]>
Commit | Line | Data |
---|---|---|
29714574 TF |
1 | '\" te |
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. | |
7 | .\" | |
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" | |
16 | .SH NAME | |
17 | zfs\-module\-parameters \- ZFS module parameters | |
18 | .SH DESCRIPTION | |
19 | .sp | |
20 | .LP | |
21 | Description of the different parameters to the ZFS module. | |
22 | ||
23 | .SS "Module parameters" | |
24 | .sp | |
25 | .LP | |
26 | ||
27 | .sp | |
28 | .ne 2 | |
29 | .na | |
30 | \fBl2arc_feed_again\fR (int) | |
31 | .ad | |
32 | .RS 12n | |
33 | Turbo L2ARC warmup | |
34 | .sp | |
35 | Use \fB1\fR for yes (default) and \fB0\fR to disable. | |
36 | .RE | |
37 | ||
38 | .sp | |
39 | .ne 2 | |
40 | .na | |
41 | \fBl2arc_feed_min_ms\fR (ulong) | |
42 | .ad | |
43 | .RS 12n | |
44 | Min feed interval in milliseconds | |
45 | .sp | |
46 | Default value: \fB200\fR. | |
47 | .RE | |
48 | ||
49 | .sp | |
50 | .ne 2 | |
51 | .na | |
52 | \fBl2arc_feed_secs\fR (ulong) | |
53 | .ad | |
54 | .RS 12n | |
55 | Seconds between L2ARC writing | |
56 | .sp | |
57 | Default value: \fB1\fR. | |
58 | .RE | |
59 | ||
60 | .sp | |
61 | .ne 2 | |
62 | .na | |
63 | \fBl2arc_headroom\fR (ulong) | |
64 | .ad | |
65 | .RS 12n | |
66 | Number of max device writes to precache | |
67 | .sp | |
68 | Default value: \fB2\fR. | |
69 | .RE | |
70 | ||
71 | .sp | |
72 | .ne 2 | |
73 | .na | |
74 | \fBl2arc_headroom_boost\fR (ulong) | |
75 | .ad | |
76 | .RS 12n | |
77 | Compressed l2arc_headroom multiplier | |
78 | .sp | |
79 | Default value: \fB200\fR. | |
80 | .RE | |
81 | ||
82 | .sp | |
83 | .ne 2 | |
84 | .na | |
85 | \fBl2arc_nocompress\fR (int) | |
86 | .ad | |
87 | .RS 12n | |
88 | Skip compressing L2ARC buffers | |
89 | .sp | |
90 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
91 | .RE | |
92 | ||
93 | .sp | |
94 | .ne 2 | |
95 | .na | |
96 | \fBl2arc_noprefetch\fR (int) | |
97 | .ad | |
98 | .RS 12n | |
99 | Skip caching prefetched buffers | |
100 | .sp | |
101 | Use \fB1\fR for yes (default) and \fB0\fR to disable. | |
102 | .RE | |
103 | ||
104 | .sp | |
105 | .ne 2 | |
106 | .na | |
107 | \fBl2arc_norw\fR (int) | |
108 | .ad | |
109 | .RS 12n | |
110 | No reads during writes | |
111 | .sp | |
112 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
113 | .RE | |
114 | ||
115 | .sp | |
116 | .ne 2 | |
117 | .na | |
118 | \fBl2arc_write_boost\fR (ulong) | |
119 | .ad | |
120 | .RS 12n | |
121 | Extra write bytes during device warmup | |
122 | .sp | |
123 | Default value: \fB8,388,608\fR. | |
124 | .RE | |
125 | ||
126 | .sp | |
127 | .ne 2 | |
128 | .na | |
129 | \fBl2arc_write_max\fR (ulong) | |
130 | .ad | |
131 | .RS 12n | |
132 | Max write bytes per interval | |
133 | .sp | |
134 | Default value: \fB8,388,608\fR. | |
135 | .RE | |
136 | ||
f3a7f661 GW |
137 | .sp |
138 | .ne 2 | |
139 | .na | |
140 | \fBmetaslab_bias_enabled\fR (int) | |
141 | .ad | |
142 | .RS 12n | |
143 | Enable metaslab group biasing based on its vdev's over- or under-utilization | |
144 | relative to the pool. | |
145 | .sp | |
146 | Use \fB1\fR for yes (default) and \fB0\fR for no. | |
147 | .RE | |
148 | ||
29714574 TF |
149 | .sp |
150 | .ne 2 | |
151 | .na | |
aa7d06a9 | 152 | \fBmetaslab_debug_load\fR (int) |
29714574 TF |
153 | .ad |
154 | .RS 12n | |
aa7d06a9 GW |
155 | Load all metaslabs during pool import. |
156 | .sp | |
157 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
158 | .RE | |
159 | ||
160 | .sp | |
161 | .ne 2 | |
162 | .na | |
163 | \fBmetaslab_debug_unload\fR (int) | |
164 | .ad | |
165 | .RS 12n | |
166 | Prevent metaslabs from being unloaded. | |
29714574 TF |
167 | .sp |
168 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
169 | .RE | |
170 | ||
f3a7f661 GW |
171 | .sp |
172 | .ne 2 | |
173 | .na | |
174 | \fBmetaslab_fragmentation_factor_enabled\fR (int) | |
175 | .ad | |
176 | .RS 12n | |
177 | Enable use of the fragmentation metric in computing metaslab weights. | |
178 | .sp | |
179 | Use \fB1\fR for yes (default) and \fB0\fR for no. | |
180 | .RE | |
181 | ||
182 | .sp | |
183 | .ne 2 | |
184 | .na | |
185 | \fBmetaslab_preload_enabled\fR (int) | |
186 | .ad | |
187 | .RS 12n | |
188 | Enable metaslab group preloading. | |
189 | .sp | |
190 | Use \fB1\fR for yes (default) and \fB0\fR for no. | |
191 | .RE | |
192 | ||
193 | .sp | |
194 | .ne 2 | |
195 | .na | |
196 | \fBmetaslab_lba_weighting_enabled\fR (int) | |
197 | .ad | |
198 | .RS 12n | |
199 | Give more weight to metaslabs with lower LBAs, assuming they have | |
200 | greater bandwidth as is typically the case on a modern constant | |
201 | angular velocity disk drive. | |
202 | .sp | |
203 | Use \fB1\fR for yes (default) and \fB0\fR for no. | |
204 | .RE | |
205 | ||
29714574 TF |
206 | .sp |
207 | .ne 2 | |
208 | .na | |
209 | \fBspa_config_path\fR (charp) | |
210 | .ad | |
211 | .RS 12n | |
212 | SPA config file | |
213 | .sp | |
214 | Default value: \fB/etc/zfs/zpool.cache\fR. | |
215 | .RE | |
216 | ||
e8b96c60 MA |
217 | .sp |
218 | .ne 2 | |
219 | .na | |
220 | \fBspa_asize_inflation\fR (int) | |
221 | .ad | |
222 | .RS 12n | |
223 | Multiplication factor used to estimate actual disk consumption from the | |
224 | size of data being written. The default value is a worst case estimate, | |
225 | but lower values may be valid for a given pool depending on its | |
226 | configuration. Pool administrators who understand the factors involved | |
227 | may wish to specify a more realistic inflation factor, particularly if | |
228 | they operate close to quota or capacity limits. | |
229 | .sp | |
230 | Default value: 24 | |
231 | .RE | |
232 | ||
dea377c0 MA |
233 | .sp |
234 | .ne 2 | |
235 | .na | |
236 | \fBspa_load_verify_data\fR (int) | |
237 | .ad | |
238 | .RS 12n | |
239 | Whether to traverse data blocks during an "extreme rewind" (\fB-X\fR) | |
240 | import. Use 0 to disable and 1 to enable. | |
241 | ||
242 | An extreme rewind import normally performs a full traversal of all | |
243 | blocks in the pool for verification. If this parameter is set to 0, | |
