xref: /netbsd-src/external/cddl/osnet/dist/cmd/zpool/zpool-features.7 (revision 3587d6f89c746bbb4f886219ddacd41ace480ecf)
1'\" te
2.\" Copyright (c) 2012, Martin Matuska <mm@FreeBSD.org>.
3.\" All Rights Reserved.
4.\"
5.\" The contents of this file are subject to the terms of the
6.\" Common Development and Distribution License (the "License").
7.\" You may not use this file except in compliance with the License.
8.\"
9.\" You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
10.\" or http://www.opensolaris.org/os/licensing.
11.\" See the License for the specific language governing permissions
12.\" and limitations under the License.
13.\"
14.\" When distributing Covered Code, include this CDDL HEADER in each
15.\" file and include the License file at usr/src/OPENSOLARIS.LICENSE.
16.\" If applicable, add the following below this CDDL HEADER, with the
17.\" fields enclosed by brackets "[]" replaced with your own identifying
18.\" information: Portions Copyright [yyyy] [name of copyright owner]
19.\"
20.\" Copyright (c) 2012 by Delphix. All rights reserved.
21.\" Copyright (c) 2013 by Saso Kiselkov. All rights reserved.
22.\" Copyright (c) 2013, Joyent, Inc. All rights reserved.
23.\"
24.\" $FreeBSD: head/cddl/contrib/opensolaris/cmd/zpool/zpool-features.7 301104 2016-06-01 06:18:34Z allanjude $
25.\"
26.Dd May 31, 2016
27.Dt ZPOOL-FEATURES 7
28.Os
29.Sh NAME
30.Nm zpool-features
31.Nd ZFS pool feature descriptions
32.Sh DESCRIPTION
33ZFS pool on\-disk format versions are specified via "features" which replace
34the old on\-disk format numbers (the last supported on\-disk format number is
3528).
36To enable a feature on a pool use the
37.Cm upgrade
38subcommand of the
39.Xr zpool 8
40command, or set the
41.Sy feature@feature_name
42property to
43.Ar enabled .
44.Pp
45The pool format does not affect file system version compatibility or the ability
46to send file systems between pools.
47.Pp
48Since most features can be enabled independently of each other the on\-disk
49format of the pool is specified by the set of all features marked as
50.Sy active
51on the pool. If the pool was created by another software version this set may
52include unsupported features.
53.Ss Identifying features
54Every feature has a guid of the form
55.Sy com.example:feature_name .
56The reverse DNS name ensures that the feature's guid is unique across all ZFS
57implementations. When unsupported features are encountered on a pool they will
58be identified by their guids.
59Refer to the documentation for the ZFS implementation that created the pool
60for information about those features.
61.Pp
62Each supported feature also has a short name.
63By convention a feature's short name is the portion of its guid which follows
64the ':' (e.g.
65.Sy com.example:feature_name
66would have the short name
67.Sy feature_name ),
68however a feature's short name may differ across ZFS implementations if
69following the convention would result in name conflicts.
70.Ss Feature states
71Features can be in one of three states:
72.Bl -tag -width "XXXXXXXX"
73.It Sy active
74This feature's on\-disk format changes are in effect on the pool.
75Support for this feature is required to import the pool in read\-write mode.
76If this feature is not read-only compatible, support is also required to
77import the pool in read\-only mode (see "Read\-only compatibility").
78.It Sy enabled
79An administrator has marked this feature as enabled on the pool, but the
80feature's on\-disk format changes have not been made yet.
81The pool can still be imported by software that does not support this feature,
82but changes may be made to the on\-disk format at any time which will move
83the feature to the
84.Sy active
85state.
86Some features may support returning to the
87.Sy enabled
88state after becoming
89.Sy active .
90See feature\-specific documentation for details.
91.It Sy disabled
92This feature's on\-disk format changes have not been made and will not be made
93unless an administrator moves the feature to the
94.Sy enabled
95state.
96Features cannot be disabled once they have been enabled.
97.El
98.Pp
99The state of supported features is exposed through pool properties of the form
100.Sy feature@short_name .
101.Ss Read\-only compatibility
102Some features may make on\-disk format changes that do not interfere with other
103software's ability to read from the pool.
104These features are referred to as "read\-only compatible".
105If all unsupported features on a pool are read\-only compatible, the pool can
106be imported in read\-only mode by setting the
107.Sy readonly
108property during import (see
109.Xr zpool 8
110for details on importing pools).
111.Ss Unsupported features
112For each unsupported feature enabled on an imported pool a pool property
113named
114.Sy unsupported@feature_guid
115will indicate why the import was allowed despite the unsupported feature.
116Possible values for this property are:
117.Bl -tag -width "XXXXXXXX"
118.It Sy inactive
119The feature is in the
120.Sy enabled
121state and therefore the pool's on\-disk format is still compatible with
122software that does not support this feature.
