man2/mount.2

   1 .\" Copyright (C) 1993 Rickard E. Faith <faith@cs.unc.edu>
   2 .\" and Copyright (C) 1994 Andries E. Brouwer <aeb@cwi.nl>
   3 .\" and Copyright (C) 2002, 2005 Michael Kerrisk <mtk.manpages@gmail.com>
   4 .\"
   5 .\" %%%LICENSE_START(VERBATIM)
   6 .\" Permission is granted to make and distribute verbatim copies of this
   7 .\" manual provided the copyright notice and this permission notice are
   8 .\" preserved on all copies.
   9 .\"
  10 .\" Permission is granted to copy and distribute modified versions of this
  11 .\" manual under the conditions for verbatim copying, provided that the
  12 .\" entire resulting derived work is distributed under the terms of a
  13 .\" permission notice identical to this one.
  14 .\"
  15 .\" Since the Linux kernel and libraries are constantly changing, this
  16 .\" manual page may be incorrect or out-of-date.  The author(s) assume no
  17 .\" responsibility for errors or omissions, or for damages resulting from
  18 .\" the use of the information contained herein.  The author(s) may not
  19 .\" have taken the same level of care in the production of this manual,
  20 .\" which is licensed free of charge, as they might when working
  21 .\" professionally.
  22 .\"
  23 .\" Formatted or processed versions of this manual, if unaccompanied by
  24 .\" the source, must acknowledge the copyright and authors of this work.
  25 .\" %%%LICENSE_END
  26 .\"
  27 .\" Modified 1996-11-04 by Eric S. Raymond <esr@thyrsus.com>
  28 .\" Modified 2001-10-13 by Michael Kerrisk <mtk.manpages@gmail.com>
  29 .\"     Added note on historical behavior of MS_NOSUID
  30 .\" Modified 2002-05-16 by Michael Kerrisk <mtk.manpages@gmail.com>
  31 .\"     Extensive changes and additions
  32 .\" Modified 2002-05-27 by aeb
  33 .\" Modified 2002-06-11 by Michael Kerrisk <mtk.manpages@gmail.com>
  34 .\"     Enhanced descriptions of MS_MOVE, MS_BIND, and MS_REMOUNT
  35 .\" Modified 2004-06-17 by Michael Kerrisk <mtk.manpages@gmail.com>
  36 .\" 2005-05-18, mtk, Added MNT_EXPIRE, plus a few other tidy-ups.
  37 .\" 2008-10-06, mtk: move umount*() material into separate umount.2 page.
  38 .\" 2008-10-06, mtk: Add discussion of namespaces.
  39 .\"
  40 .TH MOUNT 2 2016-03-15 "Linux" "Linux Programmer's Manual"
  41 .SH NAME
  42 mount \- mount filesystem
  43 .SH SYNOPSIS
  44 .nf
  45 .B "#include <sys/mount.h>"
  46 .sp
  47 .BI "int mount(const char *" source ", const char *" target ,
  48 .BI "          const char *" filesystemtype ", unsigned long " mountflags ,
  49 .BI "          const void *" data );
  50 .fi
  51 .SH DESCRIPTION
  52 .BR mount ()
  53 attaches the filesystem specified by
  54 .I source
  55 (which is often a device name, but can also be a directory name
  56 or a dummy) to the directory specified by
  57 .IR target .
  58
  59 Appropriate privilege (Linux: the
  60 .B CAP_SYS_ADMIN
  61 capability) is required to mount filesystems.
  62
  63 Since Linux 2.4 a single filesystem can be visible at
  64 multiple mount points, and multiple mounts can be stacked
  65 on the same mount point.
  66 .\" Multiple mounts on same mount point: since 2.3.99pre7.
  67
  68 Values for the
  69 .I filesystemtype
  70 argument supported by the kernel are listed in
  71 .I /proc/filesystems
  72 (e.g., "minix", "ext2", "ext3", "jfs", "xfs", "reiserfs",
  73 "msdos", "proc", "nfs", "iso9660").
  74 Further types may become available when the appropriate modules
  75 are loaded.
