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 the
37 Invariant Sections being "GNU General Public License", with the
38 Front-Cover Texts being ``A GNU Manual,'' and with the Back-Cover Texts
39 as in (a) below. A copy of the license is included in the section
40 entitled "GNU Free Documentation License".
42 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
43 this GNU Manual. Buying copies from GNU Press supports the FSF in
44 developing GNU and promoting software freedom.''
48 @dircategory Archiving
50 * Tar: (tar). Making tape (or disk) archives.
53 @dircategory Individual utilities
55 * tar: (tar)tar invocation. Invoking @GNUTAR{}.
58 @shorttitlepage @acronym{GNU} @command{tar}
61 @title @acronym{GNU} tar: an archiver tool
62 @subtitle @value{RENDITION} @value{VERSION}, @value{UPDATED}
63 @author John Gilmore, Jay Fenlason et al.
66 @vskip 0pt plus 1filll
72 @top @acronym{GNU} tar: an archiver tool
77 @cindex archiving files
79 The first part of this master menu lists the major nodes in this Info
80 document. The rest of the menu lists all the lower level nodes.
83 @c The master menu, created with texinfo-master-menu, goes here.
84 @c (However, getdate.texi's menu is interpolated by hand.)
93 * Date input formats::
100 * Configuring Help Summary::
103 * Free Software Needs Free Documentation::
104 * Copying This Manual::
105 * Index of Command Line Options::
109 --- The Detailed Node Listing ---
113 * Book Contents:: What this Book Contains
114 * Definitions:: Some Definitions
115 * What tar Does:: What @command{tar} Does
116 * Naming tar Archives:: How @command{tar} Archives are Named
117 * Authors:: @GNUTAR{} Authors
118 * Reports:: Reporting bugs or suggestions
120 Tutorial Introduction to @command{tar}
123 * stylistic conventions::
124 * basic tar options:: Basic @command{tar} Operations and Options
125 * frequent operations::
126 * Two Frequent Options::
127 * create:: How to Create Archives
128 * list:: How to List Archives
129 * extract:: How to Extract Members from an Archive
132 Two Frequently Used Options
138 How to Create Archives
140 * prepare for examples::
141 * Creating the archive::
150 How to Extract Members from an Archive
152 * extracting archives::
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
265 * after:: Operating Only on New Files
266 * recurse:: Descending into Directories
267 * one:: Crossing File System Boundaries
269 Reading Names from a File
275 * controlling pattern-matching with exclude::
276 * problems with exclude::
278 Crossing File System Boundaries
280 * directory:: Changing Directory
281 * absolute:: Absolute File Names
285 * General date syntax:: Common rules.
286 * Calendar date items:: 19 Dec 1994.
287 * Time of day items:: 9:20pm.
288 * Time zone items:: @sc{est}, @sc{pdt}, @sc{gmt}, ...
289 * Day of week items:: Monday and others.
290 * Relative items in date strings:: next tuesday, 2 years ago.
291 * Pure numbers in date strings:: 19931219, 1440.
292 * Seconds since the Epoch:: @@1078100502.
293 * Authors of get_date:: Bellovin, Eggert, Salz, Berets, et al.
295 Controlling the Archive Format
297 * Portability:: Making @command{tar} Archives More Portable
298 * Compression:: Using Less Space through Compression
299 * Attributes:: Handling File Attributes
300 * Standard:: The Standard Format
301 * Extensions:: @acronym{GNU} Extensions to the Archive Format
302 * cpio:: Comparison of @command{tar} and @command{cpio}
304 Making @command{tar} Archives More Portable
306 * Portable Names:: Portable Names
307 * dereference:: Symbolic Links
308 * old:: Old V7 Archives
309 * posix:: @acronym{POSIX} archives
310 * Checksumming:: Checksumming Problems
311 * Large or Negative Values:: Large files, negative time stamps, etc.
313 Using Less Space through Compression
315 * gzip:: Creating and Reading Compressed Archives
316 * sparse:: Archiving Sparse Files
318 Tapes and Other Archive Media
320 * Device:: Device selection and switching
321 * Remote Tape Server::
322 * Common Problems and Solutions::
323 * Blocking:: Blocking
324 * Many:: Many archives on one tape
325 * Using Multiple Tapes:: Using Multiple Tapes
326 * label:: Including a Label in the Archive
332 * Format Variations:: Format Variations
333 * Blocking Factor:: The Blocking Factor of an Archive
335 Many Archives on One Tape
337 * Tape Positioning:: Tape Positions and Tape Marks
338 * mt:: The @command{mt} Utility
342 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
343 * Tape Files:: Tape Files
344 * Tarcat:: Concatenate Volumes into a Single Archive
346 GNU tar internals and development
353 * Free Software Needs Free Documentation::
354 * GNU Free Documentation License:: License for copying this manual
360 @chapter Introduction
363 and manipulates @dfn{archives} which are actually collections of
364 many other files; the program provides users with an organized and
365 systematic method for controlling a large amount of data.
366 The name ``tar'' originally came from the phrase ``Tape ARchive'', but
367 archives need not (and these days, typically do not) reside on tapes.
370 * Book Contents:: What this Book Contains
371 * Definitions:: Some Definitions
372 * What tar Does:: What @command{tar} Does
373 * Naming tar Archives:: How @command{tar} Archives are Named
374 * Authors:: @GNUTAR{} Authors
375 * Reports:: Reporting bugs or suggestions
379 @section What this Book Contains
381 The first part of this chapter introduces you to various terms that will
382 recur throughout the book. It also tells you who has worked on @GNUTAR{}
383 and its documentation, and where you should send bug reports
386 The second chapter is a tutorial (@pxref{Tutorial}) which provides a
387 gentle introduction for people who are new to using @command{tar}. It is
388 meant to be self contained, not requiring any reading from subsequent
389 chapters to make sense. It moves from topic to topic in a logical,
390 progressive order, building on information already explained.
392 Although the tutorial is paced and structured to allow beginners to
393 learn how to use @command{tar}, it is not intended solely for beginners.
394 The tutorial explains how to use the three most frequently used
395 operations (@samp{create}, @samp{list}, and @samp{extract}) as well as
396 two frequently used options (@samp{file} and @samp{verbose}). The other
397 chapters do not refer to the tutorial frequently; however, if a section
398 discusses something which is a complex variant of a basic concept, there
399 may be a cross reference to that basic concept. (The entire book,
400 including the tutorial, assumes that the reader understands some basic
401 concepts of using a Unix-type operating system; @pxref{Tutorial}.)
403 The third chapter presents the remaining five operations, and
404 information about using @command{tar} options and option syntax.
406 @FIXME{this sounds more like a @acronym{GNU} Project Manuals Concept [tm] more
407 than the reality. should think about whether this makes sense to say
408 here, or not.} The other chapters are meant to be used as a
409 reference. Each chapter presents everything that needs to be said
410 about a specific topic.
412 One of the chapters (@pxref{Date input formats}) exists in its
413 entirety in other @acronym{GNU} manuals, and is mostly self-contained.
414 In addition, one section of this manual (@pxref{Standard}) contains a
415 big quote which is taken directly from @command{tar} sources.
417 In general, we give both long and short (abbreviated) option names
418 at least once in each section where the relevant option is covered, so
419 that novice readers will become familiar with both styles. (A few
420 options have no short versions, and the relevant sections will
424 @section Some Definitions
428 The @command{tar} program is used to create and manipulate @command{tar}
429 archives. An @dfn{archive} is a single file which contains the contents
430 of many files, while still identifying the names of the files, their
431 owner(s), and so forth. (In addition, archives record access
432 permissions, user and group, size in bytes, and data modification time.
433 Some archives also record the file names in each archived directory, as
434 well as other file and directory information.) You can use @command{tar}
435 to @dfn{create} a new archive in a specified directory.
438 @cindex archive member
441 The files inside an archive are called @dfn{members}. Within this
442 manual, we use the term @dfn{file} to refer only to files accessible in
443 the normal ways (by @command{ls}, @command{cat}, and so forth), and the term
444 @dfn{member} to refer only to the members of an archive. Similarly, a
445 @dfn{file name} is the name of a file, as it resides in the file system,
446 and a @dfn{member name} is the name of an archive member within the
451 The term @dfn{extraction} refers to the process of copying an archive
452 member (or multiple members) into a file in the file system. Extracting
453 all the members of an archive is often called @dfn{extracting the
454 archive}. The term @dfn{unpack} can also be used to refer to the
455 extraction of many or all the members of an archive. Extracting an
456 archive does not destroy the archive's structure, just as creating an
457 archive does not destroy the copies of the files that exist outside of
458 the archive. You may also @dfn{list} the members in a given archive
459 (this is often thought of as ``printing'' them to the standard output,
460 or the command line), or @dfn{append} members to a pre-existing archive.
461 All of these operations can be performed using @command{tar}.
464 @section What @command{tar} Does
467 The @command{tar} program provides the ability to create @command{tar}
468 archives, as well as various other kinds of manipulation. For example,
469 you can use @command{tar} on previously created archives to extract files,
470 to store additional files, or to update or list files which were already
473 Initially, @command{tar} archives were used to store files conveniently on
474 magnetic tape. The name @command{tar} comes from this use; it stands for
475 @code{t}ape @code{ar}chiver. Despite the utility's name, @command{tar} can
476 direct its output to available devices, files, or other programs (using
477 pipes). @command{tar} may even access remote devices or files (as archives).
479 @FIXME{the following table entries need a bit of work..}
481 You can use @command{tar} archives in many ways. We want to stress a few
482 of them: storage, backup, and transportation.
486 Often, @command{tar} archives are used to store related files for
487 convenient file transfer over a network. For example, the
488 @acronym{GNU} Project distributes its software bundled into
489 @command{tar} archives, so that all the files relating to a particular
490 program (or set of related programs) can be transferred as a single
493 A magnetic tape can store several files in sequence. However, the tape
494 has no names for these files; it only knows their relative position on
495 the tape. One way to store several files on one tape and retain their
496 names is by creating a @command{tar} archive. Even when the basic transfer
497 mechanism can keep track of names, as FTP can, the nuisance of handling
498 multiple files, directories, and multiple links makes @command{tar}
501 Archive files are also used for long-term storage. You can think of
502 this as transportation from the present into the future. (It is a
503 science-fiction idiom that you can move through time as well as in
504 space; the idea here is that @command{tar} can be used to move archives in
505 all dimensions, even time!)
508 Because the archive created by @command{tar} is capable of preserving
509 file information and directory structure, @command{tar} is commonly
510 used for performing full and incremental backups of disks. A backup
511 puts a collection of files (possibly pertaining to many users and
512 projects) together on a disk or a tape. This guards against
513 accidental destruction of the information in those files.
514 @GNUTAR{} has special features that allow it to be
515 used to make incremental and full dumps of all the files in a
519 You can create an archive on one system, transfer it to another system,
520 and extract the contents there. This allows you to transport a group of
521 files from one system to another.
524 @node Naming tar Archives
525 @section How @command{tar} Archives are Named
527 Conventionally, @command{tar} archives are given names ending with
528 @samp{.tar}. This is not necessary for @command{tar} to operate properly,
529 but this manual follows that convention in order to accustom readers to
530 it and to make examples more clear.
535 Often, people refer to @command{tar} archives as ``@command{tar} files,'' and
536 archive members as ``files'' or ``entries''. For people familiar with
537 the operation of @command{tar}, this causes no difficulty. However, in
538 this manual, we consistently refer to ``archives'' and ``archive
539 members'' to make learning to use @command{tar} easier for novice users.
542 @section @GNUTAR{} Authors
544 @GNUTAR{} was originally written by John Gilmore,
545 and modified by many people. The @acronym{GNU} enhancements were
546 written by Jay Fenlason, then Joy Kendall, and the whole package has
547 been further maintained by Thomas Bushnell, n/BSG, Fran@,{c}ois
548 Pinard, Paul Eggert, and finally Sergey Poznyakoff with the help of
549 numerous and kind users.
551 We wish to stress that @command{tar} is a collective work, and owes much to
552 all those people who reported problems, offered solutions and other
553 insights, or shared their thoughts and suggestions. An impressive, yet
554 partial list of those contributors can be found in the @file{THANKS}
555 file from the @GNUTAR{} distribution.
557 @FIXME{i want all of these names mentioned, Absolutely. BUT, i'm not
558 sure i want to spell out the history in this detail, at least not for
559 the printed book. i'm just not sure it needs to be said this way.
560 i'll think about it.}
562 @FIXME{History is more important, and surely more interesting, than
563 actual names. Quoting names without history would be meaningless. FP}
565 Jay Fenlason put together a draft of a @GNUTAR{}
566 manual, borrowing notes from the original man page from John Gilmore.
567 This was withdrawn in version 1.11. Thomas Bushnell, n/BSG and Amy
568 Gorin worked on a tutorial and manual for @GNUTAR{}.
569 Fran@,{c}ois Pinard put version 1.11.8 of the manual together by
570 taking information from all these sources and merging them. Melissa
571 Weisshaus finally edited and redesigned the book to create version
572 1.12. @FIXME{update version number as necessary; i'm being
573 optimistic!} @FIXME{Someone [maybe karl berry? maybe bob chassell?
574 maybe melissa? maybe julie sussman?] needs to properly index the
577 For version 1.12, Daniel Hagerty contributed a great deal of technical
578 consulting. In particular, he is the primary author of @ref{Backups}.
580 In July, 2003 @GNUTAR{} was put on CVS at savannah.gnu.org
581 (see @url{http://savannah.gnu.org/projects/tar}), and
582 active development and maintenance work has started
583 again. Currently @GNUTAR{} is being maintained by Paul Eggert, Sergey
584 Poznyakoff and Jeff Bailey.
586 Support for @acronym{POSIX} archives was added by Sergey Poznyakoff.
589 @section Reporting bugs or suggestions
592 @cindex reporting bugs
593 If you find problems or have suggestions about this program or manual,
594 please report them to @file{bug-tar@@gnu.org}.
596 When reporting a bug, please be sure to include as much detail as
597 possible, in order to reproduce it. @FIXME{Be more specific, I'd
598 like to make this node as detailed as 'Bug reporting' node in Emacs
602 @chapter Tutorial Introduction to @command{tar}
604 This chapter guides you through some basic examples of three @command{tar}
605 operations: @option{--create}, @option{--list}, and @option{--extract}. If
606 you already know how to use some other version of @command{tar}, then you
607 may not need to read this chapter. This chapter omits most complicated
608 details about how @command{tar} works.
612 * stylistic conventions::
613 * basic tar options:: Basic @command{tar} Operations and Options
614 * frequent operations::
615 * Two Frequent Options::
616 * create:: How to Create Archives
617 * list:: How to List Archives
618 * extract:: How to Extract Members from an Archive
623 @section Assumptions this Tutorial Makes
625 This chapter is paced to allow beginners to learn about @command{tar}
626 slowly. At the same time, we will try to cover all the basic aspects of
627 these three operations. In order to accomplish both of these tasks, we
628 have made certain assumptions about your knowledge before reading this
629 manual, and the hardware you will be using:
633 Before you start to work through this tutorial, you should understand
634 what the terms ``archive'' and ``archive member'' mean
635 (@pxref{Definitions}). In addition, you should understand something
636 about how Unix-type operating systems work, and you should know how to
637 use some basic utilities. For example, you should know how to create,
638 list, copy, rename, edit, and delete files and directories; how to
639 change between directories; and how to figure out where you are in the
640 file system. You should have some basic understanding of directory
641 structure and how files are named according to which directory they are
642 in. You should understand concepts such as standard output and standard
643 input, what various definitions of the term ``argument'' mean, and the
644 differences between relative and absolute path names. @FIXME{and what
648 This manual assumes that you are working from your own home directory
649 (unless we state otherwise). In this tutorial, you will create a
650 directory to practice @command{tar} commands in. When we show path names,
651 we will assume that those paths are relative to your home directory.
652 For example, my home directory path is @file{/home/fsf/melissa}. All of
653 my examples are in a subdirectory of the directory named by that path
654 name; the subdirectory is called @file{practice}.
657 In general, we show examples of archives which exist on (or can be
658 written to, or worked with from) a directory on a hard disk. In most
659 cases, you could write those archives to, or work with them on any other
660 device, such as a tape drive. However, some of the later examples in
661 the tutorial and next chapter will not work on tape drives.
662 Additionally, working with tapes is much more complicated than working
663 with hard disks. For these reasons, the tutorial does not cover working
664 with tape drives. @xref{Media}, for complete information on using
665 @command{tar} archives with tape drives.
667 @FIXME{this is a cop out. need to add some simple tape drive info.}
670 @node stylistic conventions
671 @section Stylistic Conventions
673 In the examples, @samp{$} represents a typical shell prompt. It
674 precedes lines you should type; to make this more clear, those lines are
675 shown in @kbd{this font}, as opposed to lines which represent the
676 computer's response; those lines are shown in @code{this font}, or
677 sometimes @samp{like this}.
679 @c When we have lines which are too long to be
680 @c displayed in any other way, we will show them like this:
682 @node basic tar options
683 @section Basic @command{tar} Operations and Options
685 @command{tar} can take a wide variety of arguments which specify and define
686 the actions it will have on the particular set of files or the archive.
687 The main types of arguments to @command{tar} fall into one of two classes:
688 operations, and options.
690 Some arguments fall into a class called @dfn{operations}; exactly one of
691 these is both allowed and required for any instance of using @command{tar};
692 you may @emph{not} specify more than one. People sometimes speak of
693 @dfn{operating modes}. You are in a particular operating mode when you
694 have specified the operation which specifies it; there are eight
695 operations in total, and thus there are eight operating modes.
697 The other arguments fall into the class known as @dfn{options}. You are
698 not required to specify any options, and you are allowed to specify more
699 than one at a time (depending on the way you are using @command{tar} at
700 that time). Some options are used so frequently, and are so useful for
701 helping you type commands more carefully that they are effectively
702 ``required''. We will discuss them in this chapter.
704 You can write most of the @command{tar} operations and options in any
705 of three forms: long (mnemonic) form, short form, and old style. Some
706 of the operations and options have no short or ``old'' forms; however,
707 the operations and options which we will cover in this tutorial have
708 corresponding abbreviations. @FIXME{make sure this is still the case,
709 at the end}We will indicate those abbreviations appropriately to get
710 you used to seeing them. (Note that the ``old style'' option forms
711 exist in @GNUTAR{} for compatibility with Unix
712 @command{tar}. In this book we present a full discussion of this way
713 of writing options and operations (@pxref{Old Options}), and we discuss
714 the other two styles of writing options (@xref{Mnemonic Options}, and
715 @pxref{Short Options}).
717 In the examples and in the text of this tutorial, we usually use the
718 long forms of operations and options; but the ``short'' forms produce
719 the same result and can make typing long @command{tar} commands easier.
720 For example, instead of typing
723 @kbd{tar --create --verbose --file=afiles.tar apple angst aspic}
729 @kbd{tar -c -v -f afiles.tar apple angst aspic}
735 @kbd{tar -cvf afiles.tar apple angst aspic}
739 For more information on option syntax, see @ref{Advanced tar}. In
740 discussions in the text, when we name an option by its long form, we
741 also give the corresponding short option in parentheses.
743 The term, ``option'', can be confusing at times, since ``operations''
744 are often lumped in with the actual, @emph{optional} ``options'' in certain
745 general class statements. For example, we just talked about ``short and
746 long forms of options and operations''. However, experienced @command{tar}
747 users often refer to these by shorthand terms such as, ``short and long
748 options''. This term assumes that the ``operations'' are included, also.
749 Context will help you determine which definition of ``options'' to use.
751 Similarly, the term ``command'' can be confusing, as it is often used in
752 two different ways. People sometimes refer to @command{tar} ``commands''.
753 A @command{tar} @dfn{command} is the entire command line of user input
754 which tells @command{tar} what to do --- including the operation, options,
755 and any arguments (file names, pipes, other commands, etc). However,
756 you will also sometimes hear the term ``the @command{tar} command''. When
757 the word ``command'' is used specifically like this, a person is usually
758 referring to the @command{tar} @emph{operation}, not the whole line.
759 Again, use context to figure out which of the meanings the speaker
762 @node frequent operations
763 @section The Three Most Frequently Used Operations
765 Here are the three most frequently used operations (both short and long
766 forms), as well as a brief description of their meanings. The rest of
767 this chapter will cover how to use these operations in detail. We will
768 present the rest of the operations in the next chapter.
773 Create a new @command{tar} archive.
776 List the contents of an archive.
779 Extract one or more members from an archive.
782 @node Two Frequent Options
783 @section Two Frequently Used Options
785 To understand how to run @command{tar} in the three operating modes listed
786 previously, you also need to understand how to use two of the options to
787 @command{tar}: @option{--file} (which takes an archive file as an argument)
788 and @option{--verbose}. (You are usually not @emph{required} to specify
789 either of these options when you run @command{tar}, but they can be very
790 useful in making things more clear and helping you avoid errors.)
799 @unnumberedsubsec The @option{--file} Option
802 @opindex file, tutorial
803 @item --file=@var{archive-name}
804 @itemx -f @var{archive-name}
805 Specify the name of an archive file.
808 You can specify an argument for the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option whenever you
809 use @command{tar}; this option determines the name of the archive file
810 that @command{tar} will work on.
813 If you don't specify this argument, then @command{tar} will examine
814 the environment variable @env{TAPE}. If it is set, its value will be
815 used as the archive name. Otherwise, @command{tar} will use the
816 default archive, determined at the compile time. Usually it is
817 standard output or some physical tape drive attached to your machine
818 (you can verify what the default is by running @kbd{tar
819 --show-defaults}, @pxref{defaults}). If there is no tape drive
820 attached, or the default is not meaningful, then @command{tar} will
821 print an error message. The error message might look roughly like one
825 tar: can't open /dev/rmt8 : No such device or address
826 tar: can't open /dev/rsmt0 : I/O error
830 To avoid confusion, we recommend that you always specify an archive file
831 name by using @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) when writing your @command{tar} commands.
832 For more information on using the @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option, see
835 @node verbose tutorial
836 @unnumberedsubsec The @option{--verbose} Option
839 @opindex verbose, introduced
842 Show the files being worked on as @command{tar} is running.
845 @option{--verbose} (@option{-v}) shows details about the results of running
846 @command{tar}. This can be especially useful when the results might not be
847 obvious. For example, if you want to see the progress of @command{tar} as
848 it writes files into the archive, you can use the @option{--verbose}
849 option. In the beginning, you may find it useful to use
850 @option{--verbose} at all times; when you are more accustomed to
851 @command{tar}, you will likely want to use it at certain times but not at
852 others. We will use @option{--verbose} at times to help make something
853 clear, and we will give many examples both using and not using
854 @option{--verbose} to show the differences.
856 Sometimes, a single instance of @option{--verbose} on the command line
857 will show a full, @samp{ls} style listing of an archive or files,
858 giving sizes, owners, and similar information. @FIXME{Describe the
859 exact output format, e.g., how hard links are displayed.}
860 Other times, @option{--verbose} will only show files or members that the particular
861 operation is operating on at the time. In the latter case, you can
862 use @option{--verbose} twice in a command to get a listing such as that
863 in the former case. For example, instead of saying
866 @kbd{tar -cvf afiles.tar apple angst aspic}
873 @kbd{tar -cvvf afiles.tar apple angst aspic}
877 This works equally well using short or long forms of options. Using
878 long forms, you would simply write out the mnemonic form of the option
882 $ @kbd{tar --create --verbose --verbose @dots{}}
886 Note that you must double the hyphens properly each time.
888 Later in the tutorial, we will give examples using @w{@option{--verbose
892 @unnumberedsubsec Getting Help: Using the @option{--help} Option
898 The @option{--help} option to @command{tar} prints out a very brief list of
899 all operations and option available for the current version of
900 @command{tar} available on your system.
904 @section How to Create Archives
907 @cindex Creation of the archive
908 @cindex Archive, creation of
909 One of the basic operations of @command{tar} is @option{--create} (@option{-c}), which
910 you use to create a @command{tar} archive. We will explain
911 @option{--create} first because, in order to learn about the other
912 operations, you will find it useful to have an archive available to
915 To make this easier, in this section you will first create a directory
916 containing three files. Then, we will show you how to create an
917 @emph{archive} (inside the new directory). Both the directory, and
918 the archive are specifically for you to practice on. The rest of this
919 chapter and the next chapter will show many examples using this
920 directory and the files you will create: some of those files may be
921 other directories and other archives.
923 The three files you will archive in this example are called
924 @file{blues}, @file{folk}, and @file{jazz}. The archive is called
925 @file{collection.tar}.
927 This section will proceed slowly, detailing how to use @option{--create}
928 in @code{verbose} mode, and showing examples using both short and long
929 forms. In the rest of the tutorial, and in the examples in the next
930 chapter, we will proceed at a slightly quicker pace. This section
931 moves more slowly to allow beginning users to understand how
935 * prepare for examples::
936 * Creating the archive::
942 @node prepare for examples
943 @subsection Preparing a Practice Directory for Examples
945 To follow along with this and future examples, create a new directory
946 called @file{practice} containing files called @file{blues}, @file{folk}
947 and @file{jazz}. The files can contain any information you like:
948 ideally, they should contain information which relates to their names,
949 and be of different lengths. Our examples assume that @file{practice}
950 is a subdirectory of your home directory.
952 Now @command{cd} to the directory named @file{practice}; @file{practice}
953 is now your @dfn{working directory}. (@emph{Please note}: Although
954 the full path name of this directory is
955 @file{/@var{homedir}/practice}, in our examples we will refer to
956 this directory as @file{practice}; the @var{homedir} is presumed.
958 In general, you should check that the files to be archived exist where
959 you think they do (in the working directory) by running @command{ls}.
960 Because you just created the directory and the files and have changed to
961 that directory, you probably don't need to do that this time.
963 It is very important to make sure there isn't already a file in the
964 working directory with the archive name you intend to use (in this case,
965 @samp{collection.tar}), or that you don't care about its contents.
966 Whenever you use @samp{create}, @command{tar} will erase the current
967 contents of the file named by @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) if it exists. @command{tar}
968 will not tell you if you are about to overwrite an archive unless you
969 specify an option which does this (@pxref{backup}, for the
970 information on how to do so). To add files to an existing archive,
971 you need to use a different option, such as @option{--append} (@option{-r}); see
972 @ref{append} for information on how to do this.
974 @node Creating the archive
975 @subsection Creating the Archive
977 @opindex create, introduced
978 To place the files @file{blues}, @file{folk}, and @file{jazz} into an
979 archive named @file{collection.tar}, use the following command:
982 $ @kbd{tar --create --file=collection.tar blues folk jazz}
985 The order of the arguments is not very important, @emph{when using long
986 option forms}. You could also say:
989 $ @kbd{tar blues --create folk --file=collection.tar jazz}
993 However, you can see that this order is harder to understand; this is
994 why we will list the arguments in the order that makes the commands
995 easiest to understand (and we encourage you to do the same when you use
996 @command{tar}, to avoid errors).
998 Note that the part of the command which says,
999 @w{@option{--file=collection.tar}} is considered to be @emph{one} argument.
1000 If you substituted any other string of characters for
1001 @kbd{collection.tar}, then that string would become the name of the
1002 archive file you create.
1004 The order of the options becomes more important when you begin to use
1005 short forms. With short forms, if you type commands in the wrong order
1006 (even if you type them correctly in all other ways), you may end up with
1007 results you don't expect. For this reason, it is a good idea to get
1008 into the habit of typing options in the order that makes inherent sense.
1009 @xref{short create}, for more information on this.
1011 In this example, you type the command as shown above: @option{--create}
1012 is the operation which creates the new archive
1013 (@file{collection.tar}), and @option{--file} is the option which lets
1014 you give it the name you chose. The files, @file{blues}, @file{folk},
1015 and @file{jazz}, are now members of the archive, @file{collection.tar}
1016 (they are @dfn{file name arguments} to the @option{--create} operation).
1017 @FIXME{xref here to the discussion of file name args?}Now that they are
1018 in the archive, they are called @emph{archive members}, not files.
1019 (@pxref{Definitions,members}).
1021 When you create an archive, you @emph{must} specify which files you
1022 want placed in the archive. If you do not specify any archive
1023 members, @GNUTAR{} will complain.
1025 If you now list the contents of the working directory (@command{ls}), you will
1026 find the archive file listed as well as the files you saw previously:
1029 blues folk jazz collection.tar
1033 Creating the archive @samp{collection.tar} did not destroy the copies of
1034 the files in the directory.
1036 Keep in mind that if you don't indicate an operation, @command{tar} will not
1037 run and will prompt you for one. If you don't name any files, @command{tar}
1038 will complain. You must have write access to the working directory,
1039 or else you will not be able to create an archive in that directory.
1041 @emph{Caution}: Do not attempt to use @option{--create} (@option{-c}) to add files to
1042 an existing archive; it will delete the archive and write a new one.
1043 Use @option{--append} (@option{-r}) instead. @xref{append}.
1045 @node create verbose
1046 @subsection Running @option{--create} with @option{--verbose}
1048 @opindex create, using with @option{--verbose}
1049 @opindex verbose, using with @option{--create}
1050 If you include the @option{--verbose} (@option{-v}) option on the command line,
1051 @command{tar} will list the files it is acting on as it is working. In
1052 verbose mode, the @code{create} example above would appear as:
1055 $ @kbd{tar --create --verbose --file=collection.tar blues folk jazz}
1061 This example is just like the example we showed which did not use
1062 @option{--verbose}, except that @command{tar} generated the remaining lines
1064 (note the different font styles).
1070 In the rest of the examples in this chapter, we will frequently use
1071 @code{verbose} mode so we can show actions or @command{tar} responses that
1072 you would otherwise not see, and which are important for you to
1076 @subsection Short Forms with @samp{create}
1078 As we said before, the @option{--create} (@option{-c}) operation is one of the most
1079 basic uses of @command{tar}, and you will use it countless times.
1080 Eventually, you will probably want to use abbreviated (or ``short'')
1081 forms of options. A full discussion of the three different forms that
1082 options can take appears in @ref{Styles}; for now, here is what the
1083 previous example (including the @option{--verbose} (@option{-v}) option) looks like
1084 using short option forms:
1087 $ @kbd{tar -cvf collection.tar blues folk jazz}
1094 As you can see, the system responds the same no matter whether you use
1095 long or short option forms.
1097 @FIXME{i don't like how this is worded:} One difference between using
1098 short and long option forms is that, although the exact placement of
1099 arguments following options is no more specific when using short forms,
1100 it is easier to become confused and make a mistake when using short
1101 forms. For example, suppose you attempted the above example in the
1105 $ @kbd{tar -cfv collection.tar blues folk jazz}
1109 In this case, @command{tar} will make an archive file called @file{v},
1110 containing the files @file{blues}, @file{folk}, and @file{jazz}, because
1111 the @samp{v} is the closest ``file name'' to the @option{-f} option, and
1112 is thus taken to be the chosen archive file name. @command{tar} will try
1113 to add a file called @file{collection.tar} to the @file{v} archive file;
1114 if the file @file{collection.tar} did not already exist, @command{tar} will
1115 report an error indicating that this file does not exist. If the file
1116 @file{collection.tar} does already exist (e.g., from a previous command
1117 you may have run), then @command{tar} will add this file to the archive.
1118 Because the @option{-v} option did not get registered, @command{tar} will not
1119 run under @samp{verbose} mode, and will not report its progress.
1121 The end result is that you may be quite confused about what happened,
1122 and possibly overwrite a file. To illustrate this further, we will show
1123 you how an example we showed previously would look using short forms.
1128 $ @kbd{tar blues --create folk --file=collection.tar jazz}
1132 is confusing as it is. When shown using short forms, however, it
1133 becomes much more so:
1136 $ @kbd{tar blues -c folk -f collection.tar jazz}
1140 It would be very easy to put the wrong string of characters
1141 immediately following the @option{-f}, but doing that could sacrifice
1144 For this reason, we recommend that you pay very careful attention to
1145 the order of options and placement of file and archive names,
1146 especially when using short option forms. Not having the option name
1147 written out mnemonically can affect how well you remember which option
1148 does what, and therefore where different names have to be placed.
1151 @subsection Archiving Directories
1153 @cindex Archiving Directories
1154 @cindex Directories, Archiving
1155 You can archive a directory by specifying its directory name as a
1156 file name argument to @command{tar}. The files in the directory will be
1157 archived relative to the working directory, and the directory will be
1158 re-created along with its contents when the archive is extracted.
1160 To archive a directory, first move to its superior directory. If you
1161 have followed the previous instructions in this tutorial, you should
1170 This will put you into the directory which contains @file{practice},
1171 i.e., your home directory. Once in the superior directory, you can
1172 specify the subdirectory, @file{practice}, as a file name argument. To
1173 store @file{practice} in the new archive file @file{music.tar}, type:
1176 $ @kbd{tar --create --verbose --file=music.tar practice}
1180 @command{tar} should output:
1187 practice/collection.tar
1190 Note that the archive thus created is not in the subdirectory
1191 @file{practice}, but rather in the current working directory---the
1192 directory from which @command{tar} was invoked. Before trying to archive a
1193 directory from its superior directory, you should make sure you have
1194 write access to the superior directory itself, not only the directory
1195 you are trying archive with @command{tar}. For example, you will probably
1196 not be able to store your home directory in an archive by invoking
1197 @command{tar} from the root directory; @xref{absolute}. (Note
1198 also that @file{collection.tar}, the original archive file, has itself
1199 been archived. @command{tar} will accept any file as a file to be
1200 archived, regardless of its content. When @file{music.tar} is
1201 extracted, the archive file @file{collection.tar} will be re-written
1202 into the file system).
1204 If you give @command{tar} a command such as
1207 $ @kbd{tar --create --file=foo.tar .}
1211 @command{tar} will report @samp{tar: ./foo.tar is the archive; not
1212 dumped}. This happens because @command{tar} creates the archive
1213 @file{foo.tar} in the current directory before putting any files into
1214 it. Then, when @command{tar} attempts to add all the files in the
1215 directory @file{.} to the archive, it notices that the file
1216 @file{./foo.tar} is the same as the archive @file{foo.tar}, and skips
1217 it. (It makes no sense to put an archive into itself.) @GNUTAR{}
1218 will continue in this case, and create the archive
1219 normally, except for the exclusion of that one file. (@emph{Please
1220 note:} Other versions of @command{tar} are not so clever; they will
1221 enter an infinite loop when this happens, so you should not depend on
1222 this behavior unless you are certain you are running @GNUTAR{}.)
1223 @FIXME{bob doesn't like this sentence, since he does
1224 it all the time, and we've been doing it in the editing passes for
1225 this manual: In general, make sure that the archive is not inside a
1226 directory being dumped.}
1229 @section How to List Archives
1232 Frequently, you will find yourself wanting to determine exactly what a
1233 particular archive contains. You can use the @option{--list} (@option{-t}) operation
1234 to get the member names as they currently appear in the archive, as well
1235 as various attributes of the files at the time they were archived. For
1236 example, you can examine the archive @file{collection.tar} that you
1237 created in the last section with the command,
1240 $ @kbd{tar --list --file=collection.tar}
1244 The output of @command{tar} would then be:
1252 @FIXME{we hope this will change. if it doesn't, need to show the
1253 creation of bfiles somewhere above!!! : }
1256 The archive @file{bfiles.tar} would list as follows:
1265 Be sure to use a @option{--file=@var{archive-name}} (@option{-f @var{archive-name}}) option just as with @option{--create} (@option{-c})
1266 to specify the name of the archive.
1268 @opindex list, using with @option{--verbose}
1269 @opindex verbose, using with @option{--list}
1270 If you use the @option{--verbose} (@option{-v}) option with @option{--list}, then
1271 @command{tar} will print out a listing reminiscent of @w{@samp{ls -l}},
1272 showing owner, file size, and so forth.
1274 If you had used @option{--verbose} (@option{-v}) mode, the example above would look
1278 $ @kbd{tar --list --verbose --file=collection.tar folk}
1279 -rw-r--r-- myself user 62 1990-05-23 10:55 folk
1282 @cindex listing member and file names
1283 @anchor{listing member and file names}
1284 It is important to notice that the output of @kbd{tar --list
1285 --verbose} does not necessarily match that produced by @kbd{tar
1286 --create --verbose} while creating the archive. It is because
1287 @GNUTAR{}, unless told explicitly not to do so, removes some directory
1288 prefixes from file names before storing them in the archive
1289 (@xref{absolute}, for more information). In other
1290 words, in verbose mode @GNUTAR{} shows @dfn{file names} when creating
1291 an archive and @dfn{member names} when listing it. Consider this
1296 $ @kbd{tar cfv archive /etc/mail}
1297 tar: Removing leading `/' from member names
1299 /etc/mail/sendmail.cf
1301 $ @kbd{tar tf archive}
1303 etc/mail/sendmail.cf
1308 @opindex show-stored-names
1309 This default behavior can sometimes be inconvenient. You can force
1310 @GNUTAR{} show member names when creating archive by supplying
1311 @option{--show-stored-names} option.
1314 @item --show-stored-names
1315 Print member (as opposed to @emph{file}) names when creating the archive.
1318 @cindex File name arguments, using @option{--list} with
1319 @opindex list, using with file name arguments
1320 You can specify one or more individual member names as arguments when
1321 using @samp{list}. In this case, @command{tar} will only list the
1322 names of members you identify. For example, @w{@kbd{tar --list
1323 --file=afiles.tar apple}} would only print @file{apple}.
1325 @FIXME{we hope the relevant aspects of this will change:}Because
1326 @command{tar} preserves paths, file names must be specified as they appear
1327 in the archive (ie., relative to the directory from which the archive
1328 was created). Therefore, it is essential when specifying member names
1329 to @command{tar} that you give the exact member names. For example,
1330 @w{@kbd{tar --list --file=bfiles birds}} would produce an error message
1331 something like @samp{tar: birds: Not found in archive}, because there is
1332 no member named @file{birds}, only one named @file{./birds}. While the
1333 names @file{birds} and @file{./birds} name the same file, @emph{member}
1334 names are compared using a simplistic name comparison, in which an exact
1335 match is necessary. @xref{absolute}.
1337 However, @w{@kbd{tar --list --file=collection.tar folk}} would respond
1338 with @file{folk}, because @file{folk} is in the archive file
1339 @file{collection.tar}. If you are not sure of the exact file name, try
1340 listing all the files in the archive and searching for the one you
1341 expect to find; remember that if you use @option{--list} with no file
1342 names as arguments, @command{tar} will print the names of all the members
1343 stored in the specified archive.
1350 @unnumberedsubsec Listing the Contents of a Stored Directory
1352 To get information about the contents of an archived directory,
1353 use the directory name as a file name argument in conjunction with
1354 @option{--list} (@option{-t}). To find out file attributes, include the
1355 @option{--verbose} (@option{-v}) option.
1357 For example, to find out about files in the directory @file{practice}, in
1358 the archive file @file{music.tar}, type:
1361 $ @kbd{tar --list --verbose --file=music.tar practice}
1364 @command{tar} responds:
1367 drwxrwxrwx myself user 0 1990-05-31 21:49 practice/
1368 -rw-r--r-- myself user 42 1990-05-21 13:29 practice/blues
1369 -rw-r--r-- myself user 62 1990-05-23 10:55 practice/folk
1370 -rw-r--r-- myself user 40 1990-05-21 13:30 practice/jazz
1371 -rw-r--r-- myself user 10240 1990-05-31 21:49 practice/collection.tar
1374 When you use a directory name as a file name argument, @command{tar} acts on
1375 all the files (including sub-directories) in that directory.
1378 @section How to Extract Members from an Archive
1381 @cindex Retrieving files from an archive
1382 @cindex Resurrecting files from an archive
1385 Creating an archive is only half the job---there is no point in storing
1386 files in an archive if you can't retrieve them. The act of retrieving
1387 members from an archive so they can be used and manipulated as
1388 unarchived files again is called @dfn{extraction}. To extract files
1389 from an archive, use the @option{--extract} (@option{--get} or
1390 @option{-x}) operation. As with @option{--create}, specify the name
1391 of the archive with @option{--file} (@option{-f}) option. Extracting
1392 an archive does not modify the archive in any way; you can extract it
1393 multiple times if you want or need to.
1395 Using @option{--extract}, you can extract an entire archive, or specific
1396 files. The files can be directories containing other files, or not. As
1397 with @option{--create} (@option{-c}) and @option{--list} (@option{-t}), you may use the short or the
1398 long form of the operation without affecting the performance.
1401 * extracting archives::
1402 * extracting files::
1404 * extracting untrusted archives::
1405 * failing commands::
1408 @node extracting archives
1409 @subsection Extracting an Entire Archive
1411 To extract an entire archive, specify the archive file name only, with
1412 no individual file names as arguments. For example,
1415 $ @kbd{tar -xvf collection.tar}
1422 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
1423 -rw-r--r-- me user 21 1996-09-23 16:44 blues
1424 -rw-r--r-- me user 20 1996-09-23 16:44 folk
1427 @node extracting files
1428 @subsection Extracting Specific Files
1430 To extract specific archive members, give their exact member names as
1431 arguments, as printed by @option{--list} (@option{-t}). If you had mistakenly deleted
1432 one of the files you had placed in the archive @file{collection.tar}
1433 earlier (say, @file{blues}), you can extract it from the archive without
1434 changing the archive's structure. Its contents will be identical to the
1435 original file @file{blues} that you deleted.
1437 First, make sure you are in the @file{practice} directory, and list the
1438 files in the directory. Now, delete the file, @samp{blues}, and list
1439 the files in the directory again.
