xref: /freebsd/sys/contrib/openzfs/man/man7/zpool-features.7 (revision 7a7741af18d6c8a804cc643cb7ecda9d730c6aa6)
1.\"
2.\" Copyright (c) 2012, 2018 by Delphix. All rights reserved.
3.\" Copyright (c) 2013 by Saso Kiselkov. All rights reserved.
4.\" Copyright (c) 2014, Joyent, Inc. All rights reserved.
5.\" The contents of this file are subject to the terms of the Common Development
6.\" and Distribution License (the "License").  You may not use this file except
7.\" in compliance with the License. You can obtain a copy of the license at
8.\" usr/src/OPENSOLARIS.LICENSE or https://opensource.org/licenses/CDDL-1.0.
9.\"
10.\" See the License for the specific language governing permissions and
11.\" limitations under the License. When distributing Covered Code, include this
12.\" CDDL HEADER in each file and include the License file at
13.\" usr/src/OPENSOLARIS.LICENSE.  If applicable, add the following below this
14.\" CDDL HEADER, with the fields enclosed by brackets "[]" replaced with your
15.\" own identifying information:
16.\" Portions Copyright [yyyy] [name of copyright owner]
17.\" Copyright (c) 2019, 2023, 2024, Klara, Inc.
18.\" Copyright (c) 2019, Allan Jude
19.\" Copyright (c) 2021, Colm Buckley <colm@tuatha.org>
20.\"
21.Dd October 2, 2024
22.Dt ZPOOL-FEATURES 7
23.Os
24.
25.Sh NAME
26.Nm zpool-features
27.Nd description of ZFS pool features
28.
29.Sh DESCRIPTION
30ZFS pool on-disk format versions are specified via
31.Dq features
32which replace the old on-disk format numbers
33.Pq the last supported on-disk format number is 28 .
34To enable a feature on a pool use the
35.Nm zpool Cm upgrade ,
36or set the
37.Sy feature Ns @ Ns Ar feature-name
38property to
39.Sy enabled .
40Please also see the
41.Sx Compatibility feature sets
42section for information on how sets of features may be enabled together.
43.Pp
44The pool format does not affect file system version compatibility or the ability
45to send file systems between pools.
46.Pp
47Since most features can be enabled independently of each other, the on-disk
48format of the pool is specified by the set of all features marked as
49.Sy active
50on the pool.
51If the pool was created by another software version
52this set may include unsupported features.
53.
54.Ss Identifying features
55Every feature has a GUID of the form
56.Ar com.example : Ns Ar feature-name .
57The reversed DNS name ensures that the feature's GUID is unique across all ZFS
58implementations.
59When unsupported features are encountered on a pool they will
60be identified by their GUIDs.
61Refer to the documentation for the ZFS
62implementation that created the pool for information about those features.
63.Pp
64Each supported feature also has a short name.
65By convention a feature's short name is the portion of its GUID which follows
66the
67.Sq \&:
68.Po
69i.e.
70.Ar com.example : Ns Ar feature-name
71would have the short name
72.Ar feature-name
73.Pc ,
74however a feature's short name may differ across ZFS implementations if
75following the convention would result in name conflicts.
76.
77.Ss Feature states
78Features can be in one of three states:
79.Bl -tag -width "disabled"
80.It Sy active
81This feature's on-disk format changes are in effect on the pool.
82Support for this feature is required to import the pool in read-write mode.
83If this feature is not read-only compatible,
84support is also required to import the pool in read-only mode
85.Pq see Sx Read-only compatibility .
86.It Sy enabled
87An administrator has marked this feature as enabled on the pool, but the
88feature's on-disk format changes have not been made yet.
89The pool can still be imported by software that does not support this feature,
90but changes may be made to the on-disk format at any time
91which will move the feature to the
92.Sy active
93state.
94Some features may support returning to the
95.Sy enabled
96state after becoming
97.Sy active .
98See feature-specific documentation for details.
99.It Sy disabled
100This feature's on-disk format changes have not been made and will not be made
101unless an administrator moves the feature to the
102.Sy enabled
103state.
104Features cannot be disabled once they have been enabled.
105.El
106.Pp
107The state of supported features is exposed through pool properties of the form
108.Sy feature Ns @ Ns Ar short-name .
109.
110.Ss Read-only compatibility
111Some features may make on-disk format changes that do not interfere with other
112software's ability to read from the pool.
113These features are referred to as
114.Dq read-only compatible .
