]> git.wh0rd.org - dump.git/blame_incremental - dump/dump.8.in
-e inode,inode and -E file implementation.
[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 1.30 2001/07/19 09:03:44 stelian Exp $
34.\"
35.Dd __DATE__
36.Dt DUMP 8
37.Os "dump __VERSION__"
38.Sh NAME
39.Nm dump
40.Nd ext2 filesystem backup
41.Sh SYNOPSIS
42.Nm dump
43.Op Fl 0123456789ackMnSu
44.Op Fl B Ar records
45.Op Fl b Ar blocksize
46.Op Fl d Ar density
47.Op Fl e Ar inode numbers
48.Op Fl E Ar file
49.Op Fl f Ar file
50.Op Fl F Ar script
51.Op Fl h Ar level
52.Op Fl L Ar label
53.Op Fl Q Ar file
54.Op Fl s Ar feet
55.Op Fl T Ar date
56.Op Fl z Ar compression level
57.Ar files-to-dump
58.Nm dump
59.Op Fl W Li \&| Fl w
60.Pp
61.in
62(The
63.Bx 4.3
64option syntax is implemented for backward compatibility but
65is not documented here.)
66.Sh DESCRIPTION
67.Nm Dump
68examines files
69on an ext2 filesystem
70and determines which files
71need to be backed up. These files
72are copied to the given disk, tape or other
73storage medium for safe keeping (see the
74.Fl f
75option below for doing remote backups).
76A dump that is larger than the output medium is broken into
77multiple volumes.
78On most media the size is determined by writing until an
79end-of-media indication is returned.
80.Pp
81On media that cannot reliably return an end-of-media indication
82(such as some cartridge tape drives), each volume is of a fixed size;
83the actual size is determined by specifying cartridge media, or via the
84tape size, density and/or block count options below.
85By default, the same output file name is used for each volume
86after prompting the operator to change media.
87.Pp
88.Ar files-to-dump
89is either a mountpoint of a filesystem
90or a list of files and directories to be backed up as a subset of a
91filesystem.
92In the former case, either the path to a mounted filesystem
93or the device of an unmounted filesystem can be used.
94In the latter case, certain restrictions are placed on the backup:
95.Fl u
96is not allowed, the only dump level that is supported is
97.Fl 0
98and all the files and directories must reside on the same filesystem.
99.Pp
100The following options are supported by
101.Nm Ns :
102.Bl -tag -width Ds
103.It Fl 0\-9
104Dump levels.
105A level 0, full backup,
106guarantees the entire file system is copied
107(but see also the
108.Fl h
109option below).
110A level number above 0,
111incremental backup,
112tells
113.Nm dump
114to
115copy all files new or modified since the
116last dump of a lower level.
117The default level is 9.
118.It Fl B Ar records
119The number of 1 kB blocks per volume.
120This option overrides the end-of-media detection, and calculation
121of tape size based on length and density. If compression is on this
122limits the size of the compressed output per volume.
123.It Fl a
124.Dq auto-size .
125Bypass all tape length calculations, and write
126until an end-of-media indication is returned. This works best
127for most modern tape drives, and is the default.
128Use of this option is particularly recommended when appending to an
129existing tape, or using a tape drive with hardware compression
130(where you can never be sure about the compression ratio).
131.It Fl b Ar blocksize
132The number of kilobytes per dump record.
133Since the IO system slices all requests into chunks of MAXBSIZE
134(typically 64kB), it is not possible to use a larger blocksize
135without having problems later with
136.Xr restore 8 .
137Therefore
138.Nm dump
139will constrain writes to MAXBSIZE.
140The default blocksize is 10.
141.It Fl c
142Change the defaults for use with a cartridge tape drive, with a density
143of 8000 bpi, and a length of 1700 feet. Specifying a cartridge drive
144overrides the end-of-media detection.
145.It Fl e Ar inodes
146Exclude
147.Ar inodes
148from the dump. The
149.Ar inodes
150parameter is a comma separated list of inode numbers (you can use
151.Ar stat
152to find the inode number for a file or directory).