244 | the traversal skips non-metadata blocks. It can be toggled once the | |
245 | import has started to stop or start the traversal of non-metadata blocks. | |
246 | .sp | |
247 | Default value: 1 | |
248 | .RE | |
249 | ||
250 | .sp | |
251 | .ne 2 | |
252 | .na | |
253 | \fBspa_load_verify_metadata\fR (int) | |
254 | .ad | |
255 | .RS 12n | |
256 | Whether to traverse blocks during an "extreme rewind" (\fB-X\fR) | |
257 | pool import. Use 0 to disable and 1 to enable. | |
258 | ||
259 | An extreme rewind import normally performs a full traversal of all | |
260 | blocks in the pool for verification. If this parameter is set to 1, | |
261 | the traversal is not performed. It can be toggled once the import has | |
262 | started to stop or start the traversal. | |
263 | .sp | |
264 | Default value: 1 | |
265 | .RE | |
266 | ||
267 | .sp | |
268 | .ne 2 | |
269 | .na | |
270 | \fBspa_load_verify_maxinflight\fR (int) | |
271 | .ad | |
272 | .RS 12n | |
273 | Maximum concurrent I/Os during the traversal performed during an "extreme | |
274 | rewind" (\fB-X\fR) pool import. | |
275 | .sp | |
276 | Default value: 10000 | |
277 | .RE | |
278 | ||
29714574 TF |
279 | .sp |
280 | .ne 2 | |
281 | .na | |
282 | \fBzfetch_array_rd_sz\fR (ulong) | |
283 | .ad | |
284 | .RS 12n | |
27b293be | 285 | If prefetching is enabled, disable prefetching for reads larger than this size. |
29714574 TF |
286 | .sp |
287 | Default value: \fB1,048,576\fR. | |
288 | .RE | |
289 | ||
290 | .sp | |
291 | .ne 2 | |
292 | .na | |
293 | \fBzfetch_block_cap\fR (uint) | |
294 | .ad | |
295 | .RS 12n | |
27b293be | 296 | Max number of blocks to prefetch at a time |
29714574 TF |
297 | .sp |
298 | Default value: \fB256\fR. | |
299 | .RE | |
300 | ||
301 | .sp | |
302 | .ne 2 | |
303 | .na | |
304 | \fBzfetch_max_streams\fR (uint) | |
305 | .ad | |
306 | .RS 12n | |
27b293be | 307 | Max number of streams per zfetch (prefetch streams per file). |
29714574 TF |
308 | .sp |
309 | Default value: \fB8\fR. | |
310 | .RE | |
311 | ||
312 | .sp | |
313 | .ne 2 | |
314 | .na | |
315 | \fBzfetch_min_sec_reap\fR (uint) | |
316 | .ad | |
317 | .RS 12n | |
27b293be | 318 | Min time before an active prefetch stream can be reclaimed |
29714574 TF |
319 | .sp |
320 | Default value: \fB2\fR. | |
321 | .RE | |
322 | ||
49ddb315 MA |
323 | .sp |
324 | .ne 2 | |
325 | .na | |
326 | \fBzfs_arc_average_blocksize\fR (int) | |
327 | .ad | |
328 | .RS 12n | |
329 | The ARC's buffer hash table is sized based on the assumption of an average | |
330 | block size of \fBzfs_arc_average_blocksize\fR (default 8K). This works out | |
331 | to roughly 1MB of hash table per 1GB of physical memory with 8-byte pointers. | |
332 | For configurations with a known larger average block size this value can be | |
333 | increased to reduce the memory footprint. | |
334 | ||
335 | .sp | |
336 | Default value: \fB8192\fR. | |
337 | .RE | |
338 | ||
29714574 TF |
339 | .sp |
340 | .ne 2 | |
341 | .na | |
342 | \fBzfs_arc_grow_retry\fR (int) | |
343 | .ad | |
344 | .RS 12n | |
345 | Seconds before growing arc size | |
346 | .sp | |
347 | Default value: \fB5\fR. | |
348 | .RE | |
349 | ||
350 | .sp | |
351 | .ne 2 | |
352 | .na | |
353 | \fBzfs_arc_max\fR (ulong) | |
354 | .ad | |
355 | .RS 12n | |
356 | Max arc size | |
357 | .sp | |
358 | Default value: \fB0\fR. | |
359 | .RE | |
360 | ||
361 | .sp | |
362 | .ne 2 | |
363 | .na | |
364 | \fBzfs_arc_memory_throttle_disable\fR (int) | |
365 | .ad | |
366 | .RS 12n | |
367 | Disable memory throttle | |
368 | .sp | |
369 | Use \fB1\fR for yes (default) and \fB0\fR to disable. | |
370 | .RE | |
371 | ||
372 | .sp | |
373 | .ne 2 | |
374 | .na | |
375 | \fBzfs_arc_meta_limit\fR (ulong) | |
376 | .ad | |
377 | .RS 12n | |
378 | Meta limit for arc size | |
379 | .sp | |
380 | Default value: \fB0\fR. | |
381 | .RE | |
382 | ||
383 | .sp | |
384 | .ne 2 | |
385 | .na | |
386 | \fBzfs_arc_meta_prune\fR (int) | |
387 | .ad | |
388 | .RS 12n | |
389 | Bytes of meta data to prune | |
390 | .sp | |
391 | Default value: \fB1,048,576\fR. | |
392 | .RE | |
393 | ||
394 | .sp | |
395 | .ne 2 | |
396 | .na | |
397 | \fBzfs_arc_min\fR (ulong) | |
398 | .ad | |
399 | .RS 12n | |
400 | Min arc size | |
401 | .sp | |
402 | Default value: \fB100\fR. | |
403 | .RE | |
404 | ||
405 | .sp | |
406 | .ne 2 | |
407 | .na | |
408 | \fBzfs_arc_min_prefetch_lifespan\fR (int) | |
409 | .ad | |
410 | .RS 12n | |
411 | Min life of prefetch block | |
412 | .sp | |
413 | Default value: \fB100\fR. | |
414 | .RE | |
415 | ||
89c8cac4 PS |
416 | .sp |
417 | .ne 2 | |
418 | .na | |
419 | \fBzfs_arc_p_aggressive_disable\fR (int) | |
420 | .ad | |
421 | .RS 12n | |
422 | Disable aggressive arc_p growth | |
423 | .sp | |
424 | Use \fB1\fR for yes (default) and \fB0\fR to disable. | |
425 | .RE | |
426 | ||
62422785 PS |
427 | .sp |
428 | .ne 2 | |
429 | .na | |
430 | \fBzfs_arc_p_dampener_disable\fR (int) | |
431 | .ad | |
432 | .RS 12n | |
433 | Disable arc_p adapt dampener | |
434 | .sp | |
435 | Use \fB1\fR for yes (default) and \fB0\fR to disable. | |
436 | .RE | |
437 | ||
29714574 TF |
438 | .sp |
439 | .ne 2 | |
440 | .na | |
441 | \fBzfs_arc_shrink_shift\fR (int) | |
442 | .ad | |
443 | .RS 12n | |
444 | log2(fraction of arc to reclaim) | |
445 | .sp | |
446 | Default value: \fB5\fR. | |
447 | .RE | |
448 | ||
449 | .sp | |
450 | .ne 2 | |
451 | .na | |
452 | \fBzfs_autoimport_disable\fR (int) | |
453 | .ad | |
454 | .RS 12n | |
27b293be | 455 | Disable pool import at module load by ignoring the cache file (typically \fB/etc/zfs/zpool.cache\fR). |
29714574 TF |
456 | .sp |
457 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
458 | .RE | |
459 | ||
460 | .sp | |
461 | .ne 2 | |
462 | .na | |
463 | \fBzfs_dbuf_state_index\fR (int) | |
464 | .ad | |
465 | .RS 12n | |
466 | Calculate arc header index | |
467 | .sp | |
468 | Default value: \fB0\fR. | |
469 | .RE | |
470 | ||
471 | .sp | |
472 | .ne 2 | |
473 | .na | |
474 | \fBzfs_deadman_enabled\fR (int) | |
475 | .ad | |
476 | .RS 12n | |
477 | Enable deadman timer | |
478 | .sp | |