1441 You can now extract the member @file{blues} from the archive file
1442 @file{collection.tar} like this:
1445 $ @kbd{tar --extract --file=collection.tar blues}
1449 If you list the files in the directory again, you will see that the file
1450 @file{blues} has been restored, with its original permissions, data modification
1451 times, and owner.@FIXME{This is only accidentally true, but not in
1452 general. In most cases, one has to be root for restoring the owner, and
1453 use a special option for restoring permissions. Here, it just happens
1454 that the restoring user is also the owner of the archived members, and
1455 that the current @code{umask} is compatible with original permissions.}
1456 (These parameters will be identical to those which
1457 the file had when you originally placed it in the archive; any changes
1458 you may have made before deleting the file from the file system,
1459 however, will @emph{not} have been made to the archive member.) The
1460 archive file, @samp{collection.tar}, is the same as it was before you
1461 extracted @samp{blues}. You can confirm this by running @command{tar} with
1462 @option{--list} (@option{-t}).
1464 @FIXME{we hope this will change:}Remember that as with other operations,
1465 specifying the exact member name is important. @w{@kbd{tar --extract
1466 --file=bfiles.tar birds}} will fail, because there is no member named
1467 @file{birds}. To extract the member named @file{./birds}, you must
1468 specify @w{@kbd{tar --extract --file=bfiles.tar ./birds}}. To find the
1469 exact member names of the members of an archive, use @option{--list} (@option{-t})
1472 You can extract a file to standard output by combining the above options
1473 with the @option{--to-stdout} (@option{-O}) option (@pxref{Writing to Standard
1476 If you give the @option{--verbose} option, then @option{--extract}
1477 will print the names of the archive members as it extracts them.
1480 @subsection Extracting Files that are Directories
1482 Extracting directories which are members of an archive is similar to
1483 extracting other files. The main difference to be aware of is that if
1484 the extracted directory has the same name as any directory already in
1485 the working directory, then files in the extracted directory will be
1486 placed into the directory of the same name. Likewise, if there are
1487 files in the pre-existing directory with the same names as the members
1488 which you extract, the files from the extracted archive will replace
1489 the files already in the working directory (and possible
1490 subdirectories). This will happen regardless of whether or not the
1491 files in the working directory were more recent than those extracted
1492 (there exist, however, special options that alter this behavior
1495 However, if a file was stored with a directory name as part of its file
1496 name, and that directory does not exist under the working directory when
1497 the file is extracted, @command{tar} will create the directory.
1499 We can demonstrate how to use @option{--extract} to extract a directory
1500 file with an example. Change to the @file{practice} directory if you
1501 weren't there, and remove the files @file{folk} and @file{jazz}. Then,
1502 go back to the parent directory and extract the archive
1503 @file{music.tar}. You may either extract the entire archive, or you may
1504 extract only the files you just deleted. To extract the entire archive,
1505 don't give any file names as arguments after the archive name
1506 @file{music.tar}. To extract only the files you deleted, use the
1510 $ @kbd{tar -xvf music.tar practice/folk practice/jazz}
1516 If you were to specify two @option{--verbose} (@option{-v}) options, @command{tar}
1517 would have displayed more detail about the extracted files, as shown
1518 in the example below:
1521 $ @kbd{tar -xvvf music.tar practice/folk practice/jazz}
1522 -rw-r--r-- me user 28 1996-10-18 16:31 practice/jazz
1523 -rw-r--r-- me user 20 1996-09-23 16:44 practice/folk
1527 Because you created the directory with @file{practice} as part of the
1528 file names of each of the files by archiving the @file{practice}
1529 directory as @file{practice}, you must give @file{practice} as part
1530 of the file names when you extract those files from the archive.
1532 @FIXME{IMPORTANT! show the final structure, here. figure out what it
1535 @node extracting untrusted archives
1536 @subsection Extracting Archives from Untrusted Sources
1538 Extracting files from archives can overwrite files that already exist.
1539 If you receive an archive from an untrusted source, you should make a
1540 new directory and extract into that directory, so that you don't have
1541 to worry about the extraction overwriting one of your existing files.
1542 For example, if @file{untrusted.tar} came from somewhere else on the
1543 Internet, and you don't necessarily trust its contents, you can
1544 extract it as follows:
1547 $ @kbd{mkdir newdir}
1549 $ @kbd{tar -xvf ../untrusted.tar}
1552 It is also a good practice to examine contents of the archive
1553 before extracting it, using @option{--list} (@option{-t}) option, possibly combined
1554 with @option{--verbose} (@option{-v}).
1556 @node failing commands
1557 @subsection Commands That Will Fail
1559 Here are some sample commands you might try which will not work, and why
1562 If you try to use this command,
1565 $ @kbd{tar -xvf music.tar folk jazz}
1569 you will get the following response:
1572 tar: folk: Not found in archive
1573 tar: jazz: Not found in archive
1578 This is because these files were not originally @emph{in} the parent
1579 directory @file{..}, where the archive is located; they were in the
1580 @file{practice} directory, and their file names reflect this:
1583 $ @kbd{tar -tvf music.tar}
1589 @FIXME{make sure the above works when going through the examples in
1593 Likewise, if you try to use this command,
1596 $ @kbd{tar -tvf music.tar folk jazz}
1600 you would get a similar response. Members with those names are not in the
1601 archive. You must use the correct member names in order to extract the
1602 files from the archive.
1604 If you have forgotten the correct names of the files in the archive,
1605 use @w{@kbd{tar --list --verbose}} to list them correctly.
1607 @FIXME{more examples, here? hag thinks it's a good idea.}
1610 @section Going Further Ahead in this Manual
1612 @FIXME{need to write up a node here about the things that are going to
1613 be in the rest of the manual.}
1615 @node tar invocation
1616 @chapter Invoking @GNUTAR{}
1619 This chapter is about how one invokes the @GNUTAR{}
1620 command, from the command synopsis (@pxref{Synopsis}). There are
1621 numerous options, and many styles for writing them. One mandatory
1622 option specifies the operation @command{tar} should perform
1623 (@pxref{Operation Summary}), other options are meant to detail how
1624 this operation should be performed (@pxref{Option Summary}).
1625 Non-option arguments are not always interpreted the same way,
1626 depending on what the operation is.
1628 You will find in this chapter everything about option styles and rules for
1629 writing them (@pxref{Styles}). On the other hand, operations and options
1630 are fully described elsewhere, in other chapters. Here, you will find
1631 only synthetic descriptions for operations and options, together with
1632 pointers to other parts of the @command{tar} manual.
1634 Some options are so special they are fully described right in this
1635 chapter. They have the effect of inhibiting the normal operation of
1636 @command{tar} or else, they globally alter the amount of feedback the user
1637 receives about what is going on. These are the @option{--help} and
1638 @option{--version} (@pxref{help}), @option{--verbose} (@pxref{verbose})
1639 and @option{--interactive} options (@pxref{interactive}).
1643 * using tar options::
1653 @section General Synopsis of @command{tar}
1655 The @GNUTAR{} program is invoked as either one of:
1658 @kbd{tar @var{option}@dots{} [@var{name}]@dots{}}
1659 @kbd{tar @var{letter}@dots{} [@var{argument}]@dots{} [@var{option}]@dots{} [@var{name}]@dots{}}
1662 The second form is for when old options are being used.
1664 You can use @command{tar} to store files in an archive, to extract them from
1665 an archive, and to do other types of archive manipulation. The primary
1666 argument to @command{tar}, which is called the @dfn{operation}, specifies
1667 which action to take. The other arguments to @command{tar} are either
1668 @dfn{options}, which change the way @command{tar} performs an operation,
1669 or file names or archive members, which specify the files or members
1670 @command{tar} is to act on.
1672 You can actually type in arguments in any order, even if in this manual
1673 the options always precede the other arguments, to make examples easier
1674 to understand. Further, the option stating the main operation mode
1675 (the @command{tar} main command) is usually given first.
1677 Each @var{name} in the synopsis above is interpreted as an archive member
1678 name when the main command is one of @option{--compare}
1679 (@option{--diff}, @option{-d}), @option{--delete}, @option{--extract}
1680 (@option{--get}, @option{-x}), @option{--list} (@option{-t}) or
1681 @option{--update} (@option{-u}). When naming archive members, you
1682 must give the exact name of the member in the archive, as it is
1683 printed by @option{--list}. For @option{--append} (@option{-r}) and
1684 @option{--create} (@option{-c}), these @var{name} arguments specify
1685 the names of either files or directory hierarchies to place in the archive.
1686 These files or hierarchies should already exist in the file system,
1687 prior to the execution of the @command{tar} command.
1689 @command{tar} interprets relative file names as being relative to the
1690 working directory. @command{tar} will make all file names relative
1691 (by removing leading slashes when archiving or restoring files),
1692 unless you specify otherwise (using the @option{--absolute-names}
1693 option). @xref{absolute}, for more information about
1694 @option{--absolute-names}.
1696 If you give the name of a directory as either a file name or a member
1697 name, then @command{tar} acts recursively on all the files and directories
1698 beneath that directory. For example, the name @file{/} identifies all
1699 the files in the file system to @command{tar}.
1701 The distinction between file names and archive member names is especially
1702 important when shell globbing is used, and sometimes a source of confusion
1703 for newcomers. @xref{Wildcards}, for more information about globbing.
1704 The problem is that shells may only glob using existing files in the
1705 file system. Only @command{tar} itself may glob on archive members, so when
1706 needed, you must ensure that wildcard characters reach @command{tar} without
1707 being interpreted by the shell first. Using a backslash before @samp{*}
1708 or @samp{?}, or putting the whole argument between quotes, is usually
1709 sufficient for this.
1711 Even if @var{name}s are often specified on the command line, they
1712 can also be read from a text file in the file system, using the
1713 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}}) option.
1715 If you don't use any file name arguments, @option{--append} (@option{-r}),
1716 @option{--delete} and @option{--concatenate} (@option{--catenate},
1717 @option{-A}) will do nothing, while @option{--create} (@option{-c})
1718 will usually yield a diagnostic and inhibit @command{tar} execution.
1719 The other operations of @command{tar} (@option{--list},
1720 @option{--extract}, @option{--compare}, and @option{--update})
1721 will act on the entire contents of the archive.
1724 @cindex return status
1725 Besides successful exits, @GNUTAR{} may fail for
1726 many reasons. Some reasons correspond to bad usage, that is, when the
1727 @command{tar} command is improperly written. Errors may be
1728 encountered later, while encountering an error processing the archive
1729 or the files. Some errors are recoverable, in which case the failure
1730 is delayed until @command{tar} has completed all its work. Some
1731 errors are such that it would not meaningful, or at least risky, to
1732 continue processing: @command{tar} then aborts processing immediately.
1733 All abnormal exits, whether immediate or delayed, should always be
1734 clearly diagnosed on @code{stderr}, after a line stating the nature of
1737 @GNUTAR{} returns only a few exit statuses. I'm really
1738 aiming simplicity in that area, for now. If you are not using the
1739 @option{--compare} @option{--diff}, @option{-d}) option, zero means
1740 that everything went well, besides maybe innocuous warnings. Nonzero
1741 means that something went wrong. Right now, as of today, ``nonzero''
1742 is almost always 2, except for remote operations, where it may be
1745 @node using tar options
1746 @section Using @command{tar} Options
1748 @GNUTAR{} has a total of eight operating modes which
1749 allow you to perform a variety of tasks. You are required to choose
1750 one operating mode each time you employ the @command{tar} program by
1751 specifying one, and only one operation as an argument to the
1752 @command{tar} command (two lists of four operations each may be found
1753 at @ref{frequent operations} and @ref{Operations}). Depending on
1754 circumstances, you may also wish to customize how the chosen operating
1755 mode behaves. For example, you may wish to change the way the output
1756 looks, or the format of the files that you wish to archive may require
1757 you to do something special in order to make the archive look right.
1759 You can customize and control @command{tar}'s performance by running
1760 @command{tar} with one or more options (such as @option{--verbose}
1761 (@option{-v}), which we used in the tutorial). As we said in the
1762 tutorial, @dfn{options} are arguments to @command{tar} which are (as
1763 their name suggests) optional. Depending on the operating mode, you
1764 may specify one or more options. Different options will have different
1765 effects, but in general they all change details of the operation, such
1766 as archive format, archive name, or level of user interaction. Some
1767 options make sense with all operating modes, while others are
1768 meaningful only with particular modes. You will likely use some
1769 options frequently, while you will only use others infrequently, or
1770 not at all. (A full list of options is available in @pxref{All Options}.)
1772 @vrindex TAR_OPTIONS, environment variable
1773 @anchor{TAR_OPTIONS}
1774 The @env{TAR_OPTIONS} environment variable specifies default options to
1775 be placed in front of any explicit options. For example, if
1776 @code{TAR_OPTIONS} is @samp{-v --unlink-first}, @command{tar} behaves as
1777 if the two options @option{-v} and @option{--unlink-first} had been
1778 specified before any explicit options. Option specifications are
1779 separated by whitespace. A backslash escapes the next character, so it
1780 can be used to specify an option containing whitespace or a backslash.
1782 Note that @command{tar} options are case sensitive. For example, the
1783 options @option{-T} and @option{-t} are different; the first requires an
1784 argument for stating the name of a file providing a list of @var{name}s,
1785 while the second does not require an argument and is another way to
1786 write @option{--list} (@option{-t}).
1788 In addition to the eight operations, there are many options to
1789 @command{tar}, and three different styles for writing both: long (mnemonic)
1790 form, short form, and old style. These styles are discussed below.
1791 Both the options and the operations can be written in any of these three
1794 @FIXME{menu at end of this node. need to think of an actual outline
1795 for this chapter; probably do that after stuff from chapter 4 is
1799 @section The Three Option Styles
1801 There are three styles for writing operations and options to the command
1802 line invoking @command{tar}. The different styles were developed at
1803 different times during the history of @command{tar}. These styles will be
1804 presented below, from the most recent to the oldest.
1806 Some options must take an argument. (For example, @option{--file}
1807 (@option{-f})) takes the name of an archive file as an argument. If
1808 you do not supply an archive file name, @command{tar} will use a
1809 default, but this can be confusing; thus, we recommend that you always
1810 supply a specific archive file name.) Where you @emph{place} the
1811 arguments generally depends on which style of options you choose. We
1812 will detail specific information relevant to each option style in the
1813 sections on the different option styles, below. The differences are
1814 subtle, yet can often be very important; incorrect option placement
1815 can cause you to overwrite a number of important files. We urge you
1816 to note these differences, and only use the option style(s) which
1817 makes the most sense to you until you feel comfortable with the others.
1819 Some options @emph{may} take an argument (currently, there are
1820 two such options: @option{--backup} and @option{--occurrence}). Such
1821 options may have at most long and short forms, they do not have old style
1822 equivalent. The rules for specifying an argument for such options
1823 are stricter than those for specifying mandatory arguments. Please,
1824 pay special attention to them.
1827 * Mnemonic Options:: Mnemonic Option Style
1828 * Short Options:: Short Option Style
1829 * Old Options:: Old Option Style
1830 * Mixing:: Mixing Option Styles
1833 @node Mnemonic Options
1834 @subsection Mnemonic Option Style
1836 @FIXME{have to decide whether or not to replace other occurrences of
1837 "mnemonic" with "long", or *ugh* vice versa.}
1839 Each option has at least one long (or mnemonic) name starting with two
1840 dashes in a row, e.g., @option{--list}. The long names are more clear than
1841 their corresponding short or old names. It sometimes happens that a
1842 single mnemonic option has many different different names which are
1843 synonymous, such as @option{--compare} and @option{--diff}. In addition,
1844 long option names can be given unique abbreviations. For example,
1845 @option{--cre} can be used in place of @option{--create} because there is no
1846 other mnemonic option which begins with @samp{cre}. (One way to find
1847 this out is by trying it and seeing what happens; if a particular
1848 abbreviation could represent more than one option, @command{tar} will tell
1849 you that that abbreviation is ambiguous and you'll know that that
1850 abbreviation won't work. You may also choose to run @samp{tar --help}
1851 to see a list of options. Be aware that if you run @command{tar} with a
1852 unique abbreviation for the long name of an option you didn't want to
1853 use, you are stuck; @command{tar} will perform the command as ordered.)
1855 Mnemonic options are meant to be obvious and easy to remember, and their
1856 meanings are generally easier to discern than those of their
1857 corresponding short options (see below). For example:
1860 $ @kbd{tar --create --verbose --blocking-factor=20 --file=/dev/rmt0}
1864 gives a fairly good set of hints about what the command does, even
1865 for those not fully acquainted with @command{tar}.
1867 Mnemonic options which require arguments take those arguments
1868 immediately following the option name. There are two ways of
1869 specifying a mandatory argument. It can be separated from the
1870 option name either by an equal sign, or by any amount of
1871 white space characters. For example, the @option{--file} option (which
1872 tells the name of the @command{tar} archive) is given a file such as
1873 @file{archive.tar} as argument by using any of the following notations:
1874 @option{--file=archive.tar} or @option{--file archive.tar}.
1876 In contrast, optional arguments must always be introduced using
1877 an equal sign. For example, the @option{--backup} option takes
1878 an optional argument specifying backup type. It must be used
1879 as @option{--backup=@var{backup-type}}.
1882 @subsection Short Option Style
1884 Most options also have a short option name. Short options start with
1885 a single dash, and are followed by a single character, e.g., @option{-t}
1886 (which is equivalent to @option{--list}). The forms are absolutely
1887 identical in function; they are interchangeable.
1889 The short option names are faster to type than long option names.
1891 Short options which require arguments take their arguments immediately
1892 following the option, usually separated by white space. It is also
1893 possible to stick the argument right after the short option name, using
1894 no intervening space. For example, you might write @w{@option{-f
1895 archive.tar}} or @option{-farchive.tar} instead of using
1896 @option{--file=archive.tar}. Both @option{--file=@var{archive-name}} and
1897 @w{@option{-f @var{archive-name}}} denote the option which indicates a
1898 specific archive, here named @file{archive.tar}.
1900 Short options which take optional arguments take their arguments
1901 immediately following the option letter, @emph{without any intervening
1902 white space characters}.
1904 Short options' letters may be clumped together, but you are not
1905 required to do this (as compared to old options; see below). When
1906 short options are clumped as a set, use one (single) dash for them
1907 all, e.g., @w{@samp{@command{tar} -cvf}}. Only the last option in
1908 such a set is allowed to have an argument@footnote{Clustering many
1909 options, the last of which has an argument, is a rather opaque way to
1910 write options. Some wonder if @acronym{GNU} @code{getopt} should not
1911 even be made helpful enough for considering such usages as invalid.}.
1913 When the options are separated, the argument for each option which requires
1914 an argument directly follows that option, as is usual for Unix programs.
1918 $ @kbd{tar -c -v -b 20 -f /dev/rmt0}
1921 If you reorder short options' locations, be sure to move any arguments
1922 that belong to them. If you do not move the arguments properly, you may
1923 end up overwriting files.
1926 @subsection Old Option Style
1929 Like short options, old options are single letters. However, old options
1930 must be written together as a single clumped set, without spaces separating
1931 them or dashes preceding them@footnote{Beware that if you precede options
1932 with a dash, you are announcing the short option style instead of the
1933 old option style; short options are decoded differently.}. This set
1934 of letters must be the first to appear on the command line, after the
1935 @command{tar} program name and some white space; old options cannot appear
1936 anywhere else. The letter of an old option is exactly the same letter as
1937 the corresponding short option. For example, the old option @samp{t} is
1938 the same as the short option @option{-t}, and consequently, the same as the
1939 mnemonic option @option{--list}. So for example, the command @w{@samp{tar
1940 cv}} specifies the option @option{-v} in addition to the operation @option{-c}.
1942 @FIXME{bob suggests having an uglier example. :-) }
1944 When options that need arguments are given together with the command,
1945 all the associated arguments follow, in the same order as the options.
1946 Thus, the example given previously could also be written in the old
1950 $ @kbd{tar cvbf 20 /dev/rmt0}
1954 Here, @samp{20} is the argument of @option{-b} and @samp{/dev/rmt0} is
1955 the argument of @option{-f}.
1957 On the other hand, this old style syntax makes it difficult to match
1958 option letters with their corresponding arguments, and is often
1959 confusing. In the command @w{@samp{tar cvbf 20 /dev/rmt0}}, for example,
1960 @samp{20} is the argument for @option{-b}, @samp{/dev/rmt0} is the
1961 argument for @option{-f}, and @option{-v} does not have a corresponding
1962 argument. Even using short options like in @w{@samp{tar -c -v -b 20 -f
1963 /dev/rmt0}} is clearer, putting all arguments next to the option they
1966 If you want to reorder the letters in the old option argument, be
1967 sure to reorder any corresponding argument appropriately.
1969 This old way of writing @command{tar} options can surprise even experienced
1970 users. For example, the two commands:
1973 @kbd{tar cfz archive.tar.gz file}
1974 @kbd{tar -cfz archive.tar.gz file}
1978 are quite different. The first example uses @file{archive.tar.gz} as
1979 the value for option @samp{f} and recognizes the option @samp{z}. The
1980 second example, however, uses @file{z} as the value for option
1981 @samp{f} --- probably not what was intended.
1983 Old options are kept for compatibility with old versions of @command{tar}.
1985 This second example could be corrected in many ways, among which the
1986 following are equivalent:
1989 @kbd{tar -czf archive.tar.gz file}
1990 @kbd{tar -cf archive.tar.gz -z file}
1991 @kbd{tar cf archive.tar.gz -z file}
1994 @FIXME{still could explain this better; it's redundant:}
1996 @cindex option syntax, traditional
1997 As far as we know, all @command{tar} programs, @acronym{GNU} and
1998 non-@acronym{GNU}, support old options. @GNUTAR{}
1999 supports them not only for historical reasons, but also because many
2000 people are used to them. For compatibility with Unix @command{tar},
2001 the first argument is always treated as containing command and option
2002 letters even if it doesn't start with @samp{-}. Thus, @samp{tar c} is
2003 equivalent to @w{@samp{tar -c}:} both of them specify the
2004 @option{--create} (@option{-c}) command to create an archive.
2007 @subsection Mixing Option Styles
2009 All three styles may be intermixed in a single @command{tar} command,
2010 so long as the rules for each style are fully
2011 respected@footnote{Before @GNUTAR{} version 1.11.6,
2012 a bug prevented intermixing old style options with mnemonic options in
2013 some cases.}. Old style options and either of the modern styles of
2014 options may be mixed within a single @command{tar} command. However,
2015 old style options must be introduced as the first arguments only,
2016 following the rule for old options (old options must appear directly
2017 after the @command{tar} command and some white space). Modern options
2018 may be given only after all arguments to the old options have been
2019 collected. If this rule is not respected, a modern option might be
2020 falsely interpreted as the value of the argument to one of the old
2023 For example, all the following commands are wholly equivalent, and
2024 illustrate the many combinations and orderings of option styles.
2027 @kbd{tar --create --file=archive.tar}
2028 @kbd{tar --create -f archive.tar}
2029 @kbd{tar --create -farchive.tar}
2030 @kbd{tar --file=archive.tar --create}
2031 @kbd{tar --file=archive.tar -c}
2032 @kbd{tar -c --file=archive.tar}
2033 @kbd{tar -c -f archive.tar}
2034 @kbd{tar -c -farchive.tar}
2035 @kbd{tar -cf archive.tar}
2036 @kbd{tar -cfarchive.tar}
2037 @kbd{tar -f archive.tar --create}
2038 @kbd{tar -f archive.tar -c}
2039 @kbd{tar -farchive.tar --create}
2040 @kbd{tar -farchive.tar -c}
2041 @kbd{tar c --file=archive.tar}
2042 @kbd{tar c -f archive.tar}
2043 @kbd{tar c -farchive.tar}
2044 @kbd{tar cf archive.tar}
2045 @kbd{tar f archive.tar --create}
2046 @kbd{tar f archive.tar -c}
2047 @kbd{tar fc archive.tar}
2050 On the other hand, the following commands are @emph{not} equivalent to
2054 @kbd{tar -f -c archive.tar}
2055 @kbd{tar -fc archive.tar}
2056 @kbd{tar -fcarchive.tar}
2057 @kbd{tar -farchive.tarc}
2058 @kbd{tar cfarchive.tar}
2062 These last examples mean something completely different from what the
2063 user intended (judging based on the example in the previous set which
2064 uses long options, whose intent is therefore very clear). The first
2065 four specify that the @command{tar} archive would be a file named
2066 @option{-c}, @samp{c}, @samp{carchive.tar} or @samp{archive.tarc},
2067 respectively. The first two examples also specify a single non-option,
2068 @var{name} argument having the value @samp{archive.tar}. The last
2069 example contains only old style option letters (repeating option
2070 @samp{c} twice), not all of which are meaningful (eg., @samp{.},
2071 @samp{h}, or @samp{i}), with no argument value. @FIXME{not sure i liked
2072 the first sentence of this paragraph..}
2075 @section All @command{tar} Options
2077 The coming manual sections contain an alphabetical listing of all
2078 @command{tar} operations and options, with brief descriptions and cross
2079 references to more in-depth explanations in the body of the manual.
2080 They also contain an alphabetically arranged table of the short option
2081 forms with their corresponding long option. You can use this table as
2082 a reference for deciphering @command{tar} commands in scripts.
2085 * Operation Summary::
2087 * Short Option Summary::
2090 @node Operation Summary
2091 @subsection Operations
2095 @opindex append, summary
2099 Appends files to the end of the archive. @xref{append}.
2101 @opindex catenate, summary
2105 Same as @option{--concatenate}. @xref{concatenate}.
2107 @opindex compare, summary
2111 Compares archive members with their counterparts in the file
2112 system, and reports differences in file size, mode, owner,
2113 modification date and contents. @xref{compare}.
2115 @opindex concatenate, summary
2119 Appends other @command{tar} archives to the end of the archive.
2122 @opindex create, summary
2126 Creates a new @command{tar} archive. @xref{create}.
2128 @opindex delete, summary
2131 Deletes members from the archive. Don't try this on a archive on a
2132 tape! @xref{delete}.
2134 @opindex diff, summary
2138 Same @option{--compare}. @xref{compare}.
2140 @opindex extract, summary
2144 Extracts members from the archive into the file system. @xref{extract}.
2146 @opindex get, summary
2150 Same as @option{--extract}. @xref{extract}.
2152 @opindex list, summary
2156 Lists the members in an archive. @xref{list}.
2158 @opindex update, summary
2162 @FIXME{It was: A combination of the @option{--compare} and
2163 @option{--append} operations. This is not true and rather misleading,
2164 as @option{--compare} (@option{--diff}, @option{-d}) does a lot more than @option{--update} (@option{-u}) for
2165 ensuring files are identical.} Adds files to the end of the archive,
2166 but only if they are newer than their counterparts already in the
2167 archive, or if they do not already exist in the archive.
2172 @node Option Summary
2173 @subsection @command{tar} Options
2177 @opindex absolute-names, summary
2178 @item --absolute-names
2181 Normally when creating an archive, @command{tar} strips an initial
2182 @samp{/} from member names. This option disables that behavior.
2185 @opindex after-date, summary
2188 (See @option{--newer}, @pxref{after})
2190 @opindex anchored, summary
2192 An exclude pattern must match an initial subsequence of the name's components.
2193 @xref{controlling pattern-matching with exclude}.
2195 @opindex atime-preserve, summary
2196 @item --atime-preserve
2197 @itemx --atime-preserve=replace
2198 @itemx --atime-preserve=system
2200 Attempt to preserve the access time of files when reading them. This
2201 option currently is effective only on files that you own, unless you
2202 have superuser privileges.
2204 @option{--atime-preserve=replace} remembers the access time of a file
2205 before reading it, and then restores the access time afterwards. This
2206 may cause problems if other programs are reading the file at the same
2207 time, as the times of their accesses will be lost. On most platforms
2208 restoring the access time also requires @command{tar} to restore the
2209 data modification time too, so this option may also cause problems if
2210 other programs are writing the file at the same time. (Tar attempts
2211 to detect this situation, but cannot do so reliably due to race
2212 conditions.) Worse, on most platforms restoring the access time also
2213 updates the status change time, which means that this option is
2214 incompatible with incremental backups.
2216 @option{--atime-preserve=system} avoids changing time stamps on files,
2217 without interfering with time stamp updates
2218 caused by other programs, so it works better with incremental backups.
2219 However, it requires a special @code{O_NOATIME} option from the
2220 underlying operating and file system implementation, and it also requires
2221 that searching directories does not update their access times. As of
2222 this writing (November 2005) this works only with Linux, and only with
2223 Linux kernels 2.6.8 and later. Worse, there is currently no reliable
2224 way to know whether this feature actually works. Sometimes
2225 @command{tar} knows that it does not work, and if you use
2226 @option{--atime-preserve=system} then @command{tar} complains and
2227 exits right away. But other times @command{tar} might think that the
2228 option works when it actually does not.
2230 Currently @option{--atime-preserve} with no operand defaults to
2231 @option{--atime-preserve=replace}, but this may change in the future
2232 as support for @option{--atime-preserve=system} improves.
2234 If your operating system does not support
2235 @option{--atime-preserve=system}, you might be able to preserve access
2236 times reliably by by using the @command{mount} command. For example,
2237 you can mount the file system read-only, or access the file system via
2238 a read-only loopback mount, or use the @samp{noatime} mount option
2239 available on some systems. However, mounting typically requires
2240 superuser privileges and can be a pain to manage.
2242 @opindex backup, summary
2243 @item --backup=@var{backup-type}
2245 Rather than deleting files from the file system, @command{tar} will
2246 back them up using simple or numbered backups, depending upon
2247 @var{backup-type}. @xref{backup}.
2249 @opindex block-number, summary
2250 @item --block-number
2253 With this option present, @command{tar} prints error messages for read errors
2254 with the block number in the archive file. @xref{block-number}.
2256 @opindex blocking-factor, summary
2257 @item --blocking-factor=@var{blocking}
2258 @itemx -b @var{blocking}
2260 Sets the blocking factor @command{tar} uses to @var{blocking} x 512 bytes per
2261 record. @xref{Blocking Factor}.
2263 @opindex bzip2, summary
2267 This option tells @command{tar} to read or write archives through
2268 @code{bzip2}. @xref{gzip}.
2270 @opindex checkpoint, summary
2273 This option directs @command{tar} to print periodic checkpoint messages as it
2274 reads through the archive. Its intended for when you want a visual
2275 indication that @command{tar} is still running, but don't want to see
2276 @option{--verbose} output. @FIXME-xref{}
2278 @opindex check-links, summary
2281 If this option was given, @command{tar} will check the number of links
2282 dumped for each processed file. If this number does not match the
2283 total number of hard links for the file, a warning message will be
2286 Future versions will take @option{-l} as a short version of
2287 @option{--check-links}. However, current release still retains the old
2288 semantics for @option{-l}.
2290 @xref{Changes}, for more information.
2292 @opindex compress, summary
2293 @opindex uncompress, summary
2298 @command{tar} will use the @command{compress} program when reading or
2299 writing the archive. This allows you to directly act on archives
2300 while saving space. @xref{gzip}.
2302 @opindex confirmation, summary
2303 @item --confirmation
2305 (See @option{--interactive}.) @xref{interactive}.
2307 @opindex delay-directory-restore, summary
2308 @item --delay-directory-restore
2310 Delay setting modification times and permissions of extracted
2311 directories until the end of extraction. @xref{Directory Modification Times and Permissions}.
2313 @opindex dereference, summary
2317 When creating a @command{tar} archive, @command{tar} will archive the
2318 file that a symbolic link points to, rather than archiving the
2319 symlink. @xref{dereference}.
2321 @opindex directory, summary
2322 @item --directory=@var{dir}
2325 When this option is specified, @command{tar} will change its current directory
2326 to @var{dir} before performing any operations. When this option is used
2327 during archive creation, it is order sensitive. @xref{directory}.
2329 @opindex exclude, summary
2330 @item --exclude=@var{pattern}
2332 When performing operations, @command{tar} will skip files that match
2333 @var{pattern}. @xref{exclude}.
2335 @opindex exclude-from, summary
2336 @item --exclude-from=@var{file}
2337 @itemx -X @var{file}
2339 Similar to @option{--exclude}, except @command{tar} will use the list of
2340 patterns in the file @var{file}. @xref{exclude}.
2342 @opindex exclude-caches, summary
2343 @item --exclude-caches
2345 Automatically excludes all directories
2346 containing a cache directory tag. @xref{exclude}.
2348 @opindex file, summary
2349 @item --file=@var{archive}
2350 @itemx -f @var{archive}
2352 @command{tar} will use the file @var{archive} as the @command{tar} archive it
2353 performs operations on, rather than @command{tar}'s compilation dependent
2354 default. @xref{file tutorial}.
2356 @opindex files-from, summary
2357 @item --files-from=@var{file}
2358 @itemx -T @var{file}
2360 @command{tar} will use the contents of @var{file} as a list of archive members
2361 or files to operate on, in addition to those specified on the
2362 command-line. @xref{files}.
2364 @opindex force-local, summary
2367 Forces @command{tar} to interpret the filename given to @option{--file}
2368 as a local file, even if it looks like a remote tape drive name.
2369 @xref{local and remote archives}.
2371 @opindex format, summary
2372 @item --format=@var{format}
2374 Selects output archive format. @var{Format} may be one of the
2379 Creates an archive that is compatible with Unix V7 @command{tar}.
2382 Creates an archive that is compatible with GNU @command{tar} version
2386 Creates archive in GNU tar 1.13 format. Basically it is the same as
2387 @samp{oldgnu} with the only difference in the way it handles long
2391 Creates a @acronym{POSIX.1-1988} compatible archive.
2394 Creates a @acronym{POSIX.1-2001 archive}.
2398 @xref{Formats}, for a detailed discussion of these formats.
2400 @opindex group, summary
2401 @item --group=@var{group}
2403 Files added to the @command{tar} archive will have a group id of @var{group},
2404 rather than the group from the source file. @var{group} is first decoded
2405 as a group symbolic name, but if this interpretation fails, it has to be
2406 a decimal numeric group ID. @FIXME-xref{}
2408 Also see the comments for the @option{--owner=@var{user}} option.
2410 @opindex gzip, summary
2411 @opindex gunzip, summary
2412 @opindex ungzip, summary
2418 This option tells @command{tar} to read or write archives through
2419 @command{gzip}, allowing @command{tar} to directly operate on several
2420 kinds of compressed archives transparently. @xref{gzip}.
2422 @opindex help, summary
2425 @command{tar} will print out a short message summarizing the operations and
2426 options to @command{tar} and exit. @xref{help}.
2428 @opindex ignore-case, summary
2430 Ignore case when excluding files. @xref{controlling pattern-matching
2433 @opindex ignore-command-error, summary
2434 @item --ignore-command-error
2435 Ignore exit codes of subprocesses. @xref{Writing to an External Program}.
2437 @opindex ignore-failed-read, summary
2438 @item --ignore-failed-read
2440 Do not exit unsuccessfully merely because an unreadable file was encountered.
2443 @opindex ignore-zeros, summary
2444 @item --ignore-zeros
2447 With this option, @command{tar} will ignore zeroed blocks in the
2448 archive, which normally signals EOF. @xref{Reading}.
2450 @opindex incremental, summary
2454 Used to inform @command{tar} that it is working with an old
2455 @acronym{GNU}-format incremental backup archive. It is intended
2456 primarily for backwards compatibility only. @FIXME{incremental and
2457 listed-incremental}.
2459 @opindex index-file, summary
2460 @item --index-file=@var{file}
2462 Send verbose output to @var{file} instead of to standard output.
2464 @opindex info-script, summary
2465 @opindex new-volume-script, summary
2466 @item --info-script=@var{script-file}
2467 @itemx --new-volume-script=@var{script-file}
2468 @itemx -F @var{script-file}
2470 When @command{tar} is performing multi-tape backups, @var{script-file} is run
2471 at the end of each tape. If @var{script-file} exits with nonzero status,
2472 @command{tar} fails immediately. @xref{info-script}, for a detailed
2473 discussion of @var{script-file}.
2475 @opindex interactive, summary
2477 @itemx --confirmation
2480 Specifies that @command{tar} should ask the user for confirmation before
2481 performing potentially destructive options, such as overwriting files.
2484 @opindex keep-newer-files, summary
2485 @item --keep-newer-files
2487 Do not replace existing files that are newer than their archive copies
2488 when extracting files from an archive.
2490 @opindex keep-old-files, summary
2491 @item --keep-old-files
2494 Do not overwrite existing files when extracting files from an archive.
2495 @xref{Keep Old Files}.
2497 @opindex label, summary
2498 @item --label=@var{name}
2499 @itemx -V @var{name}
2501 When creating an archive, instructs @command{tar} to write @var{name}
2502 as a name record in the archive. When extracting or listing archives,
2503 @command{tar} will only operate on archives that have a label matching
2504 the pattern specified in @var{name}. @xref{Tape Files}.
2506 @opindex listed-incremental, summary
2507 @item --listed-incremental=@var{snapshot-file}
2508 @itemx -g @var{snapshot-file}
2510 During a @option{--create} operation, specifies that the archive that
2511 @command{tar} creates is a new @acronym{GNU}-format incremental
2512 backup, using @var{snapshot-file} to determine which files to backup.
2513 With other operations, informs @command{tar} that the archive is in
2514 incremental format. @FIXME{incremental and listed-incremental}.
2516 @opindex mode, summary
2517 @item --mode=@var{permissions}
2519 When adding files to an archive, @command{tar} will use
2520 @var{permissions} for the archive members, rather than the permissions
2521 from the files. The program @command{chmod} and this @command{tar}
2522 option share the same syntax for what @var{permissions} might be.
2523 @xref{File permissions, Permissions, File permissions, fileutils,
2524 @acronym{GNU} file utilities}. This reference also has useful
2525 information for those not being overly familiar with the Unix
2528 Of course, @var{permissions} might be plainly specified as an octal number.
2529 However, by using generic symbolic modifications to mode bits, this allows
2530 more flexibility. For example, the value @samp{a+rw} adds read and write
2531 permissions for everybody, while retaining executable bits on directories
2532 or on any other file already marked as executable.
2534 @opindex multi-volume, summary
2535 @item --multi-volume
2538 Informs @command{tar} that it should create or otherwise operate on a
2539 multi-volume @command{tar} archive. @xref{Using Multiple Tapes}.
2541 @opindex new-volume-script, summary
2542 @item --new-volume-script
2546 @opindex seek, summary
2550 Assume that the archive media supports seeks to arbitrary
2551 locations. Usually @command{tar} determines automatically whether
2552 the archive can be seeked or not. This option is intended for use
2553 in cases when such recognition fails.
2555 @opindex newer, summary
2556 @item --newer=@var{date}
2557 @itemx --after-date=@var{date}
2560 When creating an archive, @command{tar} will only add files that have changed
2561 since @var{date}. If @var{date} begins with @samp{/} or @samp{.}, it
2562 is taken to be the name of a file whose data modification time specifies
2563 the date. @xref{after}.
2565 @opindex newer-mtime, summary
2566 @item --newer-mtime=@var{date}
2568 Like @option{--newer}, but add only files whose
2569 contents have changed (as opposed to just @option{--newer}, which will
2570 also back up files for which any status information has changed).
2572 @opindex no-anchored, summary
2574 An exclude pattern can match any subsequence of the name's components.
2575 @xref{controlling pattern-matching with exclude}.
2577 @opindex no-delay-directory-restore, summary
2578 @item --no-delay-directory-restore
2580 Setting modification times and permissions of extracted
2581 directories when all files from this directory has been
2582 extracted. This is the default. @xref{Directory Modification Times and Permissions}.
2584 @opindex no-ignore-case, summary
2585 @item --no-ignore-case
2586 Use case-sensitive matching when excluding files.
2587 @xref{controlling pattern-matching with exclude}.
2589 @opindex no-ignore-command-error, summary
2590 @item --no-ignore-command-error
2591 Print warnings about subprocesses terminated with a non-zero exit
2592 code. @xref{Writing to an External Program}.
2594 @opindex no-quote-chars, summary
2595 @item --no-quote-chars=@var{string}
2596 Do not quote characters from @var{string}, even if the selected
2597 quoting style implies they should be quoted (@FIXME-pxref{Quoting Styles}).
2599 @opindex no-recursion, summary
2600 @item --no-recursion
2602 With this option, @command{tar} will not recurse into directories.
2605 @opindex no-same-owner, summary
2606 @item --no-same-owner
2609 When extracting an archive, do not attempt to preserve the owner
2610 specified in the @command{tar} archive. This the default behavior
2613 @opindex no-same-permissions, summary
2614 @item --no-same-permissions
2616 When extracting an archive, subtract the user's umask from files from
2617 the permissions specified in the archive. This is the default behavior
2620 @opindex no-wildcards, summary
2621 @item --no-wildcards
2622 Do not use wildcards when excluding files.
2623 @xref{controlling pattern-matching with exclude}.
2625 @opindex no-wildcards-match-slash, summary
2626 @item --no-wildcards-match-slash
2627 Wildcards do not match @samp{/} when excluding files.
2628 @xref{controlling pattern-matching with exclude}.
2630 @opindex null, summary
2633 When @command{tar} is using the @option{--files-from} option, this option
2634 instructs @command{tar} to expect filenames terminated with @option{NUL}, so
2635 @command{tar} can correctly work with file names that contain newlines.
2638 @opindex numeric-owner, summary
2639 @item --numeric-owner
2641 This option will notify @command{tar} that it should use numeric user
2642 and group IDs when creating a @command{tar} file, rather than names.
2646 When extracting files, this option is a synonym for
2647 @option{--no-same-owner}, i.e. it prevents @command{tar} from
2648 restoring ownership of files being extracted.
