1LVEXTEND(8) System Manager's Manual LVEXTEND(8)
2
6 lvextend — Add space to a logical volume
7
9 lvextend option_args position_args
10 [ option_args ]
11 [ position_args ]
12 --alloc contiguous|cling|cling_by_tags|normal|anywhere|inherit
14 -A|--autobackup y|n
15 --commandprofile String
16 --config String
17 -d|--debug
18 --devices PV
19 --devicesfile String
20 --driverloaded y|n
21 -l|--extents [+]Number[PERCENT]
22 -f|--force
23 --fs String
24 --fsmode String
25 -h|--help
26 --journal String
27 --lockopt String
28 --longhelp
29 -m|--mirrors Number
30 -n|--nofsck
31 --nohints
32 --nolocking
33 --nosync
34 --noudevsync
35 --poolmetadatasize [+]Size[m|UNIT]
36 --profile String
37 -q|--quiet
38 --reportformat basic|json|json_std
39 -r|--resizefs
40 -L|--size [+]Size[m|UNIT]
41 -i|--stripes Number
42 -I|--stripesize Size[k|UNIT]
43 -t|--test
44 --type linear|striped|snapshot|raid|mirror|thin|thin-pool|vdo|
45 vdo-pool|cache|cache-pool|writecache
46 --usepolicies
47 -v|--verbose
48 --version
49 -y|--yes
50
52 lvextend extends the size of an LV. This requires allocating logical
53 extents from the VG's free physical extents. If the extension adds a
54 new LV segment, the new segment will use the existing segment type of
55 the LV.
56 Extending a copy-on-write snapshot LV adds space for COW blocks.
58 Use lvconvert(8) to change the number of data images in a RAID or mir‐
60 rored LV.
61 In the usage section below, --size Size can be replaced with --extents
63 Number. See both descriptions the options section.
64
66 Extend an LV by a specified size.
67 lvextend -L|--size [+]Size[m|UNIT] LV
69 [ -l|--extents [+]Number[PERCENT] ]
70 [ -r|--resizefs ]
71 [ -i|--stripes Number ]
72 [ -I|--stripesize Size[k|UNIT] ]
73 [ --poolmetadatasize [+]Size[m|UNIT] ]
74 [ --fs String ]
75 [ --fsmode String ]
76 [ COMMON_OPTIONS ]
77 [ PV ... ]
78 —
80 Extend an LV by specified PV extents.
82 lvextend LV PV ...
84 [ -r|--resizefs ]
85 [ -i|--stripes Number ]
86 [ -I|--stripesize Size[k|UNIT] ]
87 [ --fs String ]
88 [ --fsmode String ]
89 [ COMMON_OPTIONS ]
90 —
92 Extend a pool metadata SubLV by a specified size.
94 lvextend --poolmetadatasize [+]Size[m|UNIT] LV1
96 [ -i|--stripes Number ]
97 [ -I|--stripesize Size[k|UNIT] ]
98 [ COMMON_OPTIONS ]
99 [ PV ... ]
100 LV1 types: linear thinpool
102 —
104 Extend an LV according to a predefined policy.
106 lvextend --usepolicies LV1
108 [ -r|--resizefs ]
109 [ --fs String ]
110 [ --fsmode String ]
111 [ COMMON_OPTIONS ]
112 [ PV ... ]
113 LV1 types: snapshot thinpool vdopool
115 —
117 Common options for command:
119 [ -A|--autobackup y|n ]
120 [ -f|--force ]
121 [ -m|--mirrors Number ]
122 [ -n|--nofsck ]
123 [ --alloc contiguous|cling|cling_by_tags|normal|anywhere|inherit
124 ]
125 [ --nosync ]
126 [ --noudevsync ]
127 [ --reportformat basic|json|json_std ]
128 [ --type linear|striped|snapshot|raid|mirror|thin|thin-pool|vdo|
129 vdo-pool|cache|cache-pool|writecache ]
130 Common options for lvm:
132 [ -d|--debug ]
133 [ -h|--help ]
134 [ -q|--quiet ]
135 [ -t|--test ]
136 [ -v|--verbose ]
137 [ -y|--yes ]
138 [ --commandprofile String ]
139 [ --config String ]
140 [ --devices PV ]
141 [ --devicesfile String ]
142 [ --driverloaded y|n ]
143 [ --journal String ]
144 [ --lockopt String ]
145 [ --longhelp ]
146 [ --nohints ]
147 [ --nolocking ]
148 [ --profile String ]
149 [ --version ]
150
152 --alloc contiguous|cling|cling_by_tags|normal|anywhere|inherit
153 Determines the allocation policy when a command needs to allo‐
154 cate Physical Extents (PEs) from the VG. Each VG and LV has an
155 allocation policy which can be changed with vgchange/lvchange,
156 or overridden on the command line. normal applies common sense
157 rules such as not placing parallel stripes on the same PV. in‐
158 herit applies the VG policy to an LV. contiguous requires new
159 PEs be placed adjacent to existing PEs. cling places new PEs on
160 the same PV as existing PEs in the same stripe of the LV. If
161 there are sufficient PEs for an allocation, but normal does not
162 use them, anywhere will use them even if it reduces performance,
163 e.g. by placing two stripes on the same PV. Optional positional
164 PV args on the command line can also be used to limit which PVs
165 the command will use for allocation. See lvm(8) for more infor‐
166 mation about allocation.
