2 .\" Copyright (c) 2013 by Turbo Fredriksson <turbo@bayour.com>. All rights reserved.
3 .\" The contents of this file are subject to the terms of the Common Development
4 .\" and Distribution License (the "License"). You may not use this file except
5 .\" in compliance with the License. You can obtain a copy of the license at
6 .\" usr/src/OPENSOLARIS.LICENSE or http://www.opensolaris.org/os/licensing.
8 .\" See the License for the specific language governing permissions and
9 .\" limitations under the License. When distributing Covered Code, include this
10 .\" CDDL HEADER in each file and include the License file at
11 .\" usr/src/OPENSOLARIS.LICENSE. If applicable, add the following below this
12 .\" CDDL HEADER, with the fields enclosed by brackets "[]" replaced with your
13 .\" own identifying information:
14 .\" Portions Copyright [yyyy] [name of copyright owner]
15 .TH ZFS-MODULE-PARAMETERS 5 "Nov 16, 2013"
17 zfs\-module\-parameters \- ZFS module parameters
21 Description of the different parameters to the ZFS module.
23 .SS "Module parameters"
30 \fBignore_hole_birth\fR (int)
33 When set, the hole_birth optimization will not be used, and all holes will
34 always be sent on zfs send. Useful if you suspect your datasets are affected
35 by a bug in hole_birth.
37 Use \fB1\fR for on (default) and \fB0\fR for off.
43 \fBl2arc_feed_again\fR (int)
46 Turbo L2ARC warm-up. When the L2ARC is cold the fill interval will be set as
49 Use \fB1\fR for yes (default) and \fB0\fR to disable.
55 \fBl2arc_feed_min_ms\fR (ulong)
58 Min feed interval in milliseconds. Requires \fBl2arc_feed_again=1\fR and only
59 applicable in related situations.
61 Default value: \fB200\fR.
67 \fBl2arc_feed_secs\fR (ulong)
70 Seconds between L2ARC writing
72 Default value: \fB1\fR.
78 \fBl2arc_headroom\fR (ulong)
81 How far through the ARC lists to search for L2ARC cacheable content, expressed
82 as a multiplier of \fBl2arc_write_max\fR
84 Default value: \fB2\fR.
90 \fBl2arc_headroom_boost\fR (ulong)
93 Scales \fBl2arc_headroom\fR by this percentage when L2ARC contents are being
94 successfully compressed before writing. A value of 100 disables this feature.
96 Default value: \fB200\fR.
102 \fBl2arc_nocompress\fR (int)
105 Skip compressing L2ARC buffers
107 Use \fB1\fR for yes and \fB0\fR for no (default).
113 \fBl2arc_noprefetch\fR (int)
116 Do not write buffers to L2ARC if they were prefetched but not used by
119 Use \fB1\fR for yes (default) and \fB0\fR to disable.
125 \fBl2arc_norw\fR (int)
128 No reads during writes
130 Use \fB1\fR for yes and \fB0\fR for no (default).
136 \fBl2arc_write_boost\fR (ulong)
139 Cold L2ARC devices will have \fBl2arc_write_nax\fR increased by this amount
140 while they remain cold.
142 Default value: \fB8,388,608\fR.
148 \fBl2arc_write_max\fR (ulong)
151 Max write bytes per interval
153 Default value: \fB8,388,608\fR.
159 \fBmetaslab_aliquot\fR (ulong)
162 Metaslab granularity, in bytes. This is roughly similar to what would be
163 referred to as the "stripe size" in traditional RAID arrays. In normal
164 operation, ZFS will try to write this amount of data to a top-level vdev
165 before moving on to the next one.
167 Default value: \fB524,288\fR.
173 \fBmetaslab_bias_enabled\fR (int)
176 Enable metaslab group biasing based on its vdev's over- or under-utilization
177 relative to the pool.
179 Use \fB1\fR for yes (default) and \fB0\fR for no.
185 \fBzfs_metaslab_segment_weight_enabled\fR (int)
188 Enable/disable segment-based metaslab selection.
190 Use \fB1\fR for yes (default) and \fB0\fR for no.
196 \fBzfs_metaslab_switch_threshold\fR (int)
199 When using segment-based metaslab selection, continue allocating
200 from the active metaslab until \fBlzfs_metaslab_switch_threshold\fR
201 worth of buckets have been exhausted.
203 Default value: \fB2\fR.
209 \fBmetaslab_debug_load\fR (int)
212 Load all metaslabs during pool import.
214 Use \fB1\fR for yes and \fB0\fR for no (default).
220 \fBmetaslab_debug_unload\fR (int)
223 Prevent metaslabs from being unloaded.
225 Use \fB1\fR for yes and \fB0\fR for no (default).
231 \fBmetaslab_fragmentation_factor_enabled\fR (int)
234 Enable use of the fragmentation metric in computing metaslab weights.
236 Use \fB1\fR for yes (default) and \fB0\fR for no.
242 \fBmetaslabs_per_vdev\fR (int)
245 When a vdev is added, it will be divided into approximately (but no more than) this number of metaslabs.
247 Default value: \fB200\fR.
253 \fBmetaslab_preload_enabled\fR (int)
256 Enable metaslab group preloading.
258 Use \fB1\fR for yes (default) and \fB0\fR for no.
264 \fBmetaslab_lba_weighting_enabled\fR (int)
267 Give more weight to metaslabs with lower LBAs, assuming they have
268 greater bandwidth as is typically the case on a modern constant
269 angular velocity disk drive.
271 Use \fB1\fR for yes (default) and \fB0\fR for no.
277 \fBspa_config_path\fR (charp)
282 Default value: \fB/etc/zfs/zpool.cache\fR.
288 \fBspa_asize_inflation\fR (int)
291 Multiplication factor used to estimate actual disk consumption from the
292 size of data being written. The default value is a worst case estimate,
293 but lower values may be valid for a given pool depending on its
294 configuration. Pool administrators who understand the factors involved
295 may wish to specify a more realistic inflation factor, particularly if
296 they operate close to quota or capacity limits.
298 Default value: \fB24\fR.
304 \fBspa_load_verify_data\fR (int)
307 Whether to traverse data blocks during an "extreme rewind" (\fB-X\fR)
308 import. Use 0 to disable and 1 to enable.
