2 .\" Copyright Neil Brown and others.
3 .\" This program is free software; you can redistribute it and/or modify
4 .\" it under the terms of the GNU General Public License as published by
5 .\" the Free Software Foundation; either version 2 of the License, or
6 .\" (at your option) any later version.
7 .\" See file COPYING in distribution for details.
10 mdadm \- manage MD devices
16 .BI mdadm " [mode] <raiddevice> [options] <component-devices>"
19 RAID devices are virtual devices created from two or more
20 real block devices. This allows multiple devices (typically disk
21 drives or partitions thereof) to be combined into a single device to
22 hold (for example) a single filesystem.
23 Some RAID levels include redundancy and so can survive some degree of
26 Linux Software RAID devices are implemented through the md (Multiple
27 Devices) device driver.
29 Currently, Linux supports
46 is not a Software RAID mechanism, but does involve
48 each device is a path to one common physical storage device.
49 New installations should not use md/multipath as it is not well
50 supported and has no ongoing development. Use the Device Mapper based
51 multipath-tools instead.
54 is also not true RAID, and it only involves one device. It
55 provides a layer over a true device that can be used to inject faults.
60 is a collection of devices that are
61 managed as a set. This is similar to the set of devices connected to
62 a hardware RAID controller. The set of devices may contain a number
63 of different RAID arrays each utilising some (or all) of the blocks from a
64 number of the devices in the set. For example, two devices in a 5-device set
65 might form a RAID1 using the whole devices. The remaining three might
66 have a RAID5 over the first half of each device, and a RAID0 over the
71 there is one set of metadata that describes all of
72 the arrays in the container. So when
76 device, the device just represents the metadata. Other normal arrays (RAID1
77 etc) can be created inside the container.
80 mdadm has several major modes of operation:
83 Assemble the components of a previously created
84 array into an active array. Components can be explicitly given
85 or can be searched for.
87 checks that the components
88 do form a bona fide array, and can, on request, fiddle superblock
89 information so as to assemble a faulty array.
93 Build an array that doesn't have per-device metadata (superblocks). For these
96 cannot differentiate between initial creation and subsequent assembly
97 of an array. It also cannot perform any checks that appropriate
98 components have been requested. Because of this, the
100 mode should only be used together with a complete understanding of
105 Create a new array with per-device metadata (superblocks).
106 Appropriate metadata is written to each device, and then the array
107 comprising those devices is activated. A 'resync' process is started
108 to make sure that the array is consistent (e.g. both sides of a mirror
109 contain the same data) but the content of the device is left otherwise
111 The array can be used as soon as it has been created. There is no
112 need to wait for the initial resync to finish.
115 .B "Follow or Monitor"
116 Monitor one or more md devices and act on any state changes. This is
117 only meaningful for RAID1, 4, 5, 6, 10 or multipath arrays, as
118 only these have interesting state. RAID0 or Linear never have
119 missing, spare, or failed drives, so there is nothing to monitor.
123 Grow (or shrink) an array, or otherwise reshape it in some way.
124 Currently supported growth options including changing the active size
125 of component devices and changing the number of active devices in RAID
126 levels 1/4/5/6, changing the RAID level between 1, 5, and 6, changing
127 the chunk size and layout for RAID5 and RAID5, as well as adding or
128 removing a write-intent bitmap.
131 .B "Incremental Assembly"
132 Add a single device to an appropriate array. If the addition of the
133 device makes the array runnable, the array will be started.
134 This provides a convenient interface to a
136 system. As each device is detected,
138 has a chance to include it in some array as appropriate.
141 flag is passed in we will remove the device from any active array
142 instead of adding it.
148 in this mode, then any arrays within that container will be assembled
153 This is for doing things to specific components of an array such as
154 adding new spares and removing faulty devices.
158 This is an 'everything else' mode that supports operations on active
159 arrays, operations on component devices such as erasing old superblocks, and
160 information gathering operations.
161 .\"This mode allows operations on independent devices such as examine MD
162 .\"superblocks, erasing old superblocks and stopping active arrays.
166 This mode does not act on a specific device or array, but rather it
167 requests the Linux Kernel to activate any auto-detected arrays.
170 .SH Options for selecting a mode are:
173 .BR \-A ", " \-\-assemble
174 Assemble a pre-existing array.
177 .BR \-B ", " \-\-build
178 Build a legacy array without superblocks.
181 .BR \-C ", " \-\-create
185 .BR \-F ", " \-\-follow ", " \-\-monitor
191 .BR \-G ", " \-\-grow
192 Change the size or shape of an active array.
195 .BR \-I ", " \-\-incremental
196 Add/remove a single device to/from an appropriate array, and possibly start the array.
200 Request that the kernel starts any auto-detected arrays. This can only
203 is compiled into the kernel \(em not if it is a module.
204 Arrays can be auto-detected by the kernel if all the components are in
205 primary MS-DOS partitions with partition type
207 and all use v0.90 metadata.
208 In-kernel autodetect is not recommended for new installations. Using
210 to detect and assemble arrays \(em possibly in an
212 \(em is substantially more flexible and should be preferred.
215 If a device is given before any options, or if the first option is
220 then the MANAGE mode is assumed.
221 Anything other than these will cause the
225 .SH Options that are not mode-specific are:
228 .BR \-h ", " \-\-help
229 Display general help message or, after one of the above options, a
230 mode-specific help message.
234 Display more detailed help about command line parsing and some commonly
238 .BR \-V ", " \-\-version
239 Print version information for mdadm.
242 .BR \-v ", " \-\-verbose
243 Be more verbose about what is happening. This can be used twice to be
245 The extra verbosity currently only affects
246 .B \-\-detail \-\-scan
248 .BR "\-\-examine \-\-scan" .
251 .BR \-q ", " \-\-quiet
252 Avoid printing purely informative messages. With this,
254 will be silent unless there is something really important to report.
257 .BR \-f ", " \-\-force
258 Be more forceful about certain operations. See the various modes for
259 the exact meaning of this option in different contexts.
262 .BR \-c ", " \-\-config=
263 Specify the config file. Default is to use
264 .BR /etc/mdadm.conf ,
265 or if that is missing then
266 .BR /etc/mdadm/mdadm.conf .
267 If the config file given is
269 then nothing will be read, but
271 will act as though the config file contained exactly
272 .B "DEVICE partitions containers"
275 to find a list of devices to scan, and
277 to find a list of containers to examine.
280 is given for the config file, then
282 will act as though the config file were empty.
285 .BR \-s ", " \-\-scan
288 for missing information.
289 In general, this option gives
291 permission to get any missing information (like component devices,
292 array devices, array identities, and alert destination) from the
293 configuration file (see previous option);
294 one exception is MISC mode when using
300 says to get a list of array devices from
304 .BR \-e ", " \-\-metadata=
305 Declare the style of RAID metadata (superblock) to be used. The
306 default is {DEFAULT_METADATA} for
308 and to guess for other operations.
309 The default can be overridden by setting the
318 .ie '{DEFAULT_METADATA}'0.90'
319 .IP "0, 0.90, default"
323 Use the original 0.90 format superblock. This format limits arrays to
324 28 component devices and limits component devices of levels 1 and
325 greater to 2 terabytes. It is also possible for there to be confusion
326 about whether the superblock applies to a whole device or just the
327 last partition, if that partition starts on a 64K boundary.
328 .ie '{DEFAULT_METADATA}'0.90'
329 .IP "1, 1.0, 1.1, 1.2"
331 .IP "1, 1.0, 1.1, 1.2 default"
333 Use the new version-1 format superblock. This has fewer restrictions.
334 It can easily be moved between hosts with different endian-ness, and a
335 recovery operation can be checkpointed and restarted. The different
336 sub-versions store the superblock at different locations on the
337 device, either at the end (for 1.0), at the start (for 1.1) or 4K from
338 the start (for 1.2). "1" is equivalent to "1.0".
