xref: /netbsd-src/sbin/mount/mount.8 (revision 6cd39ddb8550f6fa1bff3fed32053d7f19fd0453)
1.\"	$NetBSD: mount.8,v 1.80 2014/12/06 12:12:17 mbalmer Exp $
2.\"
3.\" Copyright (c) 1980, 1989, 1991, 1993
4.\"	The Regents of the University of California.  All rights reserved.
5.\"
6.\" Redistribution and use in source and binary forms, with or without
7.\" modification, are permitted provided that the following conditions
8.\" are met:
9.\" 1. Redistributions of source code must retain the above copyright
10.\"    notice, this list of conditions and the following disclaimer.
11.\" 2. Redistributions in binary form must reproduce the above copyright
12.\"    notice, this list of conditions and the following disclaimer in the
13.\"    documentation and/or other materials provided with the distribution.
14.\" 3. Neither the name of the University nor the names of its contributors
15.\"    may be used to endorse or promote products derived from this software
16.\"    without specific prior written permission.
17.\"
18.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
19.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
20.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
21.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
22.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
23.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
24.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
25.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
26.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
27.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
28.\" SUCH DAMAGE.
29.\"
30.\"     @(#)mount.8	8.8 (Berkeley) 6/16/94
31.\"
32.Dd December 6, 2014
33.Dt MOUNT 8
34.Os
35.Sh NAME
36.Nm mount
37.Nd mount file systems
38.Sh SYNOPSIS
39.Nm
40.Op Fl Aadfruvw
41.Op Fl t Ar type
42.Nm
43.Op Fl dfruvw
44.Brq Ar special | Ar node
45.Nm
46.Op Fl dfruvw
47.Op Fl o Ar options
48.Op Fl t Ar type
49.Ar special node
50.Sh DESCRIPTION
51The
52.Nm
53command invokes a file system-specific program to prepare and graft the
54.Ar special
55device on to the file system tree at the point
56.Ar node ,
57or to update options for an already-mounted file system.
58.Pp
59The
60.Ar node
61argument is always interpreted as a directory in the name space of
62currently mounted file systems.
63The
64.Ar special
65argument is interpreted in different ways
66by the programs that handle different file system types;
67for example,
68.Xr mount_ffs 8
69interprets it as a device node,
70.Xr mount_null 8
71interprets it as a directory name,
72.Xr mount_nfs 8
73interprets it as reference to a remote host and a directory on that host,
74and
75.Xr mount_tmpfs 8
76ignores it.
77.Pp
78The system maintains a list of currently mounted file systems.
79This list is printed if
80.Nm
81is invoked with no arguments, and with no options that require some
82other behaviour.
83.Pp
84If exactly one of
85.Ar special
86or
87.Ar node
88is provided, then the missing information
89(including the file system type)
90is taken from the
91.Xr fstab 5
92file.
93The provided argument is looked up first in the
94.Dq fs_file ,
95then in the
96.Dq fs_spec
97column.
98If the matching entry in
99.Xr fstab 5
100has the string
101.Dq Li from_mount
102as its
103.Dq fs_spec
104field, the device or remote file system already mounted at
105the location specified by
106.Dq fs_spec
107will be used.
108.\" XXX The above paragraph doesn't address the use of "-u"
109.Pp
110If both
111.Ar special
112and
113.Ar node
114are provided, then
115.Xr fstab 5
116is not used.
117In this case, if the file system type is not specified
118via the
119.Fl t
120flag, then
121.Nm
122may determine the type from the disk label (see
123.Xr disklabel 8 ) .
124In addition, if
125.Ar special
126contains a colon
127.Pq Ql \&:
128or at sign
129.Pq Ql \&@ ,
130then the
131.Li nfs
132type is inferred, but this behaviour is deprecated, and will be
133removed in a future version of
134.Nm .
135.Pp
136In
137.Nx ,
138the file-system mounting policy is dictated by the running security models.
139The default security model may allow unprivileged mounting; see
140.Xr secmodel_suser 9
141and
142.Xr secmodel_extensions 9
143for details.
144.Pp
145The options are as follows:
146.Bl -tag -width indent
147.It Fl A
148Causes
149.Nm
150to try to mount all of the file systems listed in the
151.Xr fstab 5
152file except those for which the
153.Dq noauto
154option is specified.
