1MOUNT(8) System Administration MOUNT(8)
2
6 mount - mount a filesystem
7
9 mount [-l|-h|-V]
10 mount -a [-fFnrsvw] [-t fstype] [-O optlist]
12 mount [-fnrsvw] [-o options] device|dir
14 mount [-fnrsvw] [-t fstype] [-o options] device dir
16
18 All files accessible in a Unix system are arranged in one big tree, the
19 file hierarchy, rooted at /. These files can be spread out over sev‐
20 eral devices. The mount command serves to attach the filesystem found
21 on some device to the big file tree. Conversely, the umount(8) command
22 will detach it again. The filesystem is used to control how data is
23 stored on the device or provided in a virtual way by network or another
24 services.
25 The standard form of the mount command is:
27 mount -t type device dir
29 This tells the kernel to attach the filesystem found on device (which
31 is of type type) at the directory dir. The option -t type is optional.
32 The mount command is usually able to detect a filesystem. The root
33 permissions are necessary to mount a filesystem by default. See sec‐
34 tion "Non-superuser mounts" below for more details. The previous con‐
35 tents (if any) and owner and mode of dir become invisible, and as long
36 as this filesystem remains mounted, the pathname dir refers to the root
37 of the filesystem on device.
38 If only the directory or the device is given, for example:
40 mount /dir
42 then mount looks for a mountpoint (and if not found then for a device)
44 in the /etc/fstab file. It's possible to use the --target or --source
45 options to avoid ambivalent interpretation of the given argument. For
46 example:
47 mount --target /mountpoint
49 The same filesystem may be mounted more than once, and in some cases
52 (e.g., network filesystems) the same filesystem may be mounted on the
53 same mountpoint more times. The mount command does not implement any
54 policy to control this behavior. All behavior is controlled by the ker‐
55 nel and it is usually specific to the filesystem driver. The exception
56 is --all, in this case already mounted filesystems are ignored (see
57 --all below for more details).
58 Listing the mounts
61 The listing mode is maintained for backward compatibility only.
62 For more robust and customizable output use findmnt(8), especially in
64 your scripts. Note that control characters in the mountpoint name are
65 replaced with '?'.
66 The following command lists all mounted filesystems (of type type):
68 mount [-l] [-t type]
70 The option -l adds labels to this listing. See below.
72 Indicating the device and filesystem
75 Most devices are indicated by a filename (of a block special device),
76 like /dev/sda1, but there are other possibilities. For example, in the
77 case of an NFS mount, device may look like knuth.cwi.nl:/dir. It is
78 also possible to indicate a block special device using its filesystem
79 label or UUID (see the -L and -U options below), or its partition label
80 or UUID. Partition identifiers are supported for example for GUID Par‐
81 tition Tables (GPT).
82 The device names of disk partitions are unstable; hardware reconfigura‐
84 tion, adding or removing a device can cause changes in names. This is
85 the reason why it's strongly recommended to use filesystem or partition
86 identifiers like UUID or LABEL.
87 The command lsblk --fs provides an overview of filesystems, LABELs and
89 UUIDs on available block devices. The command blkid -p <device> pro‐
90 vides details about a filesystem on the specified device.
91 Don't forget that there is no guarantee that UUIDs and labels are
93 really unique, especially if you move, share or copy the device. Use
94 lsblk -o +UUID,PARTUUID to verify that the UUIDs are really unique in
95 your system.
96 The recommended setup is to use tags (e.g. UUID=uuid) rather than
98 /dev/disk/by-{label,uuid,partuuid,partlabel} udev symlinks in the
99 /etc/fstab file. Tags are more readable, robust and portable. The
100 mount(8) command internally uses udev symlinks, so the use of symlinks
101 in /etc/fstab has no advantage over tags. For more details see lib‐
102 blkid(3).
103 Note that mount(8) uses UUIDs as strings. The UUIDs from the command
105 line or from fstab(5) are not converted to internal binary representa‐
106 tion. The string representation of the UUID should be based on lower
107 case characters.
108 The proc filesystem is not associated with a special device, and when
110 mounting it, an arbitrary keyword, such as proc can be used instead of
111 a device specification. (The customary choice none is less fortunate:
112 the error message `none already mounted' from mount can be confusing.)
113 The files /etc/fstab, /etc/mtab and /proc/mounts
116 The file /etc/fstab (see fstab(5)), may contain lines describing what
117 devices are usually mounted where, using which options. The default
118 location of the fstab(5) file can be overridden with the --fstab path
119 command-line option (see below for more details).
120 The command
122 mount -a [-t type] [-O optlist]
124 (usually given in a bootscript) causes all filesystems mentioned in
126 fstab (of the proper type and/or having or not having the proper
127 options) to be mounted as indicated, except for those whose line con‐
128 tains the noauto keyword. Adding the -F option will make mount fork,
129 so that the filesystems are mounted simultaneously.
130 When mounting a filesystem mentioned in fstab or mtab, it suffices to
132 specify on the command line only the device, or only the mount point.
133 The programs mount and umount traditionally maintained a list of cur‐
135 rently mounted filesystems in the file /etc/mtab. The support for reg‐
136 ular classic /etc/mtab is completely disabled in compile time by
137 default, because on current Linux systems it is better to make it a
138 symlink to /proc/mounts instead. The regular mtab file maintained in
139 userspace cannot reliably work with namespaces, containers and other
140 advanced Linux features. If the regular mtab support is enabled than
141 it's possible to use the file as well as the symlink.
142 If no arguments are given to mount, the list of mounted filesystems is
144 printed.
145 If you want to override mount options from /etc/fstab you have to use
147 the -o option:
148 mount device|dir -o options
150 and then the mount options from the command line will be appended to
152 the list of options from /etc/fstab. This default behaviour is possi‐
153 ble to change by command line option --options-mode. The usual behav‐
154 ior is that the last option wins if there are conflicting ones.
155 The mount program does not read the /etc/fstab file if both device (or
157 LABEL, UUID, PARTUUID or PARTLABEL) and dir are specified. For exam‐
158 ple, to mount device foo at /dir:
159 mount /dev/foo /dir
161 This default behaviour is possible to change by command line option
163 --options-source-force to always read configuration from fstab. For
164 non-root users mount always read fstab configuration.
165 Non-superuser mounts
168 Normally, only the superuser can mount filesystems. However, when
169 fstab contains the user option on a line, anybody can mount the corre‐
170 sponding filesystem.
171 Thus, given a line
173 /dev/cdrom /cd iso9660 ro,user,noauto,unhide
175 any user can mount the iso9660 filesystem found on an inserted CDROM
177 using the command:
178 mount /cd
179 Note that mount is very strict about non-root users and all paths spec‐
181 ified on command line are verified before fstab is parsed or a helper
182 program is executed. It's strongly recommended to use a valid mount‐
183 point to specify filesystem, otherwise mount may fail. For example it's
184 bad idea to use NFS or CIFS source on command line.
185 Since version 2.35 mount command does not exit when user permissions
187 are inadequate by internal libmount security rules. It drops suid per‐
188 missions and continue as regular non-root user. It allows to support
189 use-cases where root permissions are not necessary (e.g., fuse filesys‐
190 tems, user namespaces, etc).
191 For more details, see fstab(5). Only the user that mounted a filesys‐
193 tem can unmount it again. If any user should be able to unmount it,
194 then use users instead of user in the fstab line. The owner option is
195 similar to the user option, with the restriction that the user must be
196 the owner of the special file. This may be useful e.g. for /dev/fd if
197 a login script makes the console user owner of this device. The group
198 option is similar, with the restriction that the user must be member of
199 the group of the special file.
200 Bind mount operation
203 Remount part of the file hierarchy somewhere else. The call is:
204 mount --bind olddir newdir
206 or by using this fstab entry:
208 /olddir /newdir none bind
210 After this call the same contents are accessible in two places.
212 It is important to understand that "bind" does not to create any sec‐
214 ond-class or special node in the kernel VFS. The "bind" is just another
215 operation to attach a filesystem. There is nowhere stored information
216 that the filesystem has been attached by "bind" operation. The olddir
217 and newdir are independent and the olddir may be umounted.
218 One can also remount a single file (on a single file). It's also pos‐
220 sible to use the bind mount to create a mountpoint from a regular
221 directory, for example:
222 mount --bind foo foo
224 The bind mount call attaches only (part of) a single filesystem, not
226 possible submounts. The entire file hierarchy including submounts is
227 attached a second place by using:
228 mount --rbind olddir newdir
230 Note that the filesystem mount options maintained by kernel will remain
232 the same as those on the original mount point. The userspace mount
233 options (e.g., _netdev) will not be copied by mount(8) and it's neces‐
234 sary explicitly specify the options on mount command line.
235 mount(8) since v2.27 allows to change the mount options by passing the
237 relevant options along with --bind. For example:
238 mount -o bind,ro foo foo
240 This feature is not supported by the Linux kernel; it is implemented in
242 userspace by an additional mount(2) remounting system call. This solu‐
243 tion is not atomic.
244 The alternative (classic) way to create a read-only bind mount is to
246 use the remount operation, for example:
247 mount --bind olddir newdir
249 mount -o remount,bind,ro olddir newdir
250 Note that a read-only bind will create a read-only mountpoint (VFS
252 entry), but the original filesystem superblock will still be writable,
253 meaning that the olddir will be writable, but the newdir will be read-
254 only.
