Improve zfs-module-parameters(5)

Various rewrites to the descriptions of module parameters. Corrects
spelling mistakes, makes descriptions them more user-friendly and
describes some ZFS quirks which should be understood before changing
parameter values.

Signed-off-by: DHE <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #4671

1 files changed, 95 insertions, 53 deletions

diff --git a/man/man5/zfs-module-parameters.5 b/man/man5/zfs-module-parameters.5
index 20022461c..58b8742ee 100644
--- a/man/man5/zfs-module-parameters.5
+++ b/man/man5/zfs-module-parameters.5
@@ -30,7 +30,8 @@ Description of the different parameters to the ZFS module.
 \fBl2arc_feed_again\fR (int)
 .ad
 .RS 12n
-Turbo L2ARC warmup
+Turbo L2ARC warm-up. When the L2ARC is cold the fill interval will be set as
+fast as possible.
 .sp
 Use \fB1\fR for yes (default) and \fB0\fR to disable.
 .RE
@@ -41,7 +42,8 @@ Use \fB1\fR for yes (default) and \fB0\fR to disable.
 \fBl2arc_feed_min_ms\fR (ulong)
 .ad
 .RS 12n
-Min feed interval in milliseconds
+Min feed interval in milliseconds. Requires \fBl2arc_feed_again=1\fR and only
+applicable in related situations.
 .sp
 Default value: \fB200\fR.
 .RE
@@ -63,7 +65,8 @@ Default value: \fB1\fR.
 \fBl2arc_headroom\fR (ulong)
 .ad
 .RS 12n
-Number of max device writes to precache
+How far through the ARC lists to search for L2ARC cacheable content, expressed
+as a multiplier of \fBl2arc_write_max\fR
 .sp
 Default value: \fB2\fR.
 .RE
@@ -74,7 +77,8 @@ Default value: \fB2\fR.
 \fBl2arc_headroom_boost\fR (ulong)
 .ad
 .RS 12n
-Compressed l2arc_headroom multiplier
+Scales \fBl2arc_headroom\fR by this percentage when L2ARC contents are being
+successfully compressed before writing. A value of 100 disables this feature.
 .sp
 Default value: \fB200\fR.
 .RE
@@ -110,7 +114,8 @@ Use \fB1\fR for yes and \fB0\fR for no (default).
 \fBl2arc_noprefetch\fR (int)
 .ad
 .RS 12n
-Skip caching prefetched buffers
+Do not write buffers to L2ARC if they were prefetched but not used by
+applications
 .sp
 Use \fB1\fR for yes (default) and \fB0\fR to disable.
 .RE
@@ -132,7 +137,8 @@ Use \fB1\fR for yes and \fB0\fR for no (default).
 \fBl2arc_write_boost\fR (ulong)
 .ad
 .RS 12n
-Extra write bytes during device warmup
+Cold L2ARC devices will have \fBl2arc_write_nax\fR increased by this amount
+while they remain cold.
 .sp
 Default value: \fB8,388,608\fR.
 .RE
@@ -266,7 +272,7 @@ configuration.  Pool administrators who understand the factors involved
 may wish to specify a more realistic inflation factor, particularly if
 they operate close to quota or capacity limits.
 .sp
-Default value: 24
+Default value: \fB24\fR.
 .RE
 
 .sp
@@ -283,7 +289,7 @@ blocks in the pool for verification.  If this parameter is set to 0,
 the traversal skips non-metadata blocks.  It can be toggled once the
 import has started to stop or start the traversal of non-metadata blocks.
 .sp
-Default value: 1
+Default value: \fB1\fR.
 .RE
 
 .sp
@@ -300,7 +306,7 @@ blocks in the pool for verification.  If this parameter is set to 0,
 the traversal is not performed.  It can be toggled once the import has
 started to stop or start the traversal.
 .sp
-Default value: 1
+Default value: \fB1\fR.
 .RE
 
 .sp
@@ -312,7 +318,7 @@ Default value: 1
 Maximum concurrent I/Os during the traversal performed during an "extreme
 rewind" (\fB-X\fR) pool import.
 .sp
-Default value: 10000
+Default value: \fB10000\fR.
 .RE
 
