2 .\" Copyright Neil Brown and others.
3 .\" This program is free software; you can redistribute it and/or modify
4 .\" it under the terms of the GNU General Public License as published by
5 .\" the Free Software Foundation; either version 2 of the License, or
6 .\" (at your option) any later version.
7 .\" See file COPYING in distribution for details.
10 mdadm \- manage MD devices
16 .BI mdadm " [mode] <raiddevice> [options] <component-devices>"
19 RAID devices are virtual devices created from two or more
20 real block devices. This allows multiple devices (typically disk
21 drives or partitions thereof) to be combined into a single device to
22 hold (for example) a single filesystem.
23 Some RAID levels include redundancy and so can survive some degree of
26 Linux Software RAID devices are implemented through the md (Multiple
27 Devices) device driver.
29 Currently, Linux supports
46 is not a Software RAID mechanism, but does involve
48 each device is a path to one common physical storage device.
49 New installations should not use md/multipath as it is not well
50 supported and has no ongoing development. Use the Device Mapper based
51 multipath-tools instead.
54 is also not true RAID, and it only involves one device. It
55 provides a layer over a true device that can be used to inject faults.
60 is a collection of devices that are
61 managed as a set. This is similar to the set of devices connected to
62 a hardware RAID controller. The set of devices may contain a number
63 of different RAID arrays each utilising some (or all) of the blocks from a
64 number of the devices in the set. For example, two devices in a 5-device set
65 might form a RAID1 using the whole devices. The remaining three might
66 have a RAID5 over the first half of each device, and a RAID0 over the
71 there is one set of metadata that describes all of
72 the arrays in the container. So when
76 device, the device just represents the metadata. Other normal arrays (RAID1
77 etc) can be created inside the container.
80 mdadm has several major modes of operation:
83 Assemble the components of a previously created
84 array into an active array. Components can be explicitly given
85 or can be searched for.
87 checks that the components
88 do form a bona fide array, and can, on request, fiddle superblock
89 information so as to assemble a faulty array.
93 Build an array that doesn't have per-device metadata (superblocks). For these
96 cannot differentiate between initial creation and subsequent assembly
97 of an array. It also cannot perform any checks that appropriate
98 components have been requested. Because of this, the
100 mode should only be used together with a complete understanding of
105 Create a new array with per-device metadata (superblocks).
106 Appropriate metadata is written to each device, and then the array
107 comprising those devices is activated. A 'resync' process is started
108 to make sure that the array is consistent (e.g. both sides of a mirror
109 contain the same data) but the content of the device is left otherwise
111 The array can be used as soon as it has been created. There is no
112 need to wait for the initial resync to finish.
115 .B "Follow or Monitor"
116 Monitor one or more md devices and act on any state changes. This is
117 only meaningful for RAID1, 4, 5, 6, 10 or multipath arrays, as
118 only these have interesting state. RAID0 or Linear never have
119 missing, spare, or failed drives, so there is nothing to monitor.
123 Grow (or shrink) an array, or otherwise reshape it in some way.
124 Currently supported growth options including changing the active size
125 of component devices and changing the number of active devices in RAID
126 levels 1/4/5/6, changing the RAID level between 1, 5, and 6, changing
127 the chunk size and layout for RAID5 and RAID5, as well as adding or
128 removing a write-intent bitmap.
131 .B "Incremental Assembly"
132 Add a single device to an appropriate array. If the addition of the
133 device makes the array runnable, the array will be started.
134 This provides a convenient interface to a
136 system. As each device is detected,
138 has a chance to include it in some array as appropriate.
144 in this mode, then any arrays within that container will be assembled
149 This is for doing things to specific components of an array such as
150 adding new spares and removing faulty devices.
154 This is an 'everything else' mode that supports operations on active
155 arrays, operations on component devices such as erasing old superblocks, and
156 information gathering operations.
157 .\"This mode allows operations on independent devices such as examine MD
158 .\"superblocks, erasing old superblocks and stopping active arrays.
162 This mode does not act on a specific device or array, but rather it
163 requests the Linux Kernel to activate any auto-detected arrays.