153.It Fl E Ar file
154Read list of inodes to be excluded from the dump from the text file
155.Ar file .
156The file
157.Ar file
158should be an ordinary file containing inode numbers separated by
159newlines.
160.It Fl h Ar level
161Honor the user
162.Dq nodump
163flag
164.Dp Dv UF_NODUMP
165only for dumps at or above the given
166.Ar level .
167The default honor level is 1,
168so that incremental backups omit such files
169but full backups retain them.
170.It Fl d Ar density
171Set tape density to
172.Ar density .
173The default is 1600BPI. Specifying a tape density overrides the
174end-of-media detection.
175.It Fl f Ar file
176Write the backup to
177.Ar file ;
178.Ar file
179may be a special device file
180like
181.Pa /dev/st0
182(a tape drive),
183.Pa /dev/rsd1c
184(a floppy disk drive),
185an ordinary file,
186or
187.Ql Fl
188(the standard output).
189Multiple file names may be given as a single argument separated by commas.
190Each file will be used for one dump volume in the order listed;
191if the dump requires more volumes than the number of names given,
192the last file name will used for all remaining volumes after prompting
193for media changes.
194If the name of the file is of the form
195.Dq host:file
196or
197.Dq user@host:file
198.Nm
199writes to the named file on the remote host using
200.Xr rmt 8 .
201The default path name of the remote
202.Xr rmt 8
203program is
204.\" rmt path, is the path on the remote host
205.Pa /etc/rmt ;
206this can be overridden by the environment variable
207.Ev RMT .
208.It Fl F Ar script
209Run script at the end of each tape. The device name and the
210current volume number are passed on the command line.
211The script must return 0 if
212.Nm
213should continue without asking the user to change the tape, 1 if
214.Nm
215should continue but ask the user to change the tape.
216Any other exit code will cause
217.Nm
218to abort.
219For security reasons,
220.Nm
221reverts back to the real user ID and the real group ID before
222running the script.
223.It Fl k
224Use Kerberos authentication to talk to remote tape servers. (Only
225available if this option was enabled when
226.Nm
227was compiled.)
228.It Fl L Ar label
229The user-supplied text string
230.Ar label
231is placed into the dump header, where tools like
232.Xr restore 8
233and
234.Xr file 1
235can access it.
236Note that this label is limited
237to be at most LBLSIZE (currently 16) characters, which must include
238the terminating
239.Ql \e0 .
240.It Fl M
241Enable the multi-volume feature. The name specified with
242.Fl f
243is treated as a prefix and
244.Nm
245writes in sequence to <prefix>001, <prefix>002 etc. This can be
246useful when dumping to files on an ext2 partition, in order to bypass
247the 2GB file size limitation.
248.It Fl n
249Whenever
250.Nm
251requires operator attention,
252notify all operators in the group
253.Dq operator
254by means similar to a
255.Xr wall 1 .
256.It Fl Q Ar file
257Enable the Quick File Access support. Tape positions for each
258inode are stored into the file
259.Ar file
260which is used by restore (if called with parameter Q and the filename)
261to directly position the tape at the file restore is currently working
262on. This saves hours when restoring single files from large backups,
263saves the tapes and the drive's head.
264.Pp
265It is recommended to set up the st driver to return logical tape
266positions rather than physical before calling dump/restore with
267parameter Q. Since not all tape devices support physical tape
268positions those tape devices return an error during dump/restore when
269the st driver is set to the default physical setting.
270Please see the st man page, option MTSETDRVBUFFER, or the mt man
271page, on how to set the driver to return logical tape positions.
272.Pp
273Before calling restore with parameter Q, always make sure the st
274driver is set to return the same type of tape position used during the
275call to dump. Otherwise restore may be confused.
276.It Fl s Ar feet
277Attempt to calculate the amount of tape needed at a particular density.
278If this amount is exceeded,
279.Nm
280prompts for a new tape.
