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
126 Linear and RAID levels 0/1/4/5/6,
127 changing the RAID level between 0, 1, 5, and 6, and between 0 and 10,
128 changing the chunk size and layout for RAID 0,4,5,6,10 as well as adding or
129 removing a write-intent bitmap.
132 .B "Incremental Assembly"
133 Add a single device to an appropriate array. If the addition of the
134 device makes the array runnable, the array will be started.
135 This provides a convenient interface to a
137 system. As each device is detected,
139 has a chance to include it in some array as appropriate.
142 flag is passed in we will remove the device from any active array
143 instead of adding it.
149 in this mode, then any arrays within that container will be assembled
154 This is for doing things to specific components of an array such as
155 adding new spares and removing faulty devices.
159 This is an 'everything else' mode that supports operations on active
160 arrays, operations on component devices such as erasing old superblocks, and
161 information gathering operations.
162 .\"This mode allows operations on independent devices such as examine MD
163 .\"superblocks, erasing old superblocks and stopping active arrays.
167 This mode does not act on a specific device or array, but rather it
168 requests the Linux Kernel to activate any auto-detected arrays.
171 .SH Options for selecting a mode are:
174 .BR \-A ", " \-\-assemble
175 Assemble a pre-existing array.
178 .BR \-B ", " \-\-build
179 Build a legacy array without superblocks.
182 .BR \-C ", " \-\-create
186 .BR \-F ", " \-\-follow ", " \-\-monitor
192 .BR \-G ", " \-\-grow
193 Change the size or shape of an active array.
196 .BR \-I ", " \-\-incremental
197 Add/remove a single device to/from an appropriate array, and possibly start the array.
201 Request that the kernel starts any auto-detected arrays. This can only
204 is compiled into the kernel \(em not if it is a module.
205 Arrays can be auto-detected by the kernel if all the components are in
206 primary MS-DOS partitions with partition type
208 and all use v0.90 metadata.
209 In-kernel autodetect is not recommended for new installations. Using
211 to detect and assemble arrays \(em possibly in an
213 \(em is substantially more flexible and should be preferred.
216 If a device is given before any options, or if the first option is
225 then the MANAGE mode is assumed.
226 Anything other than these will cause the
230 .SH Options that are not mode-specific are:
233 .BR \-h ", " \-\-help
234 Display general help message or, after one of the above options, a
235 mode-specific help message.
239 Display more detailed help about command line parsing and some commonly
243 .BR \-V ", " \-\-version
244 Print version information for mdadm.
247 .BR \-v ", " \-\-verbose
248 Be more verbose about what is happening. This can be used twice to be
250 The extra verbosity currently only affects
251 .B \-\-detail \-\-scan
253 .BR "\-\-examine \-\-scan" .
256 .BR \-q ", " \-\-quiet
257 Avoid printing purely informative messages. With this,
259 will be silent unless there is something really important to report.
263 .BR \-f ", " \-\-force
264 Be more forceful about certain operations. See the various modes for
265 the exact meaning of this option in different contexts.
268 .BR \-c ", " \-\-config=
269 Specify the config file or directory. Default is to use
272 .BR /etc/mdadm.conf.d ,
273 or if those are missing then
274 .B /etc/mdadm/mdadm.conf
276 .BR /etc/mdadm/mdadm.conf.d .
277 If the config file given is
279 then nothing will be read, but
281 will act as though the config file contained exactly
283 .B " DEVICE partitions containers"
287 to find a list of devices to scan, and
289 to find a list of containers to examine.
292 is given for the config file, then
294 will act as though the config file were empty.
296 If the name given is of a directory, then
298 will collect all the files contained in the directory with a name ending
301 sort them lexically, and process all of those files as config files.
304 .BR \-s ", " \-\-scan
307 for missing information.
308 In general, this option gives
310 permission to get any missing information (like component devices,
311 array devices, array identities, and alert destination) from the
312 configuration file (see previous option);
313 one exception is MISC mode when using
319 says to get a list of array devices from
323 .BR \-e ", " \-\-metadata=
324 Declare the style of RAID metadata (superblock) to be used. The
325 default is {DEFAULT_METADATA} for
327 and to guess for other operations.
328 The default can be overridden by setting the
337 .ie '{DEFAULT_METADATA}'0.90'
338 .IP "0, 0.90, default"
341 Use the original 0.90 format superblock. This format limits arrays to
342 28 component devices and limits component devices of levels 1 and
343 greater to 2 terabytes. It is also possible for there to be confusion
344 about whether the superblock applies to a whole device or just the
345 last partition, if that partition starts on a 64K boundary.
346 .ie '{DEFAULT_METADATA}'0.90'
347 .IP "1, 1.0, 1.1, 1.2"
349 .IP "1, 1.0, 1.1, 1.2 default"
350 Use the new version-1 format superblock. This has fewer restrictions.
351 It can easily be moved between hosts with different endian-ness, and a
352 recovery operation can be checkpointed and restarted. The different
353 sub-versions store the superblock at different locations on the
354 device, either at the end (for 1.0), at the start (for 1.1) or 4K from
355 the start (for 1.2). "1" is equivalent to "1.2" (the commonly
356 preferred 1.x format).
357 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
359 Use the "Industry Standard" DDF (Disk Data Format) format defined by
361 When creating a DDF array a
363 will be created, and normal arrays can be created in that container.
365 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
367 which is managed in a similar manner to DDF, and is supported by an
368 option-rom on some platforms:
370 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
376 This will override any
378 setting in the config file and provides the identity of the host which
379 should be considered the home for any arrays.
381 When creating an array, the
383 will be recorded in the metadata. For version-1 superblocks, it will
384 be prefixed to the array name. For version-0.90 superblocks, part of
385 the SHA1 hash of the hostname will be stored in the later half of the
388 When reporting information about an array, any array which is tagged
389 for the given homehost will be reported as such.
391 When using Auto-Assemble, only arrays tagged for the given homehost
392 will be allowed to use 'local' names (i.e. not ending in '_' followed
393 by a digit string). See below under
394 .BR "Auto Assembly" .
396 The special name "\fBany\fP" can be used as a wild card. If an array
399 then the name "\fBany\fP" will be stored in the array and it can be
400 assembled in the same way on any host. If an array is assembled with
401 this option, then the homehost recorded on the array will be ignored.
407 needs to print the name for a device it normally finds the name in
409 which refers to the device and is shortest. When a path component is
413 will prefer a longer name if it contains that component. For example
414 .B \-\-prefer=by-uuid
415 will prefer a name in a subdirectory of
420 This functionality is currently only provided by
425 .SH For create, build, or grow:
428 .BR \-n ", " \-\-raid\-devices=
429 Specify the number of active devices in the array. This, plus the
430 number of spare devices (see below) must equal the number of
432 (including "\fBmissing\fP" devices)
433 that are listed on the command line for
435 Setting a value of 1 is probably
436 a mistake and so requires that
438 be specified first. A value of 1 will then be allowed for linear,
439 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
441 This number can only be changed using
443 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
444 the necessary support.
447 .BR \-x ", " \-\-spare\-devices=
448 Specify the number of spare (eXtra) devices in the initial array.
449 Spares can also be added
450 and removed later. The number of component devices listed
451 on the command line must equal the number of RAID devices plus the
452 number of spare devices.
455 .BR \-z ", " \-\-size=
456 Amount (in Kibibytes) of space to use from each drive in RAID levels 1/4/5/6.
457 This must be a multiple of the chunk size, and must leave about 128Kb
458 of space at the end of the drive for the RAID superblock.
459 If this is not specified
460 (as it normally is not) the smallest drive (or partition) sets the
461 size, though if there is a variance among the drives of greater than 1%, a warning is
464 A suffix of 'M' or 'G' can be given to indicate Megabytes or
465 Gigabytes respectively.
467 Sometimes a replacement drive can be a little smaller than the
468 original drives though this should be minimised by IDEMA standards.
469 Such a replacement drive will be rejected by
471 To guard against this it can be useful to set the initial size
472 slightly smaller than the smaller device with the aim that it will
473 still be larger than any replacement.
475 This value can be set with
477 for RAID level 1/4/5/6 though
479 based arrays such as those with IMSM metadata may not be able to
481 If the array was created with a size smaller than the currently
482 active drives, the extra space can be accessed using
484 The size can be given as
486 which means to choose the largest size that fits on all current drives.
488 Before reducing the size of the array (with
489 .BR "\-\-grow \-\-size=" )
490 you should make sure that space isn't needed. If the device holds a
491 filesystem, you would need to resize the filesystem to use less space.
493 After reducing the array size you should check that the data stored in
494 the device is still available. If the device holds a filesystem, then
495 an 'fsck' of the filesystem is a minimum requirement. If there are
496 problems the array can be made bigger again with no loss with another
497 .B "\-\-grow \-\-size="
500 This value cannot be used when creating a
502 such as with DDF and IMSM metadata, though it perfectly valid when
503 creating an array inside a container.
506 .BR \-Z ", " \-\-array\-size=
507 This is only meaningful with
509 and its effect is not persistent: when the array is stopped and
510 restarted the default array size will be restored.
