xref: /freebsd/sbin/dump/dump.8 (revision 4f29da19bd44f0e99f021510460a81bf754c21d2)
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.\" 4. Neither the name of the University nor the names of its contributors
14.\"    may be used to endorse or promote products derived from this software
15.\"    without specific prior written permission.
16.\"
17.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
18.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
21.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27.\" SUCH DAMAGE.
28.\"
29.\"     @(#)dump.8	8.3 (Berkeley) 5/1/95
30.\" $FreeBSD$
31.\"
32.Dd February 24, 2006
33.Dt DUMP 8
34.Os
35.Sh NAME
36.Nm dump ,
37.Nm rdump
38.Nd file system backup
39.Sh SYNOPSIS
40.Nm
41.Op Fl 0123456789acLnSu
42.Op Fl B Ar records
43.Op Fl b Ar blocksize
44.Op Fl C Ar cachesize
45.Op Fl D Ar dumpdates
46.Op Fl d Ar density
47.Op Fl f Ar file | Fl P Ar pipecommand
48.Op Fl h Ar level
49.Op Fl s Ar feet
50.Op Fl T Ar date
51.Ar filesystem
52.Nm
53.Fl W | Fl w
54.Pp
55.Nm rdump
56is an alternate name for
57.Nm .
58.Pp
59.in \" XXX
60(The
61.Bx 4.3
62option syntax is implemented for backward compatibility, but
63is not documented here.)
64.Sh DESCRIPTION
65The
66.Nm
67utility examines files
68on a file system
69and determines which files
70need to be backed up.
71These 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.
80This can be enforced
81by using the
82.Fl a
83option.
84.Pp
85On media that cannot reliably return an end-of-media indication
86(such as some cartridge tape drives)
87each volume is of a fixed size;
88the actual size is determined by the tape size and density and/or
89.Fl B
90options.
91By default, the same output file name is used for each volume
92after prompting the operator to change media.
93.Pp
94The file system to be dumped is specified by the argument
95.Ar filesystem
96as either its device-special file or its mount point
97(if that is in a standard entry in
98.Pa /etc/fstab ) .
99.Pp
100The following options are supported by
101.Nm :
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 dump to
113copy all files new or modified since the
114last dump of any lower level.
115The default level is 0.
116.It Fl a
117.Dq auto-size .
118Bypass all tape length considerations, and enforce writing
119until an end-of-media indication is returned.
120This fits best for most modern tape drives.
121Use of this option is particularly
122recommended when appending to an existing tape, or using a tape
123drive with hardware compression (where you can never be sure about
124the compression ratio).
125.It Fl B Ar records
126The number of kilobytes per output volume, except that if it is
127not an integer multiple of the output block size,
128the command uses the next smaller such multiple.
129This option overrides the calculation of tape size
130based on length and density.
131.It Fl b Ar blocksize
132The number of kilobytes per output block.
133The default block size is 10.
134.It Fl C Ar cachesize
135Specify the cache size in megabytes.
136This will greatly improve performance
137at the cost of
138.Nm
139possibly not noticing changes in the file system between passes.
140It is
141recommended that you always use this option when dumping a snapshot.
142Beware that
143.Nm
144forks, and the actual memory use may be larger than the specified cache
145size.
146The recommended cache size is between 8 and 32 (megabytes).
147.It Fl c
148Change the defaults for use with a cartridge tape drive, with a density
149of 8000 bpi, and a length of 1700 feet.
150.It Fl D Ar dumpdates
151Specify an alternate path to the
152.Pa dumpdates
153file.
154The default is
155.Pa /etc/dumpdates .
156.It Fl d Ar density
157Set tape density to
158.Ar density .
159The default is 1600BPI.
160.It Fl f Ar file
161Write the backup to
162.Ar file ;
163.Ar file
164may be a special device file
165like
166.Pa /dev/sa0
167(a tape drive),
168.Pa /dev/fd1
169(a floppy disk drive),
170an ordinary file,
171or
172.Sq Fl
173(the standard output).
174Multiple file names may be given as a single argument separated by commas.
175Each file will be used for one dump volume in the order listed;
176if the dump requires more volumes than the number of names given,
177the last file name will used for all remaining volumes after prompting
178for media changes.
179If the name of the file is of the form
180.Dq host:file ,
181or
182.Dq user@host:file ,
183.Nm
184writes to the named file on the remote host using
185.Xr rmt 8 .
186The default path name of the remote
187.Xr rmt 8
188program is
189.\" rmt path, is the path on the remote host
190.Pa /etc/rmt ;
191this can be overridden by the environment variable
192.Ev RMT .
193.It Fl P Ar pipecommand
194Use
195.Xr popen 3
196to execute the
197.Xr sh 1
198script string defined by
199.Ar pipecommand
200for the output device of each volume.
