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.
8 .TH MDADM 8 "" v3.0-rc1
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 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 superblocks.
107 .\"in several step create-add-add-run or it can all happen with one command.
110 .B "Follow or Monitor"
111 Monitor one or more md devices and act on any state changes. This is
112 only meaningful for raid1, 4, 5, 6, 10 or multipath arrays, as
113 only these have interesting state. raid0 or linear never have
114 missing, spare, or failed drives, so there is nothing to monitor.
118 Grow (or shrink) an array, or otherwise reshape it in some way.
119 Currently supported growth options including changing the active size
120 of component devices and changing the number of active devices in RAID
121 levels 1/4/5/6, as well as adding or removing a write-intent bitmap.
124 .B "Incremental Assembly"
125 Add a single device to an appropriate array. If the addition of the
126 device makes the array runnable, the array will be started.
127 This provides a convenient interface to a
129 system. As each device is detected,
131 has a chance to include it in some array as appropriate.
137 in this mode, then any arrays within that container will be assembled
142 This is for doing things to specific components of an array such as
143 adding new spares and removing faulty devices.
147 This is an 'everything else' mode that supports operations on active
148 arrays, operations on component devices such as erasing old superblocks, and
149 information gathering operations.
150 .\"This mode allows operations on independent devices such as examine MD
151 .\"superblocks, erasing old superblocks and stopping active arrays.
155 This mode does not act on a specific device or array, but rather it
156 requests the Linux Kernel to activate any auto-detected arrays.
159 .SH Options for selecting a mode are:
162 .BR \-A ", " \-\-assemble
163 Assemble a pre-existing array.
166 .BR \-B ", " \-\-build
167 Build a legacy array without superblocks.
170 .BR \-C ", " \-\-create
174 .BR \-F ", " \-\-follow ", " \-\-monitor
180 .BR \-G ", " \-\-grow
181 Change the size or shape of an active array.
184 .BR \-I ", " \-\-incremental
185 Add a single device into an appropriate array, and possibly start the array.
189 Request that the kernel starts any auto-detected arrays. This can only
192 is compiled into the kernel \(em not if it is a module.
193 Arrays can be auto-detected by the kernel if all the components are in
194 primary MS-DOS partitions with partition type
196 In-kernel autodetect is not recommended for new installations. Using
198 to detect and assemble arrays \(em possibly in an
200 \(em is substantially more flexible and should be preferred.
203 If a device is given before any options, or if the first option is
208 then the MANAGE mode is assume.
209 Anything other than these will cause the
213 .SH Options that are not mode-specific are:
216 .BR \-h ", " \-\-help
217 Display general help message or, after one of the above options, a
218 mode-specific help message.
222 Display more detailed help about command line parsing and some commonly
226 .BR \-V ", " \-\-version
227 Print version information for mdadm.
230 .BR \-v ", " \-\-verbose
231 Be more verbose about what is happening. This can be used twice to be
233 The extra verbosity currently only affects
234 .B \-\-detail \-\-scan
236 .BR "\-\-examine \-\-scan" .
239 .BR \-q ", " \-\-quiet
240 Avoid printing purely informative messages. With this,
242 will be silent unless there is something really important to report.
245 .BR \-b ", " \-\-brief
246 Be less verbose. This is used with
254 gives an intermediate level of verbosity.
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 .B \-e ", " \-\-metadata=
305 Declare the style of superblock (raid metadata) to be used. The
308 and to guess for other operations.
309 The default can be overridden by setting the
318 .IP "0, 0.90, default"
319 Use the original 0.90 format superblock. This format limits arrays to
320 28 component devices and limits component devices of levels 1 and
321 greater to 2 terabytes.
322 .IP "1, 1.0, 1.1, 1.2"
323 Use the new version-1 format superblock. This has few restrictions.
324 The different sub-versions store the superblock at different locations
325 on the device, either at the end (for 1.0), at the start (for 1.1) or
326 4K from the start (for 1.2).
328 Use the "Industry Standard" DDF (Disk Data Format) format. When
329 creating a DDF array a
331 will be created, and normal arrays can be created in that container.
333 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
335 which is managed in a similar manner to DDF, and is supported by an
336 option-rom on some platforms:
338 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
344 This will override any
346 setting in the config file and provides the identity of the host which
347 should be considered the home for any arrays.
349 When creating an array, the
351 will be recorded in the superblock. For version-1 superblocks, it will
352 be prefixed to the array name. For version-0.90 superblocks, part of
353 the SHA1 hash of the hostname will be stored in the later half of the
356 When reporting information about an array, any array which is tagged
357 for the given homehost will be reported as such.
359 When using Auto-Assemble, only arrays tagged for the given homehost
360 will be allowed to use 'local' names (i.e. not ending in '_' followed
363 .SH For create, build, or grow:
366 .BR \-n ", " \-\-raid\-devices=
367 Specify the number of active devices in the array. This, plus the
368 number of spare devices (see below) must equal the number of
370 (including "\fBmissing\fP" devices)
371 that are listed on the command line for
373 Setting a value of 1 is probably
374 a mistake and so requires that
376 be specified first. A value of 1 will then be allowed for linear,
377 multipath, raid0 and raid1. It is never allowed for raid4 or raid5.
379 This number can only be changed using
381 for RAID1, RAID5 and RAID6 arrays, and only on kernels which provide
385 .BR \-x ", " \-\-spare\-devices=
386 Specify the number of spare (eXtra) devices in the initial array.
387 Spares can also be added
388 and removed later. The number of component devices listed
389 on the command line must equal the number of raid devices plus the
390 number of spare devices.
394 .BR \-z ", " \-\-size=
395 Amount (in Kibibytes) of space to use from each drive in RAID level 1/4/5/6.
396 This must be a multiple of the chunk size, and must leave about 128Kb
397 of space at the end of the drive for the RAID superblock.
398 If this is not specified
399 (as it normally is not) the smallest drive (or partition) sets the
400 size, though if there is a variance among the drives of greater than 1%, a warning is
403 This value can be set with
405 for RAID level 1/4/5/6. If the array was created with a size smaller
406 than the currently active drives, the extra space can be accessed
409 The size can be given as
411 which means to choose the largest size that fits on all current drives.
