1LVMTHIN(7) LVMTHIN(7)
2
6 lvmthin — LVM thin provisioning
7
10 Blocks in a standard lvm(8) Logical Volume (LV) are allocated when the
11 LV is created, but blocks in a thin provisioned LV are allocated as
12 they are written. Because of this, a thin provisioned LV is given a
13 virtual size, and can then be much larger than physically available
14 storage. The amount of physical storage provided for thin provisioned
15 LVs can be increased later as the need arises.
16 Blocks in a standard LV are allocated (during creation) from the Volume
18 Group (VG), but blocks in a thin LV are allocated (during use) from a
19 special "thin pool LV". The thin pool LV contains blocks of physical
20 storage, and blocks in thin LVs just reference blocks in the thin pool
21 LV.
22 A thin pool LV must be created before thin LVs can be created within
24 it. A thin pool LV is created by combining two standard LVs: a large
25 data LV that will hold blocks for thin LVs, and a metadata LV that will
26 hold metadata. The metadata tracks which data blocks belong to each
27 thin LV.
28 Snapshots of thin LVs are efficient because the data blocks common to a
30 thin LV and any of its snapshots are shared. Snapshots may be taken of
31 thin LVs or of other thin snapshots. Blocks common to recursive snap‐
32 shots are also shared in the thin pool. There is no limit to or degra‐
33 dation from sequences of snapshots.
34 As thin LVs or snapshot LVs are written to, they consume data blocks in
36 the thin pool. As free data blocks in the pool decrease, more free
37 blocks may need to be supplied. This is done by extending the thin
38 pool data LV with additional physical space from the VG. Removing thin
39 LVs or snapshots from the thin pool can also free blocks in the thin
40 pool. However, removing LVs is not always an effective way of freeing
41 space in a thin pool because the amount is limited to the number of
42 blocks not shared with other LVs in the pool.
43 Incremental block allocation from thin pools can cause thin LVs to
45 become fragmented. Standard LVs generally avoid this problem by allo‐
46 cating all the blocks at once during creation.
47
51 ThinDataLV
52 thin data LV
53 large LV created in a VG
54 used by thin pool to store ThinLV blocks
55 ThinMetaLV
58 thin metadata LV
59 small LV created in a VG
60 used by thin pool to track data block usage
61 ThinPoolLV
64 thin pool LV
65 combination of ThinDataLV and ThinMetaLV
66 contains ThinLVs and SnapLVs
67 ThinLV
70 thin LV
71 created from ThinPoolLV
72 appears blank after creation
73 SnapLV
76 snapshot LV
77 created from ThinPoolLV
78 appears as a snapshot of another LV after creation
79
84 The primary method for using lvm thin provisioning:
85 1. create ThinDataLV
88 Create an LV that will hold thin pool data.
89 lvcreate -n ThinDataLV -L LargeSize VG
91 Example
93 # lvcreate -n pool0 -L 10G vg
94 2. create ThinMetaLV
97 Create an LV that will hold thin pool metadata.
98 lvcreate -n ThinMetaLV -L SmallSize VG
100 Example
102 # lvcreate -n pool0meta -L 1G vg
103 # lvs
105 LV VG Attr LSize
106 pool0 vg -wi-a----- 10.00g
107 pool0meta vg -wi-a----- 1.00g
108 3. create ThinPoolLV
111 Combine the data and metadata LVs into a thin pool LV.
112 ThinDataLV is renamed to hidden ThinPoolLV_tdata.
113 ThinMetaLV is renamed to hidden ThinPoolLV_tmeta.
114 The new ThinPoolLV takes the previous name of ThinDataLV.
115 lvconvert --type thin-pool --poolmetadata VG/ThinMetaLV VG/ThinDataLV
117 Example
119 # lvconvert --type thin-pool --poolmetadata vg/pool0meta vg/pool0
120 # lvs vg/pool0
122 LV VG Attr LSize Pool Origin Data% Meta%
123 pool0 vg twi-a-tz-- 10.00g 0.00 0.00
124 # lvs -a
126 LV VG Attr LSize
127 pool0 vg twi-a-tz-- 10.00g
128 [pool0_tdata] vg Twi-ao---- 10.00g
129 [pool0_tmeta] vg ewi-ao---- 1.00g
130 4. create ThinLV
133 Create a new thin LV from the thin pool LV.
134 The thin LV is created with a virtual size.
135 Multiple new thin LVs may be created in the thin pool.
136 Thin LV names must be unique in the VG.
137 The '--type thin' option is inferred from the virtual size option.
138 The --thinpool argument specifies which thin pool will
139 contain the ThinLV.