512 Setting the array-size causes the array to appear smaller to programs
513 that access the data. This is particularly needed before reshaping an
514 array so that it will be smaller. As the reshape is not reversible,
515 but setting the size with
517 is, it is required that the array size is reduced as appropriate
518 before the number of devices in the array is reduced.
520 Before reducing the size of the array you should make sure that space
521 isn't needed. If the device holds a filesystem, you would need to
522 resize the filesystem to use less space.
524 After reducing the array size you should check that the data stored in
525 the device is still available. If the device holds a filesystem, then
526 an 'fsck' of the filesystem is a minimum requirement. If there are
527 problems the array can be made bigger again with no loss with another
528 .B "\-\-grow \-\-array\-size="
531 A suffix of 'M' or 'G' can be given to indicate Megabytes or
532 Gigabytes respectively.
535 restores the apparent size of the array to be whatever the real
536 amount of available space is.
539 .BR \-c ", " \-\-chunk=
540 Specify chunk size of kibibytes. The default when creating an
541 array is 512KB. To ensure compatibility with earlier versions, the
542 default when building an array with no persistent metadata is 64KB.
543 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
545 RAID4, RAID5, RAID6, and RAID10 require the chunk size to be a power
546 of 2. In any case it must be a multiple of 4KB.
548 A suffix of 'M' or 'G' can be given to indicate Megabytes or
549 Gigabytes respectively.
553 Specify rounding factor for a Linear array. The size of each
554 component will be rounded down to a multiple of this size.
555 This is a synonym for
557 but highlights the different meaning for Linear as compared to other
558 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
559 use, and is 0K (i.e. no rounding) in later kernels.
562 .BR \-l ", " \-\-level=
563 Set RAID level. When used with
565 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
566 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
567 Obviously some of these are synonymous.
571 metadata type is requested, only the
573 level is permitted, and it does not need to be explicitly given.
577 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
581 to change the RAID level in some cases. See LEVEL CHANGES below.
584 .BR \-p ", " \-\-layout=
585 This option configures the fine details of data layout for RAID5, RAID6,
586 and RAID10 arrays, and controls the failure modes for
589 The layout of the RAID5 parity block can be one of
590 .BR left\-asymmetric ,
591 .BR left\-symmetric ,
592 .BR right\-asymmetric ,
593 .BR right\-symmetric ,
594 .BR la ", " ra ", " ls ", " rs .
596 .BR left\-symmetric .
598 It is also possible to cause RAID5 to use a RAID4-like layout by
604 Finally for RAID5 there are DDF\-compatible layouts,
605 .BR ddf\-zero\-restart ,
606 .BR ddf\-N\-restart ,
608 .BR ddf\-N\-continue .
610 These same layouts are available for RAID6. There are also 4 layouts
611 that will provide an intermediate stage for converting between RAID5
612 and RAID6. These provide a layout which is identical to the
613 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
614 syndrome (the second 'parity' block used by RAID6) on the last device.
616 .BR left\-symmetric\-6 ,
617 .BR right\-symmetric\-6 ,
618 .BR left\-asymmetric\-6 ,
619 .BR right\-asymmetric\-6 ,
621 .BR parity\-first\-6 .
623 When setting the failure mode for level
626 .BR write\-transient ", " wt ,
627 .BR read\-transient ", " rt ,
628 .BR write\-persistent ", " wp ,
629 .BR read\-persistent ", " rp ,
631 .BR read\-fixable ", " rf ,
632 .BR clear ", " flush ", " none .
634 Each failure mode can be followed by a number, which is used as a period
635 between fault generation. Without a number, the fault is generated
636 once on the first relevant request. With a number, the fault will be
637 generated after that many requests, and will continue to be generated
638 every time the period elapses.
640 Multiple failure modes can be current simultaneously by using the
642 option to set subsequent failure modes.
644 "clear" or "none" will remove any pending or periodic failure modes,
645 and "flush" will clear any persistent faults.
647 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
648 by a small number. The default is 'n2'. The supported options are:
651 signals 'near' copies. Multiple copies of one data block are at
652 similar offsets in different devices.
655 signals 'offset' copies. Rather than the chunks being duplicated
656 within a stripe, whole stripes are duplicated but are rotated by one
657 device so duplicate blocks are on different devices. Thus subsequent
658 copies of a block are in the next drive, and are one chunk further
663 (multiple copies have very different offsets).
664 See md(4) for more detail about 'near', 'offset', and 'far'.
666 The number is the number of copies of each datablock. 2 is normal, 3
667 can be useful. This number can be at most equal to the number of
668 devices in the array. It does not need to divide evenly into that
669 number (e.g. it is perfectly legal to have an 'n2' layout for an array
670 with an odd number of devices).
672 When an array is converted between RAID5 and RAID6 an intermediate
673 RAID6 layout is used in which the second parity block (Q) is always on
674 the last device. To convert a RAID5 to RAID6 and leave it in this new
675 layout (which does not require re-striping) use
676 .BR \-\-layout=preserve .
677 This will try to avoid any restriping.
679 The converse of this is
680 .B \-\-layout=normalise
681 which will change a non-standard RAID6 layout into a more standard
688 (thus explaining the p of
692 .BR \-b ", " \-\-bitmap=
693 Specify a file to store a write-intent bitmap in. The file should not
696 is also given. The same file should be provided
697 when assembling the array. If the word
699 is given, then the bitmap is stored with the metadata on the array,
700 and so is replicated on all devices. If the word
704 mode, then any bitmap that is present is removed.
706 To help catch typing errors, the filename must contain at least one
707 slash ('/') if it is a real file (not 'internal' or 'none').
709 Note: external bitmaps are only known to work on ext2 and ext3.
710 Storing bitmap files on other filesystems may result in serious problems.
712 When creating an array on devices which are 100G or larger,
714 automatically adds an internal bitmap as it will usually be
715 beneficial. This can be suppressed with
716 .B "\-\-bitmap=none".
719 .BR \-\-bitmap\-chunk=
720 Set the chunksize of the bitmap. Each bit corresponds to that many
721 Kilobytes of storage.
722 When using a file based bitmap, the default is to use the smallest
723 size that is at-least 4 and requires no more than 2^21 chunks.
726 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
727 fit the bitmap into the available space.
729 A suffix of 'M' or 'G' can be given to indicate Megabytes or
730 Gigabytes respectively.
733 .BR \-W ", " \-\-write\-mostly
734 subsequent devices listed in a
739 command will be flagged as 'write-mostly'. This is valid for RAID1
740 only and means that the 'md' driver will avoid reading from these
741 devices if at all possible. This can be useful if mirroring over a
745 .BR \-\-write\-behind=
746 Specify that write-behind mode should be enabled (valid for RAID1
747 only). If an argument is specified, it will set the maximum number
748 of outstanding writes allowed. The default value is 256.
749 A write-intent bitmap is required in order to use write-behind
750 mode, and write-behind is only attempted on drives marked as
754 .BR \-\-assume\-clean
757 that the array pre-existed and is known to be clean. It can be useful
758 when trying to recover from a major failure as you can be sure that no
759 data will be affected unless you actually write to the array. It can
760 also be used when creating a RAID1 or RAID10 if you want to avoid the
761 initial resync, however this practice \(em while normally safe \(em is not
762 recommended. Use this only if you really know what you are doing.
764 When the devices that will be part of a new array were filled
765 with zeros before creation the operator knows the array is
766 actually clean. If that is the case, such as after running
767 badblocks, this argument can be used to tell mdadm the
768 facts the operator knows.
770 When an array is resized to a larger size with
771 .B "\-\-grow \-\-size="
772 the new space is normally resynced in that same way that the whole
773 array is resynced at creation. From Linux version 3.0,
775 can be used with that command to avoid the automatic resync.
778 .BR \-\-backup\-file=
781 is used to increase the number of raid-devices in a RAID5 or RAID6 if
782 there are no spare devices available, or to shrink, change RAID level
783 or layout. See the GROW MODE section below on RAID\-DEVICES CHANGES.
784 The file must be stored on a separate device, not on the RAID array
789 Arrays with 1.x metadata can leave a gap between the start of the
790 device and the start of array data. This gap can be used for various
791 metadata. The start of data is known as the
793 Normally an appropriate data offset is computed automatically.
794 However it can be useful to set it explicitly such as when re-creating
795 an array which was originally created using a different version of
797 which computed a different offset.
799 Setting the offset explicitly over-rides the default. The value given
800 is in Kilobytes unless an 'M' or 'G' suffix is given.
804 can also be used with
806 for some RAID levels (initially on RAID10). This allows the
807 data\-offset to be changed as part of the reshape process. When the
808 data offset is changed, no backup file is required as the difference
809 in offsets is used to provide the same functionality.
811 When the new offset is earlier than the old offset, the number of
812 devices in the array cannot shrink. When it is after the old offset,
813 the number of devices in the array cannot increase.
815 When creating an array,
819 In the case each member device is expected to have a offset appended
820 to the name, separated by a colon. This makes it possible to recreate
821 exactly an array which has varying data offsets (as can happen when
822 different versions of
824 are used to add different devices).
