]> git.wh0rd.org - dump.git/blame_incremental - dump/dump.8.in
Look also in /etc/mtab for filesystems...
[dump.git] / dump / dump.8.in
... / ...
CommitLineData
1.\" Copyright (c) 1980, 1991, 1993
2.\" Regents of the University of California.
3.\" All rights reserved.
4.\"
5.\" Redistribution and use in source and binary forms, with or without
6.\" modification, are permitted provided that the following conditions
7.\" are met:
8.\" 1. Redistributions of source code must retain the above copyright
9.\" notice, this list of conditions and the following disclaimer.
10.\" 2. Redistributions in binary form must reproduce the above copyright
11.\" notice, this list of conditions and the following disclaimer in the
12.\" documentation and/or other materials provided with the distribution.
13.\" 3. All advertising materials mentioning features or use of this software
14.\" must display the following acknowledgement:
15.\" This product includes software developed by the University of
16.\" California, Berkeley and its contributors.
17.\" 4. Neither the name of the University nor the names of its contributors
18.\" may be used to endorse or promote products derived from this software
19.\" without specific prior written permission.
20.\"
21.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
22.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
25.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31.\" SUCH DAMAGE.
32.\"
33.\" $Id: dump.8.in,v 1.46 2002/11/12 13:05:31 stelian Exp $
34.\"
35.TH DUMP 8 "version __VERSION__ of __DATE__" BSD "System management commands"
36.SH NAME
37dump \- ext2/3 filesystem backup
38.SH SYNOPSIS
39.B dump
40[\fB\-0123456789ackMnqSuv\fR]
41[\fB\-A \fIfile\fR]
42[\fB\-B \fIrecords\fR]
43[\fB\-b \fIblocksize\fR]
44[\fB\-d \fIdensity\fR]
45[\fB\-e \fIinode numbers\fR]
46[\fB\-E \fIfile\fR]
47[\fB\-f \fIfile\fR]
48[\fB\-F \fIscript\fR]
49[\fB\-h \fIlevel\fR]
50[\fB\-I \fInr errors\fR]
51[\fB\-j\fIcompression level\fR]
52[\fB\-L \fIlabel\fR]
53[\fB\-Q \fIfile\fR]
54[\fB\-s \fIfeet\fR]
55[\fB\-T \fIdate\fR]
56[\fB\-z\fIcompression level\fR]
57.I files-to-dump
58.PP
59.B dump
60[\fB\-W \fR| \fB\-w\fR]
61.PP
62(The 4.3BSD option syntax is implemented for backward compatibility but is not
63documented here.)
64.SH DESCRIPTION
65.B Dump
66examines files on an ext2/3 filesystem and determines which files need to be
67backed up. These files are copied to the given disk, tape or other storage
68medium for safe keeping (see the
69.B \-f
70option below for doing remote backups). A dump that is larger than the output
71medium is broken into multiple volumes. On most media the size is determined by
72writing until an end-of-media indication is returned.
73.PP
74On media that cannot reliably return an end-of-media indication (such as some
75cartridge tape drives), each volume is of a fixed size; the actual size is
76determined by specifying cartridge media, or via the tape size, density and/or
77block count options below. By default, the same output file name is used for
78each volume after prompting the operator to change media.
79.PP
80.I files-to-dump
81is either a mountpoint of a filesystem or a list of files and directories to be
82backed up as a subset of a filesystem. In the former case, either the path to a
83mounted filesystem or the device of an unmounted filesystem can be used. In the
84latter case, certain restrictions are placed on the backup:
85.B \-u
86is not allowed, the only dump level that is supported is
87.B 0
88and all the files and directories must reside on the same filesystem.
89.SH OPTIONS
90The following options are supported by
91.B dump:
92.TP
93.BI \-0\-9
94Dump levels. A level 0, full backup, guarantees the entire file system is
95copied (but see also the
96.B \-h
97option below). A level number above 0, incremental backup, tells
98.B dump
99to
100copy all files new or modified since the last dump of a lower level. The
101default level is 9.