140 lvcreate -n ThinLV -V VirtualSize --thinpool ThinPoolLV VG
142 Example
144 Create a thin LV in a thin pool:
145 # lvcreate -n thin1 -V 1T --thinpool pool0 vg
146 Create another thin LV in the same thin pool:
148 # lvcreate -n thin2 -V 1T --thinpool pool0 vg
149 # lvs vg/thin1 vg/thin2
151 LV VG Attr LSize Pool Origin Data%
152 thin1 vg Vwi-a-tz-- 1.00t pool0 0.00
153 thin2 vg Vwi-a-tz-- 1.00t pool0 0.00
154 5. create SnapLV
157 Create snapshots of an existing ThinLV or SnapLV.
158 Do not specify -L, --size when creating a thin snapshot.
159 A size argument will cause an old COW snapshot to be created.
160 lvcreate -n SnapLV --snapshot VG/ThinLV
162 lvcreate -n SnapLV --snapshot VG/PrevSnapLV
163 Example
165 Create first snapshot of an existing ThinLV:
166 # lvcreate -n thin1s1 -s vg/thin1
167 Create second snapshot of the same ThinLV:
169 # lvcreate -n thin1s2 -s vg/thin1
170 Create a snapshot of the first snapshot:
172 # lvcreate -n thin1s1s1 -s vg/thin1s1
173 # lvs vg/thin1s1 vg/thin1s2 vg/thin1s1s1
175 LV VG Attr LSize Pool Origin
176 thin1s1 vg Vwi---tz-k 1.00t pool0 thin1
177 thin1s2 vg Vwi---tz-k 1.00t pool0 thin1
178 thin1s1s1 vg Vwi---tz-k 1.00t pool0 thin1s1
179 6. activate SnapLV
182 Thin snapshots are created with the persistent "activation skip" flag,
183 indicated by the "k" attribute. Use -K with lvchange or vgchange to
184 activate thin snapshots with the "k" attribute.
185 lvchange -ay -K VG/SnapLV
187 Example
189 # lvchange -ay -K vg/thin1s1
190 # lvs vg/thin1s1
192 LV VG Attr LSize Pool Origin
193 thin1s1 vg Vwi-a-tz-k 1.00t pool0 thin1
194
197 Alternate syntax for specifying type thin-pool
198 Automatic pool metadata LV
199 Specify devices for data and metadata LVs
200 Tolerate device failures using raid
201 Spare metadata LV
202 Metadata check and repair
203 Activation of thin snapshots
204 Removing thin pool LVs, thin LVs and snapshots
205 Manually manage free data space of thin pool LV
206 Manually manage free metadata space of a thin pool LV
207 Using fstrim to increase free space in a thin pool LV
208 Automatically extend thin pool LV
209 Data space exhaustion
210 Metadata space exhaustion
211 Automatic extend settings
212 Zeroing
213 Discard
214 Chunk size
215 Size of pool metadata LV
216 Create a thin snapshot of an external, read only LV
217 Convert a standard LV to a thin LV with an external origin
218 Single step thin pool LV creation
219 Single step thin pool LV and thin LV creation
220 Merge thin snapshots
221 XFS on snapshots
222 Automatic pool metadata LV
227 A thin data LV can be converted to a thin pool LV without specifying a
230 thin pool metadata LV. LVM automatically creates a metadata LV from
231 the same VG.
232 lvcreate -n ThinDataLV -L LargeSize VG
234 lvconvert --type thin-pool VG/ThinDataLV
235 Example
237 # lvcreate -n pool0 -L 10G vg
238 # lvconvert --type thin-pool vg/pool0
239 # lvs -a
241 pool0 vg twi-a-tz-- 10.00g
242 [pool0_tdata] vg Twi-ao---- 10.00g
243 [pool0_tmeta] vg ewi-ao---- 16.00m
244 Specify devices for data and metadata LVs
248 The data and metadata LVs in a thin pool are best created on separate
251 physical devices. To do that, specify the device name(s) at the end of
252 the lvcreate line. It can be especially helpful to use fast devices
253 for the metadata LV.
254 lvcreate -n ThinDataLV -L LargeSize VG LargePV
256 lvcreate -n ThinMetaLV -L SmallSize VG SmallPV
257 lvconvert --type thin-pool --poolmetadata VG/ThinMetaLV VG/ThinDataLV
258 Example
260 # lvcreate -n pool0 -L 10G vg /dev/sdA
261 # lvcreate -n pool0meta -L 1G vg /dev/sdB
262 # lvconvert --type thin-pool --poolmetadata vg/pool0meta vg/pool0
263 lvm.conf(5) thin_pool_metadata_require_separate_pvs
265 controls the default PV usage for thin pool creation.
266 Tolerate device failures using raid
271 To tolerate device failures, use raid for the pool data LV and pool
274 metadata LV. This is especially recommended for pool metadata LVs.
275 lvcreate --type raid1 -m 1 -n ThinMetaLV -L SmallSize VG PVA PVB
277 lvcreate --type raid1 -m 1 -n ThinDataLV -L LargeSize VG PVC PVD
278 lvconvert --type thin-pool --poolmetadata VG/ThinMetaLV VG/ThinDataLV
279 Example
281 # lvcreate --type raid1 -m 1 -n pool0 -L 10G vg /dev/sdA /dev/sdB
282 # lvcreate --type raid1 -m 1 -n pool0meta -L 1G vg /dev/sdC /dev/sdD
283 # lvconvert --type thin-pool --poolmetadata vg/pool0meta vg/pool0
284 Spare metadata LV
288 The first time a thin pool LV is created, lvm will create a spare meta‐
291 data LV in the VG. This behavior can be controlled with the option
292 --poolmetadataspare y|n. (Future thin pool creations will also attempt
293 to create the pmspare LV if none exists.)