155.It Fl a
156Similar to the
157.Fl A
158flag, except that if a file system (other than the root file system)
159appears to be already mounted,
160.Nm
161will not try to mount it again.
162.Nm
163assumes that a file system is already mounted if a file system with
164the same type is mounted on the given mount point.
165More stringent checks are not possible because some file system types
166report strange values for the mounted-from device for mounted file
167systems.
168.It Fl d
169Causes everything to be done except for the invocation of
170the file system-specific program.
171This option is useful in conjunction with the
172.Fl v
173flag to determine what the
174.Nm
175command is trying to do.
176.It Fl f
177Forces the revocation of write access when trying to downgrade
178a file system mount status from read-write to read-only.
179.It Fl o
180Options are specified with a
181.Fl o
182flag followed by a comma separated string of options.
183The following options are available:
184.Bl -tag -width nocoredump
185.It Cm async
186All
187.Tn I/O
188to the file system should be done asynchronously.
189In the event of a crash,
190.Em "it is impossible for the system to verify the integrity of data on a file system mounted with this option" .
191You should only use this option if you have an application-specific data
192recovery mechanism, or are willing to recreate the file system from scratch.
193.It Cm noasync
194Clear
195.Cm async
196mode.
197.It Cm discard
198Use DISCARD/TRIM commands if disk and driver support it.
199EXPERIMENTAL!
200.It Cm extattr
201Enable extended attributes, if the filesystem supports them and
202does not enable them by default.
203Currently this is only the case for UFS1.
204.It Cm force
205The same as
206.Fl f ;
207forces the revocation of write access when trying to downgrade
208a file system mount status from read-write to read-only.
209.It Cm getargs
210Retrieves the file system specific mount arguments for the given
211mounted file system and prints them.
212.It Cm hidden
213By setting the
214.Dv MNT_IGNORE
215flag,
216causes the mount point to be excluded from the
217list of file systems shown by default with
218.Xr df 1 .
219.It Cm noatime
220Never update the access time field for files.
221This option is useful for optimizing read performance on file systems
222that are used as news spools.
223.It Cm noauto
224This file system should be skipped when mount is run with the
225.Fl a
226flag.
227.It Cm nocoredump
228Do not allow programs to create crash dumps (core files) on the file system.
229This option can be used to help protect sensitive
230data by keeping core files (which may contain sensitive data)
231from being created on insecure file systems.
232Only core files that would be created by program crashes are
233prevented by use of this flag; the behavior of
234.Xr savecore 8
235is not affected.
236.It Cm nodev
237Do not interpret character or block special devices on the file system.
238This option is useful for a server that has file systems containing
239special devices for architectures other than its own.
240.It Cm nodevmtime
241Do not update modification times on device special files.
242This option is useful on laptops
243or other systems that perform power management.
244.It Cm noexec
245Do not allow execution of any binaries on the mounted file system.
246This option is useful for a server that has file systems containing
247binaries for architectures other than its own.
248.It Cm nosuid
249Do not allow set-user-identifier or set-group-identifier bits to take effect.
250.It Cm port
251(NFS only) Use the specified NFS port.
252.It Cm rdonly
253The same as
254.Fl r ;
255mount the file system read-only (even the super-user may not write it).
256.It Cm reload
257Reload all incore data for a file system.
258This is used mainly after running
259.Xr fsck 8
260on the root file system and finding things to fix.
261The file system must be mounted read-only.
262All cached meta-data are invalidated, superblock and summary information
263is re-read from disk, all cached inactive vnodes and file data are
264invalidated and all inode data are re-read for all active vnodes.
265.It Cm rump
266Instead of running mount_type to mount the file system, run rump_type.
267This uses a userspace server to mount the file system and does not
268require kernel support for the specific file system type.
269See the
270.Fl t
271flag and respective rump_type manual page for more information.
272.It Cm log
273(FFS only)
274Mount the file system with
275.Xr wapbl 4
276meta-data journaling, also known simply as logging.
277It provides rapid metadata updates and eliminates the need to check
278file system consistency after a system outage.