 .sp
@@ -328,7 +334,7 @@ It also limits the worst-case time to allocate space.  If we have
 less than this amount of free space, most ZPL operations (e.g. write,
 create) will return ENOSPC.
 .sp
-Default value: 5
+Default value: \fB5\fR.
 .RE
 
 .sp
@@ -410,7 +416,8 @@ Default value: \fB10\fR.
 \fBzfs_arc_grow_retry\fR (int)
 .ad
 .RS 12n
-Seconds before growing arc size
+After a memory pressure event the ARC will wait this many seconds before trying
+to resume growth
 .sp
 Default value: \fB5\fR.
 .RE
@@ -433,7 +440,12 @@ Default value: \fB10\fR.
 \fBzfs_arc_max\fR (ulong)
 .ad
 .RS 12n
-Max arc size
+Max arc size of ARC in bytes. If set to 0 then it will consume 1/2 of system
+RAM. This value must be at least 67108864 (64 megabytes).
+.sp
+This value can be changed dynamically with some caveats. It cannot be set back
+to 0 while running and reducing it below the current ARC size will not cause
+the ARC to shrink without memory pressure to induce shrinking.
 .sp
 Default value: \fB0\fR.
 .RE
@@ -450,6 +462,9 @@ be reclaimed even if the overall arc_c_max has not been reached.  This
 value defaults to 0 which indicates that 3/4 of the ARC may be used
 for meta data.
 .sp
+This value my be changed dynamically except that it cannot be set back to 0
+for 3/4 of the ARC; it must be set to an explicit value.
+.sp
 Default value: \fB0\fR.
 .RE
 
@@ -513,9 +528,10 @@ Default value: \fB100\fR.
 \fBzfs_arc_min_prefetch_lifespan\fR (int)
 .ad
 .RS 12n
-Min life of prefetch block
+Minimum time prefetched blocks are locked in the ARC, specified in jiffies.
+A value of 0 will default to 1 second.
 .sp
-Default value: \fB100\fR.
+Default value: \fB0\fR.
 .RE
 
 .sp
@@ -529,7 +545,7 @@ of lists for both data and meta data objects.  Locking is performed at
 the level of these "sub-lists".  This parameters controls the number of
 sub-lists per ARC state.
 .sp
-Default value: 1 or the number of on-online CPUs, whichever is greater
+Default value: \fR1\fB or the number of online CPUs, whichever is greater
 .RE
 
 .sp
@@ -652,7 +668,8 @@ Default value: \fB4M\fR.
 \fBzfs_dbuf_state_index\fR (int)
 .ad
 .RS 12n
-Calculate arc header index
+This feature is currently unused. It is normally used for controlling what
+reporting is available under /proc/spl/kstat/zfs.
 .sp
 Default value: \fB0\fR.
 .RE
@@ -663,7 +680,7 @@ Default value: \fB0\fR.
 \fBzfs_deadman_enabled\fR (int)
 .ad
 .RS 12n
-Enable deadman timer
+Enable deadman timer. See description below.
 .sp
 Use \fB1\fR for yes (default) and \fB0\fR to disable.
 .RE
@@ -785,7 +802,7 @@ time, and will be ignored if \fBzfs_dirty_data_max\fR is later changed.
 The parameter \fBzfs_dirty_data_max_max\fR takes precedence over this
 one. See the section "ZFS TRANSACTION DELAY".
 .sp
-Default value: 25
+Default value: \fN25\fR.
 .RE
 