2650 When creating an archive, @option{-o} is a synonym for
2651 @option{--old-archive}. This behavior is for compatibility
2652 with previous versions of @GNUTAR{}, and will be
2653 removed in the future releases.
2655 @xref{Changes}, for more information.
2657 @opindex occurrence, summary
2658 @item --occurrence[=@var{number}]
2660 This option can be used in conjunction with one of the subcommands
2661 @option{--delete}, @option{--diff}, @option{--extract} or
2662 @option{--list} when a list of files is given either on the command
2663 line or via @option{-T} option.
2665 This option instructs @command{tar} to process only the @var{number}th
2666 occurrence of each named file. @var{Number} defaults to 1, so
2669 tar -x -f archive.tar --occurrence filename
2673 will extract the first occurrence of @file{filename} from @file{archive.tar}
2674 and will terminate without scanning to the end of the archive.
2676 @opindex old-archive, summary
2678 Synonym for @option{--format=v7}.
2680 @opindex one-file-system, summary
2681 @item --one-file-system
2683 Used when creating an archive. Prevents @command{tar} from recursing into
2684 directories that are on different file systems from the current
2687 Earlier versions of @GNUTAR{} understood @option{-l} as a
2688 synonym for @option{--one-file-system}. Although such usage is still
2689 allowed in the present version, it is @emph{strongly discouraged}.
2690 The future versions of @GNUTAR{} will use @option{-l} as
2691 a synonym for @option{--check-links}.
2693 @xref{Changes}, for more information.
2695 @opindex overwrite, summary
2698 Overwrite existing files and directory metadata when extracting files
2699 from an archive. @xref{Overwrite Old Files}.
2701 @opindex overwrite-dir, summary
2702 @item --overwrite-dir
2704 Overwrite the metadata of existing directories when extracting files
2705 from an archive. @xref{Overwrite Old Files}.
2707 @opindex owner, summary
2708 @item --owner=@var{user}
2710 Specifies that @command{tar} should use @var{user} as the owner of members
2711 when creating archives, instead of the user associated with the source
2712 file. @var{user} is first decoded as a user symbolic name, but if
2713 this interpretation fails, it has to be a decimal numeric user ID.
2716 There is no value indicating a missing number, and @samp{0} usually means
2717 @code{root}. Some people like to force @samp{0} as the value to offer in
2718 their distributions for the owner of files, because the @code{root} user is
2719 anonymous anyway, so that might as well be the owner of anonymous archives.
2721 This option does not affect extraction from archives.
2723 @opindex quote-chars, summary
2724 @item --quote-chars=@var{string}
2725 Always quote characters from @var{string}, even if the selected
2726 quoting style would not quote them (@FIXME-pxref{Quoting Styles}).
2728 @opindex quoting-style, summary
2729 @item --quoting-style=@var{style}
2730 Set quoting style to use when printing member and file names
2731 (@FIXME-pxref{Quoting Styles}). Valid @var{style} values are:
2732 @code{literal}, @code{shell}, @code{shell-always}, @code{c},
2733 @code{escape}, @code{locale}, and @code{clocale}. Default quoting
2734 style is @code{escape}, unless overridden while configuring the
2737 @opindex pax-option, summary
2738 @item --pax-option=@var{keyword-list}
2739 @FIXME{Such a detailed description does not belong there, move it elsewhere.}
2740 This option is meaningful only with @acronym{POSIX.1-2001} archives
2741 (@pxref{posix}). It modifies the way @command{tar} handles the
2742 extended header keywords. @var{Keyword-list} is a comma-separated
2743 list of keyword options, each keyword option taking one of
2744 the following forms:
2747 @item delete=@var{pattern}
2748 When used with one of archive-creation commands,
2749 this option instructs @command{tar} to omit from extended header records
2750 that it produces any keywords matching the string @var{pattern}.
2752 When used in extract or list mode, this option instructs tar
2753 to ignore any keywords matching the given @var{pattern} in the extended
2754 header records. In both cases, matching is performed using the pattern
2755 matching notation described in @acronym{POSIX 1003.2}, 3.13
2756 (See @cite{glob(7)}). For example:
2759 --pax-option delete=security.*
2762 would suppress security-related information.
2764 @item exthdr.name=@var{string}
2766 This keyword allows user control over the name that is written into the
2767 ustar header blocks for the extended headers. The name is obtained
2768 from @var{string} after substituting the following meta-characters:
2770 @multitable @columnfractions .30 .70
2771 @headitem Meta-character @tab Replaced By
2772 @item %d @tab The directory name of the file, equivalent to the
2773 result of the @command{dirname} utility on the translated pathname.
2774 @item %f @tab The filename of the file, equivalent to the result
2775 of the @command{basename} utility on the translated pathname.
2776 @item %p @tab The process ID of the @command{tar} process.
2777 @item %% @tab A @samp{%} character.
2780 Any other @samp{%} characters in @var{string} produce undefined
2783 If no option @samp{exthdr.name=string} is specified, @command{tar}
2784 will use the following default value:
2790 @item globexthdr.name=@var{string}
2791 This keyword allows user control over the name that is written into
2792 the ustar header blocks for global extended header records. The name
2793 shall will be obtained from the contents of @var{string}, after the
2794 following character substitutions have been made:
2796 @multitable @columnfractions .30 .70
2797 @headitem Meta-character @tab Replaced By
2798 @item %n @tab An integer that represents the
2799 sequence number of the global extended header record in the archive,
2801 @item %p @tab The process ID of the @command{tar} process.
2802 @item %% @tab A @samp{%} character.
2805 Any other @samp{%} characters in string produce undefined results.
2807 If no option @samp{globexthdr.name=string} is specified, @command{tar}
2808 will use the following default value:
2811 $TMPDIR/GlobalHead.%p.%n
2815 where @samp{$TMPDIR} represents the value of the @var{TMPDIR}
2816 environment variable. If @var{TMPDIR} is not set, @command{tar}
2819 @item @var{keyword}=@var{value}
2820 When used with one of archive-creation commands, these keyword/value pairs
2821 will be included at the beginning of the archive in a global extended
2822 header record. When used with one of archive-reading commands,
2823 @command{tar} will behave as if it has encountered these keyword/value
2824 pairs at the beginning of the archive in a global extended header
2827 @item @var{keyword}:=@var{value}
2828 When used with one of archive-creation commands, these keyword/value pairs
2829 will be included as records at the beginning of an extended header for
2830 each file. This is effectively equivalent to @var{keyword}=@var{value}
2831 form except that it creates no global extended header records.
2833 When used with one of archive-reading commands, @command{tar} will
2834 behave as if these keyword/value pairs were included as records at the
2835 end of each extended header; thus, they will override any global or
2836 file-specific extended header record keywords of the same names.
2837 For example, in the command:
2840 tar --format=posix --create \
2841 --file archive --pax-option gname:=user .
2844 the group name will be forced to a new value for all files
2845 stored in the archive.
2848 @opindex portability, summary
2850 @itemx --old-archive
2851 Synonym for @option{--format=v7}.
2853 @opindex posix, summary
2855 Same as @option{--format=posix}.
2857 @opindex preserve, summary
2860 Synonymous with specifying both @option{--preserve-permissions} and
2861 @option{--same-order}. @xref{Setting Access Permissions}.
2863 @opindex preserve-order, summary
2864 @item --preserve-order
2866 (See @option{--same-order}; @pxref{Reading}.)
2868 @opindex preserve-permissions, summary
2869 @opindex same-permissions, summary
2870 @item --preserve-permissions
2871 @itemx --same-permissions
2874 When @command{tar} is extracting an archive, it normally subtracts the
2875 users' umask from the permissions specified in the archive and uses
2876 that number as the permissions to create the destination file.
2877 Specifying this option instructs @command{tar} that it should use the
2878 permissions directly from the archive. @xref{Setting Access Permissions}.
2880 @opindex read-full-records, summary
2881 @item --read-full-records
2884 Specifies that @command{tar} should reblock its input, for reading
2885 from pipes on systems with buggy implementations. @xref{Reading}.
2887 @opindex record-size, summary
2888 @item --record-size=@var{size}
2890 Instructs @command{tar} to use @var{size} bytes per record when accessing the
2891 archive. @xref{Blocking Factor}.
2893 @opindex recursion, summary
2896 With this option, @command{tar} recurses into directories.
2899 @opindex recursive-unlink, summary
2900 @item --recursive-unlink
2903 directory hierarchies before extracting directories of the same name
2904 from the archive. @xref{Recursive Unlink}.
2906 @opindex remove-files, summary
2907 @item --remove-files
2909 Directs @command{tar} to remove the source file from the file system after
2910 appending it to an archive. @xref{remove files}.
2912 @opindex restrict, summary
2915 Disable use of some potentially harmful @command{tar} options.
2916 Currently this option disables shell invocaton from multi-volume menu
2917 (@pxref{Using Multiple Tapes}).
2919 @opindex rmt-command, summary
2920 @item --rmt-command=@var{cmd}
2922 Notifies @command{tar} that it should use @var{cmd} instead of
2923 the default @file{/usr/libexec/rmt} (@pxref{Remote Tape Server}).
2925 @opindex rsh-command, summary
2926 @item --rsh-command=@var{cmd}
2928 Notifies @command{tar} that is should use @var{cmd} to communicate with remote
2929 devices. @xref{Device}.
2931 @opindex same-order, summary
2933 @itemx --preserve-order
2936 This option is an optimization for @command{tar} when running on machines with
2937 small amounts of memory. It informs @command{tar} that the list of file
2938 arguments has already been sorted to match the order of files in the
2939 archive. @xref{Reading}.
2941 @opindex same-owner, summary
2944 When extracting an archive, @command{tar} will attempt to preserve the owner
2945 specified in the @command{tar} archive with this option present.
2946 This is the default behavior for the superuser; this option has an
2947 effect only for ordinary users. @xref{Attributes}.
2949 @opindex same-permissions, summary
2950 @item --same-permissions
2952 (See @option{--preserve-permissions}; @pxref{Setting Access Permissions}.)
2954 @opindex show-defaults, summary
2955 @item --show-defaults
2957 Displays the default options used by @command{tar} and exits
2958 successfully. This option is intended for use in shell scripts.
2959 Here is an example of what you can see using this option:
2962 $ tar --show-defaults
2963 --format=gnu -f- -b20 --quoting-style=escape \
2964 --rmt-command=/usr/libexec/rmt --rsh-command=/usr/bin/rsh
2967 @opindex show-omitted-dirs, summary
2968 @item --show-omitted-dirs
2970 Instructs @command{tar} to mention directories its skipping over when
2971 operating on a @command{tar} archive. @xref{show-omitted-dirs}.
2973 @opindex show-stored-names, summary
2974 @item --show-stored-names
2976 This option has effect only when used in conjunction with one of
2977 archive creation operations. It instructs tar to list the member names
2978 stored in the archive, as opposed to the actual file
2979 names. @xref{listing member and file names}.
2981 @opindex sparse, summary
2985 Invokes a @acronym{GNU} extension when adding files to an archive that handles
2986 sparse files efficiently. @xref{sparse}.
2988 @opindex starting-file, summary
2989 @item --starting-file=@var{name}
2990 @itemx -K @var{name}
2992 This option affects extraction only; @command{tar} will skip extracting
2993 files in the archive until it finds one that matches @var{name}.
2996 @opindex strip-components, summary
2997 @item --strip-components=@var{number}
2998 Strip given @var{number} of leading components from file names before
2999 extraction.@footnote{This option was called @option{--strip-path} in
3000 version 1.14.} For example, if archive @file{archive.tar} contained
3001 @file{/some/file/name}, then running
3004 tar --extract --file archive.tar --strip-components=2
3008 would extracted this file to file @file{name}.
3010 @opindex suffix, summary
3011 @item --suffix=@var{suffix}
3013 Alters the suffix @command{tar} uses when backing up files from the default
3014 @samp{~}. @xref{backup}.
3016 @opindex tape-length, summary
3017 @item --tape-length=@var{num}
3020 Specifies the length of tapes that @command{tar} is writing as being
3021 @w{@var{num} x 1024} bytes long. @xref{Using Multiple Tapes}.
3023 @opindex test-label, summary
3026 Reads the volume label. If an argument is specified, test whether it
3027 matches the volume label. @xref{--test-label option}.
3029 @opindex to-command, summary
3030 @item --to-command=@var{command}
3032 During extraction @command{tar} will pipe extracted files to the
3033 standard input of @var{command}. @xref{Writing to an External Program}.
3035 @opindex to-stdout, summary
3039 During extraction, @command{tar} will extract files to stdout rather
3040 than to the file system. @xref{Writing to Standard Output}.
3042 @opindex totals, summary
3045 Displays the total number of bytes written after creating an archive.
3048 @opindex touch, summary
3052 Sets the data modification time of extracted files to the extraction time,
3053 rather than the data modification time stored in the archive.
3054 @xref{Data Modification Times}.
3056 @opindex uncompress, summary
3059 (See @option{--compress}. @pxref{gzip})
3061 @opindex ungzip, summary
3064 (See @option{--gzip}. @pxref{gzip})
3066 @opindex unlink-first, summary
3067 @item --unlink-first
3070 Directs @command{tar} to remove the corresponding file from the file
3071 system before extracting it from the archive. @xref{Unlink First}.
3073 @opindex use-compress-program, summary
3074 @item --use-compress-program=@var{prog}
3076 Instructs @command{tar} to access the archive through @var{prog}, which is
3077 presumed to be a compression program of some sort. @xref{gzip}.
3079 @opindex utc, summary
3082 Display file modification dates in @acronym{UTC}. This option implies
3085 @opindex verbose, summary
3089 Specifies that @command{tar} should be more verbose about the operations its
3090 performing. This option can be specified multiple times for some
3091 operations to increase the amount of information displayed.
3094 @opindex verify, summary
3098 Verifies that the archive was correctly written when creating an
3099 archive. @xref{verify}.
3101 @opindex version, summary
3104 @command{tar} will print an informational message about what version
3105 it is and a copyright message, some credits, and then exit.
3108 @opindex volno-file, summary
3109 @item --volno-file=@var{file}
3111 Used in conjunction with @option{--multi-volume}. @command{tar} will keep track
3112 of which volume of a multi-volume archive its working in @var{file}.
3115 @opindex wildcards, summary
3117 Use wildcards when excluding files.
3118 @xref{controlling pattern-matching with exclude}.
3120 @opindex wildcards-match-slash, summary
3121 @item --wildcards-match-slash
3122 Wildcards match @samp{/} when excluding files.
3123 @xref{controlling pattern-matching with exclude}.
3126 @node Short Option Summary
3127 @subsection Short Options Cross Reference
3129 Here is an alphabetized list of all of the short option forms, matching
3130 them with the equivalent long option.
3136 @option{--concatenate}
3140 @option{--read-full-records}
3144 @option{--directory}
3148 @option{--info-script}
3152 @option{--incremental}
3156 @option{--starting-file}
3160 @option{--tape-length}
3164 @option{--multi-volume}
3172 @option{--to-stdout}
3176 @option{--absolute-names}
3180 @option{--block-number}
3188 @option{--files-from}
3192 @option{--unlink-first}
3204 @option{--exclude-from}
3212 @option{--blocking-factor}
3228 @option{--listed-incremental}
3232 @option{--dereference}
3236 @option{--ignore-zeros}
3244 @option{--keep-old-files}
3248 @option{--one-file-system}. Use of this short option is deprecated. It
3249 is retained for compatibility with the earlier versions of GNU
3250 @command{tar}, and will be changed in future releases.
3252 @xref{Changes}, for more information.
3260 When creating --- @option{--no-same-owner}, when extracting ---
3261 @option{--portability}.
3263 The later usage is deprecated. It is retained for compatibility with
3264 the earlier versions of @GNUTAR{}. In the future releases
3265 @option{-o} will be equivalent to @option{--no-same-owner} only.
3269 @option{--preserve-permissions}
3277 @option{--same-order}
3293 @option{--interactive}
3306 @section @GNUTAR{} documentation
3308 @cindex Getting program version number
3310 @cindex Version of the @command{tar} program
3311 Being careful, the first thing is really checking that you are using
3312 @GNUTAR{}, indeed. The @option{--version} option
3313 will generate a message giving confirmation that you are using
3314 @GNUTAR{}, with the precise version of @GNUTAR{}
3315 you are using. @command{tar} identifies itself and
3316 prints the version number to the standard output, then immediately
3317 exits successfully, without doing anything else, ignoring all other
3318 options. For example, @w{@samp{tar --version}} might return:
3321 tar (@acronym{GNU} tar) @value{VERSION}
3325 The first occurrence of @samp{tar} in the result above is the program
3326 name in the package (for example, @command{rmt} is another program),
3327 while the second occurrence of @samp{tar} is the name of the package
3328 itself, containing possibly many programs. The package is currently
3329 named @samp{tar}, after the name of the main program it
3330 contains@footnote{There are plans to merge the @command{cpio} and
3331 @command{tar} packages into a single one which would be called
3332 @code{paxutils}. So, who knows if, one of this days, the
3333 @option{--version} would not yield @w{@samp{tar (@acronym{GNU}
3336 @cindex Obtaining help
3337 @cindex Listing all @command{tar} options
3338 @opindex help, introduction
3339 Another thing you might want to do is checking the spelling or meaning
3340 of some particular @command{tar} option, without resorting to this
3341 manual, for once you have carefully read it. @GNUTAR{}
3342 has a short help feature, triggerable through the
3343 @option{--help} option. By using this option, @command{tar} will
3344 print a usage message listing all available options on standard
3345 output, then exit successfully, without doing anything else and
3346 ignoring all other options. Even if this is only a brief summary, it
3347 may be several screens long. So, if you are not using some kind of
3348 scrollable window, you might prefer to use something like:
3351 $ @kbd{tar --help | less}
3355 presuming, here, that you like using @command{less} for a pager. Other
3356 popular pagers are @command{more} and @command{pg}. If you know about some
3357 @var{keyword} which interests you and do not want to read all the
3358 @option{--help} output, another common idiom is doing:
3361 tar --help | grep @var{keyword}
3365 for getting only the pertinent lines. Notice, however, that some
3366 @command{tar} options have long description lines and the above
3367 command will list only the first of them.
3369 The exact look of the option summary displayed by @kbd{tar --help} is
3370 configurable. @xref{Configuring Help Summary}, for a detailed description.
3373 If you only wish to check the spelling of an option, running @kbd{tar
3374 --usage} may be a better choice. This will display a terse list of
3375 @command{tar} option without accompanying explanations.
3377 The short help output is quite succinct, and you might have to get
3378 back to the full documentation for precise points. If you are reading
3379 this paragraph, you already have the @command{tar} manual in some
3380 form. This manual is available in a variety of forms from
3381 @url{http://www.gnu.org/software/tar/manual}. It may printed out of the @GNUTAR{}
3382 distribution, provided you have @TeX{} already installed somewhere,
3383 and a laser printer around. Just configure the distribution, execute
3384 the command @w{@samp{make dvi}}, then print @file{doc/tar.dvi} the
3385 usual way (contact your local guru to know how). If @GNUTAR{}
3386 has been conveniently installed at your place, this
3387 manual is also available in interactive, hypertextual form as an Info
3388 file. Just call @w{@samp{info tar}} or, if you do not have the
3389 @command{info} program handy, use the Info reader provided within
3390 @acronym{GNU} Emacs, calling @samp{tar} from the main Info menu.
3392 There is currently no @code{man} page for @GNUTAR{}.
3393 If you observe such a @code{man} page on the system you are running,
3394 either it does not belong to @GNUTAR{}, or it has not
3395 been produced by @acronym{GNU}. Some package maintainers convert
3396 @kbd{tar --help} output to a man page, using @command{help2man}. In
3397 any case, please bear in mind that the authoritative source of
3398 information about @GNUTAR{} is this Texinfo documentation.
3401 @section Obtaining @GNUTAR{} default values
3403 @opindex show-defaults
3404 @GNUTAR{} has some predefined defaults that are used when you do not
3405 explicitely specify another values. To obtain a list of such
3406 defaults, use @option{--show-defaults} option. This will output the
3407 values in the form of @command{tar} command line options:
3411 @kbd{tar --show-defaults}
3412 --format=gnu -f- -b20 --rmt-command=/etc/rmt --rsh-command=/usr/bin/rsh
3417 The above output shows that this version of @GNUTAR{} defaults to
3418 using @samp{gnu} archive format (@pxref{Formats}), it uses standard
3419 output as the archive, if no @option{--file} option has been given
3420 (@pxref{file tutorial}), the default blocking factor is 20
3421 (@pxref{Blocking Factor}). It also shows the default locations where
3422 @command{tar} will look for @command{rmt} and @command{rsh} binaries.
3425 @section Checking @command{tar} progress
3427 Typically, @command{tar} performs most operations without reporting any
3428 information to the user except error messages. When using @command{tar}
3429 with many options, particularly ones with complicated or
3430 difficult-to-predict behavior, it is possible to make serious mistakes.
3431 @command{tar} provides several options that make observing @command{tar}
3432 easier. These options cause @command{tar} to print information as it
3433 progresses in its job, and you might want to use them just for being
3434 more careful about what is going on, or merely for entertaining
3435 yourself. If you have encountered a problem when operating on an
3436 archive, however, you may need more information than just an error
3437 message in order to solve the problem. The following options can be
3438 helpful diagnostic tools.
3440 @cindex Verbose operation
3442 Normally, the @option{--list} (@option{-t}) command to list an archive
3443 prints just the file names (one per line) and the other commands are
3444 silent. When used with most operations, the @option{--verbose}
3445 (@option{-v}) option causes @command{tar} to print the name of each
3446 file or archive member as it is processed. This and the other options
3447 which make @command{tar} print status information can be useful in
3448 monitoring @command{tar}.
3450 With @option{--create} or @option{--extract}, @option{--verbose} used
3451 once just prints the names of the files or members as they are processed.
3452 Using it twice causes @command{tar} to print a longer listing
3453 (reminiscent of @samp{ls -l}) for each member. Since @option{--list}
3454 already prints the names of the members, @option{--verbose} used once
3455 with @option{--list} causes @command{tar} to print an @samp{ls -l}
3456 type listing of the files in the archive. The following examples both
3457 extract members with long list output:
3460 $ @kbd{tar --extract --file=archive.tar --verbose --verbose}
3461 $ @kbd{tar xvvf archive.tar}
3464 Verbose output appears on the standard output except when an archive is
3465 being written to the standard output, as with @samp{tar --create
3466 --file=- --verbose} (@samp{tar cfv -}, or even @samp{tar cv}---if the
3467 installer let standard output be the default archive). In that case
3468 @command{tar} writes verbose output to the standard error stream.
3470 If @option{--index-file=@var{file}} is specified, @command{tar} sends
3471 verbose output to @var{file} rather than to standard output or standard
3474 @cindex Obtaining total status information
3476 The @option{--totals} option---which is only meaningful when used with
3477 @option{--create} (@option{-c})---causes @command{tar} to print the total
3478 amount written to the archive, after it has been fully created.
3480 @cindex Progress information
3482 The @option{--checkpoint} option prints an occasional message
3483 as @command{tar} reads or writes the archive. In fact, it prints
3484 a message each 10 records read or written. It is designed for
3485 those who don't need the more detailed (and voluminous) output of
3486 @option{--block-number} (@option{-R}), but do want visual confirmation that @command{tar}
3487 is actually making forward progress.
3489 @FIXME{There is some confusion here. It seems that -R once wrote a
3490 message at @samp{every} record read or written.}
3492 @opindex show-omitted-dirs
3493 @anchor{show-omitted-dirs}
3494 The @option{--show-omitted-dirs} option, when reading an archive---with
3495 @option{--list} or @option{--extract}, for example---causes a message
3496 to be printed for each directory in the archive which is skipped.
3497 This happens regardless of the reason for skipping: the directory might
3498 not have been named on the command line (implicitly or explicitly),
3499 it might be excluded by the use of the @option{--exclude=@var{pattern}} option, or
3502 @opindex block-number
3503 @cindex Block number where error occurred
3504 @anchor{block-number}
3505 If @option{--block-number} (@option{-R}) is used, @command{tar} prints, along with
3506 every message it would normally produce, the block number within the
3507 archive where the message was triggered. Also, supplementary messages
3508 are triggered when reading blocks full of NULs, or when hitting end of
3509 file on the archive. As of now, if the archive if properly terminated
3510 with a NUL block, the reading of the file may stop before end of file
3511 is met, so the position of end of file will not usually show when
3512 @option{--block-number} (@option{-R}) is used. Note that @GNUTAR{}
3513 drains the archive before exiting when reading the
3514 archive from a pipe.
3516 @cindex Error message, block number of
3517 This option is especially useful when reading damaged archives, since
3518 it helps pinpoint the damaged sections. It can also be used with
3519 @option{--list} (@option{-t}) when listing a file-system backup tape, allowing you to
3520 choose among several backup tapes when retrieving a file later, in
3521 favor of the tape where the file appears earliest (closest to the
3522 front of the tape). @xref{backup}.
3525 @section Asking for Confirmation During Operations
3526 @cindex Interactive operation
3528 Typically, @command{tar} carries out a command without stopping for
3529 further instructions. In some situations however, you may want to
3530 exclude some files and archive members from the operation (for instance
3531 if disk or storage space is tight). You can do this by excluding
3532 certain files automatically (@pxref{Choosing}), or by performing
3533 an operation interactively, using the @option{--interactive} (@option{-w}) option.
3534 @command{tar} also accepts @option{--confirmation} for this option.
3536 @opindex interactive
3537 When the @option{--interactive} (@option{-w}) option is specified, before
3538 reading, writing, or deleting files, @command{tar} first prints a message
3539 for each such file, telling what operation it intends to take, then asks
3540 for confirmation on the terminal. The actions which require
3541 confirmation include adding a file to the archive, extracting a file
3542 from the archive, deleting a file from the archive, and deleting a file
3543 from disk. To confirm the action, you must type a line of input
3544 beginning with @samp{y}. If your input line begins with anything other
3545 than @samp{y}, @command{tar} skips that file.
3547 If @command{tar} is reading the archive from the standard input,
3548 @command{tar} opens the file @file{/dev/tty} to support the interactive
3551 Verbose output is normally sent to standard output, separate from
3552 other error messages. However, if the archive is produced directly
3553 on standard output, then verbose output is mixed with errors on
3554 @code{stderr}. Producing the archive on standard output may be used
3555 as a way to avoid using disk space, when the archive is soon to be
3556 consumed by another process reading it, say. Some people felt the need
3557 of producing an archive on stdout, still willing to segregate between
3558 verbose output and error output. A possible approach would be using a
3559 named pipe to receive the archive, and having the consumer process to
3560 read from that named pipe. This has the advantage of letting standard
3561 output free to receive verbose output, all separate from errors.
3564 @chapter @GNUTAR{} Operations
3577 @section Basic @GNUTAR{} Operations
3579 The basic @command{tar} operations, @option{--create} (@option{-c}),
3580 @option{--list} (@option{-t}) and @option{--extract} (@option{--get},
3581 @option{-x}), are currently presented and described in the tutorial
3582 chapter of this manual. This section provides some complementary notes
3583 for these operations.
3586 @opindex create, complementary notes
3590 Creating an empty archive would have some kind of elegance. One can
3591 initialize an empty archive and later use @option{--append}
3592 (@option{-r}) for adding all members. Some applications would not
3593 welcome making an exception in the way of adding the first archive
3594 member. On the other hand, many people reported that it is
3595 dangerously too easy for @command{tar} to destroy a magnetic tape with
3596 an empty archive@footnote{This is well described in @cite{Unix-haters
3597 Handbook}, by Simson Garfinkel, Daniel Weise & Steven Strassmann, IDG
3598 Books, ISBN 1-56884-203-1.}. The two most common errors are:
3602 Mistakingly using @code{create} instead of @code{extract}, when the
3603 intent was to extract the full contents of an archive. This error
3604 is likely: keys @kbd{c} and @kbd{x} are right next to each other on
3605 the QWERTY keyboard. Instead of being unpacked, the archive then
3606 gets wholly destroyed. When users speak about @dfn{exploding} an
3607 archive, they usually mean something else :-).
3610 Forgetting the argument to @code{file}, when the intent was to create
3611 an archive with a single file in it. This error is likely because a
3612 tired user can easily add the @kbd{f} key to the cluster of option
3613 letters, by the mere force of habit, without realizing the full
3614 consequence of doing so. The usual consequence is that the single
3615 file, which was meant to be saved, is rather destroyed.
3618 So, recognizing the likelihood and the catastrophical nature of these
3619 errors, @GNUTAR{} now takes some distance from elegance, and
3620 cowardly refuses to create an archive when @option{--create} option is
3621 given, there are no arguments besides options, and
3622 @option{--files-from} (@option{-T}) option is @emph{not} used. To get
3623 around the cautiousness of @GNUTAR{} and nevertheless create an
3624 archive with nothing in it, one may still use, as the value for the
3625 @option{--files-from} option, a file with no names in it, as shown in
3626 the following commands:
3629 @kbd{tar --create --file=empty-archive.tar --files-from=/dev/null}
3630 @kbd{tar cfT empty-archive.tar /dev/null}
3633 @opindex extract, complementary notes
3638 A socket is stored, within a @GNUTAR{} archive, as a pipe.
3640 @item @option{--list} (@option{-t})
3642 @GNUTAR{} now shows dates as @samp{1996-08-30},
3643 while it used to show them as @samp{Aug 30 1996}. Preferably,
3644 people should get used to ISO 8601 dates. Local American dates should
3645 be made available again with full date localization support, once
3646 ready. In the meantime, programs not being localizable for dates
3647 should prefer international dates, that's really the way to go.
3649 Look up @url{http://www.ft.uni-erlangen.de/~mskuhn/iso-time.html} if you
3650 are curious, it contains a detailed explanation of the ISO 8601 standard.
3655 @section Advanced @GNUTAR{} Operations
3657 Now that you have learned the basics of using @GNUTAR{}, you may want
3658 to learn about further ways in which @command{tar} can help you.
3660 This chapter presents five, more advanced operations which you probably
3661 won't use on a daily basis, but which serve more specialized functions.
3662 We also explain the different styles of options and why you might want
3663 to use one or another, or a combination of them in your @command{tar}
3664 commands. Additionally, this chapter includes options which allow you to
3665 define the output from @command{tar} more carefully, and provide help and
3666 error correction in special circumstances.
3668 @FIXME{check this after the chapter is actually revised to make sure
3669 it still introduces the info in the chapter correctly : ).}
3681 @subsection The Five Advanced @command{tar} Operations
3684 In the last chapter, you learned about the first three operations to
3685 @command{tar}. This chapter presents the remaining five operations to
3686 @command{tar}: @option{--append}, @option{--update}, @option{--concatenate},
3687 @option{--delete}, and @option{--compare}.
3689 You are not likely to use these operations as frequently as those
3690 covered in the last chapter; however, since they perform specialized
3691 functions, they are quite useful when you do need to use them. We
3692 will give examples using the same directory and files that you created
3693 in the last chapter. As you may recall, the directory is called
3694 @file{practice}, the files are @samp{jazz}, @samp{blues}, @samp{folk},
3695 @samp{rock}, and the two archive files you created are
3696 @samp{collection.tar} and @samp{music.tar}.
3698 We will also use the archive files @samp{afiles.tar} and
3699 @samp{bfiles.tar}. @samp{afiles.tar} contains the members @samp{apple},
3700 @samp{angst}, and @samp{aspic}. @samp{bfiles.tar} contains the members
3701 @samp{./birds}, @samp{baboon}, and @samp{./box}.
3703 Unless we state otherwise, all practicing you do and examples you follow
3704 in this chapter will take place in the @file{practice} directory that
3705 you created in the previous chapter; see @ref{prepare for examples}.
3706 (Below in this section, we will remind you of the state of the examples
3707 where the last chapter left them.)
3709 The five operations that we will cover in this chapter are:
3714 Add new entries to an archive that already exists.
3717 Add more recent copies of archive members to the end of an archive, if
3722 Add one or more pre-existing archives to the end of another archive.
3724 Delete items from an archive (does not work on tapes).
3728 Compare archive members to their counterparts in the file system.
3732 @subsection How to Add Files to Existing Archives: @option{--append}
3736 If you want to add files to an existing archive, you don't need to
3737 create a new archive; you can use @option{--append} (@option{-r}).
3738 The archive must already exist in order to use @option{--append}. (A
3739 related operation is the @option{--update} operation; you can use this
3740 to add newer versions of archive members to an existing archive. To learn how to
3741 do this with @option{--update}, @pxref{update}.)
3743 If you use @option{--append} to add a file that has the same name as an
3744 archive member to an archive containing that archive member, then the
3745 old member is not deleted. What does happen, however, is somewhat
3746 complex. @command{tar} @emph{allows} you to have infinite number of files
3747 with the same name. Some operations treat these same-named members no
3748 differently than any other set of archive members: for example, if you
3749 view an archive with @option{--list} (@option{-t}), you will see all
3750 of those members listed, with their data modification times, owners, etc.
3752 Other operations don't deal with these members as perfectly as you might
3753 prefer; if you were to use @option{--extract} to extract the archive,
3754 only the most recently added copy of a member with the same name as four
3755 other members would end up in the working directory. This is because
3756 @option{--extract} extracts an archive in the order the members appeared
3757 in the archive; the most recently archived members will be extracted
3758 last. Additionally, an extracted member will @emph{replace} a file of
3759 the same name which existed in the directory already, and @command{tar}
3760 will not prompt you about this@footnote{Unless you give it
3761 @option{--keep-old-files} option, or the disk copy is newer than the
3762 the one in the archive and you invoke @command{tar} with
3763 @option{--keep-newer-files} option}. Thus, only the most recently archived
3764 member will end up being extracted, as it will replace the one
3765 extracted before it, and so on.
3767 There exists a special option that allows you to get around this
3768 behavior and extract (or list) only a particular copy of the file.
3769 This is @option{--occurrence} option. If you run @command{tar} with
3770 this option, it will extract only the first copy of the file. You
3771 may also give this option an argument specifying the number of
3772 copy to be extracted. Thus, for example if the archive
3773 @file{archive.tar} contained three copies of file @file{myfile}, then
3777 tar --extract --file archive.tar --occurrence=2 myfile
3781 would extract only the second copy. @xref{Option
3782 Summary,---occurrence}, for the description of @option{--occurrence}
3785 @FIXME{ hag -- you might want to incorporate some of the above into the
3786 MMwtSN node; not sure. i didn't know how to make it simpler...
3788 There are a few ways to get around this. (maybe xref Multiple Members
3789 with the Same Name.}
3791 @cindex Members, replacing with other members
3792 @cindex Replacing members with other members
3793 If you want to replace an archive member, use @option{--delete} to
3794 delete the member you want to remove from the archive, , and then use
3795 @option{--append} to add the member you want to be in the archive. Note
3796 that you can not change the order of the archive; the most recently
3797 added member will still appear last. In this sense, you cannot truly
3798 ``replace'' one member with another. (Replacing one member with another
3799 will not work on certain types of media, such as tapes; see @ref{delete}
3800 and @ref{Media}, for more information.)
3803 * appending files:: Appending Files to an Archive
3807 @node appending files
3808 @subsubsection Appending Files to an Archive
3810 @cindex Adding files to an Archive
3811 @cindex Appending files to an Archive
3812 @cindex Archives, Appending files to
3814 The simplest way to add a file to an already existing archive is the
3815 @option{--append} (@option{-r}) operation, which writes specified files into the
3816 archive whether or not they are already among the archived files.
3817 When you use @option{--append}, you @emph{must} specify file name
3818 arguments, as there is no default. If you specify a file that already
3819 exists in the archive, another copy of the file will be added to the
3820 end of the archive. As with other operations, the member names of the
3821 newly added files will be exactly the same as their names given on the
3822 command line. The @option{--verbose} (@option{-v}) option will print
3823 out the names of the files as they are written into the archive.
3825 @option{--append} cannot be performed on some tape drives, unfortunately,
3826 due to deficiencies in the formats those tape drives use. The archive
3827 must be a valid @command{tar} archive, or else the results of using this
3828 operation will be unpredictable. @xref{Media}.
3830 To demonstrate using @option{--append} to add a file to an archive,
3831 create a file called @file{rock} in the @file{practice} directory.
3832 Make sure you are in the @file{practice} directory. Then, run the
3833 following @command{tar} command to add @file{rock} to
3834 @file{collection.tar}:
3837 $ @kbd{tar --append --file=collection.tar rock}
3841 If you now use the @option{--list} (@option{-t}) operation, you will see that
3842 @file{rock} has been added to the archive:
3845 $ @kbd{tar --list --file=collection.tar}
3846 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
3847 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3848 -rw-r--r-- me user 20 1996-09-23 16:44 folk
3849 -rw-r--r-- me user 20 1996-09-23 16:44 rock
3852 @FIXME{in theory, dan will (soon) try to turn this node into what it's
3853 title claims it will become...}
3856 @subsubsection Multiple Files with the Same Name
3858 You can use @option{--append} (@option{-r}) to add copies of files which have been
3859 updated since the archive was created. (However, we do not recommend
3860 doing this since there is another @command{tar} option called
3861 @option{--update}; @pxref{update} for more information. We describe this
3862 use of @option{--append} here for the sake of completeness.) @FIXME{is
3863 this really a good idea, to give this whole description for something
3864 which i believe is basically a Stupid way of doing something? certain
3865 aspects of it show ways in which tar is more broken than i'd personally
3866 like to admit to, specifically the last sentence. On the other hand, i
3867 don't think it's a good idea to be saying that we explicitly don't
3868 recommend using something, but i can't see any better way to deal with
3869 the situation.}When you extract the archive, the older version will be
3870 effectively lost. This works because files are extracted from an
3871 archive in the order in which they were archived. Thus, when the
3872 archive is extracted, a file archived later in time will replace a
3873 file of the same name which was archived earlier, even though the older
3874 version of the file will remain in the archive unless you delete all
3875 versions of the file.
3877 Supposing you change the file @file{blues} and then append the changed
3878 version to @file{collection.tar}. As you saw above, the original
3879 @file{blues} is in the archive @file{collection.tar}. If you change the
3880 file and append the new version of the file to the archive, there will
3881 be two copies in the archive. When you extract the archive, the older
3882 version of the file will be extracted first, and then replaced by the
3883 newer version when it is extracted.
3885 You can append the new, changed copy of the file @file{blues} to the
3886 archive in this way:
3889 $ @kbd{tar --append --verbose --file=collection.tar blues}
3894 Because you specified the @option{--verbose} option, @command{tar} has
3895 printed the name of the file being appended as it was acted on. Now
3896 list the contents of the archive:
3899 $ @kbd{tar --list --verbose --file=collection.tar}
3900 -rw-r--r-- me user 28 1996-10-18 16:31 jazz
3901 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3902 -rw-r--r-- me user 20 1996-09-23 16:44 folk
3903 -rw-r--r-- me user 20 1996-09-23 16:44 rock
3904 -rw-r--r-- me user 58 1996-10-24 18:30 blues
3908 The newest version of @file{blues} is now at the end of the archive
3909 (note the different creation dates and file sizes). If you extract
3910 the archive, the older version of the file @file{blues} will be
3911 replaced by the newer version. You can confirm this by extracting
3912 the archive and running @samp{ls} on the directory.
3914 If you wish to extract the first occurrence of the file @file{blues}
3915 from the archive, use @option{--occurrence} option, as shown in
3916 the following example:
3919 $ @kbd{tar --extract -vv --occurrence --file=collection.tar blues}
3920 -rw-r--r-- me user 21 1996-09-23 16:44 blues
3923 @xref{Writing}, for more information on @option{--extract} and
3924 @xref{Option Summary, --occurrence}, for the description of
3925 @option{--occurrence} option.
3928 @subsection Updating an Archive
3930 @cindex Updating an archive
3933 In the previous section, you learned how to use @option{--append} to
3934 add a file to an existing archive. A related operation is
3935 @option{--update} (@option{-u}). The @option{--update} operation
3936 updates a @command{tar} archive by comparing the date of the specified
3937 archive members against the date of the file with the same name. If
3938 the file has been modified more recently than the archive member, then
3939 the newer version of the file is added to the archive (as with
3942 Unfortunately, you cannot use @option{--update} with magnetic tape drives.
3943 The operation will fail.
3945 @FIXME{other examples of media on which --update will fail? need to ask
3946 charles and/or mib/thomas/dave shevett..}
3948 Both @option{--update} and @option{--append} work by adding to the end
3949 of the archive. When you extract a file from the archive, only the
3950 version stored last will wind up in the file system, unless you use
3951 the @option{--backup} option. @FIXME-ref{Multiple Members with the
3959 @subsubsection How to Update an Archive Using @option{--update}
3961 You must use file name arguments with the @option{--update} (@option{-u}) operation.
3962 If you don't specify any files, @command{tar} won't act on any files and
3963 won't tell you that it didn't do anything (which may end up confusing
3966 @FIXME{note: the above parenthetical added because in fact, this
3967 behavior just confused the author. :-) }
3969 To see the @option{--update} option at work, create a new file,
3970 @file{classical}, in your practice directory, and some extra text to the
3971 file @file{blues}, using any text editor. Then invoke @command{tar} with
3972 the @samp{update} operation and the @option{--verbose} (@option{-v}) option specified,
3973 using the names of all the files in the practice directory as file name
3977 $ @kbd{tar --update -v -f collection.tar blues folk rock classical}
3984 Because we have specified verbose mode, @command{tar} prints out the names
3985 of the files it is working on, which in this case are the names of the
3986 files that needed to be updated. If you run @samp{tar --list} and look
3987 at the archive, you will see @file{blues} and @file{classical} at its
3988 end. There will be a total of two versions of the member @samp{blues};
3989 the one at the end will be newer and larger, since you added text before
3992 (The reason @command{tar} does not overwrite the older file when updating
3993 it is because writing to the middle of a section of tape is a difficult
3994 process. Tapes are not designed to go backward. @xref{Media}, for more
3995 information about tapes.
