| blob | 930efd9d91c8895460db638e94a727ef8a73dda1 |
1 '\" te
2 .\" Copyright (c) 2013, 2016 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 .\" Copyright (c) 2014 Integros [integros.com]
6 .\" The contents of this file are subject to the terms of the Common Development
7 .\" and Distribution License (the "License"). You may not use this file except
8 .\" in compliance with the License. You can obtain a copy of the license at
9 .\" usr/src/OPENSOLARIS.LICENSE or http://www.opensolaris.org/os/licensing.
10 .\"
11 .\" See the License for the specific language governing permissions and
12 .\" limitations under the License. When distributing Covered Code, include this
13 .\" CDDL HEADER in each file and include the License file at
14 .\" usr/src/OPENSOLARIS.LICENSE. If applicable, add the following below this
15 .\" CDDL HEADER, with the fields enclosed by brackets "[]" replaced with your
16 .\" own identifying information:
17 .\" Portions Copyright [yyyy] [name of copyright owner]
18 .TH ZPOOL-FEATURES 5 "Aug 27, 2013"
19 .SH NAME
20 zpool\-features \- ZFS pool feature descriptions
21 .SH DESCRIPTION
22 .LP
23 ZFS pool on\-disk format versions are specified via "features" which replace
24 the old on\-disk format numbers (the last supported on\-disk format number is
25 28). To enable a feature on a pool use the \fBupgrade\fR subcommand of the
26 \fBzpool\fR(1M) command, or set the \fBfeature@\fR\fIfeature_name\fR property
27 to \fBenabled\fR.
28 .sp
29 .LP
30 The pool format does not affect file system version compatibility or the ability
31 to send file systems between pools.
32 .sp
33 .LP
34 Since most features can be enabled independently of each other the on\-disk
35 format of the pool is specified by the set of all features marked as
36 \fBactive\fR on the pool. If the pool was created by another software version
37 this set may include unsupported features.
38 .SS "Identifying features"
39 .LP
40 Every feature has a guid of the form \fIcom.example:feature_name\fR. The reverse
41 DNS name ensures that the feature's guid is unique across all ZFS
42 implementations. When unsupported features are encountered on a pool they will
43 be identified by their guids. Refer to the documentation for the ZFS
44 implementation that created the pool for information about those features.
45 .sp
46 .LP
47 Each supported feature also has a short name. By convention a feature's short
48 name is the portion of its guid which follows the ':' (e.g.
49 \fIcom.example:feature_name\fR would have the short name \fIfeature_name\fR),
50 however a feature's short name may differ across ZFS implementations if
51 following the convention would result in name conflicts.
52 .SS "Feature states"
53 .LP
54 Features can be in one of three states:
55 .sp
56 .ne 2
57 .na
58 \fB\fBactive\fR\fR
59 .ad
60 .RS 12n
61 This feature's on\-disk format changes are in effect on the pool. Support for
62 this feature is required to import the pool in read\-write mode. If this
63 feature is not read-only compatible, support is also required to import the pool
64 in read\-only mode (see "Read\-only compatibility").
65 .RE
67 .sp
68 .ne 2
69 .na
70 \fB\fBenabled\fR\fR
71 .ad
72 .RS 12n
73 An administrator has marked this feature as enabled on the pool, but the
74 feature's on\-disk format changes have not been made yet. The pool can still be
75 imported by software that does not support this feature, but changes may be made
76 to the on\-disk format at any time which will move the feature to the
77 \fBactive\fR state. Some features may support returning to the \fBenabled\fR
78 state after becoming \fBactive\fR. See feature\-specific documentation for
79 details.
80 .RE
82 .sp
83 .ne 2
84 .na
85 \fBdisabled\fR
86 .ad
87 .RS 12n
88 This feature's on\-disk format changes have not been made and will not be made
89 unless an administrator moves the feature to the \fBenabled\fR state. Features
90 cannot be disabled once they have been enabled.
91 .RE
93 .sp
94 .LP
95 The state of supported features is exposed through pool properties of the form
96 \fIfeature@short_name\fR.
97 .SS "Read\-only compatibility"
98 .LP
99 Some features may make on\-disk format changes that do not interfere with other
100 software's ability to read from the pool. These features are referred to as
101 "read\-only compatible". If all unsupported features on a pool are read\-only
102 compatible, the pool can be imported in read\-only mode by setting the
103 \fBreadonly\fR property during import (see \fBzpool\fR(1M) for details on
104 importing pools).