281It is recommended to be a bit conservative on this option.
282The default tape length is 2300 feet. Specifying the tape size
283overrides end-of-media detection.
284.ne 1i
285.It Fl S
286Size estimate. Determine the amount of space
287that is needed to perform the dump without
288actually doing it, and display the estimated
289number of bytes it will take. This is useful
290with incremental dumps to determine how many
291volumes of media will be needed.
292.It Fl T Ar date
293Use the specified date as the starting time for the dump
294instead of the time determined from looking in
295.Pa __DUMPDATES__ .
296The format of
297.Ar date
298is the same as that of
299.Xr ctime 3 .
300This option is useful for automated dump scripts that wish to
301dump over a specific period of time.
302The
303.Fl T
304option is mutually exclusive from the
305.Fl u
306option.
307.It Fl u
308Update the file
309.Pa __DUMPDATES__
310after a successful dump.
311The format of
312.Pa __DUMPDATES__
313is readable by people, consisting of one
314free format record per line:
315filesystem name,
316increment level
317and
318.Xr ctime 3
319format dump date.
320There may be only one entry per filesystem at each level.
321The file
322.Pa __DUMPDATES__
323may be edited to change any of the fields,
324if necessary.
325.It Fl W
326.Nm Dump
327tells the operator what file systems need to be dumped.
328This information is gleaned from the files
329.Pa __DUMPDATES__
330and
331.Pa /etc/fstab .
332The
333.Fl W
334option causes
335.Nm
336to print out, for all file systems in
337.Pa __DUMPDATES__ ,
338and regognized file systems in
339.Pa /etc/fstab .
340the most recent dump date and level,
341and highlights those that should be dumped.
342If the
343.Fl W
344option is set, all other options are ignored, and
345.Nm
346exits immediately.
347.It Fl w
348Is like
349.Fl W ,
350but prints only recognized filesystems in
351.Pa /etc/fstab
352which need to be dumped.
353.It Fl z Ar compression level
354Compress every block to be written on the tape using zlib library. This
355option will work only when dumping to a file or pipe or, when dumping
356to a tape drive, if the tape drive is capable of writing variable
357length blocks. You will need at least the 0.4b22 version of restore in
358order to extract compressed tapes. Tapes written using compression will
359not be compatible with the BSD tape format. The (optional) parameter
360specifies the compression level zlib will use. The default compression
361level is 2.
362.El
363.Pp
364.Nm Dump
365requires operator intervention on these conditions:
366end of tape,
367end of dump,
368tape write error,
369tape open error or
370disk read error (if there is more than a threshold of 32).
371In addition to alerting all operators implied by the
372.Fl n
373key,
374.Nm
375interacts with the operator on
376.Em dump's
377control terminal at times when
378.Nm
379can no longer proceed,
380or if something is grossly wrong.
381All questions
382.Nm
383poses
384.Em must
385be answered by typing
386.Dq yes
387or
388.Dq no ,
389appropriately.
390.Pp
391Since making a dump involves a lot of time and effort for full dumps,
392.Nm
393checkpoints itself at the start of each tape volume.
394If writing that volume fails for some reason,
395.Nm
396will,
397with operator permission,
398restart itself from the checkpoint
399after the old tape has been rewound and removed,
400and a new tape has been mounted.
401.Pp
402.Nm Dump
403tells the operator what is going on at periodic intervals,
404including usually low estimates of the number of blocks to write,
405the number of tapes it will take, the time to completion, and
406the time to the tape change.
407The output is verbose,
408so that others know that the terminal
409controlling
410.Nm
411is busy,
412and will be for some time.
413.Pp
414In the event of a catastrophic disk event, the time required
415to restore all the necessary backup tapes or files to disk
416can be kept to a minimum by staggering the incremental dumps.
417An efficient method of staggering incremental dumps
418to minimize the number of tapes follows:
419.Bl -bullet -offset indent
420.It
421Always start with a level 0 backup, for example:
422.Bd -literal -offset indent
423/sbin/dump -0u -f /dev/st0 /usr/src
424.Ed
425.Pp
426This should be done at set intervals, say once a month or once every two months,
427and on a set of fresh tapes that is saved forever.
