2 .\" Copyright (c) 2013 by Turbo Fredriksson <turbo@bayour.com>. All rights reserved.
3 .\" Copyright (c) 2017 Datto Inc.
4 .\" The contents of this file are subject to the terms of the Common Development
5 .\" and Distribution License (the "License"). You may not use this file except
6 .\" in compliance with the License. You can obtain a copy of the license at
7 .\" usr/src/OPENSOLARIS.LICENSE or http://www.opensolaris.org/os/licensing.
9 .\" See the License for the specific language governing permissions and
10 .\" limitations under the License. When distributing Covered Code, include this
11 .\" CDDL HEADER in each file and include the License file at
12 .\" usr/src/OPENSOLARIS.LICENSE. If applicable, add the following below this
13 .\" CDDL HEADER, with the fields enclosed by brackets "[]" replaced with your
14 .\" own identifying information:
15 .\" Portions Copyright [yyyy] [name of copyright owner]
16 .TH ZFS-MODULE-PARAMETERS 5 "Oct 28, 2017"
18 zfs\-module\-parameters \- ZFS module parameters
22 Description of the different parameters to the ZFS module.
24 .SS "Module parameters"
31 \fBignore_hole_birth\fR (int)
34 When set, the hole_birth optimization will not be used, and all holes will
35 always be sent on zfs send. Useful if you suspect your datasets are affected
36 by a bug in hole_birth.
38 Use \fB1\fR for on (default) and \fB0\fR for off.
44 \fBl2arc_feed_again\fR (int)
47 Turbo L2ARC warm-up. When the L2ARC is cold the fill interval will be set as
50 Use \fB1\fR for yes (default) and \fB0\fR to disable.
56 \fBl2arc_feed_min_ms\fR (ulong)
59 Min feed interval in milliseconds. Requires \fBl2arc_feed_again=1\fR and only
60 applicable in related situations.
62 Default value: \fB200\fR.
68 \fBl2arc_feed_secs\fR (ulong)
71 Seconds between L2ARC writing
73 Default value: \fB1\fR.
79 \fBl2arc_headroom\fR (ulong)
82 How far through the ARC lists to search for L2ARC cacheable content, expressed
83 as a multiplier of \fBl2arc_write_max\fR
85 Default value: \fB2\fR.
91 \fBl2arc_headroom_boost\fR (ulong)
94 Scales \fBl2arc_headroom\fR by this percentage when L2ARC contents are being
95 successfully compressed before writing. A value of 100 disables this feature.
97 Default value: \fB200\fR.
103 \fBl2arc_nocompress\fR (int)
106 Skip compressing L2ARC buffers
108 Use \fB1\fR for yes and \fB0\fR for no (default).
114 \fBl2arc_noprefetch\fR (int)
117 Do not write buffers to L2ARC if they were prefetched but not used by
120 Use \fB1\fR for yes (default) and \fB0\fR to disable.
126 \fBl2arc_norw\fR (int)
129 No reads during writes
131 Use \fB1\fR for yes and \fB0\fR for no (default).
137 \fBl2arc_write_boost\fR (ulong)
140 Cold L2ARC devices will have \fBl2arc_write_max\fR increased by this amount
141 while they remain cold.
143 Default value: \fB8,388,608\fR.
149 \fBl2arc_write_max\fR (ulong)
152 Max write bytes per interval
154 Default value: \fB8,388,608\fR.
160 \fBmetaslab_aliquot\fR (ulong)
163 Metaslab granularity, in bytes. This is roughly similar to what would be
164 referred to as the "stripe size" in traditional RAID arrays. In normal
165 operation, ZFS will try to write this amount of data to a top-level vdev
166 before moving on to the next one.
168 Default value: \fB524,288\fR.
174 \fBmetaslab_bias_enabled\fR (int)
177 Enable metaslab group biasing based on its vdev's over- or under-utilization
178 relative to the pool.
180 Use \fB1\fR for yes (default) and \fB0\fR for no.
186 \fBzfs_metaslab_segment_weight_enabled\fR (int)
189 Enable/disable segment-based metaslab selection.
191 Use \fB1\fR for yes (default) and \fB0\fR for no.
197 \fBzfs_metaslab_switch_threshold\fR (int)
200 When using segment-based metaslab selection, continue allocating
201 from the active metaslab until \fBzfs_metaslab_switch_threshold\fR
202 worth of buckets have been exhausted.
204 Default value: \fB2\fR.
210 \fBmetaslab_debug_load\fR (int)
213 Load all metaslabs during pool import.
215 Use \fB1\fR for yes and \fB0\fR for no (default).
221 \fBmetaslab_debug_unload\fR (int)
224 Prevent metaslabs from being unloaded.
226 Use \fB1\fR for yes and \fB0\fR for no (default).
232 \fBmetaslab_fragmentation_factor_enabled\fR (int)
235 Enable use of the fragmentation metric in computing metaslab weights.
237 Use \fB1\fR for yes (default) and \fB0\fR for no.
243 \fBmetaslabs_per_vdev\fR (int)
246 When a vdev is added, it will be divided into approximately (but no more than) this number of metaslabs.
248 Default value: \fB200\fR.
254 \fBmetaslab_preload_enabled\fR (int)
257 Enable metaslab group preloading.
259 Use \fB1\fR for yes (default) and \fB0\fR for no.
265 \fBmetaslab_lba_weighting_enabled\fR (int)
268 Give more weight to metaslabs with lower LBAs, assuming they have
269 greater bandwidth as is typically the case on a modern constant
270 angular velocity disk drive.
272 Use \fB1\fR for yes (default) and \fB0\fR for no.
278 \fBspa_config_path\fR (charp)
283 Default value: \fB/etc/zfs/zpool.cache\fR.
289 \fBspa_asize_inflation\fR (int)
292 Multiplication factor used to estimate actual disk consumption from the
293 size of data being written. The default value is a worst case estimate,
294 but lower values may be valid for a given pool depending on its
295 configuration. Pool administrators who understand the factors involved
296 may wish to specify a more realistic inflation factor, particularly if
297 they operate close to quota or capacity limits.
299 Default value: \fB24\fR.
305 \fBspa_load_verify_data\fR (int)
308 Whether to traverse data blocks during an "extreme rewind" (\fB-X\fR)
309 import. Use 0 to disable and 1 to enable.
311 An extreme rewind import normally performs a full traversal of all
312 blocks in the pool for verification. If this parameter is set to 0,
313 the traversal skips non-metadata blocks. It can be toggled once the
314 import has started to stop or start the traversal of non-metadata blocks.
316 Default value: \fB1\fR.
322 \fBspa_load_verify_metadata\fR (int)
325 Whether to traverse blocks during an "extreme rewind" (\fB-X\fR)
326 pool import. Use 0 to disable and 1 to enable.
328 An extreme rewind import normally performs a full traversal of all
329 blocks in the pool for verification. If this parameter is set to 0,
330 the traversal is not performed. It can be toggled once the import has
331 started to stop or start the traversal.
333 Default value: \fB1\fR.
339 \fBspa_load_verify_maxinflight\fR (int)
342 Maximum concurrent I/Os during the traversal performed during an "extreme
343 rewind" (\fB-X\fR) pool import.
345 Default value: \fB10000\fR.
351 \fBspa_slop_shift\fR (int)
354 Normally, we don't allow the last 3.2% (1/(2^spa_slop_shift)) of space
355 in the pool to be consumed. This ensures that we don't run the pool
356 completely out of space, due to unaccounted changes (e.g. to the MOS).
357 It also limits the worst-case time to allocate space. If we have
358 less than this amount of free space, most ZPL operations (e.g. write,
359 create) will return ENOSPC.