255 It's also possible to change nosuid, nodev, noexec, noatime, nodiratime
257 and relatime VFS entry flags by "remount,bind" operation. The another
258 (for example filesystem specific flags) are silently ignored. It's
259 impossible to change mount options recursively (for example with -o
260 rbind,ro).
261 mount(8) since v2.31 ignores the bind flag from /etc/fstab on remount
263 operation (if "-o remount" specified on command line). This is neces‐
264 sary to fully control mount options on remount by command line. In the
265 previous versions the bind flag has been always applied and it was
266 impossible to re-define mount options without interaction with the bind
267 semantic. This mount(8) behavior does not affect situations when
268 "remount,bind" is specified in the /etc/fstab file.
269 The move operation
272 Move a mounted tree to another place (atomically). The call is:
273 mount --move olddir newdir
275 This will cause the contents which previously appeared under olddir to
277 now be accessible under newdir. The physical location of the files is
278 not changed. Note that olddir has to be a mountpoint.
279 Note also that moving a mount residing under a shared mount is invalid
281 and unsupported. Use findmnt -o TARGET,PROPAGATION to see the current
282 propagation flags.
283 Shared subtree operations
286 Since Linux 2.6.15 it is possible to mark a mount and its submounts as
287 shared, private, slave or unbindable. A shared mount provides the
288 ability to create mirrors of that mount such that mounts and unmounts
289 within any of the mirrors propagate to the other mirror. A slave mount
290 receives propagation from its master, but not vice versa. A private
291 mount carries no propagation abilities. An unbindable mount is a pri‐
292 vate mount which cannot be cloned through a bind operation. The
293 detailed semantics are documented in Documentation/filesystems/shared‐
294 subtree.txt file in the kernel source tree.
295 Supported operations are:
297 mount --make-shared mountpoint
299 mount --make-slave mountpoint
300 mount --make-private mountpoint
301 mount --make-unbindable mountpoint
302 The following commands allow one to recursively change the type of all
304 the mounts under a given mountpoint.
305 mount --make-rshared mountpoint
307 mount --make-rslave mountpoint
308 mount --make-rprivate mountpoint
309 mount --make-runbindable mountpoint
310 mount(8) does not read fstab(5) when a --make-* operation is requested.
312 All necessary information has to be specified on the command line.
313 Note that the Linux kernel does not allow to change multiple propaga‐
315 tion flags with a single mount(2) system call, and the flags cannot be
316 mixed with other mount options and operations.
317 Since util-linux 2.23 the mount command allows to do more propagation
319 (topology) changes by one mount(8) call and do it also together with
320 other mount operations. This feature is EXPERIMENTAL. The propagation
321 flags are applied by additional mount(2) system calls when the preced‐
322 ing mount operations were successful. Note that this use case is not
323 atomic. It is possible to specify the propagation flags in fstab(5) as
324 mount options (private, slave, shared, unbindable, rprivate, rslave,
325 rshared, runbindable).
326 For example:
328 mount --make-private --make-unbindable /dev/sda1 /foo
330 is the same as:
332 mount /dev/sda1 /foox
334 mount --make-private /foo
335 mount --make-unbindable /foo
336
339 The full set of mount options used by an invocation of mount is deter‐
340 mined by first extracting the mount options for the filesystem from the
341 fstab table, then applying any options specified by the -o argument,
342 and finally applying a -r or -w option, when present.
343 The command mount does not pass all command-line options to the
345 /sbin/mount.suffix mount helpers. The interface between mount and the
346 mount helpers is described below in the section EXTERNAL HELPERS.
347 Command-line options available for the mount command are:
349 -a, --all
351 Mount all filesystems (of the given types) mentioned in fstab
352 (except for those whose line contains the noauto keyword). The
353 filesystems are mounted following their order in fstab. The
354 mount command compares filesystem source, target (and fs root
355 for bind mount or btrfs) to detect already mounted filesystems.
356 The kernel table with already mounted filesystems is cached dur‐
357 ing mount --all. It means that all duplicated fstab entries will
358 be mounted.
359 The option --all is possible to use for remount operation too.
361 In this case all filters (-t and -O) are applied to the table of
362 already mounted filesystems.
363 Since version 2.35 is possible to use the command line option -o
365 to alter mount options from fstab (see also --options-mode).
366 Note that it is a bad practice to use mount -a for fstab check‐
368 ing. The recommended solution is findmnt --verify.
369 -B, --bind
371 Remount a subtree somewhere else (so that its contents are
372 available in both places). See above, under Bind mounts.
373 -c, --no-canonicalize
375 Don't canonicalize paths. The mount command canonicalizes all
376 paths (from command line or fstab) by default. This option can
377 be used together with the -f flag for already canonicalized
378 absolute paths. The option is designed for mount helpers which
379 call mount -i. It is strongly recommended to not use this com‐
380 mand-line option for normal mount operations.
381 Note that mount(8) does not pass this option to the
383 /sbin/mount.type helpers.
384 -F, --fork
386 (Used in conjunction with -a.) Fork off a new incarnation of
387 mount for each device. This will do the mounts on different
388 devices or different NFS servers in parallel. This has the
389 advantage that it is faster; also NFS timeouts go in parallel.
390 A disadvantage is that the mounts are done in undefined order.
391 Thus, you cannot use this option if you want to mount both /usr
392 and /usr/spool.
393 -f, --fake
395 Causes everything to be done except for the actual system call;
396 if it's not obvious, this ``fakes'' mounting the filesystem.
397 This option is useful in conjunction with the -v flag to deter‐
398 mine what the mount command is trying to do. It can also be
399 used to add entries for devices that were mounted earlier with
400 the -n option. The -f option checks for an existing record in
401 /etc/mtab and fails when the record already exists (with a regu‐
402 lar non-fake mount, this check is done by the kernel).
403 -i, --internal-only
405 Don't call the /sbin/mount.filesystem helper even if it exists.
406 -L, --label label
408 Mount the partition that has the specified label.
409 -l, --show-labels
411 Add the labels in the mount output. mount must have permission
412 to read the disk device (e.g. be set-user-ID root) for this to
413 work. One can set such a label for ext2, ext3 or ext4 using the
414 e2label(8) utility, or for XFS using xfs_admin(8), or for reis‐
415 erfs using reiserfstune(8).
416 -M, --move
418 Move a subtree to some other place. See above, the subsection
419 The move operation.
420 -n, --no-mtab
422 Mount without writing in /etc/mtab. This is necessary for exam‐
423 ple when /etc is on a read-only filesystem.
424 -N, --namespace ns
426 Perform mount in namespace specified by ns. ns is either PID of
427 process running in that namespace or special file representing
428 that namespace.
429 mount(8) switches to the namespace when it reads /etc/fstab,
431 writes /etc/mtab (or writes to /run/mount) and calls mount(2)
432 system call, otherwise it runs in the original namespace. It
433 means that the target namespace does not have to contain any
434 libraries or another requirements necessary to execute mount(2)
435 command.
436 See namespaces(7) for more information.
438 -O, --test-opts opts
440 Limit the set of filesystems to which the -a option applies. In
441 this regard it is like the -t option except that -O is useless
442 without -a. For example, the command:
443 mount -a -O no_netdev
445 mounts all filesystems except those which have the option _net‐
447 dev specified in the options field in the /etc/fstab file.
448 It is different from -t in that each option is matched exactly;
450 a leading no at the beginning of one option does not negate the
451 rest.
452 The -t and -O options are cumulative in effect; that is, the
454 command
455 mount -a -t ext2 -O _netdev
457 mounts all ext2 filesystems with the _netdev option, not all
459 filesystems that are either ext2 or have the _netdev option
460 specified.
461 -o, --options opts
463 Use the specified mount options. The opts argument is a comma-
464 separated list. For example:
465 mount LABEL=mydisk -o noatime,nodev,nosuid
467 For more details, see the FILESYSTEM-INDEPENDENT MOUNT OPTIONS
470 and FILESYSTEM-SPECIFIC MOUNT OPTIONS sections.
471 --options-mode mode
474 Controls how to combine options from fstab/mtab with options
475 from command line. mode can be one of ignore, append, prepend
476 or replace. For example append means that options from fstab
477 are appended to options from command line. Default value is
478 prepend -- it means command line options are evaluated after
479 fstab options. Note that the last option wins if there are con‐
480 flicting ones.
481 --options-source source
484 Source of default options. source is comma separated list of
485 fstab, mtab and disable. disable disables fstab and mtab and
486 disables --options-source-force. Default value is fstab,mtab.
487 --options-source-force
490 Use options from fstab/mtab even if both device and dir are
491 specified.
492 -R, --rbind
495 Remount a subtree and all possible submounts somewhere else (so
496 that its contents are available in both places). See above, the
497 subsection Bind mounts.
498 -r, --read-only
500 Mount the filesystem read-only. A synonym is -o ro.
501 Note that, depending on the filesystem type, state and kernel
503 behavior, the system may still write to the device. For exam‐
504 ple, ext3 and ext4 will replay the journal if the filesystem is
505 dirty. To prevent this kind of write access, you may want to
506 mount an ext3 or ext4 filesystem with the ro,noload mount
507 options or set the block device itself to read-only mode, see
508 the blockdev(8) command.