310 An extreme rewind import normally performs a full traversal of all
311 blocks in the pool for verification. If this parameter is set to 0,
312 the traversal skips non-metadata blocks. It can be toggled once the
313 import has started to stop or start the traversal of non-metadata blocks.
315 Default value: \fB1\fR.
321 \fBspa_load_verify_metadata\fR (int)
324 Whether to traverse blocks during an "extreme rewind" (\fB-X\fR)
325 pool import. Use 0 to disable and 1 to enable.
327 An extreme rewind import normally performs a full traversal of all
328 blocks in the pool for verification. If this parameter is set to 0,
329 the traversal is not performed. It can be toggled once the import has
330 started to stop or start the traversal.
332 Default value: \fB1\fR.
338 \fBspa_load_verify_maxinflight\fR (int)
341 Maximum concurrent I/Os during the traversal performed during an "extreme
342 rewind" (\fB-X\fR) pool import.
344 Default value: \fB10000\fR.
350 \fBspa_slop_shift\fR (int)
353 Normally, we don't allow the last 3.2% (1/(2^spa_slop_shift)) of space
354 in the pool to be consumed. This ensures that we don't run the pool
355 completely out of space, due to unaccounted changes (e.g. to the MOS).
356 It also limits the worst-case time to allocate space. If we have
357 less than this amount of free space, most ZPL operations (e.g. write,
358 create) will return ENOSPC.
360 Default value: \fB5\fR.
366 \fBzfetch_array_rd_sz\fR (ulong)
369 If prefetching is enabled, disable prefetching for reads larger than this size.
371 Default value: \fB1,048,576\fR.
377 \fBzfetch_max_distance\fR (uint)
380 Max bytes to prefetch per stream (default 8MB).
382 Default value: \fB8,388,608\fR.
388 \fBzfetch_max_streams\fR (uint)
391 Max number of streams per zfetch (prefetch streams per file).
393 Default value: \fB8\fR.
399 \fBzfetch_min_sec_reap\fR (uint)
402 Min time before an active prefetch stream can be reclaimed
404 Default value: \fB2\fR.
410 \fBzfs_arc_dnode_limit\fR (ulong)
413 When the number of bytes consumed by dnodes in the ARC exceeds this number of
414 bytes, try to unpin some of it in response to demand for non-metadata. This
415 value acts as a floor to the amount of dnode metadata, and defaults to 0 which
416 indicates that a percent which is based on \fBzfs_arc_dnode_limit_percent\fR of
417 the ARC meta buffers that may be used for dnodes.
419 See also \fBzfs_arc_meta_prune\fR which serves a similar purpose but is used
420 when the amount of metadata in the ARC exceeds \fBzfs_arc_meta_limit\fR rather
421 than in response to overall demand for non-metadata.
424 Default value: \fB0\fR.
430 \fBzfs_arc_dnode_limit_percent\fR (ulong)
433 Percentage that can be consumed by dnodes of ARC meta buffers.
435 See also \fBzfs_arc_dnode_limit\fR which serves a similar purpose but has a
436 higher priority if set to nonzero value.
438 Default value: \fB10\fR.
444 \fBzfs_arc_dnode_reduce_percent\fR (ulong)
447 Percentage of ARC dnodes to try to scan in response to demand for non-metadata
448 when the number of bytes consumed by dnodes exceeds \fBzfs_arc_dnode_limit\fR.
451 Default value: \fB10% of the number of dnodes in the ARC\fR.
457 \fBzfs_arc_average_blocksize\fR (int)
460 The ARC's buffer hash table is sized based on the assumption of an average
461 block size of \fBzfs_arc_average_blocksize\fR (default 8K). This works out
462 to roughly 1MB of hash table per 1GB of physical memory with 8-byte pointers.
463 For configurations with a known larger average block size this value can be
464 increased to reduce the memory footprint.
467 Default value: \fB8192\fR.
473 \fBzfs_arc_evict_batch_limit\fR (int)
476 Number ARC headers to evict per sub-list before proceeding to another sub-list.
477 This batch-style operation prevents entire sub-lists from being evicted at once
478 but comes at a cost of additional unlocking and locking.
480 Default value: \fB10\fR.
486 \fBzfs_arc_grow_retry\fR (int)
489 After a memory pressure event the ARC will wait this many seconds before trying
492 Default value: \fB5\fR.
498 \fBzfs_arc_lotsfree_percent\fR (int)
501 Throttle I/O when free system memory drops below this percentage of total
502 system memory. Setting this value to 0 will disable the throttle.
504 Default value: \fB10\fR.
510 \fBzfs_arc_max\fR (ulong)
513 Max arc size of ARC in bytes. If set to 0 then it will consume 1/2 of system
514 RAM. This value must be at least 67108864 (64 megabytes).
516 This value can be changed dynamically with some caveats. It cannot be set back
517 to 0 while running and reducing it below the current ARC size will not cause
518 the ARC to shrink without memory pressure to induce shrinking.
520 Default value: \fB0\fR.
526 \fBzfs_arc_meta_limit\fR (ulong)
529 The maximum allowed size in bytes that meta data buffers are allowed to
530 consume in the ARC. When this limit is reached meta data buffers will
531 be reclaimed even if the overall arc_c_max has not been reached. This
532 value defaults to 0 which indicates that a percent which is based on
533 \fBzfs_arc_meta_limit_percent\fR of the ARC may be used for meta data.
535 This value my be changed dynamically except that it cannot be set back to 0
536 for a specific percent of the ARC; it must be set to an explicit value.
538 Default value: \fB0\fR.
544 \fBzfs_arc_meta_limit_percent\fR (ulong)
547 Percentage of ARC buffers that can be used for meta data.
549 See also \fBzfs_arc_meta_limit\fR which serves a similar purpose but has a
550 higher priority if set to nonzero value.
553 Default value: \fB75\fR.
559 \fBzfs_arc_meta_min\fR (ulong)
562 The minimum allowed size in bytes that meta data buffers may consume in
563 the ARC. This value defaults to 0 which disables a floor on the amount
564 of the ARC devoted meta data.
566 Default value: \fB0\fR.
572 \fBzfs_arc_meta_prune\fR (int)
575 The number of dentries and inodes to be scanned looking for entries
576 which can be dropped. This may be required when the ARC reaches the
577 \fBzfs_arc_meta_limit\fR because dentries and inodes can pin buffers
578 in the ARC. Increasing this value will cause to dentry and inode caches
579 to be pruned more aggressively. Setting this value to 0 will disable
580 pruning the inode and dentry caches.