105 .SS "Unsupported features"
106 .LP
107 For each unsupported feature enabled on an imported pool a pool property
108 named \fIunsupported@feature_guid\fR will indicate why the import was allowed
109 despite the unsupported feature. Possible values for this property are:
111 .sp
112 .ne 2
113 .na
114 \fB\fBinactive\fR\fR
115 .ad
116 .RS 12n
117 The feature is in the \fBenabled\fR state and therefore the pool's on\-disk
118 format is still compatible with software that does not support this feature.
119 .RE
121 .sp
122 .ne 2
123 .na
124 \fB\fBreadonly\fR\fR
125 .ad
126 .RS 12n
127 The feature is read\-only compatible and the pool has been imported in
128 read\-only mode.
129 .RE
131 .SS "Feature dependencies"
132 .LP
133 Some features depend on other features being enabled in order to function
134 properly. Enabling a feature will automatically enable any features it
135 depends on.
136 .SH FEATURES
137 .LP
138 The following features are supported on this system:
139 .sp
140 .ne 2
141 .na
142 \fB\fBasync_destroy\fR\fR
143 .ad
144 .RS 4n
145 .TS
146 l l .
147 GUID com.delphix:async_destroy
148 READ\-ONLY COMPATIBLE yes
149 DEPENDENCIES none
150 .TE
152 Destroying a file system requires traversing all of its data in order to
153 return its used space to the pool. Without \fBasync_destroy\fR the file system
154 is not fully removed until all space has been reclaimed. If the destroy
155 operation is interrupted by a reboot or power outage the next attempt to open
156 the pool will need to complete the destroy operation synchronously.
158 When \fBasync_destroy\fR is enabled the file system's data will be reclaimed
159 by a background process, allowing the destroy operation to complete without
160 traversing the entire file system. The background process is able to resume
161 interrupted destroys after the pool has been opened, eliminating the need
162 to finish interrupted destroys as part of the open operation. The amount
163 of space remaining to be reclaimed by the background process is available
164 through the \fBfreeing\fR property.
166 This feature is only \fBactive\fR while \fBfreeing\fR is non\-zero.
167 .RE
169 .sp
170 .ne 2
171 .na
172 \fB\fBempty_bpobj\fR\fR
173 .ad
174 .RS 4n
175 .TS
176 l l .
177 GUID com.delphix:empty_bpobj
178 READ\-ONLY COMPATIBLE yes
179 DEPENDENCIES none
180 .TE
182 This feature increases the performance of creating and using a large
183 number of snapshots of a single filesystem or volume, and also reduces
184 the disk space required.
186 When there are many snapshots, each snapshot uses many Block Pointer
187 Objects (bpobj's) to track blocks associated with that snapshot.
188 However, in common use cases, most of these bpobj's are empty. This
189 feature allows us to create each bpobj on-demand, thus eliminating the
190 empty bpobjs.
192 This feature is \fBactive\fR while there are any filesystems, volumes,
193 or snapshots which were created after enabling this feature.
194 .RE
196 .sp
197 .ne 2
198 .na
199 \fB\fBfilesystem_limits\fR\fR
200 .ad
201 .RS 4n
202 .TS
203 l l .
204 GUID com.joyent:filesystem_limits
205 READ\-ONLY COMPATIBLE yes
206 DEPENDENCIES extensible_dataset
207 .TE
209 This feature enables filesystem and snapshot limits. These limits can be used
210 to control how many filesystems and/or snapshots can be created at the point in
211 the tree on which the limits are set.
213 This feature is \fBactive\fR once either of the limit properties has been
214 set on a dataset. Once activated the feature is never deactivated.
215 .RE
217 .sp
218 .ne 2
219 .na
220 \fB\fBlz4_compress\fR\fR
221 .ad
222 .RS 4n
223 .TS
224 l l .
225 GUID org.illumos:lz4_compress
226 READ\-ONLY COMPATIBLE no
227 DEPENDENCIES none
228 .TE
230 \fBlz4\fR is a high-performance real-time compression algorithm that
231 features significantly faster compression and decompression as well as a
232 higher compression ratio than the older \fBlzjb\fR compression.
233 Typically, \fBlz4\fR compression is approximately 50% faster on
234 compressible data and 200% faster on incompressible data than
235 \fBlzjb\fR. It is also approximately 80% faster on decompression, while
236 giving approximately 10% better compression ratio.