339 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
341 Use the "Industry Standard" DDF (Disk Data Format) format defined by
343 When creating a DDF array a
345 will be created, and normal arrays can be created in that container.
347 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
349 which is managed in a similar manner to DDF, and is supported by an
350 option-rom on some platforms:
352 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
358 This will override any
360 setting in the config file and provides the identity of the host which
361 should be considered the home for any arrays.
363 When creating an array, the
365 will be recorded in the metadata. For version-1 superblocks, it will
366 be prefixed to the array name. For version-0.90 superblocks, part of
367 the SHA1 hash of the hostname will be stored in the later half of the
370 When reporting information about an array, any array which is tagged
371 for the given homehost will be reported as such.
373 When using Auto-Assemble, only arrays tagged for the given homehost
374 will be allowed to use 'local' names (i.e. not ending in '_' followed
375 by a digit string). See below under
376 .BR "Auto Assembly" .
378 .SH For create, build, or grow:
381 .BR \-n ", " \-\-raid\-devices=
382 Specify the number of active devices in the array. This, plus the
383 number of spare devices (see below) must equal the number of
385 (including "\fBmissing\fP" devices)
386 that are listed on the command line for
388 Setting a value of 1 is probably
389 a mistake and so requires that
391 be specified first. A value of 1 will then be allowed for linear,
392 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
394 This number can only be changed using
396 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
397 the necessary support.
400 .BR \-x ", " \-\-spare\-devices=
401 Specify the number of spare (eXtra) devices in the initial array.
402 Spares can also be added
403 and removed later. The number of component devices listed
404 on the command line must equal the number of RAID devices plus the
405 number of spare devices.
408 .BR \-z ", " \-\-size=
409 Amount (in Kibibytes) of space to use from each drive in RAID levels 1/4/5/6.
410 This must be a multiple of the chunk size, and must leave about 128Kb
411 of space at the end of the drive for the RAID superblock.
412 If this is not specified
413 (as it normally is not) the smallest drive (or partition) sets the
414 size, though if there is a variance among the drives of greater than 1%, a warning is
417 This value can be set with
419 for RAID level 1/4/5/6. If the array was created with a size smaller
420 than the currently active drives, the extra space can be accessed
423 The size can be given as
425 which means to choose the largest size that fits on all current drives.
427 This value can not be used with
429 metadata such as DDF and IMSM.
432 .BR \-Z ", " \-\-array-size=
433 This is only meaningful with
435 and its effect is not persistent: when the array is stopped an
436 restarted the default array size will be restored.
438 Setting the array-size causes the array to appear smaller to programs
439 that access the data. This is particularly needed before reshaping an
440 array so that it will be smaller. As the reshape is not reversible,
441 but setting the size with
443 is, it is required that the array size is reduced as appropriate
444 before the number of devices in the array is reduced.
447 .BR \-c ", " \-\-chunk=
448 Specify chunk size of kibibytes. The default when creating an
449 array is 512KB. To ensure compatibility with earlier versions, the
450 default when Building and array with no persistent metadata is 64KB.
451 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
455 Specify rounding factor for a Linear array. The size of each
456 component will be rounded down to a multiple of this size.
457 This is a synonym for
459 but highlights the different meaning for Linear as compared to other
460 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
461 use, and is 0K (i.e. no rounding) in later kernels.
464 .BR \-l ", " \-\-level=
465 Set RAID level. When used with
467 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
468 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
469 Obviously some of these are synonymous.
473 metadata type is requested, only the
475 level is permitted, and it does not need to be explicitly given.
479 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
483 to change the RAID level in some cases. See LEVEL CHANGES below.
486 .BR \-p ", " \-\-layout=
487 This option configures the fine details of data layout for RAID5, RAID6,
488 and RAID10 arrays, and controls the failure modes for
491 The layout of the RAID5 parity block can be one of
492 .BR left\-asymmetric ,
493 .BR left\-symmetric ,
494 .BR right\-asymmetric ,
495 .BR right\-symmetric ,
496 .BR la ", " ra ", " ls ", " rs .
498 .BR left\-symmetric .
500 It is also possible to cause RAID5 to use a RAID4-like layout by
506 Finally for RAID5 there are DDF\-compatible layouts,
507 .BR ddf\-zero\-restart ,
508 .BR ddf\-N\-restart ,
510 .BR ddf\-N\-continue .
512 These same layouts are available for RAID6. There are also 4 layouts
513 that will provide an intermediate stage for converting between RAID5
514 and RAID6. These provide a layout which is identical to the
515 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
516 syndrome (the second 'parity' block used by RAID6) on the last device.
518 .BR left\-symmetric\-6 ,
519 .BR right\-symmetric\-6 ,
520 .BR left\-asymmetric\-6 ,
521 .BR right\-asymmetric\-6 ,
523 .BR parity\-first\-6 .
525 When setting the failure mode for level
528 .BR write\-transient ", " wt ,
529 .BR read\-transient ", " rt ,
530 .BR write\-persistent ", " wp ,
531 .BR read\-persistent ", " rp ,
533 .BR read\-fixable ", " rf ,
534 .BR clear ", " flush ", " none .
536 Each failure mode can be followed by a number, which is used as a period
537 between fault generation. Without a number, the fault is generated
538 once on the first relevant request. With a number, the fault will be
539 generated after that many requests, and will continue to be generated
540 every time the period elapses.
542 Multiple failure modes can be current simultaneously by using the
544 option to set subsequent failure modes.
546 "clear" or "none" will remove any pending or periodic failure modes,
547 and "flush" will clear any persistent faults.
549 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
550 by a small number. The default is 'n2'. The supported options are:
553 signals 'near' copies. Multiple copies of one data block are at
554 similar offsets in different devices.
557 signals 'offset' copies. Rather than the chunks being duplicated
558 within a stripe, whole stripes are duplicated but are rotated by one
559 device so duplicate blocks are on different devices. Thus subsequent
560 copies of a block are in the next drive, and are one chunk further
565 (multiple copies have very different offsets).
566 See md(4) for more detail about 'near', 'offset', and 'far'.
568 The number is the number of copies of each datablock. 2 is normal, 3
569 can be useful. This number can be at most equal to the number of
570 devices in the array. It does not need to divide evenly into that
571 number (e.g. it is perfectly legal to have an 'n2' layout for an array
572 with an odd number of devices).
574 When an array is converted between RAID5 and RAID6 an intermediate
575 RAID6 layout is used in which the second parity block (Q) is always on
576 the last device. To convert a RAID5 to RAID6 and leave it in this new
577 layout (which does not require re-striping) use
578 .BR \-\-layout=preserve .
579 This will try to avoid any restriping.
581 The converse of this is
582 .B \-\-layout=normalise
583 which will change a non-standard RAID6 layout into a more standard
590 (thus explaining the p of
594 .BR \-b ", " \-\-bitmap=
595 Specify a file to store a write-intent bitmap in. The file should not
598 is also given. The same file should be provided
599 when assembling the array. If the word
601 is given, then the bitmap is stored with the metadata on the array,
602 and so is replicated on all devices. If the word
606 mode, then any bitmap that is present is removed.
608 To help catch typing errors, the filename must contain at least one
609 slash ('/') if it is a real file (not 'internal' or 'none').
611 Note: external bitmaps are only known to work on ext2 and ext3.
612 Storing bitmap files on other filesystems may result in serious problems.
615 .BR \-\-bitmap\-chunk=
616 Set the chunksize of the bitmap. Each bit corresponds to that many
617 Kilobytes of storage.
618 When using a file based bitmap, the default is to use the smallest
619 size that is at-least 4 and requires no more than 2^21 chunks.