102.TP
103.BI \-a
104\*(lqauto-size\*(rq. Bypass all tape length calculations, and write until an
105end-of-media indication is returned. This works best for most modern tape
106drives, and is the default. Use of this option is particularly recommended when
107appending to an existing tape, or using a tape drive with hardware compression
108(where you can never be sure about the compression ratio).
109.TP
110.BI \-A " archive_file"
111Archive a dump table-of-contents in the specified
112.I archive_file
113to be used by
114.BR restore (8)
115to determine whether a file is in the dump file that is being restored.
116.TP
117.BI \-b " blocksize"
118The number of kilobytes per dump record. The default blocksize is 10
119and the maximal value is 1024.
120Note however that, since the IO system slices all requests into chunks
121of
122.B MAXBSIZE
123(which can be as low as 64kB), you can experience problems with
124.BR dump (8)
125and
126.BR restore (8)
127when using a higher value, depending on your kernel and/or libC versions.
128.TP
129.BI \-B " records"
130The number of 1 kB blocks per volume. Not normally required, as
131.B dump
132can detect end-of-media. When the specified size is reached,
133.B dump
134waits for you to change the volume. This option overrides the calculation of
135tape size based on length and density. If compression is on this limits the
136size of the compressed output per volume.
137.TP
138.BI \-c
139Change the defaults for use with a cartridge tape drive, with a density of 8000
140bpi, and a length of 1700 feet. Specifying a cartridge drive overrides the
141end-of-media detection.
142.TP
143.BI \-d " density"
144Set tape density to
145.IR density .
146The default is 1600BPI. Specifying a tape density overrides the end-of-media
147detection.
148.TP
149.BI \-e " inodes"
150Exclude
151.I inodes
152from the dump. The
153.I inodes
154parameter is a comma separated list of inode numbers (you can use
155.BR stat (1)
156to find the inode number for a file or directory).
157.TP
158.BI \-E " file"
159Read list of inodes to be excluded from the dump from the text file
160.IR file .
161The file
162.I file
163should be an ordinary file containing inode numbers separated by newlines.
164.TP
165.BI \-f " file"
166Write the backup to
167.IR file ;
168.I file
169may be a special device file like
170.I /dev/st0
171(a tape drive),
172.I /dev/rsd1c
173(a floppy disk drive), an ordinary file, or
174.I \-
175(the standard output). Multiple file names may be given as a single argument
176separated by commas. Each file will be used for one dump volume in the order
177listed; if the dump requires more volumes than the number of names given,
178the last file name will used for all remaining volumes after prompting for
179media changes. If the name of the file is of the form
180.I host:file
181or
182.I user@host:file
183.B dump
184writes to the named file on the remote host using
185.BR rmt (8).
186The default path name of the remote
187.BR rmt (8)
188program is
189.IR /etc/rmt ;
190this can be overridden by the environment variable
191.BR RMT .
192.TP
193.BI \-F " script"
194Run script at the end of each tape (except for the last one).
195The device name and the current volume number are passed on the
196command line. The script must return 0 if
197.B dump
198should continue without asking the user to change the tape, 1 if
199.B dump
200should continue but ask the user to change the tape. Any other exit code will
201cause
202.B dump
203to abort. For security reasons,
204.B dump
205reverts back to the real user ID and the real group ID before running the
206script.
207.TP
208.BI \-h " level"
209Honor the user
210.B nodump
211flag
212.B UF_NODUMP
213only for dumps at or above the given
214.IR level .
215The default honor level is 1, so that incremental backups omit such files but
216full backups retain them.
217.TP
218.BI \-I " nr errors"
219By default,
220.B dump
221will ignore the first 32 read errors on the file system before asking for
222operator intervention. You can change this using this flag to any value. This
223is useful when running
224.B dump
225on an active filesystem where read errors simply indicate an inconsistency
226between the mapping and dumping passes.