167 -A|--autobackup y|n
169 Specifies if metadata should be backed up automatically after a
170 change. Enabling this is strongly advised! See vgcfgbackup(8)
171 for more information.
172 --commandprofile String
174 The command profile to use for command configuration. See
175 lvm.conf(5) for more information about profiles.
176 --config String
178 Config settings for the command. These override lvm.conf(5) set‐
179 tings. The String arg uses the same format as lvm.conf(5), or
180 may use section/field syntax. See lvm.conf(5) for more informa‐
181 tion about config.
182 -d|--debug ...
184 Set debug level. Repeat from 1 to 6 times to increase the detail
185 of messages sent to the log file and/or syslog (if configured).
186 --devices PV
188 Restricts the devices that are visible and accessible to the
189 command. Devices not listed will appear to be missing. This op‐
190 tion can be repeated, or accepts a comma separated list of de‐
191 vices. This overrides the devices file.
192 --devicesfile String
194 A file listing devices that LVM should use. The file must exist
195 in /etc/lvm/devices/ and is managed with the lvmdevices(8) com‐
196 mand. This overrides the lvm.conf(5) devices/devicesfile and
197 devices/use_devicesfile settings.
198 --driverloaded y|n
200 If set to no, the command will not attempt to use device-mapper.
201 For testing and debugging.
202 -l|--extents [+]Number[PERCENT]
204 Specifies the new size of the LV in logical extents. The --size
205 and --extents options are alternate methods of specifying size.
206 The total number of physical extents used will be greater when
207 redundant data is needed for RAID levels. An alternate syntax
208 allows the size to be determined indirectly as a percentage of
209 the size of a related VG, LV, or set of PVs. The suffix %VG de‐
210 notes the total size of the VG, the suffix %FREE the remaining
211 free space in the VG, and the suffix %PVS the free space in the
212 specified PVs. For a snapshot, the size can be expressed as a
213 percentage of the total size of the origin LV with the suffix
214 %ORIGIN (100%ORIGIN provides space for the whole origin). When
215 expressed as a percentage, the size defines an upper limit for
216 the number of logical extents in the new LV. The precise number
217 of logical extents in the new LV is not determined until the
218 command has completed. When the plus + or minus - prefix is
219 used, the value is not an absolute size, but is relative and
220 added or subtracted from the current size.
221 -f|--force ...
223 Override various checks, confirmations and protections. Use
224 with extreme caution.
225 --fs String
227 Control file system resizing when resizing an LV. checksize:
228 Check the fs size and reduce the LV if the fs is not using the
229 reduced space (fs reduce is not needed.) If the reduced space is
230 used by the fs, then do not resize the fs or LV, and return an
231 error. (checksize only applies when reducing, and does nothing
232 for extend.) resize: Resize the fs by calling the fs-specific
233 resize command. This may also include mounting, unmounting, or
234 running fsck. See --fsmode to control mounting behavior, and
235 --nofsck to disable fsck. resize_fsadm: Use the old method of
236 calling fsadm to handle the fs (deprecated.) Warning: this op‐
237 tion does not prevent lvreduce from destroying file systems that
238 are unmounted (or mounted if prompts are skipped.) ignore: Re‐
239 size the LV without checking for or handling a file system.
240 Warning: using ignore when reducing the LV size may destroy the
241 file system.
242 --fsmode String
244 Control file system mounting behavior for fs resize. manage:
245 Mount or unmount the fs as needed to resize the fs, and attempt
246 to restore the original mount state at the end. nochange: Do
247 not mount or unmount the fs. If mounting or unmounting is re‐
248 quired to resize the fs, then do not resize the fs or the LV and
249 fail the command. offline: Unmount the fs if it is mounted, and
250 resize the fs while it is unmounted. If mounting is required to
251 resize the fs, then do not resize the fs or the LV and fail the
252 command.
253 -h|--help
255 Display help text.
256 --journal String
258 Record information in the systemd journal. This information is
259 in addition to information enabled by the lvm.conf log/journal
260 setting. command: record information about the command. out‐
261 put: record the default command output. debug: record full com‐
262 mand debugging.