 .sp
@@ -841,7 +858,7 @@ Default value: \fB100,000\fR.
 \fBzfs_vdev_async_read_max_active\fR (int)
 .ad
 .RS 12n
-Maxium asynchronous read I/Os active to each device.
+Maximum asynchronous read I/Os active to each device.
 See the section "ZFS I/O SCHEDULER".
 .sp
 Default value: \fB3\fR.
@@ -895,7 +912,7 @@ Default value: \fB30\fR.
 \fBzfs_vdev_async_write_max_active\fR (int)
 .ad
 .RS 12n
-Maxium asynchronous write I/Os active to each device.
+Maximum asynchronous write I/Os active to each device.
 See the section "ZFS I/O SCHEDULER".
 .sp
 Default value: \fB10\fR.
@@ -932,7 +949,7 @@ Default value: \fB1,000\fR.
 \fBzfs_vdev_scrub_max_active\fR (int)
 .ad
 .RS 12n
-Maxium scrub I/Os active to each device.
+Maximum scrub I/Os active to each device.
 See the section "ZFS I/O SCHEDULER".
 .sp
 Default value: \fB2\fR.
@@ -956,7 +973,7 @@ Default value: \fB1\fR.
 \fBzfs_vdev_sync_read_max_active\fR (int)
 .ad
 .RS 12n
-Maxium synchronous read I/Os active to each device.
+Maximum synchronous read I/Os active to each device.
 See the section "ZFS I/O SCHEDULER".
 .sp
 Default value: \fB10\fR.
@@ -980,7 +997,7 @@ Default value: \fB10\fR.
 \fBzfs_vdev_sync_write_max_active\fR (int)
 .ad
 .RS 12n
-Maxium synchronous write I/Os active to each device.
+Maximum synchronous write I/Os active to each device.
 See the section "ZFS I/O SCHEDULER".
 .sp
 Default value: \fB10\fR.
@@ -1125,7 +1142,8 @@ Default value: \fB0\fR.
 \fBzfs_free_min_time_ms\fR (int)
 .ad
 .RS 12n
-Min millisecs to free per txg
+During a \fRzfs destroy\fB operation using \fRfeature@async_destroy\fB a minimum
+of this much time will be spent working on freeing blocks per txg.
 .sp
 Default value: \fB1,000\fR.
 .RE
@@ -1136,7 +1154,8 @@ Default value: \fB1,000\fR.
 \fBzfs_immediate_write_sz\fR (long)
 .ad
 .RS 12n
-Largest data block to write to zil
+Largest data block to write to zil. Larger blocks will be treated as if the
+dataset being written to had the property setting \fRlogbias=throughput\fB.
 .sp
 Default value: \fB32,768\fR.
 .RE
@@ -1191,7 +1210,7 @@ Default value: \fB70\fR.
 .ad
 .RS 12n
 Metaslab groups are considered eligible for allocations if their
-fragmenation metric (measured as a percentage) is less than or equal to
+fragmentation metric (measured as a percentage) is less than or equal to
 this value. If a metaslab group exceeds this threshold then it will be
 skipped unless all metaslab groups within the metaslab class have also
 crossed this threshold.
@@ -1231,7 +1250,8 @@ Default value: \fB0\fR.
 \fBzfs_no_scrub_io\fR (int)
 .ad
 .RS 12n
-Set for no scrub I/O
+Set for no scrub I/O. This results in scrubs not actually scrubbing data and
+simply doing a metadata crawl of the pool instead.
 .sp
 Use \fB1\fR for yes and \fB0\fR for no (default).
 .RE
@@ -1242,7 +1262,7 @@ Use \fB1\fR for yes and \fB0\fR for no (default).
 \fBzfs_no_scrub_prefetch\fR (int)
 .ad
 .RS 12n
-Set for no scrub prefetching
+Set to disable block prefetching for scrubs.
 .sp
 Use \fB1\fR for yes and \fB0\fR for no (default).
 .RE
@@ -1253,7 +1273,8 @@ Use \fB1\fR for yes and \fB0\fR for no (default).
 \fBzfs_nocacheflush\fR (int)
 .ad
 .RS 12n
-Disable cache flushes
+Disable cache flush operations on disks when writing. Beware, this may cause
+corruption if disks re-order writes.
 .sp
 Use \fB1\fR for yes and \fB0\fR for no (default).
 .RE
@@ -1275,7 +1296,8 @@ Use \fB1\fR for yes (default) and \fB0\fR to disable.
 \fBzfs_pd_bytes_max\fR (int)
 .ad
 .RS 12n
-The number of bytes which should be prefetched.
+The number of bytes which should be prefetched during a pool traversal
+(eg: \fRzfs send\fB or other data crawling operations)
 .sp
 Default value: \fB52,428,800\fR.
 .RE
@@ -1311,9 +1333,10 @@ Default value: \fB1,048,576\fR.
 \fBzfs_read_history\fR (int)
 .ad
 .RS 12n
-Historic statistics for the last N reads
+Historic statistics for the last N reads will be available in
+\fR/proc/spl/kstat/zfs/POOLNAME/reads\fB
 .sp
-Default value: \fB0\fR.
+Default value: \fB0\fR (no data is kept).
 .RE
 