509 -s Tolerate sloppy mount options rather than failing. This will
511 ignore mount options not supported by a filesystem type. Not
512 all filesystems support this option. Currently it's supported
513 by the mount.nfs mount helper only.
514 --source device
516 If only one argument for the mount command is given then the
517 argument might be interpreted as target (mountpoint) or source
518 (device). This option allows to explicitly define that the
519 argument is the mount source.
520 --target directory
522 If only one argument for the mount command is given then the
523 argument might be interpreted as target (mountpoint) or source
524 (device). This option allows to explicitly define that the
525 argument is the mount target.
526 --target-prefix directory
528 Prepend specified directory to all mount targets. This option
529 allows to follow fstab, but mount operations is done on another
530 place, for example:
531 mount --all --target-prefix /chroot -o X-mount.mkdir
533 mounts all from system fstab to /chroot, all missing muontpoint
535 are created (due to X-mount.mkdir). See also --fstab to use an
536 alternative fstab.
537 -T, --fstab path
539 Specifies an alternative fstab file. If path is a directory
540 then the files in the directory are sorted by strverscmp(3);
541 files that start with "." or without an .fstab extension are
542 ignored. The option can be specified more than once. This
543 option is mostly designed for initramfs or chroot scripts where
544 additional configuration is specified beyond standard system
545 configuration.
546 Note that mount(8) does not pass the option --fstab to the
548 /sbin/mount.type helpers, meaning that the alternative fstab
549 files will be invisible for the helpers. This is no problem for
550 normal mounts, but user (non-root) mounts always require fstab
551 to verify the user's rights.
552 -t, --types fstype
554 The argument following the -t is used to indicate the filesystem
555 type. The filesystem types which are currently supported depend
556 on the running kernel. See /proc/filesystems and /lib/mod‐
557 ules/$(uname -r)/kernel/fs for a complete list of the filesys‐
558 tems. The most common are ext2, ext3, ext4, xfs, btrfs, vfat,
559 sysfs, proc, nfs and cifs.
560 The programs mount and umount support filesystem subtypes. The
562 subtype is defined by a '.subtype' suffix. For example
563 'fuse.sshfs'. It's recommended to use subtype notation rather
564 than add any prefix to the mount source (for example
565 'sshfs#example.com' is deprecated).
566 If no -t option is given, or if the auto type is specified,
568 mount will try to guess the desired type. Mount uses the blkid
569 library for guessing the filesystem type; if that does not turn
570 up anything that looks familiar, mount will try to read the file
571 /etc/filesystems, or, if that does not exist, /proc/filesystems.
572 All of the filesystem types listed there will be tried, except
573 for those that are labeled "nodev" (e.g. devpts, proc and nfs).
574 If /etc/filesystems ends in a line with a single *, mount will
575 read /proc/filesystems afterwards. While trying, all filesystem
576 types will be mounted with the mount option silent.
577 The auto type may be useful for user-mounted floppies. Creating
579 a file /etc/filesystems can be useful to change the probe order
580 (e.g., to try vfat before msdos or ext3 before ext2) or if you
581 use a kernel module autoloader.
582 More than one type may be specified in a comma-separated list,
584 for option -t as well as in an /etc/fstab entry. The list of
585 filesystem types for option -t can be prefixed with no to spec‐
586 ify the filesystem types on which no action should be taken.
587 The prefix no has no effect when specified in an /etc/fstab
588 entry.
589 The prefix no can be meaningful with the -a option. For exam‐
591 ple, the command
592 mount -a -t nomsdos,smbfs
594 mounts all filesystems except those of type msdos and smbfs.
596 For most types all the mount program has to do is issue a simple
598 mount(2) system call, and no detailed knowledge of the filesys‐
599 tem type is required. For a few types however (like nfs, nfs4,
600 cifs, smbfs, ncpfs) an ad hoc code is necessary. The nfs, nfs4,
601 cifs, smbfs, and ncpfs filesystems have a separate mount pro‐
602 gram. In order to make it possible to treat all types in a uni‐
603 form way, mount will execute the program /sbin/mount.type (if
604 that exists) when called with type type. Since different ver‐
605 sions of the smbmount program have different calling conven‐
606 tions, /sbin/mount.smbfs may have to be a shell script that sets
607 up the desired call.
608 -U, --uuid uuid
610 Mount the partition that has the specified uuid.
611 -v, --verbose
613 Verbose mode.
614 -w, --rw, --read-write
616 Mount the filesystem read/write. The read-write is kernel
617 default. A synonym is -o rw.
618 Note that specify -w on command line forces mount command to
620 never try read-only mount on write-protected devices. The
621 default is try read-only if the previous mount syscall with
622 read-write flags failed.
623 -V, --version
625 Display version information and exit.
626 -h, --help
628 Display help text and exit.
629
632 Some of these options are only useful when they appear in the
633 /etc/fstab file.
634 Some of these options could be enabled or disabled by default in the
636 system kernel. To check the current setting see the options in
637 /proc/mounts. Note that filesystems also have per-filesystem specific
638 default mount options (see for example tune2fs -l output for extN
639 filesystems).
640 The following options apply to any filesystem that is being mounted
642 (but not every filesystem actually honors them – e.g., the sync option
643 today has an effect only for ext2, ext3, ext4, fat, vfat, ufs and xfs):
644 async All I/O to the filesystem should be done asynchronously. (See
647 also the sync option.)
648 atime Do not use the noatime feature, so the inode access time is con‐
650 trolled by kernel defaults. See also the descriptions of the
651 relatime and strictatime mount options.
652 noatime
654 Do not update inode access times on this filesystem (e.g. for
655 faster access on the news spool to speed up news servers). This
656 works for all inode types (directories too), so it implies
657 nodiratime.
658 auto Can be mounted with the -a option.
660 noauto Can only be mounted explicitly (i.e., the -a option will not
662 cause the filesystem to be mounted).
663 context=context, fscontext=context, defcontext=context, and
665 rootcontext=context
666 The context= option is useful when mounting filesystems that do
667 not support extended attributes, such as a floppy or hard disk
668 formatted with VFAT, or systems that are not normally running
669 under SELinux, such as an ext3 or ext4 formatted
670 disk from a non-SELinux workstation. You can also use context=
672 on filesystems you do not trust, such as a floppy. It also
673 helps in compatibility with xattr-supporting filesystems on ear‐
674 lier 2.4.<x> kernel versions. Even where xattrs are supported,
675 you can save time not having to label every file by assigning
676 the entire disk one security context.
677 A commonly used option for removable media is
679 context="system_u:object_r:removable_t".
680 Two other options are fscontext= and defcontext=, both of which
682 are mutually exclusive of the context option. This means you
683 can use fscontext and defcontext with each other, but neither
684 can be used with context.
685 The fscontext= option works for all filesystems, regardless of
687 their xattr support. The fscontext option sets the overarching
688 filesystem label to a specific security context. This filesys‐
689 tem label is separate from the individual labels on the files.
690 It represents the entire filesystem for certain kinds of permis‐
691 sion checks, such as during mount or file creation. Individual
692 file labels are still obtained from the xattrs on the files
693 themselves. The context option actually sets the aggregate con‐
694 text that fscontext provides, in addition to supplying the same
695 label for individual files.
696 You can set the default security context for unlabeled files
698 using defcontext= option. This overrides the value set for
699 unlabeled files in the policy and requires a filesystem that
700 supports xattr labeling.
701 The rootcontext= option allows you to explicitly label the root
703 inode of a FS being mounted before that FS or inode becomes vis‐
704 ible to userspace. This was found to be useful for things like
705 stateless linux.
706 Note that the kernel rejects any remount request that includes
708 the context option, even when unchanged from the current con‐
709 text.
710 Warning: the context value might contain commas, in which case
712 the value has to be properly quoted, otherwise mount(8) will
713 interpret the comma as a separator between mount options. Don't
714 forget that the shell strips off quotes and thus double quoting
715 is required. For example:
716 mount -t tmpfs none /mnt -o \
718 'context="system_u:object_r:tmp_t:s0:c127,c456",noexec'
719 For more details, see selinux(8).
721 defaults
724 Use the default options: rw, suid, dev, exec, auto, nouser, and
725 async.
726 Note that the real set of all default mount options depends on
728 kernel and filesystem type. See the beginning of this section
729 for more details.
730 dev Interpret character or block special devices on the filesystem.
732 nodev Do not interpret character or block special devices on the file
734 system.
735 diratime
737 Update directory inode access times on this filesystem. This is
738 the default. (This option is ignored when noatime is set.)
739 nodiratime
741 Do not update directory inode access times on this filesystem.
742 (This option is implied when noatime is set.)
743 dirsync
745 All directory updates within the filesystem should be done syn‐
746 chronously. This affects the following system calls: creat,
747 link, unlink, symlink, mkdir, rmdir, mknod and rename.
748 exec Permit execution of binaries.
750 noexec Do not permit direct execution of any binaries on the mounted
752 filesystem.
753 group Allow an ordinary user to mount the filesystem if one of that
755 user's groups matches the group of the device. This option
756 implies the options nosuid and nodev (unless overridden by sub‐
757 sequent options, as in the option line group,dev,suid).