413 This value can not be used with
415 metadata such as DDF and IMSM.
418 .BR \-c ", " \-\-chunk=
419 Specify chunk size of kibibytes. The default is 64.
423 Specify rounding factor for linear array (==chunk size)
426 .BR \-l ", " \-\-level=
427 Set raid level. When used with
429 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
430 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
431 Obviously some of these are synonymous.
435 metadata type is requested, only the
437 level is permitted, and it does not need to be explicitly given.
441 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
443 Not yet supported with
447 .BR \-p ", " \-\-layout=
448 This option configures the fine details of data layout for raid5,
449 and raid10 arrays, and controls the failure modes for
452 The layout of the raid5 parity block can be one of
453 .BR left\-asymmetric ,
454 .BR left\-symmetric ,
455 .BR right\-asymmetric ,
456 .BR right\-symmetric ,
457 .BR la ", " ra ", " ls ", " rs .
459 .BR left\-symmetric .
461 When setting the failure mode for level
464 .BR write\-transient ", " wt ,
465 .BR read\-transient ", " rt ,
466 .BR write\-persistent ", " wp ,
467 .BR read\-persistent ", " rp ,
469 .BR read\-fixable ", " rf ,
470 .BR clear ", " flush ", " none .
472 Each failure mode can be followed by a number, which is used as a period
473 between fault generation. Without a number, the fault is generated
474 once on the first relevant request. With a number, the fault will be
475 generated after that many requests, and will continue to be generated
476 every time the period elapses.
478 Multiple failure modes can be current simultaneously by using the
480 option to set subsequent failure modes.
482 "clear" or "none" will remove any pending or periodic failure modes,
483 and "flush" will clear any persistent faults.
485 To set the parity with
487 the level of the array ("faulty")
488 must be specified before the fault mode is specified.
490 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
491 by a small number. The default is 'n2'. The supported options are:
494 signals 'near' copies. Multiple copies of one data block are at
495 similar offsets in different devices.
498 signals 'offset' copies. Rather than the chunks being duplicated
499 within a stripe, whole stripes are duplicated but are rotated by one
500 device so duplicate blocks are on different devices. Thus subsequent
501 copies of a block are in the next drive, and are one chunk further
506 (multiple copies have very different offsets).
507 See md(4) for more detail about 'near' and 'far'.
509 The number is the number of copies of each datablock. 2 is normal, 3
510 can be useful. This number can be at most equal to the number of
511 devices in the array. It does not need to divide evenly into that
512 number (e.g. it is perfectly legal to have an 'n2' layout for an array
513 with an odd number of devices).
519 (thus explaining the p of
523 .BR \-b ", " \-\-bitmap=
524 Specify a file to store a write-intent bitmap in. The file should not
527 is also given. The same file should be provided
528 when assembling the array. If the word
530 is given, then the bitmap is stored with the metadata on the array,
531 and so is replicated on all devices. If the word
535 mode, then any bitmap that is present is removed.
537 To help catch typing errors, the filename must contain at least one
538 slash ('/') if it is a real file (not 'internal' or 'none').
540 Note: external bitmaps are only known to work on ext2 and ext3.
541 Storing bitmap files on other filesystems may result in serious problems.
544 .BR \-\-bitmap\-chunk=
545 Set the chunksize of the bitmap. Each bit corresponds to that many
546 Kilobytes of storage.
547 When using a file based bitmap, the default is to use the smallest
548 size that is at-least 4 and requires no more than 2^21 chunks.
551 bitmap, the chunksize is automatically determined to make best use of
556 .BR \-W ", " \-\-write\-mostly
557 subsequent devices lists in a
562 command will be flagged as 'write-mostly'. This is valid for RAID1
563 only and means that the 'md' driver will avoid reading from these
564 devices if at all possible. This can be useful if mirroring over a
568 .BR \-\-write\-behind=
569 Specify that write-behind mode should be enabled (valid for RAID1
570 only). If an argument is specified, it will set the maximum number
571 of outstanding writes allowed. The default value is 256.
572 A write-intent bitmap is required in order to use write-behind
573 mode, and write-behind is only attempted on drives marked as
577 .BR \-\-assume\-clean
580 that the array pre-existed and is known to be clean. It can be useful
581 when trying to recover from a major failure as you can be sure that no
582 data will be affected unless you actually write to the array. It can
583 also be used when creating a RAID1 or RAID10 if you want to avoid the
584 initial resync, however this practice \(em while normally safe \(em is not
585 recommended. Use this only if you really know what you are doing.
588 .BR \-\-backup\-file=
591 is used to increase the number of
592 raid-devices in a RAID5 if there are no spare devices available.
593 See the section below on RAID_DEVICE CHANGES. The file should be
594 stored on a separate device, not on the raid array being reshaped.
597 .BR \-\-array-size= ", " \-Z
598 Set the size of the array which is seen by users of the device such as
599 filesystems. This can be less that the real size, but never greater.
600 The size set this way does not persist across restarts of the array.
602 This is most useful when reducing the number of devices in a RAID5 or
603 RAID6. Such arrays require the array-size to be reduced before a
604 reshape can be performed that reduces the real size.
608 restores the apparent size of the array to be whatever the real
609 amount of available space is.
612 .BR \-N ", " \-\-name=
615 for the array. This is currently only effective when creating an
616 array with a version-1 superblock. The name is a simple textual
617 string that can be used to identify array components when assembling.
623 run the array, even if some of the components
624 appear to be active in another array or filesystem. Normally
626 will ask for confirmation before including such components in an
627 array. This option causes that question to be suppressed.
630 .BR \-f ", " \-\-force
633 accept the geometry and layout specified without question. Normally
635 will not allow creation of an array with only one device, and will try
636 to create a raid5 array with one missing drive (as this makes the
637 initial resync work faster). With
640 will not try to be so clever.