263 --lockopt String
265 Used to pass options for special cases to lvmlockd. See lvm‐
266 lockd(8) for more information.
267 --longhelp
269 Display long help text.
270 -m|--mirrors Number
272 Not used.
273 -n|--nofsck
275 Do not perform fsck when resizing the file system with --re‐
276 sizefs.
277 --nohints
279 Do not use the hints file to locate devices for PVs. A command
280 may read more devices to find PVs when hints are not used. The
281 command will still perform standard hint file invalidation where
282 appropriate.
283 --nolocking
285 Disable locking. Use with caution, concurrent commands may pro‐
286 duce incorrect results.
287 --nosync
289 Causes the creation of mirror, raid1, raid4, raid5 and raid10 to
290 skip the initial synchronization. In case of mirror, raid1 and
291 raid10, any data written afterwards will be mirrored, but the
292 original contents will not be copied. In case of raid4 and
293 raid5, no parity blocks will be written, though any data written
294 afterwards will cause parity blocks to be stored. This is use‐
295 ful for skipping a potentially long and resource intensive ini‐
296 tial sync of an empty mirror/raid1/raid4/raid5 and raid10 LV.
297 This option is not valid for raid6, because raid6 relies on
298 proper parity (P and Q Syndromes) being created during initial
299 synchronization in order to reconstruct proper user date in case
300 of device failures. raid0 and raid0_meta do not provide any da‐
301 ta copies or parity support and thus do not support initial syn‐
302 chronization.
303 --noudevsync
305 Disables udev synchronization. The process will not wait for no‐
306 tification from udev. It will continue irrespective of any pos‐
307 sible udev processing in the background. Only use this if udev
308 is not running or has rules that ignore the devices LVM creates.
309 --poolmetadatasize [+]Size[m|UNIT]
311 Specifies the new size of the pool metadata LV. The plus prefix
312 + can be used, in which case the value is added to the current
313 size.
314 --profile String
316 An alias for --commandprofile or --metadataprofile, depending on
317 the command.
318 -q|--quiet ...
320 Suppress output and log messages. Overrides --debug and --ver‐
321 bose. Repeat once to also suppress any prompts with answer
322 'no'.
323 --reportformat basic|json|json_std
325 Overrides current output format for reports which is defined
326 globally by the report/output_format setting in lvm.conf(5).
327 basic is the original format with columns and rows. If there is
328 more than one report per command, each report is prefixed with
329 the report name for identification. json produces report output
330 in JSON format. json_std produces report output in JSON format
331 which is more compliant with JSON standard. See lvmreport(7)
332 for more information.
333 -r|--resizefs
335 Resize the fs using the fs-specific resize command. May include
336 mounting, unmounting, or running fsck. See --fsmode to control
337 mounting behavior, and --nofsck to disable fsck. See --fs for
338 more options (--resizefs is equivalent to --fs resize.)
339 -L|--size [+]Size[m|UNIT]
341 Specifies the new size of the LV. The --size and --extents op‐
342 tions are alternate methods of specifying size. The total num‐
343 ber of physical extents used will be greater when redundant data
344 is needed for RAID levels. When the plus + or minus - prefix is
345 used, the value is not an absolute size, but is relative and
346 added or subtracted from the current size.
347 -i|--stripes Number
349 Specifies the number of stripes in a striped LV. This is the
350 number of PVs (devices) that a striped LV is spread across. Data
351 that appears sequential in the LV is spread across multiple de‐
352 vices in units of the stripe size (see --stripesize). This does
353 not change existing allocated space, but only applies to space
354 being allocated by the command. When creating a RAID 4/5/6 LV,
355 this number does not include the extra devices that are required
356 for parity. The largest number depends on the RAID type (raid0:
357 64, raid10: 32, raid4/5: 63, raid6: 62), and when unspecified,
358 the default depends on the RAID type (raid0: 2, raid10: 2,
359 raid4/5: 3, raid6: 5.) To stripe a new raid LV across all PVs
360 by default, see lvm.conf(5) allocation/raid_stripe_all_devices.
361 -I|--stripesize Size[k|UNIT]
363 The amount of data that is written to one device before moving
364 to the next in a striped LV.
365 -t|--test
367 Run in test mode. Commands will not update metadata. This is
368 implemented by disabling all metadata writing but nevertheless
369 returning success to the calling function. This may lead to un‐
370 usual error messages in multi-stage operations if a tool relies
371 on reading back metadata it believes has changed but hasn't.
