DISKUTIL(8) BSD System Manager's Manual DISKUTIL(8)
NAME
diskutil -- modify, verify and repair local disks
SYNOPSIS
diskutil [quiet] verb [options]
DESCRIPTION
diskutil manipulates the structure of local disks. It provides information about, and allows the administration of, the partitioning
schemes, layouts, and formats of disks. This includes hard disks, solid state disks, optical discs, disk images, APFS volumes, CoreStorage
volumes, and AppleRAID sets. It generally manipulates whole volumes instead of individual files and directories.
VERBS
Each verb is listed with its description and individual arguments.
list [-plist] [internal | external] [physical | virtual] [device]
List disks, including internal and external disks, whole disks and partitions, and various kinds of virtual or offline disks.
If no argument is given, then all whole disks and their partitions are listed.
You can limit the number of disks shown by specifying filtering arguments such as internal above, and/or a device disk. When lim-
iting by a disk, you can specify either a whole disk, e.g. disk0, or any of its slices, e.g. disk0s3, but filtering is only done
at the whole disk level (disk0s3 is a synonym for disk0 in this case).
If -plist is specified, then a property list will be emitted instead of the normal user-readable output.
A script could interpret the results of diskutil list -plist and use diskutil info -plist as well as diskutil listFilesystems
-plist for more detailed information.
The top-to-bottom appearance of all whole disks is sorted in numerical order by unit (whole disk) number. However, within each
whole disk's "sublist" of partitions, the ordering indicates actual on-disk location. The first disk item listed represents the
partition which is located most near the beginning of its encompassing whole disk, and so on.
When viewed this way, the slice (partition) parts of the BSD disk identifiers may, in certain circumstances, not appear in numeri-
cal order. This is normal and is likely the result of a recent partition map editing operation in which volumes were kept
mounted.
Note that both human-readable and plist output are sorted as described above.
See the DEVICES section below for the various forms that the device specification may take for this and all of the other diskutil
verbs.
info | information [-plist] device | -all
Get detailed information about a specific whole disk or partition. If -plist is specified, then a property list instead of the
normal user-readable output will be emitted. If -all is specified, then all disks (whole disks and their partitions) are pro-
cessed.
activity
Continuously display system-wide disk manipulation activity as reported by the Disk Arbitration framework until interrupted with a
signal (e.g. by typing Control-C).
This can be useful to watch system-wide activity of disks coming on-line or being ejected, volumes on disks being mounted or
unmounted, volumes being renamed, etc. However, this output must never be parsed; programs should become Disk Arbitration clients
instead.
For debugging information, such as the monitoring of applications dissenting (attempting to deny) activities for disks for which
they have registered an interest, you must use the logging features of the diskarbitrationd daemon. Programs needing this informa-
tion must become Disk Arbitration clients.
listFilesystems [-plist]
Show the file system personalities available for formatting in diskutil when using the erasing and partitioning verbs. This is a
subset of the complete set of personalities exported by the various file system bundles that may be installed in the system. Also
shown are some shortcut aliases for common personalities. See the FORMAT section below for more details. If -plist is specified,
then a property list instead of the normal user-readable output will be emitted.
unmount | umount [force] device
Unmount a single volume. Force will force-unmount the volume (less kind to any open files; see also umount (8)).
unmountDisk | umountDisk [force] device
Given a disk containing a partition map, unmount all of its volumes. That is, unmounts are attempted for the map's partitions con-
taining file system volumes, as well as for "virtual" volumes exported by storage systems which import data from the map's parti-
tions. Storage systems supported include APFS, AppleRAID, and CoreStorage.
Force will force-unmount the volumes (less kind to any open files; see also umount (8)).
You should specify a whole disk, but all volumes of the whole disk are attempted to be unmounted even if you specify a partition.
eject device
Eject a disk. Media will become offline for the purposes of being a data store for file systems or being a member of constructs
such as software RAID or direct data. Additionally, removable media will become eligible for safe manual removal; automatically-
removable media will begin its physical (motorized) eject sequence.
mount [readOnly] [-mountPoint path] device
Mount a single volume. If readOnly is specified, then the file system is mounted read-only, even if the volume's underlying file
system and/or device and/or media supports writing; even the super-user may not write to it; this is the same as the rdonly option
to mount (8). If a -mountPoint is specified, then that path, rather than the standard path of /Volumes/VolumeName, will be used
as the view into the volume file content; a directory at that path must already exist.
mountDisk device
Mount all mountable and UI-browsable volumes on the given partition map; that is, a mount is attempted on the directly-mountable
volume, if any, on each of the whole disk's partitions. However, "virtual" volumes, such as those are implied by e.g. Core Storage
Physical Volumes, AppleRAID Members, etc., are not handled. You should specify a whole disk, but all volumes of the whole disk
are attempted to be mounted even if you specify a partition.
rename | renameVolume device name
Rename a volume. Volume names are subject to file system-specific alphabet and length restrictions.
enableJournal device
Enable journaling on an HFS+ volume. This works whether or not the volume is currently mounted (the volume is temporarily mounted
if necessary). Ownership of the affected disk is required.
disableJournal [force] device
Disable journaling on an HFS+ volume. This normally works whether or not the volume is currently mounted (the volume is temporar-
ily mounted if necessary). If the force option is specified, then journaling is disabled directly on disk; in this case, the vol-
ume must not be mounted. Ownership of the affected disk is required.
moveJournal external journalDevice device
Create a 512MB Apple_Journal partition using the journalDevice partition to serve as a journal for the volume device. For best
results, journalDevice should be a partition on a different whole-disk than the volume itself.
The journal for device will be moved externally onto the newly created Apple_Journal partition.
Since the journalDevice you specify will invariably be larger than 512MB, a new HFS+ partition will be created following the
Apple_Journal partition to fill the remaining space.
Moving the journal works whether or not the volume is mounted, provided journaling is enabled on that volume. No errors are cur-
rently supported to flag attempts to move journals on volumes that do not have journaling enabled. If you have multiple volumes
for which you want external journals, each must have its own external Apple_Journal partition. Ownership of the affected disks is
required.
moveJournal internal device
Move the journal for device back locally (onto that same device). Ownership of the affected disk is required.
enableOwnership device
Enable ownership of a volume. The on-root-disk Volume Database at /var/db/volinfo.database is manipulated such that the User and
Group ID settings of files, directories, and links (file system objects, or "FSOs") on the target volume are taken into account.
This setting for a particular volume is persistent across ejects and injects of that volume as seen by the current OS, even across
reboots of that OS, because of the entries in this OS's Volume Database. Note thus that the setting is not kept on the target
disk, nor is it in-memory.
For some locations of devices (e.g. internal hard disks), consideration of ownership settings on FSOs is the default. For others
(e.g. plug-in USB disks), it is not.
When ownership is disabled, Owner and Group ID settings on FSOs appear to the user and programs as the current user and group
instead of their actual on-disk settings, in order to make it easy to use a plug-in disk of which the user has physical posses-
sion.
When ownership is enabled, the Owner and Group ID settings that exist on the disk are taken into account for determining access,
and exact settings are written to the disk as FSOs are created. A common reason for having to enable ownership is when a disk is
to contain FSOs whose User and Group ID settings, and thus permissions behavior overall, is critically important, such as when the
plug-in disk contains system files to be changed or added to.
See also the vsdbutil(8) command. Running as root is required.
disableOwnership device
Disable ownership of a volume. See enableOwnership above. Running as root is required.
verifyVolume device
Verify the file system data structures of a volume. The appropriate fsck program is executed and the volume is attempted to be
left mounted or unmounted as it was before the command. Any underlying Storage System (e.g. Core Storage, APFS) is verified
before the target volume itself. In certain cases, "live" verify, including of the boot volume, is supported. Ownership of the
disk to be verified is required.
repairVolume device
Repair the file system data structures of a volume. The appropriate fsck program is executed and the volume is attempted to be
left mounted or unmounted as it was before the command. Any underlying Storage System (e.g. Core Storage, APFS) is repaired
before the given target volume. In most cases (e.g. except mount-read-only), the target volume must be unmountable; in all cases,
the underlying storage media must be writable. "Live" repair (e.g. of a file-writable mounted volume) is not supported. Owner-
ship of the affected disk is required.
verifyDisk device
Verify the partition map layout of a whole disk intended for booting or data use on a Macintosh. The checks further include, but
are not limited to, the integrity of the EFI System Partition, the integrity of any Core Storage Physical Volume partitions, and
provisioning of space for boot loaders. Ownership of the disk to be verified is required; it must be a whole disk and must have a
partition map.
repairDisk device
Repair the partition map layout of a whole disk intended for booting or data use on a Macintosh. The repairs further include, but
are not limited to, the repair or creation of an EFI System Partition, the integrity of any Core Storage Physical Volume parti-
tions, and the provisioning of space for boot loaders. Ownership of the affected disk is required; it must be a whole disk and
must have a partition map.
resetFusion
For Fusion Drive machine hardware configurations, reset the disk devices in the machine to a factory-like state (one empty Fusion
volume). This command requires the machine to contain exactly one internal solid-state device (SSD) and one internal rotational
device (HDD); if so, you are prompted, and if you confirm, both devices are (re)-partitioned with GPT maps and a Core Storage
Fusion Drive volume is created. No system software is installed and no user data is restored. All data on the machine is lost,
including any "extra" partitions (e.g. for Boot Camp or other "user" purposes). You generally must be booted from the Internet
Recovery System (CMD-OPT-R) or from an externally-connected macOS boot disk (e.g. a USB drive), because you cannot erase a volume
with a currently-running macOS. Ownership of the affected disks is required.
eraseDisk format name [APM[Format] | MBR[Format] | GPT[Format]] device
Erase an existing disk, removing all volumes and writing out a new partitioning scheme containing one new empty file system vol-
ume. If the partitioning scheme is not specified, then an appropriate one for the current machine is chosen. Format is discussed
below in the section for the partitionDisk verb. Ownership of the affected disk is required.
eraseVolume format name device
Write out a new empty file system volume (erasing any current file system volume) on an existing partition. The partition remains
but its data is lost. Format is discussed below in the section for the partitionDisk verb.
