man2/mount.2

   1 .\" Hey Emacs! This file is -*- nroff -*- source.
   2 .\"
   3 .\" Copyright (C) 1993 Rickard E. Faith <faith@cs.unc.edu>
   4 .\" and Copyright (C) 1994 Andries E. Brouwer <aeb@cwi.nl>
   5 .\" and Copyright (C) 2002, 2005 Michael Kerrisk <mtk-manpages@gmx.net>
   6 .\"
   7 .\" Permission is granted to make and distribute verbatim copies of this
   8 .\" manual provided the copyright notice and this permission notice are
   9 .\" preserved on all copies.
  10 .\"
  11 .\" Permission is granted to copy and distribute modified versions of this
  12 .\" manual under the conditions for verbatim copying, provided that the
  13 .\" entire resulting derived work is distributed under the terms of a
  14 .\" permission notice identical to this one.
  15 .\"
  16 .\" Since the Linux kernel and libraries are constantly changing, this
  17 .\" manual page may be incorrect or out-of-date.  The author(s) assume no
  18 .\" responsibility for errors or omissions, or for damages resulting from
  19 .\" the use of the information contained herein.  The author(s) may not
  20 .\" have taken the same level of care in the production of this manual,
  21 .\" which is licensed free of charge, as they might when working
  22 .\" professionally.
  23 .\"
  24 .\" Formatted or processed versions of this manual, if unaccompanied by
  25 .\" the source, must acknowledge the copyright and authors of this work.
  26 .\"
  27 .\" Modified 1996-11-04 by Eric S. Raymond <esr@thyrsus.com>
  28 .\" Modified 2001-10-13 by Michael Kerrisk <mtk-manpages@gmx.net>
  29 .\"     Added note on historical behaviour of MS_NOSUID
  30 .\" Modified 2002-05-16 by Michael Kerrisk <mtk-manpages@gmx.net>
  31 .\"     Extensive changes and additions
  32 .\" Modified 2002-05-27 by aeb
  33 .\" Modified 2002-06-11 by Michael Kerrisk <mtk-manpages@gmx.net>
  34 .\"     Enhanced descriptions of MS_MOVE, MS_BIND, and MS_REMOUNT
  35 .\" Modified 2004-06-17 by Michael Kerrisk <mtk-manpages@gmx.net>
  36 .\" 2005-05-18, mtk, Added MNT_EXPIRE, plus a few other tidy-ups.
  37 .\"
  38 .TH MOUNT 2 2004-05-18 "Linux 2.6.12" "Linux Programmer's Manual"
  39 .SH NAME
  40 mount, umount \- mount and unmount filesystems
  41 .SH SYNOPSIS
  42 .nf
  43 .B "#include <sys/mount.h>"
  44 .sp
  45 .BI "int mount(const char *" source ", const char *" target ,
  46 .BI "          const char *" filesystemtype ", unsigned long " mountflags ,
  47 .BI "          const void *" data );
  48 .sp
  49 .BI "int umount(const char *" target );
  50 .sp
  51 .BI "int umount2(const char *" target ", int " flags );
  52 .fi
  53 .SH DESCRIPTION
  54 .BR mount ()
  55 attaches the filesystem specified by
  56 .I source
  57 (which is often a device name, but can also be a directory name
  58 or a dummy) to the directory specified by
  59 .IR target .
  60
  61 .BR umount ()
  62 and
  63 .BR umount2 ()
  64 remove the attachment of the (topmost) filesystem mounted on
  65 .IR target .
  66
  67 Appropriate privilege (Linux: the
  68 .B CAP_SYS_ADMIN
  69 capability) is required to mount and unmount filesystems.
  70
  71 Since Linux 2.4 a single filesystem can be visible at
  72 multiple mount points, and multiple mounts can be stacked
  73 on the same mount point.
  74 .\" Multiple mounts on same mount point: since 2.3.99pre7.
