1zfs(1M) System Administration Commands zfs(1M)
2
6 zfs - configures ZFS file systems
7
9 zfs [-?]
10 zfs create [-p] [-o property=value] ... filesystem
13 zfs create [-ps] [-b blocksize] [-o property=value] ... -V size volume
16 zfs destroy [-rRf] filesystem|volume
19 zfs destroy [-rRd] snapshot
22 zfs snapshot [-r] [-o property=value]...
25 filesystem@snapname|volume@snapname
26 zfs rollback [-rRf] snapshot
29 zfs clone [-p] [-o property=value] ... snapshot filesystem|volume
32 zfs promote clone-filesystem
35 zfs rename filesystem|volume|snapshot
38 filesystem|volume|snapshot
39 zfs rename [-p] filesystem|volume filesystem|volume
42 zfs rename -r snapshot snapshot
45 zfs list [-r|-d depth][-H][-o property[,...]] [-t type[,...]]
48 [-s property] ... [-S property] ... [filesystem|volume|snapshot] ...
49 zfs set property=value filesystem|volume|snapshot ...
52 zfs get [-r|-d depth][-Hp][-o field[,...]] [-s source[,...]]
55 "all" | property[,...] filesystem|volume|snapshot ...
56 zfs inherit [-r] property filesystem|volume|snapshot ...
59 zfs upgrade [-v]
62 zfs upgrade [-r] [-V version] -a | filesystem
65 zfs userspace [-niHp] [-o field[,...]] [-sS field] ...
68 [-t type [,...]] filesystem|snapshot
69 zfs groupspace [-niHp] [-o field[,...]] [-sS field] ...
72 [-t type [,...]] filesystem|snapshot
73 zfs mount
76 zfs mount [-vO] [-o options] -a | filesystem
79 zfs unmount [-f] -a | filesystem|mountpoint
82 zfs share -a | filesystem
85 zfs unshare -a filesystem|mountpoint
88 zfs send [-vR] [-[iI] snapshot] snapshot
91 zfs receive [-vnFu] filesystem|volume|snapshot
94 zfs receive [-vnFu] -d filesystem
97 zfs allow filesystem|volume
100 zfs allow [-ldug] "everyone"|user|group[,...] perm|@setname[,...]
103 filesystem|volume
104 zfs allow [-ld] -e perm|@setname[,...] filesystem|volume
107 zfs allow -c perm|@setname[,...] filesystem|volume
110 zfs allow -s @setname perm|@setname[,...] filesystem|volume
113 zfs unallow [-rldug] "everyone"|user|group[,...] [perm|@setname[,... ]]
116 filesystem|volume
117 zfs unallow [-rld] -e [perm|@setname[,... ]] filesystem|volume
120 zfs unallow [-r] -c [perm|@setname[ ... ]] filesystem|volume
123 zfs unallow [-r] -s @setname [perm|@setname[,... ]] filesystem|volume
126 zfs hold [-r] tag snapshot...
129 zfs holds [-r] snapshot...
132 zfs release [-r] tag snapshot...
135
138 The zfs command configures ZFS datasets within a ZFS storage pool, as
139 described in zpool(1M). A dataset is identified by a unique path within
140 the ZFS namespace. For example:
141 pool/{filesystem,volume,snapshot}
143 where the maximum length of a dataset name is MAXNAMELEN (256 bytes).
148 A dataset can be one of the following:
151 file system
153 A ZFS dataset of type filesystem can be mounted within the standard
155 system namespace and behaves like other file systems. While ZFS
156 file systems are designed to be POSIX compliant, known issues exist
157 that prevent compliance in some cases. Applications that depend on
158 standards conformance might fail due to nonstandard behavior when
159 checking file system free space.
160 volume
163 A logical volume exported as a raw or block device. This type of
165 dataset should only be used under special circumstances. File sys‐
166 tems are typically used in most environments.
167 snapshot
170 A read-only version of a file system or volume at a given point in
172 time. It is specified as filesystem@name or volume@name.
173 ZFS File System Hierarchy
176 A ZFS storage pool is a logical collection of devices that provide
177 space for datasets. A storage pool is also the root of the ZFS file
178 system hierarchy.
179 The root of the pool can be accessed as a file system, such as mounting
182 and unmounting, taking snapshots, and setting properties. The physical
183 storage characteristics, however, are managed by the zpool(1M) command.
184 See zpool(1M) for more information on creating and administering pools.
187 Snapshots
189 A snapshot is a read-only copy of a file system or volume. Snapshots
190 can be created extremely quickly, and initially consume no additional
191 space within the pool. As data within the active dataset changes, the
192 snapshot consumes more data than would otherwise be shared with the
193 active dataset.
194 Snapshots can have arbitrary names. Snapshots of volumes can be cloned
197 or rolled back, but cannot be accessed independently.
198 File system snapshots can be accessed under the .zfs/snapshot directory
201 in the root of the file system. Snapshots are automatically mounted on
202 demand and may be unmounted at regular intervals. The visibility of the
203 .zfs directory can be controlled by the snapdir property.
204 Clones
206 A clone is a writable volume or file system whose initial contents are
207 the same as another dataset. As with snapshots, creating a clone is
208 nearly instantaneous, and initially consumes no additional space.
209 Clones can only be created from a snapshot. When a snapshot is cloned,
212 it creates an implicit dependency between the parent and child. Even
213 though the clone is created somewhere else in the dataset hierarchy,
214 the original snapshot cannot be destroyed as long as a clone exists.
215 The origin property exposes this dependency, and the destroy command
216 lists any such dependencies, if they exist.
217 The clone parent-child dependency relationship can be reversed by using
220 the promote subcommand. This causes the "origin" file system to become
221 a clone of the specified file system, which makes it possible to
222 destroy the file system that the clone was created from.
223 Mount Points
225 Creating a ZFS file system is a simple operation, so the number of file
226 systems per system is likely to be numerous. To cope with this, ZFS
227 automatically manages mounting and unmounting file systems without the
228 need to edit the /etc/vfstab file. All automatically managed file sys‐
229 tems are mounted by ZFS at boot time.
230 By default, file systems are mounted under /path, where path is the
233 name of the file system in the ZFS namespace. Directories are created
234 and destroyed as needed.
235 A file system can also have a mount point set in the mountpoint prop‐
238 erty. This directory is created as needed, and ZFS automatically mounts
239 the file system when the zfs mount -a command is invoked (without edit‐
240 ing /etc/vfstab). The mountpoint property can be inherited, so if
241 pool/home has a mount point of /export/stuff, then pool/home/user auto‐
242 matically inherits a mount point of /export/stuff/user.
243 A file system mountpoint property of none prevents the file system from
246 being mounted.
247 If needed, ZFS file systems can also be managed with traditional tools
250 (mount, umount, /etc/vfstab). If a file system's mount point is set to
251 legacy, ZFS makes no attempt to manage the file system, and the admin‐
252 istrator is responsible for mounting and unmounting the file system.
253 Zones
255 A ZFS file system can be added to a non-global zone by using the
256 zonecfg add fs subcommand. A ZFS file system that is added to a non-
257 global zone must have its mountpoint property set to legacy.
258 The physical properties of an added file system are controlled by the
261 global administrator. However, the zone administrator can create, mod‐
262 ify, or destroy files within the added file system, depending on how
263 the file system is mounted.
264 A dataset can also be delegated to a non-global zone by using the
267 zonecfg add dataset subcommand. You cannot delegate a dataset to one
268 zone and the children of the same dataset to another zone. The zone
269 administrator can change properties of the dataset or any of its chil‐
270 dren. However, the quota property is controlled by the global adminis‐
271 trator.
272 A ZFS volume can be added as a device to a non-global zone by using the
275 zonecfg add device subcommand. However, its physical properties can be
276 modified only by the global administrator.
277 For more information about zonecfg syntax, see zonecfg(1M).
280 After a dataset is delegated to a non-global zone, the zoned property
283 is automatically set. A zoned file system cannot be mounted in the
284 global zone, since the zone administrator might have to set the mount
285 point to an unacceptable value.
286 The global administrator can forcibly clear the zoned property, though
289 this should be done with extreme care. The global administrator should
290 verify that all the mount points are acceptable before clearing the
291 property.
292 Native Properties
294 Properties are divided into two types, native properties and user-
295 defined (or "user") properties. Native properties either export inter‐
296 nal statistics or control ZFS behavior. In addition, native properties
297 are either editable or read-only. User properties have no effect on ZFS
298 behavior, but you can use them to annotate datasets in a way that is
299 meaningful in your environment. For more information about user proper‐
300 ties, see the "User Properties" section, below.
301 Every dataset has a set of properties that export statistics about the
304 dataset as well as control various behaviors. Properties are inherited
305 from the parent unless overridden by the child. Some properties apply
306 only to certain types of datasets (file systems, volumes, or snap‐
307 shots).
308 The values of numeric properties can be specified using human-readable
311 suffixes (for example, k, KB, M, Gb, and so forth, up to Z for
312 zettabyte). The following are all valid (and equal) specifications:
313 1536M, 1.5g, 1.50GB
315 The values of non-numeric properties are case sensitive and must be
320 lowercase, except for mountpoint, sharenfs, and sharesmb.
321 The following native properties consist of read-only statistics about
324 the dataset. These properties can be neither set, nor inherited. Native
325 properties apply to all dataset types unless otherwise noted.