If you specify Free Space for format, the partition itself is deleted (removed entirely) from the partition map instead of merely
being erased. Ownership of the affected disk is required.
reformat device
Erase an existing volume by writing out a new empty file system of the same personality (type) and with the same volume name.
Ownership of the affected disk is required.
eraseOptical [quick] device
Erase optical media (CD/RW, DVD/RW, etc.). Quick specifies whether the disc recording system software should do a full erase or a
quick erase. Ownership of the affected disk is required.
zeroDisk [force] device
Erase a device, writing zeros to the media. The device can be a whole-disk or a partition. In either case, in order to be useful
again, zeroed whole-disks will need to be (re)partitioned, or zeroed partitions will need to be (re)formatted with a file system,
e.g. by using the partitionDisk, eraseDisk, or eraseVolume verbs. If you desire a more sophisticated erase algorithm or if you
need to erase only free space not in use for files, use the secureErase verb. The force parameter causes best-effort, non-error-
terminating, forced unmounts and shared-mode writes to be attempted; however, this is still no guarantee against drivers which
claim the disk exclusively. In such cases, you may have to first unmount all overlying logical volumes (e.g. CoreStorage or
AppleRAID). If a disk is partially damaged in just a certain unlucky way, you might even have to un-install a kext or erase the
disk elsewhere. Ownership of the affected disk is required.
randomDisk [times] device
Erase a whole disk, writing random data to the media. Times is the optional (defaults to 1) number of times to write random
information. The device can be a whole-disk or a partition. In either case, in order to be useful again, randomized whole-disks
will need to be (re)partitioned, or randomized partitions will need to be (re)formatted with a file system, e.g. by using the
partitionDisk, eraseDisk, or eraseVolume verbs. If you desire a more sophisticated erase algorithm or if you need to erase only
free space not in use for files, use the secureErase verb. Ownership of the affected disk is required.
secureErase [freespace] level device
Erase, using a "secure" (but see the NOTE below) method, either a whole-disk (including all of its partitions if partitioned), or,
only the free space (not in use for files) on a currently-mounted volume. Secure erasing makes it harder to recover data using
"file recovery" software.
Erasing a whole-disk will leave it useless until it is partitioned again. Erasing freespace on a volume will leave your files
intact, indeed, from an end-user perspective, it will appear unchanged, with the exception that it will have attempted to make it
impossible to recover deleted files.
If you need to erase all contents of a partition but not its hosting whole-disk, use the zeroDisk or randomDisk verbs. Ownership
of the affected disk is required.
Level should be one of the following:
o 0 - Single-pass zero-fill erase.
o 1 - Single-pass random-fill erase.
o 2 - US DoD 7-pass secure erase.
o 3 - Gutmann algorithm 35-pass secure erase.
o 4 - US DoE algorithm 3-pass secure erase.
NOTE: This kind of secure erase is no longer considered safe. Modern devices have wear-leveling, block-sparing, and possibly-per-
sistent cache hardware, which cannot be completely erased by these commands. The modern solution for quickly and securely erasing
your data is encryption. Strongly-encrypted data can be instantly "erased" by destroying (or losing) the key (password), because
this renders your data irretrievable in practical terms. Consider using APFS encryption (FileVault).
partitionDisk device [numberOfPartitions] [APM[Format] | MBR[Format] | GPT[Format]] [part1Format part1Name part1Size part2Format part2Name
part2Size part3Format part3Name part3Size ...]
(re)Partition a disk, removing all volumes. All volumes on this disk will be destroyed. The device parameter specifies which
whole disk is to be partitioned. The optional numberOfPartitions parameter specifies the number of partitions to create; if given
then the number of parameter triplets (see below) is expected to match; else, the number of triplets alone given will determine
the number of partitions created.
The optional partitioning scheme parameter forces a particular partitioning scheme; if not specified, a suitable default is cho-
sen. They are:
o APM[Format] specifies that an Apple Partition Map scheme should be used. This is the traditional Apple partitioning
scheme used to start up a PowerPC-based Macintosh computer, to use the disk as a non-startup disk with any Mac, or to
create a multiplatform compatible startup disk.
o MBR[Format] specifies that a Master Boot Record scheme should be used. This is the DOS/Windows-compatible partitioning
scheme.
o GPT[Format] specifies that a GUID Partitioning Table scheme should be used. This is the partitioning scheme used to
start up an Intel-based Macintosh computer.
For each partition, a triplet of the desired file system format, volume name, and size must be specified. Several other diskutil
verbs allow these triplets as well (and for them, the numberOfPartitions parameter is also optional). The triplets must be as
follows:
o Format names are of the form jhfs+, HFS+, MS-DOS, etc.; a list of formattable file systems (more precisely, specific
file system personalities exported by the installed file system bundles) and common aliases is available from the
listFilesystems verb.
Format guides diskutil both in what partition type to set for the partitions (slices) as well as what file system struc-
tures to initialize therein, using the file system bundle's plist's FormatExecutable setting which usually points to the
appropriate formatter program such as newfs_hfs(8).
You can specify a format of Free Space to skip an area of the disk.
You can specify the partition type manually and directly with a format of %<human-readable partition type>% such as
%Apple_HFS% or %<GPT partition type UUID constant>% such as %48465300-0000-11AA-AA11-00306543ECAC%; these imply a name
of %noformat% (below). Human-readable types must be known to the system but UUID types (GPT scheme only) can be arbi-
trary.
o Names are the initial volume names; they must conform to file system specific restrictions.
If a name of %noformat% is specified, then the partition is left blank such that the partition space is carved out, the
partition type is set according to the file system format name or explicit type, the partition space is partially erased
("wiped"), but a file system structure is not initialized with any file system's formatter program (e.g. newfs_hfs(8);
this is useful for setting up partitions that will contain user-defined (not necessarily file system) data.
For a triplet whose format is Free Space or a directly-specified partition type, its name is ignored but a dummy name
must nevertheless be present.
o Sizes are floating point numbers followed by a letter or percent sign as described in the SIZES section at the end of
this page (e.g. 165536000, 55.3T, 678M, 75%, R).
In addition to explicitly-requested partitions, space (gaps) might be allocated to satisfy certain filesystems' position and
length alignment requirements; space might be allocated for possible future booter partition insertion; and indeed, actual booter
partitions might be implicitly created.
In particular, there is a rule that unrecognized partitions 1GiB or larger automatically acquire booters. Thus, if you create an
arbitrary partition with e.g. diskutil partitionDisk disk0 gpt %11112222-1111-2222-1111-111122221111% %noformat% 3gib jhfs+
Untitled r, then a booter partition will also be created. You can always delete that booter with diskutil eraseVolume "Free Space"
dummy disk0s3.
The last partition is usually automatically lengthened to the end of the partition map (disk). You can specify an exact size for
your last partition by specifying it as the penultimate triplet and specifying an additional (last) triplet as Free Space. Or you
can use the R (remainder) size specifier for one of your middle partitions while specifying an exact size for your last partition.
Ownership of the affected disk is required.
resizeVolume device limits | mapsize [-plist] | R | size [numberOfPartitions] [part1Format part1Name part1Size part2Format part2Name
part2Size part3Format part3Name part3Size ...]
Non-destructively resize a volume (partition); you may increase or decrease its size. Alternatively, take no action and print
information.
Specifying limits instead of size takes no action, but instead prints the range of valid values for the target partition, taking
into account current file system and partition map conditions such as files in use and other (immovable) partitions following the
target.
Specifying mapsize instead of size takes no action, but instead prints the size of the encompassing whole-disk device, as well as
the size of the entire partition map (all partitions less map overhead). The whole-disk device might be larger than the partition
map if the whole-disk device has grown since the partition map was created. Growing a whole-disk device is possible with certain
enterprise disk (RAID) systems.
The -plist option will print partition or whole-disk size inquiry information in property list format.
You can grow a volume (partition) (back) to its maximum size possible, provided no new partitions have been created that are in
the way, by specifying R for the new volume size. You should use R instead of attempting an absolute value such as 100% because
the latter cannot count partition map overhead.
When decreasing the size, new partitions may optionally be created to fill the newly-freed space. To do this, specify the
numberOfPartitions, format, name, and size parameters in the same manner as the triplet description for the partitionDisk verb.
Resizing a volume that is currently set as the computer's startup disk will invalidate that setting; use the Startup Disk System
Preferences panel or bless (8) to reset the resized volume as the startup disk.