643 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
644 Instruct mdadm how to create the device file if needed, possibly allocating
645 an unused minor number. "md" causes a non-partitionable array
646 to be used (though since Linux 2.6.28, these array devices are in fact
647 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
648 later) to be used. "yes" requires the named md device to have
649 a 'standard' format, and the type and minor number will be determined
650 from this. With mdadm 3.0, device creation is normally left up to
652 so this option is unlikely to be needed.
653 See DEVICE NAMES below.
655 The argument can also come immediately after
660 is not given on the command line or in the config file, then
666 is also given, then any
668 entries in the config file will override the
670 instruction given on the command line.
672 For partitionable arrays,
674 will create the device file for the whole array and for the first 4
675 partitions. A different number of partitions can be specified at the
676 end of this option (e.g.
678 If the device name ends with a digit, the partition names add a 'p',
679 and a number, e.g. "/dev/md/home1p3". If there is no
680 trailing digit, then the partition names just have a number added,
681 e.g. "/dev/md/scratch3".
683 If the md device name is in a 'standard' format as described in DEVICE
684 NAMES, then it will be created, if necessary, with the appropriate
685 number based on that name. If the device name is not in one of these
686 formats, then a unused minor number will be allocated. The minor
687 number will be considered unused if there is no active array for that
688 number, and there is no entry in /dev for that number and with a
689 non-standard name. Name that are not in 'standard' format are only
690 allowed in "/dev/md/".
693 \".BR \-\-symlink = no
698 \"to create devices in
700 \"it will also create symlinks from
702 \"with names starting with
708 \"to suppress this, or
710 \"to enforce this even if it is suppressing
717 .BR \-u ", " \-\-uuid=
718 uuid of array to assemble. Devices which don't have this uuid are
722 .BR \-m ", " \-\-super\-minor=
723 Minor number of device that array was created for. Devices which
724 don't have this minor number are excluded. If you create an array as
725 /dev/md1, then all superblocks will contain the minor number 1, even if
726 the array is later assembled as /dev/md2.
728 Giving the literal word "dev" for
732 to use the minor number of the md device that is being assembled.
735 .B \-\-super\-minor=dev
736 will look for super blocks with a minor number of 0.
739 .BR \-N ", " \-\-name=
740 Specify the name of the array to assemble. This must be the name
741 that was specified when creating the array. It must either match
742 the name stored in the superblock exactly, or it must match
745 prefixed to the start of the given name.
748 .BR \-f ", " \-\-force
749 Assemble the array even if some superblocks appear out-of-date
753 Attempt to start the array even if fewer drives were given than were
754 present last time the array was active. Normally if not all the
755 expected drives are found and
757 is not used, then the array will be assembled but not started.
760 an attempt will be made to start it anyway.
764 This is the reverse of
766 in that it inhibits the startup of array unless all expected drives
767 are present. This is only needed with
769 and can be used if the physical connections to devices are
770 not as reliable as you would like.
773 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
774 See this option under Create and Build options.
777 .BR \-b ", " \-\-bitmap=
778 Specify the bitmap file that was given when the array was created. If
781 bitmap, there is no need to specify this when assembling the array.
784 .BR \-\-backup\-file=
787 was used to grow the number of raid-devices in a RAID5, and the system
788 crashed during the critical section, then the same
792 to allow possibly corrupted data to be restored.
795 .BR \-U ", " \-\-update=
796 Update the superblock on each device while assembling the array. The
797 argument given to this flag can be one of
811 option will adjust the superblock of an array what was created on a Sparc
812 machine running a patched 2.2 Linux kernel. This kernel got the
813 alignment of part of the superblock wrong. You can use the
814 .B "\-\-examine \-\-sparc2.2"
817 to see what effect this would have.
821 option will update the
823 field on each superblock to match the minor number of the array being
825 This can be useful if
827 reports a different "Preferred Minor" to
829 In some cases this update will be performed automatically
830 by the kernel driver. In particular the update happens automatically
831 at the first write to an array with redundancy (RAID level 1 or
832 greater) on a 2.6 (or later) kernel.
836 option will change the uuid of the array. If a UUID is given with the
838 option that UUID will be used as a new UUID and will
840 be used to help identify the devices in the array.
843 is given, a random UUID is chosen.
847 option will change the
849 of the array as stored in the superblock. This is only supported for
850 version-1 superblocks.
854 option will change the
856 as recorded in the superblock. For version-0 superblocks, this is the
857 same as updating the UUID.
858 For version-1 superblocks, this involves updating the name.
862 option will cause the array to be marked
864 meaning that any redundancy in the array (e.g. parity for raid5,
865 copies for raid1) may be incorrect. This will cause the raid system
866 to perform a "resync" pass to make sure that all redundant information
871 option allows arrays to be moved between machines with different
873 When assembling such an array for the first time after a move, giving
874 .B "\-\-update=byteorder"
877 to expect superblocks to have their byteorder reversed, and will
878 correct that order before assembling the array. This is only valid
879 with original (Version 0.90) superblocks.
883 option will correct the summaries in the superblock. That is the
884 counts of total, working, active, failed, and spare devices.
888 will rarely be of use. It applies to version 1.1 and 1.2 metadata
889 only (where the metadata is at the start of the device) and is only
890 useful when the component device has changed size (typically become
891 larger). The version 1 metadata records the amount of the device that
892 can be used to store data, so if a device in a version 1.1 or 1.2
893 array becomes larger, the metadata will still be visible, but the
894 extra space will not. In this case it might be useful to assemble the
896 .BR \-\-update=devicesize .
899 to determine the maximum usable amount of space on each device and
900 update the relevant field in the metadata.
903 .B \-\-auto\-update\-homehost
904 This flag is only meaningful with auto-assembly (see discussion below).
905 In that situation, if no suitable arrays are found for this homehost,
907 will rescan for any arrays at all and will assemble them and update the
908 homehost to match the current host.
914 hot-add listed devices.
918 re-add a device that was recently removed from an array.
921 .BR \-r ", " \-\-remove
922 remove listed devices. They must not be active. i.e. they should
923 be failed or spare devices. As well as the name of a device file
932 The first causes all failed device to be removed. The second causes
933 any device which is no longer connected to the system (i.e an 'open'
936 to be removed. This will only succeed for devices that are spares or
937 have already been marked as failed.