3997 @option{--update} (@option{-u}) is not suitable for performing backups for two
3998 reasons: it does not change directory content entries, and it
3999 lengthens the archive every time it is used. The @GNUTAR{}
4000 options intended specifically for backups are more
4001 efficient. If you need to run backups, please consult @ref{Backups}.
4004 @subsection Combining Archives with @option{--concatenate}
4006 @cindex Adding archives to an archive
4007 @cindex Concatenating Archives
4008 @opindex concatenate
4010 @c @cindex @option{-A} described
4011 Sometimes it may be convenient to add a second archive onto the end of
4012 an archive rather than adding individual files to the archive. To add
4013 one or more archives to the end of another archive, you should use the
4014 @option{--concatenate} (@option{--catenate}, @option{-A}) operation.
4016 To use @option{--concatenate}, give the first archive with
4017 @option{--file} option and name the rest of archives to be
4018 concatenated on the command line. The members, and their member
4019 names, will be copied verbatim from those archives to the first one.
4020 @FIXME-ref{This can cause multiple members to have the same name, for
4021 information on how this affects reading the archive, Multiple
4022 Members with the Same Name.}
4023 The new, concatenated archive will be called by the same name as the
4024 one given with the @option{--file} option. As usual, if you omit
4025 @option{--file}, @command{tar} will use the value of the environment
4026 variable @env{TAPE}, or, if this has not been set, the default archive name.
4028 @FIXME{There is no way to specify a new name...}
4030 To demonstrate how @option{--concatenate} works, create two small archives
4031 called @file{bluesrock.tar} and @file{folkjazz.tar}, using the relevant
4032 files from @file{practice}:
4035 $ @kbd{tar -cvf bluesrock.tar blues rock}
4038 $ @kbd{tar -cvf folkjazz.tar folk jazz}
4044 If you like, You can run @samp{tar --list} to make sure the archives
4045 contain what they are supposed to:
4048 $ @kbd{tar -tvf bluesrock.tar}
4049 -rw-r--r-- melissa user 105 1997-01-21 19:42 blues
4050 -rw-r--r-- melissa user 33 1997-01-20 15:34 rock
4051 $ @kbd{tar -tvf folkjazz.tar}
4052 -rw-r--r-- melissa user 20 1996-09-23 16:44 folk
4053 -rw-r--r-- melissa user 65 1997-01-30 14:15 jazz
4056 We can concatenate these two archives with @command{tar}:
4060 $ @kbd{tar --concatenate --file=bluesrock.tar jazzfolk.tar}
4063 If you now list the contents of the @file{bluesclass.tar}, you will see
4064 that now it also contains the archive members of @file{jazzfolk.tar}:
4067 $ @kbd{tar --list --file=bluesrock.tar}
4074 When you use @option{--concatenate}, the source and target archives must
4075 already exist and must have been created using compatible format
4076 parameters. Notice, that @command{tar} does not check whether the
4077 archives it concatenates have compatible formats, it does not
4078 even check if the files are really tar archives.
4080 Like @option{--append} (@option{-r}), this operation cannot be performed on some
4081 tape drives, due to deficiencies in the formats those tape drives use.
4083 @cindex @code{concatenate} vs @command{cat}
4084 @cindex @command{cat} vs @code{concatenate}
4085 It may seem more intuitive to you to want or try to use @command{cat} to
4086 concatenate two archives instead of using the @option{--concatenate}
4087 operation; after all, @command{cat} is the utility for combining files.
4089 However, @command{tar} archives incorporate an end-of-file marker which
4090 must be removed if the concatenated archives are to be read properly as
4091 one archive. @option{--concatenate} removes the end-of-archive marker
4092 from the target archive before each new archive is appended. If you use
4093 @command{cat} to combine the archives, the result will not be a valid
4094 @command{tar} format archive. If you need to retrieve files from an
4095 archive that was added to using the @command{cat} utility, use the
4096 @option{--ignore-zeros} (@option{-i}) option. @xref{Ignore Zeros}, for further
4097 information on dealing with archives improperly combined using the
4098 @command{cat} shell utility.
4101 @subsection Removing Archive Members Using @option{--delete}
4103 @cindex Deleting files from an archive
4104 @cindex Removing files from an archive
4107 You can remove members from an archive by using the @option{--delete}
4108 option. Specify the name of the archive with @option{--file}
4109 (@option{-f}) and then specify the names of the members to be deleted;
4110 if you list no member names, nothing will be deleted. The
4111 @option{--verbose} option will cause @command{tar} to print the names
4112 of the members as they are deleted. As with @option{--extract}, you
4113 must give the exact member names when using @samp{tar --delete}.
4114 @option{--delete} will remove all versions of the named file from the
4115 archive. The @option{--delete} operation can run very slowly.
4117 Unlike other operations, @option{--delete} has no short form.
4119 @cindex Tapes, using @option{--delete} and
4120 @cindex Deleting from tape archives
4121 This operation will rewrite the archive. You can only use
4122 @option{--delete} on an archive if the archive device allows you to
4123 write to any point on the media, such as a disk; because of this, it
4124 does not work on magnetic tapes. Do not try to delete an archive member
4125 from a magnetic tape; the action will not succeed, and you will be
4126 likely to scramble the archive and damage your tape. There is no safe
4127 way (except by completely re-writing the archive) to delete files from
4128 most kinds of magnetic tape. @xref{Media}.
4130 To delete all versions of the file @file{blues} from the archive
4131 @file{collection.tar} in the @file{practice} directory, make sure you
4132 are in that directory, and then,
4135 $ @kbd{tar --list --file=collection.tar}
4145 $ @kbd{tar --delete --file=collection.tar blues}
4146 $ @kbd{tar --list --file=collection.tar}
4153 @FIXME{I changed the order of these nodes around and haven't had a chance
4154 to fix the above example's results, yet. I have to play with this and
4155 follow it and see what it actually does!}
4157 The @option{--delete} option has been reported to work properly when
4158 @command{tar} acts as a filter from @code{stdin} to @code{stdout}.
4161 @subsection Comparing Archive Members with the File System
4162 @cindex Verifying the currency of an archive
4166 The @option{--compare} (@option{-d}), or @option{--diff} operation compares
4167 specified archive members against files with the same names, and then
4168 reports differences in file size, mode, owner, modification date and
4169 contents. You should @emph{only} specify archive member names, not file
4170 names. If you do not name any members, then @command{tar} will compare the
4171 entire archive. If a file is represented in the archive but does not
4172 exist in the file system, @command{tar} reports a difference.
4174 You have to specify the record size of the archive when modifying an
4175 archive with a non-default record size.
4177 @command{tar} ignores files in the file system that do not have
4178 corresponding members in the archive.
4180 The following example compares the archive members @file{rock},
4181 @file{blues} and @file{funk} in the archive @file{bluesrock.tar} with
4182 files of the same name in the file system. (Note that there is no file,
4183 @file{funk}; @command{tar} will report an error message.)
4186 $ @kbd{tar --compare --file=bluesrock.tar rock blues funk}
4189 tar: funk not found in archive
4192 The spirit behind the @option{--compare} (@option{--diff}, @option{-d}) option is to check whether the
4193 archive represents the current state of files on disk, more than validating
4194 the integrity of the archive media. For this later goal, @xref{verify}.
4196 @node create options
4197 @section Options Used by @option{--create}
4199 @opindex create, additional options
4200 The previous chapter described the basics of how to use
4201 @option{--create} (@option{-c}) to create an archive from a set of files.
4202 @xref{create}. This section described advanced options to be used with
4206 * Ignore Failed Read::
4209 @node Ignore Failed Read
4210 @subsection Ignore Fail Read
4213 @item --ignore-failed-read
4214 Do not exit with nonzero on unreadable files or directories.
4217 @node extract options
4218 @section Options Used by @option{--extract}
4221 @FIXME{i need to get dan to go over these options with me and see if
4222 there's a better way of organizing them.}
4224 @opindex extract, additional options
4225 The previous chapter showed how to use @option{--extract} to extract
4226 an archive into the file system. Various options cause @command{tar} to
4227 extract more information than just file contents, such as the owner,
4228 the permissions, the modification date, and so forth. This section
4229 presents options to be used with @option{--extract} when certain special
4230 considerations arise. You may review the information presented in
4231 @ref{extract} for more basic information about the
4232 @option{--extract} operation.
4235 * Reading:: Options to Help Read Archives
4236 * Writing:: Changing How @command{tar} Writes Files
4237 * Scarce:: Coping with Scarce Resources
4241 @subsection Options to Help Read Archives
4242 @cindex Options when reading archives
4245 @cindex Reading incomplete records
4246 @cindex Records, incomplete
4247 @opindex read-full-records
4248 Normally, @command{tar} will request data in full record increments from
4249 an archive storage device. If the device cannot return a full record,
4250 @command{tar} will report an error. However, some devices do not always
4251 return full records, or do not require the last record of an archive to
4252 be padded out to the next record boundary. To keep reading until you
4253 obtain a full record, or to accept an incomplete record if it contains
4254 an end-of-archive marker, specify the @option{--read-full-records} (@option{-B}) option
4255 in conjunction with the @option{--extract} or @option{--list} operations.
4258 The @option{--read-full-records} (@option{-B}) option is turned on by default when
4259 @command{tar} reads an archive from standard input, or from a remote
4260 machine. This is because on BSD Unix systems, attempting to read a
4261 pipe returns however much happens to be in the pipe, even if it is
4262 less than was requested. If this option were not enabled, @command{tar}
4263 would fail as soon as it read an incomplete record from the pipe.
4265 If you're not sure of the blocking factor of an archive, you can
4266 read the archive by specifying @option{--read-full-records} (@option{-B}) and
4267 @option{--blocking-factor=@var{512-size}} (@option{-b
4268 @var{512-size}}), using a blocking factor larger than what the archive
4269 uses. This lets you avoid having to determine the blocking factor
4270 of an archive. @xref{Blocking Factor}.
4273 * read full records::
4277 @node read full records
4278 @unnumberedsubsubsec Reading Full Records
4280 @FIXME{need sentence or so of intro here}
4283 @opindex read-full-records
4284 @item --read-full-records
4286 Use in conjunction with @option{--extract} (@option{--get},
4287 @option{-x}) to read an archive which contains incomplete records, or
4288 one which has a blocking factor less than the one specified.
4292 @unnumberedsubsubsec Ignoring Blocks of Zeros
4294 @cindex End-of-archive blocks, ignoring
4295 @cindex Ignoring end-of-archive blocks
4296 @opindex ignore-zeros
4297 Normally, @command{tar} stops reading when it encounters a block of zeros
4298 between file entries (which usually indicates the end of the archive).
4299 @option{--ignore-zeros} (@option{-i}) allows @command{tar} to
4300 completely read an archive which contains a block of zeros before the
4301 end (i.e., a damaged archive, or one that was created by concatenating
4302 several archives together).
4304 The @option{--ignore-zeros} (@option{-i}) option is turned off by default because many
4305 versions of @command{tar} write garbage after the end-of-archive entry,
4306 since that part of the media is never supposed to be read. @GNUTAR{}
4307 does not write after the end of an archive, but seeks to
4308 maintain compatiblity among archiving utilities.
4311 @item --ignore-zeros
4313 To ignore blocks of zeros (i.e., end-of-archive entries) which may be
4314 encountered while reading an archive. Use in conjunction with
4315 @option{--extract} or @option{--list}.
4319 @subsection Changing How @command{tar} Writes Files
4322 @FIXME{Introductory paragraph}
4325 * Dealing with Old Files::
4326 * Overwrite Old Files::
4328 * Keep Newer Files::
4330 * Recursive Unlink::
4331 * Data Modification Times::
4332 * Setting Access Permissions::
4333 * Directory Modification Times and Permissions::
4334 * Writing to Standard Output::
4335 * Writing to an External Program::
4339 @node Dealing with Old Files
4340 @unnumberedsubsubsec Options Controlling the Overwriting of Existing Files
4342 @opindex overwrite-dir, introduced
4343 When extracting files, if @command{tar} discovers that the extracted
4344 file already exists, it normally replaces the file by removing it before
4345 extracting it, to prevent confusion in the presence of hard or symbolic
4346 links. (If the existing file is a symbolic link, it is removed, not
4347 followed.) However, if a directory cannot be removed because it is
4348 nonempty, @command{tar} normally overwrites its metadata (ownership,
4349 permission, etc.). The @option{--overwrite-dir} option enables this
4350 default behavior. To be more cautious and preserve the metadata of
4351 such a directory, use the @option{--no-overwrite-dir} option.
4353 @cindex Overwriting old files, prevention
4354 @opindex keep-old-files, introduced
4355 To be even more cautious and prevent existing files from being replaced, use
4356 the @option{--keep-old-files} (@option{-k}) option. It causes @command{tar} to refuse
4357 to replace or update a file that already exists, i.e., a file with the
4358 same name as an archive member prevents extraction of that archive
4359 member. Instead, it reports an error.
4361 @opindex overwrite, introduced
4362 To be more aggressive about altering existing files, use the
4363 @option{--overwrite} option. It causes @command{tar} to overwrite
4364 existing files and to follow existing symbolic links when extracting.
4366 @cindex Protecting old files
4367 Some people argue that @GNUTAR{} should not hesitate
4368 to overwrite files with other files when extracting. When extracting
4369 a @command{tar} archive, they expect to see a faithful copy of the
4370 state of the file system when the archive was created. It is debatable
4371 that this would always be a proper behavior. For example, suppose one
4372 has an archive in which @file{usr/local} is a link to
4373 @file{usr/local2}. Since then, maybe the site removed the link and
4374 renamed the whole hierarchy from @file{/usr/local2} to
4375 @file{/usr/local}. Such things happen all the time. I guess it would
4376 not be welcome at all that @GNUTAR{} removes the
4377 whole hierarchy just to make room for the link to be reinstated
4378 (unless it @emph{also} simultaneously restores the full
4379 @file{/usr/local2}, of course!) @GNUTAR{} is indeed
4380 able to remove a whole hierarchy to reestablish a symbolic link, for
4381 example, but @emph{only if} @option{--recursive-unlink} is specified
4382 to allow this behavior. In any case, single files are silently
4385 @opindex unlink-first, introduced
4386 Finally, the @option{--unlink-first} (@option{-U}) option can improve performance in
4387 some cases by causing @command{tar} to remove files unconditionally
4388 before extracting them.
4390 @node Overwrite Old Files
4391 @unnumberedsubsubsec Overwrite Old Files
4396 Overwrite existing files and directory metadata when extracting files
4399 This causes @command{tar} to write extracted files into the file system without
4400 regard to the files already on the system; i.e., files with the same
4401 names as archive members are overwritten when the archive is extracted.
4402 It also causes @command{tar} to extract the ownership, permissions,
4403 and time stamps onto any preexisting files or directories.
4404 If the name of a corresponding file name is a symbolic link, the file
4405 pointed to by the symbolic link will be overwritten instead of the
4406 symbolic link itself (if this is possible). Moreover, special devices,
4407 empty directories and even symbolic links are automatically removed if
4408 they are in the way of extraction.
4410 Be careful when using the @option{--overwrite} option, particularly when
4411 combined with the @option{--absolute-names} (@option{-P}) option, as this combination
4412 can change the contents, ownership or permissions of any file on your
4413 system. Also, many systems do not take kindly to overwriting files that
4414 are currently being executed.
4416 @opindex overwrite-dir
4417 @item --overwrite-dir
4418 Overwrite the metadata of directories when extracting files from an
4419 archive, but remove other files before extracting.
4422 @node Keep Old Files
4423 @unnumberedsubsubsec Keep Old Files
4426 @opindex keep-old-files
4427 @item --keep-old-files
4429 Do not replace existing files from archive. The
4430 @option{--keep-old-files} (@option{-k}) option prevents @command{tar}
4431 from replacing existing files with files with the same name from the
4432 archive. The @option{--keep-old-files} option is meaningless with
4433 @option{--list} (@option{-t}). Prevents @command{tar} from replacing
4434 files in the file system during extraction.
4437 @node Keep Newer Files
4438 @unnumberedsubsubsec Keep Newer Files
4441 @opindex keep-newer-files
4442 @item --keep-newer-files
4443 Do not replace existing files that are newer than their archive
4444 copies. This option is meaningless with @option{--list} (@option{-t}).
4448 @unnumberedsubsubsec Unlink First
4451 @opindex unlink-first
4452 @item --unlink-first
4454 Remove files before extracting over them.
4455 This can make @command{tar} run a bit faster if you know in advance
4456 that the extracted files all need to be removed. Normally this option
4457 slows @command{tar} down slightly, so it is disabled by default.
4460 @node Recursive Unlink
4461 @unnumberedsubsubsec Recursive Unlink
4464 @opindex recursive-unlink
4465 @item --recursive-unlink
4466 When this option is specified, try removing files and directory hierarchies
4467 before extracting over them. @emph{This is a dangerous option!}
4470 If you specify the @option{--recursive-unlink} option,
4471 @command{tar} removes @emph{anything} that keeps you from extracting a file
4472 as far as current permissions will allow it. This could include removal
4473 of the contents of a full directory hierarchy.
4475 @node Data Modification Times
4476 @unnumberedsubsubsec Setting Data Modification Times
4478 @cindex Data modification times of extracted files
4479 @cindex Modification times of extracted files
4480 Normally, @command{tar} sets the data modification times of extracted
4481 files to the corresponding times recorded for the files in the archive, but
4482 limits the permissions of extracted files by the current @code{umask}
4485 To set the data modification times of extracted files to the time when
4486 the files were extracted, use the @option{--touch} (@option{-m}) option in
4487 conjunction with @option{--extract} (@option{--get}, @option{-x}).
4493 Sets the data modification time of extracted archive members to the time
4494 they were extracted, not the time recorded for them in the archive.
4495 Use in conjunction with @option{--extract} (@option{--get}, @option{-x}).
4498 @node Setting Access Permissions
4499 @unnumberedsubsubsec Setting Access Permissions
4501 @cindex Permissions of extracted files
4502 @cindex Modes of extracted files
4503 To set the modes (access permissions) of extracted files to those
4504 recorded for those files in the archive, use @option{--same-permissions}
4505 in conjunction with the @option{--extract} (@option{--get},
4506 @option{-x}) operation. @FIXME{Should be aliased to ignore-umask.}
4509 @opindex preserve-permission
4510 @opindex same-permission
4511 @item --preserve-permission
4512 @itemx --same-permission
4513 @c @itemx --ignore-umask
4515 Set modes of extracted archive members to those recorded in the
4516 archive, instead of current umask settings. Use in conjunction with
4517 @option{--extract} (@option{--get}, @option{-x}).
4520 @node Directory Modification Times and Permissions
4521 @unnumberedsubsubsec Directory Modification Times and Permissions
4523 After sucessfully extracting a file member, @GNUTAR{} normally
4524 restores its permissions and modification times, as described in the
4525 previous sections. This cannot be done for directories, because
4526 after extracting a directory @command{tar} will almost certainly
4527 extract files into that directory and this will cause the directory
4528 modification time to be updated. Moreover, restoring that directory
4529 permissions may not permit file creation within it. Thus, restoring
4530 directory permissions and modification times must be delayed at least
4531 until all files have been extracted into that directory. @GNUTAR{}
4532 restores directories using the following approach.
4534 The extracted directories are created with the mode specified in the
4535 archive, as modified by the umask of the user, which gives sufficient
4536 permissions to allow file creation. The meta-information about the
4537 directory is recorded in the temporary list of directories. When
4538 preparing to extract next archive member, @GNUTAR{} checks if the
4539 directory prefix of this file contains the remembered directory. If
4540 it does not, the program assumes that all files have been extracted
4541 into that directory, restores its modification time and permissions
4542 and removes its entry from the internal list. This approach allows
4543 to correctly restore directory meta-information in the majority of
4544 cases, while keeping memory requirements sufficiently small. It is
4545 based on the fact, that most @command{tar} archives use the predefined
4546 order of members: first the directory, then all the files and
4547 subdirectories in that directory.
4549 However, this is not always true. The most important exception are
4550 incremental archives (@pxref{Incremental Dumps}). The member order in
4551 an incremental archive is reversed: first all directory members are
4552 stored, followed by other (non-directory) members. So, when extracting
4553 from incremental archives, @GNUTAR{} alters the above procedure. It
4554 remebers all restored directories, and restores their meta-data
4555 only after the entire archive has been processed. Notice, that you do
4556 not need to specity any special options for that, as @GNUTAR{}
4557 automatically detects archives in incremental format.
4559 There may be cases, when such processing is required for normal archives
4560 too. Consider the following example:
4564 $ @kbd{tar --no-recursion -cvf archive \
4565 foo foo/file1 bar bar/file foo/file2}
4574 During the normal operation, after encountering @file{bar}
4575 @GNUTAR{} will assume that all files from the directory @file{foo}
4576 were already extracted and will therefore restore its timestamp and
4577 permission bits. However, after extracting @file{foo/file2} the
4578 directory timestamp will be offset again.
4580 To correctly restore directory meta-information in such cases, use
4581 @option{delay-directory-restore} command line option:
4584 @opindex delay-directory-restore
4585 @item --delay-directory-restore
4586 Delays restoring of the modification times and permissions of extracted
4587 directories until the end of extraction. This way, correct
4588 meta-information is restored even if the archive has unusual member
4591 @opindex no-delay-directory-restore
4592 @item --no-delay-directory-restore
4593 Cancel the effect of the previous @option{--delay-directory-restore}.
4594 Use this option if you have used @option{--delay-directory-restore} in
4595 @env{TAR_OPTIONS} variable (@pxref{TAR_OPTIONS}) and wish to
4596 temporarily disable it.
4599 @node Writing to Standard Output
4600 @unnumberedsubsubsec Writing to Standard Output
4602 @cindex Writing extracted files to standard output
4603 @cindex Standard output, writing extracted files to
4604 To write the extracted files to the standard output, instead of
4605 creating the files on the file system, use @option{--to-stdout} (@option{-O}) in
4606 conjunction with @option{--extract} (@option{--get}, @option{-x}). This option is useful if you are
4607 extracting files to send them through a pipe, and do not need to
4608 preserve them in the file system. If you extract multiple members,
4609 they appear on standard output concatenated, in the order they are
4610 found in the archive.
4616 Writes files to the standard output. Use only in conjunction with
4617 @option{--extract} (@option{--get}, @option{-x}). When this option is
4618 used, instead of creating the files specified, @command{tar} writes
4619 the contents of the files extracted to its standard output. This may
4620 be useful if you are only extracting the files in order to send them
4621 through a pipe. This option is meaningless with @option{--list}
4625 This can be useful, for example, if you have a tar archive containing
4626 a big file and don't want to store the file on disk before processing
4627 it. You can use a command like this:
4630 tar -xOzf foo.tgz bigfile | process
4633 or even like this if you want to process the concatenation of the files:
4636 tar -xOzf foo.tgz bigfile1 bigfile2 | process
4639 Hovewer, @option{--to-command} may be more convenient for use with
4640 multiple files. See the next section.
4642 @node Writing to an External Program
4643 @unnumberedsubsubsec Writing to an External Program
4645 You can instruct @command{tar} to send the contents of each extracted
4646 file to the standard input of an external program:
4650 @item --to-program=@var{command}
4651 Extract files and pipe their contents to the standard input of
4652 @var{command}. When this option is used, instead of creating the
4653 files specified, @command{tar} invokes @var{command} and pipes the
4654 contents of the files to its standard output. @var{Command} may
4655 contain command line arguments. The program is executed via
4656 @code{sh -c}. Notice, that @var{command} is executed once for each regular file
4657 extracted. Non-regular files (directories, etc.) are ignored when this
4661 The command can obtain the information about the file it processes
4662 from the following environment variables:
4665 @vrindex TAR_FILETYPE, to-command environment
4667 Type of the file. It is a single letter with the following meaning:
4669 @multitable @columnfractions 0.10 0.90
4670 @item f @tab Regular file
4671 @item d @tab Directory
4672 @item l @tab Symbolic link
4673 @item h @tab Hard link
4674 @item b @tab Block device
4675 @item c @tab Character device
4678 Currently only regular files are supported.
4680 @vrindex TAR_MODE, to-command environment
4682 File mode, an octal number.
4684 @vrindex TAR_FILENAME, to-command environment
4686 The name of the file.
4688 @vrindex TAR_REALNAME, to-command environment
4690 Name of the file as stored in the archive.
4692 @vrindex TAR_UNAME, to-command environment
4694 Name of the file owner.
4696 @vrindex TAR_GNAME, to-command environment
4698 Name of the file owner group.
4700 @vrindex TAR_ATIME, to-command environment
4702 Time of last access. It is a decimal number, representing seconds
4703 since the epoch. If the archive provides times with nanosecond
4704 precision, the nanoseconds are appended to the timestamp after a
4707 @vrindex TAR_MTIME, to-command environment
4709 Time of last modification.
4711 @vrindex TAR_CTIME, to-command environment
4713 Time of last status change.
4715 @vrindex TAR_SIZE, to-command environment
4719 @vrindex TAR_UID, to-command environment
4721 UID of the file owner.
4723 @vrindex TAR_GID, to-command environment
4725 GID of the file owner.
4728 In addition to these variables, @env{TAR_VERSION} contains the
4729 @GNUTAR{} version number.
4731 If @var{command} exits with a non-0 status, @command{tar} will print
4732 an error message similar to the following:
4735 tar: 2345: Child returned status 1
4738 Here, @samp{2345} is the PID of the finished process.
4740 If this behavior is not wanted, use @option{--ignore-command-error}:
4743 @opindex ignore-command-error
4744 @item --ignore-command-error
4745 Ignore exit codes of subprocesses. Notice that if the program
4746 exits on signal or otherwise terminates abnormally, the error message
4747 will be printed even if this option is used.
4749 @opindex no-ignore-command-error
4750 @item --no-ignore-command-error
4751 Cancel the effect of any previous @option{--ignore-command-error}
4752 option. This option is useful if you have set
4753 @option{--ignore-command-error} in @env{TAR_OPTIONS}
4754 (@pxref{TAR_OPTIONS}) and wish to temporarily cancel it.
4758 @unnumberedsubsubsec Removing Files
4760 @FIXME{the various macros in the front of the manual think that this
4761 option goes in this section. i have no idea; i only know it's nowhere
4762 else in the book...}
4765 @opindex remove-files
4766 @item --remove-files
4767 Remove files after adding them to the archive.
4771 @subsection Coping with Scarce Resources
4774 @cindex Small memory
4775 @cindex Running out of space
4783 @unnumberedsubsubsec Starting File
4786 @opindex starting-file
4787 @item --starting-file=@var{name}
4788 @itemx -K @var{name}
4789 Starts an operation in the middle of an archive. Use in conjunction
4790 with @option{--extract} (@option{--get}, @option{-x}) or @option{--list} (@option{-t}).
4793 @cindex Middle of the archive, starting in the
4794 If a previous attempt to extract files failed due to lack of disk
4795 space, you can use @option{--starting-file=@var{name}} (@option{-K
4796 @var{name}}) to start extracting only after member @var{name} of the
4797 archive. This assumes, of course, that there is now free space, or
4798 that you are now extracting into a different file system. (You could
4799 also choose to suspend @command{tar}, remove unnecessary files from
4800 the file system, and then restart the same @command{tar} operation.
4801 In this case, @option{--starting-file} is not necessary.
4802 @xref{Incremental Dumps}, @xref{interactive}, and @ref{exclude}.)
4805 @unnumberedsubsubsec Same Order
4808 @cindex Large lists of file names on small machines
4810 @opindex preserve-order
4812 @itemx --preserve-order
4814 To process large lists of file names on machines with small amounts of
4815 memory. Use in conjunction with @option{--compare} (@option{--diff},
4816 @option{-d}), @option{--list} (@option{-t}) or @option{--extract}
4817 (@option{--get}, @option{-x}).
4820 The @option{--same-order} (@option{--preserve-order}, @option{-s}) option tells @command{tar} that the list of file
4821 names to be listed or extracted is sorted in the same order as the
4822 files in the archive. This allows a large list of names to be used,
4823 even on a small machine that would not otherwise be able to hold all
4824 the names in memory at the same time. Such a sorted list can easily be
4825 created by running @samp{tar -t} on the archive and editing its output.
4827 This option is probably never needed on modern computer systems.
4830 @section Backup options
4832 @cindex backup options
4834 @GNUTAR{} offers options for making backups of files
4835 before writing new versions. These options control the details of
4836 these backups. They may apply to the archive itself before it is
4837 created or rewritten, as well as individual extracted members. Other
4838 @acronym{GNU} programs (@command{cp}, @command{install}, @command{ln},
4839 and @command{mv}, for example) offer similar options.
4841 Backup options may prove unexpectedly useful when extracting archives
4842 containing many members having identical name, or when extracting archives
4843 on systems having file name limitations, making different members appear
4844 has having similar names through the side-effect of name truncation.
4845 (This is true only if we have a good scheme for truncated backup names,
4846 which I'm not sure at all: I suspect work is needed in this area.)
4847 When any existing file is backed up before being overwritten by extraction,
4848 then clashing files are automatically be renamed to be unique, and the
4849 true name is kept for only the last file of a series of clashing files.
4850 By using verbose mode, users may track exactly what happens.
4852 At the detail level, some decisions are still experimental, and may
4853 change in the future, we are waiting comments from our users. So, please
4854 do not learn to depend blindly on the details of the backup features.
4855 For example, currently, directories themselves are never renamed through
4856 using these options, so, extracting a file over a directory still has
4857 good chances to fail. Also, backup options apply to created archives,
4858 not only to extracted members. For created archives, backups will not
4859 be attempted when the archive is a block or character device, or when it
4860 refers to a remote file.
4862 For the sake of simplicity and efficiency, backups are made by renaming old
4863 files prior to creation or extraction, and not by copying. The original
4864 name is restored if the file creation fails. If a failure occurs after a
4865 partial extraction of a file, both the backup and the partially extracted
4869 @item --backup[=@var{method}]
4871 @vindex VERSION_CONTROL
4873 Back up files that are about to be overwritten or removed.
4874 Without this option, the original versions are destroyed.
4876 Use @var{method} to determine the type of backups made.
4877 If @var{method} is not specified, use the value of the @env{VERSION_CONTROL}
4878 environment variable. And if @env{VERSION_CONTROL} is not set,
4879 use the @samp{existing} method.
4881 @vindex version-control @r{Emacs variable}
4882 This option corresponds to the Emacs variable @samp{version-control};
4883 the same values for @var{method} are accepted as in Emacs. This option
4884 also allows more descriptive names. The valid @var{method}s are:
4889 @cindex numbered @r{backup method}
4890 Always make numbered backups.
4894 @cindex existing @r{backup method}
4895 Make numbered backups of files that already have them, simple backups
4900 @cindex simple @r{backup method}
4901 Always make simple backups.
4905 @item --suffix=@var{suffix}
4907 @cindex backup suffix
4908 @vindex SIMPLE_BACKUP_SUFFIX
4909 Append @var{suffix} to each backup file made with @option{--backup}. If this
4910 option is not specified, the value of the @env{SIMPLE_BACKUP_SUFFIX}
4911 environment variable is used. And if @env{SIMPLE_BACKUP_SUFFIX} is not
4912 set, the default is @samp{~}, just as in Emacs.
4916 Some people express the desire to @emph{always} use the @option{--backup}
4917 option, by defining some kind of alias or script. This is not as easy
4918 as one may think, due to the fact that old style options should appear first
4919 and consume arguments a bit unpredictably for an alias or script. But,
4920 if you are ready to give up using old style options, you may resort to
4921 using something like (a Bourne shell function here):
4924 tar () @{ /usr/local/bin/tar --backup $*; @}
4928 @section Notable @command{tar} Usages
4931 @FIXME{Using Unix file linking capability to recreate directory
4932 structures---linking files into one subdirectory and then
4933 @command{tar}ring that directory.}
4935 @FIXME{Nice hairy example using absolute-names, newer, etc.}
4938 You can easily use archive files to transport a group of files from
4939 one system to another: put all relevant files into an archive on one
4940 computer system, transfer the archive to another system, and extract
4941 the contents there. The basic transfer medium might be magnetic tape,
4942 Internet FTP, or even electronic mail (though you must encode the
4943 archive with @command{uuencode} in order to transport it properly by
4944 mail). Both machines do not have to use the same operating system, as
4945 long as they both support the @command{tar} program.
4947 For example, here is how you might copy a directory's contents from
4948 one disk to another, while preserving the dates, modes, owners and
4949 link-structure of all the files therein. In this case, the transfer
4950 medium is a @dfn{pipe}, which is one a Unix redirection mechanism:
4953 $ @kbd{cd sourcedir; tar -cf - . | (cd targetdir; tar -xf -)}
4957 The command also works using short option forms:
4960 $ @w{@kbd{cd sourcedir; tar --create --file=- . | (cd targetdir; tar --extract --file=-)}}
4964 This is one of the easiest methods to transfer a @command{tar} archive.
4967 @section Looking Ahead: The Rest of this Manual
4969 You have now seen how to use all eight of the operations available to
4970 @command{tar}, and a number of the possible options. The next chapter
4971 explains how to choose and change file and archive names, how to use
4972 files to store names of other files which you can then call as
4973 arguments to @command{tar} (this can help you save time if you expect to
4974 archive the same list of files a number of times), and so forth.
4975 @FIXME{in case it's not obvious, i'm making this up in some sense
4976 based on my limited memory of what the next chapter *really* does. i
4977 just wanted to flesh out this final section a little bit so i'd
4978 remember to stick it in here. :-)}
4980 If there are too many files to conveniently list on the command line,
4981 you can list the names in a file, and @command{tar} will read that file.
4984 There are various ways of causing @command{tar} to skip over some files,
4985 and not archive them. @xref{Choosing}.
4988 @chapter Performing Backups and Restoring Files
4991 @GNUTAR{} is distributed along with the scripts
4992 which the Free Software Foundation uses for performing backups. There
4993 is no corresponding scripts available yet for doing restoration of
4994 files. Even if there is a good chance those scripts may be satisfying
4995 to you, they are not the only scripts or methods available for doing
4996 backups and restore. You may well create your own, or use more
4997 sophisticated packages dedicated to that purpose.
4999 Some users are enthusiastic about @code{Amanda} (The Advanced Maryland
5000 Automatic Network Disk Archiver), a backup system developed by James
5001 da Silva @file{jds@@cs.umd.edu} and available on many Unix systems.
5002 This is free software, and it is available at these places:
5005 http://www.cs.umd.edu/projects/amanda/amanda.html
5006 ftp://ftp.cs.umd.edu/pub/amanda
5011 Here is a possible plan for a future documentation about the backuping
5012 scripts which are provided within the @GNUTAR{}
5018 @item what are dumps
5019 @item different levels of dumps
5021 @item full dump = dump everything
5022 @item level 1, level 2 dumps etc
5023 A level @var{n} dump dumps everything changed since the last level
5026 @item how to use scripts for dumps (ie, the concept)
5028 @item scripts to run after editing backup specs (details)
5030 @item Backup Specs, what is it.
5032 @item how to customize
5033 @item actual text of script [/sp/dump/backup-specs]
5037 @item rsh doesn't work
5038 @item rtape isn't installed
5041 @item the @option{--incremental} option of tar
5044 @item write protection
5045 @item types of media, different sizes and types, useful for different things
5046 @item files and tape marks
5047 one tape mark between files, two at end.
5048 @item positioning the tape
5049 MT writes two at end of write,
5050 backspaces over one when writing again.
5056 This chapter documents both the provided shell scripts and @command{tar}
5057 options which are more specific to usage as a backup tool.
5059 To @dfn{back up} a file system means to create archives that contain
5060 all the files in that file system. Those archives can then be used to
5061 restore any or all of those files (for instance if a disk crashes or a
5062 file is accidentally deleted). File system @dfn{backups} are also
5066 * Full Dumps:: Using @command{tar} to Perform Full Dumps
5067 * Incremental Dumps:: Using @command{tar} to Perform Incremental Dumps
5068 * Backup Levels:: Levels of Backups
5069 * Backup Parameters:: Setting Parameters for Backups and Restoration
5070 * Scripted Backups:: Using the Backup Scripts
5071 * Scripted Restoration:: Using the Restore Script
5075 @section Using @command{tar} to Perform Full Dumps
5081 @cindex corrupted archives
5082 Full dumps should only be made when no other people or programs
5083 are modifying files in the file system. If files are modified while
5084 @command{tar} is making the backup, they may not be stored properly in
5085 the archive, in which case you won't be able to restore them if you
5086 have to. (Files not being modified are written with no trouble, and do
5087 not corrupt the entire archive.)
5089 You will want to use the @option{--label=@var{archive-label}}
5090 (@option{-V @var{archive-label}}) option to give the archive a
5091 volume label, so you can tell what this archive is even if the label
5092 falls off the tape, or anything like that.
5094 Unless the file system you are dumping is guaranteed to fit on
5095 one volume, you will need to use the @option{--multi-volume} (@option{-M}) option.
5096 Make sure you have enough tapes on hand to complete the backup.
5098 If you want to dump each file system separately you will need to use
5099 the @option{--one-file-system} (@option{-l}) option to prevent
5100 @command{tar} from crossing file system boundaries when storing
5103 The @option{--incremental} (@option{-G}) (@pxref{Incremental Dumps})
5104 option is not needed, since this is a complete copy of everything in
5105 the file system, and a full restore from this backup would only be
5106 done onto a completely
5109 Unless you are in a hurry, and trust the @command{tar} program (and your
5110 tapes), it is a good idea to use the @option{--verify} (@option{-W})
5111 option, to make sure your files really made it onto the dump properly.
5112 This will also detect cases where the file was modified while (or just
5113 after) it was being archived. Not all media (notably cartridge tapes)
5114 are capable of being verified, unfortunately.
5116 @node Incremental Dumps
5117 @section Using @command{tar} to Perform Incremental Dumps
5119 @dfn{Incremental backup} is a special form of @GNUTAR{} archive that
5120 stores additional metadata so that exact state of the file system
5121 can be restored when extracting the archive.
5123 @GNUTAR{} currently offers two options for handling incremental
5124 backups: @option{--listed-incremental=@var{snapshot-file}} (@option{-g
5125 @var{snapshot-file}}) and @option{--incremental} (@option{-G}).
5127 @opindex listed-incremental
5128 The option @option{--listed-incremental} instructs tar to operate on
5129 an incremental archive with additional metadata stored in a standalone
5130 file, called a @dfn{snapshot file}. The purpose of this file is to help
5131 determine which files have been changed, added or deleted since the
5132 last backup, so that the next incremental backup will contain only
5133 modified files. The name of the snapshot file is given as an argument
5137 @item --listed-incremental=@var{file}
5138 @itemx -g @var{file}
5139 Handle incremental backups with snapshot data in @var{file}.
5142 To create an incremental backup, you would use
5143 @option{--listed-incremental} together with @option{--create}
5144 (@pxref{create}). For example:
5147 $ @kbd{tar --create \
5148 --file=archive.1.tar \
5149 --listed-incremental=/var/log/usr.snar \
5153 This will create in @file{archive.1.tar} an incremental backup of
5154 the @file{/usr} file system, storing additional metadata in the file
5155 @file{/var/log/usr.snar}. If this file does not exist, it will be
5156 created. The created archive will then be a @dfn{level 0 backup};
5157 please see the next section for more on backup levels.
5159 Otherwise, if the file @file{/var/log/usr.snar} exists, it
5160 determines which files are modified. In this case only these files will be
5161 stored in the archive. Suppose, for example, that after running the
5162 above command, you delete file @file{/usr/doc/old} and create
5163 directory @file{/usr/local/db} with the following contents:
5166 $ @kbd{ls /usr/local/db}
5171 Some time later you create another incremental backup. You will
5175 $ @kbd{tar --create \
5176 --file=archive.2.tar \
5177 --listed-incremental=/var/log/usr.snar \
5179 tar: usr/local/db: Directory is new
5186 The created archive @file{archive.2.tar} will contain only these
5187 three members. This archive is called a @dfn{level 1 backup}. Notice
5188 that @file{/var/log/usr.snar} will be updated with the new data, so if
5189 you plan to create more @samp{level 1} backups, it is necessary to
5190 create a working copy of the snapshot file before running
5191 @command{tar}. The above example will then be modified as follows:
5194 $ @kbd{cp /var/log/usr.snar /var/log/usr.snar-1}
5195 $ @kbd{tar --create \
5196 --file=archive.2.tar \
5197 --listed-incremental=/var/log/usr.snar-1 \
5201 Incremental dumps depend crucially on time stamps, so the results are
5202 unreliable if you modify a file's time stamps during dumping (e.g.,
5203 with the @option{--atime-preserve=replace} option), or if you set the clock
5206 Metadata stored in snapshot files include device numbers, which,
5207 obviously is supposed to be a non-volatile value. However, it turns
5208 out that NFS devices have undependable values when an automounter
5209 gets in the picture. This can lead to a great deal of spurious
5210 redumping in incremental dumps, so it is somewhat useless to compare
5211 two NFS devices numbers over time. The solution implemented currently
5212 is to considers all NFS devices as being equal when it comes to
5213 comparing directories; this is fairly gross, but there does not seem
5214 to be a better way to go.