828 This option is complementary to the
829 .B \-\-freeze-reshape
830 option for assembly. It is needed when
832 operation is interrupted and it is not restarted automatically due to
833 .B \-\-freeze-reshape
834 usage during array assembly. This option is used together with
838 ) command and device for a pending reshape to be continued.
839 All parameters required for reshape continuation will be read from array metadata.
843 .BR \-\-backup\-file=
844 option to be set, continuation option will require to have exactly the same
845 backup file given as well.
847 Any other parameter passed together with
849 option will be ignored.
852 .BR \-N ", " \-\-name=
855 for the array. This is currently only effective when creating an
856 array with a version-1 superblock, or an array in a DDF container.
857 The name is a simple textual string that can be used to identify array
858 components when assembling. If name is needed but not specified, it
859 is taken from the basename of the device that is being created.
871 run the array, even if some of the components
872 appear to be active in another array or filesystem. Normally
874 will ask for confirmation before including such components in an
875 array. This option causes that question to be suppressed.
878 .BR \-f ", " \-\-force
881 accept the geometry and layout specified without question. Normally
883 will not allow creation of an array with only one device, and will try
884 to create a RAID5 array with one missing drive (as this makes the
885 initial resync work faster). With
888 will not try to be so clever.
891 .BR \-o ", " \-\-readonly
894 rather than read-write as normal. No writes will be allowed to the
895 array, and no resync, recovery, or reshape will be started.
898 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
899 Instruct mdadm how to create the device file if needed, possibly allocating
900 an unused minor number. "md" causes a non-partitionable array
901 to be used (though since Linux 2.6.28, these array devices are in fact
902 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
903 later) to be used. "yes" requires the named md device to have
904 a 'standard' format, and the type and minor number will be determined
905 from this. With mdadm 3.0, device creation is normally left up to
907 so this option is unlikely to be needed.
908 See DEVICE NAMES below.
910 The argument can also come immediately after
915 is not given on the command line or in the config file, then
921 is also given, then any
923 entries in the config file will override the
925 instruction given on the command line.
927 For partitionable arrays,
929 will create the device file for the whole array and for the first 4
930 partitions. A different number of partitions can be specified at the
931 end of this option (e.g.
933 If the device name ends with a digit, the partition names add a 'p',
935 .IR /dev/md/home1p3 .
936 If there is no trailing digit, then the partition names just have a
938 .IR /dev/md/scratch3 .
940 If the md device name is in a 'standard' format as described in DEVICE
941 NAMES, then it will be created, if necessary, with the appropriate
942 device number based on that name. If the device name is not in one of these
943 formats, then a unused device number will be allocated. The device
944 number will be considered unused if there is no active array for that
945 number, and there is no entry in /dev for that number and with a
946 non-standard name. Names that are not in 'standard' format are only
947 allowed in "/dev/md/".
949 This is meaningful with
955 .BR \-a ", " "\-\-add"
956 This option can be used in Grow mode in two cases.
958 If the target array is a Linear array, then
960 can be used to add one or more devices to the array. They
961 are simply catenated on to the end of the array. Once added, the
962 devices cannot be removed.
966 option is being used to increase the number of devices in an array,
969 can be used to add some extra devices to be included in the array.
970 In most cases this is not needed as the extra devices can be added as
971 spares first, and then the number of raid-disks can be changed.
972 However for RAID0, it is not possible to add spares. So to increase
973 the number of devices in a RAID0, it is necessary to set the new
974 number of devices, and to add the new devices, in the same command.
979 .BR \-u ", " \-\-uuid=
980 uuid of array to assemble. Devices which don't have this uuid are
984 .BR \-m ", " \-\-super\-minor=
985 Minor number of device that array was created for. Devices which
986 don't have this minor number are excluded. If you create an array as
987 /dev/md1, then all superblocks will contain the minor number 1, even if
988 the array is later assembled as /dev/md2.
990 Giving the literal word "dev" for
994 to use the minor number of the md device that is being assembled.
997 .B \-\-super\-minor=dev
998 will look for super blocks with a minor number of 0.
1001 is only relevant for v0.90 metadata, and should not normally be used.
1007 .BR \-N ", " \-\-name=
1008 Specify the name of the array to assemble. This must be the name
1009 that was specified when creating the array. It must either match
1010 the name stored in the superblock exactly, or it must match
1013 prefixed to the start of the given name.
1016 .BR \-f ", " \-\-force
1017 Assemble the array even if the metadata on some devices appears to be
1020 cannot find enough working devices to start the array, but can find
1021 some devices that are recorded as having failed, then it will mark
1022 those devices as working so that the array can be started.
1023 An array which requires
1025 to be started may contain data corruption. Use it carefully.
1028 .BR \-R ", " \-\-run
1029 Attempt to start the array even if fewer drives were given than were
1030 present last time the array was active. Normally if not all the
1031 expected drives are found and
1033 is not used, then the array will be assembled but not started.
1036 an attempt will be made to start it anyway.
1040 This is the reverse of
1042 in that it inhibits the startup of array unless all expected drives
1043 are present. This is only needed with
1045 and can be used if the physical connections to devices are
1046 not as reliable as you would like.
1049 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
1050 See this option under Create and Build options.
1053 .BR \-b ", " \-\-bitmap=
1054 Specify the bitmap file that was given when the array was created. If
1057 bitmap, there is no need to specify this when assembling the array.
1060 .BR \-\-backup\-file=
1063 was used while reshaping an array (e.g. changing number of devices or
1064 chunk size) and the system crashed during the critical section, then the same
1066 must be presented to
1068 to allow possibly corrupted data to be restored, and the reshape
1072 .BR \-\-invalid\-backup
1073 If the file needed for the above option is not available for any
1074 reason an empty file can be given together with this option to
1075 indicate that the backup file is invalid. In this case the data that
1076 was being rearranged at the time of the crash could be irrecoverably
1077 lost, but the rest of the array may still be recoverable. This option
1078 should only be used as a last resort if there is no way to recover the
1083 .BR \-U ", " \-\-update=
1084 Update the superblock on each device while assembling the array. The
1085 argument given to this flag can be one of
1103 option will adjust the superblock of an array what was created on a Sparc
1104 machine running a patched 2.2 Linux kernel. This kernel got the
1105 alignment of part of the superblock wrong. You can use the
1106 .B "\-\-examine \-\-sparc2.2"
1109 to see what effect this would have.
1113 option will update the
1114 .B "preferred minor"
1115 field on each superblock to match the minor number of the array being
1117 This can be useful if
1119 reports a different "Preferred Minor" to
1121 In some cases this update will be performed automatically
1122 by the kernel driver. In particular the update happens automatically
1123 at the first write to an array with redundancy (RAID level 1 or
1124 greater) on a 2.6 (or later) kernel.
1128 option will change the uuid of the array. If a UUID is given with the
1130 option that UUID will be used as a new UUID and will
1132 be used to help identify the devices in the array.
1135 is given, a random UUID is chosen.
1139 option will change the
1141 of the array as stored in the superblock. This is only supported for
1142 version-1 superblocks.
1146 option will change the
1148 as recorded in the superblock. For version-0 superblocks, this is the
1149 same as updating the UUID.
1150 For version-1 superblocks, this involves updating the name.
1154 option will cause the array to be marked
1156 meaning that any redundancy in the array (e.g. parity for RAID5,
1157 copies for RAID1) may be incorrect. This will cause the RAID system
1158 to perform a "resync" pass to make sure that all redundant information
1163 option allows arrays to be moved between machines with different
1165 When assembling such an array for the first time after a move, giving
1166 .B "\-\-update=byteorder"
1169 to expect superblocks to have their byteorder reversed, and will
1170 correct that order before assembling the array. This is only valid
1171 with original (Version 0.90) superblocks.
1175 option will correct the summaries in the superblock. That is the
1176 counts of total, working, active, failed, and spare devices.
1180 option will rarely be of use. It applies to version 1.1 and 1.2 metadata
1181 only (where the metadata is at the start of the device) and is only
1182 useful when the component device has changed size (typically become
1183 larger). The version 1 metadata records the amount of the device that
1184 can be used to store data, so if a device in a version 1.1 or 1.2
1185 array becomes larger, the metadata will still be visible, but the
1186 extra space will not. In this case it might be useful to assemble the
1188 .BR \-\-update=devicesize .
1191 to determine the maximum usable amount of space on each device and
1192 update the relevant field in the metadata.
1196 option only works on v0.90 metadata arrays and will convert them to
1197 v1.0 metadata. The array must not be dirty (i.e. it must not need a
1198 sync) and it must not have a write-intent bitmap.
1200 The old metadata will remain on the devices, but will appear older
1201 than the new metadata and so will usually be ignored. The old metadata
1202 (or indeed the new metadata) can be removed by giving the appropriate
1205 .BR \-\-zero\-superblock .
1209 option can be used when an array has an internal bitmap which is
1210 corrupt in some way so that assembling the array normally fails. It
1211 will cause any internal bitmap to be ignored.
1215 option will reserve space in each device for a bad block list. This
1216 will be 4K in size and positioned near the end of any free space
1217 between the superblock and the data.