294 To create the pmspare ("pool metadata spare") LV, lvm first creates an
296 LV with a default name, e.g. lvol0, and then converts this LV to a hid‐
297 den LV with the _pmspare suffix, e.g. lvol0_pmspare.
298 One pmspare LV is kept in a VG to be used for any thin pool.
300 The pmspare LV cannot be created explicitly, but may be removed explic‐
302 itly.
303 Example
305 # lvcreate -n pool0 -L 10G vg
306 # lvcreate -n pool0meta -L 1G vg
307 # lvconvert --type thin-pool --poolmetadata vg/pool0meta vg/pool0
308 # lvs -a
310 [lvol0_pmspare] vg ewi-------
311 pool0 vg twi---tz--
312 [pool0_tdata] vg Twi-------
313 [pool0_tmeta] vg ewi-------
314 The "Metadata check and repair" section describes the use of the pms‐
316 pare LV.
317 Metadata check and repair
321 If thin pool metadata is damaged, it may be repairable. Checking and
324 repairing thin pool metadata is analagous to running fsck/repair on a
325 file system.
326 When a thin pool LV is activated, lvm runs the thin_check command to
328 check the correctness of the metadata on the pool metadata LV.
329 lvm.conf(5) thin_check_executable
331 can be set to an empty string ("") to disable the thin_check step.
332 This is not recommended.
333 lvm.conf(5) thin_check_options
335 controls the command options used for the thin_check command.
336 If the thin_check command finds a problem with the metadata, the thin
338 pool LV is not activated, and the thin pool metadata needs to be
339 repaired.
340 Simple repair commands are not always successful. Advanced repair may
342 require editing thin pool metadata and lvm metadata. Newer versions of
343 the kernel and lvm tools may be more successful at repair. Report the
344 details of damaged thin metadata to get the best advice on recovery.
345 Command to repair a thin pool:
347 lvconvert --repair VG/ThinPoolLV
348 Repair performs the following steps:
350 1. Creates a new, repaired copy of the metadata.
352 lvconvert runs the thin_repair command to read damaged metadata from
353 the existing pool metadata LV, and writes a new repaired copy to the
354 VG's pmspare LV.
355 2. Replaces the thin pool metadata LV.
357 If step 1 is successful, the thin pool metadata LV is replaced with the
358 pmspare LV containing the corrected metadata. The previous thin pool
359 metadata LV, containing the damaged metadata, becomes visible with the
360 new name ThinPoolLV_tmetaN (where N is 0,1,...).
361 If the repair works, the thin pool LV and its thin LVs can be acti‐
363 vated, and the LV containing the damaged thin pool metadata can be
364 removed. It may be useful to move the new metadata LV (previously pms‐
365 pare) to a better PV.
366 If the repair does not work, the thin pool LV and its thin LVs are
368 lost.
369 If metadata is manually restored with thin_repair directly, the pool
371 metadata LV can be manually swapped with another LV containing new
372 metadata:
373 lvconvert --thinpool VG/ThinPoolLV --poolmetadata VG/NewThinMetaLV
375 Activation of thin snapshots
379 When a thin snapshot LV is created, it is by default given the "activa‐
382 tion skip" flag. This flag is indicated by the "k" attribute displayed
383 by lvs:
384 # lvs vg/thin1s1
386 LV VG Attr LSize Pool Origin
387 thin1s1 vg Vwi---tz-k 1.00t pool0 thin1
388 This flag causes the snapshot LV to be skipped, i.e. not activated, by
390 normal activation commands. The skipping behavior does not apply to
391 deactivation commands.
392 A snapshot LV with the "k" attribute can be activated using the -K (or
394 --ignoreactivationskip) option in addition to the standard -ay (or
395 --activate y) option.
396 Command to activate a thin snapshot LV:
398 lvchange -ay -K VG/SnapLV
399 The persistent "activation skip" flag can be turned off during lvcre‐
401 ate, or later with lvchange using the -kn (or --setactivationskip n)
402 option. It can be turned on again with -ky (or --setactivationskip y).
403 When the "activation skip" flag is removed, normal activation commands
405 will activate the LV, and the -K activation option is not needed.
406 Command to create snapshot LV without the activation skip flag:
408 lvcreate -kn -n SnapLV -s VG/ThinLV
409 Command to remove the activation skip flag from a snapshot LV:
411 lvchange -kn VG/SnapLV
412 lvm.conf(5) auto_set_activation_skip
414 controls the default activation skip setting used by lvcreate.
415 Removing thin pool LVs, thin LVs and snapshots
419 Removing a thin LV and its related snapshots returns the blocks it used
422 to the thin pool LV. These blocks will be reused for other thin LVs
423 and snapshots.
424 Removing a thin pool LV removes both the data LV and metadata LV and
426 returns the space to the VG.
427 lvremove of thin pool LVs, thin LVs and snapshots cannot be reversed
429 with vgcfgrestore.
430 vgcfgbackup does not back up thin pool metadata.
432 Manually manage free data space of thin pool LV
436 The available free space in a thin pool LV can be displayed with the
439 lvs command. Free space can be added by extending the thin pool LV.