201This child pipeline's
202.Dv stdin
203.Pq Pa /dev/fd/0
204is redirected from the
205.Nm
206output stream, and the environment variable
207.Ev DUMP_VOLUME
208is set to the current volume number being written.
209After every volume, the writer side of the pipe is closed and
210.Ar pipecommand
211is executed again.
212Subject to the media size specified by
213.Fl B ,
214each volume is written in this manner as if the output were a tape drive.
215.It Fl h Ar level
216Honor the user
217.Dq nodump
218flag
219.Pq Dv UF_NODUMP
220only for dumps at or above the given
221.Ar level .
222The default honor level is 1,
223so that incremental backups omit such files
224but full backups retain them.
225.It Fl L
226This option is to notify
227.Nm
228that it is dumping a live file system.
229To obtain a consistent dump image,
230.Nm
231takes a snapshot of the file system in the
232.Pa .snap
233directory in the root of the file system being dumped and
234then does a dump of the snapshot.
235The snapshot is removed when the dump is complete.
236This option is ignored for unmounted or read-only file systems.
237If the
238.Pa .snap
239directory does not exist in the root of the file system being dumped,
240a warning will be issued and the
241.Nm
242will revert to the standard behavior.
243This problem can be corrected by creating a
244.Pa .snap
245directory in the root of the file system to be dumped;
246its owner should be
247.Dq Li root ,
248its group should be
249.Dq Li operator ,
250and its mode should be
251.Dq Li 0770 .
252.It Fl n
253Whenever
254.Nm
255requires operator attention,
256notify all operators in the group
257.Dq operator
258by means similar to a
259.Xr wall 1 .
260.It Fl S
261Display an estimate of the backup size and the number of
262tapes required, and exit without actually performing the dump.
263.It Fl s Ar feet
264Attempt to calculate the amount of tape needed
265at a particular density.
266If this amount is exceeded,
267.Nm
268prompts for a new tape.
269It is recommended to be a bit conservative on this option.
270The default tape length is 2300 feet.
271.It Fl T Ar date
272Use the specified date as the starting time for the dump
273instead of the time determined from looking in
274the
275.Pa dumpdates
276file.
277The format of date is the same as that of
278.Xr ctime 3 .
279This option is useful for automated dump scripts that wish to
280dump over a specific period of time.
281The
282.Fl T
283option is mutually exclusive from the
284.Fl u
285option.
286.It Fl u
287Update the
288.Pa dumpdates
289file
290after a successful dump.
291The format of
292the
293.Pa dumpdates
294file
295is readable by people, consisting of one
296free format record per line:
297file system name,
298increment level
299and
300.Xr ctime 3
301format dump date.
302There may be only one entry per file system at each level.
303The
304.Pa dumpdates
305file
306may be edited to change any of the fields,
307if necessary.
308The default path for the
309.Pa dumpdates
310file is
311.Pa /etc/dumpdates ,
312but the
313.Fl D
314option may be used to change it.
315.It Fl W
316Tell the operator what file systems need to be dumped.
317This information is gleaned from the files
318.Pa dumpdates
319and
320.Pa /etc/fstab .
321The
322.Fl W
323option causes
324.Nm
325to print out, for each file system in
326the
327.Pa dumpdates
328file
329the most recent dump date and level,
330and highlights those file systems that should be dumped.
331If the
332.Fl W
333option is set, all other options are ignored, and
334.Nm
335exits immediately.
336.It Fl w
337Is like
338.Fl W ,
339but prints only those file systems which need to be dumped.
340.El
341.Pp
342Directories and regular files which have their
343.Dq nodump
344flag
345.Pq Dv UF_NODUMP
346set will be omitted along with everything under such directories,
347subject to the
348.Fl h
349option.
350.Pp
351The
352.Nm
353utility requires operator intervention on these conditions:
354end of tape,
355end of dump,
356tape write error,
357tape open error or
358disk read error (if there are more than a threshold of 32).
359In addition to alerting all operators implied by the
360.Fl n
361key,
362.Nm
363interacts with the operator on
364.Em dump's
365control terminal at times when
366.Nm
367can no longer proceed,
368or if something is grossly wrong.
369All questions
370.Nm
371poses
372.Em must
373be answered by typing
374.Dq yes
375or
376.Dq no ,
377appropriately.
378.Pp
379Since making a dump involves a lot of time and effort for full dumps,
380.Nm
381checkpoints itself at the start of each tape volume.
382If writing that volume fails for some reason,
383.Nm
384will,
385with operator permission,
386restart itself from the checkpoint
387after the old tape has been rewound and removed,
388and a new tape has been mounted.