582 Default value: \fB10,000\fR.
588 \fBzfs_arc_meta_adjust_restarts\fR (ulong)
591 The number of restart passes to make while scanning the ARC attempting
592 the free buffers in order to stay below the \fBzfs_arc_meta_limit\fR.
593 This value should not need to be tuned but is available to facilitate
594 performance analysis.
596 Default value: \fB4096\fR.
602 \fBzfs_arc_min\fR (ulong)
607 Default value: \fB100\fR.
613 \fBzfs_arc_min_prefetch_lifespan\fR (int)
616 Minimum time prefetched blocks are locked in the ARC, specified in jiffies.
617 A value of 0 will default to 1 second.
619 Default value: \fB0\fR.
625 \fBzfs_arc_num_sublists_per_state\fR (int)
628 To allow more fine-grained locking, each ARC state contains a series
629 of lists for both data and meta data objects. Locking is performed at
630 the level of these "sub-lists". This parameters controls the number of
631 sub-lists per ARC state.
633 Default value: \fB1\fR or the number of online CPUs, whichever is greater
639 \fBzfs_arc_overflow_shift\fR (int)
642 The ARC size is considered to be overflowing if it exceeds the current
643 ARC target size (arc_c) by a threshold determined by this parameter.
644 The threshold is calculated as a fraction of arc_c using the formula
645 "arc_c >> \fBzfs_arc_overflow_shift\fR".
647 The default value of 8 causes the ARC to be considered to be overflowing
648 if it exceeds the target size by 1/256th (0.3%) of the target size.
650 When the ARC is overflowing, new buffer allocations are stalled until
651 the reclaim thread catches up and the overflow condition no longer exists.
653 Default value: \fB8\fR.
660 \fBzfs_arc_p_min_shift\fR (int)
663 arc_c shift to calc min/max arc_p
665 Default value: \fB4\fR.
671 \fBzfs_arc_p_aggressive_disable\fR (int)
674 Disable aggressive arc_p growth
676 Use \fB1\fR for yes (default) and \fB0\fR to disable.
682 \fBzfs_arc_p_dampener_disable\fR (int)
685 Disable arc_p adapt dampener
687 Use \fB1\fR for yes (default) and \fB0\fR to disable.
693 \fBzfs_arc_shrink_shift\fR (int)
696 log2(fraction of arc to reclaim)
698 Default value: \fB5\fR.
704 \fBzfs_arc_sys_free\fR (ulong)
707 The target number of bytes the ARC should leave as free memory on the system.
708 Defaults to the larger of 1/64 of physical memory or 512K. Setting this
709 option to a non-zero value will override the default.
711 Default value: \fB0\fR.
717 \fBzfs_autoimport_disable\fR (int)
720 Disable pool import at module load by ignoring the cache file (typically \fB/etc/zfs/zpool.cache\fR).
722 Use \fB1\fR for yes (default) and \fB0\fR for no.
728 \fBzfs_dbgmsg_enable\fR (int)
731 Internally ZFS keeps a small log to facilitate debugging. By default the log
732 is disabled, to enable it set this option to 1. The contents of the log can
733 be accessed by reading the /proc/spl/kstat/zfs/dbgmsg file. Writing 0 to
734 this proc file clears the log.
736 Default value: \fB0\fR.
742 \fBzfs_dbgmsg_maxsize\fR (int)
745 The maximum size in bytes of the internal ZFS debug log.
747 Default value: \fB4M\fR.
753 \fBzfs_dbuf_state_index\fR (int)
756 This feature is currently unused. It is normally used for controlling what
757 reporting is available under /proc/spl/kstat/zfs.
759 Default value: \fB0\fR.
765 \fBzfs_deadman_enabled\fR (int)
768 Enable deadman timer. See description below.
770 Use \fB1\fR for yes (default) and \fB0\fR to disable.
776 \fBzfs_deadman_synctime_ms\fR (ulong)
779 Expiration time in milliseconds. This value has two meanings. First it is
780 used to determine when the spa_deadman() logic should fire. By default the
781 spa_deadman() will fire if spa_sync() has not completed in 1000 seconds.
782 Secondly, the value determines if an I/O is considered "hung". Any I/O that
783 has not completed in zfs_deadman_synctime_ms is considered "hung" resulting
784 in a zevent being logged.
786 Default value: \fB1,000,000\fR.
792 \fBzfs_dedup_prefetch\fR (int)
795 Enable prefetching dedup-ed blks
797 Use \fB1\fR for yes and \fB0\fR to disable (default).
803 \fBzfs_delay_min_dirty_percent\fR (int)
806 Start to delay each transaction once there is this amount of dirty data,
807 expressed as a percentage of \fBzfs_dirty_data_max\fR.
808 This value should be >= zfs_vdev_async_write_active_max_dirty_percent.
809 See the section "ZFS TRANSACTION DELAY".
811 Default value: \fB60\fR.
817 \fBzfs_delay_scale\fR (int)
820 This controls how quickly the transaction delay approaches infinity.
821 Larger values cause longer delays for a given amount of dirty data.
823 For the smoothest delay, this value should be about 1 billion divided
824 by the maximum number of operations per second. This will smoothly
825 handle between 10x and 1/10th this number.
827 See the section "ZFS TRANSACTION DELAY".
829 Note: \fBzfs_delay_scale\fR * \fBzfs_dirty_data_max\fR must be < 2^64.
831 Default value: \fB500,000\fR.
837 \fBzfs_delete_blocks\fR (ulong)
840 This is the used to define a large file for the purposes of delete. Files
841 containing more than \fBzfs_delete_blocks\fR will be deleted asynchronously
842 while smaller files are deleted synchronously. Decreasing this value will
843 reduce the time spent in an unlink(2) system call at the expense of a longer
844 delay before the freed space is available.
846 Default value: \fB20,480\fR.
852 \fBzfs_dirty_data_max\fR (int)
855 Determines the dirty space limit in bytes. Once this limit is exceeded, new
856 writes are halted until space frees up. This parameter takes precedence
857 over \fBzfs_dirty_data_max_percent\fR.