440 Command to extend thin pool data space:
442 lvextend -L Size VG/ThinPoolLV
443 Example
445 1. A thin pool LV is using 26.96% of its data blocks.
446 # lvs
447 LV VG Attr LSize Pool Origin Data%
448 pool0 vg twi-a-tz-- 10.00g 26.96
449 2. Double the amount of physical space in the thin pool LV.
451 # lvextend -L+10G vg/pool0
452 3. The percentage of used data blocks is half the previous value.
454 # lvs
455 LV VG Attr LSize Pool Origin Data%
456 pool0 vg twi-a-tz-- 20.00g 13.48
457 Other methods of increasing free data space in a thin pool LV include
459 removing a thin LV and its related snapsots, or running fstrim on the
460 file system using a thin LV.
461 Manually manage free metadata space of a thin pool LV
465 The available metadata space in a thin pool LV can be displayed with
468 the lvs -o+metadata_percent command.
469 Command to extend thin pool metadata space:
471 lvextend --poolmetadatasize Size VG/ThinPoolLV
472 Example
474 1. A thin pool LV is using 12.40% of its metadata blocks.
475 # lvs -oname,size,data_percent,metadata_percent vg/pool0
476 LV LSize Data% Meta%
477 pool0 20.00g 13.48 12.40
478 2. Display a thin pool LV with its component thin data LV and thin
480 metadata LV.
481 # lvs -a -oname,attr,size vg
482 LV Attr LSize
483 pool0 twi-a-tz-- 20.00g
484 [pool0_tdata] Twi-ao---- 20.00g
485 [pool0_tmeta] ewi-ao---- 12.00m
486 3. Double the amount of physical space in the thin metadata LV.
488 # lvextend --poolmetadatasize +12M vg/pool0
489 4. The percentage of used metadata blocks is half the previous value.
491 # lvs -a -oname,size,data_percent,metadata_percent vg
492 LV LSize Data% Meta%
493 pool0 20.00g 13.48 6.20
494 [pool0_tdata] 20.00g
495 [pool0_tmeta] 24.00m
496 Using fstrim to increase free space in a thin pool LV
500 Removing files in a file system on top of a thin LV does not generally
503 add free space back to the thin pool. Manually running the fstrim com‐
504 mand can return space back to the thin pool that had been used by
505 removed files. fstrim uses discards and will not work if the thin pool
506 LV has discards mode set to ignore.
507 Example
509 A thin pool has 10G of physical data space, and a thin LV has a virtual
510 size of 100G. Writing a 1G file to the file system reduces the free
511 space in the thin pool by 10% and increases the virtual usage of the
512 file system by 1%. Removing the 1G file restores the virtual 1% to the
513 file system, but does not restore the physical 10% to the thin pool.
514 The fstrim command restores the physical space to the thin pool.
515 # lvs -a -oname,attr,size,pool_lv,origin,data_percent,metadata_percent vg
517 LV Attr LSize Pool Origin Data% Meta%
518 pool0 twi-a-tz-- 10.00g 47.01 21.03
519 thin1 Vwi-aotz-- 100.00g pool0 2.70
520 # df -h /mnt/X
522 Filesystem Size Used Avail Use% Mounted on
523 /dev/mapper/vg-thin1 99G 1.1G 93G 2% /mnt/X
524 # dd if=/dev/zero of=/mnt/X/1Gfile bs=4096 count=262144; sync
526 # lvs
528 pool0 vg twi-a-tz-- 10.00g 57.01 25.26
529 thin1 vg Vwi-aotz-- 100.00g pool0 3.70
530 # df -h /mnt/X
532 /dev/mapper/vg-thin1 99G 2.1G 92G 3% /mnt/X
533 # rm /mnt/X/1Gfile
535 # lvs
537 pool0 vg twi-a-tz-- 10.00g 57.01 25.26
538 thin1 vg Vwi-aotz-- 100.00g pool0 3.70
539 # df -h /mnt/X
541 /dev/mapper/vg-thin1 99G 1.1G 93G 2% /mnt/X
542 # fstrim -v /mnt/X
544 # lvs
546 pool0 vg twi-a-tz-- 10.00g 47.01 21.03
547 thin1 vg Vwi-aotz-- 100.00g pool0 2.70
548 The "Discard" section covers an option for automatically freeing data
550 space in a thin pool.
551 Automatically extend thin pool LV
555 The lvm daemon dmeventd (lvm2-monitor) monitors the data usage of thin
558 pool LVs and extends them when the usage reaches a certain level. The
559 necessary free space must exist in the VG to extend thin pool LVs.
560 Monitoring and extension of thin pool LVs are controlled independently.
561 monitoring
563 When a thin pool LV is activated, dmeventd will begin monitoring it by
565 default.
566 Command to start or stop dmeventd monitoring a thin pool LV:
568 lvchange --monitor {y|n} VG/ThinPoolLV
569 The current dmeventd monitoring status of a thin pool LV can be dis‐
571 played with the command lvs -o+seg_monitor.