622 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
623 fit the bitmap into the available space.
626 .BR \-W ", " \-\-write\-mostly
627 subsequent devices listed in a
632 command will be flagged as 'write-mostly'. This is valid for RAID1
633 only and means that the 'md' driver will avoid reading from these
634 devices if at all possible. This can be useful if mirroring over a
638 .BR \-\-write\-behind=
639 Specify that write-behind mode should be enabled (valid for RAID1
640 only). If an argument is specified, it will set the maximum number
641 of outstanding writes allowed. The default value is 256.
642 A write-intent bitmap is required in order to use write-behind
643 mode, and write-behind is only attempted on drives marked as
647 .BR \-\-assume\-clean
650 that the array pre-existed and is known to be clean. It can be useful
651 when trying to recover from a major failure as you can be sure that no
652 data will be affected unless you actually write to the array. It can
653 also be used when creating a RAID1 or RAID10 if you want to avoid the
654 initial resync, however this practice \(em while normally safe \(em is not
655 recommended. Use this only if you really know what you are doing.
657 When the devices that will be part of a new array were filled
658 with zeros before creation the operator knows the array is
659 actually clean. If that is the case, such as after running
660 badblocks, this argument can be used to tell mdadm the
661 facts the operator knows.
664 .BR \-\-backup\-file=
667 is used to increase the number of raid-devices in a RAID5 or RAID6 if
668 there are no spare devices available, or to shrink, change RAID level
669 or layout. See the GROW MODE section below on RAID\-DEVICES CHANGES.
670 The file must be stored on a separate device, not on the RAID array
674 .BR \-\-array-size= ", " \-Z
675 Set the size of the array which is seen by users of the device such as
676 filesystems. This can be less that the real size, but never greater.
677 The size set this way does not persist across restarts of the array.
679 This is most useful when reducing the number of devices in a RAID5 or
680 RAID6. Such arrays require the array-size to be reduced before a
681 reshape can be performed that reduces the real size.
685 restores the apparent size of the array to be whatever the real
686 amount of available space is.
689 .BR \-N ", " \-\-name=
692 for the array. This is currently only effective when creating an
693 array with a version-1 superblock, or an array in a DDF container.
694 The name is a simple textual string that can be used to identify array
695 components when assembling. If name is needed but not specified, it
696 is taken from the basename of the device that is being created.
708 run the array, even if some of the components
709 appear to be active in another array or filesystem. Normally
711 will ask for confirmation before including such components in an
712 array. This option causes that question to be suppressed.
715 .BR \-f ", " \-\-force
718 accept the geometry and layout specified without question. Normally
720 will not allow creation of an array with only one device, and will try
721 to create a RAID5 array with one missing drive (as this makes the
722 initial resync work faster). With
725 will not try to be so clever.
728 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
729 Instruct mdadm how to create the device file if needed, possibly allocating
730 an unused minor number. "md" causes a non-partitionable array
731 to be used (though since Linux 2.6.28, these array devices are in fact
732 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
733 later) to be used. "yes" requires the named md device to have
734 a 'standard' format, and the type and minor number will be determined
735 from this. With mdadm 3.0, device creation is normally left up to
737 so this option is unlikely to be needed.
738 See DEVICE NAMES below.
740 The argument can also come immediately after
745 is not given on the command line or in the config file, then
751 is also given, then any
753 entries in the config file will override the
755 instruction given on the command line.
757 For partitionable arrays,
759 will create the device file for the whole array and for the first 4
760 partitions. A different number of partitions can be specified at the
761 end of this option (e.g.
763 If the device name ends with a digit, the partition names add a 'p',
765 .IR /dev/md/home1p3 .
766 If there is no trailing digit, then the partition names just have a
768 .IR /dev/md/scratch3 .
770 If the md device name is in a 'standard' format as described in DEVICE
771 NAMES, then it will be created, if necessary, with the appropriate
772 device number based on that name. If the device name is not in one of these
773 formats, then a unused device number will be allocated. The device
774 number will be considered unused if there is no active array for that
775 number, and there is no entry in /dev for that number and with a
776 non-standard name. Names that are not in 'standard' format are only
777 allowed in "/dev/md/".
781 .\".BR \-\-symlink = no
786 .\"to create devices in
788 .\"it will also create symlinks from
790 .\"with names starting with
796 .\"to suppress this, or
797 .\".B \-\-symlink=yes
798 .\"to enforce this even if it is suppressing
806 .BR \-u ", " \-\-uuid=
807 uuid of array to assemble. Devices which don't have this uuid are
811 .BR \-m ", " \-\-super\-minor=
812 Minor number of device that array was created for. Devices which
813 don't have this minor number are excluded. If you create an array as
814 /dev/md1, then all superblocks will contain the minor number 1, even if
815 the array is later assembled as /dev/md2.
817 Giving the literal word "dev" for
821 to use the minor number of the md device that is being assembled.
824 .B \-\-super\-minor=dev
825 will look for super blocks with a minor number of 0.
828 is only relevant for v0.90 metadata, and should not normally be used.
834 .BR \-N ", " \-\-name=
835 Specify the name of the array to assemble. This must be the name
836 that was specified when creating the array. It must either match
837 the name stored in the superblock exactly, or it must match
840 prefixed to the start of the given name.
843 .BR \-f ", " \-\-force
844 Assemble the array even if the metadata on some devices appears to be
847 cannot find enough working devices to start the array, but can find
848 some devices that are recorded as having failed, then it will mark
849 those devices as working so that the array can be started.
850 An array which requires
852 to be started may contain data corruption. Use it carefully.
856 Attempt to start the array even if fewer drives were given than were
857 present last time the array was active. Normally if not all the
858 expected drives are found and
860 is not used, then the array will be assembled but not started.
863 an attempt will be made to start it anyway.
867 This is the reverse of
869 in that it inhibits the startup of array unless all expected drives
870 are present. This is only needed with
872 and can be used if the physical connections to devices are
873 not as reliable as you would like.
876 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
877 See this option under Create and Build options.
880 .BR \-b ", " \-\-bitmap=
881 Specify the bitmap file that was given when the array was created. If
884 bitmap, there is no need to specify this when assembling the array.
887 .BR \-\-backup\-file=
890 was used when requesting a grow, shrink, RAID level change or other
891 reshape, and the system crashed during the critical section, then the
896 to allow possibly corrupted data to be restored, and the reshape
900 .BR \-U ", " \-\-update=
901 Update the superblock on each device while assembling the array. The
902 argument given to this flag can be one of
916 option will adjust the superblock of an array what was created on a Sparc
917 machine running a patched 2.2 Linux kernel. This kernel got the
918 alignment of part of the superblock wrong. You can use the
919 .B "\-\-examine \-\-sparc2.2"
922 to see what effect this would have.
926 option will update the
928 field on each superblock to match the minor number of the array being
930 This can be useful if
932 reports a different "Preferred Minor" to
934 In some cases this update will be performed automatically
935 by the kernel driver. In particular the update happens automatically
936 at the first write to an array with redundancy (RAID level 1 or
937 greater) on a 2.6 (or later) kernel.
941 option will change the uuid of the array. If a UUID is given with the
943 option that UUID will be used as a new UUID and will
945 be used to help identify the devices in the array.
948 is given, a random UUID is chosen.
952 option will change the
954 of the array as stored in the superblock. This is only supported for
955 version-1 superblocks.
959 option will change the
961 as recorded in the superblock. For version-0 superblocks, this is the
962 same as updating the UUID.
963 For version-1 superblocks, this involves updating the name.
967 option will cause the array to be marked
969 meaning that any redundancy in the array (e.g. parity for RAID5,
970 copies for RAID1) may be incorrect. This will cause the RAID system
971 to perform a "resync" pass to make sure that all redundant information
976 option allows arrays to be moved between machines with different
978 When assembling such an array for the first time after a move, giving
979 .B "\-\-update=byteorder"
982 to expect superblocks to have their byteorder reversed, and will
983 correct that order before assembling the array. This is only valid
984 with original (Version 0.90) superblocks.