5216 Note that incremental archives use @command{tar} extensions and may
5217 not be readable by non-@acronym{GNU} versions of the @command{tar} program.
5219 @opindex listed-incremental, using with @option{--extract}
5220 @opindex extract, using with @option{--listed-incremental}
5221 To extract from the incremental dumps, use
5222 @option{--listed-incremental} together with @option{--extract}
5223 option (@pxref{extracting files}). In this case, @command{tar} does
5224 not need to access snapshot file, since all the data necessary for
5225 extraction are stored in the archive itself. So, when extracting, you
5226 can give whatever argument to @option{--listed-incremental}, the usual
5227 practice is to use @option{--listed-incremental=/dev/null}.
5228 Alternatively, you can use @option{--incremental}, which needs no
5229 arguments. In general, @option{--incremental} (@option{-G}) can be
5230 used as a shortcut for @option{--listed-incremental} when listing or
5231 extracting incremental backups (for more information, regarding this
5232 option, @pxref{incremental-op}).
5234 When extracting from the incremental backup @GNUTAR{} attempts to
5235 restore the exact state the file system had when the archive was
5236 created. In particular, it will @emph{delete} those files in the file
5237 system that did not exist in their directories when the archive was
5238 created. If you have created several levels of incremental files,
5239 then in order to restore the exact contents the file system had when
5240 the last level was created, you will need to restore from all backups
5241 in turn. Continuing our example, to restore the state of @file{/usr}
5242 file system, one would do@footnote{Notice, that since both archives
5243 were created withouth @option{-P} option (@pxref{absolute}), these
5244 commands should be run from the root file system.}:
5247 $ @kbd{tar --extract \
5248 --listed-incremental=/dev/null \
5249 --file archive.1.tar}
5250 $ @kbd{tar --extract \
5251 --listed-incremental=/dev/null \
5252 --file archive.2.tar}
5255 To list the contents of an incremental archive, use @option{--list}
5256 (@pxref{list}), as usual. To obtain more information about the
5257 archive, use @option{--listed-incremental} or @option{--incremental}
5258 combined with two @option{--verbose} options@footnote{Two
5259 @option{--verbose} options were selected to avoid breaking usual
5260 verbose listing output (@option{--list --verbose}) when using in
5263 @opindex incremental, using with @option{--list}
5264 @opindex listed-incremental, using with @option{--list}
5265 @opindex list, using with @option{--incremental}
5266 @opindex list, using with @option{--listed-incremental}
5267 Versions of @GNUTAR{} up to 1.15.1 used to dump verbatim binary
5268 contents of the DUMPDIR header (with terminating nulls) when
5269 @option{--incremental} or @option{--listed-incremental} option was
5270 given, no matter what the verbosity level. This behavior, and,
5271 especially, the binary output it produced were considered incovenient
5272 and were changed in version 1.16}:
5275 @kbd{tar --list --incremental --verbose --verbose archive.tar}
5278 This command will print, for each directory in the archive, the list
5279 of files in that directory at the time the archive was created. This
5280 information is put out in a format which is both human-readable and
5281 unambiguous for a program: each file name is printed as
5288 where @var{x} is a letter describing the status of the file: @samp{Y}
5289 if the file is present in the archive, @samp{N} if the file is not
5290 included in the archive, or a @samp{D} if the file is a directory (and
5291 is included in the archive).@FIXME-xref{dumpdir format}. Each such
5292 line is terminated by a newline character. The last line is followed
5293 by an additional newline to indicate the end of the data.
5295 @anchor{incremental-op}The option @option{--incremental} (@option{-G})
5296 gives the same behavior as @option{--listed-incremental} when used
5297 with @option{--list} and @option{--extract} options. When used with
5298 @option{--create} option, it creates an incremental archive without
5299 creating snapshot file. Thus, it is impossible to create several
5300 levels of incremental backups with @option{--incremental} option.
5303 @section Levels of Backups
5305 An archive containing all the files in the file system is called a
5306 @dfn{full backup} or @dfn{full dump}. You could insure your data by
5307 creating a full dump every day. This strategy, however, would waste a
5308 substantial amount of archive media and user time, as unchanged files
5309 are daily re-archived.
5311 It is more efficient to do a full dump only occasionally. To back up
5312 files between full dumps, you can use @dfn{incremental dumps}. A @dfn{level
5313 one} dump archives all the files that have changed since the last full
5316 A typical dump strategy would be to perform a full dump once a week,
5317 and a level one dump once a day. This means some versions of files
5318 will in fact be archived more than once, but this dump strategy makes
5319 it possible to restore a file system to within one day of accuracy by
5320 only extracting two archives---the last weekly (full) dump and the
5321 last daily (level one) dump. The only information lost would be in
5322 files changed or created since the last daily backup. (Doing dumps
5323 more than once a day is usually not worth the trouble).
5325 @GNUTAR{} comes with scripts you can use to do full
5326 and level-one (actually, even level-two and so on) dumps. Using
5327 scripts (shell programs) to perform backups and restoration is a
5328 convenient and reliable alternative to typing out file name lists
5329 and @command{tar} commands by hand.
5331 Before you use these scripts, you need to edit the file
5332 @file{backup-specs}, which specifies parameters used by the backup
5333 scripts and by the restore script. This file is usually located
5334 in @file{/etc/backup} directory. @xref{Backup Parameters}, for its
5335 detailed description. Once the backup parameters are set, you can
5336 perform backups or restoration by running the appropriate script.
5338 The name of the backup script is @code{backup}. The name of the
5339 restore script is @code{restore}. The following sections describe
5340 their use in detail.
5342 @emph{Please Note:} The backup and restoration scripts are
5343 designed to be used together. While it is possible to restore files by
5344 hand from an archive which was created using a backup script, and to create
5345 an archive by hand which could then be extracted using the restore script,
5346 it is easier to use the scripts. @xref{Incremental Dumps}, before
5347 making such an attempt.
5349 @node Backup Parameters
5350 @section Setting Parameters for Backups and Restoration
5352 The file @file{backup-specs} specifies backup parameters for the
5353 backup and restoration scripts provided with @command{tar}. You must
5354 edit @file{backup-specs} to fit your system configuration and schedule
5355 before using these scripts.
5357 Syntactically, @file{backup-specs} is a shell script, containing
5358 mainly variable assignments. However, any valid shell construct
5359 is allowed in this file. Particularly, you may wish to define
5360 functions within that script (e.g., see @code{RESTORE_BEGIN} below).
5361 For more information about shell script syntax, please refer to
5362 @url{http://www.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html#ta
5363 g_02, the definition of the Shell Command Language}. See also
5364 @ref{Top,,Bash Features,bashref,Bash Reference Manual}.
5366 The shell variables controlling behavior of @code{backup} and
5367 @code{restore} are described in the following subsections.
5370 * General-Purpose Variables::
5371 * Magnetic Tape Control::
5373 * backup-specs example:: An Example Text of @file{Backup-specs}
5376 @node General-Purpose Variables
5377 @subsection General-Purpose Variables
5379 @defvr {Backup variable} ADMINISTRATOR
5380 The user name of the backup administrator. @code{Backup} scripts
5381 sends a backup report to this address.
5384 @defvr {Backup variable} BACKUP_HOUR
5385 The hour at which the backups are done. This can be a number from 0
5386 to 23, or the time specification in form @var{hours}:@var{minutes},
5387 or the string @samp{now}.
5389 This variable is used by @code{backup}. Its value may be overridden
5390 using @option{--time} option (@pxref{Scripted Backups}).
5393 @defvr {Backup variable} TAPE_FILE
5395 The device @command{tar} writes the archive to. If @var{TAPE_FILE}
5396 is a remote archive (@pxref{remote-dev}), backup script will suppose
5397 that your @command{mt} is able to access remote devices. If @var{RSH}
5398 (@pxref{RSH}) is set, @option{--rsh-command} option will be added to
5399 invocations of @command{mt}.
5402 @defvr {Backup variable} BLOCKING
5404 The blocking factor @command{tar} will use when writing the dump archive.
5405 @xref{Blocking Factor}.
5408 @defvr {Backup variable} BACKUP_DIRS
5410 A list of file systems to be dumped (for @code{backup}), or restored
5411 (for @code{restore}). You can include any directory
5412 name in the list --- subdirectories on that file system will be
5413 included, regardless of how they may look to other networked machines.
5414 Subdirectories on other file systems will be ignored.
5416 The host name specifies which host to run @command{tar} on, and should
5417 normally be the host that actually contains the file system. However,
5418 the host machine must have @GNUTAR{} installed, and
5419 must be able to access the directory containing the backup scripts and
5420 their support files using the same file name that is used on the
5421 machine where the scripts are run (ie. what @command{pwd} will print
5422 when in that directory on that machine). If the host that contains
5423 the file system does not have this capability, you can specify another
5424 host as long as it can access the file system through NFS.
5426 If the list of file systems is very long you may wish to put it
5427 in a separate file. This file is usually named
5428 @file{/etc/backup/dirs}, but this name may be overridden in
5429 @file{backup-specs} using @code{DIRLIST} variable.
5432 @defvr {Backup variable} DIRLIST
5434 A path to the file containing the list of the file systems to backup
5435 or restore. By default it is @file{/etc/backup/dirs}.
5438 @defvr {Backup variable} BACKUP_FILES
5440 A list of individual files to be dumped (for @code{backup}), or restored
5441 (for @code{restore}). These should be accessible from the machine on
5442 which the backup script is run.
5444 If the list of file systems is very long you may wish to store it
5445 in a separate file. This file is usually named
5446 @file{/etc/backup/files}, but this name may be overridden in
5447 @file{backup-specs} using @code{FILELIST} variable.
5450 @defvr {Backup variable} FILELIST
5452 A path to the file containing the list of the individual files to backup
5453 or restore. By default it is @file{/etc/backup/files}.
5456 @defvr {Backup variable} MT
5458 Full file name of @command{mt} binary.
5461 @defvr {Backup variable} RSH
5463 Full file name of @command{rsh} binary or its equivalent. You may wish to
5464 set it to @code{ssh}, to improve security. In this case you will have
5465 to use public key authentication.
5468 @defvr {Backup variable} RSH_COMMAND
5470 Full file name of @command{rsh} binary on remote mashines. This will
5471 be passed via @option{--rsh-command} option to the remote invocation
5475 @defvr {Backup variable} VOLNO_FILE
5477 Name of temporary file to hold volume numbers. This needs to be accessible
5478 by all the machines which have file systems to be dumped.
5481 @defvr {Backup variable} XLIST
5483 Name of @dfn{exclude file list}. An @dfn{exclude file list} is a file
5484 located on the remote machine and containing the list of files to
5485 be excluded from the backup. Exclude file lists are searched in
5486 /etc/tar-backup directory. A common use for exclude file lists
5487 is to exclude files containing security-sensitive information
5488 (e.g., @file{/etc/shadow} from backups).
5490 This variable affects only @code{backup}.
5493 @defvr {Backup variable} SLEEP_TIME
5495 Time to sleep between dumps of any two successive file systems
5497 This variable affects only @code{backup}.
5500 @defvr {Backup variable} DUMP_REMIND_SCRIPT
5502 Script to be run when it's time to insert a new tape in for the next
5503 volume. Administrators may want to tailor this script for their site.
5504 If this variable isn't set, @GNUTAR{} will display its built-in prompt
5505 @FIXME-xref{describe it somewhere!}, and will expect confirmation from
5509 @defvr {Backup variable} SLEEP_MESSAGE
5511 Message to display on the terminal while waiting for dump time. Usually
5512 this will just be some literal text.
5515 @defvr {Backup variable} TAR
5517 Full file name of the @GNUTAR{} executable. If this is not set, backup
5518 scripts will search @command{tar} in the current shell path.
5521 @node Magnetic Tape Control
5522 @subsection Magnetic Tape Control
5524 Backup scripts access tape device using special @dfn{hook functions}.
5525 These functions take a single argument -- the name of the tape
5526 device. Their names are kept in the following variables:
5528 @defvr {Backup variable} MT_BEGIN
5529 The name of @dfn{begin} function. This function is called before
5530 accessing the drive. By default it retensions the tape:
5536 mt -f "$1" retension
5541 @defvr {Backup variable} MT_REWIND
5542 The name of @dfn{rewind} function. The default definition is as
5555 @defvr {Backup variable} MT_OFFLINE
5556 The name of the function switching the tape off line. By default
5557 it is defined as follows:
5560 MT_OFFLINE=mt_offline
5568 @defvr {Backup variable} MT_STATUS
5569 The name of the function used to obtain the status of the archive device,
5570 including error count. Default definition:
5582 @subsection User Hooks
5584 @dfn{User hooks} are shell functions executed before and after
5585 each @command{tar} invocation. Thus, there are @dfn{backup
5586 hooks}, which are executed before and after dumping each file
5587 system, and @dfn{restore hooks}, executed before and
5588 after restoring a file system. Each user hook is a shell function
5589 taking four arguments:
5591 @deffn {User Hook Function} hook @var{level} @var{host} @var{fs} @var{fsname}
5596 Current backup or restore level.
5599 Name or IP address of the host machine being dumped or restored.
5602 Full path name to the file system being dumped or restored.
5605 File system name with directory separators replaced with colons. This
5606 is useful, e.g., for creating unique files.
5610 Following variables keep the names of user hook functions
5612 @defvr {Backup variable} DUMP_BEGIN
5613 Dump begin function. It is executed before dumping the file system.
5616 @defvr {Backup variable} DUMP_END
5617 Executed after dumping the file system.
5620 @defvr {Backup variable} RESTORE_BEGIN
5621 Executed before restoring the file system.
5624 @defvr {Backup variable} RESTORE_END
5625 Executed after restoring the file system.
5628 @node backup-specs example
5629 @subsection An Example Text of @file{Backup-specs}
5631 The following is an example of @file{backup-specs}:
5634 # site-specific parameters for file system backup.
5636 ADMINISTRATOR=friedman
5638 TAPE_FILE=/dev/nrsmt0
5640 # Use @code{ssh} instead of the less secure @code{rsh}
5642 RSH_COMMAND=/usr/bin/ssh
5644 # Override MT_STATUS function:
5650 # Disable MT_OFFLINE function
5667 apple-gunkies:/com/mailer/gnu
5668 apple-gunkies:/com/archive/gnu"
5670 BACKUP_FILES="/com/mailer/aliases /com/mailer/league*[a-z]"
5674 @node Scripted Backups
5675 @section Using the Backup Scripts
5677 The syntax for running a backup script is:
5680 backup --level=@var{level} --time=@var{time}
5683 The @option{level} option requests the dump level. Thus, to produce
5684 a full dump, specify @code{--level=0} (this is the default, so
5685 @option{--level} may be omitted if its value is @code{0}).
5686 @footnote{For backward compatibility, the @code{backup} will also
5687 try to deduce the requested dump level from the name of the
5688 script itself. If the name consists of a string @samp{level-}
5689 followed by a single decimal digit, that digit is taken as
5690 the dump level number. Thus, you may create a link from @code{backup}
5691 to @code{level-1} and then run @code{level-1} whenever you need to
5692 create a level one dump.}
5694 The @option{--time} option determines when should the backup be
5695 run. @var{Time} may take three forms:
5698 @item @var{hh}:@var{mm}
5700 The dump must be run at @var{hh} hours @var{mm} minutes.
5704 The dump must be run at @var{hh} hours
5708 The dump must be run immediately.
5711 You should start a script with a tape or disk mounted. Once you
5712 start a script, it prompts you for new tapes or disks as it
5713 needs them. Media volumes don't have to correspond to archive
5714 files --- a multi-volume archive can be started in the middle of a
5715 tape that already contains the end of another multi-volume archive.
5716 The @code{restore} script prompts for media by its archive volume,
5717 so to avoid an error message you should keep track of which tape
5718 (or disk) contains which volume of the archive (@pxref{Scripted
5721 The backup scripts write two files on the file system. The first is a
5722 record file in @file{/etc/tar-backup/}, which is used by the scripts
5723 to store and retrieve information about which files were dumped. This
5724 file is not meant to be read by humans, and should not be deleted by
5725 them. @xref{Snapshot Files}, for a more detailed explanation of this
5728 The second file is a log file containing the names of the file systems
5729 and files dumped, what time the backup was made, and any error
5730 messages that were generated, as well as how much space was left in
5731 the media volume after the last volume of the archive was written.
5732 You should check this log file after every backup. The file name is
5733 @file{log-@var{mm-dd-yyyy}-level-@var{n}}, where @var{mm-dd-yyyy}
5734 represents current date, and @var{n} represents current dump level number.
5736 The script also prints the name of each system being dumped to the
5739 Following is the full list of options accepted by @code{backup}
5743 @item -l @var{level}
5744 @itemx --level=@var{level}
5745 Do backup level @var{level} (default 0).
5749 Force backup even if today's log file already exists.
5751 @item -v[@var{level}]
5752 @itemx --verbose[=@var{level}]
5753 Set verbosity level. The higher the level is, the more debugging
5754 information will be output during execution. Devault @var{level}
5755 is 100, which means the highest debugging level.
5757 @item -t @var{start-time}
5758 @itemx --time=@var{start-time}
5759 Wait till @var{time}, then do backup.
5763 Display short help message and exit.
5767 Display program license and exit.
5771 Display program version and exit.
5775 @node Scripted Restoration
5776 @section Using the Restore Script
5778 To restore files that were archived using a scripted backup, use the
5779 @code{restore} script. Its usage is quite straightforward. In the
5780 simplest form, invoke @code{restore --all}, it will
5781 then restore all the file systems and files specified in
5782 @file{backup-specs} (@pxref{General-Purpose Variables,BACKUP_DIRS}).
5784 You may select the file systems (and/or files) to restore by
5785 giving @code{restore} list of @dfn{patterns} in its command
5786 line. For example, running
5793 will restore all file systems on the machine @samp{albert}. A more
5794 complicated example:
5797 restore 'albert:*' '*:/var'
5801 This command will restore all file systems on the machine @samp{albert}
5802 as well as @file{/var} file system on all machines.
5804 By default @code{restore} will start restoring files from the lowest
5805 available dump level (usually zero) and will continue through
5806 all available dump levels. There may be situations where such a
5807 thorough restore is not necessary. For example, you may wish to
5808 restore only files from the recent level one backup. To do so,
5809 use @option{--level} option, as shown in the example below:
5815 The full list of options accepted by @code{restore} follows:
5820 Restore all file systems and files specified in @file{backup-specs}
5822 @item -l @var{level}
5823 @itemx --level=@var{level}
5824 Start restoring from the given backup level, instead of the default 0.
5826 @item -v[@var{level}]
5827 @itemx --verbose[=@var{level}]
5828 Set verbosity level. The higher the level is, the more debugging
5829 information will be output during execution. Devault @var{level}
5830 is 100, which means the highest debugging level.
5834 Display short help message and exit.
5838 Display program license and exit.
5842 Display program version and exit.
5845 You should start the restore script with the media containing the
5846 first volume of the archive mounted. The script will prompt for other
5847 volumes as they are needed. If the archive is on tape, you don't need
5848 to rewind the tape to to its beginning---if the tape head is
5849 positioned past the beginning of the archive, the script will rewind
5850 the tape as needed. @FIXME-xref{Media, for a discussion of tape
5854 @strong{Warning:} The script will delete files from the active file
5855 system if they were not in the file system when the archive was made.
5858 @xref{Incremental Dumps}, for an explanation of how the script makes
5862 @chapter Choosing Files and Names for @command{tar}
5865 @FIXME{Melissa (still) Doesn't Really Like This ``Intro'' Paragraph!!!}
5867 Certain options to @command{tar} enable you to specify a name for your
5868 archive. Other options let you decide which files to include or exclude
5869 from the archive, based on when or whether files were modified, whether
5870 the file names do or don't match specified patterns, or whether files
5871 are in specified directories.
5874 * file:: Choosing the Archive's Name
5875 * Selecting Archive Members::
5876 * files:: Reading Names from a File
5877 * exclude:: Excluding Some Files
5879 * after:: Operating Only on New Files
5880 * recurse:: Descending into Directories
5881 * one:: Crossing File System Boundaries
5885 @section Choosing and Naming Archive Files
5888 @FIXME{should the title of this section actually be, "naming an
5891 @cindex Naming an archive
5892 @cindex Archive Name
5893 @cindex Choosing an archive file
5894 @cindex Where is the archive?
5895 By default, @command{tar} uses an archive file name that was compiled when
5896 it was built on the system; usually this name refers to some physical
5897 tape drive on the machine. However, the person who installed @command{tar}
5898 on the system may not set the default to a meaningful value as far as
5899 most users are concerned. As a result, you will usually want to tell
5900 @command{tar} where to find (or create) the archive. The @option{--file=@var{archive-name}} (@option{-f @var{archive-name}})
5901 option allows you to either specify or name a file to use as the archive
5902 instead of the default archive file location.
5905 @opindex file, short description
5906 @item --file=@var{archive-name}
5907 @itemx -f @var{archive-name}
5908 Name the archive to create or operate on. Use in conjunction with
5912 For example, in this @command{tar} command,
5915 $ @kbd{tar -cvf collection.tar blues folk jazz}
5919 @file{collection.tar} is the name of the archive. It must directly
5920 follow the @option{-f} option, since whatever directly follows @option{-f}
5921 @emph{will} end up naming the archive. If you neglect to specify an
5922 archive name, you may end up overwriting a file in the working directory
5923 with the archive you create since @command{tar} will use this file's name
5924 for the archive name.
5926 An archive can be saved as a file in the file system, sent through a
5927 pipe or over a network, or written to an I/O device such as a tape,
5928 floppy disk, or CD write drive.
5930 @cindex Writing new archives
5931 @cindex Archive creation
5932 If you do not name the archive, @command{tar} uses the value of the
5933 environment variable @env{TAPE} as the file name for the archive. If
5934 that is not available, @command{tar} uses a default, compiled-in archive
5935 name, usually that for tape unit zero (ie. @file{/dev/tu00}).
5936 @command{tar} always needs an archive name.
5938 If you use @file{-} as an @var{archive-name}, @command{tar} reads the
5939 archive from standard input (when listing or extracting files), or
5940 writes it to standard output (when creating an archive). If you use
5941 @file{-} as an @var{archive-name} when modifying an archive,
5942 @command{tar} reads the original archive from its standard input and
5943 writes the entire new archive to its standard output.
5945 @FIXME{might want a different example here; this is already used in
5946 "notable tar usages".}
5949 $ @kbd{cd sourcedir; tar -cf - . | (cd targetdir; tar -xf -)}
5954 @cindex Standard input and output
5955 @cindex tar to standard input and output
5957 To specify an archive file on a device attached to a remote machine,
5961 @kbd{--file=@var{hostname}:/@var{dev}/@var{file name}}
5965 @command{tar} will complete the remote connection, if possible, and
5966 prompt you for a username and password. If you use
5967 @option{--file=@@@var{hostname}:/@var{dev}/@var{file name}}, @command{tar}
5968 will complete the remote connection, if possible, using your username
5969 as the username on the remote machine.
5971 @cindex Local and remote archives
5972 @anchor{local and remote archives}
5973 If the archive file name includes a colon (@samp{:}), then it is assumed
5974 to be a file on another machine. If the archive file is
5975 @samp{@var{user}@@@var{host}:@var{file}}, then @var{file} is used on the
5976 host @var{host}. The remote host is accessed using the @command{rsh}
5977 program, with a username of @var{user}. If the username is omitted
5978 (along with the @samp{@@} sign), then your user name will be used.
5979 (This is the normal @command{rsh} behavior.) It is necessary for the
5980 remote machine, in addition to permitting your @command{rsh} access, to
5981 have the @file{rmt} program installed (This command is included in
5982 the @GNUTAR{} distribution and by default is installed under
5983 @file{@var{prefix}/libexec/rmt}, were @var{prefix} means your
5984 installation prefix). If you need to use a file whose name includes a
5985 colon, then the remote tape drive behavior
5986 can be inhibited by using the @option{--force-local} option.
5988 @FIXME{i know we went over this yesterday, but bob (and now i do again,
5989 too) thinks it's out of the middle of nowhere. it doesn't seem to tie
5990 into what came before it well enough <<i moved it now, is it better
5991 here?>>. bob also comments that if Amanda isn't free software, we
5992 shouldn't mention it..}
5994 When the archive is being created to @file{/dev/null}, @GNUTAR{}
5995 tries to minimize input and output operations. The
5996 Amanda backup system, when used with @GNUTAR{}, has
5997 an initial sizing pass which uses this feature.
5999 @node Selecting Archive Members
6000 @section Selecting Archive Members
6001 @cindex Specifying files to act on
6002 @cindex Specifying archive members
6004 @dfn{File Name arguments} specify which files in the file system
6005 @command{tar} operates on, when creating or adding to an archive, or which
6006 archive members @command{tar} operates on, when reading or deleting from
6007 an archive. @xref{Operations}.
6009 To specify file names, you can include them as the last arguments on
6010 the command line, as follows:
6012 @kbd{tar} @var{operation} [@var{option1} @var{option2} @dots{}] [@var{file name-1} @var{file name-2} @dots{}]
6015 If a file name begins with dash (@samp{-}), preceede it with
6016 @option{--add-file} option to preventit from being treated as an
6019 If you specify a directory name as a file name argument, all the files
6020 in that directory are operated on by @command{tar}.
6022 If you do not specify files when @command{tar} is invoked with
6023 @option{--create} (@option{-c}), @command{tar} operates on all the non-directory files in
6024 the working directory. If you specify either @option{--list} (@option{-t}) or
6025 @option{--extract} (@option{--get}, @option{-x}), @command{tar} operates on all the archive members in the
6026 archive. If you specify any operation other than one of these three,
6027 @command{tar} does nothing.
6029 By default, @command{tar} takes file names from the command line. However,
6030 there are other ways to specify file or member names, or to modify the
6031 manner in which @command{tar} selects the files or members upon which to
6032 operate. @FIXME{add xref here}In general, these methods work both for
6033 specifying the names of files and archive members.
6036 @section Reading Names from a File
6038 @cindex Reading file names from a file
6039 @cindex Lists of file names
6040 @cindex File Name arguments, alternatives
6041 Instead of giving the names of files or archive members on the command
6042 line, you can put the names into a file, and then use the
6043 @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}}) option to @command{tar}. Give the name of the file
6044 which contains the list of files to include as the argument to
6045 @option{--files-from}. In the list, the file names should be separated by
6046 newlines. You will frequently use this option when you have generated
6047 the list of files to archive with the @command{find} utility.
6051 @item --files-from=@var{file name}
6052 @itemx -T @var{file name}
6053 Get names to extract or create from file @var{file name}.
6056 If you give a single dash as a file name for @option{--files-from}, (i.e.,
6057 you specify either @code{--files-from=-} or @code{-T -}), then the file
6058 names are read from standard input.
6060 Unless you are running @command{tar} with @option{--create}, you can not use
6061 both @code{--files-from=-} and @code{--file=-} (@code{-f -}) in the same
6064 Any number of @option{-T} options can be given in the command line.
6066 @FIXME{add bob's example, from his message on 2-10-97}
6068 The following example shows how to use @command{find} to generate a list of
6069 files smaller than 400K in length and put that list into a file
6070 called @file{small-files}. You can then use the @option{-T} option to
6071 @command{tar} to specify the files from that file, @file{small-files}, to
6072 create the archive @file{little.tgz}. (The @option{-z} option to
6073 @command{tar} compresses the archive with @command{gzip}; @pxref{gzip} for
6077 $ @kbd{find . -size -400 -print > small-files}
6078 $ @kbd{tar -c -v -z -T small-files -f little.tgz}
6082 In the file list given by @option{-T} option, any file name beginning
6083 with @samp{-} character is considered a @command{tar} option and is
6084 processed accordingly.@footnote{Versions of @GNUTAR{} up to 1.15.1
6085 recognized only @option{-C} option in file lists, and only if the
6086 option and its argument occupied two consecutive lines.} For example,
6087 the common use of this feature is to change to another directory by
6088 specifying @option{-C} option:
6098 $ @kbd{tar -c -f foo.tar --files-from list}
6103 In this example, @command{tar} will first switch to @file{/etc}
6104 directory and add files @file{passwd} and @file{hosts} to the
6105 archive. Then it will change to @file{/lib} directory and will archive
6106 the file @file{libc.a}. Thus, the resulting archive @file{foo.tar} will
6111 $ @kbd{tar tf foo.tar}
6119 @opindex directory, using in @option{--files-from} argument
6120 Notice that the option parsing algorithm used with @option{-T} is
6121 stricter than the one used by shell. Namely, when specifying option
6122 arguments, you should observe the following rules:
6126 When using short (single-letter) option form, its argument must
6127 immediately follow the option letter, without any intervening
6128 whitespace. For example: @code{-Cdir}.
6131 When using long option form, the option argument must be separated
6132 from the option by a single equal sign. No whitespace is allowed on
6133 any side of the equal sign. For example: @code{--directory=dir}.
6136 For both short and long option forms, the option argument can be given
6137 on the next line after the option name, e.g.:
6158 If you happen to have a file whose name starts with @samp{-},
6159 precede it with @option{--add-file} option to prevent it from
6160 being recognized as an option. For example: @code{--add-file --my-file}.
6167 @subsection @code{NUL} Terminated File Names
6169 @cindex File names, terminated by @code{NUL}
6170 @cindex @code{NUL} terminated file names
6171 The @option{--null} option causes @option{--files-from=@var{file-of-names}} (@option{-T @var{file-of-names}}) to read file
6172 names terminated by a @code{NUL} instead of a newline, so files whose
6173 names contain newlines can be archived using @option{--files-from}.
6178 Only consider @code{NUL} terminated file names, instead of files that
6179 terminate in a newline.
6182 The @option{--null} option is just like the one in @acronym{GNU}
6183 @command{xargs} and @command{cpio}, and is useful with the
6184 @option{-print0} predicate of @acronym{GNU} @command{find}. In
6185 @command{tar}, @option{--null} also disables special handling for
6186 file names that begin with dash.
6188 This example shows how to use @command{find} to generate a list of files
6189 larger than 800K in length and put that list into a file called
6190 @file{long-files}. The @option{-print0} option to @command{find} is just
6191 like @option{-print}, except that it separates files with a @code{NUL}
6192 rather than with a newline. You can then run @command{tar} with both the
6193 @option{--null} and @option{-T} options to specify that @command{tar} get the
6194 files from that file, @file{long-files}, to create the archive
6195 @file{big.tgz}. The @option{--null} option to @command{tar} will cause
6196 @command{tar} to recognize the @code{NUL} separator between files.
6199 $ @kbd{find . -size +800 -print0 > long-files}
6200 $ @kbd{tar -c -v --null --files-from=long-files --file=big.tar}
6203 @FIXME{say anything else here to conclude the section?}
6206 @section Excluding Some Files
6209 @cindex File names, excluding files by
6210 @cindex Excluding files by name and pattern
6211 @cindex Excluding files by file system
6212 To avoid operating on files whose names match a particular pattern,
6213 use the @option{--exclude} or @option{--exclude-from} options.
6217 @item --exclude=@var{pattern}
6218 Causes @command{tar} to ignore files that match the @var{pattern}.
6222 The @option{--exclude=@var{pattern}} option prevents any file or member whose name
6223 matches the shell wildcard (@var{pattern}) from being operated on.
6224 For example, to create an archive with all the contents of the directory
6225 @file{src} except for files whose names end in @file{.o}, use the
6226 command @samp{tar -cf src.tar --exclude='*.o' src}.
6228 You may give multiple @option{--exclude} options.
6231 @opindex exclude-from
6232 @item --exclude-from=@var{file}
6233 @itemx -X @var{file}
6234 Causes @command{tar} to ignore files that match the patterns listed in
6238 @findex exclude-from
6239 Use the @option{--exclude-from} option to read a
6240 list of patterns, one per line, from @var{file}; @command{tar} will
6241 ignore files matching those patterns. Thus if @command{tar} is
6242 called as @w{@samp{tar -c -X foo .}} and the file @file{foo} contains a
6243 single line @file{*.o}, no files whose names end in @file{.o} will be
6244 added to the archive.
6246 @FIXME{do the exclude options files need to have stuff separated by
6247 newlines the same as the files-from option does?}
6250 @opindex exclude-caches
6251 @item --exclude-caches
6252 Causes @command{tar} to ignore directories containing a cache directory tag.
6255 @findex exclude-caches
6256 When creating an archive, the @option{--exclude-caches} option causes
6257 @command{tar} to exclude all directories that contain a @dfn{cache
6258 directory tag}. A cache directory tag is a short file with the
6259 well-known name @file{CACHEDIR.TAG} and having a standard header
6260 specified in @url{http://www.brynosaurus.com/cachedir/spec.html}.
6261 Various applications write cache directory tags into directories they
6262 use to hold regenerable, non-precious data, so that such data can be
6263 more easily excluded from backups.
6266 * controlling pattern-matching with exclude::
6267 * problems with exclude::
6270 @node controlling pattern-matching with exclude
6271 @unnumberedsubsec Controlling Pattern-Matching with the @code{exclude} Options
6273 Normally, a pattern matches a name if an initial subsequence of the
6274 name's components matches the pattern, where @samp{*}, @samp{?}, and
6275 @samp{[...]} are the usual shell wildcards, @samp{\} escapes wildcards,
6276 and wildcards can match @samp{/}.
6278 Other than optionally stripping leading @samp{/} from names
6279 (@pxref{absolute}), patterns and names are used as-is. For
6280 example, trailing @samp{/} is not trimmed from a user-specified name
6281 before deciding whether to exclude it.
6283 However, this matching procedure can be altered by the options listed
6284 below. These options accumulate. For example:
6287 --ignore-case --exclude='makefile' --no-ignore-case ---exclude='readme'
6290 ignores case when excluding @samp{makefile}, but not when excluding
6295 @opindex no-anchored
6297 @itemx --no-anchored
6298 If anchored, a pattern must match an initial subsequence
6299 of the name's components. Otherwise, the pattern can match any
6300 subsequence. Default is @option{--no-anchored}.
6302 @opindex ignore-case
6303 @opindex no-ignore-case
6305 @itemx --no-ignore-case
6306 When ignoring case, upper-case patterns match lower-case names and vice versa.
6307 When not ignoring case (the default), matching is case-sensitive.
6310 @opindex no-wildcards
6312 @itemx --no-wildcards
6313 When using wildcards (the default), @samp{*}, @samp{?}, and @samp{[...]}
6314 are the usual shell wildcards, and @samp{\} escapes wildcards.
6315 Otherwise, none of these characters are special, and patterns must match
6318 @opindex wildcards-match-slash
6319 @opindex no-wildcards-match-slash
6320 @item --wildcards-match-slash
6321 @itemx --no-wildcards-match-slash
6322 When wildcards match slash (the default), a wildcard like @samp{*} in
6323 the pattern can match a @samp{/} in the name. Otherwise, @samp{/} is
6324 matched only by @samp{/}.
6328 The @option{--recursion} and @option{--no-recursion} options
6329 (@pxref{recurse}) also affect how exclude patterns are interpreted. If
6330 recursion is in effect, a pattern excludes a name if it matches any of
6331 the name's parent directories.
6333 @node problems with exclude
6334 @unnumberedsubsec Problems with Using the @code{exclude} Options
6336 @opindex exclude, potential problems with
6337 Some users find @samp{exclude} options confusing. Here are some common
6342 The main operating mode of @command{tar} does not act on a path name
6343 explicitly listed on the command line if one of its file name
6344 components is excluded. In the example above, if
6345 you create an archive and exclude files that end with @samp{*.o}, but
6346 explicitly name the file @samp{dir.o/foo} after all the options have been
6347 listed, @samp{dir.o/foo} will be excluded from the archive.
6350 You can sometimes confuse the meanings of @option{--exclude} and
6351 @option{--exclude-from}. Be careful: use @option{--exclude} when files
6352 to be excluded are given as a pattern on the command line. Use
6353 @option{--exclude-from} to introduce the name of a file which contains
6354 a list of patterns, one per line; each of these patterns can exclude
6355 zero, one, or many files.
6358 When you use @option{--exclude=@var{pattern}}, be sure to quote the @var{pattern}
6359 parameter, so @GNUTAR{} sees wildcard characters
6360 like @samp{*}. If you do not do this, the shell might expand the
6361 @samp{*} itself using files at hand, so @command{tar} might receive a
6362 list of files instead of one pattern, or none at all, making the
6363 command somewhat illegal. This might not correspond to what you want.
6368 $ @kbd{tar -c -f @var{archive.tar} --exclude '*.o' @var{directory}}
6375 $ @kbd{tar -c -f @var{archive.tar} --exclude *.o @var{directory}}
6379 You must use use shell syntax, or globbing, rather than @code{regexp}
6380 syntax, when using exclude options in @command{tar}. If you try to use
6381 @code{regexp} syntax to describe files to be excluded, your command
6385 In earlier versions of @command{tar}, what is now the
6386 @option{--exclude-from} option was called @option{--exclude} instead.
6387 Now, @option{--exclude} applies to patterns listed on the command
6388 line and @option{--exclude-from} applies to patterns listed in a
6394 @section Wildcards Patterns and Matching
6396 @dfn{Globbing} is the operation by which @dfn{wildcard} characters,
6397 @samp{*} or @samp{?} for example, are replaced and expanded into all
6398 existing files matching the given pattern. However, @command{tar} often
6399 uses wildcard patterns for matching (or globbing) archive members instead
6400 of actual files in the file system. Wildcard patterns are also used for
6401 verifying volume labels of @command{tar} archives. This section has the
6402 purpose of explaining wildcard syntax for @command{tar}.
6404 @FIXME{the next few paragraphs need work.}
6406 A @var{pattern} should be written according to shell syntax, using wildcard
6407 characters to effect globbing. Most characters in the pattern stand
6408 for themselves in the matched string, and case is significant: @samp{a}
6409 will match only @samp{a}, and not @samp{A}. The character @samp{?} in the
6410 pattern matches any single character in the matched string. The character
6411 @samp{*} in the pattern matches zero, one, or more single characters in
6412 the matched string. The character @samp{\} says to take the following
6413 character of the pattern @emph{literally}; it is useful when one needs to
6414 match the @samp{?}, @samp{*}, @samp{[} or @samp{\} characters, themselves.
6416 The character @samp{[}, up to the matching @samp{]}, introduces a character
6417 class. A @dfn{character class} is a list of acceptable characters
6418 for the next single character of the matched string. For example,
6419 @samp{[abcde]} would match any of the first five letters of the alphabet.
6420 Note that within a character class, all of the ``special characters''
6421 listed above other than @samp{\} lose their special meaning; for example,
6422 @samp{[-\\[*?]]} would match any of the characters, @samp{-}, @samp{\},
6423 @samp{[}, @samp{*}, @samp{?}, or @samp{]}. (Due to parsing constraints,
6424 the characters @samp{-} and @samp{]} must either come @emph{first} or
6425 @emph{last} in a character class.)
6427 @cindex Excluding characters from a character class
6428 @cindex Character class, excluding characters from
6429 If the first character of the class after the opening @samp{[}
6430 is @samp{!} or @samp{^}, then the meaning of the class is reversed.
6431 Rather than listing character to match, it lists those characters which
6432 are @emph{forbidden} as the next single character of the matched string.
6434 Other characters of the class stand for themselves. The special
6435 construction @samp{[@var{a}-@var{e}]}, using an hyphen between two
6436 letters, is meant to represent all characters between @var{a} and
6439 @FIXME{need to add a sentence or so here to make this clear for those
6440 who don't have dan around.}
6442 Periods (@samp{.}) or forward slashes (@samp{/}) are not considered
6443 special for wildcard matches. However, if a pattern completely matches
6444 a directory prefix of a matched string, then it matches the full matched
6445 string: excluding a directory also excludes all the files beneath it.
6448 @section Operating Only on New Files
6451 @cindex Excluding file by age
6452 @cindex Data Modification time, excluding files by
6453 @cindex Modification time, excluding files by
6454 @cindex Age, excluding files by
6455 The @option{--after-date=@var{date}} (@option{--newer=@var{date}},
6456 @option{-N @var{date}}) option causes @command{tar} to only work on
6457 files whose data modification or status change times are newer than
6458 the @var{date} given. If @var{date} starts with @samp{/} or @samp{.},
6459 it is taken to be a file name; the data modification time of that file
6460 is used as the date. If you use this option when creating or appending
6461 to an archive, the archive will only include new files. If you use
6462 @option{--after-date} when extracting an archive, @command{tar} will
6463 only extract files newer than the @var{date} you specify.
6465 If you only want @command{tar} to make the date comparison based on
6466 modification of the file's data (rather than status
6467 changes), then use the @option{--newer-mtime=@var{date}} option.
6469 You may use these options with any operation. Note that these options
6470 differ from the @option{--update} (@option{-u}) operation in that they
6471 allow you to specify a particular date against which @command{tar} can
6472 compare when deciding whether or not to archive the files.
6477 @item --after-date=@var{date}
6478 @itemx --newer=@var{date}
6479 @itemx -N @var{date}
6480 Only store files newer than @var{date}.
6482 Acts on files only if their data modification or status change times are
6483 later than @var{date}. Use in conjunction with any operation.
6485 If @var{date} starts with @samp{/} or @samp{.}, it is taken to be a file
6486 name; the data modification time of that file is used as the date.
6488 @opindex newer-mtime
6489 @item --newer-mtime=@var{date}
6490 Acts like @option{--after-date}, but only looks at data modification times.
6493 These options limit @command{tar} to operate only on files which have
6494 been modified after the date specified. A file's status is considered to have
6495 changed if its contents have been modified, or if its owner,
6496 permissions, and so forth, have been changed. (For more information on
6497 how to specify a date, see @ref{Date input formats}; remember that the
6498 entire date argument must be quoted if it contains any spaces.)