361 Default value: \fB5\fR.
367 \fBzfetch_array_rd_sz\fR (ulong)
370 If prefetching is enabled, disable prefetching for reads larger than this size.
372 Default value: \fB1,048,576\fR.
378 \fBzfetch_max_distance\fR (uint)
381 Max bytes to prefetch per stream (default 8MB).
383 Default value: \fB8,388,608\fR.
389 \fBzfetch_max_streams\fR (uint)
392 Max number of streams per zfetch (prefetch streams per file).
394 Default value: \fB8\fR.
400 \fBzfetch_min_sec_reap\fR (uint)
403 Min time before an active prefetch stream can be reclaimed
405 Default value: \fB2\fR.
411 \fBzfs_arc_dnode_limit\fR (ulong)
414 When the number of bytes consumed by dnodes in the ARC exceeds this number of
415 bytes, try to unpin some of it in response to demand for non-metadata. This
416 value acts as a ceiling to the amount of dnode metadata, and defaults to 0 which
417 indicates that a percent which is based on \fBzfs_arc_dnode_limit_percent\fR of
418 the ARC meta buffers that may be used for dnodes.
420 See also \fBzfs_arc_meta_prune\fR which serves a similar purpose but is used
421 when the amount of metadata in the ARC exceeds \fBzfs_arc_meta_limit\fR rather
422 than in response to overall demand for non-metadata.
425 Default value: \fB0\fR.
431 \fBzfs_arc_dnode_limit_percent\fR (ulong)
434 Percentage that can be consumed by dnodes of ARC meta buffers.
436 See also \fBzfs_arc_dnode_limit\fR which serves a similar purpose but has a
437 higher priority if set to nonzero value.
439 Default value: \fB10\fR.
445 \fBzfs_arc_dnode_reduce_percent\fR (ulong)
448 Percentage of ARC dnodes to try to scan in response to demand for non-metadata
449 when the number of bytes consumed by dnodes exceeds \fBzfs_arc_dnode_limit\fR.
452 Default value: \fB10% of the number of dnodes in the ARC\fR.
458 \fBzfs_arc_average_blocksize\fR (int)
461 The ARC's buffer hash table is sized based on the assumption of an average
462 block size of \fBzfs_arc_average_blocksize\fR (default 8K). This works out
463 to roughly 1MB of hash table per 1GB of physical memory with 8-byte pointers.
464 For configurations with a known larger average block size this value can be
465 increased to reduce the memory footprint.
468 Default value: \fB8192\fR.
474 \fBzfs_arc_evict_batch_limit\fR (int)
477 Number ARC headers to evict per sub-list before proceeding to another sub-list.
478 This batch-style operation prevents entire sub-lists from being evicted at once
479 but comes at a cost of additional unlocking and locking.
481 Default value: \fB10\fR.
487 \fBzfs_arc_grow_retry\fR (int)
490 If set to a non zero value, it will replace the arc_grow_retry value with this value.
491 The arc_grow_retry value (default 5) is the number of seconds the ARC will wait before
492 trying to resume growth after a memory pressure event.
494 Default value: \fB0\fR.
500 \fBzfs_arc_lotsfree_percent\fR (int)
503 Throttle I/O when free system memory drops below this percentage of total
504 system memory. Setting this value to 0 will disable the throttle.
506 Default value: \fB10\fR.
512 \fBzfs_arc_max\fR (ulong)
515 Max arc size of ARC in bytes. If set to 0 then it will consume 1/2 of system
516 RAM. This value must be at least 67108864 (64 megabytes).
518 This value can be changed dynamically with some caveats. It cannot be set back
519 to 0 while running and reducing it below the current ARC size will not cause
520 the ARC to shrink without memory pressure to induce shrinking.
522 Default value: \fB0\fR.
528 \fBzfs_arc_meta_adjust_restarts\fR (ulong)
531 The number of restart passes to make while scanning the ARC attempting
532 the free buffers in order to stay below the \fBzfs_arc_meta_limit\fR.
533 This value should not need to be tuned but is available to facilitate
534 performance analysis.
536 Default value: \fB4096\fR.
542 \fBzfs_arc_meta_limit\fR (ulong)
545 The maximum allowed size in bytes that meta data buffers are allowed to
546 consume in the ARC. When this limit is reached meta data buffers will
547 be reclaimed even if the overall arc_c_max has not been reached. This
548 value defaults to 0 which indicates that a percent which is based on
549 \fBzfs_arc_meta_limit_percent\fR of the ARC may be used for meta data.
551 This value my be changed dynamically except that it cannot be set back to 0
552 for a specific percent of the ARC; it must be set to an explicit value.
554 Default value: \fB0\fR.
560 \fBzfs_arc_meta_limit_percent\fR (ulong)
563 Percentage of ARC buffers that can be used for meta data.
565 See also \fBzfs_arc_meta_limit\fR which serves a similar purpose but has a
566 higher priority if set to nonzero value.
569 Default value: \fB75\fR.
575 \fBzfs_arc_meta_min\fR (ulong)
578 The minimum allowed size in bytes that meta data buffers may consume in
579 the ARC. This value defaults to 0 which disables a floor on the amount
580 of the ARC devoted meta data.
582 Default value: \fB0\fR.
588 \fBzfs_arc_meta_prune\fR (int)
591 The number of dentries and inodes to be scanned looking for entries
592 which can be dropped. This may be required when the ARC reaches the
593 \fBzfs_arc_meta_limit\fR because dentries and inodes can pin buffers
594 in the ARC. Increasing this value will cause to dentry and inode caches
595 to be pruned more aggressively. Setting this value to 0 will disable
596 pruning the inode and dentry caches.
598 Default value: \fB10,000\fR.
604 \fBzfs_arc_meta_strategy\fR (int)
607 Define the strategy for ARC meta data buffer eviction (meta reclaim strategy).
608 A value of 0 (META_ONLY) will evict only the ARC meta data buffers.
609 A value of 1 (BALANCED) indicates that additional data buffers may be evicted if
610 that is required to in order to evict the required number of meta data buffers.
612 Default value: \fB1\fR.
618 \fBzfs_arc_min\fR (ulong)
621 Min arc size of ARC in bytes. If set to 0 then arc_c_min will default to
622 consuming the larger of 32M or 1/32 of total system memory.
624 Default value: \fB0\fR.
630 \fBzfs_arc_min_prefetch_ms\fR (int)
633 Minimum time prefetched blocks are locked in the ARC, specified in ms.
634 A value of \fB0\fR will default to 1 second.
636 Default value: \fB0\fR.
642 \fBzfs_arc_min_prescient_prefetch_ms\fR (int)
645 Minimum time "prescient prefetched" blocks are locked in the ARC, specified
646 in ms. These blocks are meant to be prefetched fairly aggresively ahead of
647 the code that may use them. A value of \fB0\fR will default to 6 seconds.
649 Default value: \fB0\fR.
655 \fBzfs_multilist_num_sublists\fR (int)
658 To allow more fine-grained locking, each ARC state contains a series
659 of lists for both data and meta data objects. Locking is performed at
660 the level of these "sub-lists". This parameters controls the number of
661 sub-lists per ARC state, and also applies to other uses of the
662 multilist data structure.
664 Default value: \fB4\fR or the number of online CPUs, whichever is greater
670 \fBzfs_arc_overflow_shift\fR (int)
673 The ARC size is considered to be overflowing if it exceeds the current
674 ARC target size (arc_c) by a threshold determined by this parameter.
675 The threshold is calculated as a fraction of arc_c using the formula
676 "arc_c >> \fBzfs_arc_overflow_shift\fR".
678 The default value of 8 causes the ARC to be considered to be overflowing
679 if it exceeds the target size by 1/256th (0.3%) of the target size.
681 When the ARC is overflowing, new buffer allocations are stalled until
682 the reclaim thread catches up and the overflow condition no longer exists.