  76
  77 The
  78 .I mountflags
  79 argument may have the magic number 0xC0ED (\fBMS_MGC_VAL\fP)
  80 in the top 16 bits (this was required in kernel versions prior to 2.4, but
  81 is no longer required and ignored if specified),
  82 and various mount flags
  83 .\" (as defined in \fI<linux/fs.h>\fP for libc4 and libc5
  84 .\" and in \fI<sys/mount.h>\fP for glibc2)
  85 in the low order 16 bits:
  86 .\" FIXME 2.6.15 added flags for "shared subtree" functionality:
  87 .\"
  88 .\"     MS_PRIVATE
  89 .\"             All mounts are private by default. Previously shared mounts
  90 .\"             can be re-marked PRIVATE.
  91 .\"     MS_SHARED
  92 .\"             Mount points that are marked SHARED propagate mount events
  93 .\"             to one another after being cloned.
  94 .\"                     mount --make-rshared ==> MS_SHARED | MS_REC
  95 .\"     MS_SLAVE
  96 .\"             A previously shared mount point can be marked SLAVE, meaning
  97 .\"             it receives propagated events, but does not propagate events.
  98 .\"     MS_UNBINDABLE
  99 .\"             mounts cannot be bound into other places, and will not be
 100 .\"             propagated into new subtrees
 101 .\"
 102 .\" These settings are visible in /proc/$$/mountinfo
 103 .\"
 104 .\" These need to be documented on this page.
 105 .\" See:
 106 .\"
 107 .\" * Documentation/filesystems/sharedsubtree.txt
 108 .\"
 109 .\" * http://lwn.net/Articles/159077/
 110 .\"
 111 .\" * https://www.kernel.org/doc/ols/2006/ols2006v2-pages-209-222.pdf
 112 .\"       Shared-Subtree Concept, Implementation, and Applications in Linux
 113 .\"       Al Viro viro@ftp.linux.org.uk
 114 .\"       Ram Pai linuxram@us.ibm.com
 115 .\"
 116 .\" * http://www.ibm.com/developerworks/linux/library/l-mount-namespaces/index.html
 117 .\"       Applying mount namespaces
 118 .\"
 119 .\"       Uncover practical applications for advanced Linux mounts features
 120 .\"       Serge E. Hallyn (sergeh@us.ibm.com), Software Engineer, IBM
 121 .\"       Ram Pai (linuxram@us.ibm.com), Software Engineer, IBM
 122 .\"       Date:  17 Sep 2007
 123 .\"
 124 .\" FIXME 2.6.25 Added MS_I_VERSION, which needs to be documented.
 125 .\"
 126 .TP
 127 .BR MS_BIND " (Linux 2.4 onward)"
 128 .\" since 2.4.0-test9
 129 Perform a bind mount, making a file or a directory subtree visible at
 130 another point within a filesystem.
 131 Bind mounts may cross filesystem boundaries and span
 132 .BR chroot (2)
 133 jails.
 134 The
 135 .IR filesystemtype
 136 and
 137 .IR data
 138 arguments are ignored.
 139 Up until Linux 2.6.26,
 140 .I mountflags
 141 was also ignored
 142 .\" with the exception of the "hidden" MS_REC mountflags bit
 143 (the bind mount has the same mount options as
 144 the underlying mount point).
 145 .TP
 146 .BR MS_DIRSYNC " (since Linux 2.5.19)"
 147 Make directory changes on this filesystem synchronous.
 148 (This property can be obtained for individual directories
 149 or subtrees using
 150 .BR chattr (1).)
 151 .TP
 152 .BR MS_LAZYTIME " (since Linux 4.0)"
 153 .\" commit 0ae45f63d4ef8d8eeec49c7d8b44a1775fff13e8
 154 .\" commit fe032c422c5ba562ba9c2d316f55e258e03259c6
 155 .\" commit a26f49926da938f47561f386be56a83dd37a496d
 156 Reduce on-disk updates of inode timestamps (atime, mtime, ctime)
 157 by maintaining these changes only in memory.
 158 The on-disk timestamps are updated only when:
 159 .RS
 160 .IP (a) 5
 161 the inode needs to be updated for some change unrelated to file timestamps;
 162 .IP (b)
 163 the application employs
 164 .BR fsync (2),
 165 .BR syncfs (2),
 166 or
 167 .BR sync (2);
 168 .IP (c)
 169 an undeleted inode is evicted from memory; or
 170 .IP (d)
 171 more than 24 hours have passed since the inode was written to disk.
 172 .RE
 173 .IP
 174 This mount option significantly reduces writes
 175 needed to update the inode's timestamps, especially mtime and atime.