6500 Gurus would say that @option{--after-date} tests both the data
6501 modification time (@code{mtime}, the time the contents of the file
6502 were last modified) and the status change time (@code{ctime}, the time
6503 the file's status was last changed: owner, permissions, etc.@:)
6504 fields, while @option{--newer-mtime} tests only the @code{mtime}
6507 To be precise, @option{--after-date} checks @emph{both} @code{mtime} and
6508 @code{ctime} and processes the file if either one is more recent than
6509 @var{date}, while @option{--newer-mtime} only checks @code{mtime} and
6510 disregards @code{ctime}. Neither does it use @code{atime} (the last time the
6511 contents of the file were looked at).
6513 Date specifiers can have embedded spaces. Because of this, you may need
6514 to quote date arguments to keep the shell from parsing them as separate
6517 @FIXME{Need example of --newer-mtime with quoted argument.}
6520 @strong{Please Note:} @option{--after-date} and @option{--newer-mtime}
6521 should not be used for incremental backups. Some files (such as those
6522 in renamed directories) are not selected properly by these options.
6523 @xref{Incremental Dumps}.
6527 @FIXME{which tells -- need to fill this in!}
6530 @section Descending into Directories
6532 @cindex Avoiding recursion in directories
6533 @cindex Descending directories, avoiding
6534 @cindex Directories, avoiding recursion
6535 @cindex Recursion in directories, avoiding
6537 @FIXME{arrggh! this is still somewhat confusing to me. :-< }
6539 @FIXME{show dan bob's comments, from 2-10-97}
6541 Usually, @command{tar} will recursively explore all directories (either
6542 those given on the command line or through the @option{--files-from}
6543 option) for the various files they contain. However, you may not always
6544 want @command{tar} to act this way.
6546 @opindex no-recursion
6547 The @option{--no-recursion} option inhibits @command{tar}'s recursive descent
6548 into specified directories. If you specify @option{--no-recursion}, you can
6549 use the @command{find} utility for hunting through levels of directories to
6550 construct a list of file names which you could then pass to @command{tar}.
6551 @command{find} allows you to be more selective when choosing which files to
6552 archive; see @ref{files} for more information on using @command{find} with
6553 @command{tar}, or look.
6556 @item --no-recursion
6557 Prevents @command{tar} from recursively descending directories.
6561 Requires @command{tar} to recursively descend directories.
6562 This is the default.
6565 When you use @option{--no-recursion}, @GNUTAR{} grabs
6566 directory entries themselves, but does not descend on them
6567 recursively. Many people use @command{find} for locating files they
6568 want to back up, and since @command{tar} @emph{usually} recursively
6569 descends on directories, they have to use the @samp{@w{! -d}} option
6570 to @command{find} @FIXME{needs more explanation or a cite to another
6571 info file}as they usually do not want all the files in a directory.
6572 They then use the @option{--files-from} option to archive the files
6573 located via @command{find}.
6575 The problem when restoring files archived in this manner is that the
6576 directories themselves are not in the archive; so the
6577 @option{--same-permissions} (@option{--preserve-permissions},
6578 @option{-p}) option does not affect them---while users might really
6579 like it to. Specifying @option{--no-recursion} is a way to tell
6580 @command{tar} to grab only the directory entries given to it, adding
6581 no new files on its own.
6583 The @option{--no-recursion} option also applies when extracting: it
6584 causes @command{tar} to extract only the matched directory entries, not
6585 the files under those directories.
6587 The @option{--no-recursion} option also affects how exclude patterns
6588 are interpreted (@pxref{controlling pattern-matching with exclude}).
6590 The @option{--no-recursion} and @option{--recursion} options apply to
6591 later options and operands, and can be overridden by later occurrences
6592 of @option{--no-recursion} and @option{--recursion}. For example:
6595 $ @kbd{tar -cf jams.tar --no-recursion grape --recursion grape/concord}
6599 creates an archive with one entry for @file{grape}, and the recursive
6600 contents of @file{grape/concord}, but no entries under @file{grape}
6601 other than @file{grape/concord}.
6604 @section Crossing File System Boundaries
6605 @cindex File system boundaries, not crossing
6608 @command{tar} will normally automatically cross file system boundaries in
6609 order to archive files which are part of a directory tree. You can
6610 change this behavior by running @command{tar} and specifying
6611 @option{--one-file-system} (@option{-l}). This option only affects files that are
6612 archived because they are in a directory that is being archived;
6613 @command{tar} will still archive files explicitly named on the command line
6614 or through @option{--files-from}, regardless of where they reside.
6617 @opindex one-file-system
6618 @item --one-file-system
6620 Prevents @command{tar} from crossing file system boundaries when
6621 archiving. Use in conjunction with any write operation.
6624 The @option{--one-file-system} option causes @command{tar} to modify its
6625 normal behavior in archiving the contents of directories. If a file in
6626 a directory is not on the same file system as the directory itself, then
6627 @command{tar} will not archive that file. If the file is a directory
6628 itself, @command{tar} will not archive anything beneath it; in other words,
6629 @command{tar} will not cross mount points.
6631 It is reported that using this option, the mount point is is archived,
6632 but nothing under it.
6634 This option is useful for making full or incremental archival backups of
6635 a file system. If this option is used in conjunction with
6636 @option{--verbose} (@option{-v}), files that are excluded are mentioned by name on the
6640 * directory:: Changing Directory
6641 * absolute:: Absolute File Names
6645 @subsection Changing the Working Directory
6648 @FIXME{need to read over this node now for continuity; i've switched
6649 things around some.}
6651 @cindex Changing directory mid-stream
6652 @cindex Directory, changing mid-stream
6653 @cindex Working directory, specifying
6654 To change the working directory in the middle of a list of file names,
6655 either on the command line or in a file specified using
6656 @option{--files-from} (@option{-T}), use @option{--directory} (@option{-C}).
6657 This will change the working directory to the specified directory
6658 after that point in the list.
6662 @item --directory=@var{directory}
6663 @itemx -C @var{directory}
6664 Changes the working directory in the middle of a command line.
6670 $ @kbd{tar -c -f jams.tar grape prune -C food cherry}
6674 will place the files @file{grape} and @file{prune} from the current
6675 directory into the archive @file{jams.tar}, followed by the file
6676 @file{cherry} from the directory @file{food}. This option is especially
6677 useful when you have several widely separated files that you want to
6678 store in the same archive.
6680 Note that the file @file{cherry} is recorded in the archive under the
6681 precise name @file{cherry}, @emph{not} @file{food/cherry}. Thus, the
6682 archive will contain three files that all appear to have come from the
6683 same directory; if the archive is extracted with plain @samp{tar
6684 --extract}, all three files will be written in the current directory.
6686 Contrast this with the command,
6689 $ @kbd{tar -c -f jams.tar grape prune -C food red/cherry}
6693 which records the third file in the archive under the name
6694 @file{red/cherry} so that, if the archive is extracted using
6695 @samp{tar --extract}, the third file will be written in a subdirectory
6696 named @file{orange-colored}.
6698 You can use the @option{--directory} option to make the archive
6699 independent of the original name of the directory holding the files.
6700 The following command places the files @file{/etc/passwd},
6701 @file{/etc/hosts}, and @file{/lib/libc.a} into the archive
6705 $ @kbd{tar -c -f foo.tar -C /etc passwd hosts -C /lib libc.a}
6709 However, the names of the archive members will be exactly what they were
6710 on the command line: @file{passwd}, @file{hosts}, and @file{libc.a}.
6711 They will not appear to be related by file name to the original
6712 directories where those files were located.
6714 Note that @option{--directory} options are interpreted consecutively. If
6715 @option{--directory} specifies a relative file name, it is interpreted
6716 relative to the then current directory, which might not be the same as
6717 the original current working directory of @command{tar}, due to a previous
6718 @option{--directory} option.
6720 When using @option{--files-from} (@pxref{files}), you can put various
6721 @command{tar} options (including @option{-C}) in the file list. Notice,
6722 however, that in this case the option and its argument may not be
6723 separated by whitespace. If you use short option, its argument must
6724 either follow the option letter immediately, without any intervening
6725 whitespace, or occupy the next line. Otherwise, if you use long
6726 option, separate its argument by an equal sign.
6728 For instance, the file list for the above example will be:
6743 To use it, you would invoke @command{tar} as follows:
6746 $ @kbd{tar -c -f foo.tar --files-from list}
6749 Notice also that you can only use the short option variant in the file
6750 list, i.e., always use @option{-C}, not @option{--directory}.
6752 The interpretation of @option{--directory} is disabled by
6753 @option{--null} option.
6756 @subsection Absolute File Names
6760 @opindex absolute-names
6761 @item --absolute-names
6763 Do not strip leading slashes from file names, and permit file names
6764 containing a @file{..} file name component.
6767 By default, @GNUTAR{} drops a leading @samp{/} on
6768 input or output, and complains about file names containing a @file{..}
6769 component. This option turns off this behavior.
6771 When @command{tar} extracts archive members from an archive, it strips any
6772 leading slashes (@samp{/}) from the member name. This causes absolute
6773 member names in the archive to be treated as relative file names. This
6774 allows you to have such members extracted wherever you want, instead of
6775 being restricted to extracting the member in the exact directory named
6776 in the archive. For example, if the archive member has the name
6777 @file{/etc/passwd}, @command{tar} will extract it as if the name were
6778 really @file{etc/passwd}.
6780 File names containing @file{..} can cause problems when extracting, so
6781 @command{tar} normally warns you about such files when creating an
6782 archive, and rejects attempts to extracts such files.
6784 Other @command{tar} programs do not do this. As a result, if you
6785 create an archive whose member names start with a slash, they will be
6786 difficult for other people with a non-@GNUTAR{}
6787 program to use. Therefore, @GNUTAR{} also strips
6788 leading slashes from member names when putting members into the
6789 archive. For example, if you ask @command{tar} to add the file
6790 @file{/bin/ls} to an archive, it will do so, but the member name will
6791 be @file{bin/ls}.@footnote{A side effect of this is that when
6792 @option{--create} is used with @option{--verbose} the resulting output
6793 is not, generally speaking, the same as the one you'd get running
6794 @kbd{tar --list} command. This may be important if you use some
6795 scripts for comparing both outputs. @xref{listing member and file names},
6796 for the information on how to handle this case.}
6798 If you use the @option{--absolute-names} (@option{-P}) option,
6799 @command{tar} will do none of these transformations.
6801 To archive or extract files relative to the root directory, specify
6802 the @option{--absolute-names} (@option{-P}) option.
6804 Normally, @command{tar} acts on files relative to the working
6805 directory---ignoring superior directory names when archiving, and
6806 ignoring leading slashes when extracting.
6808 When you specify @option{--absolute-names} (@option{-P}),
6809 @command{tar} stores file names including all superior directory
6810 names, and preserves leading slashes. If you only invoked
6811 @command{tar} from the root directory you would never need the
6812 @option{--absolute-names} option, but using this option
6813 may be more convenient than switching to root.
6815 @FIXME{Should be an example in the tutorial/wizardry section using this
6816 to transfer files between systems.}
6818 @FIXME{Is write access an issue?}
6821 @item --absolute-names
6822 Preserves full file names (including superior directory names) when
6823 archiving files. Preserves leading slash when extracting files.
6827 @FIXME{this is still horrible; need to talk with dan on monday.}
6829 @command{tar} prints out a message about removing the @samp{/} from
6830 file names. This message appears once per @GNUTAR{}
6831 invocation. It represents something which ought to be told; ignoring
6832 what it means can cause very serious surprises, later.
6834 Some people, nevertheless, do not want to see this message. Wanting to
6835 play really dangerously, one may of course redirect @command{tar} standard
6836 error to the sink. For example, under @command{sh}:
6839 $ @kbd{tar -c -f archive.tar /home 2> /dev/null}
6843 Another solution, both nicer and simpler, would be to change to
6844 the @file{/} directory first, and then avoid absolute notation.
6848 $ @kbd{(cd / && tar -c -f archive.tar home)}
6849 $ @kbd{tar -c -f archive.tar -C / home}
6852 @include getdate.texi
6855 @chapter Controlling the Archive Format
6857 @cindex Tar archive formats
6858 Due to historical reasons, there are several formats of tar archives.
6859 All of them are based on the same principles, but have some subtle
6860 differences that often make them incompatible with each other.
6862 GNU tar is able to create and handle archives in a variety of formats.
6863 The most frequently used formats are (in alphabetical order):
6867 Format used by @GNUTAR{} versions up to 1.13.25. This format derived
6868 from an early @acronym{POSIX} standard, adding some improvements such as
6869 sparse file handling and incremental archives. Unfortunately these
6870 features were implemented in a way incompatible with other archive
6873 Archives in @samp{gnu} format are able to hold pathnames of unlimited
6877 Format used by @GNUTAR{} of versions prior to 1.12.
6880 Archive format, compatible with the V7 implementation of tar. This
6881 format imposes a number of limitations. The most important of them
6885 @item The maximum length of a file name is limited to 99 characters.
6886 @item The maximum length of a symbolic link is limited to 99 characters.
6887 @item It is impossible to store special files (block and character
6888 devices, fifos etc.)
6889 @item Maximum value of user or group ID is limited to 2097151 (7777777
6891 @item V7 archives do not contain symbolic ownership information (user
6892 and group name of the file owner).
6895 This format has traditionally been used by Automake when producing
6896 Makefiles. This practice will change in the future, in the meantime,
6897 however this means that projects containing filenames more than 99
6898 characters long will not be able to use @GNUTAR{} @value{VERSION} and
6899 Automake prior to 1.9.
6902 Archive format defined by @acronym{POSIX.1-1988} specification. It stores
6903 symbolic ownership information. It is also able to store
6904 special files. However, it imposes several restrictions as well:
6907 @item The maximum length of a file name is limited to 256 characters,
6908 provided that the filename can be split at directory separator in
6909 two parts, first of them being at most 155 bytes long. So, in most
6910 cases the maximum file name length will be shorter than 256
6912 @item The maximum length of a symbolic link name is limited to
6914 @item Maximum size of a file the archive is able to accomodate
6916 @item Maximum value of UID/GID is 2097151.
6917 @item Maximum number of bits in device major and minor numbers is 21.
6921 Format used by J@"org Schilling @command{star}
6922 implementation. @GNUTAR{} is able to read @samp{star} archives but
6923 currently does not produce them.
6926 Archive format defined by @acronym{POSIX.1-2001} specification. This is the
6927 most flexible and feature-rich format. It does not impose any
6928 restrictions on file sizes or filename lengths. This format is quite
6929 recent, so not all tar implementations are able to handle it properly.
6930 However, this format is designed in such a way that any tar
6931 implementation able to read @samp{ustar} archives will be able to read
6932 most @samp{posix} archives as well, with the only exception that any
6933 additional information (such as long file names etc.) will in such
6934 case be extracted as plain text files along with the files it refers to.
6936 This archive format will be the default format for future versions
6941 The following table summarizes the limitations of each of these
6944 @multitable @columnfractions .10 .20 .20 .20 .20
6945 @headitem Format @tab UID @tab File Size @tab Path Name @tab Devn
6946 @item gnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
6947 @item oldgnu @tab 1.8e19 @tab Unlimited @tab Unlimited @tab 63
6948 @item v7 @tab 2097151 @tab 8GB @tab 99 @tab n/a
6949 @item ustar @tab 2097151 @tab 8GB @tab 256 @tab 21
6950 @item posix @tab Unlimited @tab Unlimited @tab Unlimited @tab Unlimited
6953 The default format for @GNUTAR{} is defined at compilation
6954 time. You may check it by running @command{tar --help}, and examining
6955 the last lines of its output. Usually, @GNUTAR{} is configured
6956 to create archives in @samp{gnu} format, however, future version will
6957 switch to @samp{posix}.
6960 * Portability:: Making @command{tar} Archives More Portable
6961 * Compression:: Using Less Space through Compression
6962 * Attributes:: Handling File Attributes
6963 * Standard:: The Standard Format
6964 * Extensions:: @acronym{GNU} Extensions to the Archive Format
6965 * cpio:: Comparison of @command{tar} and @command{cpio}
6969 @section Making @command{tar} Archives More Portable
6971 Creating a @command{tar} archive on a particular system that is meant to be
6972 useful later on many other machines and with other versions of @command{tar}
6973 is more challenging than you might think. @command{tar} archive formats
6974 have been evolving since the first versions of Unix. Many such formats
6975 are around, and are not always compatible with each other. This section
6976 discusses a few problems, and gives some advice about making @command{tar}
6977 archives more portable.
6979 One golden rule is simplicity. For example, limit your @command{tar}
6980 archives to contain only regular files and directories, avoiding
6981 other kind of special files. Do not attempt to save sparse files or
6982 contiguous files as such. Let's discuss a few more problems, in turn.
6984 @FIXME{Discuss GNU extensions (incremental backups, multi-volume
6985 archives and archive labels) in GNU and PAX formats.}
6988 * Portable Names:: Portable Names
6989 * dereference:: Symbolic Links
6990 * old:: Old V7 Archives
6991 * ustar:: Ustar Archives
6992 * gnu:: GNU and old GNU format archives.
6993 * posix:: @acronym{POSIX} archives
6994 * Checksumming:: Checksumming Problems
6995 * Large or Negative Values:: Large files, negative time stamps, etc.
6998 @node Portable Names
6999 @subsection Portable Names
7001 Use portable file and member names. A name is portable if it contains
7002 only ASCII letters and digits, @samp{/}, @samp{.}, @samp{_}, and
7003 @samp{-}; it cannot be empty, start with @samp{-} or @samp{//}, or
7004 contain @samp{/-}. Avoid deep directory nesting. For portability to
7005 old Unix hosts, limit your file name components to 14 characters or
7008 If you intend to have your @command{tar} archives to be read under
7009 MSDOS, you should not rely on case distinction for file names, and you
7010 might use the @acronym{GNU} @command{doschk} program for helping you
7011 further diagnosing illegal MSDOS names, which are even more limited
7015 @subsection Symbolic Links
7016 @cindex File names, using symbolic links
7017 @cindex Symbolic link as file name
7019 @opindex dereference
7020 Normally, when @command{tar} archives a symbolic link, it writes a
7021 block to the archive naming the target of the link. In that way, the
7022 @command{tar} archive is a faithful record of the file system contents.
7023 @option{--dereference} (@option{-h}) is used with @option{--create} (@option{-c}), and causes
7024 @command{tar} to archive the files symbolic links point to, instead of
7025 the links themselves. When this option is used, when @command{tar}
7026 encounters a symbolic link, it will archive the linked-to file,
7027 instead of simply recording the presence of a symbolic link.
7029 The name under which the file is stored in the file system is not
7030 recorded in the archive. To record both the symbolic link name and
7031 the file name in the system, archive the file under both names. If
7032 all links were recorded automatically by @command{tar}, an extracted file
7033 might be linked to a file name that no longer exists in the file
7036 If a linked-to file is encountered again by @command{tar} while creating
7037 the same archive, an entire second copy of it will be stored. (This
7038 @emph{might} be considered a bug.)
7040 So, for portable archives, do not archive symbolic links as such,
7041 and use @option{--dereference} (@option{-h}): many systems do not support
7042 symbolic links, and moreover, your distribution might be unusable if
7043 it contains unresolved symbolic links.
7046 @subsection Old V7 Archives
7047 @cindex Format, old style
7048 @cindex Old style format
7049 @cindex Old style archives
7050 @cindex v7 archive format
7052 Certain old versions of @command{tar} cannot handle additional
7053 information recorded by newer @command{tar} programs. To create an
7054 archive in V7 format (not ANSI), which can be read by these old
7055 versions, specify the @option{--format=v7} option in
7056 conjunction with the @option{--create} (@option{-c}) (@command{tar} also
7057 accepts @option{--portability} or @samp{op-old-archive} for this
7058 option). When you specify it,
7059 @command{tar} leaves out information about directories, pipes, fifos,
7060 contiguous files, and device files, and specifies file ownership by
7061 group and user IDs instead of group and user names.
7063 When updating an archive, do not use @option{--format=v7}
7064 unless the archive was created using this option.
7066 In most cases, a @emph{new} format archive can be read by an @emph{old}
7067 @command{tar} program without serious trouble, so this option should
7068 seldom be needed. On the other hand, most modern @command{tar}s are
7069 able to read old format archives, so it might be safer for you to
7070 always use @option{--format=v7} for your distributions.
7073 @subsection Ustar Archive Format
7075 @cindex ustar archive format
7076 Archive format defined by @acronym{POSIX}.1-1988 specification is called
7077 @code{ustar}. Although it is more flexible than the V7 format, it
7078 still has many restrictions (@xref{Formats,ustar}, for the detailed
7079 description of @code{ustar} format). Along with V7 format,
7080 @code{ustar} format is a good choice for archives intended to be read
7081 with other implementations of @command{tar}.
7083 To create archive in @code{ustar} format, use @option{--format=ustar}
7084 option in conjunction with the @option{--create} (@option{-c}).
7087 @subsection @acronym{GNU} and old @GNUTAR{} format
7089 @cindex GNU archive format
7090 @cindex Old GNU archive format
7091 @GNUTAR{} was based on an early draft of the
7092 @acronym{POSIX} 1003.1 @code{ustar} standard. @acronym{GNU} extensions to
7093 @command{tar}, such as the support for file names longer than 100
7094 characters, use portions of the @command{tar} header record which were
7095 specified in that @acronym{POSIX} draft as unused. Subsequent changes in
7096 @acronym{POSIX} have allocated the same parts of the header record for
7097 other purposes. As a result, @GNUTAR{} format is
7098 incompatible with the current @acronym{POSIX} specification, and with
7099 @command{tar} programs that follow it.
7101 In the majority of cases, @command{tar} will be configured to create
7102 this format by default. This will change in the future releases, since
7103 we plan to make @samp{posix} format the default.
7105 To force creation a @GNUTAR{} archive, use option
7106 @option{--format=gnu}.
7109 @subsection @GNUTAR{} and @acronym{POSIX} @command{tar}
7111 @cindex POSIX archive format
7112 @cindex PAX archive format
7113 The version @value{VERSION} of @GNUTAR{} is able
7114 to read and create archives conforming to @acronym{POSIX.1-2001} standard.
7116 A @acronym{POSIX} conformant archive will be created if @command{tar}
7117 was given @option{--format=posix} option.
7120 @subsection Checksumming Problems
7122 SunOS and HP-UX @command{tar} fail to accept archives created using
7123 @GNUTAR{} and containing non-ASCII file names, that
7124 is, file names having characters with the eight bit set, because they
7125 use signed checksums, while @GNUTAR{} uses unsigned
7126 checksums while creating archives, as per @acronym{POSIX} standards. On
7127 reading, @GNUTAR{} computes both checksums and
7128 accept any. It is somewhat worrying that a lot of people may go
7129 around doing backup of their files using faulty (or at least
7130 non-standard) software, not learning about it until it's time to
7131 restore their missing files with an incompatible file extractor, or
7134 @GNUTAR{} compute checksums both ways, and accept
7135 any on read, so @acronym{GNU} tar can read Sun tapes even with their
7136 wrong checksums. @GNUTAR{} produces the standard
7137 checksum, however, raising incompatibilities with Sun. That is to
7138 say, @GNUTAR{} has not been modified to
7139 @emph{produce} incorrect archives to be read by buggy @command{tar}'s.
7140 I've been told that more recent Sun @command{tar} now read standard
7141 archives, so maybe Sun did a similar patch, after all?
7143 The story seems to be that when Sun first imported @command{tar}
7144 sources on their system, they recompiled it without realizing that
7145 the checksums were computed differently, because of a change in
7146 the default signing of @code{char}'s in their compiler. So they
7147 started computing checksums wrongly. When they later realized their
7148 mistake, they merely decided to stay compatible with it, and with
7149 themselves afterwards. Presumably, but I do not really know, HP-UX
7150 has chosen that their @command{tar} archives to be compatible with Sun's.
7151 The current standards do not favor Sun @command{tar} format. In any
7152 case, it now falls on the shoulders of SunOS and HP-UX users to get
7153 a @command{tar} able to read the good archives they receive.
7155 @node Large or Negative Values
7156 @subsection Large or Negative Values
7157 @cindex large values
7158 @cindex future time stamps
7159 @cindex negative time stamps
7162 The above sections suggest to use @samp{oldest possible} archive
7163 format if in doubt. However, sometimes it is not possible. If you
7164 attempt to archive a file whose metadata cannot be represented using
7165 required format, @GNUTAR{} will print error message and ignore such a
7166 file. You will than have to switch to a format that is able to
7167 handle such values. The format summary table (@pxref{Formats}) will
7170 In particular, when trying to archive files larger than 8GB or with
7171 timestamps not in the range 1970-01-01 00:00:00 through 2242-03-16
7172 12:56:31 @sc{utc}, you will have to chose between @acronym{GNU} and
7173 @acronym{POSIX} archive formats. When considering which format to
7174 choose, bear in mind that the @acronym{GNU} format uses
7175 two's-complement base-256 notation to store values that do not fit
7176 into standard @acronym{ustar} range. Such archives can generally be
7177 read only by a @GNUTAR{} implementation. Moreover, they sometimes
7178 cannot be correctly restored on another hosts even by @GNUTAR{}. For
7179 example, using two's complement representation for negative time
7180 stamps that assumes a signed 32-bit @code{time_t} generates archives
7181 that are not portable to hosts with differing @code{time_t}
7184 On the other hand, @acronym{POSIX} archives, generally speaking, can
7185 be extracted by any tar implementation that understands older
7186 @acronym{ustar} format. The only exception are files larger than 8GB.
7188 @FIXME{Describe how @acronym{POSIX} archives are extracted by non
7192 @section Using Less Space through Compression
7195 * gzip:: Creating and Reading Compressed Archives
7196 * sparse:: Archiving Sparse Files
7200 @subsection Creating and Reading Compressed Archives
7201 @cindex Compressed archives
7202 @cindex Storing archives in compressed format
7204 @GNUTAR{} is able to create and read compressed archives. It supports
7205 @command{gzip} and @command{bzip2} compression programs. For backward
7206 compatibilty, it also supports @command{compress} command, although
7207 we strongly recommend against using it, since there is a patent
7208 covering the algorithm it uses and you could be sued for patent
7209 infringement merely by running @command{compress}! Besides, it is less
7210 effective than @command{gzip} and @command{bzip2}.
7212 Creating a compressed archive is simple: you just specify a
7213 @dfn{compression option} along with the usual archive creation
7214 commands. The compression option is @option{-z} (@option{--gzip}) to
7215 create a @command{gzip} compressed archive, @option{-j}
7216 (@option{--bzip2}) to create a @command{bzip2} compressed archive, and
7217 @option{-Z} (@option{--compress}) to use @command{compress} program.
7221 $ @kbd{tar cfz archive.tar.gz .}
7224 Reading compressed archive is even simpler: you don't need to specify
7225 any additional options as @GNUTAR{} recognizes its format
7226 automatically. Thus, the following commands will list and extract the
7227 archive created in previous example:
7230 # List the compressed archive
7231 $ @kbd{tar tf archive.tar.gz}
7232 # Extract the compressed archive
7233 $ @kbd{tar xf archive.tar.gz}
7236 The only case when you have to specify a decompression option while
7237 reading the archive is when reading from a pipe or from a tape drive
7238 that does not support random access. However, in this case @GNUTAR{}
7239 will indicate which option you should use. For example:
7242 $ @kbd{cat archive.tar.gz | tar tf -}
7243 tar: Archive is compressed. Use -z option
7244 tar: Error is not recoverable: exiting now
7247 If you see such diagnostics, just add the suggested option to the
7248 invocation of @GNUTAR{}:
7251 $ @kbd{cat archive.tar.gz | tar tfz -}
7254 Notice also, that there are several restrictions on operations on
7255 compressed archives. First of all, compressed archives cannot be
7256 modified, i.e., you cannot update (@option{--update} (@option{-u})) them or delete
7257 (@option{--delete}) members from them. Likewise, you cannot append
7258 another @command{tar} archive to a compressed archive using
7259 @option{--append} (@option{-r})). Secondly, multi-volume archives cannot be
7262 The following table summarizes compression options used by @GNUTAR{}.
7270 Filter the archive through @command{gzip}.
7272 You can use @option{--gzip} and @option{--gunzip} on physical devices
7273 (tape drives, etc.) and remote files as well as on normal files; data
7274 to or from such devices or remote files is reblocked by another copy
7275 of the @command{tar} program to enforce the specified (or default) record
7276 size. The default compression parameters are used; if you need to
7277 override them, set @env{GZIP} environment variable, e.g.:
7280 $ @kbd{GZIP=--best tar cfz archive.tar.gz subdir}
7284 Another way would be to avoid the @option{--gzip} (@option{--gunzip}, @option{--ungzip}, @option{-z}) option and run
7285 @command{gzip} explicitly:
7288 $ @kbd{tar cf - subdir | gzip --best -c - > archive.tar.gz}
7291 @cindex corrupted archives
7292 About corrupted compressed archives: @command{gzip}'ed files have no
7293 redundancy, for maximum compression. The adaptive nature of the
7294 compression scheme means that the compression tables are implicitly
7295 spread all over the archive. If you lose a few blocks, the dynamic
7296 construction of the compression tables becomes unsynchronized, and there
7297 is little chance that you could recover later in the archive.
7299 There are pending suggestions for having a per-volume or per-file
7300 compression in @GNUTAR{}. This would allow for viewing the
7301 contents without decompression, and for resynchronizing decompression at
7302 every volume or file, in case of corrupted archives. Doing so, we might
7303 lose some compressibility. But this would have make recovering easier.
7304 So, there are pros and cons. We'll see!
7309 Filter the archive through @code{bzip2}. Otherwise like @option{--gzip}.
7316 Filter the archive through @command{compress}. Otherwise like @option{--gzip}.
7318 The @acronym{GNU} Project recommends you not use
7319 @command{compress}, because there is a patent covering the algorithm it
7320 uses. You could be sued for patent infringement merely by running
7323 @opindex use-compress-program
7324 @item --use-compress-program=@var{prog}
7325 Use external compression program @var{prog}. Use this option if you
7326 have a compression program that @GNUTAR{} does not support. There
7327 are two requirements to which @var{prog} should comply:
7329 First, when called without options, it should read data from standard
7330 input, compress it and output it on standard output.
7332 Secondly, if called with @option{-d} argument, it should do exactly
7333 the opposite, i.e., read the compressed data from the standard input
7334 and produce uncompressed data on the standard output.
7337 @FIXME{I have one question, or maybe it's a suggestion if there isn't a way
7338 to do it now. I would like to use @option{--gzip}, but I'd also like
7339 the output to be fed through a program like @acronym{GNU}
7340 @command{ecc} (actually, right now that's @samp{exactly} what I'd like
7341 to use :-)), basically adding ECC protection on top of compression.
7342 It seems as if this should be quite easy to do, but I can't work out
7343 exactly how to go about it. Of course, I can pipe the standard output
7344 of @command{tar} through @command{ecc}, but then I lose (though I
7345 haven't started using it yet, I confess) the ability to have
7346 @command{tar} use @command{rmt} for it's I/O (I think).
7348 I think the most straightforward thing would be to let me specify a
7349 general set of filters outboard of compression (preferably ordered,
7350 so the order can be automatically reversed on input operations, and
7351 with the options they require specifiable), but beggars shouldn't be
7352 choosers and anything you decide on would be fine with me.
7354 By the way, I like @command{ecc} but if (as the comments say) it can't
7355 deal with loss of block sync, I'm tempted to throw some time at adding
7356 that capability. Supposing I were to actually do such a thing and
7357 get it (apparently) working, do you accept contributed changes to
7358 utilities like that? (Leigh Clayton @file{loc@@soliton.com}, May 1995).
7360 Isn't that exactly the role of the @option{--use-compress-prog=@var{program}} option?
7361 I never tried it myself, but I suspect you may want to write a
7362 @var{prog} script or program able to filter stdin to stdout to
7363 way you want. It should recognize the @option{-d} option, for when
7364 extraction is needed rather than creation.
7366 It has been reported that if one writes compressed data (through the
7367 @option{--gzip} or @option{--compress} options) to a DLT and tries to use
7368 the DLT compression mode, the data will actually get bigger and one will
7369 end up with less space on the tape.}
7372 @subsection Archiving Sparse Files
7373 @cindex Sparse Files
7380 Handle sparse files efficiently.
7383 This option causes all files to be put in the archive to be tested for
7384 sparseness, and handled specially if they are. The @option{--sparse}
7385 (@option{-S}) option is useful when many @code{dbm} files, for example, are being
7386 backed up. Using this option dramatically decreases the amount of
7387 space needed to store such a file.
7389 In later versions, this option may be removed, and the testing and
7390 treatment of sparse files may be done automatically with any special
7391 @acronym{GNU} options. For now, it is an option needing to be specified on
7392 the command line with the creation or updating of an archive.
7394 Files in the file system occasionally have ``holes.'' A hole in a file
7395 is a section of the file's contents which was never written. The
7396 contents of a hole read as all zeros. On many operating systems,
7397 actual disk storage is not allocated for holes, but they are counted
7398 in the length of the file. If you archive such a file, @command{tar}
7399 could create an archive longer than the original. To have @command{tar}
7400 attempt to recognize the holes in a file, use @option{--sparse} (@option{-S}). When
7401 you use this option, then, for any file using less disk space than
7402 would be expected from its length, @command{tar} searches the file for
7403 consecutive stretches of zeros. It then records in the archive for
7404 the file where the consecutive stretches of zeros are, and only
7405 archives the ``real contents'' of the file. On extraction (using
7406 @option{--sparse} is not needed on extraction) any such
7407 files have holes created wherever the continuous stretches of zeros
7408 were found. Thus, if you use @option{--sparse}, @command{tar} archives
7409 won't take more space than the original.
7411 A file is sparse if it contains blocks of zeros whose existence is
7412 recorded, but that have no space allocated on disk. When you specify
7413 the @option{--sparse} option in conjunction with the @option{--create}
7414 (@option{-c}) operation, @command{tar} tests all files for sparseness
7415 while archiving. If @command{tar} finds a file to be sparse, it uses a
7416 sparse representation of the file in the archive. @xref{create}, for
7417 more information about creating archives.
7419 @option{--sparse} is useful when archiving files, such as dbm files,
7420 likely to contain many nulls. This option dramatically
7421 decreases the amount of space needed to store such an archive.
7424 @strong{Please Note:} Always use @option{--sparse} when performing file
7425 system backups, to avoid archiving the expanded forms of files stored
7426 sparsely in the system.
7428 Even if your system has no sparse files currently, some may be
7429 created in the future. If you use @option{--sparse} while making file
7430 system backups as a matter of course, you can be assured the archive
7431 will never take more space on the media than the files take on disk
7432 (otherwise, archiving a disk filled with sparse files might take
7433 hundreds of tapes). @FIXME-xref{incremental when node name is set.}
7436 @command{tar} ignores the @option{--sparse} option when reading an archive.
7441 Files stored sparsely in the file system are represented sparsely in
7442 the archive. Use in conjunction with write operations.
7445 However, users should be well aware that at archive creation time,
7446 @GNUTAR{} still has to read whole disk file to
7447 locate the @dfn{holes}, and so, even if sparse files use little space
7448 on disk and in the archive, they may sometimes require inordinate
7449 amount of time for reading and examining all-zero blocks of a file.
7450 Although it works, it's painfully slow for a large (sparse) file, even
7451 though the resulting tar archive may be small. (One user reports that
7452 dumping a @file{core} file of over 400 megabytes, but with only about
7453 3 megabytes of actual data, took about 9 minutes on a Sun Sparcstation
7454 ELC, with full CPU utilization.)
7456 This reading is required in all cases and is not related to the fact
7457 the @option{--sparse} option is used or not, so by merely @emph{not}
7458 using the option, you are not saving time@footnote{Well! We should say
7459 the whole truth, here. When @option{--sparse} is selected while creating
7460 an archive, the current @command{tar} algorithm requires sparse files to be
7461 read twice, not once. We hope to develop a new archive format for saving
7462 sparse files in which one pass will be sufficient.}.
7464 Programs like @command{dump} do not have to read the entire file; by
7465 examining the file system directly, they can determine in advance
7466 exactly where the holes are and thus avoid reading through them. The
7467 only data it need read are the actual allocated data blocks.
7468 @GNUTAR{} uses a more portable and straightforward
7469 archiving approach, it would be fairly difficult that it does
7470 otherwise. Elizabeth Zwicky writes to @file{comp.unix.internals}, on
7474 What I did say is that you cannot tell the difference between a hole and an
7475 equivalent number of nulls without reading raw blocks. @code{st_blocks} at
7476 best tells you how many holes there are; it doesn't tell you @emph{where}.
7477 Just as programs may, conceivably, care what @code{st_blocks} is (care
7478 to name one that does?), they may also care where the holes are (I have
7479 no examples of this one either, but it's equally imaginable).
7481 I conclude from this that good archivers are not portable. One can
7482 arguably conclude that if you want a portable program, you can in good
7483 conscience restore files with as many holes as possible, since you can't
7488 @section Handling File Attributes
7491 When @command{tar} reads files, it updates their access times. To
7492 avoid this, use the @option{--atime-preserve[=METHOD]} option, which can either
7493 reset the access time retroactively or avoid changing it in the first
7496 Handling of file attributes
7499 @opindex atime-preserve
7500 @item --atime-preserve
7501 @itemx --atime-preserve=replace
7502 @itemx --atime-preserve=system
7503 Preserve the access times of files that are read. This works only for
7504 files that you own, unless you have superuser privileges.
7506 @option{--atime-preserve=replace} works on most systems, but it also
7507 restores the data modification time and updates the status change
7508 time. Hence it doesn't interact with incremental dumps nicely
7509 (@pxref{Backups}), and it can set access or data modification times
7510 incorrectly if other programs access the file while @command{tar} is
7513 @option{--atime-preserve=system} avoids changing the access time in
7514 the first place, if the operating system supports this.
7515 Unfortunately, this may or may not work on any given operating system
7516 or file system. If @command{tar} knows for sure it won't work, it
7517 complains right away.
7519 Currently @option{--atime-preserve} with no operand defaults to
7520 @option{--atime-preserve=replace}, but this is intended to change to
7521 @option{--atime-preserve=system} when the latter is better-supported.
7526 Do not extract data modification time.
7528 When this option is used, @command{tar} leaves the data modification times
7529 of the files it extracts as the times when the files were extracted,
7530 instead of setting it to the times recorded in the archive.
7532 This option is meaningless with @option{--list} (@option{-t}).
7536 Create extracted files with the same ownership they have in the
7539 This is the default behavior for the superuser,
7540 so this option is meaningful only for non-root users, when @command{tar}
7541 is executed on those systems able to give files away. This is
7542 considered as a security flaw by many people, at least because it
7543 makes quite difficult to correctly account users for the disk space
7544 they occupy. Also, the @code{suid} or @code{sgid} attributes of
7545 files are easily and silently lost when files are given away.
7547 When writing an archive, @command{tar} writes the user id and user name
7548 separately. If it can't find a user name (because the user id is not
7549 in @file{/etc/passwd}), then it does not write one. When restoring,
7550 and doing a @code{chmod} like when you use @option{--same-permissions},
7551 @FIXME{same-owner?}it tries to look the name (if one was written)
7552 up in @file{/etc/passwd}. If it fails, then it uses the user id
7553 stored in the archive instead.
7555 @opindex no-same-owner
7556 @item --no-same-owner
7558 Do not attempt to restore ownership when extracting. This is the
7559 default behavior for ordinary users, so this option has an effect
7560 only for the superuser.
7562 @opindex numeric-owner
7563 @item --numeric-owner
7564 The @option{--numeric-owner} option allows (ANSI) archives to be written
7565 without user/group name information or such information to be ignored
7566 when extracting. It effectively disables the generation and/or use
7567 of user/group name information. This option forces extraction using
7568 the numeric ids from the archive, ignoring the names.
7570 This is useful in certain circumstances, when restoring a backup from
7571 an emergency floppy with different passwd/group files for example.
7572 It is otherwise impossible to extract files with the right ownerships
7573 if the password file in use during the extraction does not match the
7574 one belonging to the file system(s) being extracted. This occurs,
7575 for example, if you are restoring your files after a major crash and
7576 had booted from an emergency floppy with no password file or put your
7577 disk into another machine to do the restore.
7579 The numeric ids are @emph{always} saved into @command{tar} archives.
7580 The identifying names are added at create time when provided by the
7581 system, unless @option{--old-archive} (@option{-o}) is used. Numeric ids could be
7582 used when moving archives between a collection of machines using
7583 a centralized management for attribution of numeric ids to users
7584 and groups. This is often made through using the NIS capabilities.
7586 When making a @command{tar} file for distribution to other sites, it
7587 is sometimes cleaner to use a single owner for all files in the
7588 distribution, and nicer to specify the write permission bits of the
7589 files as stored in the archive independently of their actual value on
7590 the file system. The way to prepare a clean distribution is usually
7591 to have some Makefile rule creating a directory, copying all needed
7592 files in that directory, then setting ownership and permissions as
7593 wanted (there are a lot of possible schemes), and only then making a
7594 @command{tar} archive out of this directory, before cleaning
7595 everything out. Of course, we could add a lot of options to
7596 @GNUTAR{} for fine tuning permissions and ownership.
7597 This is not the good way, I think. @GNUTAR{} is
7598 already crowded with options and moreover, the approach just explained
7599 gives you a great deal of control already.