479 | Use \fB1\fR for yes (default) and \fB0\fR to disable. | |
480 | .RE | |
481 | ||
482 | .sp | |
483 | .ne 2 | |
484 | .na | |
e8b96c60 | 485 | \fBzfs_deadman_synctime_ms\fR (ulong) |
29714574 TF |
486 | .ad |
487 | .RS 12n | |
e8b96c60 MA |
488 | Expiration time in milliseconds. This value has two meanings. First it is |
489 | used to determine when the spa_deadman() logic should fire. By default the | |
490 | spa_deadman() will fire if spa_sync() has not completed in 1000 seconds. | |
491 | Secondly, the value determines if an I/O is considered "hung". Any I/O that | |
492 | has not completed in zfs_deadman_synctime_ms is considered "hung" resulting | |
493 | in a zevent being logged. | |
29714574 | 494 | .sp |
e8b96c60 | 495 | Default value: \fB1,000,000\fR. |
29714574 TF |
496 | .RE |
497 | ||
498 | .sp | |
499 | .ne 2 | |
500 | .na | |
501 | \fBzfs_dedup_prefetch\fR (int) | |
502 | .ad | |
503 | .RS 12n | |
504 | Enable prefetching dedup-ed blks | |
505 | .sp | |
506 | Use \fB1\fR for yes (default) and \fB0\fR to disable. | |
507 | .RE | |
508 | ||
e8b96c60 MA |
509 | .sp |
510 | .ne 2 | |
511 | .na | |
512 | \fBzfs_delay_min_dirty_percent\fR (int) | |
513 | .ad | |
514 | .RS 12n | |
515 | Start to delay each transaction once there is this amount of dirty data, | |
516 | expressed as a percentage of \fBzfs_dirty_data_max\fR. | |
517 | This value should be >= zfs_vdev_async_write_active_max_dirty_percent. | |
518 | See the section "ZFS TRANSACTION DELAY". | |
519 | .sp | |
520 | Default value: \fB60\fR. | |
521 | .RE | |
522 | ||
523 | .sp | |
524 | .ne 2 | |
525 | .na | |
526 | \fBzfs_delay_scale\fR (int) | |
527 | .ad | |
528 | .RS 12n | |
529 | This controls how quickly the transaction delay approaches infinity. | |
530 | Larger values cause longer delays for a given amount of dirty data. | |
531 | .sp | |
532 | For the smoothest delay, this value should be about 1 billion divided | |
533 | by the maximum number of operations per second. This will smoothly | |
534 | handle between 10x and 1/10th this number. | |
535 | .sp | |
536 | See the section "ZFS TRANSACTION DELAY". | |
537 | .sp | |
538 | Note: \fBzfs_delay_scale\fR * \fBzfs_dirty_data_max\fR must be < 2^64. | |
539 | .sp | |
540 | Default value: \fB500,000\fR. | |
541 | .RE | |
542 | ||
543 | .sp | |
544 | .ne 2 | |
545 | .na | |
546 | \fBzfs_dirty_data_max\fR (int) | |
547 | .ad | |
548 | .RS 12n | |
549 | Determines the dirty space limit in bytes. Once this limit is exceeded, new | |
550 | writes are halted until space frees up. This parameter takes precedence | |
551 | over \fBzfs_dirty_data_max_percent\fR. | |
552 | See the section "ZFS TRANSACTION DELAY". | |
553 | .sp | |
554 | Default value: 10 percent of all memory, capped at \fBzfs_dirty_data_max_max\fR. | |
555 | .RE | |
556 | ||
557 | .sp | |
558 | .ne 2 | |
559 | .na | |
560 | \fBzfs_dirty_data_max_max\fR (int) | |
561 | .ad | |
562 | .RS 12n | |
563 | Maximum allowable value of \fBzfs_dirty_data_max\fR, expressed in bytes. | |
564 | This limit is only enforced at module load time, and will be ignored if | |
565 | \fBzfs_dirty_data_max\fR is later changed. This parameter takes | |
566 | precedence over \fBzfs_dirty_data_max_max_percent\fR. See the section | |
567 | "ZFS TRANSACTION DELAY". | |
568 | .sp | |
569 | Default value: 25% of physical RAM. | |
570 | .RE | |
571 | ||
572 | .sp | |
573 | .ne 2 | |
574 | .na | |
575 | \fBzfs_dirty_data_max_max_percent\fR (int) | |
576 | .ad | |
577 | .RS 12n | |
578 | Maximum allowable value of \fBzfs_dirty_data_max\fR, expressed as a | |
579 | percentage of physical RAM. This limit is only enforced at module load | |
580 | time, and will be ignored if \fBzfs_dirty_data_max\fR is later changed. | |
581 | The parameter \fBzfs_dirty_data_max_max\fR takes precedence over this | |
582 | one. See the section "ZFS TRANSACTION DELAY". | |
583 | .sp | |
584 | Default value: 25 | |
585 | .RE | |
586 | ||
587 | .sp | |
588 | .ne 2 | |
589 | .na | |
590 | \fBzfs_dirty_data_max_percent\fR (int) | |
591 | .ad | |
592 | .RS 12n | |
593 | Determines the dirty space limit, expressed as a percentage of all | |
594 | memory. Once this limit is exceeded, new writes are halted until space frees | |
595 | up. The parameter \fBzfs_dirty_data_max\fR takes precedence over this | |
596 | one. See the section "ZFS TRANSACTION DELAY". | |
597 | .sp | |
598 | Default value: 10%, subject to \fBzfs_dirty_data_max_max\fR. | |
599 | .RE | |
600 | ||
601 | .sp | |
602 | .ne 2 | |
603 | .na | |
604 | \fBzfs_dirty_data_sync\fR (int) | |
605 | .ad | |
606 | .RS 12n | |
607 | Start syncing out a transaction group if there is at least this much dirty data. | |
608 | .sp | |
609 | Default value: \fB67,108,864\fR. | |
610 | .RE | |
611 | ||
612 | .sp | |
613 | .ne 2 | |
614 | .na | |
615 | \fBzfs_vdev_async_read_max_active\fR (int) | |
616 | .ad | |
617 | .RS 12n | |
618 | Maxium asynchronous read I/Os active to each device. | |
619 | See the section "ZFS I/O SCHEDULER". | |
620 | .sp | |
621 | Default value: \fB3\fR. | |
622 | .RE | |
623 | ||
624 | .sp | |
625 | .ne 2 | |
626 | .na | |
627 | \fBzfs_vdev_async_read_min_active\fR (int) | |
628 | .ad | |
629 | .RS 12n | |
630 | Minimum asynchronous read I/Os active to each device. | |
631 | See the section "ZFS I/O SCHEDULER". | |
632 | .sp | |
633 | Default value: \fB1\fR. | |
634 | .RE | |
635 | ||
636 | .sp | |
637 | .ne 2 | |
638 | .na | |
639 | \fBzfs_vdev_async_write_active_max_dirty_percent\fR (int) | |
640 | .ad | |
641 | .RS 12n | |
642 | When the pool has more than | |
643 | \fBzfs_vdev_async_write_active_max_dirty_percent\fR dirty data, use | |
644 | \fBzfs_vdev_async_write_max_active\fR to limit active async writes. If | |
645 | the dirty data is between min and max, the active I/O limit is linearly | |
646 | interpolated. See the section "ZFS I/O SCHEDULER". | |
647 | .sp | |
648 | Default value: \fB60\fR. | |
649 | .RE | |
650 | ||
651 | .sp | |
652 | .ne 2 | |
653 | .na | |
654 | \fBzfs_vdev_async_write_active_min_dirty_percent\fR (int) | |
655 | .ad | |