279A file system mounted with
280.Cm log
281can not be mounted with
282.Cm async .
283It requires the
284.Dv WAPBL
285option to be enabled in the running kernel.
286See
287.Xr wapbl 4
288for more information.
289This option requires the
290.Dq UFS2
291.Pq level 4
292superblock layout, which is the default for newly created FFSv1 and
293FFSv2 file systems.
294To update an old file system with an earlier superblock format,
295use the
296.Fl c
297option of
298.Xr fsck_ffs 8 .
299.It Cm symperm
300Recognize permission of symbolic link when reading or traversing link.
301.It Cm sync
302All
303.Tn I/O
304to the file system should be done synchronously.
305This is not equivalent to the normal mode in which only
306metadata is written synchronously.
307.It Cm nosync
308Clear
309.Cm sync
310mode.
311.It Cm union
312Causes the namespace at the mount point to appear as the union
313of the mounted file system root
314(referred to as the
315.Em upper
316layer), and the existing directory
317(referred to as the
318.Em lower
319layer).
320Name lookups will be done in the upper layer first.
321If a name does not exist in the upper layer, then the name
322will be looked up in the lower layer.
323If a name exists in both the upper and lower layers, then only
324the upper instance is accessible.
325Creation of new files is done in the upper layer,
326except in the case of the fdesc file system (see
327.Xr mount_fdesc 8 ) .
328.Pp
329Note that the
330.Cm union
331option can be applied to any type of file system,
332and is fundamentally different from
333.Xr mount_union 8 ,
334which is a particular type of file system.
335Also note that the
336.Cm union
337option affects the file system name space only at the mount point
338itself; it does not apply recursively to subdirectories.
339.It Cm update
340The same as
341.Fl u ;
342indicate that the status of an already mounted file system should be changed.
343.El
344.Pp
345Any additional options specific to a given file system type (see the
346.Fl t
347option) may be passed as a comma separated list; these options are
348distinguished by a leading
349.Dq \&-
350(dash).
351Options that take a value are specified using the syntax -option=value.
352For example, the mount command:
353.Bd -literal -offset indent
354mount -t mfs -o nosuid,-N,-s=32m swap /tmp
355.Ed
356.Pp
357causes
358.Nm
359to execute the equivalent of:
360.Bd -literal -offset indent
361/sbin/mount_mfs -o nosuid -N -s 32m swap /tmp
362.Ed
363.It Fl r
364The file system is to be mounted read-only.
365Mount the file system read-only (even the super-user may not write it).
366The same as the
367.Dq rdonly
368argument to the
369.Fl o
370option.
371.It Fl t Ar type
372The argument following the
373.Fl t
374is used to indicate the file system type.
375The type
376.Ar ffs
377is the default.
378The
379.Fl t
380option can be used to indicate that the actions
381should only be taken on file systems of the specified type.
382More than one type may be specified in a comma separated list.
383The list of file system types can be prefixed with
384.Dq no
385to specify the file system types for which action should
386.Em not
387be taken.
388For example, the
389.Nm
390command:
391.Bd -literal -offset indent
392mount -a -t nonfs,mfs
393.Ed
394.Pp
395mounts all file systems except those of type
396.Tn NFS
397and
398.Tn MFS .
399.Pp
400.Nm
401will attempt to execute a program in
402.Pa /sbin/mount_ Ns Em XXX
403where
404.Em XXX
405is replaced by the type name.
406For example, nfs file systems are mounted by the program
407.Pa /sbin/mount_nfs .
408.It Fl u
409The
410.Fl u
411flag indicates that the status of an already mounted file
412system should be changed.
413Any of the options discussed above (the
414.Fl o
415option)
416may be changed;
417also a file system can be changed from read-only to read-write
418or vice versa.
419An attempt to change from read-write to read-only will fail if any
420files on the file system are currently open for writing unless the
421.Fl f
422flag is also specified.
423The set of options is determined by first extracting the options
424for the file system from the
425.Xr fstab 5
426file, then applying any options specified by the
427.Fl o
428argument,
429and finally applying the
430.Fl r
431or
432.Fl w
433option.
434.It Fl v
435Verbose mode.