858 See the section "ZFS TRANSACTION DELAY".
860 Default value: 10 percent of all memory, capped at \fBzfs_dirty_data_max_max\fR.
866 \fBzfs_dirty_data_max_max\fR (int)
869 Maximum allowable value of \fBzfs_dirty_data_max\fR, expressed in bytes.
870 This limit is only enforced at module load time, and will be ignored if
871 \fBzfs_dirty_data_max\fR is later changed. This parameter takes
872 precedence over \fBzfs_dirty_data_max_max_percent\fR. See the section
873 "ZFS TRANSACTION DELAY".
875 Default value: 25% of physical RAM.
881 \fBzfs_dirty_data_max_max_percent\fR (int)
884 Maximum allowable value of \fBzfs_dirty_data_max\fR, expressed as a
885 percentage of physical RAM. This limit is only enforced at module load
886 time, and will be ignored if \fBzfs_dirty_data_max\fR is later changed.
887 The parameter \fBzfs_dirty_data_max_max\fR takes precedence over this
888 one. See the section "ZFS TRANSACTION DELAY".
890 Default value: \fB25\fR.
896 \fBzfs_dirty_data_max_percent\fR (int)
899 Determines the dirty space limit, expressed as a percentage of all
900 memory. Once this limit is exceeded, new writes are halted until space frees
901 up. The parameter \fBzfs_dirty_data_max\fR takes precedence over this
902 one. See the section "ZFS TRANSACTION DELAY".
904 Default value: 10%, subject to \fBzfs_dirty_data_max_max\fR.
910 \fBzfs_dirty_data_sync\fR (int)
913 Start syncing out a transaction group if there is at least this much dirty data.
915 Default value: \fB67,108,864\fR.
921 \fBzfs_fletcher_4_impl\fR (string)
924 Select a fletcher 4 implementation.
926 Supported selectors are: \fBfastest\fR, \fBscalar\fR, \fBsse2\fR, \fBssse3\fR,
927 \fBavx2\fR, \fBavx512f\fR, and \fBaarch64_neon\fR.
928 All of the selectors except \fBfastest\fR and \fBscalar\fR require instruction
929 set extensions to be available and will only appear if ZFS detects that they are
930 present at runtime. If multiple implementations of fletcher 4 are available,
931 the \fBfastest\fR will be chosen using a micro benchmark. Selecting \fBscalar\fR
932 results in the original, CPU based calculation, being used. Selecting any option
933 other than \fBfastest\fR and \fBscalar\fR results in vector instructions from
934 the respective CPU instruction set being used.
936 Default value: \fBfastest\fR.
942 \fBzfs_free_bpobj_enabled\fR (int)
945 Enable/disable the processing of the free_bpobj object.
947 Default value: \fB1\fR.
953 \fBzfs_free_max_blocks\fR (ulong)
956 Maximum number of blocks freed in a single txg.
958 Default value: \fB100,000\fR.
964 \fBzfs_vdev_async_read_max_active\fR (int)
967 Maximum asynchronous read I/Os active to each device.
968 See the section "ZFS I/O SCHEDULER".
970 Default value: \fB3\fR.
976 \fBzfs_vdev_async_read_min_active\fR (int)
979 Minimum asynchronous read I/Os active to each device.
980 See the section "ZFS I/O SCHEDULER".
982 Default value: \fB1\fR.
988 \fBzfs_vdev_async_write_active_max_dirty_percent\fR (int)
991 When the pool has more than
992 \fBzfs_vdev_async_write_active_max_dirty_percent\fR dirty data, use
993 \fBzfs_vdev_async_write_max_active\fR to limit active async writes. If
994 the dirty data is between min and max, the active I/O limit is linearly
995 interpolated. See the section "ZFS I/O SCHEDULER".
997 Default value: \fB60\fR.
1003 \fBzfs_vdev_async_write_active_min_dirty_percent\fR (int)
1006 When the pool has less than
1007 \fBzfs_vdev_async_write_active_min_dirty_percent\fR dirty data, use
1008 \fBzfs_vdev_async_write_min_active\fR to limit active async writes. If
1009 the dirty data is between min and max, the active I/O limit is linearly
1010 interpolated. See the section "ZFS I/O SCHEDULER".
1012 Default value: \fB30\fR.
1018 \fBzfs_vdev_async_write_max_active\fR (int)
1021 Maximum asynchronous write I/Os active to each device.
1022 See the section "ZFS I/O SCHEDULER".
1024 Default value: \fB10\fR.
1030 \fBzfs_vdev_async_write_min_active\fR (int)
1033 Minimum asynchronous write I/Os active to each device.
1034 See the section "ZFS I/O SCHEDULER".
1036 Default value: \fB1\fR.
1042 \fBzfs_vdev_max_active\fR (int)
1045 The maximum number of I/Os active to each device. Ideally, this will be >=
1046 the sum of each queue's max_active. It must be at least the sum of each
1047 queue's min_active. See the section "ZFS I/O SCHEDULER".
1049 Default value: \fB1,000\fR.
1055 \fBzfs_vdev_scrub_max_active\fR (int)
1058 Maximum scrub I/Os active to each device.
1059 See the section "ZFS I/O SCHEDULER".
1061 Default value: \fB2\fR.
1067 \fBzfs_vdev_scrub_min_active\fR (int)
1070 Minimum scrub I/Os active to each device.
1071 See the section "ZFS I/O SCHEDULER".
1073 Default value: \fB1\fR.
1079 \fBzfs_vdev_sync_read_max_active\fR (int)
1082 Maximum synchronous read I/Os active to each device.
1083 See the section "ZFS I/O SCHEDULER".
1085 Default value: \fB10\fR.
1091 \fBzfs_vdev_sync_read_min_active\fR (int)
1094 Minimum synchronous read I/Os active to each device.
1095 See the section "ZFS I/O SCHEDULER".
1097 Default value: \fB10\fR.
1103 \fBzfs_vdev_sync_write_max_active\fR (int)
1106 Maximum synchronous write I/Os active to each device.
1107 See the section "ZFS I/O SCHEDULER".
1109 Default value: \fB10\fR.
1115 \fBzfs_vdev_sync_write_min_active\fR (int)
1118 Minimum synchronous write I/Os active to each device.
1119 See the section "ZFS I/O SCHEDULER".
1121 Default value: \fB10\fR.
1127 \fBzfs_vdev_queue_depth_pct\fR (int)
1130 The queue depth percentage for each top-level virtual device.
1131 Used in conjunction with zfs_vdev_async_max_active.
1133 Default value: \fB1000\fR.
1139 \fBzfs_disable_dup_eviction\fR (int)
1142 Disable duplicate buffer eviction
1144 Use \fB1\fR for yes and \fB0\fR for no (default).
1150 \fBzfs_expire_snapshot\fR (int)
1153 Seconds to expire .zfs/snapshot
1155 Default value: \fB300\fR.
1161 \fBzfs_admin_snapshot\fR (int)
1164 Allow the creation, removal, or renaming of entries in the .zfs/snapshot
1165 directory to cause the creation, destruction, or renaming of snapshots.