656 | .RS 12n | |
657 | When the pool has less than | |
658 | \fBzfs_vdev_async_write_active_min_dirty_percent\fR dirty data, use | |
659 | \fBzfs_vdev_async_write_min_active\fR to limit active async writes. If | |
660 | the dirty data is between min and max, the active I/O limit is linearly | |
661 | interpolated. See the section "ZFS I/O SCHEDULER". | |
662 | .sp | |
663 | Default value: \fB30\fR. | |
664 | .RE | |
665 | ||
666 | .sp | |
667 | .ne 2 | |
668 | .na | |
669 | \fBzfs_vdev_async_write_max_active\fR (int) | |
670 | .ad | |
671 | .RS 12n | |
672 | Maxium asynchronous write I/Os active to each device. | |
673 | See the section "ZFS I/O SCHEDULER". | |
674 | .sp | |
675 | Default value: \fB10\fR. | |
676 | .RE | |
677 | ||
678 | .sp | |
679 | .ne 2 | |
680 | .na | |
681 | \fBzfs_vdev_async_write_min_active\fR (int) | |
682 | .ad | |
683 | .RS 12n | |
684 | Minimum asynchronous write I/Os active to each device. | |
685 | See the section "ZFS I/O SCHEDULER". | |
686 | .sp | |
687 | Default value: \fB1\fR. | |
688 | .RE | |
689 | ||
690 | .sp | |
691 | .ne 2 | |
692 | .na | |
693 | \fBzfs_vdev_max_active\fR (int) | |
694 | .ad | |
695 | .RS 12n | |
696 | The maximum number of I/Os active to each device. Ideally, this will be >= | |
697 | the sum of each queue's max_active. It must be at least the sum of each | |
698 | queue's min_active. See the section "ZFS I/O SCHEDULER". | |
699 | .sp | |
700 | Default value: \fB1,000\fR. | |
701 | .RE | |
702 | ||
703 | .sp | |
704 | .ne 2 | |
705 | .na | |
706 | \fBzfs_vdev_scrub_max_active\fR (int) | |
707 | .ad | |
708 | .RS 12n | |
709 | Maxium scrub I/Os active to each device. | |
710 | See the section "ZFS I/O SCHEDULER". | |
711 | .sp | |
712 | Default value: \fB2\fR. | |
713 | .RE | |
714 | ||
715 | .sp | |
716 | .ne 2 | |
717 | .na | |
718 | \fBzfs_vdev_scrub_min_active\fR (int) | |
719 | .ad | |
720 | .RS 12n | |
721 | Minimum scrub I/Os active to each device. | |
722 | See the section "ZFS I/O SCHEDULER". | |
723 | .sp | |
724 | Default value: \fB1\fR. | |
725 | .RE | |
726 | ||
727 | .sp | |
728 | .ne 2 | |
729 | .na | |
730 | \fBzfs_vdev_sync_read_max_active\fR (int) | |
731 | .ad | |
732 | .RS 12n | |
733 | Maxium synchronous read I/Os active to each device. | |
734 | See the section "ZFS I/O SCHEDULER". | |
735 | .sp | |
736 | Default value: \fB10\fR. | |
737 | .RE | |
738 | ||
739 | .sp | |
740 | .ne 2 | |
741 | .na | |
742 | \fBzfs_vdev_sync_read_min_active\fR (int) | |
743 | .ad | |
744 | .RS 12n | |
745 | Minimum synchronous read I/Os active to each device. | |
746 | See the section "ZFS I/O SCHEDULER". | |
747 | .sp | |
748 | Default value: \fB10\fR. | |
749 | .RE | |
750 | ||
751 | .sp | |
752 | .ne 2 | |
753 | .na | |
754 | \fBzfs_vdev_sync_write_max_active\fR (int) | |
755 | .ad | |
756 | .RS 12n | |
757 | Maxium synchronous write I/Os active to each device. | |
758 | See the section "ZFS I/O SCHEDULER". | |
759 | .sp | |
760 | Default value: \fB10\fR. | |
761 | .RE | |
762 | ||
763 | .sp | |
764 | .ne 2 | |
765 | .na | |
766 | \fBzfs_vdev_sync_write_min_active\fR (int) | |
767 | .ad | |
768 | .RS 12n | |
769 | Minimum synchronous write I/Os active to each device. | |
770 | See the section "ZFS I/O SCHEDULER". | |
771 | .sp | |
772 | Default value: \fB10\fR. | |
773 | .RE | |
774 | ||
29714574 TF |
775 | .sp |
776 | .ne 2 | |
777 | .na | |
778 | \fBzfs_disable_dup_eviction\fR (int) | |
779 | .ad | |
780 | .RS 12n | |
781 | Disable duplicate buffer eviction | |
782 | .sp | |
783 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
784 | .RE | |
785 | ||
786 | .sp | |
787 | .ne 2 | |
788 | .na | |
789 | \fBzfs_expire_snapshot\fR (int) | |
790 | .ad | |
791 | .RS 12n | |
792 | Seconds to expire .zfs/snapshot | |
793 | .sp | |
794 | Default value: \fB300\fR. | |
795 | .RE | |
796 | ||
797 | .sp | |
798 | .ne 2 | |
799 | .na | |
800 | \fBzfs_flags\fR (int) | |
801 | .ad | |
802 | .RS 12n | |
803 | Set additional debugging flags | |
804 | .sp | |
805 | Default value: \fB1\fR. | |
806 | .RE | |
807 | ||
fbeddd60 MA |
808 | .sp |
809 | .ne 2 | |
810 | .na | |
811 | \fBzfs_free_leak_on_eio\fR (int) | |
812 | .ad | |
813 | .RS 12n | |
814 | If destroy encounters an EIO while reading metadata (e.g. indirect | |
815 | blocks), space referenced by the missing metadata can not be freed. | |
816 | Normally this causes the background destroy to become "stalled", as | |
817 | it is unable to make forward progress. While in this stalled state, | |
818 | all remaining space to free from the error-encountering filesystem is | |
819 | "temporarily leaked". Set this flag to cause it to ignore the EIO, | |
820 | permanently leak the space from indirect blocks that can not be read, | |
821 | and continue to free everything else that it can. | |
822 | ||
823 | The default, "stalling" behavior is useful if the storage partially | |
824 | fails (i.e. some but not all i/os fail), and then later recovers. In | |
825 | this case, we will be able to continue pool operations while it is | |
826 | partially failed, and when it recovers, we can continue to free the | |
827 | space, with no leaks. However, note that this case is actually | |
828 | fairly rare. | |
829 | ||
830 | Typically pools either (a) fail completely (but perhaps temporarily, | |
831 | e.g. a top-level vdev going offline), or (b) have localized, | |
832 | permanent errors (e.g. disk returns the wrong data due to bit flip or | |
833 | firmware bug). In case (a), this setting does not matter because the | |
834 | pool will be suspended and the sync thread will not be able to make | |
835 | forward progress regardless. In case (b), because the error is | |
836 | permanent, the best we can do is leak the minimum amount of space, | |
837 | which is what setting this flag will do. Therefore, it is reasonable | |
838 | for this flag to normally be set, but we chose the more conservative | |
839 | approach of not setting it, so that there is no possibility of | |
840 | leaking space in the "partial temporary" failure case. | |
841 | .sp | |
842 | Default value: \fB0\fR. | |