940 .BR \-f ", " \-\-fail
941 mark listed devices as faulty.
942 As well as the name of a device file, the word
944 can be given. This will cause any device that has been detached from
945 the system to be marked as failed. It can then be removed.
953 .BR \-\-write\-mostly
954 Subsequent devices that are added or re-added will have the 'write-mostly'
955 flag set. This is only valid for RAID! and means that the 'md' driver
956 will avoid reading from these devices if possible.
959 Subsequent devices that are added or re-added will have the 'write-mostly'
964 Each of these options require that the first device listed is the array
965 to be acted upon, and the remainder are component devices to be added,
966 removed, or marked as faulty. Several different operations can be
967 specified for different devices, e.g.
969 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
971 Each operation applies to all devices listed until the next
974 If an array is using a write-intent bitmap, then devices which have
975 been removed can be re-added in a way that avoids a full
976 reconstruction but instead just updates the blocks that have changed
977 since the device was removed. For arrays with persistent metadata
978 (superblocks) this is done automatically. For arrays created with
980 mdadm needs to be told that this device we removed recently with
983 Devices can only be removed from an array if they are not in active
984 use, i.e. that must be spares or failed devices. To remove an active
985 device, it must first be marked as
991 .BR \-Q ", " \-\-query
992 Examine a device to see
993 (1) if it is an md device and (2) if it is a component of an md
995 Information about what is discovered is presented.
998 .BR \-D ", " \-\-detail
999 Print detail of one or more md devices.
1002 .BR \-\-detail\-platform
1003 Print detail of the platform's raid capabilities (firmware / hardware
1004 topology) for a given metadata format.
1007 .BR \-Y ", " \-\-export
1012 output will be formatted as
1014 pairs for easy import into the environment.
1017 .BR \-E ", " \-\-examine
1018 Print content of md superblock on device(s).
1021 If an array was created on a 2.2 Linux kernel patched with RAID
1022 support, the superblock will have been created incorrectly, or at
1023 least incompatibly with 2.4 and later kernels. Using the
1027 will fix the superblock before displaying it. If this appears to do
1028 the right thing, then the array can be successfully assembled using
1029 .BR "\-\-assemble \-\-update=sparc2.2" .
1032 .BR \-X ", " \-\-examine\-bitmap
1033 Report information about a bitmap file.
1034 The argument is either an external bitmap file or an array component
1035 in case of an internal bitmap.
1038 .BR \-R ", " \-\-run
1039 start a partially built array.
1042 .BR \-S ", " \-\-stop
1043 deactivate array, releasing all resources.
1046 .BR \-o ", " \-\-readonly
1047 mark array as readonly.
1050 .BR \-w ", " \-\-readwrite
1051 mark array as readwrite.
1054 .B \-\-zero\-superblock
1055 If the device contains a valid md superblock, the block is
1056 overwritten with zeros. With
1058 the block where the superblock would be is overwritten even if it
1059 doesn't appear to be valid.
1062 .BR \-t ", " \-\-test
1067 is set to reflect the status of the device.
1070 .BR \-W ", " \-\-wait
1071 For each md device given, wait for any resync, recovery, or reshape
1072 activity to finish before returning.
1074 will return with success if it actually waited for every device
1075 listed, otherwise it will return failure.
1079 For each md device given, or each device in /proc/mdstat if
1081 is given, arrange for the array to be marked clean as soon as possible.
1082 Also, quiesce resync so that the monitor for external metadata arrays
1083 (mdmon) has an opportunity to checkpoint the resync position.
1085 will return with success if the array uses external metadata and we
1086 successfully waited. For native arrays this returns immediately as the
1087 kernel handles both dirty-clean transitions and resync checkpointing in
1088 the kernel at shutdown. No action is taken if safe-mode handling is
1091 .SH For Incremental Assembly mode:
1093 .BR \-\-rebuild\-map ", " \-r
1094 Rebuild the map file
1095 .RB ( /var/run/mdadm/map )
1098 uses to help track which arrays are currently being assembled.
1101 .BR \-\-run ", " \-R
1102 Run any array assembled as soon as a minimal number of devices are
1103 available, rather than waiting until all expected devices are present.
1107 This allows the hot-plug system to prevent arrays from running when it knows
1108 that more disks may arrive later in the discovery process.
1111 .BR \-\-scan ", " \-s
1112 Only meaningful with
1116 file for arrays that are being incrementally assembled and will try to
1117 start any that are not already started. If any such array is listed
1120 as requiring an external bitmap, that bitmap will be attached first.
1122 .SH For Monitor mode:
1124 .BR \-m ", " \-\-mail
1125 Give a mail address to send alerts to.
1128 .BR \-p ", " \-\-program ", " \-\-alert
1129 Give a program to be run whenever an event is detected.
1132 .BR \-y ", " \-\-syslog
1133 Cause all events to be reported through 'syslog'. The messages have
1134 facility of 'daemon' and varying priorities.
1137 .BR \-d ", " \-\-delay
1138 Give a delay in seconds.
1140 polls the md arrays and then waits this many seconds before polling
1141 again. The default is 60 seconds.
1144 .BR \-f ", " \-\-daemonise
1147 to run as a background daemon if it decides to monitor anything. This
1148 causes it to fork and run in the child, and to disconnect form the
1149 terminal. The process id of the child is written to stdout.
1152 which will only continue monitoring if a mail address or alert program
1153 is found in the config file.
1156 .BR \-i ", " \-\-pid\-file
1159 is running in daemon mode, write the pid of the daemon process to
1160 the specified file, instead of printing it on standard output.
1163 .BR \-1 ", " \-\-oneshot
1164 Check arrays only once. This will generate
1166 events and more significantly
1172 .B " mdadm \-\-monitor \-\-scan \-1"
1174 from a cron script will ensure regular notification of any degraded arrays.
1177 .BR \-t ", " \-\-test
1180 alert for every array found at startup. This alert gets mailed and
1181 passed to the alert program. This can be used for testing that alert
1182 message do get through successfully.