326 available
328 The amount of space available to the dataset and all its children,
330 assuming that there is no other activity in the pool. Because space
331 is shared within a pool, availability can be limited by any number
332 of factors, including physical pool size, quotas, reservations, or
333 other datasets within the pool.
334 This property can also be referred to by its shortened column name,
336 avail.
337 compressratio
340 The compression ratio achieved for this dataset, expressed as a
342 multiplier. Compression can be turned on by running: zfs set com‐
343 pression=on dataset. The default value is off.
344 creation
347 The time this dataset was created.
349 defer_destroy
352 This property is on if the snapshot has been marked for deferred
354 destroy by using the zfs destroy -d command. Otherwise, the prop‐
355 erty is off.
356 mounted
359 For file systems, indicates whether the file system is currently
361 mounted. This property can be either yes or no.
362 origin
365 For cloned file systems or volumes, the snapshot from which the
367 clone was created. The origin cannot be destroyed (even with the -r
368 or -f options) so long as a clone exists.
369 referenced
372 The amount of data that is accessible by this dataset, which may or
374 may not be shared with other datasets in the pool. When a snapshot
375 or clone is created, it initially references the same amount of
376 space as the file system or snapshot it was created from, since its
377 contents are identical.
378 This property can also be referred to by its shortened column name,
380 refer.
381 type
384 The type of dataset: filesystem, volume, or snapshot.
386 used
389 The amount of space consumed by this dataset and all its descen‐
391 dents. This is the value that is checked against this dataset's
392 quota and reservation. The space used does not include this
393 dataset's reservation, but does take into account the reservations
394 of any descendent datasets. The amount of space that a dataset con‐
395 sumes from its parent, as well as the amount of space that are
396 freed if this dataset is recursively destroyed, is the greater of
397 its space used and its reservation.
398 When snapshots (see the "Snapshots" section) are created, their
400 space is initially shared between the snapshot and the file system,
401 and possibly with previous snapshots. As the file system changes,
402 space that was previously shared becomes unique to the snapshot,
403 and counted in the snapshot's space used. Additionally, deleting
404 snapshots can increase the amount of space unique to (and used by)
405 other snapshots.
406 The amount of space used, available, or referenced does not take
408 into account pending changes. Pending changes are generally
409 accounted for within a few seconds. Committing a change to a disk
410 using fsync(3c) or O_SYNC does not necessarily guarantee that the
411 space usage information is updated immediately.
412 usedby*
415 The usedby* properties decompose the used properties into the vari‐
417 ous reasons that space is used. Specifically, used = usedbychildren
418 + usedbydataset + usedbyrefreservation +, usedbysnapshots. These
419 properties are only available for datasets created on zpool "ver‐
420 sion 13" pools.
421 usedbychildren
424 The amount of space used by children of this dataset, which would
426 be freed if all the dataset's children were destroyed.
427 usedbydataset
430 The amount of space used by this dataset itself, which would be
432 freed if the dataset were destroyed (after first removing any
433 refreservation and destroying any necessary snapshots or descen‐
434 dents).
435 usedbyrefreservation
438 The amount of space used by a refreservation set on this dataset,
440 which would be freed if the refreservation was removed.
441 usedbysnapshots
444 The amount of space consumed by snapshots of this dataset. In par‐
446 ticular, it is the amount of space that would be freed if all of
447 this dataset's snapshots were destroyed. Note that this is not sim‐
448 ply the sum of the snapshots' used properties because space can be
449 shared by multiple snapshots.
450 userused@user
453 The amount of space consumed by the specified user in this dataset.
455 Space is charged to the owner of each file, as displayed by ls -l.
456 The amount of space charged is displayed by du and ls -s. See the
457 zfs userspace subcommand for more information.
458 Unprivileged users can access only their own space usage. The root
460 user, or a user who has been granted the userused privilege with
461 zfs allow, can access everyone's usage.
462 The userused@... properties are not displayed by zfs get all. The
464 user's name must be appended after the @ symbol, using one of the
465 following forms:
466 o POSIX name (for example, joe)
468 o POSIX numeric ID (for example, 789)
470 o SID name (for example, joe.smith@mydomain)
472 o SID numeric ID (for example, S-1-123-456-789)
474 userrefs
477 This property is set to the number of user holds on this snapshot.
479 User holds are set by using the zfs hold command.
480 groupused@group
483 The amount of space consumed by the specified group in this
485 dataset. Space is charged to the group of each file, as displayed
486 by ls -l. See the userused@user property for more information.
487 Unprivileged users can only access their own groups' space usage.
489 The root user, or a user who has been granted the groupused privi‐
490 lege with zfs allow, can access all groups' usage.
491 volblocksize=blocksize
494 For volumes, specifies the block size of the volume. The blocksize
496 cannot be changed once the volume has been written, so it should be
497 set at volume creation time. The default blocksize for volumes is 8
498 Kbytes. Any power of 2 from 512 bytes to 128 Kbytes is valid.
499 This property can also be referred to by its shortened column name,
501 volblock.
502 The following native properties can be used to change the behavior of a
506 ZFS dataset.
507 aclinherit=discard | noallow | restricted | passthrough | passthrough-x
509 Controls how ACL entries are inherited when files and directories
511 are created. A file system with an aclinherit property of discard
512 does not inherit any ACL entries. A file system with an aclinherit
513 property value of noallow only inherits inheritable ACL entries
514 that specify "deny" permissions. The property value restricted (the
515 default) removes the write_acl and write_owner permissions when the
516 ACL entry is inherited. A file system with an aclinherit property
517 value of passthrough inherits all inheritable ACL entries without
518 any modifications made to the ACL entries when they are inherited.
519 A file system with an aclinherit property value of passthrough-x
520 has the same meaning as passthrough, except that the owner@,
521 group@, and everyone@ ACEs inherit the execute permission only if
522 the file creation mode also requests the execute bit.
523 When the property value is set to passthrough, files are created
525 with a mode determined by the inheritable ACEs. If no inheritable
526 ACEs exist that affect the mode, then the mode is set in accordance
527 to the requested mode from the application.
528 aclmode=discard | groupmask | passthrough
531 Controls how an ACL is modified during chmod(2). A file system with
533 an aclmode property of discard deletes all ACL entries that do not
534 represent the mode of the file. An aclmode property of groupmask
535 (the default) reduces user or group permissions. The permissions
536 are reduced, such that they are no greater than the group permis‐
537 sion bits, unless it is a user entry that has the same UID as the
538 owner of the file or directory. In this case, the ACL permissions
539 are reduced so that they are no greater than owner permission bits.
540 A file system with an aclmode property of passthrough indicates
541 that no changes are made to the ACL other than generating the nec‐
542 essary ACL entries to represent the new mode of the file or direc‐
543 tory.
544 atime=on | off
547 Controls whether the access time for files is updated when they are
549 read. Turning this property off avoids producing write traffic when
550 reading files and can result in significant performance gains,
551 though it might confuse mailers and other similar utilities. The
552 default value is on.
553 canmount=on | off | noauto
556 If this property is set to off, the file system cannot be mounted,
558 and is ignored by zfs mount -a. Setting this property to off is
559 similar to setting the mountpoint property to none, except that the
560 dataset still has a normal mountpoint property, which can be inher‐
561 ited. Setting this property to off allows datasets to be used
562 solely as a mechanism to inherit properties. One example of setting
563 canmount=off is to have two datasets with the same mountpoint, so
564 that the children of both datasets appear in the same directory,
565 but might have different inherited characteristics.
566 When the noauto option is set, a dataset can only be mounted and
568 unmounted explicitly. The dataset is not mounted automatically when
569 the dataset is created or imported, nor is it mounted by the zfs
570 mount -a command or unmounted by the zfs unmount -a command.
571 This property is not inherited.
573 checksum=on | off | fletcher2,| fletcher4 | sha256
576 Controls the checksum used to verify data integrity. The default
578 value is on, which automatically selects an appropriate algorithm
579 (currently, fletcher2, but this may change in future releases). The
580 value off disables integrity checking on user data. Disabling
581 checksums is NOT a recommended practice.
582 Changing this property affects only newly-written data.
584 compression=on | off | lzjb | gzip | gzip-N
587 Controls the compression algorithm used for this dataset. The lzjb
589 compression algorithm is optimized for performance while providing
590 decent data compression. Setting compression to on uses the lzjb
591 compression algorithm. The gzip compression algorithm uses the same
592 compression as the gzip(1) command. You can specify the gzip level
593 by using the value gzip-N where N is an integer from 1 (fastest) to
594 9 (best compression ratio). Currently, gzip is equivalent to gzip-6
595 (which is also the default for gzip(1)).
596 This property can also be referred to by its shortened column name
598 compress. Changing this property affects only newly-written data.
599 copies=1 | 2 | 3
602 Controls the number of copies of data stored for this dataset.
604 These copies are in addition to any redundancy provided by the
605 pool, for example, mirroring or RAID-Z. The copies are stored on
606 different disks, if possible. The space used by multiple copies is
607 charged to the associated file and dataset, changing the used prop‐
608 erty and counting against quotas and reservations.
609 Changing this property only affects newly-written data. Therefore,
611 set this property at file system creation time by using the -o
612 copies=N option.
613 devices=on | off
616 Controls whether device nodes can be opened on this file system.
618 The default value is on.
619 exec=on | off
622 Controls whether processes can be executed from within this file
624 system. The default value is on.