1166 When enabled this functionality works both locally and over NFS exports
1167 which have the 'no_root_squash' option set. This functionality is disabled
1170 Use \fB1\fR for yes and \fB0\fR for no (default).
1176 \fBzfs_flags\fR (int)
1179 Set additional debugging flags. The following flags may be bitwise-or'd
1191 Enable dprintf entries in the debug log.
1193 2 ZFS_DEBUG_DBUF_VERIFY *
1194 Enable extra dbuf verifications.
1196 4 ZFS_DEBUG_DNODE_VERIFY *
1197 Enable extra dnode verifications.
1199 8 ZFS_DEBUG_SNAPNAMES
1200 Enable snapshot name verification.
1203 Check for illegally modified ARC buffers.
1206 Enable spa_dbgmsg entries in the debug log.
1208 64 ZFS_DEBUG_ZIO_FREE
1209 Enable verification of block frees.
1211 128 ZFS_DEBUG_HISTOGRAM_VERIFY
1212 Enable extra spacemap histogram verifications.
1215 * Requires debug build.
1217 Default value: \fB0\fR.
1223 \fBzfs_free_leak_on_eio\fR (int)
1226 If destroy encounters an EIO while reading metadata (e.g. indirect
1227 blocks), space referenced by the missing metadata can not be freed.
1228 Normally this causes the background destroy to become "stalled", as
1229 it is unable to make forward progress. While in this stalled state,
1230 all remaining space to free from the error-encountering filesystem is
1231 "temporarily leaked". Set this flag to cause it to ignore the EIO,
1232 permanently leak the space from indirect blocks that can not be read,
1233 and continue to free everything else that it can.
1235 The default, "stalling" behavior is useful if the storage partially
1236 fails (i.e. some but not all i/os fail), and then later recovers. In
1237 this case, we will be able to continue pool operations while it is
1238 partially failed, and when it recovers, we can continue to free the
1239 space, with no leaks. However, note that this case is actually
1242 Typically pools either (a) fail completely (but perhaps temporarily,
1243 e.g. a top-level vdev going offline), or (b) have localized,
1244 permanent errors (e.g. disk returns the wrong data due to bit flip or
1245 firmware bug). In case (a), this setting does not matter because the
1246 pool will be suspended and the sync thread will not be able to make
1247 forward progress regardless. In case (b), because the error is
1248 permanent, the best we can do is leak the minimum amount of space,
1249 which is what setting this flag will do. Therefore, it is reasonable
1250 for this flag to normally be set, but we chose the more conservative
1251 approach of not setting it, so that there is no possibility of
1252 leaking space in the "partial temporary" failure case.
1254 Default value: \fB0\fR.
1260 \fBzfs_free_min_time_ms\fR (int)
1263 During a \fBzfs destroy\fR operation using \fBfeature@async_destroy\fR a minimum
1264 of this much time will be spent working on freeing blocks per txg.
1266 Default value: \fB1,000\fR.
1272 \fBzfs_immediate_write_sz\fR (long)
1275 Largest data block to write to zil. Larger blocks will be treated as if the
1276 dataset being written to had the property setting \fBlogbias=throughput\fR.
1278 Default value: \fB32,768\fR.
1284 \fBzfs_max_recordsize\fR (int)
1287 We currently support block sizes from 512 bytes to 16MB. The benefits of
1288 larger blocks, and thus larger IO, need to be weighed against the cost of
1289 COWing a giant block to modify one byte. Additionally, very large blocks
1290 can have an impact on i/o latency, and also potentially on the memory
1291 allocator. Therefore, we do not allow the recordsize to be set larger than
1292 zfs_max_recordsize (default 1MB). Larger blocks can be created by changing
1293 this tunable, and pools with larger blocks can always be imported and used,
1294 regardless of this setting.
1296 Default value: \fB1,048,576\fR.
1302 \fBzfs_mdcomp_disable\fR (int)
1305 Disable meta data compression
1307 Use \fB1\fR for yes and \fB0\fR for no (default).
1313 \fBzfs_metaslab_fragmentation_threshold\fR (int)
1316 Allow metaslabs to keep their active state as long as their fragmentation
1317 percentage is less than or equal to this value. An active metaslab that
1318 exceeds this threshold will no longer keep its active status allowing
1319 better metaslabs to be selected.
1321 Default value: \fB70\fR.
1327 \fBzfs_mg_fragmentation_threshold\fR (int)
1330 Metaslab groups are considered eligible for allocations if their
1331 fragmentation metric (measured as a percentage) is less than or equal to
1332 this value. If a metaslab group exceeds this threshold then it will be
1333 skipped unless all metaslab groups within the metaslab class have also
1334 crossed this threshold.
1336 Default value: \fB85\fR.
1342 \fBzfs_mg_noalloc_threshold\fR (int)
1345 Defines a threshold at which metaslab groups should be eligible for
1346 allocations. The value is expressed as a percentage of free space
1347 beyond which a metaslab group is always eligible for allocations.
1348 If a metaslab group's free space is less than or equal to the
1349 threshold, the allocator will avoid allocating to that group
1350 unless all groups in the pool have reached the threshold. Once all
1351 groups have reached the threshold, all groups are allowed to accept
1352 allocations. The default value of 0 disables the feature and causes
1353 all metaslab groups to be eligible for allocations.
1355 This parameter allows to deal with pools having heavily imbalanced
1356 vdevs such as would be the case when a new vdev has been added.
1357 Setting the threshold to a non-zero percentage will stop allocations
1358 from being made to vdevs that aren't filled to the specified percentage
1359 and allow lesser filled vdevs to acquire more allocations than they
1360 otherwise would under the old \fBzfs_mg_alloc_failures\fR facility.
1362 Default value: \fB0\fR.
1368 \fBzfs_no_scrub_io\fR (int)
1371 Set for no scrub I/O. This results in scrubs not actually scrubbing data and
1372 simply doing a metadata crawl of the pool instead.