238 When the \fBlz4_compress\fR feature is set to \fBenabled\fR, the
239 administrator can turn on \fBlz4\fR compression on any dataset on the
240 pool using the \fBzfs\fR(1M) command. Also, all newly written metadata
241 will be compressed with \fBlz4\fR algorithm. Since this feature is not
242 read-only compatible, this operation will render the pool unimportable
243 on systems without support for the \fBlz4_compress\fR feature. Booting
244 off of \fBlz4\fR-compressed root pools is supported.
246 This feature becomes \fBactive\fR as soon as it is enabled and will
247 never return to being \fBenabled\fR.
248 .RE
250 .sp
251 .ne 2
252 .na
253 \fB\fBspacemap_histogram\fR\fR
254 .ad
255 .RS 4n
256 .TS
257 l l .
258 GUID com.delphix:spacemap_histogram
259 READ\-ONLY COMPATIBLE yes
260 DEPENDENCIES none
261 .TE
263 This features allows ZFS to maintain more information about how free space
264 is organized within the pool. If this feature is \fBenabled\fR, ZFS will
265 set this feature to \fBactive\fR when a new space map object is created or
266 an existing space map is upgraded to the new format. Once the feature is
267 \fBactive\fR, it will remain in that state until the pool is destroyed.
268 .RE
270 .sp
271 .ne 2
272 .na
273 \fB\fBmulti_vdev_crash_dump\fR\fR
274 .ad
275 .RS 4n
276 .TS
277 l l .
278 GUID com.joyent:multi_vdev_crash_dump
279 READ\-ONLY COMPATIBLE no
280 DEPENDENCIES none
281 .TE
283 This feature allows a dump device to be configured with a pool comprised
284 of multiple vdevs. Those vdevs may be arranged in any mirrored or raidz
285 configuration.
287 When the \fBmulti_vdev_crash_dump\fR feature is set to \fBenabled\fR,
288 the administrator can use the \fBdumpadm\fR(1M) command to configure a
289 dump device on a pool comprised of multiple vdevs.
290 .RE
292 .sp
293 .ne 2
294 .na
295 \fB\fBextensible_dataset\fR\fR
296 .ad
297 .RS 4n
298 .TS
299 l l .
300 GUID com.delphix:extensible_dataset
301 READ\-ONLY COMPATIBLE no
302 DEPENDENCIES none
303 .TE
305 This feature allows more flexible use of internal ZFS data structures,
306 and exists for other features to depend on.
308 This feature will be \fBactive\fR when the first dependent feature uses it,
309 and will be returned to the \fBenabled\fR state when all datasets that use
310 this feature are destroyed.
312 .RE
314 .sp
315 .ne 2
316 .na
317 \fB\fBbookmarks\fR\fR
318 .ad
319 .RS 4n
320 .TS
321 l l .
322 GUID com.delphix:bookmarks
323 READ\-ONLY COMPATIBLE yes
324 DEPENDENCIES extensible_dataset
325 .TE
327 This feature enables use of the \fBzfs bookmark\fR subcommand.
329 This feature is \fBactive\fR while any bookmarks exist in the pool.
330 All bookmarks in the pool can be listed by running
331 \fBzfs list -t bookmark -r \fIpoolname\fR\fR.
333 .RE
335 .sp
336 .ne 2
337 .na
338 \fB\fBenabled_txg\fR\fR
339 .ad
340 .RS 4n
341 .TS
342 l l .
343 GUID com.delphix:enabled_txg
344 READ\-ONLY COMPATIBLE yes
345 DEPENDENCIES none
346 .TE
348 Once this feature is enabled ZFS records the transaction group number
349 in which new features are enabled. This has no user-visible impact,
350 but other features may depend on this feature.
352 This feature becomes \fBactive\fR as soon as it is enabled and will
353 never return to being \fBenabled\fR.
355 .RE
357 .sp
358 .ne 2
359 .na
360 \fB\fBhole_birth\fR\fR
361 .ad
362 .RS 4n
363 .TS
364 l l .
365 GUID com.delphix:hole_birth
366 READ\-ONLY COMPATIBLE no
367 DEPENDENCIES enabled_txg
368 .TE
370 This feature improves performance of incremental sends ("zfs send -i")
371 and receives for objects with many holes. The most common case of
372 hole-filled objects is zvols.