988 option will correct the summaries in the superblock. That is the
989 counts of total, working, active, failed, and spare devices.
993 will rarely be of use. It applies to version 1.1 and 1.2 metadata
994 only (where the metadata is at the start of the device) and is only
995 useful when the component device has changed size (typically become
996 larger). The version 1 metadata records the amount of the device that
997 can be used to store data, so if a device in a version 1.1 or 1.2
998 array becomes larger, the metadata will still be visible, but the
999 extra space will not. In this case it might be useful to assemble the
1001 .BR \-\-update=devicesize .
1004 to determine the maximum usable amount of space on each device and
1005 update the relevant field in the metadata.
1009 .B \-\-auto\-update\-homehost
1010 This flag is only meaningful with auto-assembly (see discussion below).
1011 In that situation, if no suitable arrays are found for this homehost,
1013 will rescan for any arrays at all and will assemble them and update the
1014 homehost to match the current host.
1017 .SH For Manage mode:
1020 .BR \-t ", " \-\-test
1021 Unless a more serious error occurred,
1023 will exit with a status of 2 if no changes were made to the array and
1024 0 if at least one change was made.
1025 This can be useful when an indirect specifier such as
1030 is used in requesting an operation on the array.
1032 will report failure if these specifiers didn't find any match.
1035 .BR \-a ", " \-\-add
1036 hot-add listed devices.
1037 If a device appears to have recently been part of the array
1038 (possibly it failed or was removed) the device is re-added as describe
1040 If that fails or the device was never part of the array, the device is
1041 added as a hot-spare.
1042 If the array is degraded, it will immediately start to rebuild data
1045 Note that this and the following options are only meaningful on array
1046 with redundancy. They don't apply to RAID0 or Linear.
1050 re\-add a device that was previous removed from an array.
1051 If the metadata on the device reports that it is a member of the
1052 array, and the slot that it used is still vacant, then the device will
1053 be added back to the array in the same position. This will normally
1054 cause the data for that device to be recovered. However based on the
1055 event count on the device, the recovery may only require sections that
1056 are flagged a write-intent bitmap to be recovered or may not require
1057 any recovery at all.
1059 When used on an array that has no metadata (i.e. it was built with
1061 it will be assumed that bitmap-based recovery is enough to make the
1062 device fully consistent with the array.
1064 If the device name given is
1066 then mdadm will try to find any device that looks like it should be
1067 part of the array but isn't and will try to re\-add all such devices.
1070 .BR \-r ", " \-\-remove
1071 remove listed devices. They must not be active. i.e. they should
1072 be failed or spare devices. As well as the name of a device file
1081 The first causes all failed device to be removed. The second causes
1082 any device which is no longer connected to the system (i.e an 'open'
1085 to be removed. This will only succeed for devices that are spares or
1086 have already been marked as failed.
1089 .BR \-f ", " \-\-fail
1090 mark listed devices as faulty.
1091 As well as the name of a device file, the word
1093 can be given. This will cause any device that has been detached from
1094 the system to be marked as failed. It can then be removed.
1102 .BR \-\-write\-mostly
1103 Subsequent devices that are added or re\-added will have the 'write-mostly'
1104 flag set. This is only valid for RAID1 and means that the 'md' driver
1105 will avoid reading from these devices if possible.
1108 Subsequent devices that are added or re\-added will have the 'write-mostly'
1112 Each of these options requires that the first device listed is the array
1113 to be acted upon, and the remainder are component devices to be added,
1114 removed, marked as faulty, etc. Several different operations can be
1115 specified for different devices, e.g.
1117 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1119 Each operation applies to all devices listed until the next
1122 If an array is using a write-intent bitmap, then devices which have
1123 been removed can be re\-added in a way that avoids a full
1124 reconstruction but instead just updates the blocks that have changed
1125 since the device was removed. For arrays with persistent metadata
1126 (superblocks) this is done automatically. For arrays created with
1128 mdadm needs to be told that this device we removed recently with
1131 Devices can only be removed from an array if they are not in active
1132 use, i.e. that must be spares or failed devices. To remove an active
1133 device, it must first be marked as
1139 .BR \-Q ", " \-\-query
1140 Examine a device to see
1141 (1) if it is an md device and (2) if it is a component of an md
1143 Information about what is discovered is presented.
1146 .BR \-D ", " \-\-detail
1147 Print details of one or more md devices.
1150 .BR \-\-detail\-platform
1151 Print details of the platform's RAID capabilities (firmware / hardware
1152 topology) for a given metadata format.
1155 .BR \-Y ", " \-\-export
1160 output will be formatted as
1162 pairs for easy import into the environment.
1165 .BR \-E ", " \-\-examine
1166 Print contents of the metadata stored on the named device(s).
1167 Note the contrast between
1172 applies to devices which are components of an array, while
1174 applies to a whole array which is currently active.
1177 If an array was created on a SPARC machine with a 2.2 Linux kernel
1178 patched with RAID support, the superblock will have been created
1179 incorrectly, or at least incompatibly with 2.4 and later kernels.
1184 will fix the superblock before displaying it. If this appears to do
1185 the right thing, then the array can be successfully assembled using
1186 .BR "\-\-assemble \-\-update=sparc2.2" .
1189 .BR \-X ", " \-\-examine\-bitmap
1190 Report information about a bitmap file.
1191 The argument is either an external bitmap file or an array component
1192 in case of an internal bitmap. Note that running this on an array
1195 does not report the bitmap for that array.
1198 .BR \-R ", " \-\-run
1199 start a partially assembled array. If
1201 did not find enough devices to fully start the array, it might leaving
1202 it partially assembled. If you wish, you can then use
1204 to start the array in degraded mode.
1207 .BR \-S ", " \-\-stop
1208 deactivate array, releasing all resources.
1211 .BR \-o ", " \-\-readonly
1212 mark array as readonly.
1215 .BR \-w ", " \-\-readwrite
1216 mark array as readwrite.
1219 .B \-\-zero\-superblock
1220 If the device contains a valid md superblock, the block is
1221 overwritten with zeros. With
1223 the block where the superblock would be is overwritten even if it
1224 doesn't appear to be valid.
1227 .B \-\-kill\-subarray=
1228 If the device is a container and the argument to \-\-kill\-subarray
1229 specifies an inactive subarray in the container, then the subarray is
1230 deleted. Deleting all subarrays will leave an 'empty-container' or
1231 spare superblock on the drives. See \-\-zero\-superblock for completely
1232 removing a superblock. Note that some formats depend on the subarray
1233 index for generating a UUID, this command will fail if it would change
1234 the UUID of an active subarray.
1237 .B \-\-update\-subarray=
1238 If the device is a container and the argument to \-\-update\-subarray
1239 specifies a subarray in the container, then attempt to update the given
1240 superblock field in the subarray. See below in
1245 .BR \-t ", " \-\-test
1250 is set to reflect the status of the device. See below in
1255 .BR \-W ", " \-\-wait
1256 For each md device given, wait for any resync, recovery, or reshape
1257 activity to finish before returning.
1259 will return with success if it actually waited for every device
1260 listed, otherwise it will return failure.
1264 For each md device given, or each device in /proc/mdstat if
1266 is given, arrange for the array to be marked clean as soon as possible.
1268 will return with success if the array uses external metadata and we
1269 successfully waited. For native arrays this returns immediately as the
1270 kernel handles dirty-clean transitions at shutdown. No action is taken
1271 if safe-mode handling is disabled.
