 before running @command{tar}. To do this, use the @command{mt} command.
9838 For more information on the @command{mt} command and on the organization
9839 of tapes into a sequence of tape files, see @ref{mt}.
9841 People seem to often do:
9844 @kbd{--label="@var{some-prefix} `date +@var{some-format}`"}
9847 or such, for pushing a common date in all volumes or an archive set.
9850 @subsection Concatenate Volumes into a Single Archive
9853 Sometimes it is necessary to convert existing @GNUTAR{} multi-volume
9854 archive to a single @command{tar} archive. Simply concatenating all
9855 volumes into one will not work, since each volume carries an additional
9856 information at the beginning. @GNUTAR{} is shipped with the shell
9857 script @command{tarcat} designed for this purpose.
9859 The script takes a list of files comprising a multi-volume archive
9860 and creates the resulting archive at the standard output. For example:
9863 @kbd{tarcat vol.1 vol.2 vol.3 | tar tf -}
9866 The script implements a simple heuristics to determine the format of
9867 the first volume file and to decide how to process the rest of the
9868 files. However, it makes no attempt to verify whether the files are
9869 given in order or even if they are valid @command{tar} archives.
9870 It uses @command{dd} and does not filter its standard error, so you
9871 will usually see lots of spurious messages.
9873 @FIXME{The script is not installed. Should we install it?}
9876 @section Including a Label in the Archive
9877 @cindex Labeling an archive
9878 @cindex Labels on the archive media
9882 To avoid problems caused by misplaced paper labels on the archive
9883 media, you can include a @dfn{label} entry---an archive member which
9884 contains the name of the archive---in the archive itself. Use the
9885 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9886 option in conjunction with the @option{--create} operation to include
9887 a label entry in the archive as it is being created.
9890 @item --label=@var{archive-label}
9891 @itemx -V @var{archive-label}
9892 Includes an @dfn{archive-label} at the beginning of the archive when
9893 the archive is being created, when used in conjunction with the
9894 @option{--create} operation. Checks to make sure the archive label
9895 matches the one specified (when used in conjunction with any other
9899 If you create an archive using both
9900 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9901 and @option{--multi-volume} (@option{-M}), each volume of the archive
9902 will have an archive label of the form @samp{@var{archive-label}
9903 Volume @var{n}}, where @var{n} is 1 for the first volume, 2 for the
9904 next, and so on. @xref{Using Multiple Tapes}, for information on
9905 creating multiple volume archives.
9907 @cindex Volume label, listing
9908 @cindex Listing volume label
9909 The volume label will be displayed by @option{--list} along with
9910 the file contents. If verbose display is requested, it will also be
9911 explicitely marked as in the example below:
9915 $ @kbd{tar --verbose --list --file=iamanarchive}
9916 V--------- 0 0 0 1992-03-07 12:01 iamalabel--Volume Header--
9917 -rw-r--r-- ringo user 40 1990-05-21 13:30 iamafilename
9922 @anchor{--test-label option}
9923 However, @option{--list} option will cause listing entire
9924 contents of the archive, which may be undesirable (for example, if the
9925 archive is stored on a tape). You can request checking only the volume
9926 by specifying @option{--test-label} option. This option reads only the
9927 first block of an archive, so it can be used with slow storage
9928 devices. For example:
9932 $ @kbd{tar --test-label --file=iamanarchive}
9937 If @option{--test-label} is used with a single command line
9938 argument, @command{tar} compares the volume label with the
9939 argument. It exits with code 0 if the two strings match, and with code
9940 2 otherwise. In this case no output is displayed. For example:
9944 $ @kbd{tar --test-label --file=iamanarchive 'iamalable'}
9946 $ @kbd{tar --test-label --file=iamanarchive 'iamalable' alabel}
9951 If you request any operation, other than @option{--create}, along
9952 with using @option{--label} option, @command{tar} will first check if
9953 the archive label matches the one specified and will refuse to proceed
9954 if it does not. Use this as a safety precaution to avoid accidentally
9955 overwriting existing archives. For example, if you wish to add files
9956 to @file{archive}, presumably labelled with string @samp{My volume},
9961 $ @kbd{tar -rf archive --label 'My volume' .}
9962 tar: Archive not labeled to match `My volume'
9967 in case its label does not match. This will work even if
9968 @file{archive} is not labelled at all.