684 Default value: \fB8\fR.
691 \fBzfs_arc_p_min_shift\fR (int)
694 If set to a non zero value, this will update arc_p_min_shift (default 4)
696 arc_p_min_shift is used to shift of arc_c for calculating both min and max
699 Default value: \fB0\fR.
705 \fBzfs_arc_p_aggressive_disable\fR (int)
708 Disable aggressive arc_p growth
710 Use \fB1\fR for yes (default) and \fB0\fR to disable.
716 \fBzfs_arc_p_dampener_disable\fR (int)
719 Disable arc_p adapt dampener
721 Use \fB1\fR for yes (default) and \fB0\fR to disable.
727 \fBzfs_arc_shrink_shift\fR (int)
730 If set to a non zero value, this will update arc_shrink_shift (default 7)
733 Default value: \fB0\fR.
739 \fBzfs_arc_pc_percent\fR (uint)
742 Percent of pagecache to reclaim arc to
744 This tunable allows ZFS arc to play more nicely with the kernel's LRU
745 pagecache. It can guarantee that the arc size won't collapse under scanning
746 pressure on the pagecache, yet still allows arc to be reclaimed down to
747 zfs_arc_min if necessary. This value is specified as percent of pagecache
748 size (as measured by NR_FILE_PAGES) where that percent may exceed 100. This
749 only operates during memory pressure/reclaim.
751 Default value: \fB0\fR (disabled).
757 \fBzfs_arc_sys_free\fR (ulong)
760 The target number of bytes the ARC should leave as free memory on the system.
761 Defaults to the larger of 1/64 of physical memory or 512K. Setting this
762 option to a non-zero value will override the default.
764 Default value: \fB0\fR.
770 \fBzfs_autoimport_disable\fR (int)
773 Disable pool import at module load by ignoring the cache file (typically \fB/etc/zfs/zpool.cache\fR).
775 Use \fB1\fR for yes (default) and \fB0\fR for no.
781 \fBzfs_commit_timeout_pct\fR (int)
784 This controls the amount of time that a ZIL block (lwb) will remain "open"
785 when it isn't "full", and it has a thread waiting for it to be committed to
786 stable storage. The timeout is scaled based on a percentage of the last lwb
787 latency to avoid significantly impacting the latency of each individual
788 transaction record (itx).
790 Default value: \fB5\fR.
796 \fBzfs_dbgmsg_enable\fR (int)
799 Internally ZFS keeps a small log to facilitate debugging. By default the log
800 is disabled, to enable it set this option to 1. The contents of the log can
801 be accessed by reading the /proc/spl/kstat/zfs/dbgmsg file. Writing 0 to
802 this proc file clears the log.
804 Default value: \fB0\fR.
810 \fBzfs_dbgmsg_maxsize\fR (int)
813 The maximum size in bytes of the internal ZFS debug log.
815 Default value: \fB4M\fR.
821 \fBzfs_dbuf_state_index\fR (int)
824 This feature is currently unused. It is normally used for controlling what
825 reporting is available under /proc/spl/kstat/zfs.
827 Default value: \fB0\fR.
833 \fBzfs_deadman_enabled\fR (int)
836 When a pool sync operation takes longer than \fBzfs_deadman_synctime_ms\fR
837 milliseconds, a "slow spa_sync" message is logged to the debug log
838 (see \fBzfs_dbgmsg_enable\fR). If \fBzfs_deadman_enabled\fR is set,
839 all pending IO operations are also checked and if any haven't completed
840 within \fBzfs_deadman_synctime_ms\fR milliseconds, a "SLOW IO" message
841 is logged to the debug log and a "delay" system event with the details of
842 the hung IO is posted.
844 Use \fB1\fR (default) to enable the slow IO check and \fB0\fR to disable.
850 \fBzfs_deadman_checktime_ms\fR (int)
853 Once a pool sync operation has taken longer than
854 \fBzfs_deadman_synctime_ms\fR milliseconds, continue to check for slow
855 operations every \fBzfs_deadman_checktime_ms\fR milliseconds.
857 Default value: \fB5,000\fR.
863 \fBzfs_deadman_synctime_ms\fR (ulong)
866 Interval in milliseconds after which the deadman is triggered and also
867 the interval after which an IO operation is considered to be "hung"
868 if \fBzfs_deadman_enabled\fR is set.
870 See \fBzfs_deadman_enabled\fR.
872 Default value: \fB1,000,000\fR.
878 \fBzfs_dedup_prefetch\fR (int)
881 Enable prefetching dedup-ed blks
883 Use \fB1\fR for yes and \fB0\fR to disable (default).
889 \fBzfs_delay_min_dirty_percent\fR (int)
892 Start to delay each transaction once there is this amount of dirty data,
893 expressed as a percentage of \fBzfs_dirty_data_max\fR.
894 This value should be >= zfs_vdev_async_write_active_max_dirty_percent.
895 See the section "ZFS TRANSACTION DELAY".
897 Default value: \fB60\fR.
903 \fBzfs_delay_scale\fR (int)
906 This controls how quickly the transaction delay approaches infinity.
907 Larger values cause longer delays for a given amount of dirty data.
909 For the smoothest delay, this value should be about 1 billion divided
910 by the maximum number of operations per second. This will smoothly
911 handle between 10x and 1/10th this number.
913 See the section "ZFS TRANSACTION DELAY".
915 Note: \fBzfs_delay_scale\fR * \fBzfs_dirty_data_max\fR must be < 2^64.
917 Default value: \fB500,000\fR.
923 \fBzfs_delete_blocks\fR (ulong)
926 This is the used to define a large file for the purposes of delete. Files
927 containing more than \fBzfs_delete_blocks\fR will be deleted asynchronously
928 while smaller files are deleted synchronously. Decreasing this value will
929 reduce the time spent in an unlink(2) system call at the expense of a longer
930 delay before the freed space is available.
932 Default value: \fB20,480\fR.
938 \fBzfs_dirty_data_max\fR (int)
941 Determines the dirty space limit in bytes. Once this limit is exceeded, new
942 writes are halted until space frees up. This parameter takes precedence
943 over \fBzfs_dirty_data_max_percent\fR.
944 See the section "ZFS TRANSACTION DELAY".
946 Default value: 10 percent of all memory, capped at \fBzfs_dirty_data_max_max\fR.
952 \fBzfs_dirty_data_max_max\fR (int)
955 Maximum allowable value of \fBzfs_dirty_data_max\fR, expressed in bytes.
956 This limit is only enforced at module load time, and will be ignored if
957 \fBzfs_dirty_data_max\fR is later changed. This parameter takes
958 precedence over \fBzfs_dirty_data_max_max_percent\fR. See the section
959 "ZFS TRANSACTION DELAY".
961 Default value: 25% of physical RAM.
967 \fBzfs_dirty_data_max_max_percent\fR (int)
970 Maximum allowable value of \fBzfs_dirty_data_max\fR, expressed as a
971 percentage of physical RAM. This limit is only enforced at module load
972 time, and will be ignored if \fBzfs_dirty_data_max\fR is later changed.
973 The parameter \fBzfs_dirty_data_max_max\fR takes precedence over this
974 one. See the section "ZFS TRANSACTION DELAY".
976 Default value: \fB25\fR.
982 \fBzfs_dirty_data_max_percent\fR (int)
985 Determines the dirty space limit, expressed as a percentage of all
986 memory. Once this limit is exceeded, new writes are halted until space frees
987 up. The parameter \fBzfs_dirty_data_max\fR takes precedence over this
988 one. See the section "ZFS TRANSACTION DELAY".