1273 .SH For Incremental Assembly mode:
1275 .BR \-\-rebuild\-map ", " \-r
1276 Rebuild the map file
1277 .RB ( /var/run/mdadm/map )
1280 uses to help track which arrays are currently being assembled.
1283 .BR \-\-run ", " \-R
1284 Run any array assembled as soon as a minimal number of devices are
1285 available, rather than waiting until all expected devices are present.
1288 .BR \-\-scan ", " \-s
1289 Only meaningful with
1293 file for arrays that are being incrementally assembled and will try to
1294 start any that are not already started. If any such array is listed
1297 as requiring an external bitmap, that bitmap will be attached first.
1300 .BR \-\-fail ", " \-f
1301 This allows the hot-plug system to remove devices that have fully disappeared
1302 from the kernel. It will first fail and then remove the device from any
1303 array it belongs to.
1304 The device name given should be a kernel device name such as "sda",
1308 .SH For Monitor mode:
1310 .BR \-m ", " \-\-mail
1311 Give a mail address to send alerts to.
1314 .BR \-p ", " \-\-program ", " \-\-alert
1315 Give a program to be run whenever an event is detected.
1318 .BR \-y ", " \-\-syslog
1319 Cause all events to be reported through 'syslog'. The messages have
1320 facility of 'daemon' and varying priorities.
1323 .BR \-d ", " \-\-delay
1324 Give a delay in seconds.
1326 polls the md arrays and then waits this many seconds before polling
1327 again. The default is 60 seconds. Since 2.6.16, there is no need to
1328 reduce this as the kernel alerts
1330 immediately when there is any change.
1333 .BR \-r ", " \-\-increment
1334 Give a percentage increment.
1336 will generate RebuildNN events with the given percentage increment.
1339 .BR \-f ", " \-\-daemonise
1342 to run as a background daemon if it decides to monitor anything. This
1343 causes it to fork and run in the child, and to disconnect from the
1344 terminal. The process id of the child is written to stdout.
1347 which will only continue monitoring if a mail address or alert program
1348 is found in the config file.
1351 .BR \-i ", " \-\-pid\-file
1354 is running in daemon mode, write the pid of the daemon process to
1355 the specified file, instead of printing it on standard output.
1358 .BR \-1 ", " \-\-oneshot
1359 Check arrays only once. This will generate
1361 events and more significantly
1367 .B " mdadm \-\-monitor \-\-scan \-1"
1369 from a cron script will ensure regular notification of any degraded arrays.
1372 .BR \-t ", " \-\-test
1375 alert for every array found at startup. This alert gets mailed and
1376 passed to the alert program. This can be used for testing that alert
1377 message do get through successfully.
1383 .B mdadm \-\-assemble
1384 .I md-device options-and-component-devices...
1387 .B mdadm \-\-assemble \-\-scan
1388 .I md-devices-and-options...
1391 .B mdadm \-\-assemble \-\-scan
1395 This usage assembles one or more RAID arrays from pre-existing components.
1396 For each array, mdadm needs to know the md device, the identity of the
1397 array, and a number of component-devices. These can be found in a number of ways.
1399 In the first usage example (without the
1401 the first device given is the md device.
1402 In the second usage example, all devices listed are treated as md
1403 devices and assembly is attempted.
1404 In the third (where no devices are listed) all md devices that are
1405 listed in the configuration file are assembled. If not arrays are
1406 described by the configuration file, then any arrays that
1407 can be found on unused devices will be assembled.
1409 If precisely one device is listed, but
1415 was given and identity information is extracted from the configuration file.
1417 The identity can be given with the
1423 option, will be taken from the md-device record in the config file, or
1424 will be taken from the super block of the first component-device
1425 listed on the command line.
1427 Devices can be given on the
1429 command line or in the config file. Only devices which have an md
1430 superblock which contains the right identity will be considered for
1433 The config file is only used if explicitly named with
1435 or requested with (a possibly implicit)
1440 .B /etc/mdadm/mdadm.conf
1445 is not given, then the config file will only be used to find the
1446 identity of md arrays.
1448 Normally the array will be started after it is assembled. However if
1450 is not given and not all expected drives were listed, then the array
1451 is not started (to guard against usage errors). To insist that the
1452 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1461 does not create any entries in
1465 It does record information in
1466 .B /var/run/mdadm/map
1469 to choose the correct name.
1473 detects that udev is not configured, it will create the devices in
1477 In Linux kernels prior to version 2.6.28 there were two distinctly
1478 different types of md devices that could be created: one that could be
1479 partitioned using standard partitioning tools and one that could not.
1480 Since 2.6.28 that distinction is no longer relevant as both type of
1481 devices can be partitioned.
1483 will normally create the type that originally could not be partitioned
1484 as it has a well defined major number (9).
1486 Prior to 2.6.28, it is important that mdadm chooses the correct type
1487 of array device to use. This can be controlled with the
1489 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1490 to use a partitionable device rather than the default.
1492 In the no-udev case, the value given to
1494 can be suffixed by a number. This tells
1496 to create that number of partition devices rather than the default of 4.
1500 can also be given in the configuration file as a word starting
1502 on the ARRAY line for the relevant array.
1509 and no devices are listed,
1511 will first attempt to assemble all the arrays listed in the config
1514 In no array at listed in the config (other than those marked
1516 it will look through the available devices for possible arrays and
1517 will try to assemble anything that it finds. Arrays which are tagged
1518 as belonging to the given homehost will be assembled and started
1519 normally. Arrays which do not obviously belong to this host are given
1520 names that are expected not to conflict with anything local, and are
1521 started "read-auto" so that nothing is written to any device until the
1522 array is written to. i.e. automatic resync etc is delayed.
1526 finds a consistent set of devices that look like they should comprise
1527 an array, and if the superblock is tagged as belonging to the given
1528 home host, it will automatically choose a device name and try to
1529 assemble the array. If the array uses version-0.90 metadata, then the
1531 number as recorded in the superblock is used to create a name in
1535 If the array uses version-1 metadata, then the
1537 from the superblock is used to similarly create a name in
1539 (the name will have any 'host' prefix stripped first).
1541 This behaviour can be modified by the
1545 configuration file. This line can indicate that specific metadata
1546 type should, or should not, be automatically assembled. If an array
1547 is found which is not listed in
1549 and has a metadata format that is denied by the
1551 line, then it will not be assembled.
1554 line can also request that all arrays identified as being for this
1555 homehost should be assembled regardless of their metadata type.
1558 for further details.
1563 cannot find any array for the given host at all, and if
1564 .B \-\-auto\-update\-homehost
1567 will search again for any array (not just an array created for this
1568 host) and will assemble each assuming
1569 .BR \-\-update=homehost .
1570 This will change the host tag in the superblock so that on the next run,
1571 these arrays will be found without the second pass. The intention of
1572 this feature is to support transitioning a set of md arrays to using
1575 The reason for requiring arrays to be tagged with the homehost for
1576 auto assembly is to guard against problems that can arise when moving
1577 devices from one host to another.
1588 .BI \-\-raid\-devices= Z
1592 This usage is similar to
1594 The difference is that it creates an array without a superblock. With
1595 these arrays there is no difference between initially creating the array and
1596 subsequently assembling the array, except that hopefully there is useful
1597 data there in the second case.
1599 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1600 one of their synonyms. All devices must be listed and the array will
1601 be started once complete. It will often be appropriate to use
1602 .B \-\-assume\-clean
1603 with levels raid1 or raid10.
1614 .BI \-\-raid\-devices= Z
1618 This usage will initialise a new md array, associate some devices with
1619 it, and activate the array.
1621 The named device will normally not exist when
1622 .I "mdadm \-\-create"
1623 is run, but will be created by
1625 once the array becomes active.
1627 As devices are added, they are checked to see if they contain RAID
1628 superblocks or filesystems. They are also checked to see if the variance in
1629 device size exceeds 1%.