625 mountpoint=path | none | legacy
628 Controls the mount point used for this file system. See the "Mount
630 Points" section for more information on how this property is used.
631 When the mountpoint property is changed for a file system, the file
633 system and any children that inherit the mount point are unmounted.
634 If the new value is legacy, then they remain unmounted. Otherwise,
635 they are automatically remounted in the new location if the prop‐
636 erty was previously legacy or none, or if they were mounted before
637 the property was changed. In addition, any shared file systems are
638 unshared and shared in the new location.
639 nbmand=on | off
642 Controls whether the file system should be mounted with nbmand (Non
644 Blocking mandatory locks). This is used for CIFS clients. Changes
645 to this property only take effect when the file system is umounted
646 and remounted. See mount(1M) for more information on nbmand mounts.
647 primarycache=all | none | metadata
650 Controls what is cached in the primary cache (ARC). If this prop‐
652 erty is set to all, then both user data and metadata is cached. If
653 this property is set to none, then neither user data nor metadata
654 is cached. If this property is set to metadata, then only metadata
655 is cached. The default value is all.
656 quota=size | none
659 Limits the amount of space a dataset and its descendents can con‐
661 sume. This property enforces a hard limit on the amount of space
662 used. This includes all space consumed by descendents, including
663 file systems and snapshots. Setting a quota on a descendent of a
664 dataset that already has a quota does not override the ancestor's
665 quota, but rather imposes an additional limit.
666 Quotas cannot be set on volumes, as the volsize property acts as an
668 implicit quota.
669 userquota@user=size | none
672 Limits the amount of space consumed by the specified user. User
674 space consumption is identified by the userspace@user property.
675 Enforcement of user quotas may be delayed by several seconds. This
677 delay means that a user might exceed their quota before the system
678 notices that they are over quota and begins to refuse additional
679 writes with the EDQUOT error message . See the zfs userspace sub‐
680 command for more information.
681 Unprivileged users can only access their own groups' space usage.
683 The root user, or a user who has been granted the userquota privi‐
684 lege with zfs allow, can get and set everyone's quota.
685 This property is not available on volumes, on file systems before
687 version 4, or on pools before version 15. The userquota@... proper‐
688 ties are not displayed by zfs get all. The user's name must be
689 appended after the @ symbol, using one of the following forms:
690 o POSIX name (for example, joe)
692 o POSIX numeric ID (for example, 789)
694 o SID name (for example, joe.smith@mydomain)
696 o SID numeric ID (for example, S-1-123-456-789)
698 groupquota@group=size | none
701 Limits the amount of space consumed by the specified group. Group
703 space consumption is identified by the userquota@user property.
704 Unprivileged users can access only their own groups' space usage.
706 The root user, or a user who has been granted the groupquota privi‐
707 lege with zfs allow, can get and set all groups' quotas.
708 readonly=on | off
711 Controls whether this dataset can be modified. The default value is
713 off.
714 This property can also be referred to by its shortened column name,
716 rdonly.
717 recordsize=size
720 Specifies a suggested block size for files in the file system. This
722 property is designed solely for use with database workloads that
723 access files in fixed-size records. ZFS automatically tunes block
724 sizes according to internal algorithms optimized for typical access
725 patterns.
726 For databases that create very large files but access them in small
728 random chunks, these algorithms may be suboptimal. Specifying a
729 recordsize greater than or equal to the record size of the database
730 can result in significant performance gains. Use of this property
731 for general purpose file systems is strongly discouraged, and may
732 adversely affect performance.
733 The size specified must be a power of two greater than or equal to
735 512 and less than or equal to 128 Kbytes.
736 Changing the file system's recordsize affects only files created
738 afterward; existing files are unaffected.
739 This property can also be referred to by its shortened column name,
741 recsize.
742 refquota=size | none
745 Limits the amount of space a dataset can consume. This property
747 enforces a hard limit on the amount of space used. This hard limit
748 does not include space used by descendents, including file systems
749 and snapshots.
750 refreservation=size | none
753 The minimum amount of space guaranteed to a dataset, not including
755 its descendents. When the amount of space used is below this value,
756 the dataset is treated as if it were taking up the amount of space
757 specified by refreservation. The refreservation reservation is
758 accounted for in the parent datasets' space used, and counts
759 against the parent datasets' quotas and reservations.
760 If refreservation is set, a snapshot is only allowed if there is
762 enough free pool space outside of this reservation to accommodate
763 the current number of "referenced" bytes in the dataset.
764 This property can also be referred to by its shortened column name,
766 refreserv.
767 reservation=size | none
770 The minimum amount of space guaranteed to a dataset and its descen‐
772 dents. When the amount of space used is below this value, the
773 dataset is treated as if it were taking up the amount of space
774 specified by its reservation. Reservations are accounted for in the
775 parent datasets' space used, and count against the parent datasets'
776 quotas and reservations.
777 This property can also be referred to by its shortened column name,
779 reserv.
780 secondarycache=all | none | metadata
783 Controls what is cached in the secondary cache (L2ARC). If this
785 property is set to all, then both user data and metadata is cached.
786 If this property is set to none, then neither user data nor meta‐
787 data is cached. If this property is set to metadata, then only
788 metadata is cached. The default value is all.
789 setuid=on | off
792 Controls whether the set-UID bit is respected for the file system.
794 The default value is on.
795 shareiscsi=on | off
798 Like the sharenfs property, shareiscsi indicates whether a ZFS vol‐
800 ume is exported as an iSCSI target. The acceptable values for this
801 property are on, off, and type=disk. The default value is off. In
802 the future, other target types might be supported. For example,
803 tape.
804 You might want to set shareiscsi=on for a file system so that all
806 ZFS volumes within the file system are shared by default. However,
807 setting this property on a file system has no direct effect.
808 sharesmb=on | off | opts
811 Controls whether the file system is shared by using the Solaris
813 CIFS service, and what options are to be used. A file system with
814 the sharesmb property set to off is managed through traditional
815 tools such as sharemgr(1M). Otherwise, the file system is automati‐
816 cally shared and unshared with the zfs share and zfs unshare com‐
817 mands. If the property is set to on, the sharemgr(1M) command is
818 invoked with no options. Otherwise, the sharemgr(1M) command is
819 invoked with options equivalent to the contents of this property.
820 Because SMB shares requires a resource name, a unique resource name
822 is constructed from the dataset name. The constructed name is a
823 copy of the dataset name except that the characters in the dataset
824 name, which would be illegal in the resource name, are replaced
825 with underscore (_) characters. A pseudo property "name" is also
826 supported that allows you to replace the data set name with a spec‐
827 ified name. The specified name is then used to replace the prefix
828 dataset in the case of inheritance. For example, if the dataset
829 data/home/john is set to name=john, then data/home/john has a
830 resource name of john. If a child dataset of data/home/john/back‐
831 ups, it has a resource name of john_backups.
832 When SMB shares are created, the SMB share name appears as an entry
834 in the .zfs/shares directory. You can use the ls or chmod command
835 to display the share-level ACLs on the entries in this directory.
836 When the sharesmb property is changed for a dataset, the dataset
838 and any children inheriting the property are re-shared with the new
839 options, only if the property was previously set to off, or if they
840 were shared before the property was changed. If the new property is
841 set to off, the file systems are unshared.
842 sharenfs=on | off | opts
845 Controls whether the file system is shared via NFS, and what
847 options are used. A file system with a sharenfs property of off is
848 managed through traditional tools such as share(1M), unshare(1M),
849 and dfstab(4). Otherwise, the file system is automatically shared
850 and unshared with the zfs share and zfs unshare commands. If the
851 property is set to on, the share(1M) command is invoked with no
852 options. Otherwise, the share(1M) command is invoked with options
853 equivalent to the contents of this property.
854 When the sharenfs property is changed for a dataset, the dataset
856 and any children inheriting the property are re-shared with the new
857 options, only if the property was previously off, or if they were
858 shared before the property was changed. If the new property is off,
859 the file systems are unshared.
860 logbias = latency | throughput
863 Provide a hint to ZFS about handling of synchronous requests in
865 this dataset. If logbias is set to latency (the default), ZFS will
866 use pool log devices (if configured) to handle the requests at low
867 latency. If logbias is set to throughput, ZFS will not use config‐
868 ured pool log devices. ZFS will instead optimize synchronous opera‐
869 tions for global pool throughput and efficient use of resources.
870 snapdir=hidden | visible
873 Controls whether the .zfs directory is hidden or visible in the
875 root of the file system as discussed in the "Snapshots" section.
876 The default value is hidden.
877 version=1 | 2 | current
880 The on-disk version of this file system, which is independent of
882 the pool version. This property can only be set to later supported
883 versions. See the zfs upgrade command.
884 volsize=size
887 For volumes, specifies the logical size of the volume. By default,
889 creating a volume establishes a reservation of equal size. For
890 storage pools with a version number of 9 or higher, a refreserva‐
891 tion is set instead. Any changes to volsize are reflected in an
892 equivalent change to the reservation (or refreservation). The vol‐
893 size can only be set to a multiple of volblocksize, and cannot be
894 zero.
895 The reservation is kept equal to the volume's logical size to pre‐
897 vent unexpected behavior for consumers. Without the reservation,
898 the volume could run out of space, resulting in undefined behavior
899 or data corruption, depending on how the volume is used. These
900 effects can also occur when the volume size is changed while it is
901 in use (particularly when shrinking the size). Extreme care should
902 be used when adjusting the volume size.