990 Default value: 10%, subject to \fBzfs_dirty_data_max_max\fR.
996 \fBzfs_dirty_data_sync\fR (int)
999 Start syncing out a transaction group if there is at least this much dirty data.
1001 Default value: \fB67,108,864\fR.
1007 \fBzfs_fletcher_4_impl\fR (string)
1010 Select a fletcher 4 implementation.
1012 Supported selectors are: \fBfastest\fR, \fBscalar\fR, \fBsse2\fR, \fBssse3\fR,
1013 \fBavx2\fR, \fBavx512f\fR, and \fBaarch64_neon\fR.
1014 All of the selectors except \fBfastest\fR and \fBscalar\fR require instruction
1015 set extensions to be available and will only appear if ZFS detects that they are
1016 present at runtime. If multiple implementations of fletcher 4 are available,
1017 the \fBfastest\fR will be chosen using a micro benchmark. Selecting \fBscalar\fR
1018 results in the original, CPU based calculation, being used. Selecting any option
1019 other than \fBfastest\fR and \fBscalar\fR results in vector instructions from
1020 the respective CPU instruction set being used.
1022 Default value: \fBfastest\fR.
1028 \fBzfs_free_bpobj_enabled\fR (int)
1031 Enable/disable the processing of the free_bpobj object.
1033 Default value: \fB1\fR.
1039 \fBzfs_free_max_blocks\fR (ulong)
1042 Maximum number of blocks freed in a single txg.
1044 Default value: \fB100,000\fR.
1050 \fBzfs_vdev_async_read_max_active\fR (int)
1053 Maximum asynchronous read I/Os active to each device.
1054 See the section "ZFS I/O SCHEDULER".
1056 Default value: \fB3\fR.
1062 \fBzfs_vdev_async_read_min_active\fR (int)
1065 Minimum asynchronous read I/Os active to each device.
1066 See the section "ZFS I/O SCHEDULER".
1068 Default value: \fB1\fR.
1074 \fBzfs_vdev_async_write_active_max_dirty_percent\fR (int)
1077 When the pool has more than
1078 \fBzfs_vdev_async_write_active_max_dirty_percent\fR dirty data, use
1079 \fBzfs_vdev_async_write_max_active\fR to limit active async writes. If
1080 the dirty data is between min and max, the active I/O limit is linearly
1081 interpolated. See the section "ZFS I/O SCHEDULER".
1083 Default value: \fB60\fR.
1089 \fBzfs_vdev_async_write_active_min_dirty_percent\fR (int)
1092 When the pool has less than
1093 \fBzfs_vdev_async_write_active_min_dirty_percent\fR dirty data, use
1094 \fBzfs_vdev_async_write_min_active\fR to limit active async writes. If
1095 the dirty data is between min and max, the active I/O limit is linearly
1096 interpolated. See the section "ZFS I/O SCHEDULER".
1098 Default value: \fB30\fR.
1104 \fBzfs_vdev_async_write_max_active\fR (int)
1107 Maximum asynchronous write I/Os active to each device.
1108 See the section "ZFS I/O SCHEDULER".
1110 Default value: \fB10\fR.
1116 \fBzfs_vdev_async_write_min_active\fR (int)
1119 Minimum asynchronous write I/Os active to each device.
1120 See the section "ZFS I/O SCHEDULER".
1122 Lower values are associated with better latency on rotational media but poorer
1123 resilver performance. The default value of 2 was chosen as a compromise. A
1124 value of 3 has been shown to improve resilver performance further at a cost of
1125 further increasing latency.
1127 Default value: \fB2\fR.
1133 \fBzfs_vdev_max_active\fR (int)
1136 The maximum number of I/Os active to each device. Ideally, this will be >=
1137 the sum of each queue's max_active. It must be at least the sum of each
1138 queue's min_active. See the section "ZFS I/O SCHEDULER".
1140 Default value: \fB1,000\fR.
1146 \fBzfs_vdev_scrub_max_active\fR (int)
1149 Maximum scrub I/Os active to each device.
1150 See the section "ZFS I/O SCHEDULER".
1152 Default value: \fB2\fR.
1158 \fBzfs_vdev_scrub_min_active\fR (int)
1161 Minimum scrub I/Os active to each device.
1162 See the section "ZFS I/O SCHEDULER".
1164 Default value: \fB1\fR.
1170 \fBzfs_vdev_sync_read_max_active\fR (int)
1173 Maximum synchronous read I/Os active to each device.
1174 See the section "ZFS I/O SCHEDULER".
1176 Default value: \fB10\fR.
1182 \fBzfs_vdev_sync_read_min_active\fR (int)
1185 Minimum synchronous read I/Os active to each device.
1186 See the section "ZFS I/O SCHEDULER".
1188 Default value: \fB10\fR.
1194 \fBzfs_vdev_sync_write_max_active\fR (int)
1197 Maximum synchronous write I/Os active to each device.
1198 See the section "ZFS I/O SCHEDULER".
1200 Default value: \fB10\fR.
1206 \fBzfs_vdev_sync_write_min_active\fR (int)
1209 Minimum synchronous write I/Os active to each device.
1210 See the section "ZFS I/O SCHEDULER".
1212 Default value: \fB10\fR.
1218 \fBzfs_vdev_queue_depth_pct\fR (int)
1221 Maximum number of queued allocations per top-level vdev expressed as
1222 a percentage of \fBzfs_vdev_async_write_max_active\fR which allows the
1223 system to detect devices that are more capable of handling allocations
1224 and to allocate more blocks to those devices. It allows for dynamic
1225 allocation distribution when devices are imbalanced as fuller devices
1226 will tend to be slower than empty devices.
1228 See also \fBzio_dva_throttle_enabled\fR.
1230 Default value: \fB1000\fR.
1236 \fBzfs_disable_dup_eviction\fR (int)
1239 Disable duplicate buffer eviction
1241 Use \fB1\fR for yes and \fB0\fR for no (default).
1247 \fBzfs_expire_snapshot\fR (int)
1250 Seconds to expire .zfs/snapshot
1252 Default value: \fB300\fR.
1258 \fBzfs_admin_snapshot\fR (int)
1261 Allow the creation, removal, or renaming of entries in the .zfs/snapshot
1262 directory to cause the creation, destruction, or renaming of snapshots.
1263 When enabled this functionality works both locally and over NFS exports
1264 which have the 'no_root_squash' option set. This functionality is disabled
1267 Use \fB1\fR for yes and \fB0\fR for no (default).
1273 \fBzfs_flags\fR (int)
1276 Set additional debugging flags. The following flags may be bitwise-or'd
1288 Enable dprintf entries in the debug log.
1290 2 ZFS_DEBUG_DBUF_VERIFY *
1291 Enable extra dbuf verifications.
1293 4 ZFS_DEBUG_DNODE_VERIFY *
1294 Enable extra dnode verifications.
1296 8 ZFS_DEBUG_SNAPNAMES
1297 Enable snapshot name verification.
1300 Check for illegally modified ARC buffers.
1303 Enable spa_dbgmsg entries in the debug log.
1305 64 ZFS_DEBUG_ZIO_FREE
1306 Enable verification of block frees.
1308 128 ZFS_DEBUG_HISTOGRAM_VERIFY
1309 Enable extra spacemap histogram verifications.
1311 256 ZFS_DEBUG_METASLAB_VERIFY
1312 Verify space accounting on disk matches in-core range_trees.
1314 512 ZFS_DEBUG_SET_ERROR
1315 Enable SET_ERROR and dprintf entries in the debug log.
1318 * Requires debug build.
1320 Default value: \fB0\fR.
1326 \fBzfs_free_leak_on_eio\fR (int)
1329 If destroy encounters an EIO while reading metadata (e.g. indirect
1330 blocks), space referenced by the missing metadata can not be freed.
1331 Normally this causes the background destroy to become "stalled", as
1332 it is unable to make forward progress. While in this stalled state,
1333 all remaining space to free from the error-encountering filesystem is
1334 "temporarily leaked". Set this flag to cause it to ignore the EIO,
1335 permanently leak the space from indirect blocks that can not be read,
1336 and continue to free everything else that it can.