115If all unsupported features on a pool are read-only compatible,
116the pool can be imported in read-only mode by setting the
117.Sy readonly
118property during import
119.Po see
120.Xr zpool-import 8
121for details on importing pools
122.Pc .
123.
124.Ss Unsupported features
125For each unsupported feature enabled on an imported pool, a pool property
126named
127.Sy unsupported Ns @ Ns Ar feature-name
128will indicate why the import was allowed despite the unsupported feature.
129Possible values for this property are:
130.Bl -tag -width "readonly"
131.It Sy inactive
132The feature is in the
133.Sy enabled
134state and therefore the pool's on-disk
135format is still compatible with software that does not support this feature.
136.It Sy readonly
137The feature is read-only compatible and the pool has been imported in
138read-only mode.
139.El
140.
141.Ss Feature dependencies
142Some features depend on other features being enabled in order to function.
143Enabling a feature will automatically enable any features it depends on.
144.
145.Ss Compatibility feature sets
146It is sometimes necessary for a pool to maintain compatibility with a
147specific on-disk format, by enabling and disabling particular features.
148The
149.Sy compatibility
150feature facilitates this by allowing feature sets to be read from text files.
151When set to
152.Sy off
153.Pq the default ,
154compatibility feature sets are disabled
155.Pq i.e. all features are enabled ;
156when set to
157.Sy legacy ,
158no features are enabled.
159When set to a comma-separated list of filenames
160.Po
161each filename may either be an absolute path, or relative to
162.Pa /etc/zfs/compatibility.d
163or
164.Pa /usr/share/zfs/compatibility.d
165.Pc ,
166the lists of requested features are read from those files,
167separated by whitespace and/or commas.
168Only features present in all files are enabled.
169.Pp
170Simple sanity checks are applied to the files:
171they must be between 1 B and 16 KiB in size, and must end with a newline
172character.
173.Pp
174The requested features are applied when a pool is created using
175.Nm zpool Cm create Fl o Sy compatibility Ns = Ns Ar …
176and controls which features are enabled when using
177.Nm zpool Cm upgrade .
178.Nm zpool Cm status
179will not show a warning about disabled features which are not part
180of the requested feature set.
181.Pp
182The special value
183.Sy legacy
184prevents any features from being enabled, either via
185.Nm zpool Cm upgrade
186or
187.Nm zpool Cm set Sy feature Ns @ Ns Ar feature-name Ns = Ns Sy enabled .
188This setting also prevents pools from being upgraded to newer on-disk versions.
189This is a safety measure to prevent new features from being
190accidentally enabled, breaking compatibility.
191.Pp
192By convention, compatibility files in
193.Pa /usr/share/zfs/compatibility.d
194are provided by the distribution, and include feature sets
195supported by important versions of popular distributions, and feature
196sets commonly supported at the start of each year.
197Compatibility files in
198.Pa /etc/zfs/compatibility.d ,
199if present, will take precedence over files with the same name in
200.Pa /usr/share/zfs/compatibility.d .
201.Pp
202If an unrecognized feature is found in these files, an error message will
203be shown.
204If the unrecognized feature is in a file in
205.Pa /etc/zfs/compatibility.d ,
206this is treated as an error and processing will stop.
207If the unrecognized feature is under
208.Pa /usr/share/zfs/compatibility.d ,
209this is treated as a warning and processing will continue.
210This difference is to allow distributions to include features
211which might not be recognized by the currently-installed binaries.
212.Pp
213Compatibility files may include comments:
214any text from
215.Sq #
216to the end of the line is ignored.
217.Pp
218.Sy Example :
219.Bd -literal -compact -offset 4n
220.No example# Nm cat Pa /usr/share/zfs/compatibility.d/grub2
221# Features which are supported by GRUB2 versions from v2.12 onwards.
222allocation_classes
223async_destroy
224block_cloning
225bookmarks
226device_rebuild
227embedded_data
228empty_bpobj
229enabled_txg
230extensible_dataset
231filesystem_limits
232hole_birth
233large_blocks
234livelist
235log_spacemap
236lz4_compress
237project_quota
238resilver_defer
239spacemap_histogram
240spacemap_v2
241userobj_accounting
242zilsaxattr
243zpool_checkpoint
244
245.No example# Nm cat Pa /usr/share/zfs/compatibility.d/grub2-2.06
246# Features which are supported by GRUB2 versions prior to v2.12.
247#
248# GRUB is not able to detect ZFS pool if snaphsot of top level boot pool
249# is created. This issue is observed with GRUB versions before v2.12 if
250# extensible_dataset feature is enabled on ZFS boot pool.