123.It Sy readonly
124The feature is read\-only compatible and the pool has been imported in
125read\-only mode.
126.El
127.Ss Feature dependencies
128Some features depend on other features being enabled in order to function
129properly.
130Enabling a feature will automatically enable any features it depends on.
131.Sh FEATURES
132The following features are supported on this system:
133.Bl -tag -width "XXXXXXXX"
134.It Sy async_destroy
135.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:async_destroy"
136.It GUID Ta com.delphix:async_destroy
137.It READ\-ONLY COMPATIBLE Ta yes
138.It DEPENDENCIES Ta none
139.El
140.Pp
141Destroying a file system requires traversing all of its data in order to
142return its used space to the pool.
143Without
144.Sy async_destroy
145the file system is not fully removed until all space has been reclaimed.
146If the destroy operation is interrupted by a reboot or power outage the next
147attempt to open the pool will need to complete the destroy operation
148synchronously.
149.Pp
150When
151.Sy async_destroy
152is enabled the file system's data will be reclaimed by a background process,
153allowing the destroy operation to complete without traversing the entire file
154system.
155The background process is able to resume interrupted destroys after the pool
156has been opened, eliminating the need to finish interrupted destroys as part
157of the open operation.
158The amount of space remaining to be reclaimed by the background process is
159available through the
160.Sy freeing
161property.
162.Pp
163This feature is only
164.Sy active
165while
166.Sy freeing
167is non\-zero.
168.It Sy empty_bpobj
169.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:empty_bpobj"
170.It GUID Ta com.delphix:empty_bpobj
171.It READ\-ONLY COMPATIBLE Ta yes
172.It DEPENDENCIES Ta none
173.El
174.Pp
175This feature increases the performance of creating and using a large number
176of snapshots of a single filesystem or volume, and also reduces the disk
177space required.
178.Pp
179When there are many snapshots, each snapshot uses many Block Pointer Objects
180.Pq bpobj's
181to track blocks associated with that snapshot.
182However, in common use cases, most of these bpobj's are empty.
183This feature allows us to create each bpobj on-demand, thus eliminating the
184empty bpobjs.
185.Pp
186This feature is
187.Sy active
188while there are any filesystems, volumes, or snapshots which were created
189after enabling this feature.
190.It Sy filesystem_limits
191.Bl -column "READ\-ONLY COMPATIBLE" "com.joyent:filesystem_limits"
192.It GUID Ta com.joyent:filesystem_limits
193.It READ\-ONLY COMPATIBLE Ta yes
194.It DEPENDENCIES Ta extensible_dataset
195.El
196.Pp
197This feature enables filesystem and snapshot limits.
198These limits can be used
199to control how many filesystems and/or snapshots can be created at the point in
200the tree on which the limits are set.
201.Pp
202This feature is
203.Sy active
204once either of the limit properties has been
205set on a dataset.
206Once activated the feature is never deactivated.
207.It Sy lz4_compress
208.Bl -column "READ\-ONLY COMPATIBLE" "org.illumos:lz4_compress"
209.It GUID Ta org.illumos:lz4_compress
210.It READ\-ONLY COMPATIBLE Ta no
211.It DEPENDENCIES Ta none
212.El
213.Pp
214.Sy lz4
215is a high-performance real-time compression algorithm that
216features significantly faster compression and decompression as well as a
217higher compression ratio than the older
218.Sy lzjb
219compression.
220Typically,
221.Sy lz4
222compression is approximately 50% faster on
223compressible data and 200% faster on incompressible data than
224.Sy lzjb .
225It is also approximately 80% faster on decompression, while
226giving approximately 10% better compression ratio.
227.Pp
228When the
229.Sy lz4_compress
230feature is set to
231.Sy enabled ,
232the
233administrator can turn on
234.Sy lz4
235compression on any dataset on the
236pool using the
237.Xr zfs 8
238command.
239Also, all newly written metadata
240will be compressed with
241.Sy lz4
242algorithm.
243Since this feature is not read-only compatible, this
244operation will render the pool unimportable on systems without support
245for the
246.Sy lz4_compress
247feature.
248Booting off of
249.Sy lz4
250-compressed root pools is supported.
251.Pp
252This feature becomes
253.Sy active
254as soon as it is enabled and will
255never return to being
256.Sy enabled .
257.It Sy multi_vdev_crash_dump
258.Bl -column "READ\-ONLY COMPATIBLE" "com.joyent:multi_vdev_crash_dump"
259.It GUID Ta com.joyent:multi_vdev_crash_dump
260.It READ\-ONLY COMPATIBLE Ta no
261.It DEPENDENCIES Ta none
262.El
263.Pp
264This feature allows a dump device to be configured with a pool comprised
265of multiple vdevs.