843 | .RE | |
844 | ||
29714574 TF |
845 | .sp |
846 | .ne 2 | |
847 | .na | |
848 | \fBzfs_free_min_time_ms\fR (int) | |
849 | .ad | |
850 | .RS 12n | |
851 | Min millisecs to free per txg | |
852 | .sp | |
853 | Default value: \fB1,000\fR. | |
854 | .RE | |
855 | ||
856 | .sp | |
857 | .ne 2 | |
858 | .na | |
859 | \fBzfs_immediate_write_sz\fR (long) | |
860 | .ad | |
861 | .RS 12n | |
862 | Largest data block to write to zil | |
863 | .sp | |
864 | Default value: \fB32,768\fR. | |
865 | .RE | |
866 | ||
867 | .sp | |
868 | .ne 2 | |
869 | .na | |
870 | \fBzfs_mdcomp_disable\fR (int) | |
871 | .ad | |
872 | .RS 12n | |
873 | Disable meta data compression | |
874 | .sp | |
875 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
876 | .RE | |
877 | ||
f3a7f661 GW |
878 | .sp |
879 | .ne 2 | |
880 | .na | |
881 | \fBzfs_metaslab_fragmentation_threshold\fR (int) | |
882 | .ad | |
883 | .RS 12n | |
884 | Allow metaslabs to keep their active state as long as their fragmentation | |
885 | percentage is less than or equal to this value. An active metaslab that | |
886 | exceeds this threshold will no longer keep its active status allowing | |
887 | better metaslabs to be selected. | |
888 | .sp | |
889 | Default value: \fB70\fR. | |
890 | .RE | |
891 | ||
892 | .sp | |
893 | .ne 2 | |
894 | .na | |
895 | \fBzfs_mg_fragmentation_threshold\fR (int) | |
896 | .ad | |
897 | .RS 12n | |
898 | Metaslab groups are considered eligible for allocations if their | |
899 | fragmenation metric (measured as a percentage) is less than or equal to | |
900 | this value. If a metaslab group exceeds this threshold then it will be | |
901 | skipped unless all metaslab groups within the metaslab class have also | |
902 | crossed this threshold. | |
903 | .sp | |
904 | Default value: \fB85\fR. | |
905 | .RE | |
906 | ||
f4a4046b TC |
907 | .sp |
908 | .ne 2 | |
909 | .na | |
910 | \fBzfs_mg_noalloc_threshold\fR (int) | |
911 | .ad | |
912 | .RS 12n | |
913 | Defines a threshold at which metaslab groups should be eligible for | |
914 | allocations. The value is expressed as a percentage of free space | |
915 | beyond which a metaslab group is always eligible for allocations. | |
916 | If a metaslab group's free space is less than or equal to the | |
917 | the threshold, the allocator will avoid allocating to that group | |
918 | unless all groups in the pool have reached the threshold. Once all | |
919 | groups have reached the threshold, all groups are allowed to accept | |
920 | allocations. The default value of 0 disables the feature and causes | |
921 | all metaslab groups to be eligible for allocations. | |
922 | ||
923 | This parameter allows to deal with pools having heavily imbalanced | |
924 | vdevs such as would be the case when a new vdev has been added. | |
925 | Setting the threshold to a non-zero percentage will stop allocations | |
926 | from being made to vdevs that aren't filled to the specified percentage | |
927 | and allow lesser filled vdevs to acquire more allocations than they | |
928 | otherwise would under the old \fBzfs_mg_alloc_failures\fR facility. | |
929 | .sp | |
930 | Default value: \fB0\fR. | |
931 | .RE | |
932 | ||
29714574 TF |
933 | .sp |
934 | .ne 2 | |
935 | .na | |
936 | \fBzfs_no_scrub_io\fR (int) | |
937 | .ad | |
938 | .RS 12n | |
939 | Set for no scrub I/O | |
940 | .sp | |
941 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
942 | .RE | |
943 | ||
944 | .sp | |
945 | .ne 2 | |
946 | .na | |
947 | \fBzfs_no_scrub_prefetch\fR (int) | |
948 | .ad | |
949 | .RS 12n | |
950 | Set for no scrub prefetching | |
951 | .sp | |
952 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
953 | .RE | |
954 | ||
29714574 TF |
955 | .sp |
956 | .ne 2 | |
957 | .na | |
958 | \fBzfs_nocacheflush\fR (int) | |
959 | .ad | |
960 | .RS 12n | |
961 | Disable cache flushes | |
962 | .sp | |
963 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
964 | .RE | |
965 | ||
966 | .sp | |
967 | .ne 2 | |
968 | .na | |
969 | \fBzfs_nopwrite_enabled\fR (int) | |
970 | .ad | |
971 | .RS 12n | |
972 | Enable NOP writes | |
973 | .sp | |
974 | Use \fB1\fR for yes (default) and \fB0\fR to disable. | |
975 | .RE | |
976 | ||
977 | .sp | |
978 | .ne 2 | |
979 | .na | |
980 | \fBzfs_pd_blks_max\fR (int) | |
981 | .ad | |
982 | .RS 12n | |
983 | Max number of blocks to prefetch | |
984 | .sp | |
985 | Default value: \fB100\fR. | |
986 | .RE | |
987 | ||
988 | .sp | |
989 | .ne 2 | |
990 | .na | |
991 | \fBzfs_prefetch_disable\fR (int) | |
992 | .ad | |
993 | .RS 12n | |
994 | Disable all ZFS prefetching | |
995 | .sp | |
996 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
997 | .RE | |
998 | ||
999 | .sp | |
1000 | .ne 2 | |
1001 | .na | |
1002 | \fBzfs_read_chunk_size\fR (long) | |
1003 | .ad | |
1004 | .RS 12n | |
1005 | Bytes to read per chunk | |
1006 | .sp | |
1007 | Default value: \fB1,048,576\fR. | |
1008 | .RE | |
1009 | ||
1010 | .sp | |
1011 | .ne 2 | |
1012 | .na | |
1013 | \fBzfs_read_history\fR (int) | |
1014 | .ad | |
1015 | .RS 12n | |
1016 | Historic statistics for the last N reads | |
1017 | .sp | |
1018 | Default value: \fB0\fR. | |
1019 | .RE | |
1020 | ||
1021 | .sp | |
1022 | .ne 2 | |
1023 | .na | |
1024 | \fBzfs_read_history_hits\fR (int) | |
1025 | .ad | |
1026 | .RS 12n | |
1027 | Include cache hits in read history | |
1028 | .sp | |
1029 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
1030 | .RE | |
1031 | ||
1032 | .sp | |
1033 | .ne 2 | |
1034 | .na | |
1035 | \fBzfs_recover\fR (int) | |
1036 | .ad | |
1037 | .RS 12n | |
1038 | Set to attempt to recover from fatal errors. This should only be used as a | |
1039 | last resort, as it typically results in leaked space, or worse. | |
1040 | .sp | |
1041 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
1042 | .RE | |
1043 | ||
1044 | .sp | |
1045 | .ne 2 | |
1046 | .na | |
1047 | \fBzfs_resilver_delay\fR (int) | |
1048 | .ad | |
1049 | .RS 12n | |
27b293be TC |
1050 | Number of ticks to delay prior to issuing a resilver I/O operation when |