572 autoextend
574 dmeventd should be configured to extend thin pool LVs before all data
576 space is used. Warnings are emitted through syslog when the use of a
577 thin pool reaches 80%, 85%, 90% and 95%. (See the section "Data space
578 exhaustion" for the effects of not extending a thin pool LV.) The
579 point at which dmeventd extends thin pool LVs, and the amount are con‐
580 trolled with two configuration settings:
581 lvm.conf(5) thin_pool_autoextend_threshold
583 is a percentage full value that defines when the thin pool LV should be
584 extended. Setting this to 100 disables automatic extention. The mini‐
585 mum value is 50.
586 lvm.conf(5) thin_pool_autoextend_percent
588 defines how much extra data space should be added to the thin pool LV
589 from the VG, in percent of its current size.
590 disabling
592 There are multiple ways that extension of thin pools could be pre‐
594 vented:
595 · If the dmeventd daemon is not running, no monitoring or automatic
598 extension will occur.
599 · Even when dmeventd is running, all monitoring can be disabled with
602 the lvm.conf monitoring setting.
603 · To activate or create a thin pool LV without interacting with
606 dmeventd, the --ignoremonitoring option can be used. With this
607 option, the command will not ask dmeventd to monitor the thin pool
608 LV.
609 · Setting thin_pool_autoextend_threshould to 100 disables automatic
612 extension of thin pool LVs, even if they are being monitored by
613 dmeventd.
614 Example
617 If thin_pool_autoextend_threshold is 70 and thin_pool_autoextend_per‐
618 cent is 20, whenever a pool exceeds 70% usage, it will be extended by
619 another 20%. For a 1G pool, using 700M will trigger a resize to 1.2G.
620 When the usage exceeds 840M, the pool will be extended to 1.44G, and so
621 on.
622 Data space exhaustion
626 When properly managed, thin pool data space should be extended before
629 it is all used (see the section "Automatically extend thin pool LV").
630 If thin pool data space is already exhausted, it can still be extended
631 (see the section "Manually manage free data space of thin pool LV".)
632 The behavior of a full thin pool is configurable with the --errorwhen‐
634 full y|n option to lvcreate or lvchange. The errorwhenfull setting
635 applies only to writes; reading thin LVs can continue even when data
636 space is exhausted.
637 Command to change the handling of a full thin pool:
639 lvchange --errorwhenfull {y|n} VG/ThinPoolLV
640 lvm.conf(5) error_when_full
642 controls the default error when full behavior.
643 The current setting of a thin pool LV can be displayed with the com‐
645 mand: lvs -o+lv_when_full.
646 The errorwhenfull setting does not effect the monitoring and autoextend
648 settings, and the monitoring/autoextend settings do not effect the
649 errorwhenfull setting. It is only when monitoring/autoextend are not
650 effective that the thin pool becomes full and the errorwhenfull setting
651 is applied.
652 errorwhenfull n
654 This is the default. Writes to thin LVs are accepted and queued, with
656 the expectation that pool data space will be extended soon. Once data
657 space is extended, the queued writes will be processed, and the thin
658 pool will return to normal operation.
659 While waiting to be extended, the thin pool will queue writes for up to
661 60 seconds (the default). If data space has not been extended after
662 this time, the queued writes will return an error to the caller, e.g.
663 the file system. This can result in file system corruption for non-
664 journaled file systems that may require repair. When a thin pool
665 returns errors for writes to a thin LV, any file system is subject to
666 losing unsynced user data.
667 The 60 second timeout can be changed or disabled with the dm-thin-pool
669 kernel module option no_space_timeout. This option sets the number of
670 seconds that thin pools will queue writes. If set to 0, writes will
671 not time out. Disabling timeouts can result in the system running out
672 of resources, memory exhaustion, hung tasks, and deadlocks. (The time‐
673 out applies to all thin pools on the system.)
674 errorwhenfull y
676 Writes to thin LVs immediately return an error, and no writes are
678 queued. In the case of a file system, this can result in corruption
679 that may require fs repair (the specific consequences depend on the
680 thin LV user.)
681 data percent
683 When data space is exhausted, the lvs command displays 100 under Data%
685 for the thin pool LV:
686 # lvs vg/pool0
688 LV VG Attr LSize Pool Origin Data%
689 pool0 vg twi-a-tz-- 512.00m 100.00
690 causes
692 A thin pool may run out of data space for any of the following reasons:
694 · Automatic extension of the thin pool is disabled, and the thin pool
697 is not manually extended. (Disabling automatic extension is not rec‐
698 ommended.)
699 · The dmeventd daemon is not running and the thin pool is not manually
702 extended. (Disabling dmeventd is not recommended.)
703 · Automatic extension of the thin pool is too slow given the rate of
706 writes to thin LVs in the pool. (This can be addressed by tuning the
707 thin_pool_autoextend_threshold and thin_pool_autoextend_percent. See
708 "Automatic extend settings".)
709 · The VG does not have enough free blocks to extend the thin pool.
712 Metadata space exhaustion
715 If thin pool metadata space is exhausted (or a thin pool metadata oper‐
718 ation fails), errors will be returned for IO operations on thin LVs.