227.TP
228.BI \-j "compression level"
229Compress every block to be written on the tape using bzlib library. This option
230will work only when dumping to a file or pipe or, when dumping to a tape drive,
231if the tape drive is capable of writing variable length blocks. You will need
232at least the 0.4b24 version of
233.B restore
234in order to extract compressed tapes. Tapes written using compression will not
235be compatible with the BSD tape format. The (optional) parameter specifies the
236compression level bzlib will use. The default compression level is 2. If the
237optional parameter is specified, there should be no white space between the
238option letter and the parameter.
239.TP
240.BI \-k
241Use Kerberos authentication to talk to remote tape servers. (Only available if
242this option was enabled when
243.B dump
244was compiled.)
245.TP
246.BI \-L " label"
247The user-supplied text string
248.I label
249is placed into the dump header, where tools like
250.BR restore (8)
251and
252.BR file (8)
253can access it. Note that this label is limited to be at most
254.B LBLSIZE
255(currently 16) characters, which must include the terminating \e0.
256.TP
257.BI \-m
258If this flag is specified,
259.B dump
260will optimise the output for inodes having been changed but not modified since
261the last dump ('changed' and 'modified' have the meaning defined in
262.BR stat (2)
263). For those inodes,
264.B dump
265will save only the metadata, instead of saving the entire inode contents.
266Inodes which are either directories or have been modified since the last dump
267are saved in a regular way. Uses of this flag must be consistent, meaning that
268either every dump in an incremental dump set have the flag, or no one has it.
269.IP
270Tapes written using such 'metadata only' inodes will not be compatible with the
271BSD tape format or older versions of
272.B restore.
273.TP
274.BI \-M
275Enable the multi-volume feature. The name specified with
276.B f
277is treated as a prefix and
278.B dump
279writes in sequence to
280.I <prefix>001, <prefix>002
281etc. This can be useful when dumping to files on an ext2 partition, in order to
282bypass the 2GB file size limitation.
283.TP
284.BI \-n
285Whenever
286.B dump
287requires operator attention, notify all operators in the group
288.B operator
289by means similar to a
290.BR wall (1).
291.TP
292.BI \-q
293Make
294.B dump
295abort immediately whenever operator attention is required, without prompting in
296case of write errors, tape changes etc.
297.TP
298.BI \-Q " file"
299Enable the Quick File Access support. Tape positions for each inode are stored
300into the file
301.I file
302which is used by
303.B restore
304(if called with parameter
305.B \-Q
306and the filename) to directly position the tape at the file
307.B restore
308is currently working on. This saves hours when restoring single files from
309large backups, saves the tapes and the drive's head.
310.IP
311It is recommended to set up the st driver to return logical tape positions
312rather than physical before calling
313.B dump/restore
314with parameter
315.BR \-Q .
316Since not all tape devices support physical tape positions those tape devices
317return an error during
318.B dump/restore
319when the st driver is set to the default physical setting. Please see the
320.BR st (4)
321man page, option
322.B MTSETDRVBUFFER
323, or the
324.BR mt (1)
325man page, on how to set the driver to return logical tape positions.
326.IP
327Before calling
328.B restore
329with parameter
330.BR \-Q ,
331always make sure the st driver is set to return the same type of tape position
332used during the call to
333.BR dump .
334Otherwise
335.B restore
336may be confused.
337.IP
338This option can be used when dumping to local tapes (see above) or to local
339files.
340.TP
341.BI \-s " feet"
342Attempt to calculate the amount of tape needed at a particular density. If this
343amount is exceeded,
344.B dump
345prompts for a new tape. It is recommended to be a bit conservative on this
346option. The default tape length is 2300 feet. Specifying the tape size
347overrides end-of-media detection.
348.TP
349.BI \-S
350Size estimate. Determine the amount of space that is needed to perform the dump
351without actually doing it, and display the estimated number of bytes it will
352take. This is useful with incremental dumps to determine how many volumes of
353media will be needed.