1221 option will cause any reservation of space for a bad block list to be
1222 removed. If the bad block list contains entries, this will fail, as
1223 removing the list could cause data corruption.
1226 .BR \-\-freeze\-reshape
1227 Option is intended to be used in start-up scripts during initrd boot phase.
1228 When array under reshape is assembled during initrd phase, this option
1229 stops reshape after reshape critical section is being restored. This happens
1230 before file system pivot operation and avoids loss of file system context.
1231 Losing file system context would cause reshape to be broken.
1233 Reshape can be continued later using the
1235 option for the grow command.
1237 .SH For Manage mode:
1240 .BR \-t ", " \-\-test
1241 Unless a more serious error occurred,
1243 will exit with a status of 2 if no changes were made to the array and
1244 0 if at least one change was made.
1245 This can be useful when an indirect specifier such as
1250 is used in requesting an operation on the array.
1252 will report failure if these specifiers didn't find any match.
1255 .BR \-a ", " \-\-add
1256 hot-add listed devices.
1257 If a device appears to have recently been part of the array
1258 (possibly it failed or was removed) the device is re\-added as described
1260 If that fails or the device was never part of the array, the device is
1261 added as a hot-spare.
1262 If the array is degraded, it will immediately start to rebuild data
1265 Note that this and the following options are only meaningful on array
1266 with redundancy. They don't apply to RAID0 or Linear.
1270 re\-add a device that was previously removed from an array.
1271 If the metadata on the device reports that it is a member of the
1272 array, and the slot that it used is still vacant, then the device will
1273 be added back to the array in the same position. This will normally
1274 cause the data for that device to be recovered. However based on the
1275 event count on the device, the recovery may only require sections that
1276 are flagged a write-intent bitmap to be recovered or may not require
1277 any recovery at all.
1279 When used on an array that has no metadata (i.e. it was built with
1281 it will be assumed that bitmap-based recovery is enough to make the
1282 device fully consistent with the array.
1284 When used with v1.x metadata,
1286 can be accompanied by
1287 .BR \-\-update=devicesize ,
1288 .BR \-\-update=bbl ", or"
1289 .BR \-\-update=no\-bbl .
1290 See the description of these option when used in Assemble mode for an
1291 explanation of their use.
1293 If the device name given is
1297 will try to find any device that looks like it should be
1298 part of the array but isn't and will try to re\-add all such devices.
1300 If the device name given is
1304 will find all devices in the array that are marked
1306 remove them and attempt to immediately re\-add them. This can be
1307 useful if you are certain that the reason for failure has been
1312 Add a device as a spare. This is similar to
1314 except that it does not attempt
1316 first. The device will be added as a spare even if it looks like it
1317 could be an recent member of the array.
1320 .BR \-r ", " \-\-remove
1321 remove listed devices. They must not be active. i.e. they should
1322 be failed or spare devices.
1324 As well as the name of a device file
1334 The first causes all failed device to be removed. The second causes
1335 any device which is no longer connected to the system (i.e an 'open'
1339 The third will remove a set as describe below under
1343 .BR \-f ", " \-\-fail
1344 Mark listed devices as faulty.
1345 As well as the name of a device file, the word
1349 can be given. The former will cause any device that has been detached from
1350 the system to be marked as failed. It can then be removed.
1352 For RAID10 arrays where the number of copies evenly divides the number
1353 of devices, the devices can be conceptually divided into sets where
1354 each set contains a single complete copy of the data on the array.
1355 Sometimes a RAID10 array will be configured so that these sets are on
1356 separate controllers. In this case all the devices in one set can be
1357 failed by giving a name like
1363 The appropriate set names are reported by
1373 Mark listed devices as requiring replacement. As soon as a spare is
1374 available, it will be rebuilt and will replace the marked device.
1375 This is similar to marking a device as faulty, but the device remains
1376 in service during the recovery process to increase resilience against
1377 multiple failures. When the replacement process finishes, the
1378 replaced device will be marked as faulty.
1382 This can follow a list of
1384 devices. The devices listed after
1386 will be preferentially used to replace the devices listed after
1388 These device must already be spare devices in the array.
1391 .BR \-\-write\-mostly
1392 Subsequent devices that are added or re\-added will have the 'write-mostly'
1393 flag set. This is only valid for RAID1 and means that the 'md' driver
1394 will avoid reading from these devices if possible.
1397 Subsequent devices that are added or re\-added will have the 'write-mostly'
1401 Each of these options requires that the first device listed is the array
1402 to be acted upon, and the remainder are component devices to be added,
1403 removed, marked as faulty, etc. Several different operations can be
1404 specified for different devices, e.g.
1406 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1408 Each operation applies to all devices listed until the next
1411 If an array is using a write-intent bitmap, then devices which have
1412 been removed can be re\-added in a way that avoids a full
1413 reconstruction but instead just updates the blocks that have changed
1414 since the device was removed. For arrays with persistent metadata
1415 (superblocks) this is done automatically. For arrays created with
1417 mdadm needs to be told that this device we removed recently with
1420 Devices can only be removed from an array if they are not in active
1421 use, i.e. that must be spares or failed devices. To remove an active
1422 device, it must first be marked as
1428 .BR \-Q ", " \-\-query
1429 Examine a device to see
1430 (1) if it is an md device and (2) if it is a component of an md
1432 Information about what is discovered is presented.
1435 .BR \-D ", " \-\-detail
1436 Print details of one or more md devices.
1439 .BR \-\-detail\-platform
1440 Print details of the platform's RAID capabilities (firmware / hardware
1441 topology) for a given metadata format. If used without argument, mdadm
1442 will scan all controllers looking for their capabilities. Otherwise, mdadm
1443 will only look at the controller specified by the argument in form of an
1444 absolute filepath or a link, e.g.
1445 .IR /sys/devices/pci0000:00/0000:00:1f.2 .
1448 .BR \-Y ", " \-\-export
1451 .BR \-\-detail-platform ,
1455 output will be formatted as
1457 pairs for easy import into the environment.
1463 indicates whether an array was started
1465 or not, which may include a reason
1466 .RB ( unsafe ", " nothing ", " no ).
1469 indicates if the array is expected on this host
1471 or seems to be from elsewhere
1475 .BR \-E ", " \-\-examine
1476 Print contents of the metadata stored on the named device(s).
1477 Note the contrast between
1482 applies to devices which are components of an array, while
1484 applies to a whole array which is currently active.
1487 If an array was created on a SPARC machine with a 2.2 Linux kernel
1488 patched with RAID support, the superblock will have been created
1489 incorrectly, or at least incompatibly with 2.4 and later kernels.
1494 will fix the superblock before displaying it. If this appears to do
1495 the right thing, then the array can be successfully assembled using
1496 .BR "\-\-assemble \-\-update=sparc2.2" .
1499 .BR \-X ", " \-\-examine\-bitmap
1500 Report information about a bitmap file.
1501 The argument is either an external bitmap file or an array component
1502 in case of an internal bitmap. Note that running this on an array
1505 does not report the bitmap for that array.
1508 .B \-\-examine\-badblocks
1509 List the bad-blocks recorded for the device, if a bad-blocks list has
1510 been configured. Currently only
1512 metadata supports bad-blocks lists.
1515 .BI \-\-dump= directory
1517 .BI \-\-restore= directory
1518 Save metadata from lists devices, or restore metadata to listed devices.
1521 .BR \-R ", " \-\-run
1522 start a partially assembled array. If
1524 did not find enough devices to fully start the array, it might leaving
1525 it partially assembled. If you wish, you can then use
1527 to start the array in degraded mode.
1530 .BR \-S ", " \-\-stop
1531 deactivate array, releasing all resources.
1534 .BR \-o ", " \-\-readonly
1535 mark array as readonly.
1538 .BR \-w ", " \-\-readwrite
1539 mark array as readwrite.
1542 .B \-\-zero\-superblock
1543 If the device contains a valid md superblock, the block is
1544 overwritten with zeros. With
1546 the block where the superblock would be is overwritten even if it
1547 doesn't appear to be valid.
1550 .B \-\-kill\-subarray=
1551 If the device is a container and the argument to \-\-kill\-subarray
1552 specifies an inactive subarray in the container, then the subarray is
1553 deleted. Deleting all subarrays will leave an 'empty-container' or
1554 spare superblock on the drives. See
1555 .B \-\-zero\-superblock
1557 removing a superblock. Note that some formats depend on the subarray
1558 index for generating a UUID, this command will fail if it would change
1559 the UUID of an active subarray.
1562 .B \-\-update\-subarray=
1563 If the device is a container and the argument to \-\-update\-subarray
1564 specifies a subarray in the container, then attempt to update the given
1565 superblock field in the subarray. See below in
1570 .BR \-t ", " \-\-test
1575 is set to reflect the status of the device. See below in
1580 .BR \-W ", " \-\-wait
1581 For each md device given, wait for any resync, recovery, or reshape
1582 activity to finish before returning.
1584 will return with success if it actually waited for every device
1585 listed, otherwise it will return failure.
1589 For each md device given, or each device in /proc/mdstat if
1591 is given, arrange for the array to be marked clean as soon as possible.