Device refers to a volume; the volume's file system must be journaled HFS+. Valid sizes are a number followed by a capital letter
multiplier or percent sign suffix as described in the SIZES section at the end of this page (e.g. 1.5T, 128M, 50%). Ownership of
the affected disk is required.
splitPartition device [numberOfPartitions] [part1Format part1Name part1Size part2Format part2Name part2Size part3Format part3Name part3Size
...]
Destructively split a volume into multiple partitions. You must supply a list of new partitions to create in the space of the old
partition; specify these with the numberOfPartitions, format, name, and size parameters in the same manner as the triplet descrip-
tion for the partitionDisk verb.
For one of your triplets, you can optionally specify the R meta-size in lieu of a constant number value for the size parameter:
the substituted value will be exactly the amount of space necessary to complete the re-filling of the original partition with all
of your triplets.
Device refers to a volume. Ownership of the affected disk is required.
mergePartitions [force] format name fromDevice toDevice
Merge two or more partitions on a disk. All data on merged partitions other than the first will be lost. Data on the first par-
tition will be lost as well if the force argument is given.
If force is not given, and the first partition has a resizable file system (e.g. JHFS+), the file system will be preserved and
grown in a data-preserving manner; your format and name parameters are ignored in this case. If force is not given, and the first
partition is not resizable, you are prompted if you want to format. You will also be prompted to format if the first partition
has an (HFS) Allocation Block Size which is too small to support the required growth of the first partition; see the -b option for
newfs_hfs (8).
If force is given, the final resulting partition is always (re)formatted. You should do this if you wish to (re)format to a new
file system type. You will be prompted to confirm.
Format and name must always be given, but they have an effect only when force is given.
Merged partitions are required to be ordered sequentially on disk (see diskutil list for the actual on-disk ordering). All parti-
tions in the range, except for the first one, must be unmountable. Ownership of the affected disk is required.
APFS | ap apfsVerb [...]
Apple APFS is a system of virtual volumes. APFS verbs can be used to create, manipulate and destroy APFS Containers and their
APFS Volumes. Apple APFS defines these types of objects:
o Container - An APFS Container imports one or more APFS Physical Store disks and exports zero or more APFS Volume disks.
Zero or more APFS Containers can exist in (might be attached to) the system at any one time.
While attached, the "handle" by which an APFS Container is identified is by its APFS Container Reference disk (device).
You should treat this as an opaque reference token.
The Container Reference disk is a synthesized whole-disk which is exported by APFS for identification purposes only; it
has no storage. It is associated with the AppleAPFSContainerScheme node in the IO Registry. While APFS Volume device
identifiers appear to be of a related form, you should never use the Container Reference as a basis to create device
identifiers yourself; use the listing verbs with their plist options instead.
o Physical Store - An APFS Physical Store is a disk which is imported into (that is, which backs, indeed defines) an APFS
Container. An APFS Container can import more than one Physical Store, e.g. for Fusion-style Containers.
An APFS Physical Store disk is not necessarily a disk from a partition map; it could be e.g. an AppleRAID Set disk.
Therefore, you must never assume that an APFS Physical Store's disk identifier is a 2-part form such as disk0s2.
o Volume - An APFS Volume is an [un]mountable file system volume which is exported from an APFS Container. Zero or more
APFS Volumes may be exported out of an APFS Container.
APFS Volumes have no specified "size" (capacity). Instead, all APFS Volumes consume capacity out of the remaining free
space of their parent APFS Container, consuming or returning such capacity as user file data is added or deleted. Note
that this means that all Volumes within a Container compete for the Container's remaining capacity. However, you can
manage Volume allocation with the optional reserve and quota size values.
The optional reserve size requests an assured minimum capacity for an APFS Volume. If successfully created, the Volume
is guaranteed to be able to store at least this many bytes of user file data. Note that beyond this, the Volume might be
able to store even more until constrained by reaching zero free space in its parent Container or by reaching a quota, if
any. You can use a reserve to prevent running out of capacity due to competition from other Volumes or from a Container
shrink attempt.
The optional quota size applies a maximum capacity to an APFS Volume, placing a limit on the number of bytes of user
file data which can be stored on the Volume. Note that you might not be able to reach this limit if its parent Container
becomes full first. You can use a quota to enforce accounting or to manage against "unfair" premature filling-up of the
parent Container due solely to this Volume at the expense of sibling Volumes.
Efficient file copy cloning (copy-on-write) is supported (see copyfile (3)'s COPYFILE_CLONE).
Optional file-level encryption is supported.
The format of an APFS Volume's device identifier is that of a slice disk of a special whole-disk; both disks are synthe-
sized by APFS. The "whole" identifier number (a positive possibly-multi-digit integer) is arbitrary, and the "slice"
numbers (positive possibly-multi-digit integers) count up from 1 with each new Volume. Deleting Volumes may cause gaps
in the numbering. This form appears the same as a partition (map) scheme and partitions, but it is completely unre-
lated. For example: If "disk3s2" is a Physical Store defining a Container, then "disk5s1", "disk5s2", and "disk5s3"
might be the Container's Volumes; "disk5" exists but is never used directly.
Although it has a device node, an APFS Volume's data may only be accessed through its files; you cannot open an APFS
Volume device node to "directly" access its on-disk bytes.
o Snapshot - An APFS Snapshot represents a read-only copy of its parent APFS Volume, frozen at the moment of its creation.
An APFS Volume can have zero or more associated APFS Snapshots.
APFS Snapshots are neither listed nor discoverable when their Volume is not mounted. Snapshots are uniquely identified
within their parent Volume's namespace by either a numeric identifier (preferred) or by their name; Snapshots can be
renamed, but APFS will never allow duplication of names (within a Volume) to occur.
APFS Snapshots are mountable; when this occurs, its mount point (separate from and simultaneous with its parent Volume)
provides a read-only historic version of the Volume content at Snapshot creation time.
You can revert the present state of an APFS Volume back to equality with a Snapshot in its history. This is a destruc-
tive reset/restore operation: Once a Volume is reverted, it cannot be brought forward. Any Snaphots between the revert
point and the present are lost as well.
You can delete a Snapshot; this removes the possibility of ever reverting to that Snapshot's state, but does not affect
the Volume's present-time content.
An APFS Snapshot mount point's "source device" (its statfs (2) f_mntfromname shown by mount (8)) is not a device node
(e.g. disk0s2) as is common; it is the Snapshot name followed by the '@' character and the "parent" Volume's device
node, e.g. "SnapName123@/dev/disk2s1". See mount_apfs -s (8) and fs_snapshot_create (2).
APFS itself has no provision for backing up your data. Backups should be always be performed on a regular basis and before modi-
fying any APFS Container using these commands.
The following is a list of APFS sub-verbs with their descriptions and individual arguments.
list [-plist] [containerReferenceDevice]
Display APFS objects as a tree. AFPS Container(s) are shown with their imported Physical Store(s) and exported Vol-
ume(s).
All currently-attached APFS Containers in the system are listed unless you specify a containerReferenceDevice, which
limits the output to that specific APFS Container family.
If -plist is specified, then a property list will be emitted instead of the normal user-readable output.
convert device [-dryrun] [-prebootSource yourStagingDirectory] [-noPrebootAdditions]
Non-destructively convert an HFS volume to an APFS Container with a single (but see below) APFS Volume. The APFS Con-
tainer can later be manipulated (e.g. adding and deleting APFS Volumes) as usual. This verb can be used to convert
nonbootable "data"-only volumes as well as "macOS" volumes (see below).
The source HFS volume can be located on a GPT partition or on an encrypted or non-encrypted, Fusion or non-Fusion
CoreStorage logical volume (LV). In the latter case, the CoreStorage logical volume group (LVG) is dismantled, includ-
ing automatic removal of any related Boot Camp Assistant partition(s).
If -dryrun is specified, all calculations, checks, and some data moving is performed, but your disk is left as valid
HFS (headers remain declared as HFS, partition types are left alone, etc).
For volumes currently or planned to be macOS-bearing (and bootable), you can optionally specify -prebootSource with
your own staging directory of macOS boot items; a Preboot Role APFS Volume with a UUID directory will automatically be
created as part of the conversion process to facilitate macOS bootstrap. Normally your directory should be writable;
additional (cryptographic and EFI rendering) items are automatically added to your directory prior to conversion and
are not removed afterwards. You can opt-out of automatic item addition with the -noPrebootAdditions option.
Ownership of the affected disks is required.
create device [device] name
Convenience verb which creates an empty APFS Container and then adds one APFS Volume with the given name. The APFS
Volume will have default attributes such as no encryption, no capacity reserve nor quota, etc. If you specify more
than one device, a Fusion Container is created, with the performance roles assigned automatically. This is a combina-
tion of the diskutil apfs createContainer and diskutil apfs addVolume verbs.
Ownership of the affected disks is required.
createContainer [-main] device [-secondary] [device]
Create an empty APFS Container. The device(s) specified become APFS Physical Stores.
If you specify more than one device, a Fusion Container is created, with the performance roles assigned automatically
(preferred) unless you use the -main and -secondary options, in which case, the secondary device is assumed by APFS's
performance algorithms to be on "slower" hardware. The secondary device is often used to store associated "auxiliary"
data, such as a Boot Camp Assistant partition. If you need to make a choice among multiple hardware devices, you
should assign any rotational (not solid state) and/or larger devices to be secondary.