903 Though not recommended, a "sparse volume" (also known as "thin pro‐
905 visioning") can be created by specifying the -s option to the zfs
906 create -V command, or by changing the reservation after the volume
907 has been created. A "sparse volume" is a volume where the reserva‐
908 tion is less then the volume size. Consequently, writes to a sparse
909 volume can fail with ENOSPC when the pool is low on space. For a
910 sparse volume, changes to volsize are not reflected in the reserva‐
911 tion.
912 vscan=on | off
915 Controls whether regular files should be scanned for viruses when a
917 file is opened and closed. In addition to enabling this property,
918 the virus scan service must also be enabled for virus scanning to
919 occur. The default value is off.
920 xattr=on | off
923 Controls whether extended attributes are enabled for this file sys‐
925 tem. The default value is on.
926 zoned=on | off
929 Controls whether the dataset is managed from a non-global zone. See
931 the "Zones" section for more information. The default value is off.
932 The following three properties cannot be changed after the file system
936 is created, and therefore, should be set when the file system is cre‐
937 ated. If the properties are not set with the zfs create or zpool create
938 commands, these properties are inherited from the parent dataset. If
939 the parent dataset lacks these properties due to having been created
940 prior to these features being supported, the new file system will have
941 the default values for these properties.
942 casesensitivity=sensitive | insensitive | mixed
944 Indicates whether the file name matching algorithm used by the file
946 system should be case-sensitive, case-insensitive, or allow a com‐
947 bination of both styles of matching. The default value for the cas‐
948 esensitivity property is sensitive. Traditionally, UNIX and POSIX
949 file systems have case-sensitive file names.
950 The mixed value for the casesensitivity property indicates that the
952 file system can support requests for both case-sensitive and case-
953 insensitive matching behavior. Currently, case-insensitive matching
954 behavior on a file system that supports mixed behavior is limited
955 to the Solaris CIFS server product. For more information about the
956 mixed value behavior, see the Solaris ZFS Administration Guide.
957 normalization = none | formC | formD | formKC | formKD
960 Indicates whether the file system should perform a unicode normal‐
962 ization of file names whenever two file names are compared, and
963 which normalization algorithm should be used. File names are always
964 stored unmodified, names are normalized as part of any comparison
965 process. If this property is set to a legal value other than none,
966 and the utf8only property was left unspecified, the utf8only prop‐
967 erty is automatically set to on. The default value of the normal‐
968 ization property is none. This property cannot be changed after the
969 file system is created.
970 utf8only=on | off
973 Indicates whether the file system should reject file names that
975 include characters that are not present in the UTF-8 character code
976 set. If this property is explicitly set to off, the normalization
977 property must either not be explicitly set or be set to none. The
978 default value for the utf8only property is off. This property can‐
979 not be changed after the file system is created.
980 The casesensitivity, normalization, and utf8only properties are also
984 new permissions that can be assigned to non-privileged users by using
985 the ZFS delegated administration feature.
986 Temporary Mount Point Properties
988 When a file system is mounted, either through mount(1M) for legacy
989 mounts or the zfs mount command for normal file systems, its mount
990 options are set according to its properties. The correlation between
991 properties and mount options is as follows:
992 PROPERTY MOUNT OPTION
994 devices devices/nodevices
995 exec exec/noexec
996 readonly ro/rw
997 setuid setuid/nosetuid
998 xattr xattr/noxattr
999 In addition, these options can be set on a per-mount basis using the -o
1004 option, without affecting the property that is stored on disk. The val‐
1005 ues specified on the command line override the values stored in the
1006 dataset. The -nosuid option is an alias for nodevices,nosetuid. These
1007 properties are reported as "temporary" by the zfs get command. If the
1008 properties are changed while the dataset is mounted, the new setting
1009 overrides any temporary settings.
1010 User Properties
1012 In addition to the standard native properties, ZFS supports arbitrary
1013 user properties. User properties have no effect on ZFS behavior, but
1014 applications or administrators can use them to annotate datasets (file
1015 systems, volumes, and snapshots).
1016 User property names must contain a colon (:) character to distinguish
1019 them from native properties. They may contain lowercase letters, num‐
1020 bers, and the following punctuation characters: colon (:), dash (-),
1021 period (.), and underscore (_). The expected convention is that the
1022 property name is divided into two portions such as module:property, but
1023 this namespace is not enforced by ZFS. User property names can be at
1024 most 256 characters, and cannot begin with a dash (-).
1025 When making programmatic use of user properties, it is strongly sug‐
1028 gested to use a reversed DNS domain name for the module component of
1029 property names to reduce the chance that two independently-developed
1030 packages use the same property name for different purposes. Property
1031 names beginning with com.sun. are reserved for use by Sun Microsystems.
1032 The values of user properties are arbitrary strings, are always inher‐
1035 ited, and are never validated. All of the commands that operate on
1036 properties (zfs list, zfs get, zfs set, and so forth) can be used to
1037 manipulate both native properties and user properties. Use the zfs
1038 inherit command to clear a user property . If the property is not
1039 defined in any parent dataset, it is removed entirely. Property values
1040 are limited to 1024 characters.
1041 ZFS Volumes as Swap or Dump Devices
1043 During an initial installation or a live upgrade from a UFS file sys‐
1044 tem, a swap device and dump device are created on ZFS volumes in the
1045 ZFS root pool. By default, the swap area size is based on 1/2 the size
1046 of physical memory up to 2 Gbytes. The size of the dump device depends
1047 on the kernel's requirements at installation time. Separate ZFS volumes
1048 must be used for the swap area and dump devices. Do not swap to a file
1049 on a ZFS file system. A ZFS swap file configuration is not supported.
1050 If you need to change your swap area or dump device after the system is
1053 installed or upgraded, use the swap(1M) and dumpadm(1M) commands. If
1054 you need to change the size of your swap area or dump device, see the
1055 Solaris ZFS Administration Guide.
1056
1058 All subcommands that modify state are logged persistently to the pool
1059 in their original form.
1060 zfs ?
1062 Displays a help message.
1064 zfs create [-p] [-o property=value] ... filesystem
1067 Creates a new ZFS file system. The file system is automatically
1069 mounted according to the mountpoint property inherited from the
1070 parent.
1071 -p
1073 Creates all the non-existing parent datasets. Datasets created
1075 in this manner are automatically mounted according to the
1076 mountpoint property inherited from their parent. Any property
1077 specified on the command line using the -o option is ignored.
1078 If the target filesystem already exists, the operation com‐
1079 pletes successfully.
1080 -o property=value
1083 Sets the specified property as if the command zfs set prop‐
1085 erty=value was invoked at the same time the dataset was cre‐
1086 ated. Any editable ZFS property can also be set at creation
1087 time. Multiple -o options can be specified. An error results if
1088 the same property is specified in multiple -o options.
1089 zfs create [-ps] [-b blocksize] [-o property=value] ... -V size volume
1093 Creates a volume of the given size. The volume is exported as a
1095 block device in /dev/zvol/{dsk,rdsk}/path, where path is the name
1096 of the volume in the ZFS namespace. The size represents the logical
1097 size as exported by the device. By default, a reservation of equal
1098 size is created.
1099 size is automatically rounded up to the nearest 128 Kbytes to
1101 ensure that the volume has an integral number of blocks regardless
1102 of blocksize.
1103 -p
1105 Creates all the non-existing parent datasets. Datasets created
1107 in this manner are automatically mounted according to the
1108 mountpoint property inherited from their parent. Any property
1109 specified on the command line using the -o option is ignored.
1110 If the target filesystem already exists, the operation com‐
1111 pletes successfully.
1112 -s
1115 Creates a sparse volume with no reservation. See volsize in the
1117 Native Properties section for more information about sparse
1118 volumes.
1119 -o property=value
1122 Sets the specified property as if the zfs set property=value
1124 command was invoked at the same time the dataset was created.
1125 Any editable ZFS property can also be set at creation time.
1126 Multiple -o options can be specified. An error results if the
1127 same property is specified in multiple -o options.
1128 -b blocksize
1131 Equivalent to -o volblocksize=blocksize. If this option is
1133 specified in conjunction with -o volblocksize, the resulting
1134 behavior is undefined.
1135 zfs destroy [-rRf] filesystem|volume
1139 Destroys the given dataset. By default, the command unshares any
1141 file systems that are currently shared, unmounts any file systems
1142 that are currently mounted, and refuses to destroy a dataset that
1143 has active dependents (children or clones).
1144 -r
1146 Recursively destroy all children.
1148 -R
1151 Recursively destroy all dependents, including cloned file sys‐
1153 tems outside the target hierarchy.
1154 -f
1157 Force an unmount of any file systems using the unmount -f com‐
1159 mand. This option has no effect on non-file systems or
1160 unmounted file systems.
1161 Extreme care should be taken when applying either the -r or the -f
1163 options, as they can destroy large portions of a pool and cause
1164 unexpected behavior for mounted file systems in use.
1165 zfs destroy [-rRd] snapshot
1168 The given snapshot is destroyed immediately if and only if the zfs
1170 destroy command without the -d option would have destroyed it. Such
1171 immediate destruction would occur, for example, if the snapshot had
1172 no clones and the user-initiated reference count were zero.
1173 If the snapshot does not qualify for immediate destruction, it is
1175 marked for deferred deletion. In this state, it exists as a usable,
1176 visible snapshot until both of the preconditions listed above are
1177 met, at which point it is destroyed.