1631 If any discrepancy is found, the array will not automatically be run, though
1634 can override this caution.
1636 To create a "degraded" array in which some devices are missing, simply
1637 give the word "\fBmissing\fP"
1638 in place of a device name. This will cause
1640 to leave the corresponding slot in the array empty.
1641 For a RAID4 or RAID5 array at most one slot can be
1642 "\fBmissing\fP"; for a RAID6 array at most two slots.
1643 For a RAID1 array, only one real device needs to be given. All of the
1647 When creating a RAID5 array,
1649 will automatically create a degraded array with an extra spare drive.
1650 This is because building the spare into a degraded array is in general
1651 faster than resyncing the parity on a non-degraded, but not clean,
1652 array. This feature can be overridden with the
1656 When creating an array with version-1 metadata a name for the array is
1658 If this is not given with the
1662 will choose a name based on the last component of the name of the
1663 device being created. So if
1665 is being created, then the name
1670 is being created, then the name
1674 When creating a partition based array, using
1676 with version-1.x metadata, the partition type should be set to
1678 (non fs-data). This type selection allows for greater precision since
1679 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
1680 might create problems in the event of array recovery through a live cdrom.
1682 A new array will normally get a randomly assigned 128bit UUID which is
1683 very likely to be unique. If you have a specific need, you can choose
1684 a UUID for the array by giving the
1686 option. Be warned that creating two arrays with the same UUID is a
1687 recipe for disaster. Also, using
1689 when creating a v0.90 array will silently override any
1694 .\"option is given, it is not necessary to list any component-devices in this command.
1695 .\"They can be added later, before a
1699 .\"is given, the apparent size of the smallest drive given is used.
1701 When creating an array within a
1704 can be given either the list of devices to use, or simply the name of
1705 the container. The former case gives control over which devices in
1706 the container will be used for the array. The latter case allows
1708 to automatically choose which devices to use based on how much spare
1711 The General Management options that are valid with
1716 insist on running the array even if some devices look like they might
1721 start the array readonly \(em not supported yet.
1728 .I options... devices...
1731 This usage will allow individual devices in an array to be failed,
1732 removed or added. It is possible to perform multiple operations with
1733 on command. For example:
1735 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
1741 and will then remove it from the array and finally add it back
1742 in as a spare. However only one md array can be affected by a single
1745 When a device is added to an active array, mdadm checks to see if it
1746 has metadata on it which suggests that it was recently a member of the
1747 array. If it does, it tries to "re\-add" the device. If there have
1748 been no changes since the device was removed, or if the array has a
1749 write-intent bitmap which has recorded whatever changes there were,
1750 then the device will immediately become a full member of the array and
1751 those differences recorded in the bitmap will be resolved.
1761 MISC mode includes a number of distinct operations that
1762 operate on distinct devices. The operations are:
1765 The device is examined to see if it is
1766 (1) an active md array, or
1767 (2) a component of an md array.
1768 The information discovered is reported.
1772 The device should be an active md device.
1774 will display a detailed description of the array.
1778 will cause the output to be less detailed and the format to be
1779 suitable for inclusion in
1780 .BR /etc/mdadm.conf .
1783 will normally be 0 unless
1785 failed to get useful information about the device(s); however, if the
1787 option is given, then the exit status will be:
1791 The array is functioning normally.
1794 The array has at least one failed device.
1797 The array has multiple failed devices such that it is unusable.
1800 There was an error while trying to get information about the device.
1804 .B \-\-detail\-platform
1805 Print detail of the platform's RAID capabilities (firmware / hardware
1806 topology). If the metadata is specified with
1810 then the return status will be:
1814 metadata successfully enumerated its platform components on this system
1817 metadata is platform independent
1820 metadata failed to find its platform components on this system
1824 .B \-\-update\-subarray=
1825 If the device is a container and the argument to \-\-update\-subarray
1826 specifies a subarray in the container, then attempt to update the given
1827 superblock field in the subarray. Similar to updating an array in
1828 "assemble" mode, the field to update is selected by
1832 option. Currently only
1838 option updates the subarray name in the metadata, it may not affect the
1839 device node name or the device node symlink until the subarray is
1840 re\-assembled. If updating
1842 would change the UUID of an active subarray this operation is blocked,
1843 and the command will end in an error.
1847 The device should be a component of an md array.
1849 will read the md superblock of the device and display the contents.
1854 is given, then multiple devices that are components of the one array
1855 are grouped together and reported in a single entry suitable
1857 .BR /etc/mdadm.conf .
1861 without listing any devices will cause all devices listed in the
1862 config file to be examined.
1866 The devices should be active md arrays which will be deactivated, as
1867 long as they are not currently in use.
1871 This will fully activate a partially assembled md array.
1875 This will mark an active array as read-only, providing that it is
1876 not currently being used.
1882 array back to being read/write.
1886 For all operations except
1889 will cause the operation to be applied to all arrays listed in
1894 causes all devices listed in the config file to be examined.
1897 .BR \-b ", " \-\-brief
1898 Be less verbose. This is used with
1906 gives an intermediate level of verbosity.
1912 .B mdadm \-\-monitor
1913 .I options... devices...
1918 to periodically poll a number of md arrays and to report on any events
1921 will never exit once it decides that there are arrays to be checked,
1922 so it should normally be run in the background.
1924 As well as reporting events,
1926 may move a spare drive from one array to another if they are in the
1929 and if the destination array has a failed drive but no spares.
1931 If any devices are listed on the command line,
1933 will only monitor those devices. Otherwise all arrays listed in the
1934 configuration file will be monitored. Further, if
1936 is given, then any other md devices that appear in
1938 will also be monitored.
1940 The result of monitoring the arrays is the generation of events.
1941 These events are passed to a separate program (if specified) and may
1942 be mailed to a given E-mail address.
1944 When passing events to a program, the program is run once for each event,
1945 and is given 2 or 3 command-line arguments: the first is the
1946 name of the event (see below), the second is the name of the
1947 md device which is affected, and the third is the name of a related
1948 device if relevant (such as a component device that has failed).
1952 is given, then a program or an E-mail address must be specified on the
1953 command line or in the config file. If neither are available, then
1955 will not monitor anything.
1959 will continue monitoring as long as something was found to monitor. If
1960 no program or email is given, then each event is reported to
1963 The different events are:
1967 .B DeviceDisappeared
1968 An md array which previously was configured appears to no longer be
1969 configured. (syslog priority: Critical)
1973 was told to monitor an array which is RAID0 or Linear, then it will
1975 .B DeviceDisappeared
1976 with the extra information
1978 This is because RAID0 and Linear do not support the device-failed,
1979 hot-spare and resync operations which are monitored.
1983 An md array started reconstruction. (syslog priority: Warning)
1989 is a two-digit number (ie. 05, 48). This indicates that rebuild
1990 has passed that many percent of the total. The events are generated
1991 with fixed increment since 0. Increment size may be specified with
1992 a commandline option (default is 20). (syslog priority: Warning)
1996 An md array that was rebuilding, isn't any more, either because it
1997 finished normally or was aborted. (syslog priority: Warning)
2001 An active component device of an array has been marked as
2002 faulty. (syslog priority: Critical)
2006 A spare component device which was being rebuilt to replace a faulty
2007 device has failed. (syslog priority: Critical)
2011 A spare component device which was being rebuilt to replace a faulty
2012 device has been successfully rebuilt and has been made active.
2013 (syslog priority: Info)
2017 A new md array has been detected in the
2019 file. (syslog priority: Info)
2023 A newly noticed array appears to be degraded. This message is not
2026 notices a drive failure which causes degradation, but only when
2028 notices that an array is degraded when it first sees the array.