 .sp
@@ -1358,7 +1381,8 @@ Default value: \fB2\fR.
 \fBzfs_resilver_min_time_ms\fR (int)
 .ad
 .RS 12n
-Min millisecs to resilver per txg
+Resilvers are processed by the sync thread. While resilvering it will spend
+at least this much time working on a resilver between txg flushes.
 .sp
 Default value: \fB3,000\fR.
 .RE
@@ -1383,7 +1407,8 @@ Default value: \fB50\fR.
 \fBzfs_scan_min_time_ms\fR (int)
 .ad
 .RS 12n
-Min millisecs to scrub per txg
+Scrubs are processed by the sync thread. While scrubbing it will spend
+at least this much time working on a scrub between txg flushes.
 .sp
 Default value: \fB1,000\fR.
 .RE
@@ -1407,7 +1432,7 @@ Default value: \fB4\fR.
 \fBzfs_send_corrupt_data\fR (int)
 .ad
 .RS 12n
-Allow to send corrupt data (ignore read/checksum errors when sending data)
+Allow sending of corrupt data (ignore read/checksum errors when sending data)
 .sp
 Use \fB1\fR for yes and \fB0\fR for no (default).
 .RE
@@ -1418,7 +1443,7 @@ Use \fB1\fR for yes and \fB0\fR for no (default).
 \fBzfs_sync_pass_deferred_free\fR (int)
 .ad
 .RS 12n
-Defer frees starting in this pass
+Flushing of data to disk is done in passes. Defer frees starting in this pass
 .sp
 Default value: \fB2\fR.
 .RE
@@ -1440,7 +1465,7 @@ Default value: \fB5\fR.
 \fBzfs_sync_pass_rewrite\fR (int)
 .ad
 .RS 12n
-Rewrite new bps starting in this pass
+Rewrite new block pointers starting in this pass
 .sp
 Default value: \fB2\fR.
 .RE
@@ -1451,7 +1476,8 @@ Default value: \fB2\fR.
 \fBzfs_top_maxinflight\fR (int)
 .ad
 .RS 12n
-Max I/Os per top-level vdev during scrub or resilver operations.
+Max concurrent I/Os per top-level vdev (mirrors or raidz arrays) allowed during
+scrub or resilver operations.
 .sp
 Default value: \fB32\fR.
 .RE
@@ -1462,7 +1488,8 @@ Default value: \fB32\fR.
 \fBzfs_txg_history\fR (int)
 .ad
 .RS 12n
-Historic statistics for the last N txgs
+Historic statistics for the last N txgs will be available in
+\fR/proc/spl/kstat/zfs/POOLNAME/txgs\fB
 .sp
 Default value: \fB0\fR.
 .RE
@@ -1473,7 +1500,7 @@ Default value: \fB0\fR.
 \fBzfs_txg_timeout\fR (int)
 .ad
 .RS 12n
-Max seconds worth of delta per txg
+Flush dirty data to disk at least every N seconds (maximum txg duration)
 .sp
 Default value: \fB5\fR.
 .RE
@@ -1497,7 +1524,7 @@ Default value: \fB131,072\fR.
 .RS 12n
 Shift size to inflate reads too
 .sp
-Default value: \fB16\fR.
+Default value: \fB16\fR (effectively 65536).
 .RE
 
