Use the LZ4 algorithm to compress metadata when the corresponding feature is enabled

          --- old/usr/src/man/man5/zpool-features.5
          +++ new/usr/src/man/man5/zpool-features.5

   1    1  '\" te
   2    2  .\" Copyright (c) 2013 by Delphix. All rights reserved.
   3    3  .\" Copyright (c) 2013 by Saso Kiselkov. All rights reserved.
   4    4  .\" Copyright (c) 2014, Joyent, Inc. All rights reserved.
   5    5  .\" The contents of this file are subject to the terms of the Common Development
   6    6  .\" and Distribution License (the "License").  You may not use this file except
   7    7  .\" in compliance with the License. You can obtain a copy of the license at
   8    8  .\" usr/src/OPENSOLARIS.LICENSE or http://www.opensolaris.org/os/licensing.
   9    9  .\"
  10   10  .\" See the License for the specific language governing permissions and
  11   11  .\" limitations under the License. When distributing Covered Code, include this
  12   12  .\" CDDL HEADER in each file and include the License file at
  13   13  .\" usr/src/OPENSOLARIS.LICENSE.  If applicable, add the following below this
  14   14  .\" CDDL HEADER, with the fields enclosed by brackets "[]" replaced with your
  15   15  .\" own identifying information:
  16   16  .\" Portions Copyright [yyyy] [name of copyright owner]
  17   17  .TH ZPOOL-FEATURES 5 "Aug 27, 2013"
  18   18  .SH NAME
  19   19  zpool\-features \- ZFS pool feature descriptions
  20   20  .SH DESCRIPTION
  21   21  .sp
  22   22  .LP
  23   23  ZFS pool on\-disk format versions are specified via "features" which replace
  24   24  the old on\-disk format numbers (the last supported on\-disk format number is
  25   25  28). To enable a feature on a pool use the \fBupgrade\fR subcommand of the
  26   26  \fBzpool\fR(1M) command, or set the \fBfeature@\fR\fIfeature_name\fR property
  27   27  to \fBenabled\fR.
  28   28  .sp
  29   29  .LP
  30   30  The pool format does not affect file system version compatibility or the ability
  31   31  to send file systems between pools.
  32   32  .sp
  33   33  .LP
  34   34  Since most features can be enabled independently of each other the on\-disk
  35   35  format of the pool is specified by the set of all features marked as
  36   36  \fBactive\fR on the pool. If the pool was created by another software version
  37   37  this set may include unsupported features.
  38   38  .SS "Identifying features"
  39   39  .sp
  40   40  .LP
  41   41  Every feature has a guid of the form \fIcom.example:feature_name\fR. The reverse
  42   42  DNS name ensures that the feature's guid is unique across all ZFS
  43   43  implementations. When unsupported features are encountered on a pool they will
  44   44  be identified by their guids. Refer to the documentation for the ZFS
  45   45  implementation that created the pool for information about those features.
  46   46  .sp
  47   47  .LP
  48   48  Each supported feature also has a short name. By convention a feature's short
  49   49  name is the portion of its guid which follows the ':' (e.g.
  50   50  \fIcom.example:feature_name\fR would have the short name \fIfeature_name\fR),
  51   51  however a feature's short name may differ across ZFS implementations if
  52   52  following the convention would result in name conflicts.
  53   53  .SS "Feature states"
  54   54  .sp
  55   55  .LP
  56   56  Features can be in one of three states:
  57   57  .sp
  58   58  .ne 2
  59   59  .na
  60   60  \fB\fBactive\fR\fR
  61   61  .ad
  62   62  .RS 12n
  63   63  This feature's on\-disk format changes are in effect on the pool. Support for
  64   64  this feature is required to import the pool in read\-write mode. If this
  65   65  feature is not read-only compatible, support is also required to import the pool
  66   66  in read\-only mode (see "Read\-only compatibility").
  67   67  .RE
  68   68  
  69   69  .sp
  70   70  .ne 2
  71   71  .na
  72   72  \fB\fBenabled\fR\fR
  73   73  .ad
  74   74  .RS 12n
  75   75  An administrator has marked this feature as enabled on the pool, but the
  76   76  feature's on\-disk format changes have not been made yet. The pool can still be
  77   77  imported by software that does not support this feature, but changes may be made
  78   78  to the on\-disk format at any time which will move the feature to the
  79   79  \fBactive\fR state. Some features may support returning to the \fBenabled\fR
  80   80  state after becoming \fBactive\fR. See feature\-specific documentation for
  81   81  details.