You may then add APFS Volumes with the diskutil apfs addVolume verb.
Ownership of the affected disks is required.
deleteContainer [-force] containerReferenceDevice | physicalStoreDevice [name]
Destroy an existing APFS Container, including all of its APFS Volumes. You can identify the APFS Container by its Con-
tainer Reference disk (preferred), or by its Physical Store disk.
The APFS Volumes are unmounted first; this process may not succeed if one or more is busy. If this happens, the opera-
tion is aborted and everything is left intact. Otherwise, all APFS Volumes are deleted as well as its APFS Container,
and the APFS Container's former Physical Store disks will be reformatted as HFS; data on all APFS Volumes will be lost.
You can optionally specify a new name, or else "Untitled" will be chosen. If there were multiple Physical Stores, a
space and a number suffix is added for each.
If your APFS Container is damaged, a Container Reference for it might not exist or it might not be functional. In this
case, you can reclaim your former APFS Physical Store disk(s) by specifying the -force option; this activates an alter-
nate last-resort mode. In this mode, if you had more than one Physical Store (e.g. the Fusion case) and the Container
is sufficiently damaged, you might have to delete each Physical Store manually. You should normally avoid this mode.
Ownership of the affected disks is required.
resizeContainer containerReferenceDevice | physicalStoreDevice limits [-plist] | size [part1Format part1Name part1Size part2Format
part2Name part2Size part3Format part3Name part3Size ...]
Resize an existing APFS Container by specifying either an APFS Container Reference (preferred) or an APFS Physical
Store partition, plus a proposed new size. Alternatively, take no action and print constraint information. The opera-
tion is live, non-destructive, and does not mount or unmount any APFS Volumes.
If you specify an APFS Container Reference and that Container imports more than one Physical Store (in e.g. Fusion set-
ups), the appropriate Physical Store will be chosen automatically.
Specifying limits instead of a size causes no action to be taken, but instead prints a range of valid values, taking
into account various constraints; the -plist option will print this information in property list format.
Shrinks are constrained by the amount of data usage by all APFS Volumes on the targeted or implied APFS Container.
Contributing to this data usage is the file content on the APFS Volumes, the existence of quotas and/or reserves, the
usage of APFS Snapshots (e.g. by Time Machine), and metadata overhead.
Grows are constrained by the amount of partition map free space trailing the targeted or implied Physical Store parti-
tion.
When shrinking, new partitions may optionally be created to fill the newly-freed space. To do this, specify the
format, name, and size parameters in the same manner as the triplet description for the partitionDisk verb.
You can specify a size of zero (0) to grow the targeted APFS Physical Store such that all remaining space is filled to
the next partition or the end of the partition map.
Ownership of the affected disks is required, and all APFS Volumes on the Container must be unlocked.
addVolume containerReferenceDevice filesystem name [-passprompt] | [-passphrase passphrase] | [-stdinpassphrase] [-passphraseHint
passphraseHint] [-reserve reserve] [-quota quota] [-role roles] [-nomount] [-mountpoint mountpoint]
Add a new APFS Volume to an existing APFS Container. Files can then be stored on this newly-created APFS Volume.
The filesystem parameter sets the permanent APFS personality for this new APFS Volume; you should specify APFS or
Case-sensitive APFS.
The new APFS Volume will be unencrypted unless you specify one of the passphrase options, in which case the volume will
be encrypted from the beginning of its existence (as opposed to having encryption applied later); the user which is
added will be the "Disk User". The optional passphraseHint is a user-defined string that can be displayed even while
an encrypted APFS Volume is locked.
APFS Volumes have no fixed size; they allocate backing store on an as-needed basis. You can specify the reserve param-
eter to guarantee a minimum amount of space for your volume; at least that many bytes will be available for file data.
You can also specify the quota parameter to limit your volume's file usage to a maximum amount; no more than that many
bytes will be available for file data, even if there is otherwise enough space in the parent APFS Container. You can
specify both reserve and quota simultaneously; however, the reserve is not allowed to be larger than the quota.
APFS Volumes can carry certain metadata hint flags; you can supply the role parameter with any combination of one or
more of the characters BRVI, or 0 as a no-op for scripting convenience.
The new APFS Volume is explicitly mounted after creation; you can specify -nomount to leave it unmounted, or, you can
supply a "custom" mountpoint path, in which case you must be root, the directory must already exist, and you must
delete the directory yourself when you unmount.
Ownership of the affected disks is required.
deleteVolume volumeDevice
Remove the given APFS Volume from its APFS Container. All of the Volume's data will be lost.
Ownership of the affected disks is required.
eraseVolume volumeDevice -name newName [-passprompt] | [-passphrase passphrase] | [-stdinpassphrase] [-passphraseHint
passphraseHint]
Erase the contents of an existing APFS Volume; all of its data will be lost. Unlike diskutil apfs deleteVolume, the
APFS Volume is not removed from its APFS Container.
The "new" APFS Volume will inherit the APFS file system type (Case-sensitive or not) but will not inherit attributes
such as name, reserve, quota, or encryption status.
The "new" APFS Volume be unencrypted, unless you supply passphrase options in the same manner as diskutil apfs
addVolume in which case it will be encrypted and initially accessible by the "Disk User".
If you need more control, you should delete and (re-)add the Volume instead.
Ownership of the affected disks is required.
changeVolumeRole | chrole volumeDevice roles
Change the role metadata flag bits of an existing APFS Volume.
The roles should be any combination of one or more of the characters brviBRVI in much the same manner as diskutil apfs
addVolume above, in which unspecified flags are left alone, use of lower-case causes flags to be cleared, and use of
upper-case causes flags to be set. Alternatively, clear will remove all flags, or 0 can be used as a no-op for script-
ing convenience. You should not make any assumptions about the usage or legal combinations of role bits.
Ownership of the affected disks is required.
unlockVolume | unlock volumeDevice [-user disk | -user cryptoUserUUID | -recoverykeychain file] [-passphrase passphrase] |
[-stdinpassphrase] [-nomount | -mountpoint mountpoint] [-verify] [-plist]
Unlock and mount an encrypted and locked APFS Volume or verify a passphrase.
If you do not supply the -user option, then all cryptographic users on that APFS Volume are searched for a match; if
you supply -user disk then the Disk UUID (which equals the APFS Volume UUID) user is assumed; if you supply -user with
a UUID then that specific user is assumed; if you instead supply -recoverykeychain then the Institutional Recovery user
(see below) is assumed.
You will be prompted interactively for a passphrase unless you specify a passphrase parameter with -passphrase or pipe
your passphrase into stdin and use -stdinpassphrase.
As an alternative to a passphrase, you can specify -recoverykeychain with a full path to a keychain file if an Institu-
tional Recovery Key has been previously set up on the APFS Volume. The keychain must be unlocked; see security(1) and
fdesetup(8) for more information.
You can skip the explicit mounting step with the -nomount option, or specify a "custom" mountpoint with the -mountpoint
option. If you specify your own mountpoint path, it must exist and you must have write privileges on it (e.g. usually
you must be root).
Specifying -verify will test passphrase correctness without affecting the locked or unlocked state.
If -plist is specified, then a property list will be emitted instead of the normal user-readable output; additional
details are available here.
To re-lock the volume, unmount it, e.g. with diskutil unmount or diskutil apfs lockVolume.
Ownership of the affected disks is required.
lockVolume | lock volumeDevice
Unmount and lock an encrypted unlocked APFS Volume. This is mostly a synonym for diskutil unmount.
Ownership of the affected disks is required.
listCryptoUsers | listUsers | listCryptoKeys | listKeys [-plist] volumeDevice
Show all cryptographic users and special-purpose (e.g. recovery) "users" (keys) that are currently associated with the
given APFS Volume, each by their Cryptographic User UUID and usage "type".
The usual purpose of an APFS Cryptographic User is to authenticate for unlocking its APFS Volume; any of its users can
do so.
An APFS Volume need not be encrypted in order to contain crypto users; indeed, other than the Disk User, they should be
added before encrypting.
Types of Cryptographic Users include at-most-one-per-Volume "Disk" user, whose UUID value always matches its Volume's
UUID; iCloud or personal "Recovery Keys", which are not users per se, but instead store partial crypto keys and are
associated with corresponding "Recovery Users" and have fixed-constant UUID values; and "Open Directory" users, whose
UUID values match corresponding local macOS Open Directory account user GUIDs.
If -plist is specified, then a property list will be emitted instead of the normal user-readable output.
changePassphrase | changeCryptoUserPassphrase | passwd volumeDevice -user disk | cryptoUserUUID [-oldPassphrase oldPassphrase |
-oldStdinpassphrase] [-newPassphrase newPassphrase | -newStdinpassphrase]
Change the passphrase of the given cryptographic user associated with the given APFS Volume.
The old and new passphrases are specified in the same manner as diskutil apfs addVolume; you will be interactively
prompted as necessary if you do not specify both.
Ownership of the affected disks is required.
setPassphraseHint | setCryptoUserPassphraseHint | hint volumeDevice -user disk | cryptoUserUUID -hint hintMessage | -clear
Set an arbitrary hint string to aid recall of a passphrase for the given cryptographic user associated with the given
APFS Volume. Specifying -clear will clear any existing hint (no hint is the default).