7601 @opindex same-permissions, short description
7602 @opindex preserve-permissions, short description
7604 @itemx --same-permissions
7605 @itemx --preserve-permissions
7606 Extract all protection information.
7608 This option causes @command{tar} to set the modes (access permissions) of
7609 extracted files exactly as recorded in the archive. If this option
7610 is not used, the current @code{umask} setting limits the permissions
7611 on extracted files. This option is by default enabled when
7612 @command{tar} is executed by a superuser.
7615 This option is meaningless with @option{--list} (@option{-t}).
7619 Same as both @option{--same-permissions} and @option{--same-order}.
7621 The @option{--preserve} option has no equivalent short option name.
7622 It is equivalent to @option{--same-permissions} plus @option{--same-order}.
7624 @FIXME{I do not see the purpose of such an option. (Neither I. FP.)}
7629 @section Basic Tar Format
7632 While an archive may contain many files, the archive itself is a
7633 single ordinary file. Like any other file, an archive file can be
7634 written to a storage device such as a tape or disk, sent through a
7635 pipe or over a network, saved on the active file system, or even
7636 stored in another archive. An archive file is not easy to read or
7637 manipulate without using the @command{tar} utility or Tar mode in
7638 @acronym{GNU} Emacs.
7640 Physically, an archive consists of a series of file entries terminated
7641 by an end-of-archive entry, which consists of two 512 blocks of zero
7643 entry usually describes one of the files in the archive (an
7644 @dfn{archive member}), and consists of a file header and the contents
7645 of the file. File headers contain file names and statistics, checksum
7646 information which @command{tar} uses to detect file corruption, and
7647 information about file types.
7649 Archives are permitted to have more than one member with the same
7650 member name. One way this situation can occur is if more than one
7651 version of a file has been stored in the archive. For information
7652 about adding new versions of a file to an archive, see @ref{update}.
7653 @FIXME-xref{To learn more about having more than one archive member with the
7654 same name, see -backup node, when it's written.}
7656 In addition to entries describing archive members, an archive may
7657 contain entries which @command{tar} itself uses to store information.
7658 @xref{label}, for an example of such an archive entry.
7660 A @command{tar} archive file contains a series of blocks. Each block
7661 contains @code{BLOCKSIZE} bytes. Although this format may be thought
7662 of as being on magnetic tape, other media are often used.
7664 Each file archived is represented by a header block which describes
7665 the file, followed by zero or more blocks which give the contents
7666 of the file. At the end of the archive file there are two 512-byte blocks
7667 filled with binary zeros as an end-of-file marker. A reasonable system
7668 should write such end-of-file marker at the end of an archive, but
7669 must not assume that such a block exists when reading an archive. In
7670 particular @GNUTAR{} always issues a warning if it does not encounter it.
7672 The blocks may be @dfn{blocked} for physical I/O operations.
7673 Each record of @var{n} blocks (where @var{n} is set by the
7674 @option{--blocking-factor=@var{512-size}} (@option{-b @var{512-size}}) option to @command{tar}) is written with a single
7675 @w{@samp{write ()}} operation. On magnetic tapes, the result of
7676 such a write is a single record. When writing an archive,
7677 the last record of blocks should be written at the full size, with
7678 blocks after the zero block containing all zeros. When reading
7679 an archive, a reasonable system should properly handle an archive
7680 whose last record is shorter than the rest, or which contains garbage
7681 records after a zero block.
7683 The header block is defined in C as follows. In the @GNUTAR{}
7684 distribution, this is part of file @file{src/tar.h}:
7687 @include header.texi
7690 All characters in header blocks are represented by using 8-bit
7691 characters in the local variant of ASCII. Each field within the
7692 structure is contiguous; that is, there is no padding used within
7693 the structure. Each character on the archive medium is stored
7696 Bytes representing the contents of files (after the header block
7697 of each file) are not translated in any way and are not constrained
7698 to represent characters in any character set. The @command{tar} format
7699 does not distinguish text files from binary files, and no translation
7700 of file contents is performed.
7702 The @code{name}, @code{linkname}, @code{magic}, @code{uname}, and
7703 @code{gname} are null-terminated character strings. All other fields
7704 are zero-filled octal numbers in ASCII. Each numeric field of width
7705 @var{w} contains @var{w} minus 1 digits, and a null.
7707 The @code{name} field is the file name of the file, with directory names
7708 (if any) preceding the file name, separated by slashes.
7710 @FIXME{how big a name before field overflows?}
7712 The @code{mode} field provides nine bits specifying file permissions
7713 and three bits to specify the Set UID, Set GID, and Save Text
7714 (@dfn{sticky}) modes. Values for these bits are defined above.
7715 When special permissions are required to create a file with a given
7716 mode, and the user restoring files from the archive does not hold such
7717 permissions, the mode bit(s) specifying those special permissions
7718 are ignored. Modes which are not supported by the operating system
7719 restoring files from the archive will be ignored. Unsupported modes
7720 should be faked up when creating or updating an archive; e.g., the
7721 group permission could be copied from the @emph{other} permission.
7723 The @code{uid} and @code{gid} fields are the numeric user and group
7724 ID of the file owners, respectively. If the operating system does
7725 not support numeric user or group IDs, these fields should be ignored.
7727 The @code{size} field is the size of the file in bytes; linked files
7728 are archived with this field specified as zero. @FIXME-xref{Modifiers, in
7729 particular the @option{--incremental} (@option{-G}) option.}
7731 The @code{mtime} field is the data modification time of the file at
7732 the time it was archived. It is the ASCII representation of the octal
7733 value of the last time the file's contents were modified, represented
7734 as an integer number of
7735 seconds since January 1, 1970, 00:00 Coordinated Universal Time.
7737 The @code{chksum} field is the ASCII representation of the octal value
7738 of the simple sum of all bytes in the header block. Each 8-bit
7739 byte in the header is added to an unsigned integer, initialized to
7740 zero, the precision of which shall be no less than seventeen bits.
7741 When calculating the checksum, the @code{chksum} field is treated as
7742 if it were all blanks.
7744 The @code{typeflag} field specifies the type of file archived. If a
7745 particular implementation does not recognize or permit the specified
7746 type, the file will be extracted as if it were a regular file. As this
7747 action occurs, @command{tar} issues a warning to the standard error.
7749 The @code{atime} and @code{ctime} fields are used in making incremental
7750 backups; they store, respectively, the particular file's access and
7751 status change times.
7753 The @code{offset} is used by the @option{--multi-volume} (@option{-M}) option, when
7754 making a multi-volume archive. The offset is number of bytes into
7755 the file that we need to restart at to continue the file on the next
7756 tape, i.e., where we store the location that a continued file is
7759 The following fields were added to deal with sparse files. A file
7760 is @dfn{sparse} if it takes in unallocated blocks which end up being
7761 represented as zeros, i.e., no useful data. A test to see if a file
7762 is sparse is to look at the number blocks allocated for it versus the
7763 number of characters in the file; if there are fewer blocks allocated
7764 for the file than would normally be allocated for a file of that
7765 size, then the file is sparse. This is the method @command{tar} uses to
7766 detect a sparse file, and once such a file is detected, it is treated
7767 differently from non-sparse files.
7769 Sparse files are often @code{dbm} files, or other database-type files
7770 which have data at some points and emptiness in the greater part of
7771 the file. Such files can appear to be very large when an @samp{ls
7772 -l} is done on them, when in truth, there may be a very small amount
7773 of important data contained in the file. It is thus undesirable
7774 to have @command{tar} think that it must back up this entire file, as
7775 great quantities of room are wasted on empty blocks, which can lead
7776 to running out of room on a tape far earlier than is necessary.
7777 Thus, sparse files are dealt with so that these empty blocks are
7778 not written to the tape. Instead, what is written to the tape is a
7779 description, of sorts, of the sparse file: where the holes are, how
7780 big the holes are, and how much data is found at the end of the hole.
7781 This way, the file takes up potentially far less room on the tape,
7782 and when the file is extracted later on, it will look exactly the way
7783 it looked beforehand. The following is a description of the fields
7784 used to handle a sparse file:
7786 The @code{sp} is an array of @code{struct sparse}. Each @code{struct
7787 sparse} contains two 12-character strings which represent an offset
7788 into the file and a number of bytes to be written at that offset.
7789 The offset is absolute, and not relative to the offset in preceding
7792 The header can hold four of these @code{struct sparse} at the moment;
7793 if more are needed, they are not stored in the header.
7795 The @code{isextended} flag is set when an @code{extended_header}
7796 is needed to deal with a file. Note that this means that this flag
7797 can only be set when dealing with a sparse file, and it is only set
7798 in the event that the description of the file will not fit in the
7799 allotted room for sparse structures in the header. In other words,
7800 an extended_header is needed.
7802 The @code{extended_header} structure is used for sparse files which
7803 need more sparse structures than can fit in the header. The header can
7804 fit 4 such structures; if more are needed, the flag @code{isextended}
7805 gets set and the next block is an @code{extended_header}.
7807 Each @code{extended_header} structure contains an array of 21
7808 sparse structures, along with a similar @code{isextended} flag
7809 that the header had. There can be an indeterminate number of such
7810 @code{extended_header}s to describe a sparse file.
7814 @item @code{REGTYPE}
7815 @itemx @code{AREGTYPE}
7816 These flags represent a regular file. In order to be compatible
7817 with older versions of @command{tar}, a @code{typeflag} value of
7818 @code{AREGTYPE} should be silently recognized as a regular file.
7819 New archives should be created using @code{REGTYPE}. Also, for
7820 backward compatibility, @command{tar} treats a regular file whose name
7821 ends with a slash as a directory.
7823 @item @code{LNKTYPE}
7824 This flag represents a file linked to another file, of any type,
7825 previously archived. Such files are identified in Unix by each
7826 file having the same device and inode number. The linked-to name is
7827 specified in the @code{linkname} field with a trailing null.
7829 @item @code{SYMTYPE}
7830 This represents a symbolic link to another file. The linked-to name
7831 is specified in the @code{linkname} field with a trailing null.
7833 @item @code{CHRTYPE}
7834 @itemx @code{BLKTYPE}
7835 These represent character special files and block special files
7836 respectively. In this case the @code{devmajor} and @code{devminor}
7837 fields will contain the major and minor device numbers respectively.
7838 Operating systems may map the device specifications to their own
7839 local specification, or may ignore the entry.
7841 @item @code{DIRTYPE}
7842 This flag specifies a directory or sub-directory. The directory
7843 name in the @code{name} field should end with a slash. On systems where
7844 disk allocation is performed on a directory basis, the @code{size} field
7845 will contain the maximum number of bytes (which may be rounded to
7846 the nearest disk block allocation unit) which the directory may
7847 hold. A @code{size} field of zero indicates no such limiting. Systems
7848 which do not support limiting in this manner should ignore the
7851 @item @code{FIFOTYPE}
7852 This specifies a FIFO special file. Note that the archiving of a
7853 FIFO file archives the existence of this file and not its contents.
7855 @item @code{CONTTYPE}
7856 This specifies a contiguous file, which is the same as a normal
7857 file except that, in operating systems which support it, all its
7858 space is allocated contiguously on the disk. Operating systems
7859 which do not allow contiguous allocation should silently treat this
7860 type as a normal file.
7862 @item @code{A} @dots{} @code{Z}
7863 These are reserved for custom implementations. Some of these are
7864 used in the @acronym{GNU} modified format, as described below.
7868 Other values are reserved for specification in future revisions of
7869 the P1003 standard, and should not be used by any @command{tar} program.
7871 The @code{magic} field indicates that this archive was output in
7872 the P1003 archive format. If this field contains @code{TMAGIC},
7873 the @code{uname} and @code{gname} fields will contain the ASCII
7874 representation of the owner and group of the file respectively.
7875 If found, the user and group IDs are used rather than the values in
7876 the @code{uid} and @code{gid} fields.
7878 For references, see ISO/IEC 9945-1:1990 or IEEE Std 1003.1-1990, pages
7879 169-173 (section 10.1) for @cite{Archive/Interchange File Format}; and
7880 IEEE Std 1003.2-1992, pages 380-388 (section 4.48) and pages 936-940
7881 (section E.4.48) for @cite{pax - Portable archive interchange}.
7884 @section @acronym{GNU} Extensions to the Archive Format
7887 The @acronym{GNU} format uses additional file types to describe new types of
7888 files in an archive. These are listed below.
7891 @item GNUTYPE_DUMPDIR
7893 This represents a directory and a list of files created by the
7894 @option{--incremental} (@option{-G}) option. The @code{size} field gives the total
7895 size of the associated list of files. Each file name is preceded by
7896 either a @samp{Y} (the file should be in this archive) or an @samp{N}.
7897 (The file is a directory, or is not stored in the archive.) Each file
7898 name is terminated by a null. There is an additional null after the
7901 @item GNUTYPE_MULTIVOL
7903 This represents a file continued from another volume of a multi-volume
7904 archive created with the @option{--multi-volume} (@option{-M}) option. The original
7905 type of the file is not given here. The @code{size} field gives the
7906 maximum size of this piece of the file (assuming the volume does
7907 not end before the file is written out). The @code{offset} field
7908 gives the offset from the beginning of the file where this part of
7909 the file begins. Thus @code{size} plus @code{offset} should equal
7910 the original size of the file.
7912 @item GNUTYPE_SPARSE
7914 This flag indicates that we are dealing with a sparse file. Note
7915 that archiving a sparse file requires special operations to find
7916 holes in the file, which mark the positions of these holes, along
7917 with the number of bytes of data to be found after the hole.
7919 @item GNUTYPE_VOLHDR
7921 This file type is used to mark the volume header that was given with
7922 the @option{--label=@var{archive-label}} (@option{-V @var{archive-label}}) option when the archive was created. The @code{name}
7923 field contains the @code{name} given after the @option{--label=@var{archive-label}} (@option{-V @var{archive-label}}) option.
7924 The @code{size} field is zero. Only the first file in each volume
7925 of an archive should have this type.
7929 You may have trouble reading a @acronym{GNU} format archive on a
7930 non-@acronym{GNU} system if the options @option{--incremental} (@option{-G}),
7931 @option{--multi-volume} (@option{-M}), @option{--sparse} (@option{-S}), or @option{--label=@var{archive-label}} (@option{-V @var{archive-label}}) were
7932 used when writing the archive. In general, if @command{tar} does not
7933 use the @acronym{GNU}-added fields of the header, other versions of
7934 @command{tar} should be able to read the archive. Otherwise, the
7935 @command{tar} program will give an error, the most likely one being a
7939 @section Comparison of @command{tar} and @command{cpio}
7942 @FIXME{Reorganize the following material}
7944 The @command{cpio} archive formats, like @command{tar}, do have maximum
7945 pathname lengths. The binary and old ASCII formats have a max path
7946 length of 256, and the new ASCII and CRC ASCII formats have a max
7947 path length of 1024. @acronym{GNU} @command{cpio} can read and write archives
7948 with arbitrary pathname lengths, but other @command{cpio} implementations
7949 may crash unexplainedly trying to read them.
7951 @command{tar} handles symbolic links in the form in which it comes in BSD;
7952 @command{cpio} doesn't handle symbolic links in the form in which it comes
7953 in System V prior to SVR4, and some vendors may have added symlinks
7954 to their system without enhancing @command{cpio} to know about them.
7955 Others may have enhanced it in a way other than the way I did it
7956 at Sun, and which was adopted by AT&T (and which is, I think, also
7957 present in the @command{cpio} that Berkeley picked up from AT&T and put
7958 into a later BSD release---I think I gave them my changes).
7960 (SVR4 does some funny stuff with @command{tar}; basically, its @command{cpio}
7961 can handle @command{tar} format input, and write it on output, and it
7962 probably handles symbolic links. They may not have bothered doing
7963 anything to enhance @command{tar} as a result.)
7965 @command{cpio} handles special files; traditional @command{tar} doesn't.
7967 @command{tar} comes with V7, System III, System V, and BSD source;
7968 @command{cpio} comes only with System III, System V, and later BSD
7969 (4.3-tahoe and later).
7971 @command{tar}'s way of handling multiple hard links to a file can handle
7972 file systems that support 32-bit inumbers (e.g., the BSD file system);
7973 @command{cpio}s way requires you to play some games (in its "binary"
7974 format, i-numbers are only 16 bits, and in its "portable ASCII" format,
7975 they're 18 bits---it would have to play games with the "file system ID"
7976 field of the header to make sure that the file system ID/i-number pairs
7977 of different files were always different), and I don't know which
7978 @command{cpio}s, if any, play those games. Those that don't might get
7979 confused and think two files are the same file when they're not, and
7980 make hard links between them.
7982 @command{tar}s way of handling multiple hard links to a file places only
7983 one copy of the link on the tape, but the name attached to that copy
7984 is the @emph{only} one you can use to retrieve the file; @command{cpio}s
7985 way puts one copy for every link, but you can retrieve it using any
7989 What type of check sum (if any) is used, and how is this calculated.
7992 See the attached manual pages for @command{tar} and @command{cpio} format.
7993 @command{tar} uses a checksum which is the sum of all the bytes in the
7994 @command{tar} header for a file; @command{cpio} uses no checksum.
7997 If anyone knows why @command{cpio} was made when @command{tar} was present
8001 It wasn't. @command{cpio} first showed up in PWB/UNIX 1.0; no
8002 generally-available version of UNIX had @command{tar} at the time. I don't
8003 know whether any version that was generally available @emph{within AT&T}
8004 had @command{tar}, or, if so, whether the people within AT&T who did
8005 @command{cpio} knew about it.
8007 On restore, if there is a corruption on a tape @command{tar} will stop at
8008 that point, while @command{cpio} will skip over it and try to restore the
8011 The main difference is just in the command syntax and header format.
8013 @command{tar} is a little more tape-oriented in that everything is blocked
8014 to start on a record boundary.
8017 Is there any differences between the ability to recover crashed
8018 archives between the two of them. (Is there any chance of recovering
8019 crashed archives at all.)
8022 Theoretically it should be easier under @command{tar} since the blocking
8023 lets you find a header with some variation of @samp{dd skip=@var{nn}}.
8024 However, modern @command{cpio}'s and variations have an option to just
8025 search for the next file header after an error with a reasonable chance
8026 of resyncing. However, lots of tape driver software won't allow you to
8027 continue past a media error which should be the only reason for getting
8028 out of sync unless a file changed sizes while you were writing the
8032 If anyone knows why @command{cpio} was made when @command{tar} was present
8033 at the unix scene, please tell me about this too.
8036 Probably because it is more media efficient (by not blocking everything
8037 and using only the space needed for the headers where @command{tar}
8038 always uses 512 bytes per file header) and it knows how to archive
8041 You might want to look at the freely available alternatives. The
8042 major ones are @command{afio}, @GNUTAR{}, and
8043 @command{pax}, each of which have their own extensions with some
8044 backwards compatibility.
8046 Sparse files were @command{tar}red as sparse files (which you can
8047 easily test, because the resulting archive gets smaller, and
8048 @acronym{GNU} @command{cpio} can no longer read it).
8051 @chapter Tapes and Other Archive Media
8054 A few special cases about tape handling warrant more detailed
8055 description. These special cases are discussed below.
8057 Many complexities surround the use of @command{tar} on tape drives. Since
8058 the creation and manipulation of archives located on magnetic tape was
8059 the original purpose of @command{tar}, it contains many features making
8060 such manipulation easier.
8062 Archives are usually written on dismountable media---tape cartridges,
8063 mag tapes, or floppy disks.
8065 The amount of data a tape or disk holds depends not only on its size,
8066 but also on how it is formatted. A 2400 foot long reel of mag tape
8067 holds 40 megabytes of data when formatted at 1600 bits per inch. The
8068 physically smaller EXABYTE tape cartridge holds 2.3 gigabytes.
8070 Magnetic media are re-usable---once the archive on a tape is no longer
8071 needed, the archive can be erased and the tape or disk used over.
8072 Media quality does deteriorate with use, however. Most tapes or disks
8073 should be discarded when they begin to produce data errors. EXABYTE
8074 tape cartridges should be discarded when they generate an @dfn{error
8075 count} (number of non-usable bits) of more than 10k.
8077 Magnetic media are written and erased using magnetic fields, and
8078 should be protected from such fields to avoid damage to stored data.
8079 Sticking a floppy disk to a filing cabinet using a magnet is probably
8083 * Device:: Device selection and switching
8084 * Remote Tape Server::
8085 * Common Problems and Solutions::
8086 * Blocking:: Blocking
8087 * Many:: Many archives on one tape
8088 * Using Multiple Tapes:: Using Multiple Tapes
8089 * label:: Including a Label in the Archive
8091 * Write Protection::
8095 @section Device Selection and Switching
8099 @item -f [@var{hostname}:]@var{file}
8100 @itemx --file=[@var{hostname}:]@var{file}
8101 Use archive file or device @var{file} on @var{hostname}.
8104 This option is used to specify the file name of the archive @command{tar}
8107 If the file name is @samp{-}, @command{tar} reads the archive from standard
8108 input (when listing or extracting), or writes it to standard output
8109 (when creating). If the @samp{-} file name is given when updating an
8110 archive, @command{tar} will read the original archive from its standard
8111 input, and will write the entire new archive to its standard output.
8113 If the file name contains a @samp{:}, it is interpreted as
8114 @samp{hostname:file name}. If the @var{hostname} contains an @dfn{at}
8115 sign (@samp{@@}), it is treated as @samp{user@@hostname:file name}. In
8116 either case, @command{tar} will invoke the command @command{rsh} (or
8117 @command{remsh}) to start up an @command{/usr/libexec/rmt} on the remote
8118 machine. If you give an alternate login name, it will be given to the
8120 Naturally, the remote machine must have an executable
8121 @command{/usr/libexec/rmt}. This program is free software from the
8122 University of California, and a copy of the source code can be found
8123 with the sources for @command{tar}; it's compiled and installed by default.
8124 The exact path to this utility is determined when configuring the package.
8125 It is @file{@var{prefix}/libexec/rmt}, where @var{prefix} stands for
8126 your installation prefix. This location may also be overridden at
8127 runtime by using @option{rmt-command=@var{command}} option (@xref{Option Summary,
8128 ---rmt-command}, for detailed description of this option. @xref{Remote
8129 Tape Server}, for the description of @command{rmt} command).
8131 If this option is not given, but the environment variable @env{TAPE}
8132 is set, its value is used; otherwise, old versions of @command{tar}
8133 used a default archive name (which was picked when @command{tar} was
8134 compiled). The default is normally set up to be the @dfn{first} tape
8135 drive or other transportable I/O medium on the system.
8137 Starting with version 1.11.5, @GNUTAR{} uses
8138 standard input and standard output as the default device, and I will
8139 not try anymore supporting automatic device detection at installation
8140 time. This was failing really in too many cases, it was hopeless.
8141 This is now completely left to the installer to override standard
8142 input and standard output for default device, if this seems
8143 preferable. Further, I think @emph{most} actual usages of
8144 @command{tar} are done with pipes or disks, not really tapes,
8145 cartridges or diskettes.
8147 Some users think that using standard input and output is running
8148 after trouble. This could lead to a nasty surprise on your screen if
8149 you forget to specify an output file name---especially if you are going
8150 through a network or terminal server capable of buffering large amounts
8151 of output. We had so many bug reports in that area of configuring
8152 default tapes automatically, and so many contradicting requests, that
8153 we finally consider the problem to be portably intractable. We could
8154 of course use something like @samp{/dev/tape} as a default, but this
8155 is @emph{also} running after various kind of trouble, going from hung
8156 processes to accidental destruction of real tapes. After having seen
8157 all this mess, using standard input and output as a default really
8158 sounds like the only clean choice left, and a very useful one too.
8160 @GNUTAR{} reads and writes archive in records, I
8161 suspect this is the main reason why block devices are preferred over
8162 character devices. Most probably, block devices are more efficient
8163 too. The installer could also check for @samp{DEFTAPE} in
8164 @file{<sys/mtio.h>}.
8167 @opindex force-local, short description
8169 Archive file is local even if it contains a colon.
8171 @opindex rsh-command
8172 @item --rsh-command=@var{command}
8173 Use remote @var{command} instead of @command{rsh}. This option exists
8174 so that people who use something other than the standard @command{rsh}
8175 (e.g., a Kerberized @command{rsh}) can access a remote device.
8177 When this command is not used, the shell command found when
8178 the @command{tar} program was installed is used instead. This is
8179 the first found of @file{/usr/ucb/rsh}, @file{/usr/bin/remsh},
8180 @file{/usr/bin/rsh}, @file{/usr/bsd/rsh} or @file{/usr/bin/nsh}.
8181 The installer may have overridden this by defining the environment
8182 variable @env{RSH} @emph{at installation time}.
8185 Specify drive and density.
8187 @opindex multi-volume, short description
8189 @itemx --multi-volume
8190 Create/list/extract multi-volume archive.
8192 This option causes @command{tar} to write a @dfn{multi-volume} archive---one
8193 that may be larger than will fit on the medium used to hold it.
8194 @xref{Multi-Volume Archives}.
8196 @opindex tape-length, short description
8198 @itemx --tape-length=@var{num}
8199 Change tape after writing @var{num} x 1024 bytes.
8201 This option might be useful when your tape drivers do not properly
8202 detect end of physical tapes. By being slightly conservative on the
8203 maximum tape length, you might avoid the problem entirely.
8205 @opindex info-script, short description
8206 @opindex new-volume-script, short description
8208 @itemx --info-script=@var{file}
8209 @itemx --new-volume-script=@var{file}
8210 Execute @file{file} at end of each tape. This implies
8211 @option{--multi-volume} (@option{-M}). @xref{info-script}, for a detailed
8212 description of this option.
8215 @node Remote Tape Server
8216 @section The Remote Tape Server
8218 @cindex remote tape drive
8220 In order to access the tape drive on a remote machine, @command{tar}
8221 uses the remote tape server written at the University of California at
8222 Berkeley. The remote tape server must be installed as
8223 @file{@var{prefix}/libexec/rmt} on any machine whose tape drive you
8224 want to use. @command{tar} calls @command{rmt} by running an
8225 @command{rsh} or @command{remsh} to the remote machine, optionally
8226 using a different login name if one is supplied.
8228 A copy of the source for the remote tape server is provided. It is
8229 Copyright @copyright{} 1983 by the Regents of the University of
8230 California, but can be freely distributed. It is compiled and
8231 installed by default.
8233 @cindex absolute file names
8234 Unless you use the @option{--absolute-names} (@option{-P}) option,
8235 @GNUTAR{} will not allow you to create an archive that contains
8236 absolute file names (a file name beginning with @samp{/}.) If you try,
8237 @command{tar} will automatically remove the leading @samp{/} from the
8238 file names it stores in the archive. It will also type a warning
8239 message telling you what it is doing.
8241 When reading an archive that was created with a different
8242 @command{tar} program, @GNUTAR{} automatically
8243 extracts entries in the archive which have absolute file names as if
8244 the file names were not absolute. This is an important feature. A
8245 visitor here once gave a @command{tar} tape to an operator to restore;
8246 the operator used Sun @command{tar} instead of @GNUTAR{},
8247 and the result was that it replaced large portions of
8248 our @file{/bin} and friends with versions from the tape; needless to
8249 say, we were unhappy about having to recover the file system from
8252 For example, if the archive contained a file @file{/usr/bin/computoy},
8253 @GNUTAR{} would extract the file to @file{usr/bin/computoy},
8254 relative to the current directory. If you want to extract the files in
8255 an archive to the same absolute names that they had when the archive
8256 was created, you should do a @samp{cd /} before extracting the files
8257 from the archive, or you should either use the @option{--absolute-names}
8258 option, or use the command @samp{tar -C / @dots{}}.
8260 @cindex Ultrix 3.1 and write failure
8261 Some versions of Unix (Ultrix 3.1 is known to have this problem),
8262 can claim that a short write near the end of a tape succeeded,
8263 when it actually failed. This will result in the -M option not
8264 working correctly. The best workaround at the moment is to use a
8265 significantly larger blocking factor than the default 20.
8267 In order to update an archive, @command{tar} must be able to backspace the
8268 archive in order to reread or rewrite a record that was just read (or
8269 written). This is currently possible only on two kinds of files: normal
8270 disk files (or any other file that can be backspaced with @samp{lseek}),
8271 and industry-standard 9-track magnetic tape (or any other kind of tape
8272 that can be backspaced with the @code{MTIOCTOP} @code{ioctl}.
8274 This means that the @option{--append}, @option{--concatenate}, and
8275 @option{--delete} commands will not work on any other kind of file.
8276 Some media simply cannot be backspaced, which means these commands and
8277 options will never be able to work on them. These non-backspacing
8278 media include pipes and cartridge tape drives.
8280 Some other media can be backspaced, and @command{tar} will work on them
8281 once @command{tar} is modified to do so.
8283 Archives created with the @option{--multi-volume}, @option{--label}, and
8284 @option{--incremental} (@option{-G}) options may not be readable by other version
8285 of @command{tar}. In particular, restoring a file that was split over
8286 a volume boundary will require some careful work with @command{dd}, if
8287 it can be done at all. Other versions of @command{tar} may also create
8288 an empty file whose name is that of the volume header. Some versions
8289 of @command{tar} may create normal files instead of directories archived
8290 with the @option{--incremental} (@option{-G}) option.
8292 @node Common Problems and Solutions
8293 @section Some Common Problems and their Solutions
8300 no such file or directory
8303 errors from @command{tar}:
8304 directory checksum error
8307 errors from media/system:
8318 @dfn{Block} and @dfn{record} terminology is rather confused, and it
8319 is also confusing to the expert reader. On the other hand, readers
8320 who are new to the field have a fresh mind, and they may safely skip
8321 the next two paragraphs, as the remainder of this manual uses those
8322 two terms in a quite consistent way.
8324 John Gilmore, the writer of the public domain @command{tar} from which
8325 @GNUTAR{} was originally derived, wrote (June 1995):
8328 The nomenclature of tape drives comes from IBM, where I believe
8329 they were invented for the IBM 650 or so. On IBM mainframes, what
8330 is recorded on tape are tape blocks. The logical organization of
8331 data is into records. There are various ways of putting records into
8332 blocks, including @code{F} (fixed sized records), @code{V} (variable
8333 sized records), @code{FB} (fixed blocked: fixed size records, @var{n}
8334 to a block), @code{VB} (variable size records, @var{n} to a block),
8335 @code{VSB} (variable spanned blocked: variable sized records that can
8336 occupy more than one block), etc. The @code{JCL} @samp{DD RECFORM=}
8337 parameter specified this to the operating system.
8339 The Unix man page on @command{tar} was totally confused about this.
8340 When I wrote @code{PD TAR}, I used the historically correct terminology
8341 (@command{tar} writes data records, which are grouped into blocks).
8342 It appears that the bogus terminology made it into @acronym{POSIX} (no surprise
8343 here), and now Fran@,{c}ois has migrated that terminology back
8344 into the source code too.
8347 The term @dfn{physical block} means the basic transfer chunk from or
8348 to a device, after which reading or writing may stop without anything
8349 being lost. In this manual, the term @dfn{block} usually refers to
8350 a disk physical block, @emph{assuming} that each disk block is 512
8351 bytes in length. It is true that some disk devices have different
8352 physical blocks, but @command{tar} ignore these differences in its own
8353 format, which is meant to be portable, so a @command{tar} block is always
8354 512 bytes in length, and @dfn{block} always mean a @command{tar} block.
8355 The term @dfn{logical block} often represents the basic chunk of
8356 allocation of many disk blocks as a single entity, which the operating
8357 system treats somewhat atomically; this concept is only barely used
8360 The term @dfn{physical record} is another way to speak of a physical
8361 block, those two terms are somewhat interchangeable. In this manual,
8362 the term @dfn{record} usually refers to a tape physical block,
8363 @emph{assuming} that the @command{tar} archive is kept on magnetic tape.
8364 It is true that archives may be put on disk or used with pipes,
8365 but nevertheless, @command{tar} tries to read and write the archive one
8366 @dfn{record} at a time, whatever the medium in use. One record is made
8367 up of an integral number of blocks, and this operation of putting many
8368 disk blocks into a single tape block is called @dfn{reblocking}, or
8369 more simply, @dfn{blocking}. The term @dfn{logical record} refers to
8370 the logical organization of many characters into something meaningful
8371 to the application. The term @dfn{unit record} describes a small set
8372 of characters which are transmitted whole to or by the application,
8373 and often refers to a line of text. Those two last terms are unrelated
8374 to what we call a @dfn{record} in @GNUTAR{}.
8376 When writing to tapes, @command{tar} writes the contents of the archive
8377 in chunks known as @dfn{records}. To change the default blocking
8378 factor, use the @option{--blocking-factor=@var{512-size}} (@option{-b
8379 @var{512-size}}) option. Each record will then be composed of
8380 @var{512-size} blocks. (Each @command{tar} block is 512 bytes.
8381 @xref{Standard}.) Each file written to the archive uses at least one
8382 full record. As a result, using a larger record size can result in
8383 more wasted space for small files. On the other hand, a larger record
8384 size can often be read and written much more efficiently.
8386 Further complicating the problem is that some tape drives ignore the
8387 blocking entirely. For these, a larger record size can still improve
8388 performance (because the software layers above the tape drive still
8389 honor the blocking), but not as dramatically as on tape drives that
8392 When reading an archive, @command{tar} can usually figure out the
8393 record size on itself. When this is the case, and a non-standard
8394 record size was used when the archive was created, @command{tar} will
8395 print a message about a non-standard blocking factor, and then operate
8396 normally. On some tape devices, however, @command{tar} cannot figure
8397 out the record size itself. On most of those, you can specify a
8398 blocking factor (with @option{--blocking-factor}) larger than the
8399 actual blocking factor, and then use the @option{--read-full-records}
8400 (@option{-B}) option. (If you specify a blocking factor with
8401 @option{--blocking-factor} and don't use the
8402 @option{--read-full-records} option, then @command{tar} will not
8403 attempt to figure out the recording size itself.) On some devices,
8404 you must always specify the record size exactly with
8405 @option{--blocking-factor} when reading, because @command{tar} cannot
8406 figure it out. In any case, use @option{--list} (@option{-t}) before
8407 doing any extractions to see whether @command{tar} is reading the archive
8410 @command{tar} blocks are all fixed size (512 bytes), and its scheme for
8411 putting them into records is to put a whole number of them (one or
8412 more) into each record. @command{tar} records are all the same size;
8413 at the end of the file there's a block containing all zeros, which
8414 is how you tell that the remainder of the last record(s) are garbage.
8416 In a standard @command{tar} file (no options), the block size is 512
8417 and the record size is 10240, for a blocking factor of 20. What the
8418 @option{--blocking-factor} option does is sets the blocking factor,
8419 changing the record size while leaving the block size at 512 bytes.
8420 20 was fine for ancient 800 or 1600 bpi reel-to-reel tape drives;
8421 most tape drives these days prefer much bigger records in order to
8422 stream and not waste tape. When writing tapes for myself, some tend
8423 to use a factor of the order of 2048, say, giving a record size of
8424 around one megabyte.
8426 If you use a blocking factor larger than 20, older @command{tar}
8427 programs might not be able to read the archive, so we recommend this
8428 as a limit to use in practice. @GNUTAR{}, however,
8429 will support arbitrarily large record sizes, limited only by the
8430 amount of virtual memory or the physical characteristics of the tape
8434 * Format Variations:: Format Variations
8435 * Blocking Factor:: The Blocking Factor of an Archive
8438 @node Format Variations
8439 @subsection Format Variations
8440 @cindex Format Parameters
8441 @cindex Format Options
8442 @cindex Options, archive format specifying
8443 @cindex Options, format specifying
8446 Format parameters specify how an archive is written on the archive
8447 media. The best choice of format parameters will vary depending on
8448 the type and number of files being archived, and on the media used to
8451 To specify format parameters when accessing or creating an archive,
8452 you can use the options described in the following sections.
8453 If you do not specify any format parameters, @command{tar} uses
8454 default parameters. You cannot modify a compressed archive.
8455 If you create an archive with the @option{--blocking-factor} option
8456 specified (@pxref{Blocking Factor}), you must specify that
8457 blocking-factor when operating on the archive. @xref{Formats}, for other
8458 examples of format parameter considerations.
8460 @node Blocking Factor
8461 @subsection The Blocking Factor of an Archive
8462 @cindex Blocking Factor
8464 @cindex Number of blocks per record
8465 @cindex Number of bytes per record
8466 @cindex Bytes per record
8467 @cindex Blocks per record
8470 @opindex blocking-factor
8471 The data in an archive is grouped into blocks, which are 512 bytes.
8472 Blocks are read and written in whole number multiples called
8473 @dfn{records}. The number of blocks in a record (ie. the size of a
8474 record in units of 512 bytes) is called the @dfn{blocking factor}.
8475 The @option{--blocking-factor=@var{512-size}} (@option{-b
8476 @var{512-size}}) option specifies the blocking factor of an archive.
8477 The default blocking factor is typically 20 (i.e., 10240 bytes), but
8478 can be specified at installation. To find out the blocking factor of
8479 an existing archive, use @samp{tar --list --file=@var{archive-name}}.
8480 This may not work on some devices.
8482 Records are separated by gaps, which waste space on the archive media.
8483 If you are archiving on magnetic tape, using a larger blocking factor
8484 (and therefore larger records) provides faster throughput and allows you
8485 to fit more data on a tape (because there are fewer gaps). If you are
8486 archiving on cartridge, a very large blocking factor (say 126 or more)
8487 greatly increases performance. A smaller blocking factor, on the other
8488 hand, may be useful when archiving small files, to avoid archiving lots
8489 of nulls as @command{tar} fills out the archive to the end of the record.
8490 In general, the ideal record size depends on the size of the
8491 inter-record gaps on the tape you are using, and the average size of the
8492 files you are archiving. @xref{create}, for information on
8495 @FIXME{Need example of using a cartridge with blocking factor=126 or more.}
8497 Archives with blocking factors larger than 20 cannot be read
8498 by very old versions of @command{tar}, or by some newer versions
8499 of @command{tar} running on old machines with small address spaces.
8500 With @GNUTAR{}, the blocking factor of an archive is limited
8501 only by the maximum record size of the device containing the archive,
8502 or by the amount of available virtual memory.
8504 Also, on some systems, not using adequate blocking factors, as sometimes
8505 imposed by the device drivers, may yield unexpected diagnostics. For
8506 example, this has been reported:
8509 Cannot write to /dev/dlt: Invalid argument
8513 In such cases, it sometimes happen that the @command{tar} bundled by
8514 the system is aware of block size idiosyncrasies, while @GNUTAR{}
8515 requires an explicit specification for the block size,
8516 which it cannot guess. This yields some people to consider
8517 @GNUTAR{} is misbehaving, because by comparison,
8518 @cite{the bundle @command{tar} works OK}. Adding @w{@kbd{-b 256}},
8519 for example, might resolve the problem.
8521 If you use a non-default blocking factor when you create an archive, you
8522 must specify the same blocking factor when you modify that archive. Some
8523 archive devices will also require you to specify the blocking factor when
8524 reading that archive, however this is not typically the case. Usually, you
8525 can use @option{--list} (@option{-t}) without specifying a blocking factor---@command{tar}
8526 reports a non-default record size and then lists the archive members as
8527 it would normally. To extract files from an archive with a non-standard
8528 blocking factor (particularly if you're not sure what the blocking factor
8529 is), you can usually use the @option{--read-full-records} (@option{-B}) option while
8530 specifying a blocking factor larger then the blocking factor of the archive
8531 (ie. @samp{tar --extract --read-full-records --blocking-factor=300}.
8532 @xref{list}, for more information on the @option{--list} (@option{-t})
8533 operation. @xref{Reading}, for a more detailed explanation of that option.
8536 @item --blocking-factor=@var{number}
8537 @itemx -b @var{number}
8538 Specifies the blocking factor of an archive. Can be used with any
8539 operation, but is usually not necessary with @option{--list} (@option{-t}).
8545 @item -b @var{blocks}
8546 @itemx --blocking-factor=@var{blocks}
8547 Set record size to @math{@var{blocks} * 512} bytes.
8549 This option is used to specify a @dfn{blocking factor} for the archive.
8550 When reading or writing the archive, @command{tar}, will do reads and writes
8551 of the archive in records of @math{@var{block}*512} bytes. This is true
8552 even when the archive is compressed. Some devices requires that all
8553 write operations be a multiple of a certain size, and so, @command{tar}
8554 pads the archive out to the next record boundary.
8556 The default blocking factor is set when @command{tar} is compiled, and is
8557 typically 20. Blocking factors larger than 20 cannot be read by very
8558 old versions of @command{tar}, or by some newer versions of @command{tar}
8559 running on old machines with small address spaces.
8561 With a magnetic tape, larger records give faster throughput and fit
8562 more data on a tape (because there are fewer inter-record gaps).
8563 If the archive is in a disk file or a pipe, you may want to specify
8564 a smaller blocking factor, since a large one will result in a large
8565 number of null bytes at the end of the archive.
8567 When writing cartridge or other streaming tapes, a much larger
8568 blocking factor (say 126 or more) will greatly increase performance.
8569 However, you must specify the same blocking factor when reading or
8570 updating the archive.
8572 Apparently, Exabyte drives have a physical block size of 8K bytes.
8573 If we choose our blocksize as a multiple of 8k bytes, then the problem
8574 seems to disappear. Id est, we are using block size of 112 right
8575 now, and we haven't had the problem since we switched@dots{}
8577 With @GNUTAR{} the blocking factor is limited only
8578 by the maximum record size of the device containing the archive, or by
8579 the amount of available virtual memory.