9970 Similarly, @command{tar} will refuse to list or extract the
9971 archive if its label doesn't match the @var{archive-label}
9972 specified. In those cases, @var{archive-label} argument is interpreted
9973 as a globbing-style pattern which must match the actual magnetic
9974 volume label. @xref{exclude}, for a precise description of how match
9975 is attempted@footnote{Previous versions of @command{tar} used full
9976 regular expression matching, or before that, only exact string
9977 matching, instead of wildcard matchers. We decided for the sake of
9978 simplicity to use a uniform matching device through
9979 @command{tar}.}. If the switch @option{--multi-volume} (@option{-M}) is being used,
9980 the volume label matcher will also suffix @var{archive-label} by
9981 @w{@samp{ Volume [1-9]*}} if the initial match fails, before giving
9982 up. Since the volume numbering is automatically added in labels at
9983 creation time, it sounded logical to equally help the user taking care
9984 of it when the archive is being read.
9986 The @option{--label} was once called @option{--volume}, but is not
9987 available under that name anymore.
9989 You can also use @option{--label} to get a common information on
9990 all tapes of a series. For having this information different in each
9991 series created through a single script used on a regular basis, just
9992 manage to get some date string as part of the label. For example:
9996 $ @kbd{tar cfMV /dev/tape "Daily backup for `date +%Y-%m-%d`"}
9997 $ @kbd{tar --create --file=/dev/tape --multi-volume \
9998 --volume="Daily backup for `date +%Y-%m-%d`"}
10002 Also note that each label has its own date and time, which corresponds
10003 to when @GNUTAR{} initially attempted to write it,
10004 often soon after the operator launches @command{tar} or types the
10005 carriage return telling that the next tape is ready. Comparing date
10006 labels does give an idea of tape throughput only if the delays for
10007 rewinding tapes and the operator switching them were negligible, which
10008 is usually not the case.
10011 @section Verifying Data as It is Stored
10012 @cindex Verifying a write operation
10013 @cindex Double-checking a write operation
10018 @opindex verify, short description
10019 Attempt to verify the archive after writing.
10022 This option causes @command{tar} to verify the archive after writing it.
10023 Each volume is checked after it is written, and any discrepancies
10024 are recorded on the standard error output.
10026 Verification requires that the archive be on a back-space-able medium.
10027 This means pipes, some cartridge tape drives, and some other devices
10028 cannot be verified.
10030 You can insure the accuracy of an archive by comparing files in the
10031 system with archive members. @command{tar} can compare an archive to the
10032 file system as the archive is being written, to verify a write
10033 operation, or can compare a previously written archive, to insure that
10036 @opindex verify, using with @option{--create}
10037 @opindex create, using with @option{--verify}
10038 To check for discrepancies in an archive immediately after it is
10039 written, use the @option{--verify} (@option{-W}) option in conjunction with
10040 the @option{--create} operation. When this option is
10041 specified, @command{tar} checks archive members against their counterparts
10042 in the file system, and reports discrepancies on the standard error.
10044 To verify an archive, you must be able to read it from before the end
10045 of the last written entry. This option is useful for detecting data
10046 errors on some tapes. Archives written to pipes, some cartridge tape
10047 drives, and some other devices cannot be verified.
10049 One can explicitly compare an already made archive with the file
10050 system by using the @option{--compare} (@option{--diff}, @option{-d})
10051 option, instead of using the more automatic @option{--verify} option.
10054 Note that these two options have a slightly different intent. The
10055 @option{--compare} option checks how identical are the logical contents of some
10056 archive with what is on your disks, while the @option{--verify} option is
10057 really for checking if the physical contents agree and if the recording
10058 media itself is of dependable quality. So, for the @option{--verify}
10059 operation, @command{tar} tries to defeat all in-memory cache pertaining to
10060 the archive, while it lets the speed optimization undisturbed for the
10061 @option{--compare} option. If you nevertheless use @option{--compare} for
10062 media verification, you may have to defeat the in-memory cache yourself,
10063 maybe by opening and reclosing the door latch of your recording unit,
10064 forcing some doubt in your operating system about the fact this is really
10065 the same volume as the one just written or read.