Ownership of the affected disks is required.
encryptVolume | encrypt | enableFileVault volumeDevice -user disk | existingCryptoUserUUID [-passphrase existingOrNewPassphrase |
-stdinpassphrase]
Start encryption of a currently-unencrypted APFS Volume ("Enable FileVault"). Depending on hardware, the operation may
be accomplished immediately, or it may proceed "in the background".
You can supply an existing cryptographic user UUID, in which case you must supply its corresponding passphrase, or you
can supply disk (or the Disk/Volume UUID) and the corresponding passphrase of the "Disk User", provided the "Disk User"
already exists.
Alternatively, if no users exist yet on this APFS Volume, you can still supply disk (or the Disk/Volume UUID), and a
"Disk User" will be created with a new passphrase which you supply. This is the only way using diskutil in which an
APFS Volume that has no cryptographics users on it yet can acquire the first such user.
The passphrase, interactive or not, is specified in the same manner as diskutil apfs addVolume.
Ownership of the affected disks is required.
decryptVolume | decrypt | disableFileVault volumeDevice [-user disk | existingCryptoUserUUID | -recoverykeychain file]
[-passphrase existingPassphrase | -stdinpassphrase]
Start decryption of a currently-encrypted APFS Volume ("Disable FileVault"). Depending on hardware, the operation may
be accomplished immediately, or it may proceed "in the background".
The APFS Volume must be unlocked before beginning this operation.
One (any) of the existing cryptographic user and passphrase pair(s) on the APFS Volume must be satisfied (this is not
the case prior to macOS 10.13.2). You choose that user with existingCryptoUserUUID; if you do not supply it (or supply
the word disk) then the "Disk User" is assumed.
As an alternative to a passphrase, you can specify -recoverykeychain with a full path, in the same fashion as the
unlockVolume verb.
If you do not supply a passphrase, yet one is required, you will be prompted interactively by cryptographic user UUID.
Ownership of the affected disks is required.
listSnapshots | list[Volume]Snaps[hots] [-plist] volumeDevice
Show all snapshots that are currently associated with the given APFS Volume, each with information such as its numeric
XID identifier, name, and possibly other fields.
If -plist is specified, then a property list will be emitted instead of the normal user-readable output.
deleteSnapshot volumeDevice -xid xid | -name name
Remove the given APFS Snapshot from its APFS Volume. The ability to restore the state of the APFS Volume back to that
point in its evolution will be lost.
Ownership of the affected disks is required.
defragment containerDevice | volumeDevice status | enable | disable
Manage defragmentation of APFS at the Container or Volume level. The "status" command shows whether automatic defrag-
mentation is enabled. You can set this automatic defragmentation behavior by specifying "enable" or "disable".
Ownership of the affected disks is required.
appleRAID | ar raidVerb [...]
AppleRAID verbs can be used to create, manipulate and destroy AppleRAID volumes (Software RAID). AppleRAID supports three basic
types of RAID sets:
o "stripe" - Striped Volume (RAID 0)
o "mirror" - Mirrored Volume (RAID 1)
o "concat" - Concatenated Volume (Spanning)
Of these three basic types, only the "mirror" type increases fault-tolerance. Mirrors may have more than two disks to further
increase their fault-tolerance. Striped and concatentated volumes are, in fact, more vulnerable to faults than single disk vol-
umes.
From these basic types, "stacked" or "nested" RAID volumes can be created. Stacked RAID sets that make use of mirrored RAID sets
are fault-tolerant. For example, these are some of the more common combinations of stacked RAID sets:
o RAID 50 - A striped RAID set of hardware RAID 5 disks.
o RAID 10 - A striped RAID set of mirrored RAID sets.
o RAID 0+1 - A mirrored RAID set of striped RAID sets.
o Concatenated Mirror - A concatenation of mirrored RAID sets.
When creating new RAID sets or adding disks, if possible, it is better to specify the entire disk instead of a partition on that
disk. This allows the software to reformat the entire disk using the most current partition layouts. When using whole disks, the
type of partitioning used is selected based on the platform type (PPC = APMFormat, Intel = GPTFormat). GPT and APM partition for-
mats cannot be mixed in the same RAID set.
In addition to whole disk and partition device names, AppleRAID uses UUIDs to refer to existing RAID sets and their members.
Existing RAID sets may also be specified by mount point (e.g. /Volume/raidset). In many cases, using the UUID for the device
argument is preferred because disk device names may change over time when disks are added, disks are removed or when the system is
rebooted. If RAID members have been physically disconnected from the system or are no longer responding, you must use the mem-
ber's UUID as the command argument. Messages in the system log will refer to RAID sets and their member disks by UUID. For more
information on specifying device arguments, see the DEVICES section below.
AppleRAID is not a replacement for backing up your data. Backups should be always be performed on a regular basis and before mod-
ifying any RAID set using these commands.
The following is a list of appleRAID sub-verbs with their descriptions and individual arguments.
list [-plist | UUID]
Display AppleRAID volumes with current status and associated member disks. If UUID is specified, only list the RAID
set with that AppleRAID Set UUID. If -plist is specified, then a property list will be emitted instead of user-format-
ted output. The -plist and UUID arguments may not both be specified. diskutil listRAID and diskutil checkRAID are
deprecated synonyms for diskutil appleRAID list.
create mirror | stripe | concat setName format devices ...
Create a new RAID set consisting of multiple disks and/or RAID sets. setName is used for both the name of the created
RAID volume and the RAID set itself (as displayed in list). e.g. 'diskutil createRAID stripe MyArray JHFS+ disk1 disk2
disk3 disk4'. Ownership of the affected disks is required. diskutil createRAID is a deprecated synonym for diskutil
appleRAID create.
delete raidVolume
Destroy an existing RAID set. If the RAID set is a mirror with a resizable file system, delete will attempt to convert
each of the member partitions back into a non-RAID volume while retaining the contained file system. For concatenated
RAID sets with a resizable file system, delete will attempt to shrink the file system to fit on the first member parti-
tion and convert that to a non-RAID volume. Ownership of the affected disks is required. diskutil destroyRAID is a
deprecated synonym for diskutil appleRAID delete.
repairMirror raidVolume newDevice
Repair a degraded mirror by adding a "new" disk given as newDevice to the RAID mirror set whose exported disk device or
set UUID is given as raidVolume. The new disk must be the same size or larger than the existing disks in the RAID set.
After running this command, you should manually remove the old (orphaned, failed) member(s) with diskutil appleRAID
remove. Ownership of the affected disk is required. diskutil repairMirror is a deprecated synonym for diskutil
appleRAID repairMirror.
add type newDevice raidVolume
Add a new member or hot spare to an existing RAID set. Type can be either member or spare. New disks are added live,
the RAID volume does not need to be unmounted. Mirrored volumes support adding both members and hot spares, concate-
nated volumes only support adding members. When adding to a mirrored RAID set, the new disk must be the same size or
larger than the existing disks in the RAID set. Adding a hot spare to a mirror will enable autorebuilding for that
mirror. Adding a new member to a concatenated RAID set appends the member and expands the RAID volume. Ownership of
the affected disk is required. diskutil addToRAID is a deprecated synonym for diskutil appleRAID add.
remove oldDevice raidVolume
Remove a member or spare from an existing RAID set. Old disks are removed live; the RAID volume does not need to be
unmounted. For missing devices, oldDevice must be the device's UUID. Online mirror members with a resizable file sys-
tem will be converted to non-RAID volumes, spare and offline members will be marked free. For concatenated RAID sets,
only the last member can be removed. For resizable file systems remove will first attempt to shrink the concatenated
RAID set so that the file system fits on the remaining disks. Ownership of the affected disk is required. diskutil
removeFromRAID is a deprecated synonym for diskutil appleRAID remove.
enable mirror | concat device
Convert a non-RAID disk partition containing a resizable file system (such as JHFS+) into an unpaired mirror or single
disk concatenated RAID set. Disks that were originally partitioned on Mac OS X 10.2 Jaguar or earlier or were parti-
tioned to be Mac OS 9 compatible may not be resizable. Ownership of the affected disk is required. diskutil
enableRAID is a deprecated synonym for diskutil appleRAID enable.
update key value raidVolume
Update the key value parameters of an existing RAID set. Valid keys are:
o AutoRebuild - If true, the system attempts to rebuild degraded mirrored volumes automatically. When looking
for devices for rebuild, AppleRAID first looks for hot spares and then degraded members. Use a value of "1"
for true and "0" for false.
o SetTimeout - Controls how long the system waits (in seconds) for a missing device before degrading a mirrored
raid set. Also controls the amount of time you have to disconnect all devices from an unmounted mirror with-
out degrading it.
Ownership of the affected disk is required. diskutil updateRAID is a deprecated synonym for diskutil appleRAID update.
coreStorage | cs coreStorageVerb [...]
CoreStorage verbs can be used to create, manipulate and destroy CoreStorage volumes.
CoreStorage maintains a world of virtual disks, somewhat like RAID, in which one can easily add or remove imported backing store
disks, as well as exported usable volumes, to or from a pool (or several pools). This provides the user with flexibility in allo-
cating their hardware; user or operating system data can span multiple physical disks seamlessly, for example.