719 When metadata space is exhausted, the lvs command displays 100 under
721 Meta% for the thin pool LV:
722 # lvs -o lv_name,size,data_percent,metadata_percent vg/pool0
724 LV LSize Data% Meta%
725 pool0 100.00
726 The same reasons for thin pool data space exhaustion apply to thin pool
728 metadata space.
729 Metadata space exhaustion can lead to inconsistent thin pool metadata
731 and inconsistent file systems, so the response requires offline check‐
732 ing and repair.
733 1. Deactivate the thin pool LV, or reboot the system if this is not
735 possible.
736 2. Repair thin pool with lvconvert --repair.
738 See "Metadata check and repair".
739 3. Extend pool metadata space with lvextend --poolmetadatasize.
741 See "Manually manage free metadata space of a thin pool LV".
742 4. Check and repair file system.
744 Automatic extend settings
748 Thin pool LVs can be extended according to preset values. The presets
751 determine if the LV should be extended based on how full it is, and if
752 so by how much. When dmeventd monitors thin pool LVs, it uses lvextend
753 with these presets. (See "Automatically extend thin pool LV".)
754 Command to extend a thin pool data LV using presets:
756 lvextend --use-policies VG/ThinPoolLV
757 The command uses these settings:
759 lvm.conf(5) thin_pool_autoextend_threshold
761 autoextend the LV when its usage exceeds this percent.
762 lvm.conf(5) thin_pool_autoextend_percent
764 autoextend the LV by this much additional space.
765 To see the default values of these settings, run:
767 lvmconfig --type default --withcomment
769 activation/thin_pool_autoextend_threshold
770 lvmconfig --type default --withcomment
772 activation/thin_pool_autoextend_percent
773 To change these values globally, edit lvm.conf(5).
775 To change these values on a per-VG or per-LV basis, attach a "profile"
777 to the VG or LV. A profile is a collection of config settings, saved
778 in a local text file (using the lvm.conf format). lvm looks for pro‐
779 files in the profile_dir directory, e.g. /etc/lvm/profile/. Once
780 attached to a VG or LV, lvm will process the VG or LV using the set‐
781 tings from the attached profile. A profile is named and referenced by
782 its file name.
783 To use a profile to customize the lvextend settings for an LV:
785 · Create a file containing settings, saved in profile_dir. For the
788 profile_dir location, run:
789 lvmconfig config/profile_dir
790 · Attach the profile to an LV, using the command:
793 lvchange --metadataprofile ProfileName VG/ThinPoolLV
794 · Extend the LV using the profile settings:
797 lvextend --use-policies VG/ThinPoolLV
798 Example
801 # lvmconfig config/profile_dir
802 profile_dir="/etc/lvm/profile"
803 # cat /etc/lvm/profile/pool0extend.profile
805 activation {
806 thin_pool_autoextend_threshold=50
807 thin_pool_autoextend_percent=10
808 }
809 # lvchange --metadataprofile pool0extend vg/pool0
811 # lvextend --use-policies vg/pool0
813 Notes
815 · A profile is attached to a VG or LV by name, where the name refer‐
817 ences a local file in profile_dir. If the VG is moved to another
818 machine, the file with the profile also needs to be moved.
819 · Only certain settings can be used in a VG or LV profile, see:
822 lvmconfig --type profilable-metadata.
823 · An LV without a profile of its own will inherit the VG profile.
826 · Remove a profile from an LV using the command:
829 lvchange --detachprofile VG/ThinPoolLV.
830 · Commands can also have profiles applied to them. The settings that
833 can be applied to a command are different than the settings that can
834 be applied to a VG or LV. See lvmconfig --type profilable-command.
835 To apply a profile to a command, write a profile, save it in the pro‐
836 file directory, and run the command using the option: --commandpro‐
837 file ProfileName.
838 Zeroing
842 When a thin pool provisions a new data block for a thin LV, the new
845 block is first overwritten with zeros. The zeroing mode is indicated
846 by the "z" attribute displayed by lvs. The option -Z (or --zero) can
847 be added to commands to specify the zeroing mode.
848 Command to set the zeroing mode when creating a thin pool LV:
850 lvconvert --type thin-pool -Z{y|n}
851 --poolmetadata VG/ThinMetaLV VG/ThinDataLV
852 Command to change the zeroing mode of an existing thin pool LV:
854 lvchange -Z{y|n} VG/ThinPoolLV
855 If zeroing mode is changed from "n" to "y", previously provisioned
857 blocks are not zeroed.
858 Provisioning of large zeroed chunks impacts performance.
860 lvm.conf(5) thin_pool_zero
862 controls the default zeroing mode used when creating a thin pool.
863 Discard
867 The discard behavior of a thin pool LV determines how discard requests
870 are handled. Enabling discard under a file system may adversely affect
871 the file system performance (see the section on fstrim for an alterna‐
872 tive.) Possible discard behaviors:
873 ignore: Ignore any discards that are received.
875 nopassdown: Process any discards in the thin pool itself and allow the
877 no longer needed extents to be overwritten by new data.
878 passdown: Process discards in the thin pool (as with nopassdown), and
880 pass the discards down the the underlying device. This is the default
881 mode.