1374 Use \fB1\fR for yes and \fB0\fR for no (default).
1380 \fBzfs_no_scrub_prefetch\fR (int)
1383 Set to disable block prefetching for scrubs.
1385 Use \fB1\fR for yes and \fB0\fR for no (default).
1391 \fBzfs_nocacheflush\fR (int)
1394 Disable cache flush operations on disks when writing. Beware, this may cause
1395 corruption if disks re-order writes.
1397 Use \fB1\fR for yes and \fB0\fR for no (default).
1403 \fBzfs_nopwrite_enabled\fR (int)
1408 Use \fB1\fR for yes (default) and \fB0\fR to disable.
1414 \fBzfs_pd_bytes_max\fR (int)
1417 The number of bytes which should be prefetched during a pool traversal
1418 (eg: \fBzfs send\fR or other data crawling operations)
1420 Default value: \fB52,428,800\fR.
1426 \fBzfs_prefetch_disable\fR (int)
1429 This tunable disables predictive prefetch. Note that it leaves "prescient"
1430 prefetch (e.g. prefetch for zfs send) intact. Unlike predictive prefetch,
1431 prescient prefetch never issues i/os that end up not being needed, so it
1432 can't hurt performance.
1434 Use \fB1\fR for yes and \fB0\fR for no (default).
1440 \fBzfs_read_chunk_size\fR (long)
1443 Bytes to read per chunk
1445 Default value: \fB1,048,576\fR.
1451 \fBzfs_read_history\fR (int)
1454 Historic statistics for the last N reads will be available in
1455 \fR/proc/spl/kstat/zfs/POOLNAME/reads\fB
1457 Default value: \fB0\fR (no data is kept).
1463 \fBzfs_read_history_hits\fR (int)
1466 Include cache hits in read history
1468 Use \fB1\fR for yes and \fB0\fR for no (default).
1474 \fBzfs_recover\fR (int)
1477 Set to attempt to recover from fatal errors. This should only be used as a
1478 last resort, as it typically results in leaked space, or worse.
1480 Use \fB1\fR for yes and \fB0\fR for no (default).
1486 \fBzfs_resilver_delay\fR (int)
1489 Number of ticks to delay prior to issuing a resilver I/O operation when
1490 a non-resilver or non-scrub I/O operation has occurred within the past
1491 \fBzfs_scan_idle\fR ticks.
1493 Default value: \fB2\fR.
1499 \fBzfs_resilver_min_time_ms\fR (int)
1502 Resilvers are processed by the sync thread. While resilvering it will spend
1503 at least this much time working on a resilver between txg flushes.
1505 Default value: \fB3,000\fR.
1511 \fBzfs_scan_idle\fR (int)
1514 Idle window in clock ticks. During a scrub or a resilver, if
1515 a non-scrub or non-resilver I/O operation has occurred during this
1516 window, the next scrub or resilver operation is delayed by, respectively
1517 \fBzfs_scrub_delay\fR or \fBzfs_resilver_delay\fR ticks.
1519 Default value: \fB50\fR.
1525 \fBzfs_scan_min_time_ms\fR (int)
1528 Scrubs are processed by the sync thread. While scrubbing it will spend
1529 at least this much time working on a scrub between txg flushes.
1531 Default value: \fB1,000\fR.
1537 \fBzfs_scrub_delay\fR (int)
1540 Number of ticks to delay prior to issuing a scrub I/O operation when
1541 a non-scrub or non-resilver I/O operation has occurred within the past
1542 \fBzfs_scan_idle\fR ticks.
1544 Default value: \fB4\fR.
1550 \fBzfs_send_corrupt_data\fR (int)
1553 Allow sending of corrupt data (ignore read/checksum errors when sending data)
1555 Use \fB1\fR for yes and \fB0\fR for no (default).
1561 \fBzfs_sync_pass_deferred_free\fR (int)
1564 Flushing of data to disk is done in passes. Defer frees starting in this pass
1566 Default value: \fB2\fR.
1572 \fBzfs_sync_pass_dont_compress\fR (int)
1575 Don't compress starting in this pass
1577 Default value: \fB5\fR.
1583 \fBzfs_sync_pass_rewrite\fR (int)
1586 Rewrite new block pointers starting in this pass
1588 Default value: \fB2\fR.
1594 \fBzfs_top_maxinflight\fR (int)
1597 Max concurrent I/Os per top-level vdev (mirrors or raidz arrays) allowed during
1598 scrub or resilver operations.
1600 Default value: \fB32\fR.
1606 \fBzfs_txg_history\fR (int)
1609 Historic statistics for the last N txgs will be available in
1610 \fR/proc/spl/kstat/zfs/POOLNAME/txgs\fB
1612 Default value: \fB0\fR.
1618 \fBzfs_txg_timeout\fR (int)
1621 Flush dirty data to disk at least every N seconds (maximum txg duration)
1623 Default value: \fB5\fR.
1629 \fBzfs_vdev_aggregation_limit\fR (int)
1632 Max vdev I/O aggregation size
1634 Default value: \fB131,072\fR.
1640 \fBzfs_vdev_cache_bshift\fR (int)
1643 Shift size to inflate reads too
1645 Default value: \fB16\fR (effectively 65536).
1651 \fBzfs_vdev_cache_max\fR (int)
1654 Inflate reads small than this value to meet the \fBzfs_vdev_cache_bshift\fR
1657 Default value: \fB16384\fR.
1663 \fBzfs_vdev_cache_size\fR (int)
1666 Total size of the per-disk cache in bytes.
1668 Currently this feature is disabled as it has been found to not be helpful
1669 for performance and in some cases harmful.
1671 Default value: \fB0\fR.
1677 \fBzfs_vdev_mirror_rotating_inc\fR (int)
1680 A number by which the balancing algorithm increments the load calculation for
1681 the purpose of selecting the least busy mirror member when an I/O immediately
1682 follows its predecessor on rotational vdevs for the purpose of making decisions
1685 Default value: \fB0\fR.
1691 \fBzfs_vdev_mirror_rotating_seek_inc\fR (int)
1694 A number by which the balancing algorithm increments the load calculation for
1695 the purpose of selecting the least busy mirror member when an I/O lacks
1696 locality as defined by the zfs_vdev_mirror_rotating_seek_offset. I/Os within
1697 this that are not immediately following the previous I/O are incremented by
1700 Default value: \fB5\fR.
1706 \fBzfs_vdev_mirror_rotating_seek_offset\fR (int)
1709 The maximum distance for the last queued I/O in which the balancing algorithm
1710 considers an I/O to have locality.