1051 | a non-resilver or non-scrub I/O operation has occurred within the past | |
1052 | \fBzfs_scan_idle\fR ticks. | |
29714574 TF |
1053 | .sp |
1054 | Default value: \fB2\fR. | |
1055 | .RE | |
1056 | ||
1057 | .sp | |
1058 | .ne 2 | |
1059 | .na | |
1060 | \fBzfs_resilver_min_time_ms\fR (int) | |
1061 | .ad | |
1062 | .RS 12n | |
1063 | Min millisecs to resilver per txg | |
1064 | .sp | |
1065 | Default value: \fB3,000\fR. | |
1066 | .RE | |
1067 | ||
1068 | .sp | |
1069 | .ne 2 | |
1070 | .na | |
1071 | \fBzfs_scan_idle\fR (int) | |
1072 | .ad | |
1073 | .RS 12n | |
27b293be TC |
1074 | Idle window in clock ticks. During a scrub or a resilver, if |
1075 | a non-scrub or non-resilver I/O operation has occurred during this | |
1076 | window, the next scrub or resilver operation is delayed by, respectively | |
1077 | \fBzfs_scrub_delay\fR or \fBzfs_resilver_delay\fR ticks. | |
29714574 TF |
1078 | .sp |
1079 | Default value: \fB50\fR. | |
1080 | .RE | |
1081 | ||
1082 | .sp | |
1083 | .ne 2 | |
1084 | .na | |
1085 | \fBzfs_scan_min_time_ms\fR (int) | |
1086 | .ad | |
1087 | .RS 12n | |
1088 | Min millisecs to scrub per txg | |
1089 | .sp | |
1090 | Default value: \fB1,000\fR. | |
1091 | .RE | |
1092 | ||
1093 | .sp | |
1094 | .ne 2 | |
1095 | .na | |
1096 | \fBzfs_scrub_delay\fR (int) | |
1097 | .ad | |
1098 | .RS 12n | |
27b293be TC |
1099 | Number of ticks to delay prior to issuing a scrub I/O operation when |
1100 | a non-scrub or non-resilver I/O operation has occurred within the past | |
1101 | \fBzfs_scan_idle\fR ticks. | |
29714574 TF |
1102 | .sp |
1103 | Default value: \fB4\fR. | |
1104 | .RE | |
1105 | ||
fd8febbd TF |
1106 | .sp |
1107 | .ne 2 | |
1108 | .na | |
1109 | \fBzfs_send_corrupt_data\fR (int) | |
1110 | .ad | |
1111 | .RS 12n | |
1112 | Allow to send corrupt data (ignore read/checksum errors when sending data) | |
1113 | .sp | |
1114 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
1115 | .RE | |
1116 | ||
29714574 TF |
1117 | .sp |
1118 | .ne 2 | |
1119 | .na | |
1120 | \fBzfs_sync_pass_deferred_free\fR (int) | |
1121 | .ad | |
1122 | .RS 12n | |
1123 | Defer frees starting in this pass | |
1124 | .sp | |
1125 | Default value: \fB2\fR. | |
1126 | .RE | |
1127 | ||
1128 | .sp | |
1129 | .ne 2 | |
1130 | .na | |
1131 | \fBzfs_sync_pass_dont_compress\fR (int) | |
1132 | .ad | |
1133 | .RS 12n | |
1134 | Don't compress starting in this pass | |
1135 | .sp | |
1136 | Default value: \fB5\fR. | |
1137 | .RE | |
1138 | ||
1139 | .sp | |
1140 | .ne 2 | |
1141 | .na | |
1142 | \fBzfs_sync_pass_rewrite\fR (int) | |
1143 | .ad | |
1144 | .RS 12n | |
1145 | Rewrite new bps starting in this pass | |
1146 | .sp | |
1147 | Default value: \fB2\fR. | |
1148 | .RE | |
1149 | ||
1150 | .sp | |
1151 | .ne 2 | |
1152 | .na | |
1153 | \fBzfs_top_maxinflight\fR (int) | |
1154 | .ad | |
1155 | .RS 12n | |
27b293be | 1156 | Max I/Os per top-level vdev during scrub or resilver operations. |
29714574 TF |
1157 | .sp |
1158 | Default value: \fB32\fR. | |
1159 | .RE | |
1160 | ||
1161 | .sp | |
1162 | .ne 2 | |
1163 | .na | |
1164 | \fBzfs_txg_history\fR (int) | |
1165 | .ad | |
1166 | .RS 12n | |
1167 | Historic statistics for the last N txgs | |
1168 | .sp | |
1169 | Default value: \fB0\fR. | |
1170 | .RE | |
1171 | ||
29714574 TF |
1172 | .sp |
1173 | .ne 2 | |
1174 | .na | |
1175 | \fBzfs_txg_timeout\fR (int) | |
1176 | .ad | |
1177 | .RS 12n | |
1178 | Max seconds worth of delta per txg | |
1179 | .sp | |
1180 | Default value: \fB5\fR. | |
1181 | .RE | |
1182 | ||
1183 | .sp | |
1184 | .ne 2 | |
1185 | .na | |
1186 | \fBzfs_vdev_aggregation_limit\fR (int) | |
1187 | .ad | |
1188 | .RS 12n | |
1189 | Max vdev I/O aggregation size | |
1190 | .sp | |
1191 | Default value: \fB131,072\fR. | |
1192 | .RE | |
1193 | ||
1194 | .sp | |
1195 | .ne 2 | |
1196 | .na | |
1197 | \fBzfs_vdev_cache_bshift\fR (int) | |
1198 | .ad | |
1199 | .RS 12n | |
1200 | Shift size to inflate reads too | |
1201 | .sp | |
1202 | Default value: \fB16\fR. | |
1203 | .RE | |
1204 | ||
1205 | .sp | |
1206 | .ne 2 | |
1207 | .na | |
1208 | \fBzfs_vdev_cache_max\fR (int) | |
1209 | .ad | |
1210 | .RS 12n | |
1211 | Inflate reads small than max | |
1212 | .RE | |
1213 | ||
1214 | .sp | |
1215 | .ne 2 | |
1216 | .na | |
1217 | \fBzfs_vdev_cache_size\fR (int) | |
1218 | .ad | |
1219 | .RS 12n | |
1220 | Total size of the per-disk cache | |
1221 | .sp | |
1222 | Default value: \fB0\fR. | |
1223 | .RE | |
1224 | ||
29714574 TF |
1225 | .sp |
1226 | .ne 2 | |
1227 | .na | |
1228 | \fBzfs_vdev_mirror_switch_us\fR (int) | |
1229 | .ad | |
1230 | .RS 12n | |
1231 | Switch mirrors every N usecs | |
1232 | .sp | |
1233 | Default value: \fB10,000\fR. | |
1234 | .RE | |
1235 | ||
29714574 TF |
1236 | .sp |
1237 | .ne 2 | |
1238 | .na | |
1239 | \fBzfs_vdev_read_gap_limit\fR (int) | |
1240 | .ad | |
1241 | .RS 12n | |
1242 | Aggregate read I/O over gap | |
1243 | .sp | |
1244 | Default value: \fB32,768\fR. | |
1245 | .RE | |
1246 | ||
1247 | .sp | |
1248 | .ne 2 | |
1249 | .na | |
1250 | \fBzfs_vdev_scheduler\fR (charp) | |
1251 | .ad | |
1252 | .RS 12n | |
1253 | I/O scheduler | |
1254 | .sp | |
1255 | Default value: \fBnoop\fR. | |
1256 | .RE | |
1257 | ||
29714574 TF |
1258 | .sp |
1259 | .ne 2 | |
1260 | .na | |
1261 | \fBzfs_vdev_write_gap_limit\fR (int) | |
1262 | .ad | |
1263 | .RS 12n | |
1264 | Aggregate write I/O over gap | |
1265 | .sp | |
1266 | Default value: \fB4,096\fR. | |
1267 | .RE | |
1268 | ||
29714574 TF |
1269 | .sp |
1270 | .ne 2 | |
1271 | .na | |
1272 | \fBzfs_zevent_cols\fR (int) | |
1273 | .ad | |
1274 | .RS 12n | |
1275 | Max event column width | |
1276 | .sp | |
1277 | Default value: \fB80\fR. | |
1278 | .RE | |
1279 | ||
1280 | .sp | |
1281 | .ne 2 | |
1282 | .na | |
1283 | \fBzfs_zevent_console\fR (int) | |
1284 | .ad | |
1285 | .RS 12n | |
1286 | Log events to the console | |
1287 | .sp | |
1288 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
1289 | .RE | |
1290 | ||
1291 | .sp | |
1292 | .ne 2 | |
1293 | .na | |
1294 | \fBzfs_zevent_len_max\fR (int) | |
1295 | .ad | |
1296 | .RS 12n | |