  75
  76 Values for the
  77 .IR filesystemtype
  78 argument supported by the kernel are listed in
  79 .I /proc/filesystems
  80 (like "minix", "ext2", "msdos", "proc", "nfs", "iso9660" etc.).
  81 Further types may become available when the appropriate modules
  82 are loaded.
  83
  84 The
  85 .IR mountflags
  86 argument may have the magic number 0xC0ED (\fBMS_MGC_VAL\fP)
  87 in the top 16 bits (this was required in kernel versions prior to 2.4, but
  88 is no longer required and ignored if specified),
  89 and various mount flags (as defined in <linux/fs.h> for libc4 and libc5
  90 and in <sys/mount.h> for glibc2) in the low order 16 bits:
  91 .\" FIXME 2.6.15 added flags for "shared sub-tree" functionality:
  92 .\" MS_UNBINDABLE, MS_PRIVATE, MS_SHARED, MS_SLAVE
  93 .\" These need to be documented on this page.
  94 .\" See Documentation/sharedsubtree.txt
  95 .TP
  96 .B MS_BIND
  97 (Linux 2.4 onwards)
  98 .\" since 2.4.0-test9
  99 Perform a bind mount, making a file or a directory subtree visible at
 100 another point within a file system.
 101 Bind mounts may cross file system boundaries and span
 102 .BR chroot (2)
 103 jails.
 104 The
 105 .IR filesystemtype ", " mountflags ", and " data
 106 arguments are ignored.
 107 .\" with the exception of the "hidden" MS_REC mountflags bit
 108 .TP
 109 .BR MS_DIRSYNC " (since Linux 2.5.19)"
 110 Make directory changes on this file system synchronous.
 111 (This property can be obtained for individual directories
 112 or subtrees using
 113 .BR chattr (8).)
 114 .TP
 115 .B MS_MANDLOCK
 116 Permit mandatory locking on files in this file system.
 117 (Mandatory locking must still be enabled on a per-file basis,
 118 as described in
 119 .BR fcntl (2).)
 120 .\" FIXME Say more about MS_MOVE
 121 .TP
 122 .B MS_MOVE
 123 Move a subtree.
 124 .I source
 125 specifies an existing mount point and
 126 .I target
 127 specifies the new location.
 128 The move is atomic: at no point is the subtree unmounted.
 129 The
 130 .IR filesystemtype ", " mountflags ", and " data
 131 arguments are ignored.
 132 .TP
 133 .B MS_NOATIME
 134 Do not update access times for (all types of) files on this file system.
 135 .TP
 136 .B MS_NODEV
 137 Do not allow access to devices (special files) on this file system.
 138 .TP
 139 .B MS_NODIRATIME
 140 Do not update access times for directories on this file system.
 141 .TP
 142 .B MS_NOEXEC
 143 Do not allow programs to be executed from this file system.
 144 .\" (Possibly useful for a file system that contains non-Linux executables.
 145 .\" Often used as a security feature, e.g. to make sure that restricted
 146 .\" users cannot execute files uploaded using ftp or so.)
 147 .TP
 148 .B MS_NOSUID
 149 Do not honour set-user-ID and set-group-ID bits when executing
 150 programs from this file system.
 151 .\" (This is a security feature to prevent users executing set-user-ID and
 152 .\" set-group-ID programs from removable disk devices.)
 153 .TP
 154 .B MS_RDONLY
 155 Mount file system read-only.
 156 .\"
 157 .\" FIXME Document MS_REC, available since 2.4.11.
 158 .\" This flag has meaning in conjunction with MS_BIND and
 159 .\" also with the shared sub-tree flags.
 160 .TP
 161 .B MS_REMOUNT
 162 Remount an existing mount.  This is allows you to change the
 163 .I mountflags
 164 and
 165 .I data
 166 of an existing mount without having to unmount and remount the file system.
 167 .I source
 168 and
 169 .I target
 170 should be the same values specified in the initial
 171 .BR mount ()
 172 call;
 173 .I filesystemtype
 174 is ignored.