251#
252# This file lists all read-only comaptible features except
253# extensible_dataset and any other feature that depends on it.
254#
255allocation_classes
256async_destroy
257block_cloning
258device_rebuild
259embedded_data
260empty_bpobj
261enabled_txg
262hole_birth
263log_spacemap
264lz4_compress
265resilver_defer
266spacemap_histogram
267spacemap_v2
268zpool_checkpoint
269
270.No example# Nm zpool Cm create Fl o Sy compatibility Ns = Ns Ar grub2 Ar bootpool Ar vdev
271.Ed
272.Pp
273See
274.Xr zpool-create 8
275and
276.Xr zpool-upgrade 8
277for more information on how these commands are affected by feature sets.
278.
279.de feature
280.It Sy \\$2
281.Bl -tag -compact -width "READ-ONLY COMPATIBLE"
282.It GUID
283.Sy \\$1:\\$2
284.if !"\\$4"" \{\
285.It DEPENDENCIES
286\fB\\$4\fP\c
287.if !"\\$5"" , \fB\\$5\fP\c
288.if !"\\$6"" , \fB\\$6\fP\c
289.if !"\\$7"" , \fB\\$7\fP\c
290.if !"\\$8"" , \fB\\$8\fP\c
291.if !"\\$9"" , \fB\\$9\fP\c
292.\}
293.It READ-ONLY COMPATIBLE
294\\$3
295.El
296.Pp
297..
298.
299.ds instant-never \
300.No This feature becomes Sy active No as soon as it is enabled \
301and will never return to being Sy enabled .
302.
303.ds remount-upgrade \
304.No Each filesystem will be upgraded automatically when remounted, \
305or when a new file is created under that filesystem. \
306The upgrade can also be triggered on filesystems via \
307Nm zfs Cm set Sy version Ns = Ns Sy current Ar fs . \
308No The upgrade process runs in the background and may take a while to complete \
309for filesystems containing large amounts of files .
310.
311.de checksum-spiel
312When the
313.Sy \\$1
314feature is set to
315.Sy enabled ,
316the administrator can turn on the
317.Sy \\$1
318checksum on any dataset using
319.Nm zfs Cm set Sy checksum Ns = Ns Sy \\$1 Ar dset
320.Po see Xr zfs-set 8 Pc .
321This feature becomes
322.Sy active
323once a
324.Sy checksum
325property has been set to
326.Sy \\$1 ,
327and will return to being
328.Sy enabled
329once all filesystems that have ever had their checksum set to
330.Sy \\$1
331are destroyed.
332..
333.
334.Sh FEATURES
335The following features are supported on this system:
336.Bl -tag -width Ds
337.feature org.zfsonlinux allocation_classes yes
338This feature enables support for separate allocation classes.
339.Pp
340This feature becomes
341.Sy active
342when a dedicated allocation class vdev
343.Pq dedup or special
344is created with the
345.Nm zpool Cm create No or Nm zpool Cm add No commands .
346With device removal, it can be returned to the
347.Sy enabled
348state if all the dedicated allocation class vdevs are removed.
349.
350.feature com.delphix async_destroy yes
351Destroying a file system requires traversing all of its data in order to
352return its used space to the pool.
353Without
354.Sy async_destroy ,
355the file system is not fully removed until all space has been reclaimed.
356If the destroy operation is interrupted by a reboot or power outage,
357the next attempt to open the pool will need to complete the destroy
358operation synchronously.
359.Pp
360When
361.Sy async_destroy
362is enabled, the file system's data will be reclaimed by a background process,
363allowing the destroy operation to complete
364without traversing the entire file system.
365The background process is able to resume
366interrupted destroys after the pool has been opened, eliminating the need
367to finish interrupted destroys as part of the open operation.
368The amount of space remaining to be reclaimed by the background process
369is available through the
370.Sy freeing
371property.
372.Pp
373This feature is only
374.Sy active
375while
376.Sy freeing
377is non-zero.
378.
379.feature org.openzfs blake3 no extensible_dataset
380This feature enables the use of the BLAKE3 hash algorithm for checksum and
381dedup.
382BLAKE3 is a secure hash algorithm focused on high performance.
383.Pp
384.checksum-spiel blake3
385.
386.feature com.fudosecurity block_cloning yes
387When this feature is enabled ZFS will use block cloning for operations like
388.Fn copy_file_range 2 .
389Block cloning allows to create multiple references to a single block.
390It is much faster than copying the data (as the actual data is neither read nor
391written) and takes no additional space.