 176 However, in the event of a system crash, the atime and mtime fields
 177 on disk might be out of date by up to 24 hours.
 178
 179 Examples of workloads where this option could be of significant benefit
 180 include frequent random writes to preallocated files,
 181 as well as cases where the
 182 .B MS_STRICTATIME
 183 mount option is also enabled.
 184 (The advantage of combining
 185 .BR MS_STRICTATIME
 186 and
 187 .BR MS_LAZYTIME
 188 is that
 189 .BR stat (2)
 190 will return the correctly updated atime, but the atime updates
 191 will be flushed to disk only in the cases listed above.)
 192 .TP
 193 .B MS_MANDLOCK
 194 Permit mandatory locking on files in this filesystem.
 195 (Mandatory locking must still be enabled on a per-file basis,
 196 as described in
 197 .BR fcntl (2).)
 198 Since Linux 4.5,
 199 .\" commit 95ace75414f312f9a7b93d873f386987b92a5301
 200 this mount option requires the
 201 .B CAP_SYS_ADMIN
 202 capability.
 203 .\" FIXME Describe the MS_MOVE flag in more detail
 204 .TP
 205 .B MS_MOVE
 206 Move a subtree.
 207 .I source
 208 specifies an existing mount point and
 209 .I target
 210 specifies the new location.
 211 The move is atomic: at no point is the subtree unmounted.
 212 The
 213 .IR filesystemtype ", " mountflags ", and " data
 214 arguments are ignored.
 215 .TP
 216 .B MS_NOATIME
 217 Do not update access times for (all types of) files on this filesystem.
 218 .TP
 219 .B MS_NODEV
 220 Do not allow access to devices (special files) on this filesystem.
 221 .TP
 222 .B MS_NODIRATIME
 223 Do not update access times for directories on this filesystem.
 224 This flag provides a subset of the functionality provided by
 225 .BR MS_NOATIME ;
 226 that is,
 227 .BR MS_NOATIME
 228 implies
 229 .BR MS_NODIRATIME .
 230 .TP
 231 .B MS_NOEXEC
 232 Do not allow programs to be executed from this filesystem.
 233 .\" (Possibly useful for a filesystem that contains non-Linux executables.
 234 .\" Often used as a security feature, e.g., to make sure that restricted
 235 .\" users cannot execute files uploaded using ftp or so.)
 236 .TP
 237 .B MS_NOSUID
 238 Do not honor set-user-ID and set-group-ID bits when executing
 239 programs from this filesystem.
 240 .\" (This is a security feature to prevent users executing set-user-ID and
 241 .\" set-group-ID programs from removable disk devices.)
 242 .TP
 243 .B MS_RDONLY
 244 Mount filesystem read-only.
 245 .\"
 246 .\" FIXME Document MS_REC, available since 2.4.11.
 247 .\" This flag has meaning in conjunction with MS_BIND and
 248 .\" also with the shared subtree flags.
 249 .TP
 250 .BR MS_RELATIME " (since Linux 2.6.20)"
 251 When a file on this filesystem is accessed,
 252 update the file's last access time (atime) only if the current value
 253 of atime is less than or equal to the file's last modification time (mtime)
 254 or last status change time (ctime).
 255 This option is useful for programs, such as
 256 .BR mutt (1),
 257 that need to know when a file has been read since it was last modified.
 258 Since Linux 2.6.30, the kernel defaults to the behavior provided
 259 by this flag (unless
 260 .BR MS_NOATIME
 261 was specified), and the
 262 .B MS_STRICTATIME
 263 flag is required to obtain traditional semantics.
 264 In addition, since Linux 2.6.30,
 265 the file's last access time is always updated if it
 266 is more than 1 day old.
 267 .\" Matthew Garrett notes in the patch that added this behavior
 268 .\" that this lets utilities such as tmpreaper (which deletes
 269 .\" files based on last access time) work correctly.
 270 .TP
 271 .B MS_REMOUNT
 272 Remount an existing mount.
 273 This allows you to change the
 274 .I mountflags
 275 and
 276 .I data
 277 of an existing mount without having to unmount and remount the filesystem.
 278 .I target
 279 should be the same value specified in the initial
 280 .BR mount ()
 281 call;
 282 .I source
 283 and
 284 .I filesystemtype
 285 are ignored.