436If this flag is specified more than once, then the
437file system-specific mount arguments are printed for the given mounted
438file system.
439.It Fl w
440The file system object is to be read and write.
441.El
442.Pp
443The options specific to the various file system types are
444described in the manual pages for those file systems'
445.Nm mount_XXX
446commands;
447for instance, the options specific to Berkeley
448Fast File System (FFS) are described in the
449.Xr mount_ffs 8
450manual page.
451.Pp
452The particular type of file system in each partition of a disk can
453be found by examining the disk label with the
454.Xr disklabel 8
455command.
456.Sh FILES
457.Bl -tag -width /etc/fstab -compact
458.It Pa /etc/fstab
459file system table
460.El
461.Sh EXAMPLES
462Some useful examples:
463.Bl -hang -offset indent -width "MS-DOS"
464.It Tn CD-ROM
465.br
466mount -t cd9660 -r /dev/cd0a /cdrom
467.It Tn MS-DOS
468.br
469mount -t msdos /dev/fd0a /floppy
470.It Tn NFS
471.br
472mount -t nfs nfs-server-host:/directory/path /mount-point
473.It Tn MFS (32 megabyte)
474.br
475mount -t mfs -o nosuid,-s=32m swap /tmp
476.El
477.Pp
478The
479.Dq noauto
480directive in
481.Pa /etc/fstab
482can be used to make it easy to manually mount and unmount removable
483media using just the mountpoint filename, with an entry like this:
484.Pp
485.Dl /dev/cd0a  /cdrom  cd9660 ro,noauto 0 0
486.Pp
487That would allow a simple command like
488.Qq mount /cdrom
489or
490.Qq umount /cdrom
491for media using the
492.Tn ISO-9660
493file system format in the first
494.Tn CD-ROM
495drive.
496.Sh DIAGNOSTICS
497The error
498.Dq Operation not supported by device
499indicates that the mount for the specified file-system type cannot be
500completed because the kernel lacks support for the said file-system.
501See
502.Xr options 4 .
503.Pp
504The error
505.Dq Operation not permitted
506may indicate that the mount options include privileged options and/or
507don't include options that exclude privileged options.
508One should try using at least
509.Dq nodev
510and
511.Dq nosuid
512in such cases:
513.Bd -literal -offset indent
514mount -t cd9660 -o nodev,nosuid /dev/cd0a /mnt
515.Ed
516.Sh SEE ALSO
517.Xr df 1 ,
518.Xr mount 2 ,
519.Xr options 4 ,
520.Xr wapbl 4 ,
521.Xr fstab 5 ,
522.Xr disklabel 8 ,
523.Xr fsck 8 ,
524.Xr mount_ados 8 ,
525.Xr mount_cd9660 8 ,
526.Xr mount_chfs 8 ,
527.Xr mount_ext2fs 8 ,
528.Xr mount_fdesc 8 ,
529.Xr mount_ffs 8 ,
530.Xr mount_filecore 8 ,
531.Xr mount_kernfs 8 ,
532.Xr mount_lfs 8 ,
533.Xr mount_mfs 8 ,
534.Xr mount_msdos 8 ,
535.Xr mount_nfs 8 ,
536.Xr mount_ntfs 8 ,
537.Xr mount_null 8 ,
538.Xr mount_overlay 8 ,
539.Xr mount_portal 8 ,
540.Xr mount_procfs 8 ,
541.Xr mount_tmpfs 8 ,
542.Xr mount_udf 8 ,
543.Xr mount_umap 8 ,
544.Xr mount_union 8 ,
545.Xr rump_cd9660 8 ,
546.Xr rump_efs 8 ,
547.Xr rump_ext2fs 8 ,
548.Xr rump_ffs 8 ,
549.Xr rump_hfs 8 ,
550.Xr rump_lfs 8 ,
551.Xr rump_msdos 8 ,
552.Xr rump_nfs 8 ,
553.Xr rump_ntfs 8 ,
554.Xr rump_smbfs 8 ,
555.Xr rump_sysvbfs 8 ,
556.Xr rump_tmpfs 8 ,
557.Xr rump_udf 8 ,
558.Xr umount 8
559.Sh HISTORY
560A
561.Nm
562command appeared in
563.At v6 .
564