392Blocks can be cloned across datasets under some conditions (like equal
393.Nm recordsize ,
394the same master encryption key, etc.).
395ZFS tries its best to clone across datasets including encrypted ones.
396This is limited for various (nontrivial) reasons depending on the OS
397and/or ZFS internals.
398.Pp
399This feature becomes
400.Sy active
401when first block is cloned.
402When the last cloned block is freed, it goes back to the enabled state.
403.feature com.delphix bookmarks yes extensible_dataset
404This feature enables use of the
405.Nm zfs Cm bookmark
406command.
407.Pp
408This feature is
409.Sy active
410while any bookmarks exist in the pool.
411All bookmarks in the pool can be listed by running
412.Nm zfs Cm list Fl t Sy bookmark Fl r Ar poolname .
413.
414.feature com.datto bookmark_v2 no bookmark extensible_dataset
415This feature enables the creation and management of larger bookmarks which are
416needed for other features in ZFS.
417.Pp
418This feature becomes
419.Sy active
420when a v2 bookmark is created and will be returned to the
421.Sy enabled
422state when all v2 bookmarks are destroyed.
423.
424.feature com.delphix bookmark_written no bookmark extensible_dataset bookmark_v2
425This feature enables additional bookmark accounting fields, enabling the
426.Sy written Ns # Ns Ar bookmark
427property
428.Pq space written since a bookmark
429and estimates of send stream sizes for incrementals from bookmarks.
430.Pp
431This feature becomes
432.Sy active
433when a bookmark is created and will be
434returned to the
435.Sy enabled
436state when all bookmarks with these fields are destroyed.
437.
438.feature org.openzfs device_rebuild yes
439This feature enables the ability for the
440.Nm zpool Cm attach
441and
442.Nm zpool Cm replace
443commands to perform sequential reconstruction
444.Pq instead of healing reconstruction
445when resilvering.
446.Pp
447Sequential reconstruction resilvers a device in LBA order without immediately
448verifying the checksums.
449Once complete, a scrub is started, which then verifies the checksums.
450This approach allows full redundancy to be restored to the pool
451in the minimum amount of time.
452This two-phase approach will take longer than a healing resilver
453when the time to verify the checksums is included.
454However, unless there is additional pool damage,
455no checksum errors should be reported by the scrub.
456This feature is incompatible with raidz configurations.
457.
458This feature becomes
459.Sy active
460while a sequential resilver is in progress, and returns to
461.Sy enabled
462when the resilver completes.
463.
464.feature com.delphix device_removal no
465This feature enables the
466.Nm zpool Cm remove
467command to remove top-level vdevs,
468evacuating them to reduce the total size of the pool.
469.Pp
470This feature becomes
471.Sy active
472when the
473.Nm zpool Cm remove
474command is used
475on a top-level vdev, and will never return to being
476.Sy enabled .
477.
478.feature org.openzfs draid no
479This feature enables use of the
480.Sy draid
481vdev type.
482dRAID is a variant of RAID-Z which provides integrated distributed
483hot spares that allow faster resilvering while retaining the benefits of RAID-Z.
484Data, parity, and spare space are organized in redundancy groups
485and distributed evenly over all of the devices.
486.Pp
487This feature becomes
488.Sy active
489when creating a pool which uses the
490.Sy draid
491vdev type, or when adding a new
492.Sy draid
493vdev to an existing pool.
494.
495.feature org.illumos edonr no extensible_dataset
496This feature enables the use of the Edon-R hash algorithm for checksum,
497including for nopwrite
498.Po if compression is also enabled, an overwrite of
499a block whose checksum matches the data being written will be ignored
500.Pc .
501In an abundance of caution, Edon-R requires verification when used with
502dedup:
503.Nm zfs Cm set Sy dedup Ns = Ns Sy edonr , Ns Sy verify
504.Po see Xr zfs-set 8 Pc .
505.Pp
506Edon-R is a very high-performance hash algorithm that was part
507of the NIST SHA-3 competition.
508It provides extremely high hash performance
509.Pq over 350% faster than SHA-256 ,
510but was not selected because of its unsuitability
511as a general purpose secure hash algorithm.
512This implementation utilizes the new salted checksumming functionality
513in ZFS, which means that the checksum is pre-seeded with a secret
514256-bit random key
515.Pq stored on the pool
516before being fed the data block to be checksummed.
517Thus the produced checksums are unique to a given pool,
518preventing hash collision attacks on systems with dedup.
519.Pp
520.checksum-spiel edonr
521.
522.feature com.delphix embedded_data no
523This feature improves the performance and compression ratio of
524highly-compressible blocks.