 .sp
@@ -1506,7 +1533,10 @@ Default value: \fB16\fR.
 \fBzfs_vdev_cache_max\fR (int)
 .ad
 .RS 12n
-Inflate reads small than max
+Inflate reads small than this value to meet the \fBzfs_vdev_cache_bshift\fR
+size.
+.sp
+Default value: \fB16384\fR.
 .RE
 
 .sp
@@ -1515,7 +1545,10 @@ Inflate reads small than max
 \fBzfs_vdev_cache_size\fR (int)
 .ad
 .RS 12n
-Total size of the per-disk cache
+Total size of the per-disk cache in bytes.
+.sp
+Currently this feature is disabled as it has been found to not be helpful
+for performance and in some cases harmful.
 .sp
 Default value: \fB0\fR.
 .RE
@@ -1596,7 +1629,8 @@ Default value: \fB1\fR.
 \fBzfs_vdev_read_gap_limit\fR (int)
 .ad
 .RS 12n
-Aggregate read I/O over gap
+Aggregate read I/O operations if the gap on-disk between them is within this
+threshold.
 .sp
 Default value: \fB32,768\fR.
 .RE
@@ -1607,7 +1641,7 @@ Default value: \fB32,768\fR.
 \fBzfs_vdev_scheduler\fR (charp)
 .ad
 .RS 12n
-I/O scheduler
+Set the Linux I/O scheduler on whole disk vdevs to this scheduler
 .sp
 Default value: \fBnoop\fR.
 .RE
@@ -1629,7 +1663,7 @@ Default value: \fB4,096\fR.
 \fBzfs_zevent_cols\fR (int)
 .ad
 .RS 12n
-Max event column width
+When zevents are logged to the console use this as the word wrap width.
 .sp
 Default value: \fB80\fR.
 .RE
@@ -1651,7 +1685,9 @@ Use \fB1\fR for yes and \fB0\fR for no (default).
 \fBzfs_zevent_len_max\fR (int)
 .ad
 .RS 12n
-Max event queue length
+Max event queue length. A value of 0 will result in a calculated value which
+increases with the number of CPUs in the system (minimum 64 events). Events
+in the queue can be viewed with the \fBzpool events\fR command.
 .sp
 Default value: \fB0\fR.
 .RE
@@ -1662,7 +1698,8 @@ Default value: \fB0\fR.
 \fBzil_replay_disable\fR (int)
 .ad
 .RS 12n
-Disable intent logging replay
+Disable intent logging replay. Can be disabled for recovery from corrupted
+ZIL
 .sp
 Use \fB1\fR for yes and \fB0\fR for no (default).
 .RE
@@ -1684,7 +1721,9 @@ Default value: \fB1,048,576\fR.
 \fBzio_delay_max\fR (int)
 .ad
 .RS 12n
-Max zio millisecond delay before posting event
+A zevent will be logged if a ZIO operation takes more than N milliseconds to
+complete. Note that this is only a logging facility, not a timeout on 
+operations.
 .sp
 Default value: \fB30,000\fR.
 .RE
@@ -1723,7 +1762,8 @@ Default value: \fB75\fR.
 \fBzvol_inhibit_dev\fR (uint)
 .ad
 .RS 12n
-Do not create zvol device nodes
+Do not create zvol device nodes. This may slightly improve startup time on
+systems with a very large number of zvols.
 .sp
 Use \fB1\fR for yes and \fB0\fR for no (default).
 .RE
@@ -1734,7 +1774,7 @@ Use \fB1\fR for yes and \fB0\fR for no (default).
 \fBzvol_major\fR (uint)
 .ad
 .RS 12n
-Major number for zvol device
+Major number for zvol block devices
 .sp
 Default value: \fB230\fR.
 .RE
@@ -1745,7 +1785,9 @@ Default value: \fB230\fR.
 \fBzvol_max_discard_blocks\fR (ulong)
 .ad
 .RS 12n
-Max number of blocks to discard at once
+Discard (aka TRIM) operations done on zvols will be done in batches of this
+many blocks, where block size is determined by the \fBvolblocksize\fR property
+of a zvol.
 .sp
 Default value: \fB16,384\fR.
 .RE