 286 The
 287 .I mountflags
 288 and
 289 .I data
 290 arguments should match the values used in the original
 291 .BR mount ()
 292 call, except for those parameters that are being deliberately changed.
 293
 294 The following
 295 .I mountflags
 296 can be changed:
 297 .BR MS_RDONLY ,
 298 .BR MS_SYNCHRONOUS ,
 299 .BR MS_MANDLOCK ;
 300 before kernel 2.6.16, the following could also be changed:
 301 .B MS_NOATIME
 302 and
 303 .BR MS_NODIRATIME ;
 304 and, additionally, before kernel 2.4.10, the following could also be changed:
 305 .BR MS_NOSUID ,
 306 .BR MS_NODEV ,
 307 .BR MS_NOEXEC .
 308 .TP
 309 .BR MS_SILENT " (since Linux 2.6.17)"
 310 Suppress the display of certain
 311 .RI ( printk ())
 312 warning messages in the kernel log.
 313 This flag supersedes the misnamed and obsolete
 314 .BR MS_VERBOSE
 315 flag (available since Linux 2.4.12), which has the same meaning.
 316 .TP
 317 .BR MS_STRICTATIME " (since Linux 2.6.30)"
 318 Always update the last access time (atime) when files on this
 319 filesystem are accessed.
 320 (This was the default behavior before Linux 2.6.30.)
 321 Specifying this flag overrides the effect of setting the
 322 .BR MS_NOATIME
 323 and
 324 .BR MS_RELATIME
 325 flags.
 326 .TP
 327 .B MS_SYNCHRONOUS
 328 Make writes on this filesystem synchronous (as though
 329 the
 330 .B O_SYNC
 331 flag to
 332 .BR open (2)
 333 was specified for all file opens to this filesystem).
 334 .PP
 335 From Linux 2.4 onward, the
 336 .BR MS_NODEV ", " MS_NOEXEC ", and " MS_NOSUID
 337 flags are settable on a per-mount-point basis.
 338 From kernel 2.6.16 onward,
 339 .B MS_NOATIME
 340 and
 341 .B MS_NODIRATIME
 342 are also settable on a per-mount-point basis.
 343 The
 344 .B MS_RELATIME
 345 flag is also settable on a per-mount-point basis.
 346 .PP
 347 The
 348 .I data
 349 argument is interpreted by the different filesystems.
 350 Typically it is a string of comma-separated options
 351 understood by this filesystem.
 352 See
 353 .BR mount (8)
 354 for details of the options available for each filesystem type.
 355 .SH RETURN VALUE
 356 On success, zero is returned.
 357 On error, \-1 is returned, and
 358 .I errno
 359 is set appropriately.
 360 .SH ERRORS
 361 The error values given below result from filesystem type independent
 362 errors.
 363 Each filesystem type may have its own special errors and its
 364 own special behavior.
 365 See the Linux kernel source code for details.
 366 .TP
 367 .B EACCES
 368 A component of a path was not searchable.
 369 (See also
 370 .BR path_resolution (7).)
 371 Or, mounting a read-only filesystem was attempted without giving the
 372 .B MS_RDONLY
 373 flag.
 374 Or, the block device
 375 .I source
 376 is located on a filesystem mounted with the
 377 .B MS_NODEV
 378 option.
 379 .\" mtk: Probably: write permission is required for MS_BIND, with
 380 .\" the error EPERM if not present; CAP_DAC_OVERRIDE is required.
 381 .TP
 382 .B EBUSY
 383 .I source
 384 is already mounted.
 385 Or, it cannot be remounted read-only,
 386 because it still holds files open for writing.
 387 Or, it cannot be mounted on
 388 .I target
 389 because
 390 .I target
 391 is still busy (it is the working directory of some thread,
 392 the mount point of another device, has open files, etc.).
 393 .TP
 394 .B EFAULT
 395 One of the pointer arguments points outside the user address space.
 396 .TP
 397 .B EINVAL
 398 .I source
 399 had an invalid superblock.
 400 Or, a remount
 401 .RB ( MS_REMOUNT )
 402 was attempted, but
 403 .I source
 404 was not already mounted on
 405 .IR target .
 406 Or, a move
 407 .RB ( MS_MOVE )
 408 was attempted, but
 409 .I source
 410 was not a mount point, or was \(aq/\(aq.
 411 .TP
 412 .B ELOOP
 413 Too many links encountered during pathname resolution.
 414 Or, a move was attempted, while
 415 .I target
 416 is a descendant of
 417 .IR source .