1297 | Max event queue length | |
1298 | .sp | |
1299 | Default value: \fB0\fR. | |
1300 | .RE | |
1301 | ||
1302 | .sp | |
1303 | .ne 2 | |
1304 | .na | |
1305 | \fBzil_replay_disable\fR (int) | |
1306 | .ad | |
1307 | .RS 12n | |
1308 | Disable intent logging replay | |
1309 | .sp | |
1310 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
1311 | .RE | |
1312 | ||
1313 | .sp | |
1314 | .ne 2 | |
1315 | .na | |
1316 | \fBzil_slog_limit\fR (ulong) | |
1317 | .ad | |
1318 | .RS 12n | |
1319 | Max commit bytes to separate log device | |
1320 | .sp | |
1321 | Default value: \fB1,048,576\fR. | |
1322 | .RE | |
1323 | ||
1324 | .sp | |
1325 | .ne 2 | |
1326 | .na | |
1327 | \fBzio_bulk_flags\fR (int) | |
1328 | .ad | |
1329 | .RS 12n | |
1330 | Additional flags to pass to bulk buffers | |
1331 | .sp | |
1332 | Default value: \fB0\fR. | |
1333 | .RE | |
1334 | ||
1335 | .sp | |
1336 | .ne 2 | |
1337 | .na | |
1338 | \fBzio_delay_max\fR (int) | |
1339 | .ad | |
1340 | .RS 12n | |
1341 | Max zio millisec delay before posting event | |
1342 | .sp | |
1343 | Default value: \fB30,000\fR. | |
1344 | .RE | |
1345 | ||
1346 | .sp | |
1347 | .ne 2 | |
1348 | .na | |
1349 | \fBzio_injection_enabled\fR (int) | |
1350 | .ad | |
1351 | .RS 12n | |
1352 | Enable fault injection | |
1353 | .sp | |
1354 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
1355 | .RE | |
1356 | ||
1357 | .sp | |
1358 | .ne 2 | |
1359 | .na | |
1360 | \fBzio_requeue_io_start_cut_in_line\fR (int) | |
1361 | .ad | |
1362 | .RS 12n | |
1363 | Prioritize requeued I/O | |
1364 | .sp | |
1365 | Default value: \fB0\fR. | |
1366 | .RE | |
1367 | ||
1368 | .sp | |
1369 | .ne 2 | |
1370 | .na | |
1371 | \fBzvol_inhibit_dev\fR (uint) | |
1372 | .ad | |
1373 | .RS 12n | |
1374 | Do not create zvol device nodes | |
1375 | .sp | |
1376 | Use \fB1\fR for yes and \fB0\fR for no (default). | |
1377 | .RE | |
1378 | ||
1379 | .sp | |
1380 | .ne 2 | |
1381 | .na | |
1382 | \fBzvol_major\fR (uint) | |
1383 | .ad | |
1384 | .RS 12n | |
1385 | Major number for zvol device | |
1386 | .sp | |
1387 | Default value: \fB230\fR. | |
1388 | .RE | |
1389 | ||
1390 | .sp | |
1391 | .ne 2 | |
1392 | .na | |
1393 | \fBzvol_max_discard_blocks\fR (ulong) | |
1394 | .ad | |
1395 | .RS 12n | |
1396 | Max number of blocks to discard at once | |
1397 | .sp | |
1398 | Default value: \fB16,384\fR. | |
1399 | .RE | |
1400 | ||
1401 | .sp | |
1402 | .ne 2 | |
1403 | .na | |
1404 | \fBzvol_threads\fR (uint) | |
1405 | .ad | |
1406 | .RS 12n | |
1407 | Number of threads for zvol device | |
1408 | .sp | |
1409 | Default value: \fB32\fR. | |
1410 | .RE | |
1411 | ||
e8b96c60 MA |
1412 | .SH ZFS I/O SCHEDULER |
1413 | ZFS issues I/O operations to leaf vdevs to satisfy and complete I/Os. | |
1414 | The I/O scheduler determines when and in what order those operations are | |
1415 | issued. The I/O scheduler divides operations into five I/O classes | |
1416 | prioritized in the following order: sync read, sync write, async read, | |
1417 | async write, and scrub/resilver. Each queue defines the minimum and | |
1418 | maximum number of concurrent operations that may be issued to the | |
1419 | device. In addition, the device has an aggregate maximum, | |
1420 | \fBzfs_vdev_max_active\fR. Note that the sum of the per-queue minimums | |
1421 | must not exceed the aggregate maximum. If the sum of the per-queue | |
1422 | maximums exceeds the aggregate maximum, then the number of active I/Os | |
1423 | may reach \fBzfs_vdev_max_active\fR, in which case no further I/Os will | |
1424 | be issued regardless of whether all per-queue minimums have been met. | |
1425 | .sp | |
1426 | For many physical devices, throughput increases with the number of | |
1427 | concurrent operations, but latency typically suffers. Further, physical | |
1428 | devices typically have a limit at which more concurrent operations have no | |
1429 | effect on throughput or can actually cause it to decrease. | |
1430 | .sp | |
1431 | The scheduler selects the next operation to issue by first looking for an | |
1432 | I/O class whose minimum has not been satisfied. Once all are satisfied and | |
1433 | the aggregate maximum has not been hit, the scheduler looks for classes | |
1434 | whose maximum has not been satisfied. Iteration through the I/O classes is | |
1435 | done in the order specified above. No further operations are issued if the | |
1436 | aggregate maximum number of concurrent operations has been hit or if there | |
1437 | are no operations queued for an I/O class that has not hit its maximum. | |
1438 | Every time an I/O is queued or an operation completes, the I/O scheduler | |
1439 | looks for new operations to issue. | |
1440 | .sp | |
1441 | In general, smaller max_active's will lead to lower latency of synchronous | |
1442 | operations. Larger max_active's may lead to higher overall throughput, | |
1443 | depending on underlying storage. | |
1444 | .sp | |
1445 | The ratio of the queues' max_actives determines the balance of performance | |
1446 | between reads, writes, and scrubs. E.g., increasing | |
1447 | \fBzfs_vdev_scrub_max_active\fR will cause the scrub or resilver to complete | |
1448 | more quickly, but reads and writes to have higher latency and lower throughput. | |
1449 | .sp | |
1450 | All I/O classes have a fixed maximum number of outstanding operations | |
1451 | except for the async write class. Asynchronous writes represent the data | |
1452 | that is committed to stable storage during the syncing stage for | |
1453 | transaction groups. Transaction groups enter the syncing state | |
1454 | periodically so the number of queued async writes will quickly burst up | |
1455 | and then bleed down to zero. Rather than servicing them as quickly as | |
1456 | possible, the I/O scheduler changes the maximum number of active async | |
1457 | write I/Os according to the amount of dirty data in the pool. Since | |
1458 | both throughput and latency typically increase with the number of | |
1459 | concurrent operations issued to physical devices, reducing the | |