1593 will return with success if the array uses external metadata and we
1594 successfully waited. For native arrays this returns immediately as the
1595 kernel handles dirty-clean transitions at shutdown. No action is taken
1596 if safe-mode handling is disabled.
1600 Set the "sync_action" for all md devices given to one of
1607 will abort any currently running action though some actions will
1608 automatically restart.
1611 will abort any current action and ensure no other action starts
1621 .BR "SCRUBBING AND MISMATCHES" .
1623 .SH For Incremental Assembly mode:
1625 .BR \-\-rebuild\-map ", " \-r
1626 Rebuild the map file
1630 uses to help track which arrays are currently being assembled.
1633 .BR \-\-run ", " \-R
1634 Run any array assembled as soon as a minimal number of devices are
1635 available, rather than waiting until all expected devices are present.
1638 .BR \-\-scan ", " \-s
1639 Only meaningful with
1643 file for arrays that are being incrementally assembled and will try to
1644 start any that are not already started. If any such array is listed
1647 as requiring an external bitmap, that bitmap will be attached first.
1650 .BR \-\-fail ", " \-f
1651 This allows the hot-plug system to remove devices that have fully disappeared
1652 from the kernel. It will first fail and then remove the device from any
1653 array it belongs to.
1654 The device name given should be a kernel device name such as "sda",
1660 Only used with \-\-fail. The 'path' given will be recorded so that if
1661 a new device appears at the same location it can be automatically
1662 added to the same array. This allows the failed device to be
1663 automatically replaced by a new device without metadata if it appears
1664 at specified path. This option is normally only set by a
1668 .SH For Monitor mode:
1670 .BR \-m ", " \-\-mail
1671 Give a mail address to send alerts to.
1674 .BR \-p ", " \-\-program ", " \-\-alert
1675 Give a program to be run whenever an event is detected.
1678 .BR \-y ", " \-\-syslog
1679 Cause all events to be reported through 'syslog'. The messages have
1680 facility of 'daemon' and varying priorities.
1683 .BR \-d ", " \-\-delay
1684 Give a delay in seconds.
1686 polls the md arrays and then waits this many seconds before polling
1687 again. The default is 60 seconds. Since 2.6.16, there is no need to
1688 reduce this as the kernel alerts
1690 immediately when there is any change.
1693 .BR \-r ", " \-\-increment
1694 Give a percentage increment.
1696 will generate RebuildNN events with the given percentage increment.
1699 .BR \-f ", " \-\-daemonise
1702 to run as a background daemon if it decides to monitor anything. This
1703 causes it to fork and run in the child, and to disconnect from the
1704 terminal. The process id of the child is written to stdout.
1707 which will only continue monitoring if a mail address or alert program
1708 is found in the config file.
1711 .BR \-i ", " \-\-pid\-file
1714 is running in daemon mode, write the pid of the daemon process to
1715 the specified file, instead of printing it on standard output.
1718 .BR \-1 ", " \-\-oneshot
1719 Check arrays only once. This will generate
1721 events and more significantly
1727 .B " mdadm \-\-monitor \-\-scan \-1"
1729 from a cron script will ensure regular notification of any degraded arrays.
1732 .BR \-t ", " \-\-test
1735 alert for every array found at startup. This alert gets mailed and
1736 passed to the alert program. This can be used for testing that alert
1737 message do get through successfully.
1741 This inhibits the functionality for moving spares between arrays.
1742 Only one monitoring process started with
1744 but without this flag is allowed, otherwise the two could interfere
1751 .B mdadm \-\-assemble
1752 .I md-device options-and-component-devices...
1755 .B mdadm \-\-assemble \-\-scan
1756 .I md-devices-and-options...
1759 .B mdadm \-\-assemble \-\-scan
1763 This usage assembles one or more RAID arrays from pre-existing components.
1764 For each array, mdadm needs to know the md device, the identity of the
1765 array, and a number of component-devices. These can be found in a number of ways.
1767 In the first usage example (without the
1769 the first device given is the md device.
1770 In the second usage example, all devices listed are treated as md
1771 devices and assembly is attempted.
1772 In the third (where no devices are listed) all md devices that are
1773 listed in the configuration file are assembled. If no arrays are
1774 described by the configuration file, then any arrays that
1775 can be found on unused devices will be assembled.
1777 If precisely one device is listed, but
1783 was given and identity information is extracted from the configuration file.
1785 The identity can be given with the
1791 option, will be taken from the md-device record in the config file, or
1792 will be taken from the super block of the first component-device
1793 listed on the command line.
1795 Devices can be given on the
1797 command line or in the config file. Only devices which have an md
1798 superblock which contains the right identity will be considered for
1801 The config file is only used if explicitly named with
1803 or requested with (a possibly implicit)
1808 .B /etc/mdadm/mdadm.conf
1813 is not given, then the config file will only be used to find the
1814 identity of md arrays.
1816 Normally the array will be started after it is assembled. However if
1818 is not given and not all expected drives were listed, then the array
1819 is not started (to guard against usage errors). To insist that the
1820 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1829 does not create any entries in
1833 It does record information in
1837 to choose the correct name.
1841 detects that udev is not configured, it will create the devices in
1845 In Linux kernels prior to version 2.6.28 there were two distinctly
1846 different types of md devices that could be created: one that could be
1847 partitioned using standard partitioning tools and one that could not.
1848 Since 2.6.28 that distinction is no longer relevant as both type of
1849 devices can be partitioned.
1851 will normally create the type that originally could not be partitioned
1852 as it has a well defined major number (9).
1854 Prior to 2.6.28, it is important that mdadm chooses the correct type
1855 of array device to use. This can be controlled with the
1857 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1858 to use a partitionable device rather than the default.
1860 In the no-udev case, the value given to
1862 can be suffixed by a number. This tells
1864 to create that number of partition devices rather than the default of 4.
1868 can also be given in the configuration file as a word starting
1870 on the ARRAY line for the relevant array.
1877 and no devices are listed,
1879 will first attempt to assemble all the arrays listed in the config
1882 If no arrays are listed in the config (other than those marked
1884 it will look through the available devices for possible arrays and
1885 will try to assemble anything that it finds. Arrays which are tagged
1886 as belonging to the given homehost will be assembled and started
1887 normally. Arrays which do not obviously belong to this host are given
1888 names that are expected not to conflict with anything local, and are
1889 started "read-auto" so that nothing is written to any device until the
1890 array is written to. i.e. automatic resync etc is delayed.
1894 finds a consistent set of devices that look like they should comprise
1895 an array, and if the superblock is tagged as belonging to the given
1896 home host, it will automatically choose a device name and try to
1897 assemble the array. If the array uses version-0.90 metadata, then the
1899 number as recorded in the superblock is used to create a name in
1903 If the array uses version-1 metadata, then the
1905 from the superblock is used to similarly create a name in
1907 (the name will have any 'host' prefix stripped first).
1909 This behaviour can be modified by the
1913 configuration file. This line can indicate that specific metadata
1914 type should, or should not, be automatically assembled. If an array
1915 is found which is not listed in
1917 and has a metadata format that is denied by the
1919 line, then it will not be assembled.
1922 line can also request that all arrays identified as being for this
1923 homehost should be assembled regardless of their metadata type.
1926 for further details.
1928 Note: Auto assembly cannot be used for assembling and activating some
1929 arrays which are undergoing reshape. In particular as the
1931 cannot be given, any reshape which requires a backup-file to continue
1932 cannot be started by auto assembly. An array which is growing to more
1933 devices and has passed the critical section can be assembled using
1944 .BI \-\-raid\-devices= Z
1948 This usage is similar to
1950 The difference is that it creates an array without a superblock. With
1951 these arrays there is no difference between initially creating the array and
1952 subsequently assembling the array, except that hopefully there is useful
1953 data there in the second case.
1955 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1956 one of their synonyms. All devices must be listed and the array will
1957 be started once complete. It will often be appropriate to use
1958 .B \-\-assume\-clean
1959 with levels raid1 or raid10.
1970 .BI \-\-raid\-devices= Z
1974 This usage will initialise a new md array, associate some devices with
1975 it, and activate the array.
1977 The named device will normally not exist when
1978 .I "mdadm \-\-create"
1979 is run, but will be created by
1981 once the array becomes active.
1983 As devices are added, they are checked to see if they contain RAID
1984 superblocks or filesystems. They are also checked to see if the variance in
1985 device size exceeds 1%.
1987 If any discrepancy is found, the array will not automatically be run, though
1990 can override this caution.
1992 To create a "degraded" array in which some devices are missing, simply
1993 give the word "\fBmissing\fP"
1994 in place of a device name. This will cause
1996 to leave the corresponding slot in the array empty.
1997 For a RAID4 or RAID5 array at most one slot can be
1998 "\fBmissing\fP"; for a RAID6 array at most two slots.
1999 For a RAID1 array, only one real device needs to be given. All of the
2003 When creating a RAID5 array,
2005 will automatically create a degraded array with an extra spare drive.