10067 The @option{--verify} option would not be necessary if drivers were indeed
10068 able to detect dependably all write failures. This sometimes require many
10069 magnetic heads, some able to read after the writes occurred. One would
10070 not say that drivers unable to detect all cases are necessarily flawed,
10071 as long as programming is concerned.
10073 The @option{--verify} (@option{-W}) option will not work in
10074 conjunction with the @option{--multi-volume} (@option{-M}) option or
10075 the @option{--append} (@option{-r}), @option{--update} (@option{-u})
10076 and @option{--delete} operations. @xref{Operations}, for more
10077 information on these operations.
10079 Also, since @command{tar} normally strips leading @samp{/} from file
10080 names (@pxref{absolute}), a command like @samp{tar --verify -cf
10081 /tmp/foo.tar /etc} will work as desired only if the working directory is
10082 @file{/}, as @command{tar} uses the archive's relative member names
10083 (e.g., @file{etc/motd}) when verifying the archive.
10085 @node Write Protection
10086 @section Write Protection
10088 Almost all tapes and diskettes, and in a few rare cases, even disks can
10089 be @dfn{write protected}, to protect data on them from being changed.
10090 Once an archive is written, you should write protect the media to prevent
10091 the archive from being accidentally overwritten or deleted. (This will
10092 protect the archive from being changed with a tape or floppy drive---it
10093 will not protect it from magnet fields or other physical hazards).
10095 The write protection device itself is usually an integral part of the
10096 physical media, and can be a two position (write enabled/write
10097 disabled) switch, a notch which can be popped out or covered, a ring
10098 which can be removed from the center of a tape reel, or some other
10099 changeable feature.
10104 This appendix lists some important user-visible changes between
10105 version @GNUTAR{} @value{VERSION} and previous versions. An up-to-date
10106 version of this document is available at
10107 @uref{http://www.gnu.org/@/software/@/tar/manual/changes.html,the
10108 @GNUTAR{} documentation page}.
10111 @item Use of globbing patterns when listing and extracting.
10113 Previous versions of GNU tar assumed shell-style globbing when
10114 extracting from or listing an archive. For example:
10117 $ @kbd{tar xf foo.tar '*.c'}
10120 would extract all files whose names end in @samp{.c}. This behavior
10121 was not documented and was incompatible with traditional tar
10122 implementations. Therefore, starting from version 1.15.91, GNU tar
10123 no longer uses globbing by default. For example, the above invocation
10124 is now interpreted as a request to extract from the archive the file
10127 To facilitate transition to the new behavior for those users who got
10128 used to the previous incorrect one, @command{tar} will print a warning
10129 if it finds out that a requested member was not found in the archive
10130 and its name looks like a globbing pattern. For example:
10133 $ @kbd{tar xf foo.tar '*.c'}
10134 tar: Pattern matching characters used in file names. Please,
10135 tar: use --wildcards to enable pattern matching, or --no-wildcards to
10136 tar: suppress this warning.
10137 tar: *.c: Not found in archive
10138 tar: Error exit delayed from previous errors
10141 To treat member names as globbing patterns, use --wildcards option.
10142 If you want to tar to mimic the behavior of versions prior to 1.15.91,
10143 add this option to your @env{TAR_OPTIONS} variable.
10145 @xref{wildcards}, for the detailed discussion of the use of globbing
10146 patterns by @GNUTAR{}.
10148 @item Use of short option @option{-o}.
10150 Earlier versions of @GNUTAR{} understood @option{-o} command line
10151 option as a synonym for @option{--old-archive}.
10153 @GNUTAR{} starting from version 1.13.90 understands this option as
10154 a synonym for @option{--no-same-owner}. This is compatible with
10155 UNIX98 @command{tar} implementations.
10157 However, to facilitate transition, @option{-o} option retains its
10158 old semantics when it is used with one of archive-creation commands.
10159 Users are encouraged to use @option{--format=oldgnu} instead.
10161 It is especially important, since versions of @acronym{GNU} Automake
10162 up to and including 1.8.4 invoke tar with this option to produce
10163 distribution tarballs. @xref{Formats,v7}, for the detailed discussion
10164 of this issue and its implications.