Apple CoreStorage defines four types of objects, instances of which are uniquely represented by a UUID:
o Logical Volume Group (LVG)
o Physical Volume (PV)
o Logical Volume Family (LVF)
o Logical Volume (LV)
The Logical Volume Group (LVG) is the top or "pool" level; zero or more may exist during any OS boot time session.
An LVG imports one or more Physical Volumes (PVs). A PV represents a device that feeds the LVG storage space; a PV is normally
real media but it can be a disk image or even an AppleRAID Set. A disk offered to be a PV must be a partition and the encompassing
scheme must be GPT.
An LVG exports zero or more Logical Volume Families (LVFs). An LVF contains properties which govern and bind together all of its
descendant Logical Volumes (LVs). These properties provide settings for Full Disk Encryption (FDE) (such as whether the LVG is
encrypted, which users have access, etc) and other services. However, at the present time, for new LVF creation, only zero or one
LVF per LVG is supported.
A Logical Volume Family (LVF) exports one or more Logical Volumes (LVs). However, at the present time, only and exactly one LV
per LVF is supported.
A Logical Volume (LV) exports a dev node, upon which a file system (such as Journaled HFS+) resides.
For more information on specifying device arguments, see the DEVICES section below.
CoreStorage is not a replacement for backing up your data. Backups should be always be performed on a regular basis and before
modifying any CoreStorage volumes using these commands.
The following is a list of coreStorage sub-verbs with their descriptions and individual arguments.
list [-plist | UUID]
Display a tree view of the CoreStorage world for all current logical volume groups (LVGs) with member disks (PVs) and
exported volumes (LVFs and LVs), with properties and status for each level. If -plist is specified then a property
list will be emitted instead of the formatted tree output; the UUIDs can be used with the diskutil coreStorage
information verb to get properties for the object represented by that UUID. If UUID is specified then an attempt is
made to list only that UUID (whatever type of CoreStorage object it may represent). The -plist and UUID arguments may
not both be specified.
info | information [-plist] UUID | device
Display properties of the CoreStorage object (LVG, PV, LVF, or LV) associated with the given CoreStorage UUID or disk.
convert device [-stdinpassphrase | -passphrase [passphrase]]
Convert a regular Journaled HFS+ or Case-sensitive Journaled HFS+ volume (must be on a partition and within a GPT par-
titioning scheme) into a CoreStorage logical volume.
If -passphrase is specified, the on-disk bytes will be encrypted. You will be prompted for a new passphrase interac-
tively, or you can specify the passphrase on the command line. Alternatively, if you specify -stdinpassphrase the stan-
dard input is read for the passphrase so that a program could execute diskutil and send the passphrase through a pipe
without having to expose it as a command-line parameter.
The volume is encrypted with an FDE "Disk" passphrase, which is distinct from the "User" ID and passphrase combination
which FileVault associates with a volume. Therefore, if you want to encrypt a macOS "OS-bearing" volume (with its user
accounts), you must use FileVault in Security Preferences or the contextual menu in the Finder.
The volume must be resizable (the above types are) and also mounted. Conversion is done live and in-place; targeting
the boot volume is supported; as much of the conversion as possible is done before an eject or reboot is necessary.
After slightly shrinking the source volume to make room for CoreStorage data structures at the end, its partition type
is changed to Apple_CoreStorage and it becomes a CoreStorage Physical Volume. A new CoreStorage Logical Volume Group
is then created with this Physical Volume as the backing store, followed by the creation of a Logical Volume Family and
Logical Volume pair.
At this point, the new CoreStorage PV/LVG/LVF/LV stack is ready for use, although the "old" mountpoint must first be
unmounted; yet it might not be unmountable. This will occur if the target (now the PV) is the current boot volume.
Just before exiting, diskutil coreStorage convert will try to unmount the target disk (which is now the "old" mount
point and the new PV). If successful (target is not the boot disk), the volume now becomes mounted from the LV. If
unsuccessful (target is the boot disk), a reboot is necessary.
At this point, if no encryption was specified, all is done. Otherwise, the bytes-on-disk will begin to be encrypted in-
place by CoreStorage automatically "in the background" while the PV/LVG/LVF/LV stack continues to be usable. Encryption
progress may be monitored with diskutil coreStorage list.
When encryption is finished, a Disk passphrase will be required the next time the LV is ejected and re-attached. If
the LV is hosting the boot volume, this passphrase requirement will thus occur at the next reboot.
Note that all on-disk data is not secured immediately; it is a deliberate process of encrypting all on-disk bytes while
the CoreStorage driver keeps publishing the (usable) LVG/LV.
Ownership of the affected disk is required.
revert device | lvUUID [-stdinpassphrase] | [-passphrase passphrase] | [-recoverykeychain file]
Convert a CoreStorage logical volume back to its native type. The volume must have been created by means of conver-
sion, e.g. with diskutil coreStorage convert.
If the volume was not created with a passphrase, then simple ownership of the affected disk is required; otherwise, a
Disk passphrase must be supplied, either interactively or via one of the parameters or a keychain file in the same man-
ner as diskutil coreStorage unlockVolume.
create | createLVG lvgName devices ...
Create a CoreStorage logical volume group. The disks specified will become the (initial) set of physical volumes; more
than one may be specified. You can specify partitions (which will be re-typed to be Apple_CoreStorage) or whole-disks
(which will be partitioned as GPT and will contain an Apple_CoreStorage partition). The resulting LVG UUID can then be
used with createVolume below. All existing data on the drive(s) will be lost. Ownership of the affected disk is
required.
delete | deleteLVG lvgUUID | lvgName
Delete a CoreStorage logical volume group. All logical volume families with their logical volumes are removed, the log-
ical volume group is destroyed, and the now-orphaned physical volumes are erased and partition-typed as Journaled HFS+.
rename | renameLVG lvgUUID | lvgName newName
Rename a CoreStorage logical volume group. Do not confuse this name with the LV name or the volume name of the file
system volume on the LV.
createVolume | createLV lvgUUID | lvgName type name size [-stdinpassphrase | -passphrase [passphrase]]
Export a new logical volume family, with a new logical volume under it, out of a CoreStorage logical volume group.
Type is the file system personality to initialize on the new logical volume. Valid types are Journaled HFS+ or Case-
sensitive Journaled HFS+ or their aliases. Size is the amount of space to allocate from the parent LVG. It is given in
the same manner as the triplet description for the partitionDisk verb, and you can also specify with % a percentage of
the currently remaining unallocated space in the LVG.
If -passphrase or -stdinpassphrase is specified, in the same manner as with diskutil coreStorage convert above, on-disk
data will be stored in an encrypted form as the Logical Volume is filled; otherwise, the data will remain plain.
deleteVolume | deleteLV lvUUID | device
Remove an exported logical volume (and its logical volume family as appropriate) from a CoreStorage logical volume
group. Any data on that logical volume will be lost. This operation will thus result in an increase in free space in
the logical volume group.
It is assumed that the logical volume is used as a backing store for a file system; therefore, an unmount attempt is
made which must succeed before the removal of the logical volume is done.
encryptVolume | encryptLV lvUUID | device [-stdinpassphrase] | [-passphrase passphrase]
Begin a live background process of encrypting the on-disk backing bytes of an existing plain CoreStorage logical volume
(LV).
That is, the on-disk bytes that are backing the user data are all visited, read, and re-written in an encrypted form;
this process can take a long time (minutes to hours). This process continues seamlessly across reboots. The logical
volume remains usable at all times. When this command returns, the operation will be ongoing; you can check progress
with diskutil coreStorage list.
The entire logical volume family (LVF) is affected since all LVs in an LVF share the same encryption settings.
Any new user data written while this background operation is in progress will be in encrypted form.
Specifying -passphrase or -stdinpassphrase or interactively entering a passphrase is mandatory; you do so in the same
manner as with diskutil coreStorage convert above.
The volume is encrypted with an FDE "Disk" passphrase, which is distinct from the "User" ID and passphrase combination
which FileVault associates with a volume. Therefore, if you want to encrypt a macOS "OS-bearing" volume (with its user
accounts), you must use FileVault in Security Preferences or the contextual menu in the Finder.
decryptVolume | decryptLV lvUUID | device [-stdinpassphrase] | [-passphrase passphrase]
Begin a live background process of decrypting the on-disk backing bytes of an existing encrypted CoreStorage logical
volume (LV). Bytes are read, decrypted, and written back to disk in plain form. The LV must be unlocked before begin-
ning this operation.
Like as in diskutil coreStorage encryptVolume above, all on-disk bytes are visited and converted, the process is seam-
less across reboots, the logical volume remains usable at all times, the entire logical volume family (LVF) is
affected, any new user data written will be in plain form, and the operation will be ongoing when this command returns.
Credentials must be supplied; you can use -passphrase or -stdinpassphrase to supply a Disk passphrase, or you can spec-
ify that a recovery keychain file be used, in the same manner as diskutil coreStorage unlockVolume.
unlockVolume | unlockLV [-nomount] lvUUID [-stdinpassphrase] | [-passphrase passphrase] | [-recoverykeychain file]
Unlock a logical volume and file system, causing it to be attached and mounted.
Data is then accessible in plain form to the file system and applications, while the on-physical-disk backing bytes
remain in encrypted form.
The locked state means that the CoreStorage driver has not been given authentication information (a passphrase) to
interpret the encrypted bytes on disk and thus export a dev node. This verb unlocks a logical volume family (LVF) and
its logical volumes (LVs) by providing that authentication; as the LVs thus appear as dev nodes, any file systems upon
them are automatically mounted unless the -nomount option is given.