1338 The default, "stalling" behavior is useful if the storage partially
1339 fails (i.e. some but not all i/os fail), and then later recovers. In
1340 this case, we will be able to continue pool operations while it is
1341 partially failed, and when it recovers, we can continue to free the
1342 space, with no leaks. However, note that this case is actually
1345 Typically pools either (a) fail completely (but perhaps temporarily,
1346 e.g. a top-level vdev going offline), or (b) have localized,
1347 permanent errors (e.g. disk returns the wrong data due to bit flip or
1348 firmware bug). In case (a), this setting does not matter because the
1349 pool will be suspended and the sync thread will not be able to make
1350 forward progress regardless. In case (b), because the error is
1351 permanent, the best we can do is leak the minimum amount of space,
1352 which is what setting this flag will do. Therefore, it is reasonable
1353 for this flag to normally be set, but we chose the more conservative
1354 approach of not setting it, so that there is no possibility of
1355 leaking space in the "partial temporary" failure case.
1357 Default value: \fB0\fR.
1363 \fBzfs_free_min_time_ms\fR (int)
1366 During a \fBzfs destroy\fR operation using \fBfeature@async_destroy\fR a minimum
1367 of this much time will be spent working on freeing blocks per txg.
1369 Default value: \fB1,000\fR.
1375 \fBzfs_immediate_write_sz\fR (long)
1378 Largest data block to write to zil. Larger blocks will be treated as if the
1379 dataset being written to had the property setting \fBlogbias=throughput\fR.
1381 Default value: \fB32,768\fR.
1387 \fBzfs_max_recordsize\fR (int)
1390 We currently support block sizes from 512 bytes to 16MB. The benefits of
1391 larger blocks, and thus larger IO, need to be weighed against the cost of
1392 COWing a giant block to modify one byte. Additionally, very large blocks
1393 can have an impact on i/o latency, and also potentially on the memory
1394 allocator. Therefore, we do not allow the recordsize to be set larger than
1395 zfs_max_recordsize (default 1MB). Larger blocks can be created by changing
1396 this tunable, and pools with larger blocks can always be imported and used,
1397 regardless of this setting.
1399 Default value: \fB1,048,576\fR.
1405 \fBzfs_mdcomp_disable\fR (int)
1408 Disable meta data compression
1410 Use \fB1\fR for yes and \fB0\fR for no (default).
1416 \fBzfs_metaslab_fragmentation_threshold\fR (int)
1419 Allow metaslabs to keep their active state as long as their fragmentation
1420 percentage is less than or equal to this value. An active metaslab that
1421 exceeds this threshold will no longer keep its active status allowing
1422 better metaslabs to be selected.
1424 Default value: \fB70\fR.
1430 \fBzfs_mg_fragmentation_threshold\fR (int)
1433 Metaslab groups are considered eligible for allocations if their
1434 fragmentation metric (measured as a percentage) is less than or equal to
1435 this value. If a metaslab group exceeds this threshold then it will be
1436 skipped unless all metaslab groups within the metaslab class have also
1437 crossed this threshold.
1439 Default value: \fB85\fR.
1445 \fBzfs_mg_noalloc_threshold\fR (int)
1448 Defines a threshold at which metaslab groups should be eligible for
1449 allocations. The value is expressed as a percentage of free space
1450 beyond which a metaslab group is always eligible for allocations.
1451 If a metaslab group's free space is less than or equal to the
1452 threshold, the allocator will avoid allocating to that group
1453 unless all groups in the pool have reached the threshold. Once all
1454 groups have reached the threshold, all groups are allowed to accept
1455 allocations. The default value of 0 disables the feature and causes
1456 all metaslab groups to be eligible for allocations.
1458 This parameter allows one to deal with pools having heavily imbalanced
1459 vdevs such as would be the case when a new vdev has been added.
1460 Setting the threshold to a non-zero percentage will stop allocations
1461 from being made to vdevs that aren't filled to the specified percentage
1462 and allow lesser filled vdevs to acquire more allocations than they
1463 otherwise would under the old \fBzfs_mg_alloc_failures\fR facility.
1465 Default value: \fB0\fR.
1471 \fBzfs_multihost_history\fR (int)
1474 Historical statistics for the last N multihost updates will be available in
1475 \fB/proc/spl/kstat/zfs/<pool>/multihost\fR
1477 Default value: \fB0\fR.
1483 \fBzfs_multihost_interval\fR (ulong)
1486 Used to control the frequency of multihost writes which are performed when the
1487 \fBmultihost\fR pool property is on. This is one factor used to determine
1488 the length of the activity check during import.
1490 The multihost write period is \fBzfs_multihost_interval / leaf-vdevs\fR milliseconds.
1491 This means that on average a multihost write will be issued for each leaf vdev every
1492 \fBzfs_multihost_interval\fR milliseconds. In practice, the observed period can
1493 vary with the I/O load and this observed value is the delay which is stored in
1496 On import the activity check waits a minimum amount of time determined by
1497 \fBzfs_multihost_interval * zfs_multihost_import_intervals\fR. The activity
1498 check time may be further extended if the value of mmp delay found in the best
1499 uberblock indicates actual multihost updates happened at longer intervals than
1500 \fBzfs_multihost_interval\fR. A minimum value of \fB100ms\fR is enforced.
1502 Default value: \fB1000\fR.
1508 \fBzfs_multihost_import_intervals\fR (uint)
1511 Used to control the duration of the activity test on import. Smaller values of
1512 \fBzfs_multihost_import_intervals\fR will reduce the import time but increase
1513 the risk of failing to detect an active pool. The total activity check time is
1514 never allowed to drop below one second. A value of 0 is ignored and treated as
1517 Default value: \fB10\fR.
1523 \fBzfs_multihost_fail_intervals\fR (uint)
1526 Controls the behavior of the pool when multihost write failures are detected.
1528 When \fBzfs_multihost_fail_intervals = 0\fR then multihost write failures are ignored.
1529 The failures will still be reported to the ZED which depending on its
1530 configuration may take action such as suspending the pool or offlining a device.
1532 When \fBzfs_multihost_fail_intervals > 0\fR then sequential multihost write failures
1533 will cause the pool to be suspended. This occurs when
1534 \fBzfs_multihost_fail_intervals * zfs_multihost_interval\fR milliseconds have
1535 passed since the last successful multihost write. This guarantees the activity test
1536 will see multihost writes if the pool is imported.
1538 Default value: \fB5\fR.
1544 \fBzfs_no_scrub_io\fR (int)
1547 Set for no scrub I/O. This results in scrubs not actually scrubbing data and
1548 simply doing a metadata crawl of the pool instead.
1550 Use \fB1\fR for yes and \fB0\fR for no (default).
1556 \fBzfs_no_scrub_prefetch\fR (int)
1559 Set to disable block prefetching for scrubs.
1561 Use \fB1\fR for yes and \fB0\fR for no (default).
1567 \fBzfs_nocacheflush\fR (int)
1570 Disable cache flush operations on disks when writing. Beware, this may cause
1571 corruption if disks re-order writes.
1573 Use \fB1\fR for yes and \fB0\fR for no (default).
1579 \fBzfs_nopwrite_enabled\fR (int)
1584 Use \fB1\fR for yes (default) and \fB0\fR to disable.
1590 \fBzfs_dmu_offset_next_sync\fR (int)
1593 Enable forcing txg sync to find holes. When enabled forces ZFS to act
1594 like prior versions when SEEK_HOLE or SEEK_DATA flags are used, which
1595 when a dnode is dirty causes txg's to be synced so that this data can be
1598 Use \fB1\fR for yes and \fB0\fR to disable (default).
1604 \fBzfs_pd_bytes_max\fR (int)
1607 The number of bytes which should be prefetched during a pool traversal
1608 (eg: \fBzfs send\fR or other data crawling operations)
1610 Default value: \fB52,428,800\fR.
1616 \fBzfs_per_txg_dirty_frees_percent \fR (ulong)
1619 Tunable to control percentage of dirtied blocks from frees in one TXG.
1620 After this threshold is crossed, additional dirty blocks from frees
1621 wait until the next TXG.