10166 @FIXME{Change the first argument to tar-formats if and when Automake
10167 people accept my patch to the documentation, and the new Automake is
10168 out --Sergey 2006-05-25}.
10169 @xref{Options, tar-v7, Changing Automake's Behavior,
10170 automake, GNU Automake}, for a description on how to use various
10171 archive formats with @command{automake}.
10173 Future versions of @GNUTAR{} will understand @option{-o} only as a
10174 synonym for @option{--no-same-owner}.
10176 @item Use of short option @option{-l}
10178 Earlier versions of @GNUTAR{} understood @option{-l} option as a
10179 synonym for @option{--one-file-system}. Since such usage contradicted
10180 to UNIX98 specification and harmed compatibility with other
10181 implementation, it was declared deprecated in version 1.14. However,
10182 to facilitate transition to its new semantics, it was supported by
10183 versions 1.15 and 1.15.90. The present use of @option{-l} as a short
10184 variant of @option{--check-links} was introduced in version 1.15.91.
10186 @item Use of options @option{--portability} and @option{--old-archive}
10188 These options are deprecated. Please use @option{--format=v7} instead.
10190 @item Use of option @option{--posix}
10192 This option is deprecated. Please use @option{--format=posix} instead.
10195 @node Configuring Help Summary
10196 @appendix Configuring Help Summary
10198 Running @kbd{tar --help} displays the short @command{tar} option
10199 summary (@pxref{help}). This summary is organised by @dfn{groups} of
10200 semantically close options. The options within each group are printed
10201 in the following order: a short option, eventually followed by a list
10202 of corresponding long option names, followed by a short description of
10203 the option. For example, here is an excerpt from the actual @kbd{tar
10207 Main operation mode:
10209 -A, --catenate, --concatenate append tar files to an archive
10210 -c, --create create a new archive
10211 -d, --diff, --compare find differences between archive and
10213 --delete delete from the archive
10216 @vrindex ARGP_HELP_FMT, environment variable
10217 The exact visual representation of the help output is configurable via
10218 @env{ARGP_HELP_FMT} environment variable. The value of this variable
10219 is a comma-separated list of @dfn{format variable} assignments. There
10220 are two kinds of format variables. An @dfn{offset variable} keeps the
10221 offset of some part of help output text from the leftmost column on
10222 the screen. A @dfn{boolean} variable is a flag that toggles some
10223 output feature on or off. Depending on the type of the corresponding
10224 variable, there are two kinds of assignments:
10227 @item Offset assignment
10229 The assignment to an offset variable has the following syntax:
10232 @var{variable}=@var{value}
10236 where @var{variable} is the variable name, and @var{value} is a
10237 numeric value to be assigned to the variable.
10239 @item Boolean assignment
10241 To assign @code{true} value to a variable, simply put this variable name. To
10242 assign @code{false} value, prefix the variable name with @samp{no-}. For
10247 # Assign @code{true} value:
10249 # Assign @code{false} value:
10255 Following variables are declared:
10257 @deftypevr {Help Output} boolean dup-args
10258 If true, arguments for an option are shown with both short and long
10259 options, even when a given option has both forms, for example:
10262 -f ARCHIVE, --file=ARCHIVE use archive file or device ARCHIVE
10265 If false, then if an option has both short and long forms, the
10266 argument is only shown with the long one, for example:
10269 -f, --file=ARCHIVE use archive file or device ARCHIVE
10273 and a message indicating that the argument is applicable to both
10274 forms is printed below the options. This message can be disabled
10275 using @code{dup-args-note} (see below).
10277 The default is false.
10280 @deftypevr {Help Output} boolean dup-args-note
10281 If this variable is true, which is the default, the following notice
10282 is displayed at the end of the help output:
10285 Mandatory or optional arguments to long options are also mandatory or
10286 optional for any corresponding short options.
10289 Setting @code{no-dup-args-note} inhibits this message. Normally, only one of
10290 variables @code{dup-args} or @code{dup-args-note} should be set.