525Blocks whose contents can compress to 112 bytes
526or smaller can take advantage of this feature.
527.Pp
528When this feature is enabled, the contents of highly-compressible blocks are
529stored in the block
530.Dq pointer
531itself
532.Po a misnomer in this case, as it contains
533the compressed data, rather than a pointer to its location on disk
534.Pc .
535Thus the space of the block
536.Pq one sector, typically 512 B or 4 KiB
537is saved, and no additional I/O is needed to read and write the data block.
538.
539\*[instant-never]
540.
541.feature com.delphix empty_bpobj yes
542This feature increases the performance of creating and using a large
543number of snapshots of a single filesystem or volume, and also reduces
544the disk space required.
545.Pp
546When there are many snapshots, each snapshot uses many Block Pointer
547Objects
548.Pq bpobjs
549to track blocks associated with that snapshot.
550However, in common use cases, most of these bpobjs are empty.
551This feature allows us to create each bpobj on-demand,
552thus eliminating the empty bpobjs.
553.Pp
554This feature is
555.Sy active
556while there are any filesystems, volumes,
557or snapshots which were created after enabling this feature.
558.
559.feature com.delphix enabled_txg yes
560Once this feature is enabled, ZFS records the transaction group number
561in which new features are enabled.
562This has no user-visible impact, but other features may depend on this feature.
563.Pp
564This feature becomes
565.Sy active
566as soon as it is enabled and will never return to being
567.Sy enabled .
568.
569.feature com.datto encryption no bookmark_v2 extensible_dataset
570This feature enables the creation and management of natively encrypted datasets.
571.Pp
572This feature becomes
573.Sy active
574when an encrypted dataset is created and will be returned to the
575.Sy enabled
576state when all datasets that use this feature are destroyed.
577.
578.feature com.klarasystems fast_dedup yes
579This feature allows more advanced deduplication features to be enabled on new
580dedup tables.
581.Pp
582This feature will be
583.Sy active
584when the first deduplicated block is written after a new dedup table is created
585(ie after a new pool creation, or new checksum used on a dataset with
586.Sy dedup
587enabled).
588It will be returned to the
589.Sy enabled
590state when all deduplicated blocks using it are freed.
591.
592.feature com.delphix extensible_dataset no
593This feature allows more flexible use of internal ZFS data structures,
594and exists for other features to depend on.
595.Pp
596This feature will be
597.Sy active
598when the first dependent feature uses it, and will be returned to the
599.Sy enabled
600state when all datasets that use this feature are destroyed.
601.
602.feature com.joyent filesystem_limits yes extensible_dataset
603This feature enables filesystem and snapshot limits.
604These limits can be used to control how many filesystems and/or snapshots
605can be created at the point in the tree on which the limits are set.
606.Pp
607This feature is
608.Sy active
609once either of the limit properties has been set on a dataset
610and will never return to being
611.Sy enabled .
612.
613.feature com.delphix head_errlog no
614This feature enables the upgraded version of errlog, which required an on-disk
615error log format change.
616Now the error log of each head dataset is stored separately in the zap object
617and keyed by the head id.
618With this feature enabled, every dataset affected by an error block is listed
619in the output of
620.Nm zpool Cm status .
621In case of encrypted filesystems with unloaded keys we are unable to check
622their snapshots or clones for errors and these will not be reported.
623An "access denied" error will be reported.
624.Pp
625\*[instant-never]
626.
627.feature com.delphix hole_birth no enabled_txg
628This feature has/had bugs, the result of which is that, if you do a
629.Nm zfs Cm send Fl i
630.Pq or Fl R , No since it uses Fl i
631from an affected dataset, the receiving party will not see any checksum
632or other errors, but the resulting destination snapshot
633will not match the source.
634Its use by
635.Nm zfs Cm send Fl i
636has been disabled by default
637.Po
638see
639.Sy send_holes_without_birth_time
640in
641.Xr zfs 4
642.Pc .
643.Pp
644This feature improves performance of incremental sends
645.Pq Nm zfs Cm send Fl i
646and receives for objects with many holes.
647The most common case of hole-filled objects is zvols.
648.Pp
649An incremental send stream from snapshot
650.Sy A No to snapshot Sy B
651contains information about every block that changed between
652.Sy A No and Sy B .
653Blocks which did not change between those snapshots can be
654identified and omitted from the stream using a piece of metadata called
655the
656.Dq block birth time ,
657but birth times are not recorded for holes
658.Pq blocks filled only with zeroes .