  82   82  .RE
  83   83  
  84   84  .sp
  85   85  .ne 2
  86   86  .na
  87   87  \fBdisabled\fR
  88   88  .ad
  89   89  .RS 12n
  90   90  This feature's on\-disk format changes have not been made and will not be made
  91   91  unless an administrator moves the feature to the \fBenabled\fR state. Features
  92   92  cannot be disabled once they have been enabled.
  93   93  .RE
  94   94  
  95   95  .sp
  96   96  .LP
  97   97  The state of supported features is exposed through pool properties of the form
  98   98  \fIfeature@short_name\fR.
  99   99  .SS "Read\-only compatibility"
 100  100  .sp
 101  101  .LP
 102  102  Some features may make on\-disk format changes that do not interfere with other
 103  103  software's ability to read from the pool. These features are referred to as
 104  104  "read\-only compatible". If all unsupported features on a pool are read\-only
 105  105  compatible, the pool can be imported in read\-only mode by setting the
 106  106  \fBreadonly\fR property during import (see \fBzpool\fR(1M) for details on
 107  107  importing pools).
 108  108  .SS "Unsupported features"
 109  109  .sp
 110  110  .LP
 111  111  For each unsupported feature enabled on an imported pool a pool property
 112  112  named \fIunsupported@feature_guid\fR will indicate why the import was allowed
 113  113  despite the unsupported feature. Possible values for this property are:
 114  114  
 115  115  .sp
 116  116  .ne 2
 117  117  .na
 118  118  \fB\fBinactive\fR\fR
 119  119  .ad
 120  120  .RS 12n
 121  121  The feature is in the \fBenabled\fR state and therefore the pool's on\-disk
 122  122  format is still compatible with software that does not support this feature.
 123  123  .RE
 124  124  
 125  125  .sp
 126  126  .ne 2
 127  127  .na
 128  128  \fB\fBreadonly\fR\fR
 129  129  .ad
 130  130  .RS 12n
 131  131  The feature is read\-only compatible and the pool has been imported in
 132  132  read\-only mode.
 133  133  .RE
 134  134  
 135  135  .SS "Feature dependencies"
 136  136  .sp
 137  137  .LP
 138  138  Some features depend on other features being enabled in order to function
 139  139  properly. Enabling a feature will automatically enable any features it
 140  140  depends on.
 141  141  .SH FEATURES
 142  142  .sp
 143  143  .LP
 144  144  The following features are supported on this system:
 145  145  .sp
 146  146  .ne 2
 147  147  .na
 148  148  \fB\fBasync_destroy\fR\fR
 149  149  .ad
 150  150  .RS 4n
 151  151  .TS
 152  152  l l .
 153  153  GUID    com.delphix:async_destroy
 154  154  READ\-ONLY COMPATIBLE   yes
 155  155  DEPENDENCIES    none
 156  156  .TE
 157  157  
 158  158  Destroying a file system requires traversing all of its data in order to
 159  159  return its used space to the pool. Without \fBasync_destroy\fR the file system
 160  160  is not fully removed until all space has been reclaimed. If the destroy
 161  161  operation is interrupted by a reboot or power outage the next attempt to open
 162  162  the pool will need to complete the destroy operation synchronously.
 163  163  
 164  164  When \fBasync_destroy\fR is enabled the file system's data will be reclaimed
 165  165  by a background process, allowing the destroy operation to complete without
 166  166  traversing the entire file system. The background process is able to resume
 167  167  interrupted destroys after the pool has been opened, eliminating the need
 168  168  to finish interrupted destroys as part of the open operation. The amount
 169  169  of space remaining to be reclaimed by the background process is available
 170  170  through the \fBfreeing\fR property.
 171  171  
 172  172  This feature is only \fBactive\fR while \fBfreeing\fR is non\-zero.
 173  173  .RE
 174  174  
 175  175  .sp
 176  176  .ne 2
 177  177  .na
 178  178  \fB\fBempty_bpobj\fR\fR
 179  179  .ad
 180  180  .RS 4n
 181  181  .TS
 182  182  l l .
 183  183  GUID    com.delphix:empty_bpobj
 184  184  READ\-ONLY COMPATIBLE   yes
 185  185  DEPENDENCIES    none
 186  186  .TE
 187  187  
 188  188  This feature increases the performance of creating and using a large
 189  189  number of snapshots of a single filesystem or volume, and also reduces
 190  190  the disk space required.
 191  191  
 192  192  When there are many snapshots, each snapshot uses many Block Pointer
 193  193  Objects (bpobj's) to track blocks associated with that snapshot.
 194  194  However, in common use cases, most of these bpobj's are empty.  This
 195  195  feature allows us to create each bpobj on-demand, thus eliminating the
 196  196  empty bpobjs.