1622 A value of zero will disable this throttle.
1624 Default value: \fB30\fR and \fB0\fR to disable.
1632 \fBzfs_prefetch_disable\fR (int)
1635 This tunable disables predictive prefetch. Note that it leaves "prescient"
1636 prefetch (e.g. prefetch for zfs send) intact. Unlike predictive prefetch,
1637 prescient prefetch never issues i/os that end up not being needed, so it
1638 can't hurt performance.
1640 Use \fB1\fR for yes and \fB0\fR for no (default).
1646 \fBzfs_read_chunk_size\fR (long)
1649 Bytes to read per chunk
1651 Default value: \fB1,048,576\fR.
1657 \fBzfs_read_history\fR (int)
1660 Historical statistics for the last N reads will be available in
1661 \fB/proc/spl/kstat/zfs/<pool>/reads\fR
1663 Default value: \fB0\fR (no data is kept).
1669 \fBzfs_read_history_hits\fR (int)
1672 Include cache hits in read history
1674 Use \fB1\fR for yes and \fB0\fR for no (default).
1680 \fBzfs_recover\fR (int)
1683 Set to attempt to recover from fatal errors. This should only be used as a
1684 last resort, as it typically results in leaked space, or worse.
1686 Use \fB1\fR for yes and \fB0\fR for no (default).
1692 \fBzfs_resilver_min_time_ms\fR (int)
1695 Resilvers are processed by the sync thread. While resilvering it will spend
1696 at least this much time working on a resilver between txg flushes.
1698 Default value: \fB3,000\fR.
1704 \fBzfs_scrub_min_time_ms\fR (int)
1707 Scrubs are processed by the sync thread. While scrubbing it will spend
1708 at least this much time working on a scrub between txg flushes.
1710 Default value: \fB1,000\fR.
1716 \fBzfs_scan_checkpoint_intval\fR (int)
1719 To preserve progress across reboots the sequential scan algorithm periodically
1720 needs to stop metadata scanning and issue all the verifications I/Os to disk.
1721 The frequency of this flushing is determined by the
1722 \fBfBzfs_scan_checkpoint_intval\fR tunable.
1724 Default value: \fB7200\fR seconds (every 2 hours).
1730 \fBzfs_scan_fill_weight\fR (int)
1733 This tunable affects how scrub and resilver I/O segments are ordered. A higher
1734 number indicates that we care more about how filled in a segment is, while a
1735 lower number indicates we care more about the size of the extent without
1736 considering the gaps within a segment. This value is only tunable upon module
1737 insertion. Changing the value afterwards will have no affect on scrub or
1738 resilver performance.
1740 Default value: \fB3\fR.
1746 \fBzfs_scan_issue_strategy\fR (int)
1749 Determines the order that data will be verified while scrubbing or resilvering.
1750 If set to \fB1\fR, data will be verified as sequentially as possible, given the
1751 amount of memory reserved for scrubbing (see \fBzfs_scan_mem_lim_fact\fR). This
1752 may improve scrub performance if the pool's data is very fragmented. If set to
1753 \fB2\fR, the largest mostly-contiguous chunk of found data will be verified
1754 first. By deferring scrubbing of small segments, we may later find adjacent data
1755 to coalesce and increase the segment size. If set to \fB0\fR, zfs will use
1756 strategy \fB1\fR during normal verification and strategy \fB2\fR while taking a
1759 Default value: \fB0\fR.
1765 \fBzfs_scan_legacy\fR (int)
1768 A value of 0 indicates that scrubs and resilvers will gather metadata in
1769 memory before issuing sequential I/O. A value of 1 indicates that the legacy
1770 algorithm will be used where I/O is initiated as soon as it is discovered.
1771 Changing this value to 0 will not affect scrubs or resilvers that are already
1774 Default value: \fB0\fR.
1780 \fBzfs_scan_max_ext_gap\fR (int)
1783 Indicates the largest gap in bytes between scrub / resilver I/Os that will still
1784 be considered sequential for sorting purposes. Changing this value will not
1785 affect scrubs or resilvers that are already in progress.
1787 Default value: \fB2097152 (2 MB)\fR.
1793 \fBzfs_scan_mem_lim_fact\fR (int)
1796 Maximum fraction of RAM used for I/O sorting by sequential scan algorithm.
1797 This tunable determines the hard limit for I/O sorting memory usage.
1798 When the hard limit is reached we stop scanning metadata and start issuing
1799 data verification I/O. This is done until we get below the soft limit.
1801 Default value: \fB20\fR which is 5% of RAM (1/20).
1807 \fBzfs_scan_mem_lim_soft_fact\fR (int)
1810 The fraction of the hard limit used to determined the soft limit for I/O sorting
1811 by the sequential scan algorithm. When we cross this limit from bellow no action
1812 is taken. When we cross this limit from above it is because we are issuing
1813 verification I/O. In this case (unless the metadata scan is done) we stop
1814 issuing verification I/O and start scanning metadata again until we get to the
1817 Default value: \fB20\fR which is 5% of the hard limit (1/20).
1823 \fBzfs_scan_vdev_limit\fR (int)
1826 Maximum amount of data that can be concurrently issued at once for scrubs and
1827 resilvers per leaf device, given in bytes.
1829 Default value: \fB41943040\fR.
1835 \fBzfs_send_corrupt_data\fR (int)
1838 Allow sending of corrupt data (ignore read/checksum errors when sending data)
1840 Use \fB1\fR for yes and \fB0\fR for no (default).
1846 \fBzfs_sync_pass_deferred_free\fR (int)
1849 Flushing of data to disk is done in passes. Defer frees starting in this pass
1851 Default value: \fB2\fR.
1857 \fBzfs_sync_pass_dont_compress\fR (int)
1860 Don't compress starting in this pass
1862 Default value: \fB5\fR.
1868 \fBzfs_sync_pass_rewrite\fR (int)
1871 Rewrite new block pointers starting in this pass
1873 Default value: \fB2\fR.
1879 \fBzfs_sync_taskq_batch_pct\fR (int)
1882 This controls the number of threads used by the dp_sync_taskq. The default
1883 value of 75% will create a maximum of one thread per cpu.
1885 Default value: \fB75\fR.
1891 \fBzfs_txg_history\fR (int)
1894 Historical statistics for the last N txgs will be available in
1895 \fB/proc/spl/kstat/zfs/<pool>/txgs\fR
1897 Default value: \fB0\fR.
1903 \fBzfs_txg_timeout\fR (int)
1906 Flush dirty data to disk at least every N seconds (maximum txg duration)
1908 Default value: \fB5\fR.
1914 \fBzfs_vdev_aggregation_limit\fR (int)
1917 Max vdev I/O aggregation size
1919 Default value: \fB131,072\fR.
1925 \fBzfs_vdev_cache_bshift\fR (int)
1928 Shift size to inflate reads too
1930 Default value: \fB16\fR (effectively 65536).
1936 \fBzfs_vdev_cache_max\fR (int)
1939 Inflate reads smaller than this value to meet the \fBzfs_vdev_cache_bshift\fR
1942 Default value: \fB16384\fR.
1948 \fBzfs_vdev_cache_size\fR (int)
1951 Total size of the per-disk cache in bytes.
1953 Currently this feature is disabled as it has been found to not be helpful
1954 for performance and in some cases harmful.