659Since holes created after
660.Sy A No cannot be distinguished from holes created before Sy A ,
661information about every hole in the entire filesystem or zvol
662is included in the send stream.
663.Pp
664For workloads where holes are rare this is not a problem.
665However, when incrementally replicating filesystems or zvols with many holes
666.Pq for example a zvol formatted with another filesystem
667a lot of time will be spent sending and receiving unnecessary information
668about holes that already exist on the receiving side.
669.Pp
670Once the
671.Sy hole_birth
672feature has been enabled the block birth times
673of all new holes will be recorded.
674Incremental sends between snapshots created after this feature is enabled
675will use this new metadata to avoid sending information about holes that
676already exist on the receiving side.
677.Pp
678\*[instant-never]
679.
680.feature org.open-zfs large_blocks no extensible_dataset
681This feature allows the record size on a dataset to be set larger than 128 KiB.
682.Pp
683This feature becomes
684.Sy active
685once a dataset contains a file with a block size larger than 128 KiB,
686and will return to being
687.Sy enabled
688once all filesystems that have ever had their recordsize larger than 128 KiB
689are destroyed.
690.
691.feature org.zfsonlinux large_dnode no extensible_dataset
692This feature allows the size of dnodes in a dataset to be set larger than 512 B.
693.
694This feature becomes
695.Sy active
696once a dataset contains an object with a dnode larger than 512 B,
697which occurs as a result of setting the
698.Sy dnodesize
699dataset property to a value other than
700.Sy legacy .
701The feature will return to being
702.Sy enabled
703once all filesystems that have ever contained a dnode larger than 512 B
704are destroyed.
705Large dnodes allow more data to be stored in the bonus buffer,
706thus potentially improving performance by avoiding the use of spill blocks.
707.
708.feature com.klarasystems large_microzap yes extensible_dataset large_blocks
709This feature allows "micro" ZAPs to grow larger than 128 KiB without being
710upgraded to "fat" ZAPs.
711.Pp
712This feature becomes
713.Sy active
714the first time a micro ZAP grows larger than 128KiB.
715It will only be returned to the
716.Sy enabled
717state when all datasets that ever had a large micro ZAP are destroyed.
718.Pp
719Note that even when this feature is enabled, micro ZAPs cannot grow larger
720than 128 KiB without also changing the
721.Sy zap_micro_max_size
722module parameter.
723See
724.Xr zfs 4 .
725.
726.feature com.delphix livelist yes extensible_dataset
727This feature allows clones to be deleted faster than the traditional method
728when a large number of random/sparse writes have been made to the clone.
729All blocks allocated and freed after a clone is created are tracked by the
730the clone's livelist which is referenced during the deletion of the clone.
731The feature is activated when a clone is created and remains
732.Sy active
733until all clones have been destroyed.
734.
735.feature com.delphix log_spacemap yes com.delphix:spacemap_v2
736This feature improves performance for heavily-fragmented pools,
737especially when workloads are heavy in random-writes.
738It does so by logging all the metaslab changes on a single spacemap every TXG
739instead of scattering multiple writes to all the metaslab spacemaps.
740.Pp
741\*[instant-never]
742.
743.feature org.zfsonlinux longname no extensible_dataset
744This feature allows creating files and directories with name up to 1023 bytes
745in length.
746A new dataset property
747.Sy longname
748is also introduced to toggle longname support for each dataset individually.
749This property can be disabled even if it contains longname files.
750In such case, new file cannot be created with longname but existing longname
751files can still be looked up.
752.Pp
753This feature becomes
754.Sy active
755when a file name greater than 255 is created in a dataset, and returns to
756being
757.Sy enabled
758when all such datasets are destroyed.
759.
760.feature org.illumos lz4_compress no
761.Sy lz4
762is a high-performance real-time compression algorithm that
763features significantly faster compression and decompression as well as a
764higher compression ratio than the older
765.Sy lzjb
766compression.
767Typically,
768.Sy lz4
769compression is approximately 50% faster on compressible data and 200% faster
770on incompressible data than
771.Sy lzjb .
772It is also approximately 80% faster on decompression,
773while giving approximately a 10% better compression ratio.
774.Pp
775When the
776.Sy lz4_compress
777feature is set to
778.Sy enabled ,
779the administrator can turn on
780.Sy lz4
781compression on any dataset on the pool using the
782.Xr zfs-set 8
783command.
784All newly written metadata will be compressed with the
785.Sy lz4
786algorithm.
787.Pp
788\*[instant-never]
789.