266Those vdevs may be arranged in any mirrored or raidz
267configuration.
268.\" TODO: this is not yet supported on FreeBSD.
269.\" .Pp
270.\" When the
271.\" .Sy multi_vdev_crash_dump
272.\" feature is set to
273.\" .Sy enabled ,
274.\" the administrator can use the
275.\" .Xr dumpon 8
276.\" command to configure a
277.\" dump device on a pool comprised of multiple vdevs.
278.It Sy spacemap_histogram
279.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:spacemap_histogram"
280.It GUID Ta com.delphix:spacemap_histogram
281.It READ\-ONLY COMPATIBLE Ta yes
282.It DEPENDENCIES Ta none
283.El
284.Pp
285This features allows ZFS to maintain more information about how free space
286is organized within the pool. If this feature is
287.Sy enabled ,
288ZFS will
289set this feature to
290.Sy active
291when a new space map object is created or
292an existing space map is upgraded to the new format.
293Once the feature is
294.Sy active ,
295it will remain in that state until the pool is destroyed.
296.It Sy extensible_dataset
297.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:extensible_dataset"
298.It GUID Ta com.delphix:extensible_dataset
299.It READ\-ONLY COMPATIBLE Ta no
300.It DEPENDENCIES Ta none
301.El
302.Pp
303This feature allows more flexible use of internal ZFS data structures,
304and exists for other features to depend on.
305.Pp
306This feature will be
307.Sy active
308when the first dependent feature uses it,
309and will be returned to the
310.Sy enabled
311state when all datasets that use
312this feature are destroyed.
313.It Sy bookmarks
314.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:bookmarks"
315.It GUID Ta com.delphix:bookmarks
316.It READ\-ONLY COMPATIBLE Ta yes
317.It DEPENDENCIES Ta extensible_dataset
318.El
319.Pp
320This feature enables use of the
321.Nm zfs
322.Cm bookmark
323subcommand.
324.Pp
325This feature is
326.Sy active
327while any bookmarks exist in the pool.
328All bookmarks in the pool can be listed by running
329.Nm zfs
330.Cm list
331.Fl t No bookmark Fl r Ar poolname .
332.It Sy enabled_txg
333.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:enabled_txg"
334.It GUID Ta com.delphix:enabled_txg
335.It READ\-ONLY COMPATIBLE Ta yes
336.It DEPENDENCIES Ta none
337.El
338.Pp
339Once this feature is enabled ZFS records the transaction group number
340in which new features are enabled. This has no user-visible impact,
341but other features may depend on this feature.
342.Pp
343This feature becomes
344.Sy active
345as soon as it is enabled and will
346never return to being
347.Sy enabled .
348.It Sy hole_birth
349.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:hole_birth"
350.It GUID Ta com.delphix:hole_birth
351.It READ\-ONLY COMPATIBLE Ta no
352.It DEPENDENCIES Ta enabled_txg
353.El
354.Pp
355This feature improves performance of incremental sends
356.Pq Dq zfs send -i
357and receives for objects with many holes.
358The most common case of
359hole-filled objects is zvols.
360.Pp
361An incremental send stream from snapshot
362.Sy A
363to snapshot
364.Sy B
365contains information about every block that changed between
366.Sy A
367and
368.Sy B .
369Blocks which did not change between those snapshots can be
370identified and omitted from the stream using a piece of metadata called
371the 'block birth time', but birth times are not recorded for holes
372.Pq blocks filled only with zeroes .
373Since holes created after
374.Sy A
375cannot be
376distinguished from holes created before
377.Sy A ,
378information about every
379hole in the entire filesystem or zvol is included in the send stream.
380.Pp
381For workloads where holes are rare this is not a problem.
382However, when
383incrementally replicating filesystems or zvols with many holes
384.Pq for example a zvol formatted with another filesystem
385a lot of time will
386be spent sending and receiving unnecessary information about holes that
387already exist on the receiving side.
388.Pp
389Once the
390.Sy hole_birth
391feature has been enabled the block birth times
392of all new holes will be recorded.
393Incremental sends between snapshots
394created after this feature is enabled will use this new metadata to avoid
395sending information about holes that already exist on the receiving side.
396.Pp
397This feature becomes
398.Sy active
399as soon as it is enabled and will
400never return to being
401.Sy enabled .
402.It Sy embedded_data
403.Bl -column "READ\-ONLY COMPATIBLE" "com.delphix:embedded_data"
404.It GUID Ta com.delphix:embedded_data
405.It READ\-ONLY COMPATIBLE Ta no
406.It DEPENDENCIES Ta none
407.El
408.Pp
409This feature improves the performance and compression ratio of
410highly-compressible blocks.