2029 (syslog priority: Critical)
2033 A spare drive has been moved from one array in a
2035 to another to allow a failed drive to be replaced.
2036 (syslog priority: Info)
2042 has been told, via the config file, that an array should have a certain
2043 number of spare devices, and
2045 detects that it has fewer than this number when it first sees the
2046 array, it will report a
2049 (syslog priority: Warning)
2053 An array was found at startup, and the
2056 (syslog priority: Info)
2066 cause Email to be sent. All events cause the program to be run.
2067 The program is run with two or three arguments: the event
2068 name, the array device and possibly a second device.
2070 Each event has an associated array device (e.g.
2072 and possibly a second device. For
2077 the second device is the relevant component device.
2080 the second device is the array that the spare was moved from.
2084 to move spares from one array to another, the different arrays need to
2085 be labeled with the same
2087 in the configuration file. The
2089 name can be any string; it is only necessary that different spare
2090 groups use different names.
2094 detects that an array in a spare group has fewer active
2095 devices than necessary for the complete array, and has no spare
2096 devices, it will look for another array in the same spare group that
2097 has a full complement of working drive and a spare. It will then
2098 attempt to remove the spare from the second drive and add it to the
2100 If the removal succeeds but the adding fails, then it is added back to
2104 The GROW mode is used for changing the size or shape of an active
2106 For this to work, the kernel must support the necessary change.
2107 Various types of growth are being added during 2.6 development,
2108 including restructuring a RAID5 array to have more active devices.
2110 Currently the only support available is to
2112 change the "size" attribute
2113 for RAID1, RAID5 and RAID6.
2115 increase or decrease the "raid\-devices" attribute of RAID1, RAID5,
2118 change the chunk-size and layout of RAID5 and RAID6.
2120 convert between RAID1 and RAID5, and between RAID5 and RAID6.
2122 add a write-intent bitmap to any array which supports these bitmaps, or
2123 remove a write-intent bitmap from such an array.
2126 GROW mode is not currently supported for
2128 or arrays inside containers.
2131 Normally when an array is built the "size" it taken from the smallest
2132 of the drives. If all the small drives in an arrays are, one at a
2133 time, removed and replaced with larger drives, then you could have an
2134 array of large drives with only a small amount used. In this
2135 situation, changing the "size" with "GROW" mode will allow the extra
2136 space to start being used. If the size is increased in this way, a
2137 "resync" process will start to make sure the new parts of the array
2140 Note that when an array changes size, any filesystem that may be
2141 stored in the array will not automatically grow to use the space. The
2142 filesystem will need to be explicitly told to use the extra space.
2144 Also the size of an array cannot be changed while it has an active
2145 bitmap. If an array has a bitmap, it must be removed before the size
2146 can be changed. Once the change it complete a new bitmap can be created.
2148 .SS RAID\-DEVICES CHANGES
2150 A RAID1 array can work with any number of devices from 1 upwards
2151 (though 1 is not very useful). There may be times which you want to
2152 increase or decrease the number of active devices. Note that this is
2153 different to hot-add or hot-remove which changes the number of
2156 When reducing the number of devices in a RAID1 array, the slots which
2157 are to be removed from the array must already be vacant. That is, the
2158 devices which were in those slots must be failed and removed.
2160 When the number of devices is increased, any hot spares that are
2161 present will be activated immediately.
2163 Changing the number of active devices in a RAID5 or RAID6 is much more
2164 effort. Every block in the array will need to be read and written
2165 back to a new location. From 2.6.17, the Linux Kernel is able to
2166 increase the number of devices in a RAID5 safely, including restarting
2167 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2168 increase or decrease the number of devices in a RAID5 or RAID6.
2170 When decreasing the number of devices, the size of the array will also
2171 decrease. If there was data in the array, it could get destroyed and
2172 this is not reversible. To help prevent accidents,
2174 requires that the size of the array be decreased first with
2175 .BR "mdadm --grow --array-size" .
2176 This is a reversible change which simply makes the end of the array
2177 inaccessible. The integrity of any data can then be checked before
2178 the non-reversible reduction in the number of devices is request.
2180 When relocating the first few stripes on a RAID5 or RAID6, it is not
2181 possible to keep the data on disk completely consistent and
2182 crash-proof. To provide the required safety, mdadm disables writes to
2183 the array while this "critical section" is reshaped, and takes a
2184 backup of the data that is in that section. For grows, this backup may be
2185 stored in any spare devices that the array has, however it can also be
2186 stored in a separate file specified with the
2188 option, and is required to be specified for shrinks, RAID level
2189 changes and layout changes. If this option is used, and the system
2190 does crash during the critical period, the same file must be passed to
2192 to restore the backup and reassemble the array. When shrinking rather
2193 than growing the array, the reshape is done from the end towards the
2194 beginning, so the "critical section" is at the end of the reshape.
2198 Changing the RAID level of any array happens instantaneously. However
2199 in the RAID5 to RAID6 case this requires a non-standard layout of the
2200 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2201 required before the change can be accomplished. So while the level
2202 change is instant, the accompanying layout change can take quite a
2205 is required. If the array is not simultaneously being grown or
2206 shrunk, so that the array size will remain the same - for example,
2207 reshaping a 3-drive RAID5 into a 4-drive RAID6 - the backup file will
2208 be used not just for a "cricital section" but throughout the reshape
2209 operation, as described below under LAYOUT CHANGES.
2211 .SS CHUNK-SIZE AND LAYOUT CHANGES
2213 Changing the chunk-size of layout without also changing the number of
2214 devices as the same time will involve re-writing all blocks in-place.
2215 To ensure against data loss in the case of a crash, a
2217 must be provided for these changes. Small sections of the array will
2218 be copied to the backup file while they are being rearranged. This
2219 means that all the data is copied twice, once to the backup and once
2220 to the new layout on the array, so this type of reshape will go very
2223 If the reshape is interrupted for any reason, this backup file must be
2225 .B "mdadm --assemble"
2226 so the array can be reassembled. Consequently the file cannot be
2227 stored on the device being reshaped.
2232 A write-intent bitmap can be added to, or removed from, an active
2233 array. Either internal bitmaps, or bitmaps stored in a separate file,
2234 can be added. Note that if you add a bitmap stored in a file which is
2235 in a filesystem that is on the RAID array being affected, the system
2236 will deadlock. The bitmap must be on a separate filesystem.
2238 .SH INCREMENTAL MODE
2242 .B mdadm \-\-incremental
2248 .B mdadm \-\-incremental \-\-fail
2252 .B mdadm \-\-incremental \-\-rebuild\-map
2255 .B mdadm \-\-incremental \-\-run \-\-scan
2258 This mode is designed to be used in conjunction with a device
2259 discovery system. As devices are found in a system, they can be
2261 .B "mdadm \-\-incremental"
2262 to be conditionally added to an appropriate array.
2264 Conversely, it can also be used with the
2266 flag to do just the opposite and find whatever array a particular device
2267 is part of and remove the device from that array.
2269 If the device passed is a
2271 device created by a previous call to
2273 then rather than trying to add that device to an array, all the arrays
2274 described by the metadata of the container will be started.
2277 performs a number of tests to determine if the device is part of an
2278 array, and which array it should be part of. If an appropriate array
2279 is found, or can be created,
2281 adds the device to the array and conditionally starts the array.
2285 will only add devices to an array which were previously working
2286 (active or spare) parts of that array. It does not currently support
2287 automatic inclusion of a new drive as a spare in some array.
2291 makes are as follow:
2293 Is the device permitted by
2295 That is, is it listed in a
2297 line in that file. If
2299 is absent then the default it to allow any device. Similar if
2301 contains the special word
2303 then any device is allowed. Otherwise the device name given to
2305 must match one of the names or patterns in a
2310 Does the device have a valid md superblock. If a specific metadata
2311 version is request with
2315 then only that style of metadata is accepted, otherwise
2317 finds any known version of metadata. If no
2319 metadata is found, the device is rejected.