790.feature com.joyent multi_vdev_crash_dump no
791This feature allows a dump device to be configured with a pool comprised
792of multiple vdevs.
793Those vdevs may be arranged in any mirrored or raidz configuration.
794.Pp
795When the
796.Sy multi_vdev_crash_dump
797feature is set to
798.Sy enabled ,
799the administrator can use
800.Xr dumpadm 8
801to configure a dump device on a pool comprised of multiple vdevs.
802.Pp
803Under
804.Fx
805and Linux this feature is unused, but registered for compatibility.
806New pools created on these systems will have the feature
807.Sy enabled
808but will never transition to
809.Sy active ,
810as this functionality is not required for crash dump support.
811Existing pools where this feature is
812.Sy active
813can be imported.
814.
815.feature com.delphix obsolete_counts yes device_removal
816This feature is an enhancement of
817.Sy device_removal ,
818which will over time reduce the memory used to track removed devices.
819When indirect blocks are freed or remapped,
820we note that their part of the indirect mapping is
821.Dq obsolete
822– no longer needed.
823.Pp
824This feature becomes
825.Sy active
826when the
827.Nm zpool Cm remove
828command is used on a top-level vdev, and will never return to being
829.Sy enabled .
830.
831.feature org.zfsonlinux project_quota yes extensible_dataset
832This feature allows administrators to account the spaces and objects usage
833information against the project identifier
834.Pq ID .
835.Pp
836The project ID is an object-based attribute.
837When upgrading an existing filesystem,
838objects without a project ID will be assigned a zero project ID.
839When this feature is enabled, newly created objects inherit
840their parent directories' project ID if the parent's inherit flag is set
841.Pq via Nm chattr Sy [+-]P No or Nm zfs Cm project Fl s Ns | Ns Fl C .
842Otherwise, the new object's project ID will be zero.
843An object's project ID can be changed at any time by the owner
844.Pq or privileged user
845via
846.Nm chattr Fl p Ar prjid
847or
848.Nm zfs Cm project Fl p Ar prjid .
849.Pp
850This feature will become
851.Sy active
852as soon as it is enabled and will never return to being
853.Sy disabled .
854\*[remount-upgrade]
855.
856.feature org.openzfs raidz_expansion no none
857This feature enables the
858.Nm zpool Cm attach
859subcommand to attach a new device to a RAID-Z group, expanding the total
860amount usable space in the pool.
861See
862.Xr zpool-attach 8 .
863.
864.feature com.delphix redaction_bookmarks no bookmarks extensible_dataset
865This feature enables the use of redacted
866.Nm zfs Cm send Ns s ,
867which create redaction bookmarks storing the list of blocks
868redacted by the send that created them.
869For more information about redacted sends, see
870.Xr zfs-send 8 .
871.
872.feature com.delphix redacted_datasets no extensible_dataset
873This feature enables the receiving of redacted
874.Nm zfs Cm send
875streams, which create redacted datasets when received.
876These datasets are missing some of their blocks,
877and so cannot be safely mounted, and their contents cannot be safely read.
878For more information about redacted receives, see
879.Xr zfs-send 8 .
880.
881.feature com.delphix redaction_list_spill no redaction_bookmarks
882This feature enables the redaction list created by zfs redact to store
883many more entries.
884It becomes
885.Sy active
886when a redaction list is created with more than 36 entries,
887and returns to being
888.Sy enabled
889when no long redaction lists remain in the pool.
890For more information about redacted sends, see
891.Xr zfs-send 8 .
892.
893.feature com.datto resilver_defer yes
894This feature allows ZFS to postpone new resilvers if an existing one is already
895in progress.
896Without this feature, any new resilvers will cause the currently
897running one to be immediately restarted from the beginning.
898.Pp
899This feature becomes
900.Sy active
901once a resilver has been deferred, and returns to being
902.Sy enabled
903when the deferred resilver begins.
904.
905.feature org.illumos sha512 no extensible_dataset
906This feature enables the use of the SHA-512/256 truncated hash algorithm
907.Pq FIPS 180-4
908for checksum and dedup.
909The native 64-bit arithmetic of SHA-512 provides an approximate 50%
910performance boost over SHA-256 on 64-bit hardware
911and is thus a good minimum-change replacement candidate
912for systems where hash performance is important,
913but these systems cannot for whatever reason utilize the faster
914.Sy skein No and Sy edonr
915algorithms.
916.Pp
917.checksum-spiel sha512
918.
919.feature org.illumos skein no extensible_dataset
920This feature enables the use of the Skein hash algorithm for checksum and dedup.