428.It
429After a level 0, dumps of active file
430systems are taken on a daily basis,
431using a modified Tower of Hanoi algorithm,
432with this sequence of dump levels:
433.Bd -literal -offset indent
4343 2 5 4 7 6 9 8 9 9 ...
435.Ed
436.Pp
437For the daily dumps, it should be possible to use a fixed number of tapes
438for each day, used on a weekly basis.
439Each week, a level 1 dump is taken, and
440the daily Hanoi sequence repeats beginning with 3.
441For weekly dumps, another fixed set of tapes per dumped file system is
442used, also on a cyclical basis.
443.El
444.Pp
445After several months or so, the daily and weekly tapes should get
446rotated out of the dump cycle and fresh tapes brought in.
447.Sh ENVIRONMENT
448.Bl -tag -width Fl
449.It Ev TAPE
450If no -f option was specified,
451.Nm
452will use the device specified via
453.Ev TAPE
454as the dump device.
455.Ev TAPE
456may be of the form
457.Qq tapename ,
458.Qq host:tapename ,
459or
460.Qq user@host:tapename .
461.It Ev RMT
462The environment variable
463.Ev RMT
464will be used to determine the pathname of the remote
465.Xr rmt 8
466program.
467.It Ev RSH
468.Nm Dump
469uses the contents of this variable to determine the name of the
470remote shell command to use when doing remote backups (rsh, ssh etc.).
471If this variable is not set,
472.Xr rcmd 3
473will be used, but only root will be able to do remote backups.
474.Sh FILES
475.Bl -tag -width __DUMPDATES__ -compact
476.It Pa /dev/st0
477default tape unit to dump to
478.It Pa __DUMPDATES__
479dump date records
480.It Pa /etc/fstab
481dump table: file systems and frequency
482.It Pa /etc/group
483to find group
484.Em operator
485.El
486.Sh SEE ALSO
487.Xr fstab 5 ,
488.Xr restore 8 ,
489.Xr rmt 8
490.Sh DIAGNOSTICS
491Many, and verbose.
492.Pp
493.Nm Dump
494exits with zero status on success.
495Startup errors are indicated with an exit code of 1;
496abnormal termination is indicated with an exit code of 3.
497.Sh BUGS
498It might be considered a bug that this version of dump can only handle ext2
499filesystems. Specifically, it does not work with FAT filesystems.
500.Pp
501Fewer than 32 read errors on the filesystem are ignored. If noticing
502read errors is important, the output from dump can be parsed to look for lines
503that contain the text 'read error'.
504.Pp
505Each reel requires a new process, so parent processes for
506reels already written just hang around until the entire tape
507is written.
508.Pp
509The estimated number of tapes is not correct if compression is on.
510.Pp
511It would be nice if
512.Nm
513knew about the dump sequence,
514kept track of the tapes scribbled on,
515told the operator which tape to mount when,
516and provided more assistance
517for the operator running
518.Xr restore .
519.Pp
520.Nm Dump
521cannot do remote backups without being run as root, due to its
522security history.
523Presently, it works if you set it setuid (like it used to be), but this
524might constitute a security risk. Note that you can set RSH to use
525a remote shell program instead.
526.Sh AUTHOR
527The
528.Nm dump/restore
529backup suite was ported to Linux's Second Extended File System
530by Remy Card <card@Linux.EU.Org>. He maintained the initial versions
531of dump (up and including 0.4b4, released in january 1997).
532.Pp
533Starting with 0.4b5, the new maintainer is Stelian Pop
534.br
535<pop@noos.fr>.
536.Sh AVAILABILITY
537The
538.Nm dump/restore
539backup suite is available from
540.br
541http://dump.sourceforge.net
542.Sh HISTORY
543A
544.Nm
545command appeared in
546.At v6 .