758 iversion
760 Every time the inode is modified, the i_version field will be
761 incremented.
762 noiversion
764 Do not increment the i_version inode field.
765 mand Allow mandatory locks on this filesystem. See fcntl(2).
767 nomand Do not allow mandatory locks on this filesystem.
769 _netdev
771 The filesystem resides on a device that requires network access
772 (used to prevent the system from attempting to mount these
773 filesystems until the network has been enabled on the system).
774 nofail Do not report errors for this device if it does not exist.
776 relatime
778 Update inode access times relative to modify or change time.
779 Access time is only updated if the previous access time was ear‐
780 lier than the current modify or change time. (Similar to
781 noatime, but it doesn't break mutt or other applications that
782 need to know if a file has been read since the last time it was
783 modified.)
784 Since Linux 2.6.30, the kernel defaults to the behavior provided
786 by this option (unless noatime was specified), and the
787 strictatime option is required to obtain traditional semantics.
788 In addition, since Linux 2.6.30, the file's last access time is
789 always updated if it is more than 1 day old.
790 norelatime
792 Do not use the relatime feature. See also the strictatime mount
793 option.
794 strictatime
796 Allows to explicitly request full atime updates. This makes it
797 possible for the kernel to default to relatime or noatime but
798 still allow userspace to override it. For more details about
799 the default system mount options see /proc/mounts.
800 nostrictatime
802 Use the kernel's default behavior for inode access time updates.
803 lazytime
805 Only update times (atime, mtime, ctime) on the in-memory version
806 of the file inode.
807 This mount option significantly reduces writes to the inode ta‐
809 ble for workloads that perform frequent random writes to preal‐
810 located files.
811 The on-disk timestamps are updated only when:
813 - the inode needs to be updated for some change unrelated to
815 file timestamps
816 - the application employs fsync(2), syncfs(2), or sync(2)
818 - an undeleted inode is evicted from memory
820 - more than 24 hours have passed since the i-node was written to
822 disk.
823 nolazytime
826 Do not use the lazytime feature.
827 suid Honor set-user-ID and set-group-ID bits or file capabilities
829 when executing programs from this filesystem.
830 nosuid Do not honor set-user-ID and set-group-ID bits or file capabili‐
832 ties when executing programs from this filesystem.
833 silent Turn on the silent flag.
835 loud Turn off the silent flag.
837 owner Allow an ordinary user to mount the filesystem if that user is
839 the owner of the device. This option implies the options nosuid
840 and nodev (unless overridden by subsequent options, as in the
841 option line owner,dev,suid).
842 remount
844 Attempt to remount an already-mounted filesystem. This is com‐
845 monly used to change the mount flags for a filesystem, espe‐
846 cially to make a readonly filesystem writable. It does not
847 change device or mount point.
848 The remount operation together with the bind flag has special
850 semantic. See above, the subsection Bind mounts.
851 The remount functionality follows the standard way the mount
853 command works with options from fstab. This means that mount
854 does not read fstab (or mtab) only when both device and dir are
855 specified.
856 mount -o remount,rw /dev/foo /dir
858 After this call all old mount options are replaced and arbitrary
860 stuff from fstab (or mtab) is ignored, except the loop= option
861 which is internally generated and maintained by the mount com‐
862 mand.
863 mount -o remount,rw /dir
865 After this call, mount reads fstab and merges these options with
867 the options from the command line (-o). If no mountpoint is
868 found in fstab, then a remount with unspecified source is
869 allowed.
870 mount(8) allows to use --all to remount all already mounted
872 filesystems which match a specified filter (-O and -t). For
873 example:
874 mount --all -o remount,ro -t vfat
876 remounts all already mounted vfat filesystems in read-only mode.
878 The each of the filesystems is remounted by "mount -o remount,ro
879 /dir" semantic. It means the mount command reads fstab or mtab
880 and merges these options with the options from the command line.
881 ro Mount the filesystem read-only.
883 rw Mount the filesystem read-write.
885 sync All I/O to the filesystem should be done synchronously. In the
887 case of media with a limited number of write cycles (e.g. some
888 flash drives), sync may cause life-cycle shortening.
889 user Allow an ordinary user to mount the filesystem. The name of the
891 mounting user is written to the mtab file (or to the private
892 libmount file in /run/mount on systems without a regular mtab)
893 so that this same user can unmount the filesystem again. This
894 option implies the options noexec, nosuid, and nodev (unless
895 overridden by subsequent options, as in the option line
896 user,exec,dev,suid).
897 nouser Forbid an ordinary user to mount the filesystem. This is the
899 default; it does not imply any other options.
900 users Allow any user to mount and to unmount the filesystem, even when
902 some other ordinary user mounted it. This option implies the
903 options noexec, nosuid, and nodev (unless overridden by subse‐
904 quent options, as in the option line users,exec,dev,suid).
905 X-* All options prefixed with "X-" are interpreted as comments or as
907 userspace application-specific options. These options are not
908 stored in the user space (e.g., mtab file), nor sent to the
909 mount.type helpers nor to the mount(2) system call. The sug‐
910 gested format is X-appname.option.
911 x-* The same as X-* options, but stored permanently in the user
913 space. It means the options are also available for umount or
914 another operations. Note that maintain mount options in user
915 space is tricky, because it's necessary use libmount based tools
916 and there is no guarantee that the options will be always avail‐
917 able (for example after a move mount operation or in unshared
918 namespace).
919 Note that before util-linux v2.30 the x-* options have not been
921 maintained by libmount and stored in user space (functionality
922 was the same as have X-* now), but due to growing number of use-
923 cases (in initrd, systemd etc.) the functionality have been
924 extended to keep existing fstab configurations usable without a
925 change.
926 X-mount.mkdir[=mode]
928 Allow to make a target directory (mountpoint) if it does not
929 exit yet. The optional argument mode specifies the filesystem
930 access mode used for mkdir(2) in octal notation. The default
931 mode is 0755. This functionality is supported only for root
932 users or when mount executed without suid permissions. The
933 option is also supported as x-mount.mkdir, this notation is dep‐
934 recated since v2.30.
935
938 You should consult the respective man page for the filesystem first.
939 If you want to know what options the ext4 filesystem supports, then
940 check the ext4(5) man page. If that doesn't exist, you can also check
941 the corresponding mount page like mount.cifs(8). Note that you might
942 have to install the respective userland tools.
943 The following options apply only to certain filesystems. We sort them
945 by filesystem. They all follow the -o flag.
946 What options are supported depends a bit on the running kernel. More
948 info may be found in the kernel source subdirectory Documenta‐
949 tion/filesystems.
950 Mount options for adfs
953 uid=value and gid=value
954 Set the owner and group of the files in the filesystem (default:
955 uid=gid=0).
956 ownmask=value and othmask=value
958 Set the permission mask for ADFS 'owner' permissions and 'other'
959 permissions, respectively (default: 0700 and 0077, respec‐
960 tively). See also /usr/src/linux/Documentation/filesys‐
961 tems/adfs.txt.
962 Mount options for affs
965 uid=value and gid=value
966 Set the owner and group of the root of the filesystem (default:
967 uid=gid=0, but with option uid or gid without specified value,
968 the UID and GID of the current process are taken).
969 setuid=value and setgid=value
971 Set the owner and group of all files.
972 mode=value
974 Set the mode of all files to value & 0777 disregarding the orig‐
975 inal permissions. Add search permission to directories that
976 have read permission. The value is given in octal.
977 protect
979 Do not allow any changes to the protection bits on the filesys‐
980 tem.
981 usemp Set UID and GID of the root of the filesystem to the UID and GID
983 of the mount point upon the first sync or umount, and then clear
984 this option. Strange...
985 verbose
987 Print an informational message for each successful mount.
988 prefix=string
990 Prefix used before volume name, when following a link.
991 volume=string
993 Prefix (of length at most 30) used before '/' when following a
994 symbolic link.
995 reserved=value
997 (Default: 2.) Number of unused blocks at the start of the
998 device.
999 root=value
1001 Give explicitly the location of the root block.
1002 bs=value
1004 Give blocksize. Allowed values are 512, 1024, 2048, 4096.
1005 grpquota|noquota|quota|usrquota
1007 These options are accepted but ignored. (However, quota utili‐
1008 ties may react to such strings in /etc/fstab.)
1009 Mount options for debugfs
1012 The debugfs filesystem is a pseudo filesystem, traditionally mounted on
1013 /sys/kernel/debug. As of kernel version 3.4, debugfs has the following
1014 options:
1015 uid=n, gid=n
1017 Set the owner and group of the mountpoint.
1018 mode=value
1020 Sets the mode of the mountpoint.
1021 Mount options for devpts
1024 The devpts filesystem is a pseudo filesystem, traditionally mounted on
1025 /dev/pts. In order to acquire a pseudo terminal, a process opens
1026 /dev/ptmx; the number of the pseudo terminal is then made available to
1027 the process and the pseudo terminal slave can be accessed as
1028 /dev/pts/<number>.