 175
 176 The following
 177 .I mountflags
 178 can be changed:
 179 .BR MS_RDONLY ,
 180 .BR MS_SYNCHRONOUS ,
 181 .BR MS_MANDLOCK ;
 182 before kernel 2.6.16, the following could also be changed:
 183 .BR MS_NOATIME
 184 and
 185 .BR MS_NODIRATIME;
 186 and, additionally, before kernel 2.4, the following could also be changed:
 187 .BR MS_NOSUID ,
 188 .BR MS_NODEV ,
 189 .BR MS_NOEXEC .
 190 .TP
 191 .B MS_SYNCHRONOUS
 192 Make writes on this file system synchronous (as though
 193 the
 194 .B O_SYNC
 195 flag to
 196 .BR open (2)
 197 was specified for all file opens to this file system).
 198 .PP
 199 From Linux 2.4 onwards, the
 200 .BR MS_NODEV ", " MS_NOEXEC ", and " MS_NOSUID
 201 flags are settable on a per-mount-point basis.
 202 From kernel 2.6.16 onwards,
 203 .B MS_NOATIME
 204 and
 205 .B MS_NODIRATIME
 206 are also settable on a per-mount-point basis.
 207 .PP
 208 The
 209 .IR data
 210 argument is interpreted by the different file systems.
 211 Typically it is a string of comma-separated options
 212 understood by this file system.
 213 See
 214 .BR mount (8)
 215 for details of the options available for each filesystem type.
 216 .PP
 217 .\" Note: the kernel naming differs from the glibc naming
 218 .\" umount2 is the glibc name for what the kernel now calls umount
 219 .\" and umount is the glibc name for oldumount
 220 Linux 2.1.116 added the
 221 .BR umount2 ()
 222 system call, which, like
 223 .BR umount (),
 224 unmounts a target, but allows additional
 225 .I flags
 226 controlling the behaviour of the operation:
 227 .TP
 228 .BR MNT_FORCE " (since Linux 2.1.116)"
 229 Force unmount even if busy.
 230 (Only for NFS mounts.)
 231 .\" FIXME Can MNT_FORCE result in data loss?  According to
 232 .\" the Solaris manual page it can cause data loss on Solaris.
 233 .\" If the same holds on Linux, then this should be documented.
 234 .TP
 235 .BR MNT_DETACH " (since Linux 2.4.11)"
 236 Perform a lazy unmount: make the mount point unavailable for
 237 new accesses, and actually perform the unmount when the mount point
 238 ceases to be busy.
 239 .TP
 240 .BR MNT_EXPIRE " (since Linux 2.6.8)"
 241 Mark the mount point as expired.
 242 If a mount point is not currently in use, then an initial call to
 243 .BR umount2 ()
 244 with this flag fails with the error
 245 .BR EAGAIN ,
 246 but marks the mount point as expired.
 247 The mount point remains expired as long as it isn't accessed
 248 by any process.
 249 A second
 250 .BR umount2 ()
 251 call specifying
 252 .B MNT_EXPIRE
 253 unmounts an expired mount point.
 254 This flag cannot be specified with either
 255 .B MNT_FORCE
 256 or
 257 .BR MNT_DETACH .
 258 .SH "RETURN VALUE"
 259 On success, zero is returned.  On error, \-1 is returned, and
 260 .I errno
 261 is set appropriately.
 262 .SH ERRORS
 263 The error values given below result from filesystem type independent
 264 errors. Each filesystem type may have its own special errors and its
 265 own special behavior.  See the kernel source code for details.
 266
 267 .TP
 268 .B EACCES
 269 A component of a path was not searchable. (See also
 270 .BR path_resolution (2).)
 271 Or, mounting a read-only filesystem was attempted without giving the
 272 .B MS_RDONLY
 273 flag.
 274 Or, the block device
 275 .I source
 276 is located on a filesystem mounted with the
 277 .B MS_NODEV
 278 option.
 279 .\" mtk: Probably: write permission is required for MS_BIND, with
 280 .\" the error EPERM if not present; CAP_DAC_OVERRIDE is required.