1178 -d
1180 Defer snapshot deletion.
1182 -r
1185 Destroy (or mark for deferred deletion) all snapshots with this
1187 name in descendent file systems.
1188 -R
1191 Recursively destroy all dependents.
1193 zfs snapshot [-r] [-o property=value] ... filesystem@snapname|vol‐
1197 ume@snapname
1198 Creates a snapshot with the given name. All previous modifications
1200 by successful system calls to the file system are part of the snap‐
1201 shot. See the "Snapshots" section for details.
1202 -r
1204 Recursively create snapshots of all descendent datasets. Snap‐
1206 shots are taken atomically, so that all recursive snapshots
1207 correspond to the same moment in time.
1208 -o property=value
1211 Sets the specified property; see zfs create for details.
1213 zfs rollback [-rRf] snapshot
1217 Roll back the given dataset to a previous snapshot. When a dataset
1219 is rolled back, all data that has changed since the snapshot is
1220 discarded, and the dataset reverts to the state at the time of the
1221 snapshot. By default, the command refuses to roll back to a snap‐
1222 shot other than the most recent one. In order to do so, all inter‐
1223 mediate snapshots must be destroyed by specifying the -r option.
1224 The -rR options do not recursively destroy the child snapshots of a
1226 recursive snapshot. Only the top-level recursive snapshot is
1227 destroyed by either of these options. To completely roll back a
1228 recursive snapshot, you must rollback the individual child snap‐
1229 shots.
1230 -r
1232 Recursively destroy any snapshots more recent than the one
1234 specified.
1235 -R
1238 Recursively destroy any more recent snapshots, as well as any
1240 clones of those snapshots.
1241 -f
1244 Used with the -R option to force an unmount of any clone file
1246 systems that are to be destroyed.
1247 zfs clone [-p] [-o property=value] ... snapshot filesystem|volume
1251 Creates a clone of the given snapshot. See the "Clones" section for
1253 details. The target dataset can be located anywhere in the ZFS
1254 hierarchy, and is created as the same type as the original.
1255 -p
1257 Creates all the non-existing parent datasets. Datasets created
1259 in this manner are automatically mounted according to the
1260 mountpoint property inherited from their parent. If the target
1261 filesystem or volume already exists, the operation completes
1262 successfully.
1263 -o property=value
1266 Sets the specified property; see zfs create for details.
1268 zfs promote clone-filesystem
1272 Promotes a clone file system to no longer be dependent on its "ori‐
1274 gin" snapshot. This makes it possible to destroy the file system
1275 that the clone was created from. The clone parent-child dependency
1276 relationship is reversed, so that the origin file system becomes a
1277 clone of the specified file system.
1278 The snapshot that was cloned, and any snapshots previous to this
1280 snapshot, are now owned by the promoted clone. The space they use
1281 moves from the origin file system to the promoted clone, so enough
1282 space must be available to accommodate these snapshots. No new
1283 space is consumed by this operation, but the space accounting is
1284 adjusted. The promoted clone must not have any conflicting snapshot
1285 names of its own. The rename subcommand can be used to rename any
1286 conflicting snapshots.
1287 zfs rename filesystem|volume|snapshot
1290 filesystem|volume|snapshot
1291 zfs rename [-p] filesystem|volume filesystem|volume
1292 Renames the given dataset. The new target can be located anywhere
1294 in the ZFS hierarchy, with the exception of snapshots. Snapshots
1295 can only be renamed within the parent file system or volume. When
1296 renaming a snapshot, the parent file system of the snapshot does
1297 not need to be specified as part of the second argument. Renamed
1298 file systems can inherit new mount points, in which case they are
1299 unmounted and remounted at the new mount point.
1300 -p
1302 Creates all the nonexistent parent datasets. Datasets created
1304 in this manner are automatically mounted according to the
1305 mountpoint property inherited from their parent.
1306 zfs rename -r snapshot snapshot
1310 Recursively rename the snapshots of all descendent datasets. Snap‐
1312 shots are the only dataset that can be renamed recursively.
1313 zfs list [-r|-d depth] [-H] [-o property[,...]] [ -t type[,...]] [ -s
1316 property ] ... [ -S property ] ... [filesystem|volume|snapshot] ...
1317 Lists the property information for the given datasets in tabular
1319 form. If specified, you can list property information by the abso‐
1320 lute pathname or the relative pathname. By default, all file sys‐
1321 tems and volumes are displayed. Snapshots are displayed if the
1322 listsnaps property is on (the default is off) . The following
1323 fields are displayed, name,used,available,referenced,mountpoint.
1324 -H
1326 Used for scripting mode. Do not print headers and separate
1328 fields by a single tab instead of arbitrary white space.
1329 -r
1332 Recursively display any children of the dataset on the command
1334 line.
1335 -d depth
1338 Recursively display any children of the dataset, limiting the
1340 recursion to depth. A depth of 1 will display only the dataset
1341 and its direct children.
1342 -o property
1345 A comma-separated list of properties to display. The property
1347 must be:
1348 o One of the properties described in the "Native Prop‐
1350 erties" section
1351 o A user property
1353 o The value name to display the dataset name
1355 o The value space to display space usage properties on
1357 file systems and volumes. This is a shortcut for
1358 specifying -o name,avail,used,usedsnap,usedds,use‐
1359 drefreserv,usedchild -t filesystem,volume syntax.
1360 -s property
1363 A property for sorting the output by column in ascending order
1365 based on the value of the property. The property must be one of
1366 the properties described in the "Properties" section, or the
1367 special value name to sort by the dataset name. Multiple prop‐
1368 erties can be specified at one time using multiple -s property
1369 options. Multiple -s options are evaluated from left to right
1370 in decreasing order of importance.
1371 The following is a list of sorting criteria:
1373 o Numeric types sort in numeric order.
1375 o String types sort in alphabetical order.
1377 o Types inappropriate for a row sort that row to the
1379 literal bottom, regardless of the specified order‐
1380 ing.
1381 o If no sorting options are specified the existing
1383 behavior of zfs list is preserved.
1384 -S property
1387 Same as the -s option, but sorts by property in descending
1389 order.
1390 -t type
1393 A comma-separated list of types to display, where type is one
1395 of filesystem, snapshot , volume, or all. For example, specify‐
1396 ing -t snapshot displays only snapshots.
1397 zfs set property=value filesystem|volume|snapshot ...
1401 Sets the property to the given value for each dataset. Only some
1403 properties can be edited. See the "Properties" section for more
1404 information on what properties can be set and acceptable values.
1405 Numeric values can be specified as exact values, or in a human-
1406 readable form with a suffix of B, K, M, G, T, P, E, Z (for bytes,
1407 kilobytes, megabytes, gigabytes, terabytes, petabytes, exabytes, or
1408 zettabytes, respectively). User properties can be set on snapshots.
1409 For more information, see the "User Properties" section.
1410 zfs get [-r|-d depth] [-Hp] [-o field[,...] [-s source[,...] "all" |
1413 property[,...] filesystem|volume|snapshot ...
1414 Displays properties for the given datasets. If no datasets are
1416 specified, then the command displays properties for all datasets on
1417 the system. For each property, the following columns are displayed:
1418 name Dataset name
1420 property Property name
1421 value Property value
1422 source Property source. Can either be local, default,
1423 temporary, inherited, or none (-).
1424 All columns are displayed by default, though this can be controlled
1427 by using the -o option. This command takes a comma-separated list
1428 of properties as described in the "Native Properties" and "User
1429 Properties" sections.
1430 The special value all can be used to display all properties that
1432 apply to the given dataset's type (filesystem, volume, or snap‐
1433 shot).
1434 -r
1436 Recursively display properties for any children.
1438 -d depth
1441 Recursively display any children of the dataset, limiting the
1443 recursion to depth. A depth of 1 will display only the dataset
1444 and its direct children.
1445 -H
1448 Display output in a form more easily parsed by scripts. Any
1450 headers are omitted, and fields are explicitly separated by a
1451 single tab instead of an arbitrary amount of space.
1452 -o field
1455 A comma-separated list of columns to display. name,prop‐
1457 erty,value,source is the default value.
1458 -s source
1461 A comma-separated list of sources to display. Those properties
1463 coming from a source other than those in this list are ignored.
1464 Each source must be one of the following: local,default,inher‐
1465 ited,temporary,none. The default value is all sources.
1466 -p
1469 Display numbers in parseable (exact) values.
1471 zfs inherit [-r] property filesystem|volume|snapshot ...
1475 Clears the specified property, causing it to be inherited from an
1477 ancestor. If no ancestor has the property set, then the default
1478 value is used. See the "Properties" section for a listing of
1479 default values, and details on which properties can be inherited.
1480 -r
1482 Recursively inherit the given property for all children.
1484 zfs upgrade [-v]
1488 Displays a list of file systems that are not the most recent ver‐
1490 sion.
1491 zfs upgrade [-r] [-V version] [-a | filesystem]
1494 Upgrades file systems to a new on-disk version. Once this is done,
1496 the file systems will no longer be accessible on systems running
1497 older versions of the software. zfs send streams generated from new
1498 snapshots of these file systems cannot be accessed on systems run‐
1499 ning older versions of the software.
1500 In general, the file system version is independent of the pool ver‐
1502 sion. See zpool(1M) for information on the zpool upgrade command.
1503 In some cases, the file system version and the pool version are
1505 interrelated and the pool version must be upgraded before the file
1506 system version can be upgraded.