1029 uid=value and gid=value
1031 This sets the owner or the group of newly created PTYs to the
1032 specified values. When nothing is specified, they will be set
1033 to the UID and GID of the creating process. For example, if
1034 there is a tty group with GID 5, then gid=5 will cause newly
1035 created PTYs to belong to the tty group.
1036 mode=value
1038 Set the mode of newly created PTYs to the specified value. The
1039 default is 0600. A value of mode=620 and gid=5 makes "mesg y"
1040 the default on newly created PTYs.
1041 newinstance
1043 Create a private instance of devpts filesystem, such that
1044 indices of ptys allocated in this new instance are independent
1045 of indices created in other instances of devpts.
1046 All mounts of devpts without this newinstance option share the
1048 same set of pty indices (i.e., legacy mode). Each mount of
1049 devpts with the newinstance option has a private set of pty
1050 indices.
1051 This option is mainly used to support containers in the linux
1053 kernel. It is implemented in linux kernel versions starting
1054 with 2.6.29. Further, this mount option is valid only if CON‐
1055 FIG_DEVPTS_MULTIPLE_INSTANCES is enabled in the kernel configu‐
1056 ration.
1057 To use this option effectively, /dev/ptmx must be a symbolic
1059 link to pts/ptmx. See Documentation/filesystems/devpts.txt in
1060 the linux kernel source tree for details.
1061 ptmxmode=value
1063 Set the mode for the new ptmx device node in the devpts filesys‐
1065 tem.
1066 With the support for multiple instances of devpts (see newin‐
1068 stance option above), each instance has a private ptmx node in
1069 the root of the devpts filesystem (typically /dev/pts/ptmx).
1070 For compatibility with older versions of the kernel, the default
1072 mode of the new ptmx node is 0000. ptmxmode=value specifies a
1073 more useful mode for the ptmx node and is highly recommended
1074 when the newinstance option is specified.
1075 This option is only implemented in linux kernel versions start‐
1077 ing with 2.6.29. Further, this option is valid only if CON‐
1078 FIG_DEVPTS_MULTIPLE_INSTANCES is enabled in the kernel configu‐
1079 ration.
1080 Mount options for fat
1083 (Note: fat is not a separate filesystem, but a common part of the
1084 msdos, umsdos and vfat filesystems.)
1085 blocksize={512|1024|2048}
1087 Set blocksize (default 512). This option is obsolete.
1088 uid=value and gid=value
1090 Set the owner and group of all files. (Default: the UID and GID
1091 of the current process.)
1092 umask=value
1094 Set the umask (the bitmask of the permissions that are not
1095 present). The default is the umask of the current process. The
1096 value is given in octal.
1097 dmask=value
1099 Set the umask applied to directories only. The default is the
1100 umask of the current process. The value is given in octal.
1101 fmask=value
1103 Set the umask applied to regular files only. The default is the
1104 umask of the current process. The value is given in octal.
1105 allow_utime=value
1107 This option controls the permission check of mtime/atime.
1108 20 If current process is in group of file's group ID, you
1110 can change timestamp.
1111 2 Other users can change timestamp.
1113 The default is set from `dmask' option. (If the directory is
1115 writable, utime(2) is also allowed. I.e. ~dmask & 022)
1116 Normally utime(2) checks current process is owner of the file,
1118 or it has CAP_FOWNER capability. But FAT filesystem doesn't
1119 have UID/GID on disk, so normal check is too inflexible. With
1120 this option you can relax it.
1121 check=value
1123 Three different levels of pickiness can be chosen:
1124 r[elaxed]
1126 Upper and lower case are accepted and equivalent, long
1127 name parts are truncated (e.g. verylongname.foobar
1128 becomes verylong.foo), leading and embedded spaces are
1129 accepted in each name part (name and extension).
1130 n[ormal]
1132 Like "relaxed", but many special characters (*, ?, <,
1133 spaces, etc.) are rejected. This is the default.
1134 s[trict]
1136 Like "normal", but names that contain long parts or spe‐
1137 cial characters that are sometimes used on Linux but are
1138 not accepted by MS-DOS (+, =, etc.) are rejected.
1139 codepage=value
1141 Sets the codepage for converting to shortname characters on FAT
1142 and VFAT filesystems. By default, codepage 437 is used.
1143 conv=mode
1145 This option is obsolete and may fail or being ignored.
1146 cvf_format=module
1148 Forces the driver to use the CVF (Compressed Volume File) module
1149 cvf_module instead of auto-detection. If the kernel supports
1150 kmod, the cvf_format=xxx option also controls on-demand CVF mod‐
1151 ule loading. This option is obsolete.
1152 cvf_option=option
1154 Option passed to the CVF module. This option is obsolete.
1155 debug Turn on the debug flag. A version string and a list of filesys‐
1157 tem parameters will be printed (these data are also printed if
1158 the parameters appear to be inconsistent).
1159 discard
1161 If set, causes discard/TRIM commands to be issued to the block
1162 device when blocks are freed. This is useful for SSD devices
1163 and sparse/thinly-provisioned LUNs.
1164 dos1xfloppy
1166 If set, use a fallback default BIOS Parameter Block configura‐
1167 tion, determined by backing device size. These static parame‐
1168 ters match defaults assumed by DOS 1.x for 160 kiB, 180 kiB, 320
1169 kiB, and 360 kiB floppies and floppy images.
1170 errors={panic|continue|remount-ro}
1172 Specify FAT behavior on critical errors: panic, continue without
1173 doing anything, or remount the partition in read-only mode
1174 (default behavior).
1175 fat={12|16|32}
1177 Specify a 12, 16 or 32 bit fat. This overrides the automatic
1178 FAT type detection routine. Use with caution!
1179 iocharset=value
1181 Character set to use for converting between 8 bit characters and
1182 16 bit Unicode characters. The default is iso8859-1. Long
1183 filenames are stored on disk in Unicode format.
1184 nfs={stale_rw|nostale_ro}
1186 Enable this only if you want to export the FAT filesystem over
1187 NFS.
1188 stale_rw: This option maintains an index (cache) of directory
1190 inodes which is used by the nfs-related code to improve look-
1191 ups. Full file operations (read/write) over NFS are supported
1192 but with cache eviction at NFS server, this could result in spu‐
1193 rious ESTALE errors.
1194 nostale_ro: This option bases the inode number and file handle
1196 on the on-disk location of a file in the FAT directory entry.
1197 This ensures that ESTALE will not be returned after a file is
1198 evicted from the inode cache. However, it means that operations
1199 such as rename, create and unlink could cause file handles that
1200 previously pointed at one file to point at a different file,
1201 potentially causing data corruption. For this reason, this
1202 option also mounts the filesystem readonly.
1203 To maintain backward compatibility, '-o nfs' is also accepted,
1205 defaulting to stale_rw.
1206 tz=UTC This option disables the conversion of timestamps between local
1208 time (as used by Windows on FAT) and UTC (which Linux uses
1209 internally). This is particularly useful when mounting devices
1210 (like digital cameras) that are set to UTC in order to avoid the
1211 pitfalls of local time.
1212 time_offset=minutes
1214 Set offset for conversion of timestamps from local time used by
1215 FAT to UTC. I.e., minutes will be subtracted from each time‐
1216 stamp to convert it to UTC used internally by Linux. This is
1217 useful when the time zone set in the kernel via settimeofday(2)
1218 is not the time zone used by the filesystem. Note that this
1219 option still does not provide correct time stamps in all cases
1220 in presence of DST - time stamps in a different DST setting will
1221 be off by one hour.
1222 quiet Turn on the quiet flag. Attempts to chown or chmod files do not
1224 return errors, although they fail. Use with caution!
1225 rodir FAT has the ATTR_RO (read-only) attribute. On Windows, the
1227 ATTR_RO of the directory will just be ignored, and is used only
1228 by applications as a flag (e.g. it's set for the customized
1229 folder).
1230 If you want to use ATTR_RO as read-only flag even for the direc‐
1232 tory, set this option.
1233 showexec
1235 If set, the execute permission bits of the file will be allowed
1236 only if the extension part of the name is .EXE, .COM, or .BAT.
1237 Not set by default.
1238 sys_immutable
1240 If set, ATTR_SYS attribute on FAT is handled as IMMUTABLE flag
1241 on Linux. Not set by default.
1242 flush If set, the filesystem will try to flush to disk more early than
1244 normal. Not set by default.
1245 usefree
1247 Use the "free clusters" value stored on FSINFO. It'll be used
1248 to determine number of free clusters without scanning disk. But
1249 it's not used by default, because recent Windows don't update it
1250 correctly in some case. If you are sure the "free clusters" on
1251 FSINFO is correct, by this option you can avoid scanning disk.
1252 dots, nodots, dotsOK=[yes|no]
1254 Various misguided attempts to force Unix or DOS conventions onto
1255 a FAT filesystem.
1256 Mount options for hfs
1259 creator=cccc, type=cccc
1260 Set the creator/type values as shown by the MacOS finder used
1261 for creating new files. Default values: '????'.
1262 uid=n, gid=n
1264 Set the owner and group of all files. (Default: the UID and GID
1265 of the current process.)
1266 dir_umask=n, file_umask=n, umask=n
1268 Set the umask used for all directories, all regular files, or
1269 all files and directories. Defaults to the umask of the current
1270 process.