1460 | burstiness in the number of concurrent operations also stabilizes the | |
1461 | response time of operations from other -- and in particular synchronous | |
1462 | -- queues. In broad strokes, the I/O scheduler will issue more | |
1463 | concurrent operations from the async write queue as there's more dirty | |
1464 | data in the pool. | |
1465 | .sp | |
1466 | Async Writes | |
1467 | .sp | |
1468 | The number of concurrent operations issued for the async write I/O class | |
1469 | follows a piece-wise linear function defined by a few adjustable points. | |
1470 | .nf | |
1471 | ||
1472 | | o---------| <-- zfs_vdev_async_write_max_active | |
1473 | ^ | /^ | | |
1474 | | | / | | | |
1475 | active | / | | | |
1476 | I/O | / | | | |
1477 | count | / | | | |
1478 | | / | | | |
1479 | |-------o | | <-- zfs_vdev_async_write_min_active | |
1480 | 0|_______^______|_________| | |
1481 | 0% | | 100% of zfs_dirty_data_max | |
1482 | | | | |
1483 | | `-- zfs_vdev_async_write_active_max_dirty_percent | |
1484 | `--------- zfs_vdev_async_write_active_min_dirty_percent | |
1485 | ||
1486 | .fi | |
1487 | Until the amount of dirty data exceeds a minimum percentage of the dirty | |
1488 | data allowed in the pool, the I/O scheduler will limit the number of | |
1489 | concurrent operations to the minimum. As that threshold is crossed, the | |
1490 | number of concurrent operations issued increases linearly to the maximum at | |
1491 | the specified maximum percentage of the dirty data allowed in the pool. | |
1492 | .sp | |
1493 | Ideally, the amount of dirty data on a busy pool will stay in the sloped | |
1494 | part of the function between \fBzfs_vdev_async_write_active_min_dirty_percent\fR | |
1495 | and \fBzfs_vdev_async_write_active_max_dirty_percent\fR. If it exceeds the | |
1496 | maximum percentage, this indicates that the rate of incoming data is | |
1497 | greater than the rate that the backend storage can handle. In this case, we | |
1498 | must further throttle incoming writes, as described in the next section. | |
1499 | ||
1500 | .SH ZFS TRANSACTION DELAY | |
1501 | We delay transactions when we've determined that the backend storage | |
1502 | isn't able to accommodate the rate of incoming writes. | |
1503 | .sp | |
1504 | If there is already a transaction waiting, we delay relative to when | |
1505 | that transaction will finish waiting. This way the calculated delay time | |
1506 | is independent of the number of threads concurrently executing | |
1507 | transactions. | |
1508 | .sp | |
1509 | If we are the only waiter, wait relative to when the transaction | |
1510 | started, rather than the current time. This credits the transaction for | |
1511 | "time already served", e.g. reading indirect blocks. | |
1512 | .sp | |
1513 | The minimum time for a transaction to take is calculated as: | |
1514 | .nf | |
1515 | min_time = zfs_delay_scale * (dirty - min) / (max - dirty) | |
1516 | min_time is then capped at 100 milliseconds. | |
1517 | .fi | |
1518 | .sp | |
1519 | The delay has two degrees of freedom that can be adjusted via tunables. The | |
1520 | percentage of dirty data at which we start to delay is defined by | |
1521 | \fBzfs_delay_min_dirty_percent\fR. This should typically be at or above | |
1522 | \fBzfs_vdev_async_write_active_max_dirty_percent\fR so that we only start to | |
1523 | delay after writing at full speed has failed to keep up with the incoming write | |
1524 | rate. The scale of the curve is defined by \fBzfs_delay_scale\fR. Roughly speaking, | |
1525 | this variable determines the amount of delay at the midpoint of the curve. | |
1526 | .sp | |
1527 | .nf | |
1528 | delay | |
1529 | 10ms +-------------------------------------------------------------*+ | |
1530 | | *| | |
1531 | 9ms + *+ | |
1532 | | *| | |
1533 | 8ms + *+ | |
1534 | | * | | |
1535 | 7ms + * + | |
1536 | | * | | |
1537 | 6ms + * + | |
1538 | | * | | |
1539 | 5ms + * + | |
1540 | | * | | |
1541 | 4ms + * + | |
1542 | | * | | |
1543 | 3ms + * + | |
1544 | | * | | |
1545 | 2ms + (midpoint) * + | |
1546 | | | ** | | |
1547 | 1ms + v *** + | |
1548 | | zfs_delay_scale ----------> ******** | | |
1549 | 0 +-------------------------------------*********----------------+ | |
1550 | 0% <- zfs_dirty_data_max -> 100% | |
1551 | .fi | |
1552 | .sp | |
1553 | Note that since the delay is added to the outstanding time remaining on the | |
1554 | most recent transaction, the delay is effectively the inverse of IOPS. | |
1555 | Here the midpoint of 500us translates to 2000 IOPS. The shape of the curve | |
1556 | was chosen such that small changes in the amount of accumulated dirty data | |
1557 | in the first 3/4 of the curve yield relatively small differences in the | |
1558 | amount of delay. | |
1559 | .sp | |
1560 | The effects can be easier to understand when the amount of delay is | |
1561 | represented on a log scale: | |
1562 | .sp | |
1563 | .nf | |
1564 | delay | |
1565 | 100ms +-------------------------------------------------------------++ | |
1566 | + + | |
1567 | | | | |
1568 | + *+ | |
1569 | 10ms + *+ | |
1570 | + ** + | |
1571 | | (midpoint) ** | | |
1572 | + | ** + | |
1573 | 1ms + v **** + | |
1574 | + zfs_delay_scale ----------> ***** + | |
1575 | | **** | | |
1576 | + **** + | |
1577 | 100us + ** + | |
1578 | + * + | |
1579 | | * | | |
1580 | + * + | |
1581 | 10us + * + | |
1582 | + + | |
1583 | | | | |
1584 | + + | |
1585 | +--------------------------------------------------------------+ | |
1586 | 0% <- zfs_dirty_data_max -> 100% | |
1587 | .fi | |
1588 | .sp | |
1589 | Note here that only as the amount of dirty data approaches its limit does | |
1590 | the delay start to increase rapidly. The goal of a properly tuned system | |
1591 | should be to keep the amount of dirty data out of that range by first | |
1592 | ensuring that the appropriate limits are set for the I/O scheduler to reach | |
1593 | optimal throughput on the backend storage, and then by changing the value | |
1594 | of \fBzfs_delay_scale\fR to increase the steepness of the curve. |