166 .SH Options for selecting a mode are:
169 .BR \-A ", " \-\-assemble
170 Assemble a pre-existing array.
173 .BR \-B ", " \-\-build
174 Build a legacy array without superblocks.
177 .BR \-C ", " \-\-create
181 .BR \-F ", " \-\-follow ", " \-\-monitor
187 .BR \-G ", " \-\-grow
188 Change the size or shape of an active array.
191 .BR \-I ", " \-\-incremental
192 Add a single device into an appropriate array, and possibly start the array.
196 Request that the kernel starts any auto-detected arrays. This can only
199 is compiled into the kernel \(em not if it is a module.
200 Arrays can be auto-detected by the kernel if all the components are in
201 primary MS-DOS partitions with partition type
203 and all use v0.90 metadata.
204 In-kernel autodetect is not recommended for new installations. Using
206 to detect and assemble arrays \(em possibly in an
208 \(em is substantially more flexible and should be preferred.
211 If a device is given before any options, or if the first option is
216 then the MANAGE mode is assumed.
217 Anything other than these will cause the
221 .SH Options that are not mode-specific are:
224 .BR \-h ", " \-\-help
225 Display general help message or, after one of the above options, a
226 mode-specific help message.
230 Display more detailed help about command line parsing and some commonly
234 .BR \-V ", " \-\-version
235 Print version information for mdadm.
238 .BR \-v ", " \-\-verbose
239 Be more verbose about what is happening. This can be used twice to be
241 The extra verbosity currently only affects
242 .B \-\-detail \-\-scan
244 .BR "\-\-examine \-\-scan" .
247 .BR \-q ", " \-\-quiet
248 Avoid printing purely informative messages. With this,
250 will be silent unless there is something really important to report.
253 .BR \-b ", " \-\-brief
254 Be less verbose. This is used with
262 gives an intermediate level of verbosity.
265 .BR \-f ", " \-\-force
266 Be more forceful about certain operations. See the various modes for
267 the exact meaning of this option in different contexts.
270 .BR \-c ", " \-\-config=
271 Specify the config file. Default is to use
272 .BR /etc/mdadm.conf ,
273 or if that is missing then
274 .BR /etc/mdadm/mdadm.conf .
275 If the config file given is
277 then nothing will be read, but
279 will act as though the config file contained exactly
280 .B "DEVICE partitions containers"
283 to find a list of devices to scan, and
285 to find a list of containers to examine.
288 is given for the config file, then
290 will act as though the config file were empty.
293 .BR \-s ", " \-\-scan
296 for missing information.
297 In general, this option gives
299 permission to get any missing information (like component devices,
300 array devices, array identities, and alert destination) from the
301 configuration file (see previous option);
302 one exception is MISC mode when using
308 says to get a list of array devices from
312 .B \-e ", " \-\-metadata=
313 Declare the style of RAID metadata (superblock) to be used. The
316 and to guess for other operations.
317 The default can be overridden by setting the
326 .IP "0, 0.90, default"
327 Use the original 0.90 format superblock. This format limits arrays to
328 28 component devices and limits component devices of levels 1 and
329 greater to 2 terabytes.
330 .IP "1, 1.0, 1.1, 1.2"
331 Use the new version-1 format superblock. This has few restrictions.
332 The different sub-versions store the superblock at different locations
333 on the device, either at the end (for 1.0), at the start (for 1.1) or
334 4K from the start (for 1.2).
336 Use the "Industry Standard" DDF (Disk Data Format) format defined by
338 When creating a DDF array a
340 will be created, and normal arrays can be created in that container.
342 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
344 which is managed in a similar manner to DDF, and is supported by an
345 option-rom on some platforms:
347 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
353 This will override any
355 setting in the config file and provides the identity of the host which
356 should be considered the home for any arrays.
358 When creating an array, the
360 will be recorded in the metadata. For version-1 superblocks, it will
361 be prefixed to the array name. For version-0.90 superblocks, part of
362 the SHA1 hash of the hostname will be stored in the later half of the
365 When reporting information about an array, any array which is tagged
366 for the given homehost will be reported as such.
368 When using Auto-Assemble, only arrays tagged for the given homehost
369 will be allowed to use 'local' names (i.e. not ending in '_' followed
370 by a digit string). See below under
371 .BR "Auto Assembly" .
373 .SH For create, build, or grow:
376 .BR \-n ", " \-\-raid\-devices=
377 Specify the number of active devices in the array. This, plus the
378 number of spare devices (see below) must equal the number of
380 (including "\fBmissing\fP" devices)
381 that are listed on the command line for
383 Setting a value of 1 is probably
384 a mistake and so requires that
386 be specified first. A value of 1 will then be allowed for linear,
387 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
389 This number can only be changed using
391 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
392 the necessary support.
395 .BR \-x ", " \-\-spare\-devices=
396 Specify the number of spare (eXtra) devices in the initial array.
397 Spares can also be added
398 and removed later. The number of component devices listed
399 on the command line must equal the number of RAID devices plus the
400 number of spare devices.
403 .BR \-z ", " \-\-size=
404 Amount (in Kibibytes) of space to use from each drive in RAID levels 1/4/5/6.
405 This must be a multiple of the chunk size, and must leave about 128Kb
406 of space at the end of the drive for the RAID superblock.
407 If this is not specified
408 (as it normally is not) the smallest drive (or partition) sets the
409 size, though if there is a variance among the drives of greater than 1%, a warning is
412 This value can be set with
414 for RAID level 1/4/5/6. If the array was created with a size smaller
415 than the currently active drives, the extra space can be accessed
418 The size can be given as
420 which means to choose the largest size that fits on all current drives.
422 This value can not be used with
424 metadata such as DDF and IMSM.
427 .BR \-Z ", " \-\-array-size=
428 This is only meaningful with
430 and its effect is not persistent: when the array is stopped an
431 restarted the default array size will be restored.
433 Setting the array-size causes the array to appear smaller to programs
434 that access the data. This is particularly needed before reshaping an
435 array so that it will be smaller. As the reshape is not reversible,
436 but setting the size with
438 is, it is required that the array size is reduced as appropriate
439 before the number of devices in the array is reduced.
442 .BR \-c ", " \-\-chunk=
443 Specify chunk size of kibibytes. The default is 64.
444 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
448 Specify rounding factor for a Linear array. The size of each
449 component will be rounded down to a multiple of this size.
450 This is a synonym for
452 but highlights the different meaning for Linear as compared to other
456 .BR \-l ", " \-\-level=
457 Set RAID level. When used with
459 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
460 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
461 Obviously some of these are synonymous.
465 metadata type is requested, only the
467 level is permitted, and it does not need to be explicitly given.
471 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
473 Not yet supported with
477 .BR \-p ", " \-\-layout=
478 This option configures the fine details of data layout for RAID5, RAID6,
479 and RAID10 arrays, and controls the failure modes for
482 The layout of the RAID5 parity block can be one of
483 .BR left\-asymmetric ,
484 .BR left\-symmetric ,
485 .BR right\-asymmetric ,
486 .BR right\-symmetric ,
487 .BR la ", " ra ", " ls ", " rs .
489 .BR left\-symmetric .
491 It is also possibly to cause RAID5 to use a RAID4-like layout by
497 Finally for RAID5 there are DDF\-compatible layouts,
498 .BR ddf\-zero\-restart ,
499 .BR ddf\-N\-restart ,
501 .BR ddf\-N\-continue .
503 These same layouts are available for RAID6. There are also 4 layouts
504 that will provide an intermediate stage for converting between RAID5
505 and RAID6. These provide a layout which is identical to the
506 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
507 syndrome (the second 'parity' block used by RAID6) on the last device.
509 .BR left\-symmetric\-6 ,
510 .BR right\-symmetric\-6 ,
511 .BR left\-asymmetric\-6 ,
512 .BR right\-asymmetric\-6 ,
514 .BR pairty\-first\-6 .
516 When setting the failure mode for level
519 .BR write\-transient ", " wt ,
520 .BR read\-transient ", " rt ,
521 .BR write\-persistent ", " wp ,
522 .BR read\-persistent ", " rp ,
524 .BR read\-fixable ", " rf ,
525 .BR clear ", " flush ", " none .
527 Each failure mode can be followed by a number, which is used as a period
528 between fault generation. Without a number, the fault is generated
529 once on the first relevant request. With a number, the fault will be
530 generated after that many requests, and will continue to be generated
531 every time the period elapses.
533 Multiple failure modes can be current simultaneously by using the
535 option to set subsequent failure modes.
537 "clear" or "none" will remove any pending or periodic failure modes,
538 and "flush" will clear any persistent faults.
540 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
541 by a small number. The default is 'n2'. The supported options are:
544 signals 'near' copies. Multiple copies of one data block are at
545 similar offsets in different devices.
548 signals 'offset' copies. Rather than the chunks being duplicated
549 within a stripe, whole stripes are duplicated but are rotated by one
550 device so duplicate blocks are on different devices. Thus subsequent
551 copies of a block are in the next drive, and are one chunk further
556 (multiple copies have very different offsets).
