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  16 .\" manual page may be incorrect or out-of-date.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