 197  197  
 198  198  This feature is \fBactive\fR while there are any filesystems, volumes,
 199  199  or snapshots which were created after enabling this feature.
 200  200  .RE
 201  201  
 202  202  .sp
 203  203  .ne 2
 204  204  .na
 205  205  \fB\fBfilesystem_limits\fR\fR
 206  206  .ad
 207  207  .RS 4n
 208  208  .TS
 209  209  l l .
 210  210  GUID    com.joyent:filesystem_limits
 211  211  READ\-ONLY COMPATIBLE   yes
 212  212  DEPENDENCIES    extensible_dataset
 213  213  .TE
 214  214  
 215  215  This feature enables filesystem and snapshot limits. These limits can be used
 216  216  to control how many filesystems and/or snapshots can be created at the point in
 217  217  the tree on which the limits are set.
 218  218  
 219  219  This feature is \fBactive\fR once either of the limit properties has been
 220  220  set on a dataset. Once activated the feature is never deactivated.
 221  221  .RE
 222  222  
 223  223  .sp
 224  224  .ne 2
 225  225  .na
 226  226  \fB\fBlz4_compress\fR\fR
 227  227  .ad
 228  228  .RS 4n
 229  229  .TS
 230  230  l l .
 231  231  GUID    org.illumos:lz4_compress
 232  232  READ\-ONLY COMPATIBLE   no
 233  233  DEPENDENCIES    none
 234  234  .TE
 235  235  

 236  236  \fBlz4\fR is a high-performance real-time compression algorithm that
 237  237  features significantly faster compression and decompression as well as a
 238  238  higher compression ratio than the older \fBlzjb\fR compression.
 239  239  Typically, \fBlz4\fR compression is approximately 50% faster on
 240  240  compressible data and 200% faster on incompressible data than
 241  241  \fBlzjb\fR. It is also approximately 80% faster on decompression, while
 242  242  giving approximately 10% better compression ratio.
 243  243  
 244  244  When the \fBlz4_compress\fR feature is set to \fBenabled\fR, the
 245  245  administrator can turn on \fBlz4\fR compression on any dataset on the
 246      -pool using the \fBzfs\fR(1M) command. Please note that doing so will
 247      -immediately activate the \fBlz4_compress\fR feature on the underlying
 248      -pool (even before any data is written), and the feature will not be
 249      -deactivated. Since this feature is not read-only compatible, this
 250      -operation will render the pool unimportable on systems without support
 251      -for the \fBlz4_compress\fR feature. Booting off of \fBlz4\fR-compressed
 252      -root pools is supported.
      246 +pool using the \fBzfs\fR(1M) command. Also, all newly written metadata
      247 +will be compressed with \fBlz4\fR algorithm. Since this feature is not
      248 +read-only compatible, this operation will render the pool unimportable
      249 +on systems without support for the \fBlz4_compress\fR feature. Booting
      250 +off of \fBlz4\fR-compressed root pools is supported.
      251 +
      252 +This feature becomes \fBactive\fR as soon as it is enabled and will
      253 +never return to being \fBenabled\fB.
 253  254  .RE
 254  255  
 255  256  .sp
 256  257  .ne 2
 257  258  .na
 258  259  \fB\fBspacemap_histogram\fR\fR
 259  260  .ad
 260  261  .RS 4n
 261  262  .TS
 262  263  l l .

 263  264  GUID    com.delphix:spacemap_histogram
 264  265  READ\-ONLY COMPATIBLE   yes
 265  266  DEPENDENCIES    none
 266  267  .TE
 267  268  
 268  269  This features allows ZFS to maintain more information about how free space
 269  270  is organized within the pool. If this feature is \fBenabled\fR, ZFS will
 270  271  set this feature to \fBactive\fR when a new space map object is created or
 271  272  an existing space map is upgraded to the new format. Once the feature is
 272  273  \fBactive\fR, it will remain in that state until the pool is destroyed.
 273  274  .RE
 274  275  
 275  276  .sp
 276  277  .ne 2
 277  278  .na
 278  279  \fB\fBmulti_vdev_crash_dump\fR\fR
 279  280  .ad
 280  281  .RS 4n
 281  282  .TS
 282  283  l l .
 283  284  GUID    com.joyent:multi_vdev_crash_dump
 284  285  READ\-ONLY COMPATIBLE   no
 285  286  DEPENDENCIES    none
 286  287  .TE
 287  288  
 288  289  This feature allows a dump device to be configured with a pool comprised
 289  290  of multiple vdevs.  Those vdevs may be arranged in any mirrored or raidz
 290  291  configuration.
 291  292  
 292  293  When the \fBmulti_vdev_crash_dump\fR feature is set to \fBenabled\fR,
 293  294  the administrator can use the \fBdumpadm\fR(1M) command to configure a
 294  295  dump device on a pool comprised of multiple vdevs.