1271 session=n
1273 Select the CDROM session to mount. Defaults to leaving that
1274 decision to the CDROM driver. This option will fail with any‐
1275 thing but a CDROM as underlying device.
1276 part=n Select partition number n from the device. Only makes sense for
1278 CDROMs. Defaults to not parsing the partition table at all.
1279 quiet Don't complain about invalid mount options.
1281 Mount options for hpfs
1284 uid=value and gid=value
1285 Set the owner and group of all files. (Default: the UID and GID
1286 of the current process.)
1287 umask=value
1289 Set the umask (the bitmask of the permissions that are not
1290 present). The default is the umask of the current process. The
1291 value is given in octal.
1292 case={lower|asis}
1294 Convert all files names to lower case, or leave them. (Default:
1295 case=lower.)
1296 conv=mode
1298 This option is obsolete and may fail or being ignored.
1299 nocheck
1301 Do not abort mounting when certain consistency checks fail.
1302 Mount options for iso9660
1305 ISO 9660 is a standard describing a filesystem structure to be used on
1306 CD-ROMs. (This filesystem type is also seen on some DVDs. See also the
1307 udf filesystem.)
1308 Normal iso9660 filenames appear in an 8.3 format (i.e., DOS-like
1310 restrictions on filename length), and in addition all characters are in
1311 upper case. Also there is no field for file ownership, protection,
1312 number of links, provision for block/character devices, etc.
1313 Rock Ridge is an extension to iso9660 that provides all of these UNIX-
1315 like features. Basically there are extensions to each directory record
1316 that supply all of the additional information, and when Rock Ridge is
1317 in use, the filesystem is indistinguishable from a normal UNIX filesys‐
1318 tem (except that it is read-only, of course).
1319 norock Disable the use of Rock Ridge extensions, even if available.
1321 Cf. map.
1322 nojoliet
1324 Disable the use of Microsoft Joliet extensions, even if avail‐
1325 able. Cf. map.
1326 check={r[elaxed]|s[trict]}
1328 With check=relaxed, a filename is first converted to lower case
1329 before doing the lookup. This is probably only meaningful
1330 together with norock and map=normal. (Default: check=strict.)
1331 uid=value and gid=value
1333 Give all files in the filesystem the indicated user or group id,
1334 possibly overriding the information found in the Rock Ridge
1335 extensions. (Default: uid=0,gid=0.)
1336 map={n[ormal]|o[ff]|a[corn]}
1338 For non-Rock Ridge volumes, normal name translation maps upper
1339 to lower case ASCII, drops a trailing `;1', and converts `;' to
1340 `.'. With map=off no name translation is done. See norock.
1341 (Default: map=normal.) map=acorn is like map=normal but also
1342 apply Acorn extensions if present.
1343 mode=value
1345 For non-Rock Ridge volumes, give all files the indicated mode.
1346 (Default: read and execute permission for everybody.) Octal
1347 mode values require a leading 0.
1348 unhide Also show hidden and associated files. (If the ordinary files
1350 and the associated or hidden files have the same filenames, this
1351 may make the ordinary files inaccessible.)
1352 block={512|1024|2048}
1354 Set the block size to the indicated value. (Default:
1355 block=1024.)
1356 conv=mode
1358 This option is obsolete and may fail or being ignored.
1359 cruft If the high byte of the file length contains other garbage, set
1361 this mount option to ignore the high order bits of the file
1362 length. This implies that a file cannot be larger than 16 MB.
1363 session=x
1365 Select number of session on multisession CD.
1366 sbsector=xxx
1368 Session begins from sector xxx.
1369 The following options are the same as for vfat and specifying them only
1371 makes sense when using discs encoded using Microsoft's Joliet exten‐
1372 sions.
1373 iocharset=value
1375 Character set to use for converting 16 bit Unicode characters on
1376 CD to 8 bit characters. The default is iso8859-1.
1377 utf8 Convert 16 bit Unicode characters on CD to UTF-8.
1379 Mount options for jfs
1382 iocharset=name
1383 Character set to use for converting from Unicode to ASCII. The
1384 default is to do no conversion. Use iocharset=utf8 for UTF8
1385 translations. This requires CONFIG_NLS_UTF8 to be set in the
1386 kernel .config file.
1387 resize=value
1389 Resize the volume to value blocks. JFS only supports growing a
1390 volume, not shrinking it. This option is only valid during a
1391 remount, when the volume is mounted read-write. The resize key‐
1392 word with no value will grow the volume to the full size of the
1393 partition.
1394 nointegrity
1396 Do not write to the journal. The primary use of this option is
1397 to allow for higher performance when restoring a volume from
1398 backup media. The integrity of the volume is not guaranteed if
1399 the system abnormally ends.
1400 integrity
1402 Default. Commit metadata changes to the journal. Use this
1403 option to remount a volume where the nointegrity option was pre‐
1404 viously specified in order to restore normal behavior.
1405 errors={continue|remount-ro|panic}
1407 Define the behavior when an error is encountered. (Either
1408 ignore errors and just mark the filesystem erroneous and con‐
1409 tinue, or remount the filesystem read-only, or panic and halt
1410 the system.)
1411 noquota|quota|usrquota|grpquota
1413 These options are accepted but ignored.
1414 Mount options for msdos
1417 See mount options for fat. If the msdos filesystem detects an incon‐
1418 sistency, it reports an error and sets the file system read-only. The
1419 filesystem can be made writable again by remounting it.
1420 Mount options for ncpfs
1423 Just like nfs, the ncpfs implementation expects a binary argument (a
1424 struct ncp_mount_data) to the mount system call. This argument is con‐
1425 structed by ncpmount(8) and the current version of mount (2.12) does
1426 not know anything about ncpfs.
1427 Mount options for ntfs
1430 iocharset=name
1431 Character set to use when returning file names. Unlike VFAT,
1432 NTFS suppresses names that contain nonconvertible characters.
1433 Deprecated.
1434 nls=name
1436 New name for the option earlier called iocharset.
1437 utf8 Use UTF-8 for converting file names.
1439 uni_xlate={0|1|2}
1441 For 0 (or `no' or `false'), do not use escape sequences for
1442 unknown Unicode characters. For 1 (or `yes' or `true') or 2,
1443 use vfat-style 4-byte escape sequences starting with ":". Here
1444 2 give a little-endian encoding and 1 a byteswapped bigendian
1445 encoding.
1446 posix=[0|1]
1448 If enabled (posix=1), the filesystem distinguishes between upper
1449 and lower case. The 8.3 alias names are presented as hard links
1450 instead of being suppressed. This option is obsolete.
1451 uid=value, gid=value and umask=value
1453 Set the file permission on the filesystem. The umask value is
1454 given in octal. By default, the files are owned by root and not
1455 readable by somebody else.
1456 Mount options for overlay
1459 Since Linux 3.18 the overlay pseudo filesystem implements a union mount
1460 for other filesystems.
1461 An overlay filesystem combines two filesystems - an upper filesystem
1463 and a lower filesystem. When a name exists in both filesystems, the
1464 object in the upper filesystem is visible while the object in the lower
1465 filesystem is either hidden or, in the case of directories, merged with
1466 the upper object.
1467 The lower filesystem can be any filesystem supported by Linux and does
1469 not need to be writable. The lower filesystem can even be another
1470 overlayfs. The upper filesystem will normally be writable and if it is
1471 it must support the creation of trusted.* extended attributes, and must
1472 provide a valid d_type in readdir responses, so NFS is not suitable.
1473 A read-only overlay of two read-only filesystems may use any filesystem
1475 type. The options lowerdir and upperdir are combined into a merged
1476 directory by using:
1477 mount -t overlay overlay \
1479 -olowerdir=/lower,upperdir=/upper,workdir=/work /merged
1480 lowerdir=directory
1483 Any filesystem, does not need to be on a writable filesystem.
1484 upperdir=directory
1486 The upperdir is normally on a writable filesystem.
1487 workdir=directory
1489 The workdir needs to be an empty directory on the same filesys‐
1490 tem as upperdir.
1491 Mount options for reiserfs
1494 Reiserfs is a journaling filesystem.
1495 conv Instructs version 3.6 reiserfs software to mount a version 3.5
1497 filesystem, using the 3.6 format for newly created objects.
1498 This filesystem will no longer be compatible with reiserfs 3.5
1499 tools.
1500 hash={rupasov|tea|r5|detect}
1502 Choose which hash function reiserfs will use to find files
1503 within directories.
1504 rupasov
1506 A hash invented by Yury Yu. Rupasov. It is fast and pre‐
1507 serves locality, mapping lexicographically close file
1508 names to close hash values. This option should not be
1509 used, as it causes a high probability of hash collisions.
1510 tea A Davis-Meyer function implemented by Jeremy
1512 Fitzhardinge. It uses hash permuting bits in the name.
1513 It gets high randomness and, therefore, low probability
1514 of hash collisions at some CPU cost. This may be used if
1515 EHASHCOLLISION errors are experienced with the r5 hash.
1516 r5 A modified version of the rupasov hash. It is used by
1518 default and is the best choice unless the filesystem has
1519 huge directories and unusual file-name patterns.
1520 detect Instructs mount to detect which hash function is in use
1522 by examining the filesystem being mounted, and to write
1523 this information into the reiserfs superblock. This is
1524 only useful on the first mount of an old format filesys‐
1525 tem.