 281 .TP
 282 .B EAGAIN
 283 A call to
 284 .BR umount2 ()
 285 specifying
 286 .B MNT_EXPIRE
 287 successfully marked an unbusy file system as expired.
 288 .TP
 289 .B EBUSY
 290 .I source
 291 is already mounted. Or, it cannot be remounted read-only,
 292 because it still holds files open for writing.
 293 Or, it cannot be mounted on
 294 .I target
 295 because
 296 .I target
 297 is still busy (it is the working directory of some task,
 298 the mount point of another device, has open files, etc.).
 299 Or, it could not be unmounted because it is busy.
 300 .TP
 301 .B EFAULT
 302 One of the pointer arguments points outside the user address space.
 303 .TP
 304 .B EINVAL
 305 .I source
 306 had an invalid superblock.
 307 Or, a remount
 308 .RB ( MS_REMOUNT )
 309 was attempted, but
 310 .I source
 311 was not already mounted on
 312 .IR target .
 313 Or, a move
 314 .RB ( MS_MOVE )
 315 was attempted, but
 316 .I source
 317 was not a mount point, or was '/'.
 318 Or, an unmount was attempted, but
 319 .I target
 320 was not a mount point.
 321 Or,
 322 .BR umount2 ()
 323 was called with
 324 .B MNT_EXPIRE
 325 and either
 326 .B MNT_DETACH
 327 or
 328 .BR MNT_FORCE .
 329 .TP
 330 .B ELOOP
 331 Too many link encountered during pathname resolution.
 332 Or, a move was attempted, while
 333 .I target
 334 is a descendant of
 335 .IR source .
 336 .TP
 337 .B EMFILE
 338 (In case no block device is required:)
 339 Table of dummy devices is full.
 340 .TP
 341 .B ENAMETOOLONG
 342 A pathname was longer than MAXPATHLEN.
 343 .TP
 344 .B ENODEV
 345 .I filesystemtype
 346 not configured in the kernel.
 347 .TP
 348 .B ENOENT
 349 A pathname was empty or had a nonexistent component.
 350 .TP
 351 .B ENOMEM
 352 The kernel could not allocate a free page to copy filenames or data into.
 353 .TP
 354 .B ENOTBLK
 355 .I source
 356 is not a block device (and a device was required).
 357 .TP
 358 .B ENOTDIR
 359 The second argument, or a prefix of the first argument, is not
 360 a directory.
 361 .TP
 362 .B ENXIO
 363 The major number of the block device
 364 .I source
 365 is out of range.
 366 .TP
 367 .B EPERM
 368 The caller does not have the required privileges.
 369 .SH "CONFORMING TO"
 370 These functions are Linux-specific and should not be used in
 371 programs intended to be portable.
 372 .SH HISTORY
 373 The original
 374 .BR umount ()
 375 function was called as \fIumount(device)\fP and would return ENOTBLK
 376 when called with something other than a block device.
 377 In Linux 0.98p4 a call \fIumount(dir)\fP was added, in order to
 378 support anonymous devices.
 379 In Linux 2.3.99-pre7 the call \fIumount(device)\fP was removed,
 380 leaving only \fIumount(dir)\fP (since now devices can be mounted
 381 in more than one place, so specifying the device does not suffice).
 382 .LP
 383 The original MS_SYNC flag was renamed MS_SYNCHRONOUS in 1.1.69
 384 when a different MS_SYNC was added to <mman.h>.
 385 .LP
 386 Before Linux 2.4 an attempt to execute a set-user-ID or set-group-ID program
 387 on a filesystem mounted with
 388 .B MS_NOSUID
 389 would fail with
 390 .BR EPERM .
 391 Since Linux 2.4 the set-user-ID and set-group-ID bits are
 392 just silently ignored in this case.
 393 .\" The change is in patch-2.4.0-prerelease.
 394 .SH "SEE ALSO"
 395 .BR path_resolution (2),
 396 .BR mount (8),
 397 .BR umount (8)