2006 This is because building the spare into a degraded array is in general
2007 faster than resyncing the parity on a non-degraded, but not clean,
2008 array. This feature can be overridden with the
2012 When creating an array with version-1 metadata a name for the array is
2014 If this is not given with the
2018 will choose a name based on the last component of the name of the
2019 device being created. So if
2021 is being created, then the name
2026 is being created, then the name
2030 When creating a partition based array, using
2032 with version-1.x metadata, the partition type should be set to
2034 (non fs-data). This type selection allows for greater precision since
2035 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
2036 might create problems in the event of array recovery through a live cdrom.
2038 A new array will normally get a randomly assigned 128bit UUID which is
2039 very likely to be unique. If you have a specific need, you can choose
2040 a UUID for the array by giving the
2042 option. Be warned that creating two arrays with the same UUID is a
2043 recipe for disaster. Also, using
2045 when creating a v0.90 array will silently override any
2050 .\"option is given, it is not necessary to list any component-devices in this command.
2051 .\"They can be added later, before a
2055 .\"is given, the apparent size of the smallest drive given is used.
2057 If the array type supports a write-intent bitmap, and if the devices
2058 in the array exceed 100G is size, an internal write-intent bitmap
2059 will automatically be added unless some other option is explicitly
2062 option. In any case space for a bitmap will be reserved so that one
2063 can be added layer with
2064 .BR "\-\-grow \-\-bitmap=internal" .
2066 If the metadata type supports it (currently only 1.x metadata), space
2067 will be allocated to store a bad block list. This allows a modest
2068 number of bad blocks to be recorded, allowing the drive to remain in
2069 service while only partially functional.
2071 When creating an array within a
2074 can be given either the list of devices to use, or simply the name of
2075 the container. The former case gives control over which devices in
2076 the container will be used for the array. The latter case allows
2078 to automatically choose which devices to use based on how much spare
2081 The General Management options that are valid with
2086 insist on running the array even if some devices look like they might
2091 start the array readonly \(em not supported yet.
2098 .I options... devices...
2101 This usage will allow individual devices in an array to be failed,
2102 removed or added. It is possible to perform multiple operations with
2103 on command. For example:
2105 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
2111 and will then remove it from the array and finally add it back
2112 in as a spare. However only one md array can be affected by a single
2115 When a device is added to an active array, mdadm checks to see if it
2116 has metadata on it which suggests that it was recently a member of the
2117 array. If it does, it tries to "re\-add" the device. If there have
2118 been no changes since the device was removed, or if the array has a
2119 write-intent bitmap which has recorded whatever changes there were,
2120 then the device will immediately become a full member of the array and
2121 those differences recorded in the bitmap will be resolved.
2131 MISC mode includes a number of distinct operations that
2132 operate on distinct devices. The operations are:
2135 The device is examined to see if it is
2136 (1) an active md array, or
2137 (2) a component of an md array.
2138 The information discovered is reported.
2142 The device should be an active md device.
2144 will display a detailed description of the array.
2148 will cause the output to be less detailed and the format to be
2149 suitable for inclusion in
2153 will normally be 0 unless
2155 failed to get useful information about the device(s); however, if the
2157 option is given, then the exit status will be:
2161 The array is functioning normally.
2164 The array has at least one failed device.
2167 The array has multiple failed devices such that it is unusable.
2170 There was an error while trying to get information about the device.
2174 .B \-\-detail\-platform
2175 Print detail of the platform's RAID capabilities (firmware / hardware
2176 topology). If the metadata is specified with
2180 then the return status will be:
2184 metadata successfully enumerated its platform components on this system
2187 metadata is platform independent
2190 metadata failed to find its platform components on this system
2194 .B \-\-update\-subarray=
2195 If the device is a container and the argument to \-\-update\-subarray
2196 specifies a subarray in the container, then attempt to update the given
2197 superblock field in the subarray. Similar to updating an array in
2198 "assemble" mode, the field to update is selected by
2202 option. Currently only
2208 option updates the subarray name in the metadata, it may not affect the
2209 device node name or the device node symlink until the subarray is
2210 re\-assembled. If updating
2212 would change the UUID of an active subarray this operation is blocked,
2213 and the command will end in an error.
2217 The device should be a component of an md array.
2219 will read the md superblock of the device and display the contents.
2224 is given, then multiple devices that are components of the one array
2225 are grouped together and reported in a single entry suitable
2231 without listing any devices will cause all devices listed in the
2232 config file to be examined.
2235 .BI \-\-dump= directory
2236 If the device contains RAID metadata, a file will be created in the
2238 and the metadata will be written to it. The file will be the same
2239 size as the device and have the metadata written in the file at the
2240 same locate that it exists in the device. However the file will be "sparse" so
2241 that only those blocks containing metadata will be allocated. The
2242 total space used will be small.
2244 The file name used in the
2246 will be the base name of the device. Further if any links appear in
2248 which point to the device, then hard links to the file will be created
2255 Multiple devices can be listed and their metadata will all be stored
2256 in the one directory.
2259 .BI \-\-restore= directory
2260 This is the reverse of
2263 will locate a file in the directory that has a name appropriate for
2264 the given device and will restore metadata from it. Names that match
2266 names are preferred, however if two of those refer to different files,
2268 will not choose between them but will abort the operation.
2270 If a file name is given instead of a
2274 will restore from that file to a single device, always provided the
2275 size of the file matches that of the device, and the file contains
2279 The devices should be active md arrays which will be deactivated, as
2280 long as they are not currently in use.
2284 This will fully activate a partially assembled md array.
2288 This will mark an active array as read-only, providing that it is
2289 not currently being used.
2295 array back to being read/write.
2299 For all operations except
2302 will cause the operation to be applied to all arrays listed in
2307 causes all devices listed in the config file to be examined.
2310 .BR \-b ", " \-\-brief
2311 Be less verbose. This is used with
2319 gives an intermediate level of verbosity.
2325 .B mdadm \-\-monitor
2326 .I options... devices...
2331 to periodically poll a number of md arrays and to report on any events
2334 will never exit once it decides that there are arrays to be checked,
2335 so it should normally be run in the background.
2337 As well as reporting events,
2339 may move a spare drive from one array to another if they are in the
2344 and if the destination array has a failed drive but no spares.
2346 If any devices are listed on the command line,
2348 will only monitor those devices. Otherwise all arrays listed in the
2349 configuration file will be monitored. Further, if
2351 is given, then any other md devices that appear in
2353 will also be monitored.
2355 The result of monitoring the arrays is the generation of events.
2356 These events are passed to a separate program (if specified) and may
2357 be mailed to a given E-mail address.
2359 When passing events to a program, the program is run once for each event,
2360 and is given 2 or 3 command-line arguments: the first is the
2361 name of the event (see below), the second is the name of the
2362 md device which is affected, and the third is the name of a related
2363 device if relevant (such as a component device that has failed).
2367 is given, then a program or an E-mail address must be specified on the
2368 command line or in the config file. If neither are available, then
2370 will not monitor anything.
2374 will continue monitoring as long as something was found to monitor. If
2375 no program or email is given, then each event is reported to
2378 The different events are:
2382 .B DeviceDisappeared
2383 An md array which previously was configured appears to no longer be
2384 configured. (syslog priority: Critical)
2388 was told to monitor an array which is RAID0 or Linear, then it will
2390 .B DeviceDisappeared
2391 with the extra information
2393 This is because RAID0 and Linear do not support the device-failed,
2394 hot-spare and resync operations which are monitored.
2398 An md array started reconstruction (e.g. recovery, resync, reshape,
2399 check, repair). (syslog priority: Warning)
2405 is a two-digit number (ie. 05, 48). This indicates that rebuild
2406 has passed that many percent of the total. The events are generated
2407 with fixed increment since 0. Increment size may be specified with
2408 a commandline option (default is 20). (syslog priority: Warning)
2412 An md array that was rebuilding, isn't any more, either because it
2413 finished normally or was aborted. (syslog priority: Warning)
2417 An active component device of an array has been marked as
2418 faulty. (syslog priority: Critical)
2422 A spare component device which was being rebuilt to replace a faulty
2423 device has failed. (syslog priority: Critical)
2427 A spare component device which was being rebuilt to replace a faulty
2428 device has been successfully rebuilt and has been made active.
2429 (syslog priority: Info)
2433 A new md array has been detected in the
2435 file. (syslog priority: Info)
2439 A newly noticed array appears to be degraded. This message is not
2442 notices a drive failure which causes degradation, but only when
2444 notices that an array is degraded when it first sees the array.
2445 (syslog priority: Critical)
2449 A spare drive has been moved from one array in a
2453 to another to allow a failed drive to be replaced.
2454 (syslog priority: Info)
2460 has been told, via the config file, that an array should have a certain
2461 number of spare devices, and
2463 detects that it has fewer than this number when it first sees the
2464 array, it will report a
2467 (syslog priority: Warning)
2471 An array was found at startup, and the
2474 (syslog priority: Info)
2484 cause Email to be sent. All events cause the program to be run.
2485 The program is run with two or three arguments: the event
2486 name, the array device and possibly a second device.
2488 Each event has an associated array device (e.g.
2490 and possibly a second device. For
2495 the second device is the relevant component device.
2498 the second device is the array that the spare was moved from.