1188 .B mdadm \-\-assemble
1189 .I md-device options-and-component-devices...
1192 .B mdadm \-\-assemble \-\-scan
1193 .I md-devices-and-options...
1196 .B mdadm \-\-assemble \-\-scan
1200 This usage assembles one or more raid arrays from pre-existing components.
1201 For each array, mdadm needs to know the md device, the identity of the
1202 array, and a number of component-devices. These can be found in a number of ways.
1204 In the first usage example (without the
1206 the first device given is the md device.
1207 In the second usage example, all devices listed are treated as md
1208 devices and assembly is attempted.
1209 In the third (where no devices are listed) all md devices that are
1210 listed in the configuration file are assembled. Then any arrays that
1211 can be found on unused devices will also be assembled.
1213 If precisely one device is listed, but
1219 was given and identity information is extracted from the configuration file.
1221 The identity can be given with the
1225 option, will be taken from the md-device record in the config file, or
1226 will be taken from the super block of the first component-device
1227 listed on the command line.
1229 Devices can be given on the
1231 command line or in the config file. Only devices which have an md
1232 superblock which contains the right identity will be considered for
1235 The config file is only used if explicitly named with
1237 or requested with (a possibly implicit)
1242 .B /etc/mdadm/mdadm.conf
1247 is not given, then the config file will only be used to find the
1248 identity of md arrays.
1250 Normally the array will be started after it is assembled. However if
1252 is not given and insufficient drives were listed to start a complete
1253 (non-degraded) array, then the array is not started (to guard against
1254 usage errors). To insist that the array be started in this case (as
1255 may work for RAID1, 4, 5, 6, or 10), give the
1259 If the md device does not exist, then it will be created providing the
1260 intent is clear. i.e. the name must be in a standard form, or the
1262 option must be given to clarify how and whether the device should be
1264 This can be useful for handling partitioned devices (which don't have
1265 a stable device number \(em it can change after a reboot) and when using
1266 "udev" to manage your
1268 tree (udev cannot handle md devices because of the unusual device
1269 initialisation conventions).
1271 If the option to "auto" is "mdp" or "part" or (on the command line
1272 only) "p", then mdadm will create a partitionable array, using the
1273 first free one that is not in use and does not already have an entry
1274 in /dev (apart from numeric /dev/md* entries).
1276 If the option to "auto" is "yes" or "md" or (on the command line)
1277 nothing, then mdadm will create a traditional, non-partitionable md
1280 It is expected that the "auto" functionality will be used to create
1281 device entries with meaningful names such as "/dev/md/home" or
1282 "/dev/md/root", rather than names based on the numerical array number.
1284 When using option "auto" to create a partitionable array, the device
1285 files for the first 4 partitions are also created. If a different
1286 number is required it can be simply appended to the auto option.
1287 e.g. "auto=part8". Partition names are created by appending a digit
1288 string to the device name, with an intervening "p" if the device name
1293 option is also available in Build and Create modes. As those modes do
1294 not use a config file, the "auto=" config option does not apply to
1302 and no devices are listed,
1304 will first attempt to assemble all the arrays listed in the config
1307 It will then look further for possible arrays and will try to assemble
1308 anything that it finds. Arrays which are tagged as belonging to the given
1309 homehost will be assembled and started normally. Arrays which do not
1310 obviously belong to this host are given names that are expected not to
1311 conflict with anything local, and are started "read-auto" so that
1312 nothing is written to any device until the array is written to. i.e.
1313 automatic resync etc is delayed.
1317 finds a consistent set of devices that look like they should comprise
1318 an array, and if the superblock is tagged as belonging to the given
1319 home host, it will automatically choose a device name and try to
1320 assemble the array. If the array uses version-0.90 metadata, then the
1322 number as recorded in the superblock is used to create a name in
1326 If the array uses version-1 metadata, then the
1328 from the superblock is used to similarly create a name in
1330 (the name will have any 'host' prefix stripped first).
1334 cannot find any array for the given host at all, and if
1335 .B \-\-auto\-update\-homehost
1338 will search again for any array (not just an array created for this
1339 host) and will assemble each assuming
1340 .BR \-\-update=homehost .
1341 This will change the host tag in the superblock so that on the next run,
1342 these arrays will be found without the second pass. The intention of
1343 this feature is to support transitioning a set of md arrays to using
1346 The reason for requiring arrays to be tagged with the homehost for
1347 auto assembly is to guard against problems that can arise when moving
1348 devices from one host to another.
1358 .BI \-\-raid\-devices= Z
1362 This usage is similar to
1364 The difference is that it creates an array without a superblock. With
1365 these arrays there is no difference between initially creating the array and
1366 subsequently assembling the array, except that hopefully there is useful
1367 data there in the second case.
1369 The level may raid0, linear, multipath, or faulty, or one of their
1370 synonyms. All devices must be listed and the array will be started
1382 .BI \-\-raid\-devices= Z
1386 This usage will initialise a new md array, associate some devices with
1387 it, and activate the array.
1391 option is given (as described in more detail in the section on
1392 Assemble mode), then the md device will be created with a suitable
1393 device number if necessary.
1395 As devices are added, they are checked to see if they contain raid
1396 superblocks or filesystems. They are also checked to see if the variance in
1397 device size exceeds 1%.
1399 If any discrepancy is found, the array will not automatically be run, though
1402 can override this caution.
1404 To create a "degraded" array in which some devices are missing, simply
1405 give the word "\fBmissing\fP"
1406 in place of a device name. This will cause
1408 to leave the corresponding slot in the array empty.
1409 For a RAID4 or RAID5 array at most one slot can be
1410 "\fBmissing\fP"; for a RAID6 array at most two slots.
1411 For a RAID1 array, only one real device needs to be given. All of the
1415 When creating a RAID5 array,
1417 will automatically create a degraded array with an extra spare drive.