1526 hashed_relocation
1528 Tunes the block allocator. This may provide performance
1529 improvements in some situations.
1530 no_unhashed_relocation
1532 Tunes the block allocator. This may provide performance
1533 improvements in some situations.
1534 noborder
1536 Disable the border allocator algorithm invented by Yury Yu.
1537 Rupasov. This may provide performance improvements in some sit‐
1538 uations.
1539 nolog Disable journaling. This will provide slight performance
1541 improvements in some situations at the cost of losing reiserfs's
1542 fast recovery from crashes. Even with this option turned on,
1543 reiserfs still performs all journaling operations, save for
1544 actual writes into its journaling area. Implementation of nolog
1545 is a work in progress.
1546 notail By default, reiserfs stores small files and `file tails'
1548 directly into its tree. This confuses some utilities such as
1549 LILO(8). This option is used to disable packing of files into
1550 the tree.
1551 replayonly
1553 Replay the transactions which are in the journal, but do not
1554 actually mount the filesystem. Mainly used by reiserfsck.
1555 resize=number
1557 A remount option which permits online expansion of reiserfs par‐
1558 titions. Instructs reiserfs to assume that the device has num‐
1559 ber blocks. This option is designed for use with devices which
1560 are under logical volume management (LVM). There is a special
1561 resizer utility which can be obtained from
1562 ftp://ftp.namesys.com/pub/reiserfsprogs.
1563 user_xattr
1565 Enable Extended User Attributes. See the attr(1) manual page.
1566 acl Enable POSIX Access Control Lists. See the acl(5) manual page.
1568 barrier=none / barrier=flush
1570 This disables / enables the use of write barriers in the jour‐
1571 naling code. barrier=none disables, barrier=flush enables
1572 (default). This also requires an IO stack which can support
1573 barriers, and if reiserfs gets an error on a barrier write, it
1574 will disable barriers again with a warning. Write barriers
1575 enforce proper on-disk ordering of journal commits, making
1576 volatile disk write caches safe to use, at some performance
1577 penalty. If your disks are battery-backed in one way or
1578 another, disabling barriers may safely improve performance.
1579 Mount options for ubifs
1582 UBIFS is a flash filesystem which works on top of UBI volumes. Note
1583 that atime is not supported and is always turned off.
1584 The device name may be specified as
1586 ubiX_Y UBI device number X, volume number Y
1587 ubiY UBI device number 0, volume number Y
1589 ubiX:NAME
1591 UBI device number X, volume with name NAME
1592 ubi:NAME
1594 UBI device number 0, volume with name NAME
1595 Alternative ! separator may be used instead of :.
1596 The following mount options are available:
1598 bulk_read
1600 Enable bulk-read. VFS read-ahead is disabled because it slows
1601 down the file system. Bulk-Read is an internal optimization.
1602 Some flashes may read faster if the data are read at one go,
1603 rather than at several read requests. For example, OneNAND can
1604 do "read-while-load" if it reads more than one NAND page.
1605 no_bulk_read
1607 Do not bulk-read. This is the default.
1608 chk_data_crc
1610 Check data CRC-32 checksums. This is the default.
1611 no_chk_data_crc.
1613 Do not check data CRC-32 checksums. With this option, the
1614 filesystem does not check CRC-32 checksum for data, but it does
1615 check it for the internal indexing information. This option
1616 only affects reading, not writing. CRC-32 is always calculated
1617 when writing the data.
1618 compr={none|lzo|zlib}
1620 Select the default compressor which is used when new files are
1621 written. It is still possible to read compressed files if
1622 mounted with the none option.
1623 Mount options for udf
1626 UDF is the "Universal Disk Format" filesystem defined by OSTA, the
1627 Optical Storage Technology Association, and is often used for DVD-ROM,
1628 frequently in the form of a hybrid UDF/ISO-9660 filesystem. It is, how‐
1629 ever, perfectly usable by itself on disk drives, flash drives and other
1630 block devices. See also iso9660.
1631 uid= Make all files in the filesystem belong to the given user.
1633 uid=forget can be specified independently of (or usually in
1634 addition to) uid=<user> and results in UDF not storing uids to
1635 the media. In fact the recorded uid is the 32-bit overflow uid
1636 -1 as defined by the UDF standard. The value is given as either
1637 <user> which is a valid user name or the corresponding decimal
1638 user id, or the special string "forget".
1639 gid= Make all files in the filesystem belong to the given group.
1641 gid=forget can be specified independently of (or usually in
1642 addition to) gid=<group> and results in UDF not storing gids to
1643 the media. In fact the recorded gid is the 32-bit overflow gid
1644 -1 as defined by the UDF standard. The value is given as either
1645 <group> which is a valid group name or the corresponding decimal
1646 group id, or the special string "forget".
1647 umask= Mask out the given permissions from all inodes read from the
1649 filesystem. The value is given in octal.
1650 mode= If mode= is set the permissions of all non-directory inodes read
1652 from the filesystem will be set to the given mode. The value is
1653 given in octal.
1654 dmode= If dmode= is set the permissions of all directory inodes read
1656 from the filesystem will be set to the given dmode. The value is
1657 given in octal.
1658 bs= Set the block size. Default value prior to kernel version 2.6.30
1660 was 2048. Since 2.6.30 and prior to 4.11 it was logical device
1661 block size with fallback to 2048. Since 4.11 it is logical block
1662 size with fallback to any valid block size between logical
1663 device block size and 4096.
1664 For other details see the mkudffs(8) 2.0+ manpage, sections COM‐
1666 PATIBILITY and BLOCK SIZE.
1667 unhide Show otherwise hidden files.
1669 undelete
1671 Show deleted files in lists.
1672 adinicb
1674 Embed data in the inode. (default)
1675 noadinicb
1677 Don't embed data in the inode.
1678 shortad
1680 Use short UDF address descriptors.
1681 longad Use long UDF address descriptors. (default)
1683 nostrict
1685 Unset strict conformance.
1686 iocharset=
1688 Set the NLS character set. This requires kernel compiled with
1689 CONFIG_UDF_NLS option.
1690 utf8 Set the UTF-8 character set.
1692 Mount options for debugging and disaster recovery
1694 novrs Ignore the Volume Recognition Sequence and attempt to mount any‐
1695 way.
1696 session=
1698 Select the session number for multi-session recorded optical
1699 media. (default= last session)
1700 anchor=
1702 Override standard anchor location. (default= 256)
1703 lastblock=
1705 Set the last block of the filesystem.
1706 Unused historical mount options that may be encountered and should be
1708 removed
1709 uid=ignore
1710 Ignored, use uid=<user> instead.
1711 gid=ignore
1713 Ignored, use gid=<group> instead.
1714 volume=
1716 Unimplemented and ignored.
1717 partition=
1719 Unimplemented and ignored.
1720 fileset=
1722 Unimplemented and ignored.
1723 rootdir=
1725 Unimplemented and ignored.
1726 Mount options for ufs
1729 ufstype=value
1730 UFS is a filesystem widely used in different operating systems.
1731 The problem are differences among implementations. Features of
1732 some implementations are undocumented, so its hard to recognize
1733 the type of ufs automatically. That's why the user must specify
1734 the type of ufs by mount option. Possible values are:
1735 old Old format of ufs, this is the default, read only.
1737 (Don't forget to give the -r option.)
1738 44bsd For filesystems created by a BSD-like system (NetBSD,
1740 FreeBSD, OpenBSD).
1741 ufs2 Used in FreeBSD 5.x supported as read-write.
1743 5xbsd Synonym for ufs2.
1745 sun For filesystems created by SunOS or Solaris on Sparc.
1747 sunx86 For filesystems created by Solaris on x86.
1749 hp For filesystems created by HP-UX, read-only.
1751 nextstep
1753 For filesystems created by NeXTStep (on NeXT station)
1754 (currently read only).
1755 nextstep-cd
1757 For NextStep CDROMs (block_size == 2048), read-only.
1758 openstep
1760 For filesystems created by OpenStep (currently read
1761 only). The same filesystem type is also used by Mac OS
1762 X.
1763 onerror=value
1766 Set behavior on error:
1767 panic If an error is encountered, cause a kernel panic.
1769 [lock|umount|repair]
1771 These mount options don't do anything at present; when an
1772 error is encountered only a console message is printed.
1773 Mount options for umsdos
1776 See mount options for msdos. The dotsOK option is explicitly killed by
1777 umsdos.
1778 Mount options for vfat
1781 First of all, the mount options for fat are recognized. The dotsOK
1782 option is explicitly killed by vfat. Furthermore, there are
1783 uni_xlate
1785 Translate unhandled Unicode characters to special escaped
1786 sequences. This lets you backup and restore filenames that are
1787 created with any Unicode characters. Without this option, a '?'
1788 is used when no translation is possible. The escape character
1789 is ':' because it is otherwise invalid on the vfat filesystem.
1790 The escape sequence that gets used, where u is the Unicode char‐
1791 acter, is: ':', (u & 0x3f), ((u>>6) & 0x3f), (u>>12).
1792 posix Allow two files with names that only differ in case. This
1794 option is obsolete.