2502 to move spares from one array to another, the different arrays need to
2503 be labeled with the same
2505 or the spares must be allowed to migrate through matching POLICY domains
2506 in the configuration file. The
2508 name can be any string; it is only necessary that different spare
2509 groups use different names.
2513 detects that an array in a spare group has fewer active
2514 devices than necessary for the complete array, and has no spare
2515 devices, it will look for another array in the same spare group that
2516 has a full complement of working drive and a spare. It will then
2517 attempt to remove the spare from the second drive and add it to the
2519 If the removal succeeds but the adding fails, then it is added back to
2522 If the spare group for a degraded array is not defined,
2524 will look at the rules of spare migration specified by POLICY lines in
2526 and then follow similar steps as above if a matching spare is found.
2529 The GROW mode is used for changing the size or shape of an active
2531 For this to work, the kernel must support the necessary change.
2532 Various types of growth are being added during 2.6 development.
2534 Currently the supported changes include
2536 change the "size" attribute for RAID1, RAID4, RAID5 and RAID6.
2538 increase or decrease the "raid\-devices" attribute of RAID0, RAID1, RAID4,
2541 change the chunk-size and layout of RAID0, RAID4, RAID5, RAID6 and RAID10.
2543 convert between RAID1 and RAID5, between RAID5 and RAID6, between
2544 RAID0, RAID4, and RAID5, and between RAID0 and RAID10 (in the near-2 mode).
2546 add a write-intent bitmap to any array which supports these bitmaps, or
2547 remove a write-intent bitmap from such an array.
2550 Using GROW on containers is currently supported only for Intel's IMSM
2551 container format. The number of devices in a container can be
2552 increased - which affects all arrays in the container - or an array
2553 in a container can be converted between levels where those levels are
2554 supported by the container, and the conversion is on of those listed
2555 above. Resizing arrays in an IMSM container with
2557 is not yet supported.
2559 Grow functionality (e.g. expand a number of raid devices) for Intel's
2560 IMSM container format has an experimental status. It is guarded by the
2561 .B MDADM_EXPERIMENTAL
2562 environment variable which must be set to '1' for a GROW command to
2564 This is for the following reasons:
2567 Intel's native IMSM check-pointing is not fully tested yet.
2568 This can causes IMSM incompatibility during the grow process: an array
2569 which is growing cannot roam between Microsoft Windows(R) and Linux
2573 Interrupting a grow operation is not recommended, because it
2574 has not been fully tested for Intel's IMSM container format yet.
2577 Note: Intel's native checkpointing doesn't use
2579 option and it is transparent for assembly feature.
2582 Normally when an array is built the "size" is taken from the smallest
2583 of the drives. If all the small drives in an arrays are, one at a
2584 time, removed and replaced with larger drives, then you could have an
2585 array of large drives with only a small amount used. In this
2586 situation, changing the "size" with "GROW" mode will allow the extra
2587 space to start being used. If the size is increased in this way, a
2588 "resync" process will start to make sure the new parts of the array
2591 Note that when an array changes size, any filesystem that may be
2592 stored in the array will not automatically grow or shrink to use or
2593 vacate the space. The
2594 filesystem will need to be explicitly told to use the extra space
2595 after growing, or to reduce its size
2597 to shrinking the array.
2599 Also the size of an array cannot be changed while it has an active
2600 bitmap. If an array has a bitmap, it must be removed before the size
2601 can be changed. Once the change is complete a new bitmap can be created.
2603 .SS RAID\-DEVICES CHANGES
2605 A RAID1 array can work with any number of devices from 1 upwards
2606 (though 1 is not very useful). There may be times which you want to
2607 increase or decrease the number of active devices. Note that this is
2608 different to hot-add or hot-remove which changes the number of
2611 When reducing the number of devices in a RAID1 array, the slots which
2612 are to be removed from the array must already be vacant. That is, the
2613 devices which were in those slots must be failed and removed.
2615 When the number of devices is increased, any hot spares that are
2616 present will be activated immediately.
2618 Changing the number of active devices in a RAID5 or RAID6 is much more
2619 effort. Every block in the array will need to be read and written
2620 back to a new location. From 2.6.17, the Linux Kernel is able to
2621 increase the number of devices in a RAID5 safely, including restarting
2622 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2623 increase or decrease the number of devices in a RAID5 or RAID6.
2625 From 2.6.35, the Linux Kernel is able to convert a RAID0 in to a RAID4
2628 uses this functionality and the ability to add
2629 devices to a RAID4 to allow devices to be added to a RAID0. When
2630 requested to do this,
2632 will convert the RAID0 to a RAID4, add the necessary disks and make
2633 the reshape happen, and then convert the RAID4 back to RAID0.
2635 When decreasing the number of devices, the size of the array will also
2636 decrease. If there was data in the array, it could get destroyed and
2637 this is not reversible, so you should firstly shrink the filesystem on
2638 the array to fit within the new size. To help prevent accidents,
2640 requires that the size of the array be decreased first with
2641 .BR "mdadm --grow --array-size" .
2642 This is a reversible change which simply makes the end of the array
2643 inaccessible. The integrity of any data can then be checked before
2644 the non-reversible reduction in the number of devices is request.
2646 When relocating the first few stripes on a RAID5 or RAID6, it is not
2647 possible to keep the data on disk completely consistent and
2648 crash-proof. To provide the required safety, mdadm disables writes to
2649 the array while this "critical section" is reshaped, and takes a
2650 backup of the data that is in that section. For grows, this backup may be
2651 stored in any spare devices that the array has, however it can also be
2652 stored in a separate file specified with the
2654 option, and is required to be specified for shrinks, RAID level
2655 changes and layout changes. If this option is used, and the system
2656 does crash during the critical period, the same file must be passed to
2658 to restore the backup and reassemble the array. When shrinking rather
2659 than growing the array, the reshape is done from the end towards the
2660 beginning, so the "critical section" is at the end of the reshape.
2664 Changing the RAID level of any array happens instantaneously. However
2665 in the RAID5 to RAID6 case this requires a non-standard layout of the
2666 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2667 required before the change can be accomplished. So while the level
2668 change is instant, the accompanying layout change can take quite a
2671 is required. If the array is not simultaneously being grown or
2672 shrunk, so that the array size will remain the same - for example,
2673 reshaping a 3-drive RAID5 into a 4-drive RAID6 - the backup file will
2674 be used not just for a "cricital section" but throughout the reshape
2675 operation, as described below under LAYOUT CHANGES.
2677 .SS CHUNK-SIZE AND LAYOUT CHANGES
2679 Changing the chunk-size of layout without also changing the number of
2680 devices as the same time will involve re-writing all blocks in-place.
2681 To ensure against data loss in the case of a crash, a
2683 must be provided for these changes. Small sections of the array will
2684 be copied to the backup file while they are being rearranged. This
2685 means that all the data is copied twice, once to the backup and once
2686 to the new layout on the array, so this type of reshape will go very
2689 If the reshape is interrupted for any reason, this backup file must be
2691 .B "mdadm --assemble"
2692 so the array can be reassembled. Consequently the file cannot be
2693 stored on the device being reshaped.
2698 A write-intent bitmap can be added to, or removed from, an active
2699 array. Either internal bitmaps, or bitmaps stored in a separate file,
2700 can be added. Note that if you add a bitmap stored in a file which is
2701 in a filesystem that is on the RAID array being affected, the system
2702 will deadlock. The bitmap must be on a separate filesystem.
2704 .SH INCREMENTAL MODE
2708 .B mdadm \-\-incremental
2712 .RI [ optional-aliases-for-device ]
2715 .B mdadm \-\-incremental \-\-fail
2719 .B mdadm \-\-incremental \-\-rebuild\-map
2722 .B mdadm \-\-incremental \-\-run \-\-scan
2725 This mode is designed to be used in conjunction with a device
2726 discovery system. As devices are found in a system, they can be
2728 .B "mdadm \-\-incremental"
2729 to be conditionally added to an appropriate array.
2731 Conversely, it can also be used with the
2733 flag to do just the opposite and find whatever array a particular device
2734 is part of and remove the device from that array.
2736 If the device passed is a
2738 device created by a previous call to
2740 then rather than trying to add that device to an array, all the arrays
2741 described by the metadata of the container will be started.
2744 performs a number of tests to determine if the device is part of an
2745 array, and which array it should be part of. If an appropriate array
2746 is found, or can be created,
2748 adds the device to the array and conditionally starts the array.
2752 will normally only add devices to an array which were previously working
2753 (active or spare) parts of that array. The support for automatic
2754 inclusion of a new drive as a spare in some array requires
2755 a configuration through POLICY in config file.
2759 makes are as follow:
2761 Is the device permitted by
2763 That is, is it listed in a
2765 line in that file. If
2767 is absent then the default it to allow any device. Similarly if
2769 contains the special word
2771 then any device is allowed. Otherwise the device name given to
2773 or one of the aliases given, or an alias found in the filesystem,
2774 must match one of the names or patterns in a
2778 This is the only context where the aliases are used. They are
2779 usually provided by a
2785 Does the device have a valid md superblock? If a specific metadata
2786 version is requested with
2790 then only that style of metadata is accepted, otherwise
2792 finds any known version of metadata. If no
2794 metadata is found, the device may be still added to an array
2795 as a spare if POLICY allows.