882 Command to display the current discard mode of a thin pool LV:
884 lvs -o+discards VG/ThinPoolLV
885 Command to set the discard mode when creating a thin pool LV:
887 lvconvert --discards {ignore|nopassdown|passdown}
888 --type thin-pool --poolmetadata VG/ThinMetaLV VG/ThinDataLV
889 Command to change the discard mode of an existing thin pool LV:
891 lvchange --discards {ignore|nopassdown|passdown} VG/ThinPoolLV
892 Example
894 # lvs -o name,discards vg/pool0
895 pool0 passdown
896 # lvchange --discards ignore vg/pool0
898 lvm.conf(5) thin_pool_discards
900 controls the default discards mode used when creating a thin pool.
901 Chunk size
905 The size of data blocks managed by a thin pool can be specified with
908 the --chunksize option when the thin pool LV is created. The default
909 unit is KiB. The value must be a multiple of 64KiB between 64KiB and
910 1GiB.
911 When a thin pool is used primarily for the thin provisioning feature, a
913 larger value is optimal. To optimize for many snapshots, a smaller
914 value reduces copying time and consumes less space.
915 Command to display the thin pool LV chunk size:
917 lvs -o+chunksize VG/ThinPoolLV
918 Example
920 # lvs -o name,chunksize
921 pool0 64.00k
922 lvm.conf(5) thin_pool_chunk_size
924 controls the default chunk size used when creating a thin pool.
925 The default value is shown by:
927 lvmconfig --type default allocation/thin_pool_chunk_size
928 Size of pool metadata LV
932 The amount of thin metadata depends on how many blocks are shared
935 between thin LVs (i.e. through snapshots). A thin pool with many snap‐
936 shots may need a larger metadata LV. Thin pool metadata LV sizes can
937 be from 2MiB to 16GiB.
938 When using lvcreate to create what will become a thin metadata LV, the
940 size is specified with the -L--size option.
941 When an LVM command automatically creates a thin metadata LV, the size
943 is specified with the --poolmetadatasize option. When this option is
944 not given, LVM automatically chooses a size based on the data size and
945 chunk size.
946 It can be hard to predict the amount of metadata space that will be
948 needed, so it is recommended to start with a size of 1GiB which should
949 be enough for all practical purposes. A thin pool metadata LV can
950 later be manually or automatically extended if needed.
951 Create a thin snapshot of an external, read only LV
955 Thin snapshots are typically taken of other thin LVs or other thin
958 snapshot LVs within the same thin pool. It is also possible to take
959 thin snapshots of external, read only LVs. Writes to the snapshot are
960 stored in the thin pool, and the external LV is used to read unwritten
961 parts of the thin snapshot.
962 lvcreate -n SnapLV -s VG/ExternalOriginLV --thinpool VG/ThinPoolLV
964 Example
966 # lvchange -an vg/lve
967 # lvchange --permission r vg/lve
968 # lvcreate -n snaplve -s vg/lve --thinpool vg/pool0
969 # lvs vg/lve vg/snaplve
971 LV VG Attr LSize Pool Origin Data%
972 lve vg ori------- 10.00g
973 snaplve vg Vwi-a-tz-- 10.00g pool0 lve 0.00
974 Convert a standard LV to a thin LV with an external origin
978 A new thin LV can be created and given the name of an existing standard
981 LV. At the same time, the existing LV is converted to a read only
982 external LV with a new name. Unwritten portions of the thin LV are
983 read from the external LV. The new name given to the existing LV can
984 be specified with --originname, otherwise the existing LV will be given
985 a default name, e.g. lvol#.
986 Convert ExampleLV into a read only external LV with the new name NewEx‐
988 ternalOriginLV, and create a new thin LV that is given the previous
989 name of ExampleLV.
990 lvconvert --type thin --thinpool VG/ThinPoolLV
992 --originname NewExternalOriginLV VG/ExampleLV
993 Example
995 # lvcreate -n lv_example -L 10G vg
996 # lvs
998 lv_example vg -wi-a----- 10.00g
999 # lvconvert --type thin --thinpool vg/pool0
1001 --originname lv_external --thin vg/lv_example
1002 # lvs
1004 LV VG Attr LSize Pool Origin
1005 lv_example vg Vwi-a-tz-- 10.00g pool0 lv_external
1006 lv_external vg ori------- 10.00g
1007 Single step thin pool LV creation
1011 A thin pool LV can be created with a single lvcreate command, rather
1014 than using lvconvert on existing LVs. This one command creates a thin
1015 data LV, a thin metadata LV, and combines the two into a thin pool LV.
1016 lvcreate --type thin-pool -L LargeSize -n ThinPoolLV VG
1018 Example
1020 # lvcreate --type thin-pool -L8M -n pool0 vg
1021 # lvs vg/pool0
1023 LV VG Attr LSize Pool Origin Data%
1024 pool0 vg twi-a-tz-- 8.00m 0.00
1025 # lvs -a
1027 pool0 vg twi-a-tz-- 8.00m
1028 [pool0_tdata] vg Twi-ao---- 8.00m
1029 [pool0_tmeta] vg ewi-ao---- 8.00m
1030 Single step thin pool LV and thin LV creation
1034 A thin pool LV and a thin LV can be created with a single lvcreate com‐
1037 mand. This one command creates a thin data LV, a thin metadata LV,
1038 combines the two into a thin pool LV, and creates a thin LV in the new
1039 pool.