354.TP
355.BI \-T " date"
356Use the specified date as the starting time for the dump instead of the time
357determined from looking in
358.I __DUMPDATES__ .
359The format of
360.I date
361is the same as that of
362.BR ctime (3).
363This option is useful for automated dump scripts that wish to dump over a
364specific period of time. The
365.B \-T
366option is mutually exclusive from the
367.B \-u
368option.
369.TP
370.BI \-u
371Update the file
372.I __DUMPDATES__
373after a successful dump. The format of
374.I __DUMPDATES__
375is readable by people, consisting of one free format record per line:
376filesystem name, increment level and
377.BR ctime (3)
378format dump date. There may be only one entry per filesystem at each level. The
379file
380.I __DUMPDATES__
381may be edited to change any of the fields, if necessary.
382.TP
383.BI \-v
384The
385.B \-v
386(verbose) makes
387.B dump
388to print extra information which could be helpful in debug sessions.
389.TP
390.BI \-W
391.B Dump
392tells the operator what file systems need to be dumped. This information is
393gleaned from the files
394.I __DUMPDATES__
395and
396.IR /etc/fstab .
397The
398.B \-W
399option causes
400.B dump
401to print out, for all file systems in
402.I __DUMPDATES__ ,
403and regognized file systems in
404.IR /etc/fstab .
405the most recent dump date and level, and highlights those that should be
406dumped. If the
407.B \-W
408option is set, all other options are ignored, and
409.B dump
410exits immediately.
411.TP
412.BI \-w
413Is like
414.BR \-W ,
415but prints only recognized filesystems in
416.I /etc/fstab
417which need to be dumped.
418.TP
419.BI \-z "compression level"
420Compress every block to be written on the tape using zlib library. This option
421will work only when dumping to a file or pipe or, when dumping to a tape drive,
422if the tape drive is capable of writing variable length blocks. You will need
423at least the 0.4b22 version of
424.B restore
425in order to extract compressed tapes. Tapes written using compression will not
426be compatible with the BSD tape format. The (optional) parameter specifies the
427compression level zlib will use. The default compression level is 2. If the
428optional parameter is specified, there should be no white space between the
429option letter and the parameter.
430.PP
431.B Dump
432requires operator intervention on these conditions: end of tape, end of dump,
433tape write error, tape open error or disk read error (if there is more than a
434threshold of nr errors). In addition to alerting all operators implied by the
435.B \-n
436key,
437.B dump
438interacts with the operator on dump's control terminal at times when
439.B dump
440can no longer proceed, or if something is grossly wrong. All questions
441.B dump
442poses
443.I must
444be answered by typing \*(lqyes\*(rq or \*(lqno\*(rq, appropriately.
445.PP
446Since making a dump involves a lot of time and effort for full dumps,
447.B dump
448checkpoints itself at the start of each tape volume. If writing that volume
449fails for some reason,
450.B dump
451will, with operator permission, restart itself from the checkpoint after the
452old tape has been rewound and removed, and a new tape has been mounted.
453.PP
454.B Dump
455tells the operator what is going on at periodic intervals, including usually
456low estimates of the number of blocks to write, the number of tapes it will
457take, the time to completion, and the time to the tape change. The output is
458verbose, so that others know that the terminal controlling
459.B dump
460is busy, and will be for some time.
461.PP
462In the event of a catastrophic disk event, the time required to restore all the
463necessary backup tapes or files to disk can be kept to a minimum by staggering
464the incremental dumps. An efficient method of staggering incremental dumps to
465minimize the number of tapes follows:
466.IP \(em
467Always start with a level 0 backup, for example:
468.RS 14
469.B /sbin/dump -0u -f /dev/st0 /usr/src
470.RE
471.IP
472This should be done at set intervals, say once a month or once every two months,
473and on a set of fresh tapes that is saved forever.