374 An incremental send stream from snapshot \fBA\fR to snapshot \fBB\fR
375 contains information about every block that changed between \fBA\fR and
376 \fBB\fR. Blocks which did not change between those snapshots can be
377 identified and omitted from the stream using a piece of metadata called
378 the 'block birth time', but birth times are not recorded for holes (blocks
379 filled only with zeroes). Since holes created after \fBA\fR cannot be
380 distinguished from holes created before \fBA\fR, information about every
381 hole in the entire filesystem or zvol is included in the send stream.
383 For workloads where holes are rare this is not a problem. However, when
384 incrementally replicating filesystems or zvols with many holes (for
385 example a zvol formatted with another filesystem) a lot of time will
386 be spent sending and receiving unnecessary information about holes that
387 already exist on the receiving side.
389 Once the \fBhole_birth\fR feature has been enabled the block birth times
390 of all new holes will be recorded. Incremental sends between snapshots
391 created after this feature is enabled will use this new metadata to avoid
392 sending information about holes that already exist on the receiving side.
394 This feature becomes \fBactive\fR as soon as it is enabled and will
395 never return to being \fBenabled\fR.
397 .RE
399 .sp
400 .ne 2
401 .na
402 \fB\fBembedded_data\fR\fR
403 .ad
404 .RS 4n
405 .TS
406 l l .
407 GUID com.delphix:embedded_data
408 READ\-ONLY COMPATIBLE no
409 DEPENDENCIES none
410 .TE
412 This feature improves the performance and compression ratio of
413 highly-compressible blocks. Blocks whose contents can compress to 112 bytes
414 or smaller can take advantage of this feature.
416 When this feature is enabled, the contents of highly-compressible blocks are
417 stored in the block "pointer" itself (a misnomer in this case, as it contains
418 the compresseed data, rather than a pointer to its location on disk). Thus
419 the space of the block (one sector, typically 512 bytes or 4KB) is saved,
420 and no additional i/o is needed to read and write the data block.
422 This feature becomes \fBactive\fR as soon as it is enabled and will
423 never return to being \fBenabled\fR.
425 .RE
427 .sp
428 .ne 2
429 .na
430 \fB\fBdevice_removal\fR\fR
431 .ad
432 .RS 4n
433 .TS
434 l l .
435 GUID com.delphix:device_removal
436 READ\-ONLY COMPATIBLE no
437 DEPENDENCIES none
438 .TE
440 This feature enables the "zpool remove" subcommand to remove top-level
441 vdevs, evacuating them to reduce the total size of the pool.
443 This feature becomes \fBactive\fR when the "zpool remove" command is used
444 on a top-level vdev, and will never return to being \fBenabled\fR.
446 .RE
447 .sp
448 .ne 2
449 .na
450 \fB\fBobsolete_counts\fR\fR
451 .ad
452 .RS 4n
453 .TS
454 l l .
455 GUID com.delphix:obsolete_counts
456 READ\-ONLY COMPATIBLE yes
457 DEPENDENCIES device_removal
458 .TE
460 This feature is an enhancement of device_removal, which will over time
461 reduce the memory used to track removed devices. When indirect blocks
462 are freed or remapped, we note that their part of the indirect mapping
463 is "obsolete", i.e. no longer needed. See also the \fBzfs remap\fR
464 subcommand in \fBzfs\fR(1M).
466 This feature becomes \fBactive\fR when the "zpool remove" command is
467 used on a top-level vdev, and will never return to being \fBenabled\fR.
469 .RE
470 .sp
471 .ne 2
472 .na
473 \fB\fBlarge_blocks\fR\fR
474 .ad
475 .RS 4n
476 .TS
477 l l .
478 GUID org.open-zfs:large_block
479 READ\-ONLY COMPATIBLE no
480 DEPENDENCIES extensible_dataset
481 .TE
483 The \fBlarge_block\fR feature allows the record size on a dataset to be
484 set larger than 128KB.
486 This feature becomes \fBactive\fR once a \fBrecordsize\fR property has been
487 set larger than 128KB, and will return to being \fBenabled\fR once all
488 filesystems that have ever had their recordsize larger than 128KB are destroyed.
489 .RE
491 .sp
492 .ne 2
493 .na
494 \fB\fBsha512\fR\fR
495 .ad
496 .RS 4n
497 .TS
498 l l .