1040 -L LargeSize specifies the physical size of the thin pool LV.
1041 -V VirtualSize specifies the virtual size of the thin LV.
1042 lvcreate --type thin -V VirtualSize -L LargeSize
1044 -n ThinLV --thinpool VG/ThinPoolLV
1045 Equivalent to:
1047 lvcreate --type thin-pool -L LargeSize VG/ThinPoolLV
1048 lvcreate -n ThinLV -V VirtualSize --thinpool VG/ThinPoolLV
1049 Example
1051 # lvcreate -L8M -V2G -n thin1 --thinpool vg/pool0
1052 # lvs -a
1054 pool0 vg twi-a-tz-- 8.00m
1055 [pool0_tdata] vg Twi-ao---- 8.00m
1056 [pool0_tmeta] vg ewi-ao---- 8.00m
1057 thin1 vg Vwi-a-tz-- 2.00g pool0
1058 Merge thin snapshots
1062 A thin snapshot can be merged into its origin thin LV using the lvcon‐
1065 vert --merge command. The result of a snapshot merge is that the ori‐
1066 gin thin LV takes the content of the snapshot LV, and the snapshot LV
1067 is removed. Any content that was unique to the origin thin LV is lost
1068 after the merge.
1069 Because a merge changes the content of an LV, it cannot be done while
1071 the LVs are open, e.g. mounted. If a merge is initiated while the LVs
1072 are open, the effect of the merge is delayed until the origin thin LV
1073 is next activated.
1074 lvconvert --merge VG/SnapLV
1076 Example
1078 # lvs vg
1079 LV VG Attr LSize Pool Origin
1080 pool0 vg twi-a-tz-- 10.00g
1081 thin1 vg Vwi-a-tz-- 100.00g pool0
1082 thin1s1 vg Vwi-a-tz-k 100.00g pool0 thin1
1083 # lvconvert --merge vg/thin1s1
1085 # lvs vg
1087 LV VG Attr LSize Pool Origin
1088 pool0 vg twi-a-tz-- 10.00g
1089 thin1 vg Vwi-a-tz-- 100.00g pool0
1090 Example
1092 Delayed merging of open LVs.
1093 # lvs vg
1095 LV VG Attr LSize Pool Origin
1096 pool0 vg twi-a-tz-- 10.00g
1097 thin1 vg Vwi-aotz-- 100.00g pool0
1098 thin1s1 vg Vwi-aotz-k 100.00g pool0 thin1
1099 # df
1101 /dev/mapper/vg-thin1 100G 33M 100G 1% /mnt/X
1102 /dev/mapper/vg-thin1s1 100G 33M 100G 1% /mnt/Xs
1103 # ls /mnt/X
1105 file1 file2 file3
1106 # ls /mnt/Xs
1107 file3 file4 file5
1108 # lvconvert --merge vg/thin1s1
1110 Logical volume vg/thin1s1 contains a filesystem in use.
1111 Delaying merge since snapshot is open.
1112 Merging of thin snapshot thin1s1 will occur on next activation.
1113 # umount /mnt/X
1115 # umount /mnt/Xs
1116 # lvs -a vg
1118 LV VG Attr LSize Pool Origin
1119 pool0 vg twi-a-tz-- 10.00g
1120 [pool0_tdata] vg Twi-ao---- 10.00g
1121 [pool0_tmeta] vg ewi-ao---- 1.00g
1122 thin1 vg Owi-a-tz-- 100.00g pool0
1123 [thin1s1] vg Swi-a-tz-k 100.00g pool0 thin1
1124 # lvchange -an vg/thin1
1126 # lvchange -ay vg/thin1
1127 # mount /dev/vg/thin1 /mnt/X
1129 # ls /mnt/X
1131 file3 file4 file5
1132 XFS on snapshots
1136 Mounting an XFS file system on a new snapshot LV requires attention to
1139 the file system's log state and uuid. On the snapshot LV, the xfs log
1140 will contain a dummy transaction, and the xfs uuid will match the uuid
1141 from the file system on the origin LV.
1142 If the snapshot LV is writable, mounting will recover the log to clear
1144 the dummy transaction, but will require skipping the uuid check:
1145 mount /dev/VG/SnapLV /mnt -o nouuid
1147 Or, the uuid can be changed on disk before mounting:
1149 xfs_admin -U generate /dev/VG/SnapLV
1151 mount /dev/VG/SnapLV /mnt
1152 If the snapshot LV is readonly, the log recovery and uuid check need to
1154 be skipped while mounting readonly:
1155 mount /dev/VG/SnapLV /mnt -o ro,nouuid,norecovery
1157
1160 lvm(8), lvm.conf(5), lvmconfig(8), lvcreate(8), lvconvert(8),
1161 lvchange(8), lvextend(8), lvremove(8), lvs(8), thin_dump(8),
1162 thin_repair(8) thin_restore(8)
1163Red Hat, Inc LVM TOOLS 2.02.183(2) (2018-12-07) LVMTHIN(7)