1418 This is because building the spare into a degraded array is in general faster than resyncing
1419 the parity on a non-degraded, but not clean, array. This feature can
1420 be overridden with the
1424 When creating an array with version-1 metadata a name for the array is
1426 If this is not given with the
1430 will choose a name based on the last component of the name of the
1431 device being created. So if
1433 is being created, then the name
1438 is being created, then the name
1442 When creating a partition based array, using
1444 with version-1.x metadata, the partition type should be set to
1446 (non fs-data). This type selection allows for greater precision since
1447 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
1448 might create problems in the event of array recovery through a live cdrom.
1450 A new array will normally get a randomly assigned 128bit UUID which is
1451 very likely to be unique. If you have a specific need, you can choose
1452 a UUID for the array by giving the
1454 option. Be warned that creating two arrays with the same UUID is a
1455 recipe for disaster. Also, using
1457 when creating a v0.90 array will silently override any
1462 .\"option is given, it is not necessary to list any component-devices in this command.
1463 .\"They can be added later, before a
1467 .\"is given, the apparent size of the smallest drive given is used.
1469 When creating an array within a
1472 can be given either the list of devices to use, or simply the name of
1473 the container. The former case gives control over which devices in
1474 the container will be used for the array. The latter case allows
1476 to automatically choose which devices to use based on how much spare
1479 The General Management options that are valid with
1484 insist on running the array even if some devices look like they might
1489 start the array readonly \(em not supported yet.
1497 .I options... devices...
1500 This usage will allow individual devices in an array to be failed,
1501 removed or added. It is possible to perform multiple operations with
1502 on command. For example:
1504 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
1510 and will then remove it from the array and finally add it back
1511 in as a spare. However only one md array can be affected by a single
1522 MISC mode includes a number of distinct operations that
1523 operate on distinct devices. The operations are:
1526 The device is examined to see if it is
1527 (1) an active md array, or
1528 (2) a component of an md array.
1529 The information discovered is reported.
1533 The device should be an active md device.
1535 will display a detailed description of the array.
1539 will cause the output to be less detailed and the format to be
1540 suitable for inclusion in
1541 .BR /etc/mdadm.conf .
1544 will normally be 0 unless
1546 failed to get useful information about the device(s); however, if the
1548 option is given, then the exit status will be:
1552 The array is functioning normally.
1555 The array has at least one failed device.
1558 The array has multiple failed devices such that it is unusable.
1561 There was an error while trying to get information about the device.
1565 .B \-\-detail\-platform
1566 Print detail of the platform's raid capabilities (firmware / hardware
1567 topology). If the metadata is specified with
1571 then the return status will be:
1575 metadata successfully enumerated its platform components on this system
1578 metadata is platform independent
1581 metadata failed to find its platform components on this system
1586 The device should be a component of an md array.
1588 will read the md superblock of the device and display the contents.
1593 is given, then multiple devices that are components of the one array
1594 are grouped together and reported in a single entry suitable
1596 .BR /etc/mdadm.conf .
1600 without listing any devices will cause all devices listed in the
1601 config file to be examined.
1605 The devices should be active md arrays which will be deactivated, as
1606 long as they are not currently in use.
1610 This will fully activate a partially assembled md array.
1614 This will mark an active array as read-only, providing that it is
1615 not currently being used.
1621 array back to being read/write.
1625 For all operations except
1628 will cause the operation to be applied to all arrays listed in
1633 causes all devices listed in the config file to be examined.
1640 .B mdadm \-\-monitor
1641 .I options... devices...
1646 to periodically poll a number of md arrays and to report on any events
1649 will never exit once it decides that there are arrays to be checked,
1650 so it should normally be run in the background.
1652 As well as reporting events,
1654 may move a spare drive from one array to another if they are in the
1657 and if the destination array has a failed drive but no spares.
1659 If any devices are listed on the command line,
1661 will only monitor those devices. Otherwise all arrays listed in the
1662 configuration file will be monitored. Further, if
1664 is given, then any other md devices that appear in
1666 will also be monitored.
1668 The result of monitoring the arrays is the generation of events.
1669 These events are passed to a separate program (if specified) and may
1670 be mailed to a given E-mail address.
1672 When passing events to a program, the program is run once for each event,
1673 and is given 2 or 3 command-line arguments: the first is the
1674 name of the event (see below), the second is the name of the
1675 md device which is affected, and the third is the name of a related
1676 device if relevant (such as a component device that has failed).
1680 is given, then a program or an E-mail address must be specified on the
1681 command line or in the config file. If neither are available, then
1683 will not monitor anything.
1687 will continue monitoring as long as something was found to monitor. If
1688 no program or email is given, then each event is reported to
1691 The different events are:
1695 .B DeviceDisappeared
1696 An md array which previously was configured appears to no longer be
1697 configured. (syslog priority: Critical)
1701 was told to monitor an array which is RAID0 or Linear, then it will
1703 .B DeviceDisappeared
1704 with the extra information
1706 This is because RAID0 and Linear do not support the device-failed,
1707 hot-spare and resync operations which are monitored.
1711 An md array started reconstruction. (syslog priority: Warning)
1717 is 20, 40, 60, or 80, this indicates that rebuild has passed that many
1718 percentage of the total. (syslog priority: Warning)
1722 An md array that was rebuilding, isn't any more, either because it
1723 finished normally or was aborted. (syslog priority: Warning)
1727 An active component device of an array has been marked as
1728 faulty. (syslog priority: Critical)
1732 A spare component device which was being rebuilt to replace a faulty
1733 device has failed. (syslog priority: Critical)
1737 A spare component device which was being rebuilt to replace a faulty
1738 device has been successfully rebuilt and has been made active.
1739 (syslog priority: Info)
1743 A new md array has been detected in the
1745 file. (syslog priority: Info)
1749 A newly noticed array appears to be degraded. This message is not
1752 notices a drive failure which causes degradation, but only when
1754 notices that an array is degraded when it first sees the array.
1755 (syslog priority: Critical)
1759 A spare drive has been moved from one array in a
1761 to another to allow a failed drive to be replaced.