389.Pp
390The
391.Nm
392utility tells the operator what is going on at periodic intervals
393(every 5 minutes, or promptly after receiving
394.Dv SIGINFO ) ,
395including usually low estimates of the number of blocks to write,
396the number of tapes it will take, the time to completion, and
397the time to the tape change.
398The output is verbose,
399so that others know that the terminal
400controlling
401.Nm
402is busy,
403and will be for some time.
404.Pp
405In the event of a catastrophic disk event, the time required
406to restore all the necessary backup tapes or files to disk
407can be kept to a minimum by staggering the incremental dumps.
408An efficient method of staggering incremental dumps
409to minimize the number of tapes follows:
410.Bl -bullet -offset indent
411.It
412Always start with a level 0 backup, for example:
413.Bd -literal -offset indent
414/sbin/dump -0u -f /dev/nsa0 /usr/src
415.Ed
416.Pp
417This should be done at set intervals, say once a month or once every two months,
418and on a set of fresh tapes that is saved forever.
419.It
420After a level 0, dumps of active file systems (file systems with files
421that change, depending on your partition layout some file systems may
422contain only data that does not change) are taken on a daily basis,
423using a modified Tower of Hanoi algorithm,
424with this sequence of dump levels:
425.Bd -literal -offset indent
4263 2 5 4 7 6 9 8 9 9 ...
427.Ed
428.Pp
429For the daily dumps, it should be possible to use a fixed number of tapes
430for each day, used on a weekly basis.
431Each week, a level 1 dump is taken, and
432the daily Hanoi sequence repeats beginning with 3.
433For weekly dumps, another fixed set of tapes per dumped file system is
434used, also on a cyclical basis.
435.El
436.Pp
437After several months or so, the daily and weekly tapes should get
438rotated out of the dump cycle and fresh tapes brought in.
439.Sh ENVIRONMENT
440.Bl -tag -width ".Ev TAPE"
441.It Ev TAPE
442The
443.Ar file
444or device to dump to if the
445.Fl f
446option is not used.
447.It Ev RMT
448Pathname of the remote
449.Xr rmt 8
450program.
451.It Ev RSH
452Pathname of a remote shell program, if not
453.Xr rsh 1 .
454.El
455.Sh EXAMPLES
456Dumps the "/u" file system to dvds using growisofs.
457Uses a 16MB cache and creates a snapshot of the dump and records the
458dumpdates file.
459.Bd -literal
460/sbin/dump -0u  -L -C16 -B4589840 -P 'growisofs -Z /dev/cd0=/dev/fd/0' /u
461.Ed
462.Sh FILES
463.Bl -tag -width /etc/dumpdates -compact
464.It Pa /dev/sa0
465default tape unit to dump to
466.It Pa /etc/dumpdates
467dump date records
468(this can be changed;
469see the
470.Fl D
471option)
472.It Pa /etc/fstab
473dump table: file systems and frequency
474.It Pa /etc/group
475to find group
476.Em operator
477.El
478.Sh EXIT STATUS
479Dump exits with zero status on success.
480Startup errors are indicated with an exit code of 1;
481abnormal termination is indicated with an exit code of 3.
482.Sh DIAGNOSTICS
483Many, and verbose.
484.Sh SEE ALSO
485.Xr chflags 1 ,
486.Xr fstab 5 ,
487.Xr restore 8 ,
488.Xr rmt 8
489.Sh HISTORY
490A
491.Nm
492utility appeared in
493.At v6 .
494.Sh BUGS
495Fewer than 32 read errors on the file system are ignored, though all
496errors will generate a warning message.
497This is a bit of a compromise.
498In practice, it is possible to generate read errors when doing dumps
499on mounted partitions if the file system is being modified while the
500.Nm
501is running.
502Since dumps are often done in an unattended fashion using
503.Xr cron 8
504jobs asking for Operator intervention would result in the
505.Nm
506dying.
507However, there is nothing wrong with a dump tape written when this sort
508of read error occurs, and there is no reason to terminate the
509.Nm .
510.Pp
511Each reel requires a new process, so parent processes for
512reels already written just hang around until the entire tape
513is written.
514.Pp
515The
516.Nm
517utility with the
518.Fl W
519or
520.Fl w
521options does not report file systems that have never been recorded
522in the
523.Pa dumpdates
524file,
525even if listed in
526.Pa /etc/fstab .
527.Pp
528It would be nice if
529.Nm
530knew about the dump sequence,
531kept track of the tapes scribbled on,
532told the operator which tape to mount when,
533and provided more assistance
534for the operator running
535.Xr restore 8 .
536.Pp
537The
538.Nm
539utility cannot do remote backups without being run as root, due to its
540security history.
541This will be fixed in a later version of
542.Fx .
543Presently, it works if you set it setuid (like it used to be), but this
544might constitute a security risk.
545