To re-lock the volume, make it offline again by ejecting it, e.g. with diskutil eject.
Credentials must be supplied. You must either supply a Disk passphrase interactively, with one of the -passphrase or
-stdinpassphrase parameters in the same manner as with diskutil coreStorage convert above, or you must specify that a
recovery keychain file be used.
You can specify -recoverykeychain with a path to a keychain file. The keychain must be unlocked; see security(1) for
more information.
changeVolumePassphrase | passwd lvUUID [-recoverykeychain file] [-oldpassphrase oldpassphrase] [-newpassphrase newpassphrase]
[-stdinpassphrase]
Change the Disk passphrase of an existing encrypted volume. It need not be unlocked nor mounted. The parameters, while
variously optional, must be given in the above order.
You must authenticate either via the -oldpassphrase parameter, via the -stdinpassphrase parameter (with newline or eof-
terminated data given to stdin), or via an interactive prompt (if no parameters are given), in the same manner as
diskutil coreStorage convert above. Alternatively, you can authenticate by specifying -recoverykeychain with a path to
a keychain file.
A new passphrase must then be supplied, again via one of the three methods above (interactive, -newpassphrase, or
-stdinpassphrase).
If you are supplying both the old and new passphrases via stdin, they must be separated with a newline character.
Only the Disk passphrase is supported; you cannot change credentials for various users that were set up with FileVault
or the Finder; to edit credentials for such users, you should use fdesetup (8).
resizeVolume | resizeLV lvUUID | device size
Resize a logical volume (LV). If you shrink an LV, more space becomes available in its logical volume group (LVG); if
you grow an LV, less space becomes available. You can check the free space with diskutil coreStorage list. The file
system volume which resides inside the LV is grown or shrunk as needed.
You can specify a size of zero (0) to fill up all remaining space in the parent LVG with the given LV.
resizeDisk | resizePV pvUUID size [part1Format part1Name part1Size part2Format part2Name part2Size part3Format part3Name part3Size
...]
Resize a physical volume (PV). If you shrink a PV, less space becomes available in its logical volume group (LVG); if
you grow a PV, more space becomes available. The partition in which the PV resides is changed to accommodate, and the
associated booter partition, if present, is automatically moved.
Note that you cannot ordinarily grow a PV unless there is free space in the partition map beyond it; note also that you
cannot ordinarily shrink a PV unless the LVG has some free space in it (e.g. by shrinking an overlying LV first).
When decreasing the size (shrinking), new partitions may optionally be created to fill the newly-freed space. To do
this, specify the format, name, and size parameters in the same manner as the triplet description for the partitionDisk
verb.
You can specify a size of zero (0) to fill up all remaining space to the next partition or the end of the partition
map, if possible.
resizeStack lvUUID | device [pvUUID] size [part1Format part1Name part1Size part2Format part2Name part2Size part3Format part3Name
part3Size ...]
Combine the actions of diskutil coreStorage resizePV and diskutil coreStorage resizeLV in the correct sequence in order
to effect a shrink or a grow in an entire LVG setup.
This is done by making a change to the size of a logical volume (LV), after or before which (one of its) physical vol-
ume(s) (PV) also changes its size accordingly. The (HFS) file system "on top of" the LV and the disk partition "below"
the PV, as well as the location of the PV's associated booter partition, are automatically adjusted.
When decreasing the size (shrinking), new partitions may optionally be created to fill the newly-freed space. To do
this, specify the format, name, and size parameters in the same manner as the triplet description for the partitionDisk
verb.
Since an LVG might have one (e.g. Full Disk Encryption (FDE), aka FileVault), two (e.g. Fusion), or even three (certain
Boot Camp configurations) PVs, a specific PV must be chosen. You can have this command choose one for you, or you can
specify the PV UUID directly. If you do not specify a PV, the one which has previously been marked for this purpose is
used; if no mark, a policy algorithm is applied.
If your new LV size represents a grow of the existing LV size, then the PV size will take up more space on disk, thus
creating a larger LVG for the larger LV to live in. If your new LV size represents a shrink, then the PV size will
take up less space on disk, thus creating a smaller LVG, which is enough for the smaller LV to live in. The magnitude
of the size change you specify (which is for the LV) causes an exact size change in the PV if you conform to partition
rounding (alignment) restrictions; the corresponding LV change may be greater because it is under additional alignment
restrictions imposed by CoreStorage and HFS.
The "spilling over" of size change effects from one PV onto another is not supported; only and exactly one PV is
affected by this operation. Grows or shrinks whose effects don't "fit" the designated PV will result in an error mes-
sage and no effect. For example, you can't do a shrink on a multi-PV setup such that the designated PV should shrink
to zero size and so effectively should disappear. Nor can you do a grow which would necessitate the growth of some
other PV or the addition of new PVs.
As in diskutil coreStorage resizePV, note that you cannot grow unless there is free space in the partition map beyond
the designated PV, which is not normally the case because you usually don't leave gaps of free space on your disk.
You can specify a size of zero (0) to fill up all remaining space to the partition following the designated PV's booter
or to the end of the partition map, if possible.
DEVICES
A device parameter for any of the above commands (except where explicitly required otherwise) can usually be any of the following:
o The disk identifier (see below). Any entry of the form of disk*, e.g. disk1s9.
o The device node entry containing the disk identifier. Any entry of the form of /dev/[r]disk*, e.g. /dev/disk2.
o The volume mount point. Any entry of the form of /Volumes/*, e.g. /Volumes/Untitled. In most cases, a "custom" mount point e.g.
/your/custom/mountpoint/here is also accepted.
o The URL form of any of the volume mount point forms described above. E.g. file:///Volumes/Untitled or file:///.
o A UUID. Any entry of the form of e.g. 11111111-2222-3333-4444-555555555555. The UUID can be a "media" UUID which IOKit places in
an IOMedia node as derived from e.g. a GPT map's partition UUID, or it can be an AppleRAID (or CoreStorage) set (LV) or member (PV)
UUID.
o A volume name, e.g. Untitled. This match is only attempted if the given device is not of the form [/dev/][r]disk*, nor
[/Volumes/]*. The match attempt is against the intrinsic volume label, not against the terminal component, if mounted, of its
mount point.
DISK IDENTIFIER
The (BSD) disk identifier string variously identifies a physical or logical device unit, a session (if any) upon that device, a partition
(slice) upon that session (if any), or a virtual logical volume. It may take the form of diskU, diskUsS, diskUsQ, or diskCsV, where U, C, S,
Q and V are positive decimal integers (possibly multi-digit), and where:
o U is the device unit. It may refer to hardware (e.g. a hard drive, optical drive, or memory card) or a virtual "drive" constructed
by software (e.g. an AppleRAID Set, disk image, CoreStorage LV, etc).
o C is an APFS Container. This is a a virtual disk constructed by APFS to represent a collection of APFS Volumes.
o Q is the session and is only included for optical media; it refers to the number of times recording has taken place on the cur-
rently-inserted medium (disc).
o S is the "slice"; it refers to a partition. Upon this partition, the raw data that underlies a user-visible file system is usually
present, but it may also contain specialized data for certain 3rd-party database programs, or data required for the system software
(e.g. EFI partitions, booter partitions, APM partition map data, etc), or, notably, it might contain backing-store physical volumes
for AppleRAID, CoreStorage, APFS, or 3rd-party Storage Systems.
o V is an APFS Volume; it refers to a virtual logical volume that is shared out of an APFS Container. For example, if an Apple_APFS-
typed partition is on disk5s2, then disk5s2 is termed the APFS Physical Store which is imported into an APFS Container. The APFS
Container might then export e.g. disk8s1, which is termed an APFS Volume, which is mountable as a file system. Multiple APFS Vol-
umes can be exported from a single APFS Container.
Some units (e.g. floppy disks, RAID sets) contain file system data upon their "whole" device instead of containing a partitioning scheme with
partitions.
Note that some of the forms appear the same and must be distinguished by context. For example, diskUsQ, diskUsS, and diskCsV are all 2-part
forms that can mean different things: For non-optical media, it identifies a partition (on a partition map) upon which (file system) data is
stored; for optical media, it identifies a session upon which an entire partition map (with its partitions with file systems) is stored; for
an APFS setup, it identifies an APFS Volume. As another example, in "stacked" cases (CoreStorage on AppleRAID or APFS on AppleRAID), the
1-part diskU form becomes a CoreStorage PV or APFS PhysicalStore, in contrast with the more-common 2-part form.
It is important for software to avoid relying on numerical ordering of any of the parts. Activities including but not limited to partition
deletions and insertions, partition resizing, virtual volume deletions and additions, device ejects and attachments due to media insertion
cycles, plug cycles, authentication lock cycles or reboots, can all cause (temporary) gaps and non-increments in the numerical ordering of
any of the parts. You must rely on more persistent means of identification, such as the various UUIDs.
SIZES
Wherever a size is emitted as an output, it is presented as a base-ten approximation to the precision of one fractional decimal digit and a
base-ten SI multiplier, often accompanied by a precise count in bytes. Scripts should refrain from parsing this human-readable output and use
the -plist option instead.