1762 (syslog priority: Info)
1768 has been told, via the config file, that an array should have a certain
1769 number of spare devices, and
1771 detects that it has fewer than this number when it first sees the
1772 array, it will report a
1775 (syslog priority: Warning)
1779 An array was found at startup, and the
1782 (syslog priority: Info)
1792 cause Email to be sent. All events cause the program to be run.
1793 The program is run with two or three arguments: the event
1794 name, the array device and possibly a second device.
1796 Each event has an associated array device (e.g.
1798 and possibly a second device. For
1803 the second device is the relevant component device.
1806 the second device is the array that the spare was moved from.
1810 to move spares from one array to another, the different arrays need to
1811 be labeled with the same
1813 in the configuration file. The
1815 name can be any string; it is only necessary that different spare
1816 groups use different names.
1820 detects that an array in a spare group has fewer active
1821 devices than necessary for the complete array, and has no spare
1822 devices, it will look for another array in the same spare group that
1823 has a full complement of working drive and a spare. It will then
1824 attempt to remove the spare from the second drive and add it to the
1826 If the removal succeeds but the adding fails, then it is added back to
1830 The GROW mode is used for changing the size or shape of an active
1832 For this to work, the kernel must support the necessary change.
1833 Various types of growth are being added during 2.6 development,
1834 including restructuring a raid5 array to have more active devices.
1836 Currently the only support available is to
1838 change the "size" attribute
1839 for RAID1, RAID5 and RAID6.
1841 increase the "raid\-devices" attribute of RAID1, RAID5, and RAID6.
1843 add a write-intent bitmap to any array which supports these bitmaps, or
1844 remove a write-intent bitmap from such an array.
1847 GROW mode is not currently supported for
1849 or arrays inside containers.
1852 Normally when an array is built the "size" it taken from the smallest
1853 of the drives. If all the small drives in an arrays are, one at a
1854 time, removed and replaced with larger drives, then you could have an
1855 array of large drives with only a small amount used. In this
1856 situation, changing the "size" with "GROW" mode will allow the extra
1857 space to start being used. If the size is increased in this way, a
1858 "resync" process will start to make sure the new parts of the array
1861 Note that when an array changes size, any filesystem that may be
1862 stored in the array will not automatically grow to use the space. The
1863 filesystem will need to be explicitly told to use the extra space.
1865 .SS RAID-DEVICES CHANGES
1867 A RAID1 array can work with any number of devices from 1 upwards
1868 (though 1 is not very useful). There may be times which you want to
1869 increase or decrease the number of active devices. Note that this is
1870 different to hot-add or hot-remove which changes the number of
1873 When reducing the number of devices in a RAID1 array, the slots which
1874 are to be removed from the array must already be vacant. That is, the
1875 devices which were in those slots must be failed and removed.
1877 When the number of devices is increased, any hot spares that are
1878 present will be activated immediately.
1880 Increasing the number of active devices in a RAID5 is much more
1881 effort. Every block in the array will need to be read and written
1882 back to a new location. From 2.6.17, the Linux Kernel is able to do
1883 this safely, including restart and interrupted "reshape".
1885 When relocating the first few stripes on a raid5, it is not possible
1886 to keep the data on disk completely consistent and crash-proof. To
1887 provide the required safety, mdadm disables writes to the array while
1888 this "critical section" is reshaped, and takes a backup of the data
1889 that is in that section. This backup is normally stored in any spare
1890 devices that the array has, however it can also be stored in a
1891 separate file specified with the
1893 option. If this option is used, and the system does crash during the
1894 critical period, the same file must be passed to
1896 to restore the backup and reassemble the array.
1900 A write-intent bitmap can be added to, or removed from, an active
1901 array. Either internal bitmaps, or bitmaps stored in a separate file,
1902 can be added. Note that if you add a bitmap stored in a file which is
1903 in a filesystem that is on the raid array being affected, the system
1904 will deadlock. The bitmap must be on a separate filesystem.
1906 .SH INCREMENTAL MODE
1910 .B mdadm \-\-incremental
1916 .B mdadm \-\-incremental \-\-rebuild
1919 .B mdadm \-\-incremental \-\-run \-\-scan
1923 This mode is designed to be used in conjunction with a device
1924 discovery system. As devices are found in a system, they can be
1926 .B "mdadm \-\-incremental"
1927 to be conditionally added to an appropriate array.
1929 If the device passed is a
1931 device created by a previous call to
1933 then rather than trying to add that device to an array, all the arrays
1934 described by the metadata of the container will be started.
1937 performs a number of tests to determine if the device is part of an
1938 array, and which array it should be part of. If an appropriate array
1939 is found, or can be created,
1941 adds the device to the array and conditionally starts the array.
1945 will only add devices to an array which were previously working
1946 (active or spare) parts of that array. It does not currently support
1947 automatic inclusion of a new drive as a spare in some array.
1951 makes are as follow:
1953 Is the device permitted by
1955 That is, is it listed in a
1957 line in that file. If
1959 is absent then the default it to allow any device. Similar if
1961 contains the special word
1963 then any device is allowed. Otherwise the device name given to
1965 must match one of the names or patterns in a
1970 Does the device have a valid md superblock. If a specific metadata
1971 version is request with
1975 then only that style of metadata is accepted, otherwise
1977 finds any known version of metadata. If no
1979 metadata is found, the device is rejected.
1982 Does the metadata match an expected array?
1983 The metadata can match in two ways. Either there is an array listed
1986 which identifies the array (either by UUID, by name, by device list,
1987 or by minor-number), or the array was created with a
1993 or on the command line.
1996 is not able to positively identify the array as belonging to the
1997 current host, the device will be rejected.
2001 keeps a list of arrays that it has partially assembled in
2002 .B /var/run/mdadm/map
2004 .B /var/run/mdadm.map
2005 if the directory doesn't exist). If no array exists which matches
2006 the metadata on the new device,
2008 must choose a device name and unit number. It does this based on any
2011 or any name information stored in the metadata. If this name
2012 suggests a unit number, that number will be used, otherwise a free
2013 unit number will be chosen. Normally
2015 will prefer to create a partitionable array, however if the
2019 suggests that a non-partitionable array is preferred, that will be
2023 Once an appropriate array is found or created and the device is added,
2025 must decide if the array is ready to be started. It will
2026 normally compare the number of available (non-spare) devices to the
2027 number of devices that the metadata suggests need to be active. If
2028 there are at least that many, the array will be started. This means
2029 that if any devices are missing the array will not be restarted.