 295  296  .RE
 296  297  
 297  298  .sp
 298  299  .ne 2
 299  300  .na
 300  301  \fB\fBextensible_dataset\fR\fR
 301  302  .ad
 302  303  .RS 4n
 303  304  .TS
 304  305  l l .
 305  306  GUID    com.delphix:extensible_dataset
 306  307  READ\-ONLY COMPATIBLE   no
 307  308  DEPENDENCIES    none
 308  309  .TE
 309  310  
 310  311  This feature allows more flexible use of internal ZFS data structures,
 311  312  and exists for other features to depend on.
 312  313  
 313  314  This feature will be \fBactive\fR when the first dependent feature uses it,
 314  315  and will be returned to the \fBenabled\fR state when all datasets that use
 315  316  this feature are destroyed.
 316  317  
 317  318  .RE
 318  319  
 319  320  .sp
 320  321  .ne 2
 321  322  .na
 322  323  \fB\fBbookmarks\fR\fR
 323  324  .ad
 324  325  .RS 4n
 325  326  .TS
 326  327  l l .
 327  328  GUID    com.delphix:bookmarks
 328  329  READ\-ONLY COMPATIBLE   yes
 329  330  DEPENDENCIES    extensible_dataset
 330  331  .TE
 331  332  
 332  333  This feature enables use of the \fBzfs bookmark\fR subcommand.
 333  334  
 334  335  This feature is \fBactive\fR while any bookmarks exist in the pool.
 335  336  All bookmarks in the pool can be listed by running
 336  337  \fBzfs list -t bookmark -r \fIpoolname\fR\fR.
 337  338  
 338  339  .RE
 339  340  
 340  341  .sp
 341  342  .ne 2
 342  343  .na
 343  344  \fB\fBenabled_txg\fR\fR
 344  345  .ad
 345  346  .RS 4n
 346  347  .TS
 347  348  l l .
 348  349  GUID    com.delphix:enabled_txg
 349  350  READ\-ONLY COMPATIBLE   yes
 350  351  DEPENDENCIES    none
 351  352  .TE
 352  353  
 353  354  Once this feature is enabled ZFS records the transaction group number
 354  355  in which new features are enabled. This has no user-visible impact,
 355  356  but other features may depend on this feature.
 356  357  
 357  358  This feature becomes \fBactive\fR as soon as it is enabled and will
 358  359  never return to being \fBenabled\fB.
 359  360  
 360  361  .RE
 361  362  
 362  363  .sp
 363  364  .ne 2
 364  365  .na
 365  366  \fB\fBhole_birth\fR\fR
 366  367  .ad
 367  368  .RS 4n
 368  369  .TS
 369  370  l l .
 370  371  GUID    com.delphix:hole_birth
 371  372  READ\-ONLY COMPATIBLE   no
 372  373  DEPENDENCIES    enabled_txg
 373  374  .TE
 374  375  
 375  376  This feature improves performance of incremental sends ("zfs send -i")
 376  377  and receives for objects with many holes. The most common case of
 377  378  hole-filled objects is zvols.
 378  379  
 379  380  An incremental send stream from snapshot \fBA\fR to snapshot \fBB\fR
 380  381  contains information about every block that changed between \fBA\fR and
 381  382  \fBB\fR. Blocks which did not change between those snapshots can be
 382  383  identified and omitted from the stream using a piece of metadata called
 383  384  the 'block birth time', but birth times are not recorded for holes (blocks
 384  385  filled only with zeroes). Since holes created after \fBA\fR cannot be
 385  386  distinguished from holes created before \fBA\fR, information about every
 386  387  hole in the entire filesystem or zvol is included in the send stream.
 387  388  
 388  389  For workloads where holes are rare this is not a problem. However, when
 389  390  incrementally replicating filesystems or zvols with many holes (for
 390  391  example a zvol formatted with another filesystem) a lot of time will
 391  392  be spent sending and receiving unnecessary information about holes that
 392  393  already exist on the receiving side.
 393  394  
 394  395  Once the \fBhole_birth\fR feature has been enabled the block birth times
 395  396  of all new holes will be recorded. Incremental sends between snapshots
 396  397  created after this feature is enabled will use this new metadata to avoid
 397  398  sending information about holes that already exist on the receiving side.
 398  399  
 399  400  This feature becomes \fBactive\fR as soon as it is enabled and will
 400  401  never return to being \fBenabled\fB.
 401  402  
 402  403  .RE
 403  404  
 404  405  
 405  406  .SH "SEE ALSO"
 406  407  \fBzpool\fR(1M)

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX