xref: /freebsd/sbin/fsck_ffs/fsck_ffs.8 (revision b00ab754)
1.\"
2.\" Copyright (c) 1980, 1989, 1991, 1993
3.\"	The Regents of the University of California.  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. 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.\"	@(#)fsck.8	8.4 (Berkeley) 5/9/95
30.\" $FreeBSD$
31.\"
32.Dd January 13, 2018
33.Dt FSCK_FFS 8
34.Os
35.Sh NAME
36.Nm fsck_ffs ,
37.Nm fsck_ufs
38.Nd file system consistency check and interactive repair
39.Sh SYNOPSIS
40.Nm
41.Op Fl BCdEFfnpRrSyZ
42.Op Fl b Ar block
43.Op Fl c Ar level
44.Op Fl m Ar mode
45.Ar filesystem
46.Ar ...
47.Sh DESCRIPTION
48The specified disk partitions and/or file systems are checked.
49In "preen" or "check clean" mode the clean flag of each file system's
50superblock is examined and only those file systems that are not marked clean
51are checked.
52File systems are marked clean when they are unmounted,
53when they have been mounted read-only, or when
54.Nm
55runs on them successfully.
56If the
57.Fl f
58option is specified, the file systems
59will be checked regardless of the state of their clean flag.
60.Pp
61The kernel takes care that only a restricted class of innocuous file system
62inconsistencies can happen unless hardware or software failures intervene.
63These are limited to the following:
64.Pp
65.Bl -item -compact -offset indent
66.It
67Unreferenced inodes
68.It
69Link counts in inodes too large
70.It
71Missing blocks in the free map
72.It
73Blocks in the free map also in files
74.It
75Counts in the super-block wrong
76.El
77.Pp
78These are the only inconsistencies that
79.Nm
80with the
81.Fl p
82option will correct; if it encounters other inconsistencies, it exits
83with an abnormal return status and an automatic reboot will then fail.
84For each corrected inconsistency one or more lines will be printed
85identifying the file system on which the correction will take place,
86and the nature of the correction.
87After successfully correcting a file system,
88.Nm
89will print the number of files on that file system,
90the number of used and free blocks,
91and the percentage of fragmentation.
92.Pp
93If sent a
94.Dv QUIT
95signal,
96.Nm
97will finish the file system checks, then exit with an abnormal
98return status that causes an automatic reboot to fail.
99This is useful when you want to finish the file system checks during an
100automatic reboot,
101but do not want the machine to come up multiuser after the checks complete.
102.Pp
103If
104.Nm
105receives a
106.Dv SIGINFO
107(see the
108.Dq status
109argument for
110.Xr stty 1 )
111signal, a line will be written to the standard output indicating
112the name of the device currently being checked, the current phase
113number and phase-specific progress information.
114.Pp
115Without the
116.Fl p
117option,
118.Nm
119audits and interactively repairs inconsistent conditions for file systems.
120If the file system is inconsistent the operator is prompted for concurrence
121before each correction is attempted.
122It should be noted that some of the corrective actions which are not
123correctable under the
124.Fl p
125option will result in some loss of data.
126The amount and severity of data lost may be determined from the diagnostic
127output.
128The default action for each consistency correction
129is to wait for the operator to respond
130.Li yes
131or
132.Li no .
133If the operator does not have write permission on the file system
134.Nm
135will default to a
136.Fl n
137action.
138.Pp
139The following flags are interpreted by
140.Nm :
141.Bl -tag -width indent
142.It Fl B
143A check is done on the specified and possibly active file system.
144The set of corrections that can be done is limited to those done
145when running in preen mode (see the
146.Fl p
147flag).
148If unexpected errors are found,
149the file system is marked as needing a foreground check and
150.Nm
151exits without attempting any further cleaning.
152.It Fl b
153Use the block specified immediately after the flag as
154the super block for the file system.
155An alternate super block is usually located at block 32 for UFS1,
156and block 160 for UFS2.