921Skein is a high-performance secure hash algorithm that was a
922finalist in the NIST SHA-3 competition.
923It provides a very high security margin and high performance on 64-bit hardware
924.Pq 80% faster than SHA-256 .
925This implementation also utilizes the new salted checksumming
926functionality in ZFS, which means that the checksum is pre-seeded with a
927secret 256-bit random key
928.Pq stored on the pool
929before being fed the data block to be checksummed.
930Thus the produced checksums are unique to a given pool,
931preventing hash collision attacks on systems with dedup.
932.Pp
933.checksum-spiel skein
934.
935.feature com.delphix spacemap_histogram yes
936This features allows ZFS to maintain more information about how free space
937is organized within the pool.
938If this feature is
939.Sy enabled ,
940it will be activated when a new space map object is created, or
941an existing space map is upgraded to the new format,
942and never returns back to being
943.Sy enabled .
944.
945.feature com.delphix spacemap_v2 yes
946This feature enables the use of the new space map encoding which
947consists of two words
948.Pq instead of one
949whenever it is advantageous.
950The new encoding allows space maps to represent large regions of
951space more efficiently on-disk while also increasing their maximum
952addressable offset.
953.Pp
954This feature becomes
955.Sy active
956once it is
957.Sy enabled ,
958and never returns back to being
959.Sy enabled .
960.
961.feature org.zfsonlinux userobj_accounting yes extensible_dataset
962This feature allows administrators to account the object usage information
963by user and group.
964.Pp
965\*[instant-never]
966\*[remount-upgrade]
967.
968.feature com.klarasystems vdev_zaps_v2 no
969This feature creates a ZAP object for the root vdev.
970.Pp
971This feature becomes active after the next
972.Nm zpool Cm import
973or
974.Nm zpool reguid .
975.
976Properties can be retrieved or set on the root vdev using
977.Nm zpool Cm get
978and
979.Nm zpool Cm set
980with
981.Sy root
982as the vdev name which is an alias for
983.Sy root-0 .
984.feature org.openzfs zilsaxattr yes extensible_dataset
985This feature enables
986.Sy xattr Ns = Ns Sy sa
987extended attribute logging in the ZIL.
988If enabled, extended attribute changes
989.Pq both Sy xattrdir Ns = Ns Sy dir No and Sy xattr Ns = Ns Sy sa
990are guaranteed to be durable if either the dataset had
991.Sy sync Ns = Ns Sy always
992set at the time the changes were made, or
993.Xr sync 2
994is called on the dataset after the changes were made.
995.Pp
996This feature becomes
997.Sy active
998when a ZIL is created for at least one dataset and will be returned to the
999.Sy enabled
1000state when it is destroyed for all datasets that use this feature.
1001.
1002.feature com.delphix zpool_checkpoint yes
1003This feature enables the
1004.Nm zpool Cm checkpoint
1005command that can checkpoint the state of the pool
1006at the time it was issued and later rewind back to it or discard it.
1007.Pp
1008This feature becomes
1009.Sy active
1010when the
1011.Nm zpool Cm checkpoint
1012command is used to checkpoint the pool.
1013The feature will only return back to being
1014.Sy enabled
1015when the pool is rewound or the checkpoint has been discarded.
1016.
1017.feature org.freebsd zstd_compress no extensible_dataset
1018.Sy zstd
1019is a high-performance compression algorithm that features a
1020combination of high compression ratios and high speed.
1021Compared to
1022.Sy gzip ,
1023.Sy zstd
1024offers slightly better compression at much higher speeds.
1025Compared to
1026.Sy lz4 ,
1027.Sy zstd
1028offers much better compression while being only modestly slower.
1029Typically,
1030.Sy zstd
1031compression speed ranges from 250 to 500 MB/s per thread
1032and decompression speed is over 1 GB/s per thread.
1033.Pp
1034When the
1035.Sy zstd
1036feature is set to
1037.Sy enabled ,
1038the administrator can turn on
1039.Sy zstd
1040compression of any dataset using
1041.Nm zfs Cm set Sy compress Ns = Ns Sy zstd Ar dset
1042.Po see Xr zfs-set 8 Pc .
1043This feature becomes
1044.Sy active
1045once a
1046.Sy compress
1047property has been set to
1048.Sy zstd ,
1049and will return to being
1050.Sy enabled
1051once all filesystems that have ever had their
1052.Sy compress
1053property set to
1054.Sy zstd
1055are destroyed.
1056.El
1057.
1058.Sh SEE ALSO
1059.Xr zfs 8 ,
1060.Xr zpool 8
1061