2035 in which case the array will be run as soon as there are enough
2036 devices present for the data to be accessible. For a raid1, that
2037 means one device will start the array. For a clean raid5, the array
2038 will be started as soon as all but one drive is present.
2040 Note that neither of these approaches is really ideal. If it can
2041 be known that all device discovery has completed, then
2045 can be run which will try to start all arrays that are being
2046 incrementally assembled. They are started in "read-auto" mode in
2047 which they are read-only until the first write request. This means
2048 that no metadata updates are made and no attempt at resync or recovery
2049 happens. Further devices that are found before the first write can
2050 still be added safely.
2054 This section describes environment variables that affect how mdadm
2059 Setting this value to 1 will prevent mdadm from automatically launching
2060 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2066 does not create any device nodes in /dev, but leaves that task to
2070 appears not to be configured, or if this environment variable is set
2073 will create and devices that are needed.
2077 .B " mdadm \-\-query /dev/name-of-device"
2079 This will find out if a given device is a raid array, or is part of
2080 one, and will provide brief information about the device.
2082 .B " mdadm \-\-assemble \-\-scan"
2084 This will assemble and start all arrays listed in the standard config
2085 file. This command will typically go in a system startup file.
2087 .B " mdadm \-\-stop \-\-scan"
2089 This will shut down all arrays that can be shut down (i.e. are not
2090 currently in use). This will typically go in a system shutdown script.
2092 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2094 If (and only if) there is an Email address or program given in the
2095 standard config file, then
2096 monitor the status of all arrays listed in that file by
2097 polling them ever 2 minutes.
2099 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2101 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2104 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2106 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2108 This will create a prototype config file that describes currently
2109 active arrays that are known to be made from partitions of IDE or SCSI drives.
2110 This file should be reviewed before being used as it may
2111 contain unwanted detail.
2113 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2115 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
2117 This will find arrays which could be assembled from existing IDE and
2118 SCSI whole drives (not partitions), and store the information in the
2119 format of a config file.
2120 This file is very likely to contain unwanted detail, particularly
2123 entries. It should be reviewed and edited before being used as an
2126 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
2128 .B " mdadm \-Ebsc partitions"
2130 Create a list of devices by reading
2131 .BR /proc/partitions ,
2132 scan these for RAID superblocks, and printout a brief listing of all
2135 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
2137 Scan all partitions and devices listed in
2138 .BR /proc/partitions
2141 out of all such devices with a RAID superblock with a minor number of 0.
2143 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /var/run/mdadm"
2145 If config file contains a mail address or alert program, run mdadm in
2146 the background in monitor mode monitoring all md devices. Also write
2147 pid of mdadm daemon to
2148 .BR /var/run/mdadm .
2150 .B " mdadm \-Iq /dev/somedevice"
2152 Try to incorporate newly discovered device into some array as
2155 .B " mdadm \-\-incremental \-\-rebuild \-\-run \-\-scan"
2157 Rebuild the array map from any current arrays, and then start any that
2160 .B " mdadm /dev/md4 --fail detached --remove detached"
2162 Any devices which are components of /dev/md4 will be marked as faulty
2163 and then remove from the array.
2165 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
2167 Create a DDF array over 6 devices.
2169 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
2171 Create a raid5 array over any 3 devices in the given DDF set. Use
2172 only 30 gigabytes of each device.
2174 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
2176 Assemble a pre-exist ddf array.
2178 .B " mdadm -I /dev/md/ddf1"
2180 Assemble all arrays contained in the ddf array, assigning names as
2183 .B " mdadm \-\-create \-\-help"
2185 Provide help about the Create mode.
2187 .B " mdadm \-\-config \-\-help"
2189 Provide help about the format of the config file.
2191 .B " mdadm \-\-help"
2193 Provide general help.
2204 lists all active md devices with information about them.
2206 uses this to find arrays when
2208 is given in Misc mode, and to monitor array reconstruction
2214 The config file lists which devices may be scanned to see if
2215 they contain MD super block, and gives identifying information
2216 (e.g. UUID) about known MD arrays. See
2220 .SS /var/run/mdadm/map
2223 mode is used, this file gets a list of arrays currently being created.
2226 does not exist as a directory, then
2227 .B /var/run/mdadm.map
2233 understand two sorts of names for array devices.
2235 The first is the so-called 'standard' format name, which matches the
2236 names used by the kernel and which appear in
2239 The second sort can be freely chosen, but must reside in
2241 When giving a device name to
2243 to create or assemble an array, either full path name such as
2247 can be given, or just the suffix of the second sort of name, such as
2253 chooses device names during auto-assembly, it will normally add a
2254 small sequence number to the end of the name to avoid conflicted
2255 between multiple arrays that have the same name. If
2257 can reasonably determine that the array really is meant for this host,
2258 either by a hostname in the metadata, or by the presence of the array
2259 in /etc/mdadm.conf, then it will leave of the suffix if possible.
2261 The standard names for non-partitioned arrays (the only sort of md
2262 array available in 2.4 and earlier) are of the form
2266 where NN is a number.
2267 The standard names for partitionable arrays (as available from 2.6
2268 onwards) are of the form
2272 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2274 From kernel version, 2.6.28 the "non-partitioned array" can actually
2275 be partitioned. So the "md_dNN" names are no longer needed, and
2276 partitions such as "/dev/mdNNpXX" are possible.
2280 was previously known as
2284 is completely separate from the
2286 package, and does not use the
2288 configuration file at all.
2291 For further information on mdadm usage, MD and the various levels of
2294 .B http://linux\-raid.osdl.org/
2296 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
2298 .\"for new releases of the RAID driver check out:
2301 .\".UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2302 .\"ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2307 .\".UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2308 .\"http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2311 The latest version of
2313 should always be available from
2315 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/