1711 See the section "ZFS I/O SCHEDULER".
1713 Default value: \fB1048576\fR.
1719 \fBzfs_vdev_mirror_non_rotating_inc\fR (int)
1722 A number by which the balancing algorithm increments the load calculation for
1723 the purpose of selecting the least busy mirror member on non-rotational vdevs
1724 when I/Os do not immediately follow one another.
1726 Default value: \fB0\fR.
1732 \fBzfs_vdev_mirror_non_rotating_seek_inc\fR (int)
1735 A number by which the balancing algorithm increments the load calculation for
1736 the purpose of selecting the least busy mirror member when an I/O lacks
1737 locality as defined by the zfs_vdev_mirror_rotating_seek_offset. I/Os within
1738 this that are not immediately following the previous I/O are incremented by
1741 Default value: \fB1\fR.
1747 \fBzfs_vdev_read_gap_limit\fR (int)
1750 Aggregate read I/O operations if the gap on-disk between them is within this
1753 Default value: \fB32,768\fR.
1759 \fBzfs_vdev_scheduler\fR (charp)
1762 Set the Linux I/O scheduler on whole disk vdevs to this scheduler
1764 Default value: \fBnoop\fR.
1770 \fBzfs_vdev_write_gap_limit\fR (int)
1773 Aggregate write I/O over gap
1775 Default value: \fB4,096\fR.
1781 \fBzfs_vdev_raidz_impl\fR (string)
1784 Parameter for selecting raidz parity implementation to use.
1786 Options marked (always) below may be selected on module load as they are
1787 supported on all systems.
1788 The remaining options may only be set after the module is loaded, as they
1789 are available only if the implementations are compiled in and supported
1790 on the running system.
1792 Once the module is loaded, the content of
1793 /sys/module/zfs/parameters/zfs_vdev_raidz_impl will show available options
1794 with the currently selected one enclosed in [].
1795 Possible options are:
1796 fastest - (always) implementation selected using built-in benchmark
1797 original - (always) original raidz implementation
1798 scalar - (always) scalar raidz implementation
1799 sse2 - implementation using SSE2 instruction set (64bit x86 only)
1800 ssse3 - implementation using SSSE3 instruction set (64bit x86 only)
1801 avx2 - implementation using AVX2 instruction set (64bit x86 only)
1802 avx512f - implementation using AVX512F instruction set (64bit x86 only)
1803 avx512bw - implementation using AVX512F & AVX512BW instruction sets (64bit x86 only)
1804 aarch64_neon - implementation using NEON (Aarch64/64 bit ARMv8 only)
1805 aarch64_neonx2 - implementation using NEON with more unrolling (Aarch64/64 bit ARMv8 only)
1807 Default value: \fBfastest\fR.
1813 \fBzfs_zevent_cols\fR (int)
1816 When zevents are logged to the console use this as the word wrap width.
1818 Default value: \fB80\fR.
1824 \fBzfs_zevent_console\fR (int)
1827 Log events to the console
1829 Use \fB1\fR for yes and \fB0\fR for no (default).
1835 \fBzfs_zevent_len_max\fR (int)
1838 Max event queue length. A value of 0 will result in a calculated value which
1839 increases with the number of CPUs in the system (minimum 64 events). Events
1840 in the queue can be viewed with the \fBzpool events\fR command.
1842 Default value: \fB0\fR.
1848 \fBzil_replay_disable\fR (int)
1851 Disable intent logging replay. Can be disabled for recovery from corrupted
1854 Use \fB1\fR for yes and \fB0\fR for no (default).
1860 \fBzil_slog_limit\fR (ulong)
1863 Max commit bytes to separate log device
1865 Default value: \fB1,048,576\fR.
1871 \fBzio_delay_max\fR (int)
1874 A zevent will be logged if a ZIO operation takes more than N milliseconds to
1875 complete. Note that this is only a logging facility, not a timeout on
1878 Default value: \fB30,000\fR.
1884 \fBzio_dva_throttle_enabled\fR (int)
1887 Throttle block allocations in the ZIO pipeline. This allows for
1888 dynamic allocation distribution when devices are imbalanced.
1890 Default value: \fB1\fR.
1896 \fBzio_requeue_io_start_cut_in_line\fR (int)
1899 Prioritize requeued I/O
1901 Default value: \fB0\fR.
1907 \fBzio_taskq_batch_pct\fR (uint)
1910 Percentage of online CPUs (or CPU cores, etc) which will run a worker thread
1911 for IO. These workers are responsible for IO work such as compression and
1912 checksum calculations. Fractional number of CPUs will be rounded down.
1914 The default value of 75 was chosen to avoid using all CPUs which can result in
1915 latency issues and inconsistent application performance, especially when high
1916 compression is enabled.
1918 Default value: \fB75\fR.
1924 \fBzvol_inhibit_dev\fR (uint)
1927 Do not create zvol device nodes. This may slightly improve startup time on
1928 systems with a very large number of zvols.
1930 Use \fB1\fR for yes and \fB0\fR for no (default).
1936 \fBzvol_major\fR (uint)
1939 Major number for zvol block devices
1941 Default value: \fB230\fR.
1947 \fBzvol_max_discard_blocks\fR (ulong)
1950 Discard (aka TRIM) operations done on zvols will be done in batches of this
1951 many blocks, where block size is determined by the \fBvolblocksize\fR property
1954 Default value: \fB16,384\fR.
1960 \fBzvol_prefetch_bytes\fR (uint)
1963 When adding a zvol to the system prefetch \fBzvol_prefetch_bytes\fR
1964 from the start and end of the volume. Prefetching these regions
1965 of the volume is desirable because they are likely to be accessed
1966 immediately by \fBblkid(8)\fR or by the kernel scanning for a partition
1969 Default value: \fB131,072\fR.
1972 .SH ZFS I/O SCHEDULER
1973 ZFS issues I/O operations to leaf vdevs to satisfy and complete I/Os.
1974 The I/O scheduler determines when and in what order those operations are
1975 issued. The I/O scheduler divides operations into five I/O classes
1976 prioritized in the following order: sync read, sync write, async read,
1977 async write, and scrub/resilver. Each queue defines the minimum and
1978 maximum number of concurrent operations that may be issued to the
1979 device. In addition, the device has an aggregate maximum,
1980 \fBzfs_vdev_max_active\fR. Note that the sum of the per-queue minimums
1981 must not exceed the aggregate maximum. If the sum of the per-queue
1982 maximums exceeds the aggregate maximum, then the number of active I/Os
1983 may reach \fBzfs_vdev_max_active\fR, in which case no further I/Os will
1984 be issued regardless of whether all per-queue minimums have been met.