1507 -a
1509 Upgrade all file systems on all imported pools.
1511 filesystem
1514 Upgrade the specified file system.
1516 -r
1519 Upgrade the specified file system and all descendent file sys‐
1521 tems
1522 -V version
1525 Upgrade to the specified version. If the -V flag is not speci‐
1527 fied, this command upgrades to the most recent version. This
1528 option can only be used to increase the version number, and
1529 only up to the most recent version supported by this software.
1530 zfs userspace [-niHp] [-o field[,...]] [-sS field]... [-t type [,...]]
1534 filesystem | snapshot
1535 Displays space consumed by, and quotas on, each user in the speci‐
1537 fied filesystem or snapshot. This corresponds to the userused@user
1538 and userquota@user properties.
1539 -n
1541 Print numeric ID instead of user/group name.
1543 -H
1546 Do not print headers, use tab-delimited output.
1548 -p
1551 Use exact (parseable) numeric output.
1553 -o field[,...]
1556 Display only the specified fields from the following set,
1558 type,name,used,quota.The default is to display all fields.
1559 -s field
1562 Sort output by this field. The s and S flags may be specified
1564 multiple times to sort first by one field, then by another. The
1565 default is -s type -s name.
1566 -S field
1569 Sort by this field in reverse order. See -s.
1571 -t type[,...]
1574 Print only the specified types from the following set,
1576 all,posixuser,smbuser,posixgroup,smbgroup.
1577 The default is -t posixuser,smbuser
1579 The default can be changed to include group types.
1581 -i
1584 Translate SID to POSIX ID. The POSIX ID may be ephemeral if no
1586 mapping exists. Normal POSIX interfaces (for example, stat(2),
1587 ls -l) perform this translation, so the -i option allows the
1588 output from zfs userspace to be compared directly with those
1589 utilities. However, -i may lead to confusion if some files were
1590 created by an SMB user before a SMB-to-POSIX name mapping was
1591 established. In such a case, some files are owned by the SMB
1592 entity and some by the POSIX entity. However, the -i option
1593 will report that the POSIX entity has the total usage and quota
1594 for both.
1595 zfs groupspace [-niHp] [-o field[,...]] [-sS field]... [-t type [,...]]
1599 filesystem | snapshot
1600 Displays space consumed by, and quotas on, each group in the speci‐
1602 fied filesystem or snapshot. This subcommand is identical to zfs
1603 userspace, except that the default types to display are -t posix‐
1604 group,smbgroup.
1605 -
1607 zfs mount
1612 Displays all ZFS file systems currently mounted.
1614 zfs mount [-vO] [-o options] -a | filesystem
1617 Mounts ZFS file systems. Invoked automatically as part of the boot
1619 process.
1620 -o options
1622 An optional, comma-separated list of mount options to use tem‐
1624 porarily for the duration of the mount. See the "Temporary
1625 Mount Point Properties" section for details.
1626 -O
1629 Perform an overlay mount. See mount(1M) for more information.
1631 -v
1634 Report mount progress.
1636 -a
1639 Mount all available ZFS file systems. Invoked automatically as
1641 part of the boot process.
1642 filesystem
1645 Mount the specified filesystem.
1647 zfs unmount [-f] -a | filesystem|mountpoint
1651 Unmounts currently mounted ZFS file systems. Invoked automatically
1653 as part of the shutdown process.
1654 -f
1656 Forcefully unmount the file system, even if it is currently in
1658 use.
1659 -a
1662 Unmount all available ZFS file systems. Invoked automatically
1664 as part of the boot process.
1665 filesystem|mountpoint
1668 Unmount the specified filesystem. The command can also be given
1670 a path to a ZFS file system mount point on the system.
1671 zfs share -a | filesystem
1675 Shares available ZFS file systems.
1677 -a
1679 Share all available ZFS file systems. Invoked automatically as
1681 part of the boot process.
1682 filesystem
1685 Share the specified filesystem according to the sharenfs and
1687 sharesmb properties. File systems are shared when the sharenfs
1688 or sharesmb property is set.
1689 zfs unshare -a | filesystem|mountpoint
1693 Unshares currently shared ZFS file systems. This is invoked auto‐
1695 matically as part of the shutdown process.
1696 -a
1698 Unshare all available ZFS file systems. Invoked automatically
1700 as part of the boot process.
1701 filesystem|mountpoint
1704 Unshare the specified filesystem. The command can also be given
1706 a path to a ZFS file system shared on the system.
1707 zfs send [-vR] [-[iI] snapshot] snapshot
1711 Creates a stream representation of the second snapshot, which is
1713 written to standard output. The output can be redirected to a file
1714 or to a different system (for example, using ssh(1). By default, a
1715 full stream is generated.
1716 -i snapshot
1718 Generate an incremental stream from the first snapshot to the
1720 second snapshot. The incremental source (the first snapshot)
1721 can be specified as the last component of the snapshot name
1722 (for example, the part after the @), and it is assumed to be
1723 from the same file system as the second snapshot.
1724 If the destination is a clone, the source may be the origin
1726 snapshot, which must be fully specified (for example,
1727 pool/fs@origin, not just @origin).
1728 -I snapshot
1731 Generate a stream package that sends all intermediary snapshots
1733 from the first snapshot to the second snapshot. For example, -I
1734 @a fs@d is similar to -i @a fs@b; -i @b fs@c; -i @c fs@d. The
1735 incremental source snapshot may be specified as with the -i
1736 option.
1737 -R
1740 Generate a replication stream package, which will replicate the
1742 specified filesystem, and all descendent file systems, up to
1743 the named snapshot. When received, all properties, snapshots,
1744 descendent file systems, and clones are preserved.
1745 If the -i or -I flags are used in conjunction with the -R flag,
1747 an incremental replication stream is generated. The current
1748 values of properties, and current snapshot and file system
1749 names are set when the stream is received. If the -F flag is
1750 specified when this stream is received, snapshots and file sys‐
1751 tems that do not exist on the sending side are destroyed.
1752 -v
1755 Print verbose information about the stream package generated.
1757 The format of the stream is committed. You will be able to receive
1759 your streams on future versions of ZFS.
1760 zfs receive [-vnFu] filesystem|volume|snapshot
1763 zfs receive [-vnFu] -d filesystem
1764 Creates a snapshot whose contents are as specified in the stream
1766 provided on standard input. If a full stream is received, then a
1767 new file system is created as well. Streams are created using the
1768 zfs send subcommand, which by default creates a full stream. zfs
1769 recv can be used as an alias for zfs receive.
1770 If an incremental stream is received, then the destination file
1772 system must already exist, and its most recent snapshot must match
1773 the incremental stream's source. For zvols, the destination device
1774 link is destroyed and recreated, which means the zvol cannot be
1775 accessed during the receive operation.
1776 When a snapshot replication package stream that is generated by
1778 using the zfs send -R command is received, any snapshots that do
1779 not exist on the sending location are destroyed by using the zfs
1780 destroy -d command.
1781 The name of the snapshot (and file system, if a full stream is
1783 received) that this subcommand creates depends on the argument type
1784 and the -d option.
1785 If the argument is a snapshot name, the specified snapshot is cre‐
1787 ated. If the argument is a file system or volume name, a snapshot
1788 with the same name as the sent snapshot is created within the spec‐
1789 ified filesystem or volume. If the -d option is specified, the
1790 snapshot name is determined by appending the sent snapshot's name
1791 to the specified filesystem. If the -d option is specified, any
1792 required file systems within the specified one are created.
1793 -d
1795 Use the name of the sent snapshot to determine the name of the
1797 new snapshot as described in the paragraph above.
1798 -u
1801 File system that is associated with the received stream is not
1803 mounted.
1804 -v
1807 Print verbose information about the stream and the time
1809 required to perform the receive operation.
1810 -n
1813 Do not actually receive the stream. This can be useful in con‐
1815 junction with the -v option to verify the name the receive
1816 operation would use.
1817 -F
1820 Force a rollback of the file system to the most recent snapshot
1822 before performing the receive operation. If receiving an incre‐
1823 mental replication stream (for example, one generated by zfs
1824 send -R -[iI]), destroy snapshots and file systems that do not
1825 exist on the sending side.
1826 zfs allow filesystem | volume
1830 Displays permissions that have been delegated on the specified
1832 filesystem or volume. See the other forms of zfs allow for more
1833 information.
1834 zfs allow [-ldug] "everyone"|user|group[,...] perm|@setname[,...]
1837 filesystem| volume
1838 zfs allow [-ld] -e perm|@setname[,...] filesystem | volume
1839 Delegates ZFS administration permission for the file systems to
1841 non-privileged users.
1842 [-ug] "everyone"|user|group[,...]
1844 Specifies to whom the permissions are delegated. Multiple enti‐
1846 ties can be specified as a comma-separated list. If neither of
1847 the -ug options are specified, then the argument is interpreted
1848 preferentially as the keyword "everyone", then as a user name,
1849 and lastly as a group name. To specify a user or group named
1850 "everyone", use the -u or -g options. To specify a group with
1851 the same name as a user, use the -g options.
1852 [-e] perm|@setname[,...]
1855 Specifies that the permissions be delegated to "everyone." Mul‐
1857 tiple permissions may be specified as a comma-separated list.
1858 Permission names are the same as ZFS subcommand and property
1859 names. See the property list below. Property set names, which
1860 begin with an at sign (@) , may be specified. See the -s form
1861 below for details.