1795 nonumtail
1797 First try to make a short name without sequence number, before
1798 trying name~num.ext.
1799 utf8 UTF8 is the filesystem safe 8-bit encoding of Unicode that is
1801 used by the console. It can be enabled for the filesystem with
1802 this option or disabled with utf8=0, utf8=no or utf8=false. If
1803 `uni_xlate' gets set, UTF8 gets disabled.
1804 shortname=mode
1806 Defines the behavior for creation and display of filenames which
1807 fit into 8.3 characters. If a long name for a file exists, it
1808 will always be the preferred one for display. There are four
1809 modes:
1810 lower Force the short name to lower case upon display; store a
1812 long name when the short name is not all upper case.
1813 win95 Force the short name to upper case upon display; store a
1815 long name when the short name is not all upper case.
1816 winnt Display the short name as is; store a long name when the
1818 short name is not all lower case or all upper case.
1819 mixed Display the short name as is; store a long name when the
1821 short name is not all upper case. This mode is the
1822 default since Linux 2.6.32.
1823 Mount options for usbfs
1826 devuid=uid and devgid=gid and devmode=mode
1827 Set the owner and group and mode of the device files in the
1828 usbfs filesystem (default: uid=gid=0, mode=0644). The mode is
1829 given in octal.
1830 busuid=uid and busgid=gid and busmode=mode
1832 Set the owner and group and mode of the bus directories in the
1833 usbfs filesystem (default: uid=gid=0, mode=0555). The mode is
1834 given in octal.
1835 listuid=uid and listgid=gid and listmode=mode
1837 Set the owner and group and mode of the file devices (default:
1838 uid=gid=0, mode=0444). The mode is given in octal.
1839
1842 The device-mapper verity target provides read-only transparent
1843 integrity checking of block devices using kernel crypto API. The mount
1844 command can open the dm-verity device and do the integrity verification
1845 before on the device filesystem is mounted. Requires libcryptsetup
1846 with in libmount. If libcryptsetup supports extracting the root hash
1847 of an already mounted device, existing devices will be automatically
1848 reused in case of a match. Mount options for dm-verity:
1849 verity.hashdevice=path
1851 Path to the hash tree device associated with the source volume
1852 to pass to dm-verity.
1853 verity.roothash=hex
1855 Hex-encoded hash of the root of verity.hashdevice Mutually
1856 exclusive with verity.roothashfile.
1857 verity.roothashfile=path
1859 Path to file containing the hex-encoded hash of the root of ver‐
1860 ity.hashdevice. Mutually exclusive with verity.roothash.
1861 verity.hashoffset=offset
1863 If the hash tree device is embedded in the source volume, offset
1864 (default: 0) is used by dm-verity to get to the tree.
1865 verity.fecdevice=path
1867 Path to the Forward Error Correction (FEC) device associated
1868 with the source volume to pass to dm-verity. Optional. Requires
1869 kernel built with CONFIG_DM_VERITY_FEC.
1870 verity.fecoffset=offset
1872 If the FEC device is embedded in the source volume, offset
1873 (default: 0) is used by dm-verity to get to the FEC area.
1874 Optional.
1875 verity.fecroots=value
1877 Parity bytes for FEC (default: 2). Optional.
1878 Supported since util-linux v2.35.
1880 For example commands:
1882 mksquashfs /etc /tmp/etc.squashfs
1884 dd if=/dev/zero of=/tmp/etc.hash bs=1M count=10
1885 veritysetup format /tmp/etc.squashfs /tmp/etc.hash
1886 mount -o verity.hashdevice=/tmp/etc.hash,verity.roothash=<hash> /tmp/etc.squashfs /mnt
1887 create squashfs image from /etc directory, verity hash device and mount
1889 verified filesystem image to /mnt.
1890
1893 One further possible type is a mount via the loop device. For example,
1894 the command
1895 mount /tmp/disk.img /mnt -t vfat -o loop=/dev/loop3
1897 will set up the loop device /dev/loop3 to correspond to the file
1899 /tmp/disk.img, and then mount this device on /mnt.
1900 If no explicit loop device is mentioned (but just an option `-o loop'
1902 is given), then mount will try to find some unused loop device and use
1903 that, for example
1904 mount /tmp/disk.img /mnt -o loop
1906 The mount command automatically creates a loop device from a regular
1908 file if a filesystem type is not specified or the filesystem is known
1909 for libblkid, for example:
1910 mount /tmp/disk.img /mnt
1912 mount -t ext4 /tmp/disk.img /mnt
1914 This type of mount knows about three options, namely loop, offset and
1916 sizelimit, that are really options to losetup(8). (These options can
1917 be used in addition to those specific to the filesystem type.)
1918 Since Linux 2.6.25 auto-destruction of loop devices is supported, mean‐
1920 ing that any loop device allocated by mount will be freed by umount
1921 independently of /etc/mtab.
1922 You can also free a loop device by hand, using losetup -d or umount -d.
1924 Since util-linux v2.29 mount command re-uses the loop device rather
1926 than initialize a new device if the same backing file is already used
1927 for some loop device with the same offset and sizelimit. This is neces‐
1928 sary to avoid a filesystem corruption.
1929
1932 mount has the following return codes (the bits can be ORed):
1933 0 success
1935 1 incorrect invocation or permissions
1937 2 system error (out of memory, cannot fork, no more loop devices)
1939 4 internal mount bug
1941 8 user interrupt
1943 16 problems writing or locking /etc/mtab
1945 32 mount failure
1947 64 some mount succeeded
1949 The command mount -a returns 0 (all succeeded), 32 (all failed),
1951 or 64 (some failed, some succeeded).
1952
1955 The syntax of external mount helpers is:
1956 /sbin/mount.suffix spec dir [-sfnv] [-N namespace] [-o options] [-t
1958 type.subtype]
1959 where the suffix is the filesystem type and the -sfnvoN options have
1961 the same meaning as the normal mount options. The -t option is used
1962 for filesystems with subtypes support (for example /sbin/mount.fuse -t
1963 fuse.sshfs).
1964 The command mount does not pass the mount options unbindable, runbind‐
1966 able, private, rprivate, slave, rslave, shared, rshared, auto, noauto,
1967 comment, x-*, loop, offset and sizelimit to the mount.<suffix> helpers.
1968 All other options are used in a comma-separated list as argument to the
1969 -o option.
1970
1973 See also "The files /etc/fstab, /etc/mtab and /proc/mounts" section
1974 above.
1975 /etc/fstab filesystem table
1977 /run/mount libmount private runtime directory
1979 /etc/mtab table of mounted filesystems or symlink to
1981 /proc/mounts
1982 /etc/mtab~ lock file (unused on systems with mtab symlink)
1984 /etc/mtab.tmp temporary file (unused on systems with mtab symlink)
1986 /etc/filesystems a list of filesystem types to try
1988
1990 LIBMOUNT_FSTAB=<path>
1991 overrides the default location of the fstab file (ignored for
1992 suid)
1993 LIBMOUNT_MTAB=<path>
1995 overrides the default location of the mtab file (ignored for
1996 suid)
1997 LIBMOUNT_DEBUG=all
1999 enables libmount debug output
2000 LIBBLKID_DEBUG=all
2002 enables libblkid debug output
2003 LOOPDEV_DEBUG=all
2005 enables loop device setup debug output
2006
2008 mount(2), umount(2), umount(8), fstab(5), nfs(5), xfs(5), e2label(8),
2009 findmnt(8), losetup(8), mke2fs(8), mountd(8), nfsd(8), swapon(8),
2010 tune2fs(8), xfs_admin(8)
2011
2013 It is possible for a corrupted filesystem to cause a crash.
2014 Some Linux filesystems don't support -o sync nor -o dirsync (the ext2,
2016 ext3, ext4, fat and vfat filesystems do support synchronous updates (a
2017 la BSD) when mounted with the sync option).
2018 The -o remount may not be able to change mount parameters (all ext2fs-
2020 specific parameters, except sb, are changeable with a remount, for
2021 example, but you can't change gid or umask for the fatfs).
2022 It is possible that the files /etc/mtab and /proc/mounts don't match on
2024 systems with a regular mtab file. The first file is based only on the
2025 mount command options, but the content of the second file also depends
2026 on the kernel and others settings (e.g. on a remote NFS server -- in
2027 certain cases the mount command may report unreliable information about
2028 an NFS mount point and the /proc/mounts file usually contains more
2029 reliable information.) This is another reason to replace the mtab file
2030 with a symlink to the /proc/mounts file.
2031 Checking files on NFS filesystems referenced by file descriptors (i.e.
2033 the fcntl and ioctl families of functions) may lead to inconsistent
2034 results due to the lack of a consistency check in the kernel even if
2035 noac is used.
2036 The loop option with the offset or sizelimit options used may fail when
2038 using older kernels if the mount command can't confirm that the size of
2039 the block device has been configured as requested. This situation can
2040 be worked around by using the losetup command manually before calling
2041 mount with the configured loop device.
2042
2044 A mount command existed in Version 5 AT&T UNIX.
2045
2047 Karel Zak <kzak@redhat.com>
2048
2050 The mount command is part of the util-linux package and is available
2051 from https://www.kernel.org/pub/linux/utils/util-linux/.
2052util-linux August 2015 MOUNT(8)