2323 Does the metadata match an expected array?
2324 The metadata can match in two ways. Either there is an array listed
2327 which identifies the array (either by UUID, by name, by device list,
2328 or by minor-number), or the array was created with a
2334 or on the command line.
2337 is not able to positively identify the array as belonging to the
2338 current host, the device will be rejected.
2342 keeps a list of arrays that it has partially assembled in
2343 .B /var/run/mdadm/map
2345 .B /var/run/mdadm.map
2346 if the directory doesn't exist. Or maybe even
2347 .BR /dev/.mdadm.map ).
2348 If no array exists which matches
2349 the metadata on the new device,
2351 must choose a device name and unit number. It does this based on any
2354 or any name information stored in the metadata. If this name
2355 suggests a unit number, that number will be used, otherwise a free
2356 unit number will be chosen. Normally
2358 will prefer to create a partitionable array, however if the
2362 suggests that a non-partitionable array is preferred, that will be
2365 If the array is not found in the config file and its metadata does not
2366 identify it as belonging to the "homehost", then
2368 will choose a name for the array which is certain not to conflict with
2369 any array which does belong to this host. It does this be adding an
2370 underscore and a small number to the name preferred by the metadata.
2372 Once an appropriate array is found or created and the device is added,
2374 must decide if the array is ready to be started. It will
2375 normally compare the number of available (non-spare) devices to the
2376 number of devices that the metadata suggests need to be active. If
2377 there are at least that many, the array will be started. This means
2378 that if any devices are missing the array will not be restarted.
2384 in which case the array will be run as soon as there are enough
2385 devices present for the data to be accessible. For a RAID1, that
2386 means one device will start the array. For a clean RAID5, the array
2387 will be started as soon as all but one drive is present.
2389 Note that neither of these approaches is really ideal. If it can
2390 be known that all device discovery has completed, then
2394 can be run which will try to start all arrays that are being
2395 incrementally assembled. They are started in "read-auto" mode in
2396 which they are read-only until the first write request. This means
2397 that no metadata updates are made and no attempt at resync or recovery
2398 happens. Further devices that are found before the first write can
2399 still be added safely.
2402 This section describes environment variables that affect how mdadm
2407 Setting this value to 1 will prevent mdadm from automatically launching
2408 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2414 does not create any device nodes in /dev, but leaves that task to
2418 appears not to be configured, or if this environment variable is set
2421 will create and devices that are needed.
2425 .B " mdadm \-\-query /dev/name-of-device"
2427 This will find out if a given device is a RAID array, or is part of
2428 one, and will provide brief information about the device.
2430 .B " mdadm \-\-assemble \-\-scan"
2432 This will assemble and start all arrays listed in the standard config
2433 file. This command will typically go in a system startup file.
2435 .B " mdadm \-\-stop \-\-scan"
2437 This will shut down all arrays that can be shut down (i.e. are not
2438 currently in use). This will typically go in a system shutdown script.
2440 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2442 If (and only if) there is an Email address or program given in the
2443 standard config file, then
2444 monitor the status of all arrays listed in that file by
2445 polling them ever 2 minutes.
2447 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2449 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2452 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2454 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2456 This will create a prototype config file that describes currently
2457 active arrays that are known to be made from partitions of IDE or SCSI drives.
2458 This file should be reviewed before being used as it may
2459 contain unwanted detail.
2461 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2463 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
2465 This will find arrays which could be assembled from existing IDE and
2466 SCSI whole drives (not partitions), and store the information in the
2467 format of a config file.
2468 This file is very likely to contain unwanted detail, particularly
2471 entries. It should be reviewed and edited before being used as an
2474 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
2476 .B " mdadm \-Ebsc partitions"
2478 Create a list of devices by reading
2479 .BR /proc/partitions ,
2480 scan these for RAID superblocks, and printout a brief listing of all
2483 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
2485 Scan all partitions and devices listed in
2486 .BR /proc/partitions
2489 out of all such devices with a RAID superblock with a minor number of 0.
2491 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /var/run/mdadm"
2493 If config file contains a mail address or alert program, run mdadm in
2494 the background in monitor mode monitoring all md devices. Also write
2495 pid of mdadm daemon to
2496 .BR /var/run/mdadm .
2498 .B " mdadm \-Iq /dev/somedevice"
2500 Try to incorporate newly discovered device into some array as
2503 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
2505 Rebuild the array map from any current arrays, and then start any that
2508 .B " mdadm /dev/md4 --fail detached --remove detached"
2510 Any devices which are components of /dev/md4 will be marked as faulty
2511 and then remove from the array.
2513 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4
2517 which is currently a RAID5 array will be converted to RAID6. There
2518 should normally already be a spare drive attached to the array as a
2519 RAID6 needs one more drive than a matching RAID5.
2521 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
2523 Create a DDF array over 6 devices.
2525 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
2527 Create a RAID5 array over any 3 devices in the given DDF set. Use
2528 only 30 gigabytes of each device.
2530 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
2532 Assemble a pre-exist ddf array.
2534 .B " mdadm -I /dev/md/ddf1"
2536 Assemble all arrays contained in the ddf array, assigning names as
2539 .B " mdadm \-\-create \-\-help"
2541 Provide help about the Create mode.
2543 .B " mdadm \-\-config \-\-help"
2545 Provide help about the format of the config file.
2547 .B " mdadm \-\-help"
2549 Provide general help.
2559 lists all active md devices with information about them.
2561 uses this to find arrays when
2563 is given in Misc mode, and to monitor array reconstruction
2568 The config file lists which devices may be scanned to see if
2569 they contain MD super block, and gives identifying information
2570 (e.g. UUID) about known MD arrays. See
2574 .SS /var/run/mdadm/map
2577 mode is used, this file gets a list of arrays currently being created.
2580 does not exist as a directory, then
2581 .B /var/run/mdadm.map
2584 is not available (as may be the case during early boot),
2586 is used on the basis that
2588 is usually available very early in boot.
2593 understand two sorts of names for array devices.
2595 The first is the so-called 'standard' format name, which matches the
2596 names used by the kernel and which appear in
2599 The second sort can be freely chosen, but must reside in
2601 When giving a device name to
2603 to create or assemble an array, either full path name such as
2607 can be given, or just the suffix of the second sort of name, such as
2613 chooses device names during auto-assembly or incremental assembly, it
2614 will sometimes add a small sequence number to the end of the name to
2615 avoid conflicted between multiple arrays that have the same name. If
2617 can reasonably determine that the array really is meant for this host,
2618 either by a hostname in the metadata, or by the presence of the array
2619 in /etc/mdadm.conf, then it will leave off the suffix if possible.
2620 Also if the homehost is specified as
2623 will only use a suffix if a different array of the same name already
2624 exists or is listed in the config file.
2626 The standard names for non-partitioned arrays (the only sort of md
2627 array available in 2.4 and earlier) are of the form
2631 where NN is a number.
2632 The standard names for partitionable arrays (as available from 2.6
2633 onwards) are of the form
2637 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2639 From kernel version, 2.6.28 the "non-partitioned array" can actually
2640 be partitioned. So the "md_dNN" names are no longer needed, and
2641 partitions such as "/dev/mdNNpXX" are possible.
2645 was previously known as
2649 is completely separate from the
2651 package, and does not use the
2653 configuration file at all.
2656 For further information on mdadm usage, MD and the various levels of
2659 .B http://linux\-raid.osdl.org/
2661 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
2663 .\"for new releases of the RAID driver check out:
2666 .\".UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2667 .\"ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2672 .\".UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2673 .\"http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2676 The latest version of
2678 should always be available from
2680 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/