499 GUID org.illumos:sha512
500 READ\-ONLY COMPATIBLE no
501 DEPENDENCIES extensible_dataset
502 .TE
504 This feature enables the use of the SHA-512/256 truncated hash algorithm
505 (FIPS 180-4) for checksum and dedup. The native 64-bit arithmetic of
506 SHA-512 provides an approximate 50% performance boost over SHA-256 on
507 64-bit hardware and is thus a good minimum-change replacement candidate
508 for systems where hash performance is important, but these systems
509 cannot for whatever reason utilize the faster \fBskein\fR and
510 \fBedonr\fR algorithms.
512 When the \fBsha512\fR feature is set to \fBenabled\fR, the administrator
513 can turn on the \fBsha512\fR checksum on any dataset using the
514 \fBzfs set checksum=sha512\fR command. This feature becomes
515 \fBactive\fR once a \fBchecksum\fR property has been set to \fBsha512\fR,
516 and will return to being \fBenabled\fR once all filesystems that have
517 ever had their checksum set to \fBsha512\fR are destroyed.
519 Booting off of pools utilizing SHA-512/256 is supported (provided that
520 the updated GRUB stage2 module is installed).
522 .RE
524 .sp
525 .ne 2
526 .na
527 \fB\fBskein\fR\fR
528 .ad
529 .RS 4n
530 .TS
531 l l .
532 GUID org.illumos:skein
533 READ\-ONLY COMPATIBLE no
534 DEPENDENCIES extensible_dataset
535 .TE
537 This feature enables the use of the Skein hash algorithm for checksum
538 and dedup. Skein is a high-performance secure hash algorithm that was a
539 finalist in the NIST SHA-3 competition. It provides a very high security
540 margin and high performance on 64-bit hardware (80% faster than
541 SHA-256). This implementation also utilizes the new salted checksumming
542 functionality in ZFS, which means that the checksum is pre-seeded with a
543 secret 256-bit random key (stored on the pool) before being fed the data
544 block to be checksummed. Thus the produced checksums are unique to a
545 given pool, preventing hash collision attacks on systems with dedup.
547 When the \fBskein\fR feature is set to \fBenabled\fR, the administrator
548 can turn on the \fBskein\fR checksum on any dataset using the
549 \fBzfs set checksum=skein\fR command. This feature becomes
550 \fBactive\fR once a \fBchecksum\fR property has been set to \fBskein\fR,
551 and will return to being \fBenabled\fR once all filesystems that have
552 ever had their checksum set to \fBskein\fR are destroyed.
554 Booting off of pools using \fBskein\fR is \fBNOT\fR supported
555 -- any attempt to enable \fBskein\fR on a root pool will fail with an
556 error.
558 .RE
560 .sp
561 .ne 2
562 .na
563 \fB\fBedonr\fR\fR
564 .ad
565 .RS 4n
566 .TS
567 l l .
568 GUID org.illumos:edonr
569 READ\-ONLY COMPATIBLE no
570 DEPENDENCIES extensible_dataset
571 .TE
573 This feature enables the use of the Edon-R hash algorithm for checksum,
574 including for nopwrite (if compression is also enabled, an overwrite of
575 a block whose checksum matches the data being written will be ignored).
576 In an abundance of caution, Edon-R can not be used with dedup
577 (without verification).
579 Edon-R is a very high-performance hash algorithm that was part
580 of the NIST SHA-3 competition. It provides extremely high hash
581 performance (over 350% faster than SHA-256), but was not selected
582 because of its unsuitability as a general purpose secure hash algorithm.
583 This implementation utilizes the new salted checksumming functionality
584 in ZFS, which means that the checksum is pre-seeded with a secret
585 256-bit random key (stored on the pool) before being fed the data block
586 to be checksummed. Thus the produced checksums are unique to a given
587 pool.
589 When the \fBedonr\fR feature is set to \fBenabled\fR, the administrator
590 can turn on the \fBedonr\fR checksum on any dataset using the
591 \fBzfs set checksum=edonr\fR command. This feature becomes
592 \fBactive\fR once a \fBchecksum\fR property has been set to \fBedonr\fR,
593 and will return to being \fBenabled\fR once all filesystems that have
594 ever had their checksum set to \fBedonr\fR are destroyed.
596 Booting off of pools using \fBedonr\fR is \fBNOT\fR supported
597 -- any attempt to enable \fBedonr\fR on a root pool will fail with an
598 error.
600 .SH "SEE ALSO"
601 \fBzfs\fR(1M), \fBzpool\fR(1M)