1956 Default value: \fB0\fR.
1962 \fBzfs_vdev_mirror_rotating_inc\fR (int)
1965 A number by which the balancing algorithm increments the load calculation for
1966 the purpose of selecting the least busy mirror member when an I/O immediately
1967 follows its predecessor on rotational vdevs for the purpose of making decisions
1970 Default value: \fB0\fR.
1976 \fBzfs_vdev_mirror_rotating_seek_inc\fR (int)
1979 A number by which the balancing algorithm increments the load calculation for
1980 the purpose of selecting the least busy mirror member when an I/O lacks
1981 locality as defined by the zfs_vdev_mirror_rotating_seek_offset. I/Os within
1982 this that are not immediately following the previous I/O are incremented by
1985 Default value: \fB5\fR.
1991 \fBzfs_vdev_mirror_rotating_seek_offset\fR (int)
1994 The maximum distance for the last queued I/O in which the balancing algorithm
1995 considers an I/O to have locality.
1996 See the section "ZFS I/O SCHEDULER".
1998 Default value: \fB1048576\fR.
2004 \fBzfs_vdev_mirror_non_rotating_inc\fR (int)
2007 A number by which the balancing algorithm increments the load calculation for
2008 the purpose of selecting the least busy mirror member on non-rotational vdevs
2009 when I/Os do not immediately follow one another.
2011 Default value: \fB0\fR.
2017 \fBzfs_vdev_mirror_non_rotating_seek_inc\fR (int)
2020 A number by which the balancing algorithm increments the load calculation for
2021 the purpose of selecting the least busy mirror member when an I/O lacks
2022 locality as defined by the zfs_vdev_mirror_rotating_seek_offset. I/Os within
2023 this that are not immediately following the previous I/O are incremented by
2026 Default value: \fB1\fR.
2032 \fBzfs_vdev_read_gap_limit\fR (int)
2035 Aggregate read I/O operations if the gap on-disk between them is within this
2038 Default value: \fB32,768\fR.
2044 \fBzfs_vdev_scheduler\fR (charp)
2047 Set the Linux I/O scheduler on whole disk vdevs to this scheduler. Valid options
2048 are noop, cfq, bfq & deadline
2050 Default value: \fBnoop\fR.
2056 \fBzfs_vdev_write_gap_limit\fR (int)
2059 Aggregate write I/O over gap
2061 Default value: \fB4,096\fR.
2067 \fBzfs_vdev_raidz_impl\fR (string)
2070 Parameter for selecting raidz parity implementation to use.
2072 Options marked (always) below may be selected on module load as they are
2073 supported on all systems.
2074 The remaining options may only be set after the module is loaded, as they
2075 are available only if the implementations are compiled in and supported
2076 on the running system.
2078 Once the module is loaded, the content of
2079 /sys/module/zfs/parameters/zfs_vdev_raidz_impl will show available options
2080 with the currently selected one enclosed in [].
2081 Possible options are:
2082 fastest - (always) implementation selected using built-in benchmark
2083 original - (always) original raidz implementation
2084 scalar - (always) scalar raidz implementation
2085 sse2 - implementation using SSE2 instruction set (64bit x86 only)
2086 ssse3 - implementation using SSSE3 instruction set (64bit x86 only)
2087 avx2 - implementation using AVX2 instruction set (64bit x86 only)
2088 avx512f - implementation using AVX512F instruction set (64bit x86 only)
2089 avx512bw - implementation using AVX512F & AVX512BW instruction sets (64bit x86 only)
2090 aarch64_neon - implementation using NEON (Aarch64/64 bit ARMv8 only)
2091 aarch64_neonx2 - implementation using NEON with more unrolling (Aarch64/64 bit ARMv8 only)
2093 Default value: \fBfastest\fR.
2099 \fBzfs_zevent_cols\fR (int)
2102 When zevents are logged to the console use this as the word wrap width.
2104 Default value: \fB80\fR.
2110 \fBzfs_zevent_console\fR (int)
2113 Log events to the console
2115 Use \fB1\fR for yes and \fB0\fR for no (default).
2121 \fBzfs_zevent_len_max\fR (int)
2124 Max event queue length. A value of 0 will result in a calculated value which
2125 increases with the number of CPUs in the system (minimum 64 events). Events
2126 in the queue can be viewed with the \fBzpool events\fR command.
2128 Default value: \fB0\fR.
2134 \fBzfs_zil_clean_taskq_maxalloc\fR (int)
2137 The maximum number of taskq entries that are allowed to be cached. When this
2138 limit is exceeded transaction records (itxs) will be cleaned synchronously.
2140 Default value: \fB1048576\fR.
2146 \fBzfs_zil_clean_taskq_minalloc\fR (int)
2149 The number of taskq entries that are pre-populated when the taskq is first
2150 created and are immediately available for use.
2152 Default value: \fB1024\fR.
2158 \fBzfs_zil_clean_taskq_nthr_pct\fR (int)
2161 This controls the number of threads used by the dp_zil_clean_taskq. The default
2162 value of 100% will create a maximum of one thread per cpu.
2164 Default value: \fB100\fR.
2170 \fBzil_replay_disable\fR (int)
2173 Disable intent logging replay. Can be disabled for recovery from corrupted
2176 Use \fB1\fR for yes and \fB0\fR for no (default).
2182 \fBzil_slog_bulk\fR (ulong)
2185 Limit SLOG write size per commit executed with synchronous priority.
2186 Any writes above that will be executed with lower (asynchronous) priority
2187 to limit potential SLOG device abuse by single active ZIL writer.
2189 Default value: \fB786,432\fR.
2195 \fBzio_delay_max\fR (int)
2198 A zevent will be logged if a ZIO operation takes more than N milliseconds to
2199 complete. Note that this is only a logging facility, not a timeout on
2202 Default value: \fB30,000\fR.
2208 \fBzio_dva_throttle_enabled\fR (int)
2211 Throttle block allocations in the ZIO pipeline. This allows for
2212 dynamic allocation distribution when devices are imbalanced.
2213 When enabled, the maximum number of pending allocations per top-level vdev
2214 is limited by \fBzfs_vdev_queue_depth_pct\fR.
2216 Default value: \fB1\fR.
2222 \fBzio_requeue_io_start_cut_in_line\fR (int)
2225 Prioritize requeued I/O
2227 Default value: \fB0\fR.
2233 \fBzio_taskq_batch_pct\fR (uint)
2236 Percentage of online CPUs (or CPU cores, etc) which will run a worker thread
2237 for IO. These workers are responsible for IO work such as compression and
2238 checksum calculations. Fractional number of CPUs will be rounded down.
2240 The default value of 75 was chosen to avoid using all CPUs which can result in
2241 latency issues and inconsistent application performance, especially when high
2242 compression is enabled.
2244 Default value: \fB75\fR.
2250 \fBzvol_inhibit_dev\fR (uint)
2253 Do not create zvol device nodes. This may slightly improve startup time on
2254 systems with a very large number of zvols.
2256 Use \fB1\fR for yes and \fB0\fR for no (default).
2262 \fBzvol_major\fR (uint)
2265 Major number for zvol block devices
2267 Default value: \fB230\fR.
2273 \fBzvol_max_discard_blocks\fR (ulong)
2276 Discard (aka TRIM) operations done on zvols will be done in batches of this
2277 many blocks, where block size is determined by the \fBvolblocksize\fR property
2280 Default value: \fB16,384\fR.
2286 \fBzvol_prefetch_bytes\fR (uint)
2289 When adding a zvol to the system prefetch \fBzvol_prefetch_bytes\fR
2290 from the start and end of the volume. Prefetching these regions
2291 of the volume is desirable because they are likely to be accessed
2292 immediately by \fBblkid(8)\fR or by the kernel scanning for a partition
2295 Default value: \fB131,072\fR.
2301 \fBzvol_request_sync\fR (uint)
2304 When processing I/O requests for a zvol submit them synchronously. This
2305 effectively limits the queue depth to 1 for each I/O submitter. When set
2306 to 0 requests are handled asynchronously by a thread pool. The number of
2307 requests which can be handled concurrently is controller by \fBzvol_threads\fR.