157.Pp
158See the
159.Fl N
160flag of
161.Xr newfs 8 .
162.It Fl C
163Check if file system was dismounted cleanly.
164If so, skip file system checks (like "preen").
165However, if the file system was not cleanly dismounted, do full checks,
166as if
167.Nm
168was invoked without
169.Fl C .
170.It Fl c
171Convert the file system to the specified level.
172Note that the level of a file system can only be raised.
173There are currently four levels defined:
174.Bl -tag -width indent
175.It 0
176The file system is in the old (static table) format.
177.It 1
178The file system is in the new (dynamic table) format.
179.It 2
180The file system supports 32-bit uid's and gid's,
181short symbolic links are stored in the inode,
182and directories have an added field showing the file type.
183.It 3
184If maxcontig is greater than one,
185build the free segment maps to aid in finding contiguous sets of blocks.
186If maxcontig is equal to one, delete any existing segment maps.
187.El
188.Pp
189In interactive mode,
190.Nm
191will list the conversion to be made
192and ask whether the conversion should be done.
193If a negative answer is given,
194no further operations are done on the file system.
195In preen mode,
196the conversion is listed and done if
197possible without user interaction.
198Conversion in preen mode is best used when all the file systems
199are being converted at once.
200The format of a file system can be determined from the
201first line of output from
202.Xr dumpfs 8 .
203.Pp
204This option implies the
205.Fl f
206flag.
207.It Fl d
208Enable debugging messages.
209.It Fl E
210Clear unallocated blocks, notifying the underlying device that they
211are not used and that their contents may be discarded.
212This is useful for filesystems which have been mounted on systems
213without TRIM support, or with TRIM support disabled, as well as
214filesystems which have been copied from one device to another.
215.Pp
216See the
217.Fl E
218and
219.Fl t
220flags of
221.Xr newfs 8 ,
222and
223the
224.Fl t
225flag of
226.Xr tunefs 8 .
227.It Fl F
228Determine whether the file system needs to be cleaned immediately
229in foreground, or if its cleaning can be deferred to background.
230To be eligible for background cleaning it must have been running
231with soft updates, not have been marked as needing a foreground check,
232and be mounted and writable when the background check is to be done.
233If these conditions are met, then
234.Nm
235exits with a zero exit status.
236Otherwise it exits with a non-zero exit status.
237If the file system is clean,
238it will exit with a non-zero exit status so that the clean status
239of the file system can be verified and reported during the foreground
240checks.
241Note that when invoked with the
242.Fl F
243flag, no cleanups are done.
244The only thing that
245.Nm
246does is to determine whether a foreground or background
247check is needed and exit with an appropriate status code.
248.It Fl f
249Force
250.Nm
251to check
252.Sq clean
253file systems when preening.
254.It Fl m
255Use the mode specified in octal immediately after the flag as the
256permission bits to use when creating the
257.Pa lost+found
258directory rather than the default 1777.
259In particular, systems that do not wish to have lost files accessible
260by all users on the system should use a more restrictive
261set of permissions such as 700.
262.It Fl n
263Assume a no response to all questions asked by
264.Nm
265except for
266.Ql CONTINUE? ,
267which is assumed to be affirmative;
268do not open the file system for writing.
269.It Fl p
270Preen file systems (see above).
271.It Fl R
272Instruct fsck_ffs to restart itself if it encounters certain errors that
273warrant another run.
274It will limit itself to a maximum of 10 restarts in a given run in order
275to avoid an endless loop with extremely corrupted filesystems.
276.It Fl r
277Free up excess unused inodes.
278Decreasing the number of preallocated inodes reduces the
279running time of future runs of
280.Nm
281and frees up space that can allocated to files.
282The
283.Fl r
284option is ignored when running in preen mode.
285.It Fl S
286Surrender on error.
287With this flag enabled, a hard error returned on disk i/o will cause
288.Nm
289to abort instead of continuing on and possibly tripping over more i/o errors.