411Blocks whose contents can compress to 112 bytes
412or smaller can take advantage of this feature.
413.Pp
414When this feature is enabled, the contents of highly-compressible blocks are
415stored in the block "pointer" itself
416.Po a misnomer in this case, as it contains
417the compressed data, rather than a pointer to its location on disk
418.Pc .
419Thus
420the space of the block
421.Pq one sector, typically 512 bytes or 4KB
422is saved,
423and no additional i/o is needed to read and write the data block.
424.Pp
425This feature becomes
426.Sy active
427as soon as it is enabled and will
428never return to being
429.Sy enabled .
430.It Sy large_blocks
431.Bl -column "READ\-ONLY COMPATIBLE" "org.open-zfs:large_block"
432.It GUID Ta org.open-zfs:large_block
433.It READ\-ONLY COMPATIBLE Ta no
434.It DEPENDENCIES Ta extensible_dataset
435.El
436.Pp
437The
438.Sy large_block
439feature allows the record size on a dataset to be
440set larger than 128KB.
441.Pp
442This feature becomes
443.Sy active
444once a
445.Sy recordsize
446property has been set larger than 128KB, and will return to being
447.Sy enabled
448once all filesystems that have ever had their recordsize larger than 128KB
449are destroyed.
450.Pp
451Please note that booting from datasets that have recordsize greater than
452128KB is
453.Em NOT
454supported by the
455.Fx
456boot loader.
457.It Sy sha512
458.Bl -column "READ\-ONLY COMPATIBLE" "org.illumos:sha512"
459.It GUID Ta org.illumos:sha512
460.It READ\-ONLY COMPATIBLE Ta no
461.It DEPENDENCIES Ta none
462.El
463.Pp
464The
465.Sy sha512
466feature enables the use of the SHA-512/256 truncated hash algorithm
467.Pq FIPS 180-4
468for checksum and dedup.
469The native 64-bit arithmetic of SHA-512 provides an approximate 50%
470performance boost over SHA-256 on 64-bit hardware and is thus a good
471minimum-change replacement candidate for systems where hash performance is
472important, but these systems cannot for whatever reason utilize the faster
473.Sy skein
474algorithms.
475.Pp
476When the
477.Sy sha512
478feature is set to
479.Sy enabled ,
480the administrator can turn on the
481.Sy sha512
482checksum on any dataset using the
483.Dl # zfs set checksum=sha512 Ar dataset
484command.
485This feature becomes
486.Sy active
487once a
488.Sy checksum
489property has been set to
490.Sy sha512 ,
491and will return to being
492.Sy enabled
493once all filesystems that have ever had their checksum set to
494.Sy sha512
495are destroyed.
496.Pp
497Booting off of a pools utilizing SHA-512/256 is
498.Em NOT
499yet supported.
500.It Sy skein
501.Bl -column "READ\-ONLY COMPATIBLE" "org.illumos:skein"
502.It GUID Ta org.illumos:skein
503.It READ\-ONLY COMPATIBLE Ta no
504.It DEPENDENCIES Ta none
505.El
506.Pp
507The
508.Sy skein
509feature enables the use of the Skein hash algorithm for checksum and dedup.
510Skein is a high-performance secure hash algorithm that was a finalist in the
511NIST SHA-3 competition.
512It provides a very high security margin and high performance on 64-bit hardware
513.Pq 80% faster than SHA-256 .
514This implementation also utilizes the new salted checksumming functionality in
515ZFS, which means that the checksum is pre-seeded with a secret 256-bit random
516key
517.Pq stored on the pool
518before being fed the data block to be checksummed.
519Thus the produced checksums are unique to a given pool, preventing hash
520collision attacks on systems with dedup.
521.Pp
522When the
523.Sy skein
524feature is set to
525.Sy enabled ,
526the administrator can turn on the
527.Sy skein
528checksum on any dataset using the
529.Dl # zfs set checksum=skein Ar dataset
530command.
531This feature becomes
532.Sy active
533once a
534.Sy checksum
535property has been set to
536.Sy skein ,
537and will return to being
538.Sy enabled
539once all filesystems that have ever had their checksum set to
540.Sy skein
541are destroyed.
542.Pp
543Booting off of pools using
544.Sy skein
545is
546.Em NOT
547supported.
548.El
549.Sh SEE ALSO
550.Xr zpool 8
551.Sh AUTHORS
552This manual page is a
553.Xr mdoc 7
554reimplementation of the
555.Tn illumos
556manual page
557.Em zpool-features(5) ,
558modified and customized for
559.Fx
560and licensed under the Common Development and Distribution License
561.Pq Tn CDDL .
562.Pp
563The
564.Xr mdoc 7
565implementation of this manual page was initially written by
566.An Martin Matuska Aq mm@FreeBSD.org .
567