8581 However, deblocking or reblocking is virtually avoided in a special
8582 case which often occurs in practice, but which requires all the
8583 following conditions to be simultaneously true:
8586 the archive is subject to a compression option,
8588 the archive is not handled through standard input or output, nor
8589 redirected nor piped,
8591 the archive is directly handled to a local disk, instead of any special
8594 @option{--blocking-factor} is not explicitly specified on the @command{tar}
8598 If the output goes directly to a local disk, and not through
8599 stdout, then the last write is not extended to a full record size.
8600 Otherwise, reblocking occurs. Here are a few other remarks on this
8606 @command{gzip} will complain about trailing garbage if asked to
8607 uncompress a compressed archive on tape, there is an option to turn
8608 the message off, but it breaks the regularity of simply having to use
8609 @samp{@var{prog} -d} for decompression. It would be nice if gzip was
8610 silently ignoring any number of trailing zeros. I'll ask Jean-loup
8611 Gailly, by sending a copy of this message to him.
8614 @command{compress} does not show this problem, but as Jean-loup pointed
8615 out to Michael, @samp{compress -d} silently adds garbage after
8616 the result of decompression, which tar ignores because it already
8617 recognized its end-of-file indicator. So this bug may be safely
8621 @samp{gzip -d -q} will be silent about the trailing zeros indeed,
8622 but will still return an exit status of 2 which tar reports in turn.
8623 @command{tar} might ignore the exit status returned, but I hate doing
8624 that, as it weakens the protection @command{tar} offers users against
8625 other possible problems at decompression time. If @command{gzip} was
8626 silently skipping trailing zeros @emph{and} also avoiding setting the
8627 exit status in this innocuous case, that would solve this situation.
8630 @command{tar} should become more solid at not stopping to read a pipe at
8631 the first null block encountered. This inelegantly breaks the pipe.
8632 @command{tar} should rather drain the pipe out before exiting itself.
8635 @opindex ignore-zeros, short description
8637 @itemx --ignore-zeros
8638 Ignore blocks of zeros in archive (means EOF).
8640 The @option{--ignore-zeros} (@option{-i}) option causes @command{tar} to ignore blocks
8641 of zeros in the archive. Normally a block of zeros indicates the
8642 end of the archive, but when reading a damaged archive, or one which
8643 was created by concatenating several archives together, this option
8644 allows @command{tar} to read the entire archive. This option is not on
8645 by default because many versions of @command{tar} write garbage after
8648 Note that this option causes @command{tar} to read to the end of the
8649 archive file, which may sometimes avoid problems when multiple files
8650 are stored on a single physical tape.
8652 @opindex read-full-records, short description
8654 @itemx --read-full-records
8655 Reblock as we read (for reading 4.2BSD pipes).
8657 If @option{--read-full-records} is used, @command{tar}
8658 will not panic if an attempt to read a record from the archive does
8659 not return a full record. Instead, @command{tar} will keep reading
8660 until it has obtained a full
8663 This option is turned on by default when @command{tar} is reading
8664 an archive from standard input, or from a remote machine. This is
8665 because on BSD Unix systems, a read of a pipe will return however
8666 much happens to be in the pipe, even if it is less than @command{tar}
8667 requested. If this option was not used, @command{tar} would fail as
8668 soon as it read an incomplete record from the pipe.
8670 This option is also useful with the commands for updating an archive.
8676 @FIXME{Appropriate options should be moved here from elsewhere.}
8678 @cindex blocking factor
8679 @cindex tape blocking
8681 When handling various tapes or cartridges, you have to take care of
8682 selecting a proper blocking, that is, the number of disk blocks you
8683 put together as a single tape block on the tape, without intervening
8684 tape gaps. A @dfn{tape gap} is a small landing area on the tape
8685 with no information on it, used for decelerating the tape to a
8686 full stop, and for later regaining the reading or writing speed.
8687 When the tape driver starts reading a record, the record has to
8688 be read whole without stopping, as a tape gap is needed to stop the
8689 tape motion without loosing information.
8691 @cindex Exabyte blocking
8692 @cindex DAT blocking
8693 Using higher blocking (putting more disk blocks per tape block) will use
8694 the tape more efficiently as there will be less tape gaps. But reading
8695 such tapes may be more difficult for the system, as more memory will be
8696 required to receive at once the whole record. Further, if there is a
8697 reading error on a huge record, this is less likely that the system will
8698 succeed in recovering the information. So, blocking should not be too
8699 low, nor it should be too high. @command{tar} uses by default a blocking of
8700 20 for historical reasons, and it does not really matter when reading or
8701 writing to disk. Current tape technology would easily accommodate higher
8702 blockings. Sun recommends a blocking of 126 for Exabytes and 96 for DATs.
8703 We were told that for some DLT drives, the blocking should be a multiple
8704 of 4Kb, preferably 64Kb (@w{@kbd{-b 128}}) or 256 for decent performance.
8705 Other manufacturers may use different recommendations for the same tapes.
8706 This might also depends of the buffering techniques used inside modern
8707 tape controllers. Some imposes a minimum blocking, or a maximum blocking.
8708 Others request blocking to be some exponent of two.
8710 So, there is no fixed rule for blocking. But blocking at read time
8711 should ideally be the same as blocking used at write time. At one place
8712 I know, with a wide variety of equipment, they found it best to use a
8713 blocking of 32 to guarantee that their tapes are fully interchangeable.
8715 I was also told that, for recycled tapes, prior erasure (by the same
8716 drive unit that will be used to create the archives) sometimes lowers
8717 the error rates observed at rewriting time.
8719 I might also use @option{--number-blocks} instead of
8720 @option{--block-number}, so @option{--block} will then expand to
8721 @option{--blocking-factor} unambiguously.
8724 @section Many Archives on One Tape
8726 @FIXME{Appropriate options should be moved here from elsewhere.}
8728 @findex ntape @r{device}
8729 Most tape devices have two entries in the @file{/dev} directory, or
8730 entries that come in pairs, which differ only in the minor number for
8731 this device. Let's take for example @file{/dev/tape}, which often
8732 points to the only or usual tape device of a given system. There might
8733 be a corresponding @file{/dev/nrtape} or @file{/dev/ntape}. The simpler
8734 name is the @emph{rewinding} version of the device, while the name
8735 having @samp{nr} in it is the @emph{no rewinding} version of the same
8738 A rewinding tape device will bring back the tape to its beginning point
8739 automatically when this device is opened or closed. Since @command{tar}
8740 opens the archive file before using it and closes it afterwards, this
8741 means that a simple:
8744 $ @kbd{tar cf /dev/tape @var{directory}}
8748 will reposition the tape to its beginning both prior and after saving
8749 @var{directory} contents to it, thus erasing prior tape contents and
8750 making it so that any subsequent write operation will destroy what has
8753 @cindex tape positioning
8754 So, a rewinding device is normally meant to hold one and only one file.
8755 If you want to put more than one @command{tar} archive on a given tape, you
8756 will need to avoid using the rewinding version of the tape device. You
8757 will also have to pay special attention to tape positioning. Errors in
8758 positioning may overwrite the valuable data already on your tape. Many
8759 people, burnt by past experiences, will only use rewinding devices and
8760 limit themselves to one file per tape, precisely to avoid the risk of
8761 such errors. Be fully aware that writing at the wrong position on a
8762 tape loses all information past this point and most probably until the
8763 end of the tape, and this destroyed information @emph{cannot} be
8766 To save @var{directory-1} as a first archive at the beginning of a
8767 tape, and leave that tape ready for a second archive, you should use:
8770 $ @kbd{mt -f /dev/nrtape rewind}
8771 $ @kbd{tar cf /dev/nrtape @var{directory-1}}
8775 @dfn{Tape marks} are special magnetic patterns written on the tape
8776 media, which are later recognizable by the reading hardware. These
8777 marks are used after each file, when there are many on a single tape.
8778 An empty file (that is to say, two tape marks in a row) signal the
8779 logical end of the tape, after which no file exist. Usually,
8780 non-rewinding tape device drivers will react to the close request issued
8781 by @command{tar} by first writing two tape marks after your archive, and by
8782 backspacing over one of these. So, if you remove the tape at that time
8783 from the tape drive, it is properly terminated. But if you write
8784 another file at the current position, the second tape mark will be
8785 erased by the new information, leaving only one tape mark between files.
8787 So, you may now save @var{directory-2} as a second archive after the
8788 first on the same tape by issuing the command:
8791 $ @kbd{tar cf /dev/nrtape @var{directory-2}}
8795 and so on for all the archives you want to put on the same tape.
8797 Another usual case is that you do not write all the archives the same
8798 day, and you need to remove and store the tape between two archive
8799 sessions. In general, you must remember how many files are already
8800 saved on your tape. Suppose your tape already has 16 files on it, and
8801 that you are ready to write the 17th. You have to take care of skipping
8802 the first 16 tape marks before saving @var{directory-17}, say, by using
8806 $ @kbd{mt -f /dev/nrtape rewind}
8807 $ @kbd{mt -f /dev/nrtape fsf 16}
8808 $ @kbd{tar cf /dev/nrtape @var{directory-17}}
8811 In all the previous examples, we put aside blocking considerations, but
8812 you should do the proper things for that as well. @xref{Blocking}.
8815 * Tape Positioning:: Tape Positions and Tape Marks
8816 * mt:: The @command{mt} Utility
8819 @node Tape Positioning
8820 @subsection Tape Positions and Tape Marks
8823 Just as archives can store more than one file from the file system,
8824 tapes can store more than one archive file. To keep track of where
8825 archive files (or any other type of file stored on tape) begin and
8826 end, tape archive devices write magnetic @dfn{tape marks} on the
8827 archive media. Tape drives write one tape mark between files,
8828 two at the end of all the file entries.
8830 If you think of data as a series of records "rrrr"'s, and tape marks as
8831 "*"'s, a tape might look like the following:
8834 rrrr*rrrrrr*rrrrr*rr*rrrrr**-------------------------
8837 Tape devices read and write tapes using a read/write @dfn{tape
8838 head}---a physical part of the device which can only access one
8839 point on the tape at a time. When you use @command{tar} to read or
8840 write archive data from a tape device, the device will begin reading
8841 or writing from wherever on the tape the tape head happens to be,
8842 regardless of which archive or what part of the archive the tape
8843 head is on. Before writing an archive, you should make sure that no
8844 data on the tape will be overwritten (unless it is no longer needed).
8845 Before reading an archive, you should make sure the tape head is at
8846 the beginning of the archive you want to read. You can do it manually
8847 via @code{mt} utility (@pxref{mt}). The @code{restore} script does
8848 that automatically (@pxref{Scripted Restoration}).
8850 If you want to add new archive file entries to a tape, you should
8851 advance the tape to the end of the existing file entries, backspace
8852 over the last tape mark, and write the new archive file. If you were
8853 to add two archives to the example above, the tape might look like the
8857 rrrr*rrrrrr*rrrrr*rr*rrrrr*rrr*rrrr**----------------
8861 @subsection The @command{mt} Utility
8864 @FIXME{Is it true that this only works on non-block devices?
8865 should explain the difference, (fixed or variable).}
8866 @xref{Blocking Factor}.
8868 You can use the @command{mt} utility to advance or rewind a tape past a
8869 specified number of archive files on the tape. This will allow you
8870 to move to the beginning of an archive before extracting or reading
8871 it, or to the end of all the archives before writing a new one.
8872 @FIXME{Why isn't there an "advance 'til you find two tape marks
8875 The syntax of the @command{mt} command is:
8878 @kbd{mt [-f @var{tapename}] @var{operation} [@var{number}]}
8881 where @var{tapename} is the name of the tape device, @var{number} is
8882 the number of times an operation is performed (with a default of one),
8883 and @var{operation} is one of the following:
8885 @FIXME{is there any use for record operations?}
8890 Writes @var{number} tape marks at the current position on the tape.
8893 Moves tape position forward @var{number} files.
8896 Moves tape position back @var{number} files.
8899 Rewinds the tape. (Ignores @var{number}).
8903 Rewinds the tape and takes the tape device off-line. (Ignores @var{number}).
8906 Prints status information about the tape unit.
8910 @FIXME{Is there a better way to frob the spacing on the list?}
8912 If you don't specify a @var{tapename}, @command{mt} uses the environment
8913 variable @env{TAPE}; if @env{TAPE} is not set, @command{mt} uses the device
8916 @command{mt} returns a 0 exit status when the operation(s) were
8917 successful, 1 if the command was unrecognized, and 2 if an operation
8920 @node Using Multiple Tapes
8921 @section Using Multiple Tapes
8924 Often you might want to write a large archive, one larger than will fit
8925 on the actual tape you are using. In such a case, you can run multiple
8926 @command{tar} commands, but this can be inconvenient, particularly if you
8927 are using options like @option{--exclude=@var{pattern}} or dumping entire file systems.
8928 Therefore, @command{tar} supports multiple tapes automatically.
8930 Use @option{--multi-volume} (@option{-M}) on the command line, and
8931 then @command{tar} will, when it reaches the end of the tape, prompt
8932 for another tape, and continue the archive. Each tape will have an
8933 independent archive, and can be read without needing the other. (As
8934 an exception to this, the file that @command{tar} was archiving when
8935 it ran out of tape will usually be split between the two archives; in
8936 this case you need to extract from the first archive, using
8937 @option{--multi-volume}, and then put in the second tape when
8938 prompted, so @command{tar} can restore both halves of the file.)
8940 @GNUTAR{} multi-volume archives do not use a truly portable format.
8941 You need @GNUTAR{} at both ends to process them properly.
8943 When prompting for a new tape, @command{tar} accepts any of the following
8948 Request @command{tar} to explain possible responses
8950 Request @command{tar} to exit immediately.
8951 @item n @var{file name}
8952 Request @command{tar} to write the next volume on the file @var{file name}.
8954 Request @command{tar} to run a subshell. This option can be disabled
8955 by giving @option{--restrict} command line option to @command{tar}.
8957 Request @command{tar} to begin writing the next volume.
8960 (You should only type @samp{y} after you have changed the tape;
8961 otherwise @command{tar} will write over the volume it just finished.)
8963 @cindex End-of-archive info script
8965 @anchor{info-script}
8966 @opindex info-script
8967 @opindex new-volume-script
8968 If you want more elaborate behavior than this, give @command{tar} the
8969 @option{--info-script=@var{script-name}}
8970 (@option{--new-volume-script=@var{script-name}}, @option{-F
8971 @var{script-name}}) option. The file @var{script-name} is expected to
8972 be a program (or shell script) to be run instead of the normal
8973 prompting procedure. It is executed without any command line
8974 arguments. Additional data is passed to it via the following
8975 environment variables:
8978 @vrindex TAR_VERSION, info script environment variable
8980 @GNUTAR{} version number.
8982 @vrindex TAR_ARCHIVE, info script environment variable
8984 The name of the archive @command{tar} is processing.
8986 @vrindex TAR_VOLUME, info script environment variable
8988 Ordinal number of the volume @command{tar} is about to start.
8990 @vrindex TAR_SUBCOMMAND, info script environment variable
8991 @item TAR_SUBCOMMAND
8992 Short option describing the operation @command{tar} is executed.
8993 @xref{Operations}, for a complete list of subcommand options.
8995 @vrindex TAR_FORMAT, info script environment variable
8997 Format of the archive being processed. @xref{Formats}, for a complete
8998 list of archive format names.
9001 The info script can instruct @command{tar} to use new archive name,
9002 by writing in to file descriptor 3 (see below for an
9005 If the info script fails, @command{tar} exits; otherwise, it begins
9006 writing the next volume.
9008 The method @command{tar} uses to detect end of tape is not perfect, and
9009 fails on some operating systems or on some devices. You can use the
9010 @option{--tape-length=@var{size}} (@option{-L @var{size}}) option if
9011 @command{tar} can't detect the end of the tape itself. This option
9012 selects @option{--multi-volume} (@option{-M}) automatically. The
9013 @var{size} argument should then be the usable size of the tape in
9014 units of 1024 bytes. But for many devices, and floppy disks in
9015 particular, this option is never required for real, as far as we know.
9017 @cindex Volume number file
9021 The volume number used by @command{tar} in its tape-change prompt
9022 can be changed; if you give the
9023 @option{--volno-file=@var{file-of-number}} option, then
9024 @var{file-of-number} should be an unexisting file to be created, or
9025 else, a file already containing a decimal number. That number will be
9026 used as the volume number of the first volume written. When
9027 @command{tar} is finished, it will rewrite the file with the
9028 now-current volume number. (This does not change the volume number
9029 written on a tape label, as per @ref{label}, it @emph{only} affects
9030 the number used in the prompt.)
9032 If you want @command{tar} to cycle through a series of files or tape
9033 drives, there are three approaches to choose from. First of all, you
9034 can give @command{tar} multiple @option{--file} options. In this case
9035 the specified files will be used, in sequence, as the successive
9036 volumes of the archive. Only when the first one in the sequence needs
9037 to be used again will @command{tar} prompt for a tape change (or run
9038 the info script). Secondly, you can use the @samp{n} response to the
9039 tape-change prompt, and, finally, you can use an info script, that
9040 writes new archive name to file descriptor. The following example
9041 illustrates this approach:
9046 echo Preparing volume $TAR_VOLUME of $TAR_ARCHIVE.
9048 name=`expr $TAR_ARCHIVE : '\(.*\)-.*'`
9049 case $TAR_SUBCOMMAND in
9051 -d|-x|-t) test -r $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME || exit 1
9056 echo $@{name:-$TAR_ARCHIVE@}-$TAR_VOLUME >&3
9060 Each volume of a multi-volume archive is an independent @command{tar}
9061 archive, complete in itself. For example, you can list or extract any
9062 volume alone; just don't specify @option{--multi-volume}
9063 (@option{-M}). However, if one file in the archive is split across
9064 volumes, the only way to extract it successfully is with a
9065 multi-volume extract command @option{--extract --multi-volume}
9066 (@option{-xM}) starting on or before the volume where the file begins.
9068 For example, let's presume someone has two tape drives on a system
9069 named @file{/dev/tape0} and @file{/dev/tape1}. For having @GNUTAR{}
9070 to switch to the second drive when it needs to write the
9071 second tape, and then back to the first tape, etc., just do either of:
9074 $ @kbd{tar --create --multi-volume --file=/dev/tape0 --file=/dev/tape1 @var{files}}
9075 $ @kbd{tar cMff /dev/tape0 /dev/tape1 @var{files}}
9079 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
9080 * Tape Files:: Tape Files
9081 * Tarcat:: Concatenate Volumes into a Single Archive
9085 @node Multi-Volume Archives
9086 @subsection Archives Longer than One Tape or Disk
9087 @cindex Multi-volume archives
9090 @opindex multi-volume
9091 To create an archive that is larger than will fit on a single unit of
9092 the media, use the @option{--multi-volume} (@option{-M}) option in conjunction with
9093 the @option{--create} option (@pxref{create}). A @dfn{multi-volume}
9094 archive can be manipulated like any other archive (provided the
9095 @option{--multi-volume} option is specified), but is stored on more
9096 than one tape or disk.
9098 When you specify @option{--multi-volume}, @command{tar} does not report an
9099 error when it comes to the end of an archive volume (when reading), or
9100 the end of the media (when writing). Instead, it prompts you to load
9101 a new storage volume. If the archive is on a magnetic tape, you
9102 should change tapes when you see the prompt; if the archive is on a
9103 floppy disk, you should change disks; etc.
9105 You can read each individual volume of a multi-volume archive as if it
9106 were an archive by itself. For example, to list the contents of one
9107 volume, use @option{--list}, without @option{--multi-volume} specified.
9108 To extract an archive member from one volume (assuming it is described
9109 that volume), use @option{--extract}, again without
9110 @option{--multi-volume}.
9112 If an archive member is split across volumes (ie. its entry begins on
9113 one volume of the media and ends on another), you need to specify
9114 @option{--multi-volume} to extract it successfully. In this case, you
9115 should load the volume where the archive member starts, and use
9116 @samp{tar --extract --multi-volume}---@command{tar} will prompt for later
9117 volumes as it needs them. @xref{extracting archives}, for more
9118 information about extracting archives.
9120 @option{--info-script=@var{script-name}}
9121 (@option{--new-volume-script=@var{script-name}}, @option{-F
9122 @var{script-name}}) (@pxref{info-script}) is like
9123 @option{--multi-volume} (@option{-M}), except that @command{tar} does
9124 not prompt you directly to change media volumes when a volume is
9125 full---instead, @command{tar} runs commands you have stored in
9126 @var{script-name}. For example, this option can be used to eject
9127 cassettes, or to broadcast messages such as @samp{Someone please come
9128 change my tape} when performing unattended backups. When
9129 @var{script-name} is done, @command{tar} will assume that the media
9132 Multi-volume archives can be modified like any other archive. To add
9133 files to a multi-volume archive, you need to only mount the last
9134 volume of the archive media (and new volumes, if needed). For all
9135 other operations, you need to use the entire archive.
9137 If a multi-volume archive was labeled using
9138 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9139 (@pxref{label}) when it was created, @command{tar} will not
9140 automatically label volumes which are added later. To label
9141 subsequent volumes, specify @option{--label=@var{archive-label}} again
9142 in conjunction with the @option{--append}, @option{--update} or
9143 @option{--concatenate} operation.
9145 @cindex Labeling multi-volume archives
9148 @FIXME{There should be a sample program here, including an exit
9149 before end. Is the exit status even checked in tar? :-(}
9152 @item --multi-volume
9154 Creates a multi-volume archive, when used in conjunction with
9155 @option{--create} (@option{-c}). To perform any other operation on a multi-volume
9156 archive, specify @option{--multi-volume} in conjunction with that
9159 @item --info-script=@var{program-file}
9160 @itemx --new-volume-script=@var{program-file}
9161 @itemx -F @var{program-file}
9162 Creates a multi-volume archive via a script. Used in conjunction with
9163 @option{--create} (@option{-c}). @xref{info-script}, dor a detailed discussion.
9166 Beware that there is @emph{no} real standard about the proper way, for
9167 a @command{tar} archive, to span volume boundaries. If you have a
9168 multi-volume created by some vendor's @command{tar}, there is almost
9169 no chance you could read all the volumes with @GNUTAR{}.
9170 The converse is also true: you may not expect
9171 multi-volume archives created by @GNUTAR{} to be
9172 fully recovered by vendor's @command{tar}. Since there is little
9173 chance that, in mixed system configurations, some vendor's
9174 @command{tar} will work on another vendor's machine, and there is a
9175 great chance that @GNUTAR{} will work on most of
9176 them, your best bet is to install @GNUTAR{} on all
9177 machines between which you know exchange of files is possible.
9180 @subsection Tape Files
9183 To give the archive a name which will be recorded in it, use the
9184 @option{--label=@var{volume-label}} (@option{-V @var{volume-label}})
9185 option. This will write a special block identifying
9186 @var{volume-label} as the name of the archive to the front of the
9187 archive which will be displayed when the archive is listed with
9188 @option{--list}. If you are creating a multi-volume archive with
9189 @option{--multi-volume} (@pxref{Using Multiple Tapes}), then the
9190 volume label will have @samp{Volume @var{nnn}} appended to the name
9191 you give, where @var{nnn} is the number of the volume of the archive.
9192 (If you use the @option{--label=@var{volume-label}}) option when
9193 reading an archive, it checks to make sure the label on the tape
9194 matches the one you give. @xref{label}.
9196 When @command{tar} writes an archive to tape, it creates a single
9197 tape file. If multiple archives are written to the same tape, one
9198 after the other, they each get written as separate tape files. When
9199 extracting, it is necessary to position the tape at the right place
9200 before running @command{tar}. To do this, use the @command{mt} command.
9201 For more information on the @command{mt} command and on the organization
9202 of tapes into a sequence of tape files, see @ref{mt}.
9204 People seem to often do:
9207 @kbd{--label="@var{some-prefix} `date +@var{some-format}`"}
9210 or such, for pushing a common date in all volumes or an archive set.
9213 @subsection Concatenate Volumes into a Single Archive
9216 Sometimes it is necessary to convert existing @GNUTAR{} multi-volume
9217 archive to a single @command{tar} archive. Simply concatenating all
9218 volumes into one will not work, since each volume carries an additional
9219 information at the beginning. @GNUTAR{} is shipped with the shell
9220 script @command{tarcat} designed for this purpose.
9222 The script takes a list of files comprising a multi-volume archive
9223 and creates the resulting archive at the standard output. For example:
9226 @kbd{tarcat vol.1 vol.2 vol.3 | tar tf -}
9229 The script implements a simple heuristics to determine the format of
9230 the first volume file and to decide how to process the rest of the
9231 files. However, it makes no attempt to verify whether the files are
9232 given in order or even if they are valid @command{tar} archives.
9233 It uses @command{dd} and does not filter its standard error, so you
9234 will usually see lots of spurious messages.
9236 @FIXME{The script is not installed. Should we install it?}
9239 @section Including a Label in the Archive
9240 @cindex Labeling an archive
9241 @cindex Labels on the archive media
9245 To avoid problems caused by misplaced paper labels on the archive
9246 media, you can include a @dfn{label} entry---an archive member which
9247 contains the name of the archive---in the archive itself. Use the
9248 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9249 option in conjunction with the @option{--create} operation to include
9250 a label entry in the archive as it is being created.
9253 @item --label=@var{archive-label}
9254 @itemx -V @var{archive-label}
9255 Includes an @dfn{archive-label} at the beginning of the archive when
9256 the archive is being created, when used in conjunction with the
9257 @option{--create} operation. Checks to make sure the archive label
9258 matches the one specified (when used in conjunction with any other
9262 If you create an archive using both
9263 @option{--label=@var{archive-label}} (@option{-V @var{archive-label}})
9264 and @option{--multi-volume} (@option{-M}), each volume of the archive
9265 will have an archive label of the form @samp{@var{archive-label}
9266 Volume @var{n}}, where @var{n} is 1 for the first volume, 2 for the
9267 next, and so on. @xref{Using Multiple Tapes}, for information on
9268 creating multiple volume archives.
9270 @cindex Volume label, listing
9271 @cindex Listing volume label
9272 The volume label will be displayed by @option{--list} along with
9273 the file contents. If verbose display is requested, it will also be
9274 explicitely marked as in the example below:
9278 $ @kbd{tar --verbose --list --file=iamanarchive}
9279 V--------- 0 0 0 1992-03-07 12:01 iamalabel--Volume Header--
9280 -rw-r--r-- ringo user 40 1990-05-21 13:30 iamafilename
9285 @anchor{--test-label option}
9286 However, @option{--list} option will cause listing entire
9287 contents of the archive, which may be undesirable (for example, if the
9288 archive is stored on a tape). You can request checking only the volume
9289 by specifying @option{--test-label} option. This option reads only the
9290 first block of an archive, so it can be used with slow storage
9291 devices. For example:
9295 $ @kbd{tar --test-label --file=iamanarchive}
9300 If @option{--test-label} is used with a single command line
9301 argument, @command{tar} compares the volume label with the
9302 argument. It exits with code 0 if the two strings match, and with code
9303 2 otherwise. In this case no output is displayed. For example:
9307 $ @kbd{tar --test-label --file=iamanarchive 'iamalable'}
9309 $ @kbd{tar --test-label --file=iamanarchive 'iamalable' alabel}
9314 If you request any operation, other than @option{--create}, along
9315 with using @option{--label} option, @command{tar} will first check if
9316 the archive label matches the one specified and will refuse to proceed
9317 if it does not. Use this as a safety precaution to avoid accidentally
9318 overwriting existing archives. For example, if you wish to add files
9319 to @file{archive}, presumably labelled with string @samp{My volume},
9324 $ @kbd{tar -rf archive --label 'My volume' .}
9325 tar: Archive not labeled to match `My volume'
9330 in case its label does not match. This will work even if
9331 @file{archive} is not labelled at all.
9333 Similarly, @command{tar} will refuse to list or extract the
9334 archive if its label doesn't match the @var{archive-label}
9335 specified. In those cases, @var{archive-label} argument is interpreted
9336 as a globbing-style pattern which must match the actual magnetic
9337 volume label. @xref{exclude}, for a precise description of how match
9338 is attempted@footnote{Previous versions of @command{tar} used full
9339 regular expression matching, or before that, only exact string
9340 matching, instead of wildcard matchers. We decided for the sake of
9341 simplicity to use a uniform matching device through
9342 @command{tar}.}. If the switch @option{--multi-volume} (@option{-M}) is being used,
9343 the volume label matcher will also suffix @var{archive-label} by
9344 @w{@samp{ Volume [1-9]*}} if the initial match fails, before giving
9345 up. Since the volume numbering is automatically added in labels at
9346 creation time, it sounded logical to equally help the user taking care
9347 of it when the archive is being read.
9349 The @option{--label} was once called @option{--volume}, but is not
9350 available under that name anymore.
9352 You can also use @option{--label} to get a common information on
9353 all tapes of a series. For having this information different in each
9354 series created through a single script used on a regular basis, just
9355 manage to get some date string as part of the label. For example:
9359 $ @kbd{tar cfMV /dev/tape "Daily backup for `date +%Y-%m-%d`"}
9360 $ @kbd{tar --create --file=/dev/tape --multi-volume \
9361 --volume="Daily backup for `date +%Y-%m-%d`"}
9365 Also note that each label has its own date and time, which corresponds
9366 to when @GNUTAR{} initially attempted to write it,
9367 often soon after the operator launches @command{tar} or types the
9368 carriage return telling that the next tape is ready. Comparing date
9369 labels does give an idea of tape throughput only if the delays for
9370 rewinding tapes and the operator switching them were negligible, which
9371 is usually not the case.
9374 @section Verifying Data as It is Stored
9375 @cindex Verifying a write operation
9376 @cindex Double-checking a write operation
9381 @opindex verify, short description
9382 Attempt to verify the archive after writing.
9385 This option causes @command{tar} to verify the archive after writing it.
9386 Each volume is checked after it is written, and any discrepancies
9387 are recorded on the standard error output.
9389 Verification requires that the archive be on a back-space-able medium.
9390 This means pipes, some cartridge tape drives, and some other devices
9393 You can insure the accuracy of an archive by comparing files in the
9394 system with archive members. @command{tar} can compare an archive to the
9395 file system as the archive is being written, to verify a write
9396 operation, or can compare a previously written archive, to insure that
9399 @opindex verify, using with @option{--create}
9400 @opindex create, using with @option{--verify}
9401 To check for discrepancies in an archive immediately after it is
9402 written, use the @option{--verify} (@option{-W}) option in conjunction with
9403 the @option{--create} operation. When this option is
9404 specified, @command{tar} checks archive members against their counterparts
9405 in the file system, and reports discrepancies on the standard error.
9407 To verify an archive, you must be able to read it from before the end
9408 of the last written entry. This option is useful for detecting data
9409 errors on some tapes. Archives written to pipes, some cartridge tape
9410 drives, and some other devices cannot be verified.
9412 One can explicitly compare an already made archive with the file
9413 system by using the @option{--compare} (@option{--diff}, @option{-d})
9414 option, instead of using the more automatic @option{--verify} option.
9417 Note that these two options have a slightly different intent. The
9418 @option{--compare} option checks how identical are the logical contents of some
9419 archive with what is on your disks, while the @option{--verify} option is
9420 really for checking if the physical contents agree and if the recording
9421 media itself is of dependable quality. So, for the @option{--verify}
9422 operation, @command{tar} tries to defeat all in-memory cache pertaining to
9423 the archive, while it lets the speed optimization undisturbed for the
9424 @option{--compare} option. If you nevertheless use @option{--compare} for
9425 media verification, you may have to defeat the in-memory cache yourself,
9426 maybe by opening and reclosing the door latch of your recording unit,
9427 forcing some doubt in your operating system about the fact this is really
9428 the same volume as the one just written or read.
9430 The @option{--verify} option would not be necessary if drivers were indeed
9431 able to detect dependably all write failures. This sometimes require many
9432 magnetic heads, some able to read after the writes occurred. One would
9433 not say that drivers unable to detect all cases are necessarily flawed,
9434 as long as programming is concerned.
9436 The @option{--verify} (@option{-W}) option will not work in
9437 conjunction with the @option{--multi-volume} (@option{-M}) option or
9438 the @option{--append} (@option{-r}), @option{--update} (@option{-u})
9439 and @option{--delete} operations. @xref{Operations}, for more
9440 information on these operations.
9442 Also, since @command{tar} normally strips leading @samp{/} from file
9443 names (@pxref{absolute}), a command like @samp{tar --verify -cf
9444 /tmp/foo.tar /etc} will work as desired only if the working directory is
9445 @file{/}, as @command{tar} uses the archive's relative member names
9446 (e.g., @file{etc/motd}) when verifying the archive.
9448 @node Write Protection
9449 @section Write Protection
9451 Almost all tapes and diskettes, and in a few rare cases, even disks can
9452 be @dfn{write protected}, to protect data on them from being changed.
9453 Once an archive is written, you should write protect the media to prevent
9454 the archive from being accidentally overwritten or deleted. (This will
9455 protect the archive from being changed with a tape or floppy drive---it
9456 will not protect it from magnet fields or other physical hazards).
9458 The write protection device itself is usually an integral part of the
9459 physical media, and can be a two position (write enabled/write
9460 disabled) switch, a notch which can be popped out or covered, a ring
9461 which can be removed from the center of a tape reel, or some other
9467 This appendix lists some important user-visible changes between
9468 version @GNUTAR{} @value{VERSION} and previous versions. An up-to-date
9469 version of this document is available at
9470 @uref{http://www.gnu.org/@/software/@/tar/manual/changes.html,the
9471 @GNUTAR{} documentation page}.
9474 @item Use of short option @option{-o}.
9476 Earlier versions of @GNUTAR{} understood @option{-o} command line
9477 option as a synonym for @option{--old-archive}.
9479 @GNUTAR{} starting from version 1.13.90 understands this option as
9480 a synonym for @option{--no-same-owner}. This is compatible with
9481 UNIX98 @command{tar} implementations.
9483 However, to facilitate transition, @option{-o} option retains its
9484 old semantics when it is used with one of archive-creation commands.
9485 Users are encouraged to use @option{--format=oldgnu} instead.
9487 It is especially important, since versions of @acronym{GNU} Automake
9488 up to and including 1.8.4 invoke tar with this option to produce
9489 distribution tarballs. @xref{Formats,v7}, for the detailed discussion
9490 of this issue and its implications.
9492 Future versions of @GNUTAR{} will understand @option{-o} only as a
9493 synonym for @option{--no-same-owner}.
9495 @item Use of short option @option{-l}
9497 Earlier versions of @GNUTAR{} understood @option{-l} option as a
9498 synonym for @option{--one-file-system}. Such usage is deprecated.
9499 For compatibility with other implementations future versions of
9500 @GNUTAR{} will understand this option as a synonym for
9501 @option{--check-links}.
9503 @item Use of options @option{--portability} and @option{--old-archive}
9505 These options are deprecated. Please use @option{--format=v7} instead.
9507 @item Use of option @option{--posix}
9509 This option is deprecated. Please use @option{--format=posix} instead.
9512 @node Configuring Help Summary
9513 @appendix Configuring Help Summary
9515 Running @kbd{tar --help} displays the short @command{tar} option
9516 summary (@pxref{help}). This summary is organised by @dfn{groups} of
9517 semantically close options. The options within each group are printed
9518 in the following order: a short option, eventually followed by a list
9519 of corresponding long option names, followed by a short description of
9520 the option. For example, here is an excerpt from the actual @kbd{tar
9524 Main operation mode:
9526 -A, --catenate, --concatenate append tar files to an archive
9527 -c, --create create a new archive
9528 -d, --diff, --compare find differences between archive and
9530 --delete delete from the archive
9533 @vrindex ARGP_HELP_FMT, environment variable
9534 The exact visual representation of the help output is configurable via
9535 @env{ARGP_HELP_FMT} environment variable. The value of this variable
9536 is a comma-separated list of @dfn{format variable} assignments. There
9537 are two kinds of format variables. An @dfn{offset variable} keeps the
9538 offset of some part of help output text from the leftmost column on
9539 the screen. A @dfn{boolean} variable is a flag that toggles some
9540 output feature on or off. Depending on the type of the corresponding
9541 variable, there are two kinds of assignments:
9544 @item Offset assignment
9546 The assignment to an offset variable has the following syntax:
9549 @var{variable}=@var{value}
9553 where @var{variable} is the variable name, and @var{value} is a
9554 numeric value to be assigned to the variable.
9556 @item Boolean assignment
9558 To assign @code{true} value to a variable, simply put this variable name. To
9559 assign @code{false} value, prefix the variable name with @samp{no-}. For
9564 # Assign @code{true} value:
9566 # Assign @code{false} value:
9572 Following variables are declared:
9574 @deftypevr {Help Output} boolean dup-args
9575 If true, arguments for an option are shown with both short and long
9576 options, even when a given option has both forms, for example:
9579 -f ARCHIVE, --file=ARCHIVE use archive file or device ARCHIVE
9582 If false, then if an option has both short and long forms, the
9583 argument is only shown with the long one, for example:
9586 -f, --file=ARCHIVE use archive file or device ARCHIVE
9590 and a message indicating that the argument is applicable to both
9591 forms is printed below the options. This message can be disabled
9592 using @code{dup-args-note} (see below).
9594 The default is false.
9597 @deftypevr {Help Output} boolean dup-args-note
9598 If this variable is true, which is the default, the following notice
9599 is displayed at the end of the help output:
9602 Mandatory or optional arguments to long options are also mandatory or
9603 optional for any corresponding short options.
9606 Setting @code{no-dup-args-note} inhibits this message. Normally, only one of
9607 variables @code{dup-args} or @code{dup-args-note} should be set.
9610 @deftypevr {Help Output} offset short-opt-col
9611 Column in which short options start. Default is 2.
9615 $ @kbd{tar --help|grep ARCHIVE}
9616 -f, --file=ARCHIVE use archive file or device ARCHIVE
9617 $ @kbd{ARGP_HELP_FMT=short-opt-col=6 tar --help|grep ARCHIVE}
9618 -f, --file=ARCHIVE use archive file or device ARCHIVE
9623 @deftypevr {Help Output} offset long-opt-col
9624 Column in which long options start. Default is 6. For example:
9628 $ @kbd{tar --help|grep ARCHIVE}
9629 -f, --file=ARCHIVE use archive file or device ARCHIVE
9630 $ @kbd{ARGP_HELP_FMT=long-opt-col=16 tar --help|grep ARCHIVE}
9631 -f, --file=ARCHIVE use archive file or device ARCHIVE
9636 @deftypevr {Help Output} offset doc-opt-col
9637 Column in which @dfn{doc options} start. A doc option isn't actually
9638 an option, but rather an arbitrary piece of documentation that is
9639 displayed in much the same manner as the options. For example, in
9640 the description of @option{--format} option:
9644 -H, --format=FORMAT create archive of the given format.
9646 FORMAT is one of the following:
9648 gnu GNU tar 1.13.x format
9649 oldgnu GNU format as per tar <= 1.12
9650 pax POSIX 1003.1-2001 (pax) format
9652 ustar POSIX 1003.1-1988 (ustar) format
9653 v7 old V7 tar format
9658 the format names are doc options. Thus, if you set
9659 @kbd{ARGP_HELP_FMT=doc-opt-col=6} the above part of the help output
9660 will look as follows:
9664 -H, --format=FORMAT create archive of the given format.
9666 FORMAT is one of the following:
9668 gnu GNU tar 1.13.x format
9669 oldgnu GNU format as per tar <= 1.12
9670 pax POSIX 1003.1-2001 (pax) format
9672 ustar POSIX 1003.1-1988 (ustar) format
9673 v7 old V7 tar format
9678 @deftypevr {Help Output} offset opt-doc-col
9679 Column in which option description starts. Default is 29.
9683 $ @kbd{tar --help|grep ARCHIVE}
9684 -f, --file=ARCHIVE use archive file or device ARCHIVE
9685 $ @kbd{ARGP_HELP_FMT=opt-doc-col=19 tar --help|grep ARCHIVE}
9686 -f, --file=ARCHIVE use archive file or device ARCHIVE
9687 $ @kbd{ARGP_HELP_FMT=opt-doc-col=9 tar --help|grep ARCHIVE}
9689 use archive file or device ARCHIVE
9694 Notice, that the description starts on a separate line if
9695 @code{opt-doc-col} value is too small.
9698 @deftypevr {Help Output} offset header-col
9699 Column in which @dfn{group headers} are printed. A group header is a
9700 descriptive text preceding an option group. For example, in the
9704 Main operation mode:
9706 -A, --catenate, --concatenate append tar files to
9708 -c, --create create a new archive
9711 @samp{Main operation mode:} is the group header.
9713 The default value is 1.
9716 @deftypevr {Help Output} offset usage-indent
9717 Indentation of wrapped usage lines. Affects @option{--usage}
9718 output. Default is 12.
9721 @deftypevr {Help Output} offset rmargin
9722 Right margin of the text output. Used for wrapping.
9727 @include genfile.texi
9729 @node Snapshot Files
9730 @appendix Format of the Incremental Snapshot Files
9731 @include snapshot.texi
9733 @node Free Software Needs Free Documentation
9734 @appendix Free Software Needs Free Documentation
9735 @include freemanuals.texi
9737 @node Copying This Manual
9738 @appendix Copying This Manual
9741 * GNU Free Documentation License:: License for copying this manual
9746 @node Index of Command Line Options
9747 @appendix Index of Command Line Options
9749 This appendix contains an index of all @GNUTAR{} long command line
9750 options. The options are listed without the preceeding double-dash.
9753 @item Make sure @emph{all} options are indexed.
9754 @item Provide an index of short options
9769 @c texinfo-column-for-description: 32