290.It Fl y
291Assume a yes response to all questions asked by
292.Nm ;
293this should be used with great caution as this is a free license
294to continue after essentially unlimited trouble has been encountered.
295.It Fl Z
296Similar to
297.Fl E ,
298but overwrites unused blocks with zeroes.
299If both
300.Fl E
301and
302.Fl Z
303are specified, blocks are first zeroed and then erased.
304.El
305.Pp
306Inconsistencies checked are as follows:
307.Pp
308.Bl -enum -compact
309.It
310Blocks claimed by more than one inode or the free map.
311.It
312Blocks claimed by an inode outside the range of the file system.
313.It
314Incorrect link counts.
315.It
316Size checks:
317.Bl -item -offset indent -compact
318.It
319Directory size not a multiple of DIRBLKSIZ.
320.It
321Partially truncated file.
322.El
323.It
324Bad inode format.
325.It
326Blocks not accounted for anywhere.
327.It
328Directory checks:
329.Bl -item -offset indent -compact
330.It
331File pointing to unallocated inode.
332.It
333Inode number out of range.
334.It
335Directories with unallocated blocks (holes).
336.It
337Dot or dot-dot not the first two entries of a directory
338or having the wrong inode number.
339.El
340.It
341Super Block checks:
342.Bl -item -offset indent -compact
343.It
344More blocks for inodes than there are in the file system.
345.It
346Bad free block map format.
347.It
348Total free block and/or free inode count incorrect.
349.El
350.El
351.Pp
352Orphaned files and directories (allocated but unreferenced) are,
353with the operator's concurrence, reconnected by
354placing them in the
355.Pa lost+found
356directory.
357The name assigned is the inode number.
358If the
359.Pa lost+found
360directory does not exist, it is created.
361If there is insufficient space its size is increased.
362.Pp
363The full foreground
364.Nm
365checks for many more problems that may occur after an
366unrecoverable disk write error.
367Thus, it is recommended that you perform foreground
368.Nm
369on your systems periodically and whenever you encounter
370unrecoverable disk write errors or file-system\-related panics.
371.Sh FILES
372.Bl -tag -width /etc/fstab -compact
373.It Pa /etc/fstab
374contains default list of file systems to check.
375.El
376.Sh EXIT STATUS
377.Ex -std
378.Pp
379Specific non-zero exit status values used are:
380.Bl -tag -width indent
381.It 1
382Usage error (missing or invalid command arguments).
383.It 2
384The
385.Fl p
386option was used and a
387.Dv SIGQUIT
388was received, indicating that the system should be returned to single
389user mode after the file system check.
390.It 3
391The file system superblock cannot be read.
392This could indicate that the file system device does not exist or is not yet
393ready.
394.It 4
395A mounted file system was modified; the system should be rebooted.
396.It 5
397The
398.Fl B
399option was used and soft updates are not enabled on the file system.
400.It 6
401The
402.Fl B
403option was used and the kernel lacks needed support.
404.It 7
405The
406.Fl F
407option was used and the file system is clean.
408.It 8
409General error exit.
410.It 16
411The file system could not be completely repaired.
412The file system may be able to be repaired by running
413.Nm
414on the file system again.
415.El
416.Sh DIAGNOSTICS
417The diagnostics produced by
418.Nm
419are fully enumerated and explained in Appendix A of
420.Rs
421.%T "Fsck \- The UNIX File System Check Program"
422.Re
423.Sh SEE ALSO
424.Xr fs 5 ,
425.Xr fstab 5 ,
426.Xr fsck 8 ,
427.Xr fsdb 8 ,
428.Xr newfs 8 ,
429.Xr reboot 8
430.Sh HISTORY
431A
432.Nm fsck
433utility appeared in
434.Bx 4.0 .
435It became
436.Nm
437in
438.Fx 5.0
439with the introduction of the filesystem independent wrapper as
440.Nm fsck .
441