1862 [-ld] filesystem|volume
1865 Specifies where the permissions are delegated. If neither of
1867 the -ld options are specified, or both are, then the permis‐
1868 sions are allowed for the file system or volume, and all of its
1869 descendents. If only the -l option is used, then is allowed
1870 "locally" only for the specified file system. If only the -d
1871 option is used, then is allowed only for the descendent file
1872 systems.
1873 Permissions are generally the ability to use a ZFS subcommand or change
1878 a ZFS property. The following permissions are available:
1879 NAME TYPE NOTES
1881 allow subcommand Must also have the permission that is being
1882 allowed
1883 clone subcommand Must also have the 'create' ability and 'mount'
1884 ability in the origin file system
1885 create subcommand Must also have the 'mount' ability
1886 destroy subcommand Must also have the 'mount' ability
1887 mount subcommand Allows mount/umount of ZFS datasets
1888 promote subcommand Must also have the 'mount'
1889 and 'promote' ability in the origin file system
1890 receive subcommand Must also have the 'mount' and 'create' ability
1891 rename subcommand Must also have the 'mount' and 'create'
1892 ability in the new parent
1893 rollback subcommand Must also have the 'mount' ability
1894 send subcommand
1895 share subcommand Allows sharing file systems over NFS or SMB
1896 protocols
1897 snapshot subcommand Must also have the 'mount' ability
1898 groupquota other Allows accessing any groupquota@... property
1899 groupused other Allows reading any groupused@... property
1900 userprop other Allows changing any user property
1901 userquota other Allows accessing any userquota@... property
1902 userused other Allows reading any userused@... property
1903 aclinherit property
1905 aclmode property
1906 atime property
1907 canmount property
1908 casesensitivity property
1909 checksum property
1910 compression property
1911 copies property
1912 devices property
1913 exec property
1914 mountpoint property
1915 nbmand property
1916 normalization property
1917 primarycache property
1918 quota property
1919 readonly property
1920 recordsize property
1921 refquota property
1922 refreservation property
1923 reservation property
1924 secondarycache property
1925 setuid property
1926 shareiscsi property
1927 sharenfs property
1928 sharesmb property
1929 snapdir property
1930 utf8only property
1931 version property
1932 volblocksize property
1933 volsize property
1934 vscan property
1935 xattr property
1936 zoned property
1937 zfs allow -c perm|@setname[,...] filesystem|volume
1941 Sets "create time" permissions. These permissions are granted
1943 (locally) to the creator of any newly-created descendent file sys‐
1944 tem.
1945 zfs allow -s @setname perm|@setname[,...] filesystem|volume
1948 Defines or adds permissions to a permission set. The set can be
1950 used by other zfs allow commands for the specified file system and
1951 its descendents. Sets are evaluated dynamically, so changes to a
1952 set are immediately reflected. Permission sets follow the same nam‐
1953 ing restrictions as ZFS file systems, but the name must begin with
1954 an "at sign" (@), and can be no more than 64 characters long.
1955 zfs unallow [-rldug] "everyone"|user|group[,...] [perm|@setname[, ...]]
1958 filesystem|volume
1959 zfs unallow [-rld] -e [perm|@setname [,...]] filesystem|volume
1960 zfs unallow [-r] -c [perm|@setname[,...]]
1961 filesystem|volume
1962 Removes permissions that were granted with the zfs allow command.
1964 No permissions are explicitly denied, so other permissions granted
1965 are still in effect. For example, if the permission is granted by
1966 an ancestor. If no permissions are specified, then all permissions
1967 for the specified user, group, or everyone are removed. Specifying
1968 "everyone" (or using the -e option) only removes the permissions
1969 that were granted to "everyone", not all permissions for every user
1970 and group. See the zfs allow command for a description of the
1971 -ldugec options.
1972 -r
1974 Recursively remove the permissions from this file system and
1976 all descendents.
1977 zfs unallow [-r] -s @setname [perm|@setname[,...]]
1981 filesystem|volume
1982 Removes permissions from a permission set. If no permissions are
1984 specified, then all permissions are removed, thus removing the set
1985 entirely.
1986 zfs hold [-r] tag snapshot...
1989 Adds a single reference, named with the tag argument, to the speci‐
1991 fied snapshot or snapshots. Each snapshot has its own tag names‐
1992 pace, and tags must be unique within that space.
1993 If a hold exists on a snapshot, attempts to destroy that snapshot
1995 by using the zfs destroy command return EBUSY.
1996 -r
1998 Specifies that a hold with the given tag is applied recursively
2000 to the snapshots of all descendent file systems.
2001 zfs holds [-r] snapshot...
2005 Lists all existing user references for the given snapshot or snap‐
2007 shots.
2008 -r
2010 Lists the holds that are set on the named descendent snapshots,
2012 in addition to listing the holds on the named snapshot.
2013 zfs release [-r] tag snapshot...
2017 Removes a single reference, named with the tag argument, from the
2019 specified snapshot or snapshots. The tag must already exist for
2020 each snapshot.
2021 If a hold exists on a snapshot, attempts to destroy that snapshot
2023 by using the zfs destroy command return EBUSY.
2024 -r
2026 Recursively releases a hold with the given tag on the snapshots
2028 of all descendent file systems.
2029
2033 Example 1 Creating a ZFS File System Hierarchy
2034 The following commands create a file system named pool/home and a file
2037 system named pool/home/bob. The mount point /export/home is set for the
2038 parent file system, and is automatically inherited by the child file
2039 system.
2040 # zfs create pool/home
2043 # zfs set mountpoint=/export/home pool/home
2044 # zfs create pool/home/bob
2045 Example 2 Creating a ZFS Snapshot
2049 The following command creates a snapshot named yesterday. This snapshot
2052 is mounted on demand in the .zfs/snapshot directory at the root of the
2053 pool/home/bob file system.
2054 # zfs snapshot pool/home/bob@yesterday
2057 Example 3 Creating and Destroying Multiple Snapshots
2061 The following command creates snapshots named yesterday of pool/home
2064 and all of its descendent file systems. Each snapshot is mounted on
2065 demand in the .zfs/snapshot directory at the root of its file system.
2066 The second command destroys the newly created snapshots.
2067 # zfs snapshot -r pool/home@yesterday
2070 # zfs destroy -r pool/home@yesterday
2071 Example 4 Disabling and Enabling File System Compression
2075 The following command disables the compression property for all file
2078 systems under pool/home. The next command explicitly enables compres‐
2079 sion for pool/home/anne.
2080 # zfs set compression=off pool/home
2083 # zfs set compression=on pool/home/anne
2084 Example 5 Listing ZFS Datasets
2088 The following command lists all active file systems and volumes in the
2091 system. Snapshots are displayed if the listsnaps property is on. The
2092 default is off. See zpool(1M) for more information on pool properties.
2093 # zfs list
2096 NAME USED AVAIL REFER MOUNTPOINT
2097 pool 450K 457G 18K /pool
2098 pool/home 315K 457G 21K /export/home
2099 pool/home/anne 18K 457G 18K /export/home/anne
2100 pool/home/bob 276K 457G 276K /export/home/bob
2101 Example 6 Setting a Quota on a ZFS File System
2105 The following command sets a quota of 50 Gbytes for pool/home/bob.
2108 # zfs set quota=50G pool/home/bob
2111 Example 7 Listing ZFS Properties
2115 The following command lists all properties for pool/home/bob.
2118 # zfs get all pool/home/bob
2121 NAME PROPERTY VALUE SOURCE
2122 pool/home/bob type filesystem -
2123 pool/home/bob creation Tue Jul 21 15:53 2009 -
2124 pool/home/bob used 21K -
2125 pool/home/bob available 20.0G -
2126 pool/home/bob referenced 21K -
2127 pool/home/bob compressratio 1.00x -
2128 pool/home/bob mounted yes -
2129 pool/home/bob quota 20G local
2130 pool/home/bob reservation none default
2131 pool/home/bob recordsize 128K default
2132 pool/home/bob mountpoint /pool/home/bob default
2133 pool/home/bob sharenfs off default
2134 pool/home/bob checksum on default
2135 pool/home/bob compression on local
2136 pool/home/bob atime on default
2137 pool/home/bob devices on default
2138 pool/home/bob exec on default
2139 pool/home/bob setuid on default
2140 pool/home/bob readonly off default
2141 pool/home/bob zoned off default
2142 pool/home/bob snapdir hidden default
2143 pool/home/bob aclmode groupmask default
2144 pool/home/bob aclinherit restricted default
2145 pool/home/bob canmount on default
2146 pool/home/bob shareiscsi off default
2147 pool/home/bob xattr on default
2148 pool/home/bob copies 1 default
2149 pool/home/bob version 4 -
2150 pool/home/bob utf8only off -
2151 pool/home/bob normalization none -
2152 pool/home/bob casesensitivity sensitive -
2153 pool/home/bob vscan off default
2154 pool/home/bob nbmand off default
2155 pool/home/bob sharesmb off default
2156 pool/home/bob refquota none default
2157 pool/home/bob refreservation none default
2158 pool/home/bob primarycache all default
2159 pool/home/bob secondarycache all default
2160 pool/home/bob usedbysnapshots 0 -
2161 pool/home/bob usedbydataset 21K -
2162 pool/home/bob usedbychildren 0 -
2163 pool/home/bob usedbyrefreservation 0 -
2164 The following command gets a single property value.
2169 # zfs get -H -o value compression pool/home/bob
2172 on
2173 The following command lists all properties with local settings for
2178 pool/home/bob.