372 --type linear|striped|snapshot|raid|mirror|thin|thin-pool|vdo|vdo-pool|
374 cache|cache-pool|writecache
375 The LV type, also known as "segment type" or "segtype". See us‐
376 age descriptions for the specific ways to use these types. For
377 more information about redundancy and performance (raid<N>, mir‐
378 ror, striped, linear) see lvmraid(7). For thin provisioning
379 (thin, thin-pool) see lvmthin(7). For performance caching
380 (cache, cache-pool) see lvmcache(7). For copy-on-write snap‐
381 shots (snapshot) see usage definitions. For VDO (vdo) see lvmv‐
382 do(7). Several commands omit an explicit type option because
383 the type is inferred from other options or shortcuts (e.g.
384 --stripes, --mirrors, --snapshot, --virtualsize, --thin,
385 --cache, --vdo). Use inferred types with care because it can
386 lead to unexpected results.
387 --usepolicies
389 Perform an operation according to the policy configured in
390 lvm.conf(5) or a profile.
391 -v|--verbose ...
393 Set verbose level. Repeat from 1 to 4 times to increase the de‐
394 tail of messages sent to stdout and stderr.
395 --version
397 Display version information.
398 -y|--yes
400 Do not prompt for confirmation interactively but always assume
401 the answer yes. Use with extreme caution. (For automatic no,
402 see -qq.)
403
405 LV Logical Volume name. See lvm(8) for valid names. An LV posi‐
406 tional arg generally includes the VG name and LV name, e.g.
407 VG/LV. LV1 indicates the LV must have a specific type, where
408 the accepted LV types are listed. (raid represents raid<N>
409 type).
410 PV Physical Volume name, a device path under /dev. For commands
412 managing physical extents, a PV positional arg generally accepts
413 a suffix indicating a range (or multiple ranges) of physical ex‐
414 tents (PEs). When the first PE is omitted, it defaults to the
415 start of the device, and when the last PE is omitted it defaults
416 to end. Start and end range (inclusive): PV[:PE-PE]... Start
417 and length range (counting from 0): PV[:PE+PE]...
418 String See the option description for information about the string con‐
420 tent.
421 Size[UNIT]
423 Size is an input number that accepts an optional unit. Input
424 units are always treated as base two values, regardless of capi‐
425 talization, e.g. 'k' and 'K' both refer to 1024. The default
426 input unit is specified by letter, followed by |UNIT. UNIT rep‐
427 resents other possible input units: b|B is bytes, s|S is sectors
428 of 512 bytes, k|K is KiB, m|M is MiB, g|G is GiB, t|T is TiB,
429 p|P is PiB, e|E is EiB. (This should not be confused with the
430 output control --units, where capital letters mean multiple of
431 1000.)
432
434 See lvm(8) for information about environment variables used by lvm.
435 For example, LVM_VG_NAME can generally be substituted for a required VG
436 parameter.
437
439 Extend the size of an LV by 54MiB, using a specific PV.
440 lvextend -L +54 vg01/lvol10 /dev/sdk3
441 Extend the size of an LV by the amount of free space on PV /dev/sdk3.
443 This is equivalent to specifying "-l +100%PVS" on the command line.
444 lvextend vg01/lvol01 /dev/sdk3
445 Extend an LV by 16MiB using specific physical extents.
447 lvextend -L+16m vg01/lvol01 /dev/sda:8-9 /dev/sdb:8-9
448 Extend an LV to use all remaining free space in volume group and all
450 resize its filesystem with fsadm(8).
451 lvextend -l+100%FREE -r vg01/lvol01
452
454 lvm(8), lvm.conf(5), lvmconfig(8), lvmdevices(8),
455 pvchange(8), pvck(8), pvcreate(8), pvdisplay(8), pvmove(8),
457 pvremove(8), pvresize(8), pvs(8), pvscan(8),
458 vgcfgbackup(8), vgcfgrestore(8), vgchange(8), vgck(8), vgcreate(8),
460 vgconvert(8), vgdisplay(8), vgexport(8), vgextend(8), vgimport(8),
461 vgimportclone(8), vgimportdevices(8), vgmerge(8), vgmknodes(8),
462 vgreduce(8), vgremove(8), vgrename(8), vgs(8), vgscan(8), vgsplit(8),
463 lvcreate(8), lvchange(8), lvconvert(8), lvdisplay(8), lvextend(8),
465 lvreduce(8), lvremove(8), lvrename(8), lvresize(8), lvs(8), lvscan(8),
466 lvm-fullreport(8), lvm-lvpoll(8), blkdeactivate(8), lvmdump(8),
468 dmeventd(8), lvmpolld(8), lvmlockd(8), lvmlockctl(8), cmirrord(8),
470 lvmdbusd(8), fsadm(8),
471 lvmsystemid(7), lvmreport(7), lvmcache(7), lvmraid(7), lvmthin(7),
473 lvmvdo(7), lvmautoactivation(7)
474Red Hat, Inc. LVM TOOLS 2.03.22(2) (2023-08-02) LVEXTEND(8)