1986 For many physical devices, throughput increases with the number of
1987 concurrent operations, but latency typically suffers. Further, physical
1988 devices typically have a limit at which more concurrent operations have no
1989 effect on throughput or can actually cause it to decrease.
1991 The scheduler selects the next operation to issue by first looking for an
1992 I/O class whose minimum has not been satisfied. Once all are satisfied and
1993 the aggregate maximum has not been hit, the scheduler looks for classes
1994 whose maximum has not been satisfied. Iteration through the I/O classes is
1995 done in the order specified above. No further operations are issued if the
1996 aggregate maximum number of concurrent operations has been hit or if there
1997 are no operations queued for an I/O class that has not hit its maximum.
1998 Every time an I/O is queued or an operation completes, the I/O scheduler
1999 looks for new operations to issue.
2001 In general, smaller max_active's will lead to lower latency of synchronous
2002 operations. Larger max_active's may lead to higher overall throughput,
2003 depending on underlying storage.
2005 The ratio of the queues' max_actives determines the balance of performance
2006 between reads, writes, and scrubs. E.g., increasing
2007 \fBzfs_vdev_scrub_max_active\fR will cause the scrub or resilver to complete
2008 more quickly, but reads and writes to have higher latency and lower throughput.
2010 All I/O classes have a fixed maximum number of outstanding operations
2011 except for the async write class. Asynchronous writes represent the data
2012 that is committed to stable storage during the syncing stage for
2013 transaction groups. Transaction groups enter the syncing state
2014 periodically so the number of queued async writes will quickly burst up
2015 and then bleed down to zero. Rather than servicing them as quickly as
2016 possible, the I/O scheduler changes the maximum number of active async
2017 write I/Os according to the amount of dirty data in the pool. Since
2018 both throughput and latency typically increase with the number of
2019 concurrent operations issued to physical devices, reducing the
2020 burstiness in the number of concurrent operations also stabilizes the
2021 response time of operations from other -- and in particular synchronous
2022 -- queues. In broad strokes, the I/O scheduler will issue more
2023 concurrent operations from the async write queue as there's more dirty
2028 The number of concurrent operations issued for the async write I/O class
2029 follows a piece-wise linear function defined by a few adjustable points.
2032 | o---------| <-- zfs_vdev_async_write_max_active
2039 |-------o | | <-- zfs_vdev_async_write_min_active
2040 0|_______^______|_________|
2041 0% | | 100% of zfs_dirty_data_max
2043 | `-- zfs_vdev_async_write_active_max_dirty_percent
2044 `--------- zfs_vdev_async_write_active_min_dirty_percent
2047 Until the amount of dirty data exceeds a minimum percentage of the dirty
2048 data allowed in the pool, the I/O scheduler will limit the number of
2049 concurrent operations to the minimum. As that threshold is crossed, the
2050 number of concurrent operations issued increases linearly to the maximum at
2051 the specified maximum percentage of the dirty data allowed in the pool.
2053 Ideally, the amount of dirty data on a busy pool will stay in the sloped
2054 part of the function between \fBzfs_vdev_async_write_active_min_dirty_percent\fR
2055 and \fBzfs_vdev_async_write_active_max_dirty_percent\fR. If it exceeds the
2056 maximum percentage, this indicates that the rate of incoming data is
2057 greater than the rate that the backend storage can handle. In this case, we
2058 must further throttle incoming writes, as described in the next section.
2060 .SH ZFS TRANSACTION DELAY
2061 We delay transactions when we've determined that the backend storage
2062 isn't able to accommodate the rate of incoming writes.
2064 If there is already a transaction waiting, we delay relative to when
2065 that transaction will finish waiting. This way the calculated delay time
2066 is independent of the number of threads concurrently executing
2069 If we are the only waiter, wait relative to when the transaction
2070 started, rather than the current time. This credits the transaction for
2071 "time already served", e.g. reading indirect blocks.
2073 The minimum time for a transaction to take is calculated as:
2075 min_time = zfs_delay_scale * (dirty - min) / (max - dirty)
2076 min_time is then capped at 100 milliseconds.
2079 The delay has two degrees of freedom that can be adjusted via tunables. The
2080 percentage of dirty data at which we start to delay is defined by
2081 \fBzfs_delay_min_dirty_percent\fR. This should typically be at or above
2082 \fBzfs_vdev_async_write_active_max_dirty_percent\fR so that we only start to
2083 delay after writing at full speed has failed to keep up with the incoming write
2084 rate. The scale of the curve is defined by \fBzfs_delay_scale\fR. Roughly speaking,
2085 this variable determines the amount of delay at the midpoint of the curve.
2089 10ms +-------------------------------------------------------------*+
2105 2ms + (midpoint) * +
2108 | zfs_delay_scale ----------> ******** |
2109 0 +-------------------------------------*********----------------+
2110 0% <- zfs_dirty_data_max -> 100%
2113 Note that since the delay is added to the outstanding time remaining on the
2114 most recent transaction, the delay is effectively the inverse of IOPS.
2115 Here the midpoint of 500us translates to 2000 IOPS. The shape of the curve
2116 was chosen such that small changes in the amount of accumulated dirty data
2117 in the first 3/4 of the curve yield relatively small differences in the
2120 The effects can be easier to understand when the amount of delay is
2121 represented on a log scale:
2125 100ms +-------------------------------------------------------------++
2134 + zfs_delay_scale ----------> ***** +
2145 +--------------------------------------------------------------+
2146 0% <- zfs_dirty_data_max -> 100%
2149 Note here that only as the amount of dirty data approaches its limit does
2150 the delay start to increase rapidly. The goal of a properly tuned system
2151 should be to keep the amount of dirty data out of that range by first
2152 ensuring that the appropriate limits are set for the I/O scheduler to reach
2153 optimal throughput on the backend storage, and then by changing the value
2154 of \fBzfs_delay_scale\fR to increase the steepness of the curve.