 418 .TP
 419 .B EMFILE
 420 (In case no block device is required:)
 421 Table of dummy devices is full.
 422 .TP
 423 .B ENAMETOOLONG
 424 A pathname was longer than
 425 .BR MAXPATHLEN .
 426 .TP
 427 .B ENODEV
 428 .I filesystemtype
 429 not configured in the kernel.
 430 .TP
 431 .B ENOENT
 432 A pathname was empty or had a nonexistent component.
 433 .TP
 434 .B ENOMEM
 435 The kernel could not allocate a free page to copy filenames or data into.
 436 .TP
 437 .B ENOTBLK
 438 .I source
 439 is not a block device (and a device was required).
 440 .TP
 441 .B ENOTDIR
 442 .IR target ,
 443 or a prefix of
 444 .IR source ,
 445 is not a directory.
 446 .TP
 447 .B ENXIO
 448 The major number of the block device
 449 .I source
 450 is out of range.
 451 .TP
 452 .B EPERM
 453 The caller does not have the required privileges.
 454 .SH VERSIONS
 455 The definitions of
 456 .BR MS_DIRSYNC ,
 457 .BR MS_MOVE ,
 458 .BR MS_REC ,
 459 .BR MS_RELATIME ,
 460 and
 461 .BR MS_STRICTATIME
 462 were added to glibc headers in version 2.12.
 463 .\" FIXME . Definitions of the so-far-undocumented MS_UNBINDABLE, MS_PRIVATE,
 464 .\" MS_SHARED, and MS_SLAVE were (also) only added to glibc headers in 2.12.
 465 .SH CONFORMING TO
 466 This function is Linux-specific and should not be used in
 467 programs intended to be portable.
 468 .SH NOTES
 469 The original
 470 .B MS_SYNC
 471 flag was renamed
 472 .B MS_SYNCHRONOUS
 473 in 1.1.69
 474 when a different
 475 .B MS_SYNC
 476 was added to \fI<mman.h>\fP.
 477 .LP
 478 Before Linux 2.4 an attempt to execute a set-user-ID or set-group-ID program
 479 on a filesystem mounted with
 480 .B MS_NOSUID
 481 would fail with
 482 .BR EPERM .
 483 Since Linux 2.4 the set-user-ID and set-group-ID bits are
 484 just silently ignored in this case.
 485 .\" The change is in patch-2.4.0-prerelease.
 486 .SS Per-process namespaces
 487 Starting with kernel 2.4.19, Linux provides
 488 per-process mount namespaces.
 489 A mount namespace is the set of filesystem mounts that
 490 are visible to a process.
 491 Mount-point namespaces can be (and usually are)
 492 shared between multiple processes,
 493 and changes to the namespace (i.e., mounts and unmounts) by one process
 494 are visible to all other processes sharing the same namespace.
 495 (The pre-2.4.19 Linux situation can be considered as one in which
 496 a single namespace was shared by every process on the system.)
 497
 498 A child process created by
 499 .BR fork (2)
 500 shares its parent's mount namespace;
 501 the mount namespace is preserved across an
 502 .BR execve (2).
 503
 504 A process can obtain a private mount namespace if:
 505 it was created using the
 506 .BR clone (2)
 507 .BR CLONE_NEWNS
 508 flag,
 509 in which case its new namespace is initialized to be a
 510 .I copy
 511 of the namespace of the process that called
 512 .BR clone (2);
 513 or it calls
 514 .BR unshare (2)
 515 with the
 516 .BR CLONE_NEWNS
 517 flag,
 518 which causes the caller's mount namespace to obtain a private copy
 519 of the namespace that it was previously sharing with other processes,
 520 so that future mounts and unmounts by the caller are invisible
 521 to other processes (except child processes that the caller
 522 subsequently creates) and vice versa.
 523
 524 The Linux-specific
 525 .I /proc/PID/mounts
 526 file exposes the list of mount points in the mount
 527 namespace of the process with the specified ID; see
 528 .BR proc (5)
 529 for details.
 530 .SH SEE ALSO
 531 .BR umount (2),
 532 .BR namespaces (7),
 533 .BR path_resolution (7),
 534 .BR lsblk (8),
 535 .BR findmnt (8),
 536 .BR mount (8),
 537 .BR umount (8)