2799 Does the metadata match an expected array?
2800 The metadata can match in two ways. Either there is an array listed
2803 which identifies the array (either by UUID, by name, by device list,
2804 or by minor-number), or the array was created with a
2810 or on the command line.
2813 is not able to positively identify the array as belonging to the
2814 current host, the device will be rejected.
2819 keeps a list of arrays that it has partially assembled in
2821 If no array exists which matches
2822 the metadata on the new device,
2824 must choose a device name and unit number. It does this based on any
2827 or any name information stored in the metadata. If this name
2828 suggests a unit number, that number will be used, otherwise a free
2829 unit number will be chosen. Normally
2831 will prefer to create a partitionable array, however if the
2835 suggests that a non-partitionable array is preferred, that will be
2838 If the array is not found in the config file and its metadata does not
2839 identify it as belonging to the "homehost", then
2841 will choose a name for the array which is certain not to conflict with
2842 any array which does belong to this host. It does this be adding an
2843 underscore and a small number to the name preferred by the metadata.
2845 Once an appropriate array is found or created and the device is added,
2847 must decide if the array is ready to be started. It will
2848 normally compare the number of available (non-spare) devices to the
2849 number of devices that the metadata suggests need to be active. If
2850 there are at least that many, the array will be started. This means
2851 that if any devices are missing the array will not be restarted.
2857 in which case the array will be run as soon as there are enough
2858 devices present for the data to be accessible. For a RAID1, that
2859 means one device will start the array. For a clean RAID5, the array
2860 will be started as soon as all but one drive is present.
2862 Note that neither of these approaches is really ideal. If it can
2863 be known that all device discovery has completed, then
2867 can be run which will try to start all arrays that are being
2868 incrementally assembled. They are started in "read-auto" mode in
2869 which they are read-only until the first write request. This means
2870 that no metadata updates are made and no attempt at resync or recovery
2871 happens. Further devices that are found before the first write can
2872 still be added safely.
2875 This section describes environment variables that affect how mdadm
2880 Setting this value to 1 will prevent mdadm from automatically launching
2881 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2887 does not create any device nodes in /dev, but leaves that task to
2891 appears not to be configured, or if this environment variable is set
2894 will create and devices that are needed.
2897 .B MDADM_NO_SYSTEMCTL
2902 is in use it will normally request
2904 to start various background tasks (particularly
2906 rather than forking and running them in the background. This can be
2907 suppressed by setting
2908 .BR MDADM_NO_SYSTEMCTL=1 .
2912 A key value of IMSM metadata is that it allows interoperability with
2913 boot ROMs on Intel platforms, and with other major operating systems.
2916 will only allow an IMSM array to be created or modified if detects
2917 that it is running on an Intel platform which supports IMSM, and
2918 supports the particular configuration of IMSM that is being requested
2919 (some functionality requires newer OROM support).
2921 These checks can be suppressed by setting IMSM_NO_PLATFORM=1 in the
2922 environment. This can be useful for testing or for disaster
2923 recovery. You should be aware that interoperability may be
2924 compromised by setting this value.
2927 .B MDADM_GROW_ALLOW_OLD
2928 If an array is stopped while it is performing a reshape and that
2929 reshape was making use of a backup file, then when the array is
2932 will sometimes complain that the backup file is too old. If this
2933 happens and you are certain it is the right backup file, you can
2934 over-ride this check by setting
2935 .B MDADM_GROW_ALLOW_OLD=1
2940 Any string given in this variable is added to the start of the
2942 line in the config file, or treated as the whole
2944 line if none is given. It can be used to disable certain metadata
2947 is called from a boot script. For example
2949 .B " export MDADM_CONF_AUTO='-ddf -imsm'
2953 does not automatically assemble any DDF or
2954 IMSM arrays that are found. This can be useful on systems configured
2955 to manage such arrays with
2961 .B " mdadm \-\-query /dev/name-of-device"
2963 This will find out if a given device is a RAID array, or is part of
2964 one, and will provide brief information about the device.
2966 .B " mdadm \-\-assemble \-\-scan"
2968 This will assemble and start all arrays listed in the standard config
2969 file. This command will typically go in a system startup file.
2971 .B " mdadm \-\-stop \-\-scan"
2973 This will shut down all arrays that can be shut down (i.e. are not
2974 currently in use). This will typically go in a system shutdown script.
2976 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2978 If (and only if) there is an Email address or program given in the
2979 standard config file, then
2980 monitor the status of all arrays listed in that file by
2981 polling them ever 2 minutes.
2983 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2985 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2988 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2990 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2992 This will create a prototype config file that describes currently
2993 active arrays that are known to be made from partitions of IDE or SCSI drives.
2994 This file should be reviewed before being used as it may
2995 contain unwanted detail.
2997 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2999 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
3001 This will find arrays which could be assembled from existing IDE and
3002 SCSI whole drives (not partitions), and store the information in the
3003 format of a config file.
3004 This file is very likely to contain unwanted detail, particularly
3007 entries. It should be reviewed and edited before being used as an
3010 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
3012 .B " mdadm \-Ebsc partitions"
3014 Create a list of devices by reading
3015 .BR /proc/partitions ,
3016 scan these for RAID superblocks, and printout a brief listing of all
3019 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
3021 Scan all partitions and devices listed in
3022 .BR /proc/partitions
3025 out of all such devices with a RAID superblock with a minor number of 0.
3027 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /run/mdadm/mon.pid"
3029 If config file contains a mail address or alert program, run mdadm in
3030 the background in monitor mode monitoring all md devices. Also write
3031 pid of mdadm daemon to
3032 .BR /run/mdadm/mon.pid .
3034 .B " mdadm \-Iq /dev/somedevice"
3036 Try to incorporate newly discovered device into some array as
3039 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
3041 Rebuild the array map from any current arrays, and then start any that
3044 .B " mdadm /dev/md4 --fail detached --remove detached"
3046 Any devices which are components of /dev/md4 will be marked as faulty
3047 and then remove from the array.
3049 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4"
3053 which is currently a RAID5 array will be converted to RAID6. There
3054 should normally already be a spare drive attached to the array as a
3055 RAID6 needs one more drive than a matching RAID5.
3057 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
3059 Create a DDF array over 6 devices.
3061 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
3063 Create a RAID5 array over any 3 devices in the given DDF set. Use
3064 only 30 gigabytes of each device.
3066 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
3068 Assemble a pre-exist ddf array.
3070 .B " mdadm -I /dev/md/ddf1"
3072 Assemble all arrays contained in the ddf array, assigning names as
3075 .B " mdadm \-\-create \-\-help"
3077 Provide help about the Create mode.
3079 .B " mdadm \-\-config \-\-help"
3081 Provide help about the format of the config file.
3083 .B " mdadm \-\-help"
3085 Provide general help.
3095 lists all active md devices with information about them.
3097 uses this to find arrays when
3099 is given in Misc mode, and to monitor array reconstruction
3104 The config file lists which devices may be scanned to see if
3105 they contain MD super block, and gives identifying information
3106 (e.g. UUID) about known MD arrays. See
3110 .SS /etc/mdadm.conf.d
3112 A directory containing configuration files which are read in lexical
3118 mode is used, this file gets a list of arrays currently being created.
3123 understand two sorts of names for array devices.
3125 The first is the so-called 'standard' format name, which matches the
3126 names used by the kernel and which appear in
3129 The second sort can be freely chosen, but must reside in
3131 When giving a device name to
3133 to create or assemble an array, either full path name such as
3137 can be given, or just the suffix of the second sort of name, such as
3143 chooses device names during auto-assembly or incremental assembly, it
3144 will sometimes add a small sequence number to the end of the name to
3145 avoid conflicted between multiple arrays that have the same name. If
3147 can reasonably determine that the array really is meant for this host,
3148 either by a hostname in the metadata, or by the presence of the array
3151 then it will leave off the suffix if possible.
3152 Also if the homehost is specified as
3155 will only use a suffix if a different array of the same name already
3156 exists or is listed in the config file.
3158 The standard names for non-partitioned arrays (the only sort of md
3159 array available in 2.4 and earlier) are of the form
3163 where NN is a number.
3164 The standard names for partitionable arrays (as available from 2.6
3165 onwards) are of the form:
3169 Partition numbers should be indicated by adding "pMM" to these, thus "/dev/md/d1p2".
3171 From kernel version 2.6.28 the "non-partitioned array" can actually
3172 be partitioned. So the "md_d\fBNN\fP"
3173 names are no longer needed, and
3174 partitions such as "/dev/md\fBNN\fPp\fBXX\fP"
3177 From kernel version 2.6.29 standard names can be non-numeric following
3184 is any string. These names are supported by
3186 since version 3.3 provided they are enabled in
3191 was previously known as
3195 For further information on mdadm usage, MD and the various levels of
3198 .B http://raid.wiki.kernel.org/
3200 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
3202 The latest version of
3204 should always be available from
3206 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/