474.IP \(em
475After a level 0, dumps of active file systems are taken on a daily basis, using
476a modified Tower of Hanoi algorithm, with this sequence of dump levels:
477.RS 14
478.B 3 2 5 4 7 6 9 8 9 9 ...
479.RE
480.IP
481For the daily dumps, it should be possible to use a fixed number of tapes for
482each day, used on a weekly basis. Each week, a level 1 dump is taken, and the
483daily Hanoi sequence repeats beginning with 3. For weekly dumps, another fixed
484set of tapes per dumped file system is used, also on a cyclical basis.
485.PP
486After several months or so, the daily and weekly tapes should get rotated out
487of the dump cycle and fresh tapes brought in.
488.SH ENVIRONMENT
489.TP
490.B TAPE
491If no
492.B \-f
493option was specified,
494.B dump
495will use the device specified via
496.B TAPE
497as the dump device.
498.B TAPE
499may be of the form
500.IR tapename ,
501.IR host:tapename ,
502or
503.IR user@host:tapename .
504.TP
505.B RMT
506The environment variable
507.B RMT
508will be used to determine the pathname of the remote
509.BR rmt (8)
510program.
511.TP
512.B RSH
513.B Dump
514uses the contents of this variable to determine the name of the remote shell
515command to use when doing remote backups (rsh, ssh etc.). If this variable is
516not set,
517.BR rcmd (3)
518will be used, but only root will be able to do remote backups.
519.SH FILES
520.TP
521.I /dev/st0
522default tape unit to dump to
523.TP
524.I __DUMPDATES__
525dump date records
526.TP
527.I /etc/fstab
528dump table: file systems and frequency
529.TP
530.I /etc/group
531to find group
532.I operator
533.SH SEE ALSO
534.BR fstab (5),
535.BR restore (8),
536.BR rmt (8)
537.SH DIAGNOSTICS
538Many, and verbose.
539.SH EXIT STATUS
540.B Dump
541exits with zero status on success. Startup errors are indicated with an exit
542code of 1; abnormal termination is indicated with an exit code of 3.
543.SH BUGS
544It might be considered a bug that this version of dump can only handle ext2/3
545filesystems. Specifically, it does not work with FAT filesystems.
546.PP
547Fewer than 32 read errors (change this with
548.BR \-I )
549on the filesystem are ignored. If noticing read errors is important, the output
550from dump can be parsed to look for lines that contain the text 'read error'.
551.PP
552When a read error occurs,
553.B dump
554prints out the corresponding physical disk block and sector number and the
555ext2/3 logical block number. It doesn't print out the corresponing file name or
556even the inode number. The user has to use
557.BR debugfs (8),
558commands
559.B ncheck
560and
561.B icheck
562to translate the
563.B ext2blk
564number printed out by
565.B dump
566into an inode number, then into a file name.
567.PP
568Each reel requires a new process, so parent processes for reels already written
569just hang around until the entire tape is written.
570.PP
571The estimated number of tapes is not correct if compression is on.
572.PP
573It would be nice if
574.B dump
575knew about the dump sequence, kept track of the tapes scribbled on, told the
576operator which tape to mount when, and provided more assistance for the
577operator running
578.BR restore .
579.PP
580.B Dump
581cannot do remote backups without being run as root, due to its security history.
582Presently, it works if you set it setuid (like it used to be), but this might
583constitute a security risk. Note that you can set
584.B RSH
585to use a remote shell program instead.
586.SH AUTHOR
587The
588.B dump/restore
589backup suite was ported to Linux's Second Extended File System by Remy Card
590<card@Linux.EU.Org>. He maintained the initial versions of
591.B dump
592(up and including 0.4b4, released in january 1997).
593.PP
594Starting with 0.4b5, the new maintainer is Stelian Pop <stelian@popies.net>.
595.SH AVAILABILITY
596The
597.B dump/restore
598backup suite is available from <http://dump.sourceforge.net>
599.SH HISTORY
600A
601.B dump
602command appeared in
603.B Version 6 AT&T UNIX.