Wherever a size is to be supplied by you as an input, you can provide values in a number of different ways, some absolute and some context-
sensitive. Values are interpreted as base ten and must be positive with no preceding "+". An integer without a suffix is taken to mean an
exact number of bytes (e.g. 5368709120). Multiplier suffixes are optional, must follow your value immediately without whitespace, and allow
your value to be a real number (e.g. 5.1234t). Some of the specifiers below should not have a preceding value at all (e.g. the letter R for
"remainder").
Power-of-ten suffixes:
o B is bytes (not blocks) where the multiplier is 1. This suffix may be omitted.
o K[B] is power of ten kilobytes where the multiplier is 1000 (1 x 10^3).
o M[B] is power of ten megabytes where the multiplier is 1000000 (1 x 10^6).
o G[B] is power of ten gigabytes where the multiplier is 1000000000 (1 x 10^9).
o T[B] is power of ten terabytes where the multiplier is 1000000000000 (1 x 10^12).
o P[B] is power of ten petabytes where the multiplier is 1000000000000000 (1 x 10^15).
o E[B] is power of ten exabytes where the multiplier is 1000000000000000000 (1 x 10^18).
Power-of-two suffixes:
o Ki[B] is power of two kibibytes where the multiplier is 1024 (1 x 2^10).
o Mi[B] is power of two mebibytes where the multiplier is 1048576 (1 x 2^20).
o Gi[B] is power of two gibibytes where the multiplier is 1073741824 (1 x 2^30).
o Ti[B] is power of two tebibytes where the multiplier is 1099511627776 (1 x 2^40).
o Pi[B] is power of two pebibytes where the multiplier is 1125899906842624 (1 x 2^50).
o Ei[B] is power of two exbibytes where the multiplier is 1152921504606846976 (1 x 2^60).
The following are useful when working with devices and partition maps:
o S | UAM ("sectors") is 512-byte units (device-independent) where the multiplier is always 512.
o DBS ("device block size") is the device-dependent native block size of the encompassing whole disk, if applicable, where the multi-
plier is often 512, but not always; indeed it might not be a power of two.
In certain contexts (e.g. when asking to "use all space available", or when building partition triplets) you can provide a relative value as
follows:
o 0 (the number zero) is a request to allocate "all possible". This may mean different things in different contexts. For partition
maps, this requests allocation until the start of the following partition or the end of the partition map's allocatable space.
o % (with a preceding number) is a percentage of the whole-disk size, the partition map size, or other allocatable size, as appropri-
ate by context. Use of % is not supported in all situations.
o R (with no preceding number) specifies the remainder of the whole-disk size or other allocatable size after all other triplets in
the group are taken into account. It need not be in the last triplet. It must only appear in at most one triplet among all
triplets. Use of R is not supported in all situations; in some such cases, a value of 0 is more appropriate.
You can provide an operating system-defined constant value as follows:
o %recovery% (with no preceding number) is the customary size of pre-macOS-13.0 Recovery Partitions.
Note again that B refers to bytes and S and UAM refer to a constant multiplier of 512; the latter are useful when working with tools such as
gpt (8) or df (1). Note also that this multiplier is not a "block" size as actually implemented by the underlying device driver and/or hard-
ware, nor is it an "allocation block", which is a file system's minimum unit of backing store usage, often formatting-option-dependent.
Examples: 10G (10 gigabytes), 4.23tb (4.23 terabytes), 5M (5 megabytes), 4GiB (exactly 2^32 bytes), 126000 (exactly 126000 bytes), 25.4%
(25.4 percent of whole disk size).
FORMAT
The format parameter for the erasing and partitioning verbs is the file system personality name. You can determine this name by looking in a
file system bundle's /System/Library/Filesystems/<fs>.fs/Contents/Info.plist and looking at the keys for the FSPersonalities dictionary, or
by using the listFilesystems verb, which also lists shortcut aliases for common personalities (these shortcuts are defined by diskutil for
use with it only).
Common examples include JHFS+, JHFSX, MS-DOS, etc, as nicknames for the canonical forms from the file system bundles such as "Case-sensitive
HFS+".
EXAMPLES
Erase a whole disk (device)
diskutil eraseDisk JHFS+ Untitled disk3
Erase a volume (or format a partition or virtual disk)
diskutil eraseVolume jhfs+ UntitledHFS /Volumes/SomeDisk
Erase and (re)-partition a disk (device) with three partitions
diskutil partitionDisk disk3 HFSX Foo1 10G JHFS+ Foo2 10G MS-DOS FOO3 0
Erase and format with a different volume file system
diskutil eraseVolume ExFAT FOO disk3s2
Remove a partition from a partition map altogether
diskutil eraseVolume free free disk3s2
Convert a HFS disk to APFS
diskutil apfs convert disk3s2
Create a new APFS Container with three new APFS Volumes
diskutil apfs createContainer disk0s2
diskutil apfs addVolume disk8 APFS MyVolume1
diskutil apfs addVolume disk8 APFS MyVolume2 -passprompt
diskutil apfs addVolume disk8 APFS MyVolume3 -quota 10g
diskutil apfs list
Encrypt an APFS Volume (enable FileVault)
diskutil apfs encryptVolume disk8s1 -user disk
Lock or unlock an APFS Volume
diskutil apfs list disk8
diskutil apfs lockVolume disk8s1
diskutil apfs unlockVolume disk8s1 (tries all users)
diskutil apfs unlockVolume disk8s2 -user USERUUID (tries specific user)
Decrypt an APFS Volume (disable FileVault)
diskutil apfs listUsers disk8s1
diskutil apfs decryptVolume disk8s1 -user USERUUID
Remove an APFS Volume from its APFS Container altogether
diskutil apfs deleteVolume disk8s3
Resize an HFS volume and create a volume after it
diskutil resizeVolume /Volumes/SomeDisk 50g MS-DOS DOS 0
Resize an HFS volume and leave all remaining space as unused
diskutil resizeVolume /Volumes/SomeDisk 12g
Shrink a Core Storage PV in order to make space for a Boot Camp volume
diskutil coreStorage list
diskutil coreStorage resizeStack LVUUID PVUUID 150g ms-dos BOOTCAMP 0
Revert a disk from Core Storage back to plain HFS, possibly decrypting
diskutil coreStorage revert disk5
Merge two partitions into a new partition
diskutil mergePartitions JHFS+ not disk1s3 disk1s5
Split a partition into three new ones
diskutil splitPartition /Volumes/SomeDisk JHFS+ vol1 12g MS-DOS VOL2 8g JHFS+ vol3 0
Create an AppleRAID
diskutil createRAID mirror MirroredVolume JHFS+ disk1 disk2
Destroy an AppleRAID
diskutil destroyRAID /Volumes/MirroredVolume
Repair a damaged AppleRAID
diskutil repairMirror /Volumes/MirroredVolume disk3
Convert volume into an AppleRAID volume
diskutil enableRAID mirror /Volumes/ExistingVolume
Erase a partition, shrink, associate a pre-macOS-13.0 Recovery Partition
diskutil splitPartition disk8s2 JHFS+ MacHD R %Apple_Boot% %noformat% %recovery%
Partition a disk with the MBR partitioning scheme (e.g. for a camera)
diskutil partitionDisk disk3 MBR MS-DOS CAM1 0
Partition a disk with the (deprecated) APM partitioning scheme
diskutil partitionDisk disk3 APM HFS+ vol1 15% Journaled HFS+ vol2 R Journaled HFS+ vol3 25% Free Space volX 10g
SEE ALSO
hdiutil(1), mount(8), umount(8), diskmanagementd(8), diskmanagementstartup(8), diskarbitrationd(8), corestoraged(8), fdesetup(8), ioreg(8),
newfs_hfs(8), fsck_hfs(8), authopen(1), hfs.util(8), msdos.util(8), ufs.util(8), drutil(1), vsdbutil(8)
ERRORS
diskutil will exit with status 0 if successful or 1 if it cannot complete the requested operation; this includes cases in which usage text is
printed. Before diskutil returns with status 1, it prints a message which might include an explanation local to diskutil, an error string
from the DiskManagement or MediaKit frameworks, an underlying POSIX error, or some combination.
HISTORY
The eraseDisk and partitionDisk verbs had an option to add Mac OS 9 drivers (in partitions designated for that purpose); there was also a
repairOS9Permissions verb. These have been removed.
Starting with Mac OS X 10.6, the input and output notation of disk and partition sizes use power-of-10 suffixes. In the past this had been
power-of-2, regardless of the suffix (e.g. G, Gi, GiB) used for display or accepted as input. Starting with Mac OS X 10.11, the B suffix is
optional even for "bare" numeric values.
Starting with Mac OS X 10.11, the verify- and repairPermissions verbs have been removed.
Starting with macOS 10.12, the plist output of partitions from diskutil list -plist is presented in on-disk (not BSD slice name, e.g.
disk0s2) order. This mimics the order of outputs from programs such as gpt (1). The human-readable output always has been, and remains, in
on-disk order.
Starting with macOS 10.13.2, APFS cryptographic user authentication is required even when disabling FileVault.
Starting with macOS 10.14, partitions on all media above 1GiB in size will default to 1MiB alignment, regardless of the partitioning scheme.
This is significant for MBR partition maps and their use in appliances such as cameras. Free-space requests will not be aligned.
macOS 14 June 2018 macOS