2179 # zfs get -r -s local -o name,property,value all pool/home/bob
2182 NAME PROPERTY VALUE
2183 pool/home/bob quota 20G
2184 pool/home/bob compression on
2185 Example 8 Rolling Back a ZFS File System
2189 The following command reverts the contents of pool/home/anne to the
2192 snapshot named yesterday, deleting all intermediate snapshots.
2193 # zfs rollback -r pool/home/anne@yesterday
2196 Example 9 Creating a ZFS Clone
2200 The following command creates a writable file system whose initial con‐
2203 tents are the same as pool/home/bob@yesterday.
2204 # zfs clone pool/home/bob@yesterday pool/clone
2207 Example 10 Promoting a ZFS Clone
2211 The following commands illustrate how to test out changes to a file
2214 system, and then replace the original file system with the changed one,
2215 using clones, clone promotion, and renaming:
2216 # zfs create pool/project/production
2219 populate /pool/project/production with data
2220 # zfs snapshot pool/project/production@today
2221 # zfs clone pool/project/production@today pool/project/beta
2222 make changes to /pool/project/beta and test them
2223 # zfs promote pool/project/beta
2224 # zfs rename pool/project/production pool/project/legacy
2225 # zfs rename pool/project/beta pool/project/production
2226 once the legacy version is no longer needed, it can be destroyed
2227 # zfs destroy pool/project/legacy
2228 Example 11 Inheriting ZFS Properties
2232 The following command causes pool/home/bob and pool/home/anne to
2235 inherit the checksum property from their parent.
2236 # zfs inherit checksum pool/home/bob pool/home/anne
2239 Example 12 Remotely Replicating ZFS Data
2243 The following commands send a full stream and then an incremental
2246 stream to a remote machine, restoring them into poolB/received/fs@aand
2247 poolB/received/fs@b, respectively. poolB must contain the file system
2248 poolB/received, and must not initially contain poolB/received/fs.
2249 # zfs send pool/fs@a | \
2252 ssh host zfs receive poolB/received/fs@a
2253 # zfs send -i a pool/fs@b | ssh host \
2254 zfs receive poolB/received/fs
2255 Example 13 Using the zfs receive -d Option
2259 The following command sends a full stream of poolA/fsA/fsB@snap to a
2262 remote machine, receiving it into poolB/received/fsA/fsB@snap. The
2263 fsA/fsB@snap portion of the received snapshot's name is determined from
2264 the name of the sent snapshot. poolB must contain the file system
2265 poolB/received. If poolB/received/fsA does not exist, it is created as
2266 an empty file system.
2267 # zfs send poolA/fsA/fsB@snap | \
2270 ssh host zfs receive -d poolB/received
2271 Example 14 Setting User Properties
2275 The following example sets the user-defined com.example:department
2278 property for a dataset.
2279 # zfs set com.example:department=12345 tank/accounting
2282 Example 15 Creating a ZFS Volume as an iSCSI Target Device
2286 The following example shows how to create a ZFS volume as an iSCSI tar‐
2289 get.
2290 # zfs create -V 2g pool/volumes/vol1
2293 # zfs set shareiscsi=on pool/volumes/vol1
2294 # iscsitadm list target
2295 Target: pool/volumes/vol1
2296 iSCSI Name:
2297 iqn.1986-03.com.sun:02:7b4b02a6-3277-eb1b-e686-a24762c52a8c
2298 Connections: 0
2299 After the iSCSI target is created, set up the iSCSI initiator. For more
2304 information about the Solaris iSCSI initiator, see iscsitadm(1M).
2305 Example 16 Performing a Rolling Snapshot
2307 The following example shows how to maintain a history of snapshots with
2310 a consistent naming scheme. To keep a week's worth of snapshots, the
2311 user destroys the oldest snapshot, renames the remaining snapshots, and
2312 then creates a new snapshot, as follows:
2313 # zfs destroy -r pool/users@7daysago
2316 # zfs rename -r pool/users@6daysago @7daysago
2317 # zfs rename -r pool/users@5daysago @6daysago
2318 # zfs rename -r pool/users@yesterday @5daysago
2319 # zfs rename -r pool/users@yesterday @4daysago
2320 # zfs rename -r pool/users@yesterday @3daysago
2321 # zfs rename -r pool/users@yesterday @2daysago
2322 # zfs rename -r pool/users@today @yesterday
2323 # zfs snapshot -r pool/users@today
2324 Example 17 Setting sharenfs Property Options on a ZFS File System
2328 The following commands show how to set sharenfs property options to
2331 enable rw access for a set of IP addresses and to enable root access
2332 for system neo on the tank/home file system.
2333 # # zfs set sharenfs='rw=@123.123.0.0/16,root=neo' tank/home
2336 If you are using DNS for host name resolution, specify the fully quali‐
2341 fied hostname.
2342 Example 18 Delegating ZFS Administration Permissions on a ZFS Dataset
2345 The following example shows how to set permissions so that user cindys
2348 can create, destroy, mount, and take snapshots on tank/cindys. The per‐
2349 missions on tank/cindys are also displayed.
2350 # zfs allow cindys create,destroy,mount,snapshot tank/cindys
2353 # zfs allow tank/cindys
2354 -------------------------------------------------------------
2355 Local+Descendent permissions on (tank/cindys)
2356 user cindys create,destroy,mount,snapshot
2357 -------------------------------------------------------------
2358 Because the tank/cindys mount point permission is set to 755 by
2363 default, user cindys will be unable to mount file systems under
2364 tank/cindys. Set an ACL similar to the following syntax to provide
2365 mount point access:
2366 # chmod A+user:cindys:add_subdirectory:allow /tank/cindys
2368 Example 19 Delegating Create Time Permissions on a ZFS Dataset
2372 The following example shows how to grant anyone in the group staff to
2375 create file systems in tank/users. This syntax also allows staff mem‐
2376 bers to destroy their own file systems, but not destroy anyone else's
2377 file system. The permissions on tank/users are also displayed.
2378 # # zfs allow staff create,mount tank/users
2381 # zfs allow -c destroy tank/users
2382 # zfs allow tank/users
2383 -------------------------------------------------------------
2384 Create time permissions on (tank/users)
2385 create,destroy
2386 Local+Descendent permissions on (tank/users)
2387 group staff create,mount
2388 -------------------------------------------------------------
2389 Example 20 Defining and Granting a Permission Set on a ZFS Dataset
2393 The following example shows how to define and grant a permission set on
2396 the tank/users file system. The permissions on tank/users are also dis‐
2397 played.
2398 # zfs allow -s @pset create,destroy,snapshot,mount tank/users
2401 # zfs allow staff @pset tank/users
2402 # zfs allow tank/users
2403 -------------------------------------------------------------
2404 Permission sets on (tank/users)
2405 @pset create,destroy,mount,snapshot
2406 Create time permissions on (tank/users)
2407 create,destroy
2408 Local+Descendent permissions on (tank/users)
2409 group staff @pset,create,mount
2410 -------------------------------------------------------------
2411 Example 21 Delegating Property Permissions on a ZFS Dataset
2415 The following example shows to grant the ability to set quotas and
2418 reservations on the users/home file system. The permissions on
2419 users/home are also displayed.
2420 # zfs allow cindys quota,reservation users/home
2423 # zfs allow users/home
2424 -------------------------------------------------------------
2425 Local+Descendent permissions on (users/home)
2426 user cindys quota,reservation
2427 -------------------------------------------------------------
2428 cindys% zfs set quota=10G users/home/marks
2429 cindys% zfs get quota users/home/marks
2430 NAME PROPERTY VALUE SOURCE
2431 users/home/marks quota 10G local
2432 Example 22 Removing ZFS Delegated Permissions on a ZFS Dataset
2436 The following example shows how to remove the snapshot permission from
2439 the staff group on the tank/users file system. The permissions on
2440 tank/users are also displayed.
2441 # zfs unallow staff snapshot tank/users
2444 # zfs allow tank/users
2445 -------------------------------------------------------------
2446 Permission sets on (tank/users)
2447 @pset create,destroy,mount,snapshot
2448 Create time permissions on (tank/users)
2449 create,destroy
2450 Local+Descendent permissions on (tank/users)
2451 group staff @pset,create,mount
2452 -------------------------------------------------------------
2453
2457 The following exit values are returned:
2458 0
2460 Successful completion.
2462 1
2465 An error occurred.
2467 2
2470 Invalid command line options were specified.
2472
2475 See attributes(5) for descriptions of the following attributes:
2476 ┌─────────────────────────────┬─────────────────────────────┐
2481 │ ATTRIBUTE TYPE │ ATTRIBUTE VALUE │
2482 ├─────────────────────────────┼─────────────────────────────┤
2483 │Availability │SUNWzfsu │
2484 ├─────────────────────────────┼─────────────────────────────┤
2485 │Interface Stability │Committed │
2486 └─────────────────────────────┴─────────────────────────────┘
2487
2489 ssh(1), iscsitadm(1M), mount(1M), share(1M), sharemgr(1M), unshare(1M),
2490 zonecfg(1M), zpool(1M), chmod(2), stat(2), write(2), fsync(3C),
2491 dfstab(4), attributes(5)
2492 See the gzip(1) man page, which is not part of the SunOS man page col‐
2495 lection.
2496 For information about using the ZFS web-based management tool and other
2499 ZFS features, see the Solaris ZFS Administration Guide.
2500SunOS 5.11 24 Sep 2009 zfs(1M)