2309 Default value: \fB0\fR.
2315 \fBzvol_threads\fR (uint)
2318 Max number of threads which can handle zvol I/O requests concurrently.
2320 Default value: \fB32\fR.
2326 \fBzvol_volmode\fR (uint)
2329 Defines zvol block devices behaviour when \fBvolmode\fR is set to \fBdefault\fR.
2330 Valid values are \fB1\fR (full), \fB2\fR (dev) and \fB3\fR (none).
2332 Default value: \fB1\fR.
2338 \fBzfs_qat_disable\fR (int)
2341 This tunable disables qat hardware acceleration for gzip compression.
2342 It is available only if qat acceleration is compiled in and qat driver
2345 Use \fB1\fR for yes and \fB0\fR for no (default).
2348 .SH ZFS I/O SCHEDULER
2349 ZFS issues I/O operations to leaf vdevs to satisfy and complete I/Os.
2350 The I/O scheduler determines when and in what order those operations are
2351 issued. The I/O scheduler divides operations into five I/O classes
2352 prioritized in the following order: sync read, sync write, async read,
2353 async write, and scrub/resilver. Each queue defines the minimum and
2354 maximum number of concurrent operations that may be issued to the
2355 device. In addition, the device has an aggregate maximum,
2356 \fBzfs_vdev_max_active\fR. Note that the sum of the per-queue minimums
2357 must not exceed the aggregate maximum. If the sum of the per-queue
2358 maximums exceeds the aggregate maximum, then the number of active I/Os
2359 may reach \fBzfs_vdev_max_active\fR, in which case no further I/Os will
2360 be issued regardless of whether all per-queue minimums have been met.
2362 For many physical devices, throughput increases with the number of
2363 concurrent operations, but latency typically suffers. Further, physical
2364 devices typically have a limit at which more concurrent operations have no
2365 effect on throughput or can actually cause it to decrease.
2367 The scheduler selects the next operation to issue by first looking for an
2368 I/O class whose minimum has not been satisfied. Once all are satisfied and
2369 the aggregate maximum has not been hit, the scheduler looks for classes
2370 whose maximum has not been satisfied. Iteration through the I/O classes is
2371 done in the order specified above. No further operations are issued if the
2372 aggregate maximum number of concurrent operations has been hit or if there
2373 are no operations queued for an I/O class that has not hit its maximum.
2374 Every time an I/O is queued or an operation completes, the I/O scheduler
2375 looks for new operations to issue.
2377 In general, smaller max_active's will lead to lower latency of synchronous
2378 operations. Larger max_active's may lead to higher overall throughput,
2379 depending on underlying storage.
2381 The ratio of the queues' max_actives determines the balance of performance
2382 between reads, writes, and scrubs. E.g., increasing
2383 \fBzfs_vdev_scrub_max_active\fR will cause the scrub or resilver to complete
2384 more quickly, but reads and writes to have higher latency and lower throughput.
2386 All I/O classes have a fixed maximum number of outstanding operations
2387 except for the async write class. Asynchronous writes represent the data
2388 that is committed to stable storage during the syncing stage for
2389 transaction groups. Transaction groups enter the syncing state
2390 periodically so the number of queued async writes will quickly burst up
2391 and then bleed down to zero. Rather than servicing them as quickly as
2392 possible, the I/O scheduler changes the maximum number of active async
2393 write I/Os according to the amount of dirty data in the pool. Since
2394 both throughput and latency typically increase with the number of
2395 concurrent operations issued to physical devices, reducing the
2396 burstiness in the number of concurrent operations also stabilizes the
2397 response time of operations from other -- and in particular synchronous
2398 -- queues. In broad strokes, the I/O scheduler will issue more
2399 concurrent operations from the async write queue as there's more dirty
2404 The number of concurrent operations issued for the async write I/O class
2405 follows a piece-wise linear function defined by a few adjustable points.
2408 | o---------| <-- zfs_vdev_async_write_max_active
2415 |-------o | | <-- zfs_vdev_async_write_min_active
2416 0|_______^______|_________|
2417 0% | | 100% of zfs_dirty_data_max
2419 | `-- zfs_vdev_async_write_active_max_dirty_percent
2420 `--------- zfs_vdev_async_write_active_min_dirty_percent
2423 Until the amount of dirty data exceeds a minimum percentage of the dirty
2424 data allowed in the pool, the I/O scheduler will limit the number of
2425 concurrent operations to the minimum. As that threshold is crossed, the
2426 number of concurrent operations issued increases linearly to the maximum at
2427 the specified maximum percentage of the dirty data allowed in the pool.
2429 Ideally, the amount of dirty data on a busy pool will stay in the sloped
2430 part of the function between \fBzfs_vdev_async_write_active_min_dirty_percent\fR
2431 and \fBzfs_vdev_async_write_active_max_dirty_percent\fR. If it exceeds the
2432 maximum percentage, this indicates that the rate of incoming data is
2433 greater than the rate that the backend storage can handle. In this case, we
2434 must further throttle incoming writes, as described in the next section.
2436 .SH ZFS TRANSACTION DELAY
2437 We delay transactions when we've determined that the backend storage
2438 isn't able to accommodate the rate of incoming writes.
2440 If there is already a transaction waiting, we delay relative to when
2441 that transaction will finish waiting. This way the calculated delay time
2442 is independent of the number of threads concurrently executing
2445 If we are the only waiter, wait relative to when the transaction
2446 started, rather than the current time. This credits the transaction for
2447 "time already served", e.g. reading indirect blocks.
2449 The minimum time for a transaction to take is calculated as:
2451 min_time = zfs_delay_scale * (dirty - min) / (max - dirty)
2452 min_time is then capped at 100 milliseconds.
2455 The delay has two degrees of freedom that can be adjusted via tunables. The
2456 percentage of dirty data at which we start to delay is defined by
2457 \fBzfs_delay_min_dirty_percent\fR. This should typically be at or above
2458 \fBzfs_vdev_async_write_active_max_dirty_percent\fR so that we only start to
2459 delay after writing at full speed has failed to keep up with the incoming write
2460 rate. The scale of the curve is defined by \fBzfs_delay_scale\fR. Roughly speaking,
2461 this variable determines the amount of delay at the midpoint of the curve.
2465 10ms +-------------------------------------------------------------*+
2481 2ms + (midpoint) * +
2484 | zfs_delay_scale ----------> ******** |
2485 0 +-------------------------------------*********----------------+
2486 0% <- zfs_dirty_data_max -> 100%
2489 Note that since the delay is added to the outstanding time remaining on the
2490 most recent transaction, the delay is effectively the inverse of IOPS.
2491 Here the midpoint of 500us translates to 2000 IOPS. The shape of the curve
2492 was chosen such that small changes in the amount of accumulated dirty data
2493 in the first 3/4 of the curve yield relatively small differences in the
2496 The effects can be easier to understand when the amount of delay is
2497 represented on a log scale:
2501 100ms +-------------------------------------------------------------++
2510 + zfs_delay_scale ----------> ***** +
2521 +--------------------------------------------------------------+
2522 0% <- zfs_dirty_data_max -> 100%
2525 Note here that only as the amount of dirty data approaches its limit does
2526 the delay start to increase rapidly. The goal of a properly tuned system
2527 should be to keep the amount of dirty data out of that range by first
2528 ensuring that the appropriate limits are set for the I/O scheduler to reach
2529 optimal throughput on the backend storage, and then by changing the value
2530 of \fBzfs_delay_scale\fR to increase the steepness of the curve.
projects / mirror_zfs.git / blob