10293 @deftypevr {Help Output} offset short-opt-col
10294 Column in which short options start. Default is 2.
10298 $ @kbd{tar --help|grep ARCHIVE}
10299 -f, --file=ARCHIVE use archive file or device ARCHIVE
10300 $ @kbd{ARGP_HELP_FMT=short-opt-col=6 tar --help|grep ARCHIVE}
10301 -f, --file=ARCHIVE use archive file or device ARCHIVE
10306 @deftypevr {Help Output} offset long-opt-col
10307 Column in which long options start. Default is 6. For example:
10311 $ @kbd{tar --help|grep ARCHIVE}
10312 -f, --file=ARCHIVE use archive file or device ARCHIVE
10313 $ @kbd{ARGP_HELP_FMT=long-opt-col=16 tar --help|grep ARCHIVE}
10314 -f, --file=ARCHIVE use archive file or device ARCHIVE
10319 @deftypevr {Help Output} offset doc-opt-col
10320 Column in which @dfn{doc options} start. A doc option isn't actually
10321 an option, but rather an arbitrary piece of documentation that is
10322 displayed in much the same manner as the options. For example, in
10323 the description of @option{--format} option:
10327 -H, --format=FORMAT create archive of the given format.
10329 FORMAT is one of the following:
10331 gnu GNU tar 1.13.x format
10332 oldgnu GNU format as per tar <= 1.12
10333 pax POSIX 1003.1-2001 (pax) format
10335 ustar POSIX 1003.1-1988 (ustar) format
10336 v7 old V7 tar format
10341 the format names are doc options. Thus, if you set
10342 @kbd{ARGP_HELP_FMT=doc-opt-col=6} the above part of the help output
10343 will look as follows:
10347 -H, --format=FORMAT create archive of the given format.
10349 FORMAT is one of the following:
10351 gnu GNU tar 1.13.x format
10352 oldgnu GNU format as per tar <= 1.12
10353 pax POSIX 1003.1-2001 (pax) format
10355 ustar POSIX 1003.1-1988 (ustar) format
10356 v7 old V7 tar format
10361 @deftypevr {Help Output} offset opt-doc-col
10362 Column in which option description starts. Default is 29.
10366 $ @kbd{tar --help|grep ARCHIVE}
10367 -f, --file=ARCHIVE use archive file or device ARCHIVE
10368 $ @kbd{ARGP_HELP_FMT=opt-doc-col=19 tar --help|grep ARCHIVE}
10369 -f, --file=ARCHIVE use archive file or device ARCHIVE
10370 $ @kbd{ARGP_HELP_FMT=opt-doc-col=9 tar --help|grep ARCHIVE}
10372 use archive file or device ARCHIVE
10377 Notice, that the description starts on a separate line if
10378 @code{opt-doc-col} value is too small.
10381 @deftypevr {Help Output} offset header-col
10382 Column in which @dfn{group headers} are printed. A group header is a
10383 descriptive text preceding an option group. For example, in the
10387 Main operation mode:
10389 -A, --catenate, --concatenate append tar files to
10391 -c, --create create a new archive
10394 @samp{Main operation mode:} is the group header.
10396 The default value is 1.
10399 @deftypevr {Help Output} offset usage-indent
10400 Indentation of wrapped usage lines. Affects @option{--usage}
10401 output. Default is 12.
10404 @deftypevr {Help Output} offset rmargin
10405 Right margin of the text output. Used for wrapping.
10410 @include genfile.texi
10412 @node Snapshot Files
10413 @appendix Format of the Incremental Snapshot Files
10414 @include snapshot.texi
10418 @include dumpdir.texi
10420 @node Free Software Needs Free Documentation
10421 @appendix Free Software Needs Free Documentation
10422 @include freemanuals.texi
10424 @node Copying This Manual
10425 @appendix Copying This Manual
10428 * GNU Free Documentation License:: License for copying this manual
10433 @node Index of Command Line Options
10434 @appendix Index of Command Line Options
10436 This appendix contains an index of all @GNUTAR{} long command line
10437 options. The options are listed without the preceeding double-dash.
10440 @item Make sure @emph{all} options are indexed.
10441 @item Provide an index of short options
10455 @c Local variables:
10456 @c texinfo-column-for-description: 32