557 See md(4) for more detail about 'near', 'offset', and 'far'.
559 The number is the number of copies of each datablock. 2 is normal, 3
560 can be useful. This number can be at most equal to the number of
561 devices in the array. It does not need to divide evenly into that
562 number (e.g. it is perfectly legal to have an 'n2' layout for an array
563 with an odd number of devices).
565 When an array is converted between RAID5 and RAID6 an intermediate
566 RAID6 layout is used in which the second parity block (Q) is always on
567 the last device. To convert a RAID5 to RAID6 and leave it in this new
568 layout (which does not require re-striping) use
569 .BR \-\-layout=preserve .
570 This will try to avoid any restriping.
572 The converse of this is
573 .B \-\-layout=normalise
574 which will change a non-standard RAID6 layout into a more standard
581 (thus explaining the p of
585 .BR \-b ", " \-\-bitmap=
586 Specify a file to store a write-intent bitmap in. The file should not
589 is also given. The same file should be provided
590 when assembling the array. If the word
592 is given, then the bitmap is stored with the metadata on the array,
593 and so is replicated on all devices. If the word
597 mode, then any bitmap that is present is removed.
599 To help catch typing errors, the filename must contain at least one
600 slash ('/') if it is a real file (not 'internal' or 'none').
602 Note: external bitmaps are only known to work on ext2 and ext3.
603 Storing bitmap files on other filesystems may result in serious problems.
606 .BR \-\-bitmap\-chunk=
607 Set the chunksize of the bitmap. Each bit corresponds to that many
608 Kilobytes of storage.
609 When using a file based bitmap, the default is to use the smallest
610 size that is at-least 4 and requires no more than 2^21 chunks.
613 bitmap, the chunksize is automatically determined to make best use of
617 .BR \-W ", " \-\-write\-mostly
618 subsequent devices listed in a
623 command will be flagged as 'write-mostly'. This is valid for RAID1
624 only and means that the 'md' driver will avoid reading from these
625 devices if at all possible. This can be useful if mirroring over a
629 .BR \-\-write\-behind=
630 Specify that write-behind mode should be enabled (valid for RAID1
631 only). If an argument is specified, it will set the maximum number
632 of outstanding writes allowed. The default value is 256.
633 A write-intent bitmap is required in order to use write-behind
634 mode, and write-behind is only attempted on drives marked as
638 .BR \-\-assume\-clean
641 that the array pre-existed and is known to be clean. It can be useful
642 when trying to recover from a major failure as you can be sure that no
643 data will be affected unless you actually write to the array. It can
644 also be used when creating a RAID1 or RAID10 if you want to avoid the
645 initial resync, however this practice \(em while normally safe \(em is not
646 recommended. Use this only if you really know what you are doing.
649 .BR \-\-backup\-file=
652 is used to increase the number of
653 raid-devices in a RAID5 if there are no spare devices available.
654 See the GROW MODE section below on RAID\-DEVICES CHANGES. The file
655 should be stored on a separate device, not on the RAID array being
659 .BR \-\-array-size= ", " \-Z
660 Set the size of the array which is seen by users of the device such as
661 filesystems. This can be less that the real size, but never greater.
662 The size set this way does not persist across restarts of the array.
664 This is most useful when reducing the number of devices in a RAID5 or
665 RAID6. Such arrays require the array-size to be reduced before a
666 reshape can be performed that reduces the real size.
670 restores the apparent size of the array to be whatever the real
671 amount of available space is.
674 .BR \-N ", " \-\-name=
677 for the array. This is currently only effective when creating an
678 array with a version-1 superblock, or an array in a DDF container.
679 The name is a simple textual string that can be used to identify array
680 components when assembling. If name is needed but not specified, it
681 is taken from the basename of the device that is being created.
693 run the array, even if some of the components
694 appear to be active in another array or filesystem. Normally
696 will ask for confirmation before including such components in an
697 array. This option causes that question to be suppressed.
700 .BR \-f ", " \-\-force
703 accept the geometry and layout specified without question. Normally
705 will not allow creation of an array with only one device, and will try
706 to create a RAID5 array with one missing drive (as this makes the
707 initial resync work faster). With
710 will not try to be so clever.
713 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
714 Instruct mdadm how to create the device file if needed, possibly allocating
715 an unused minor number. "md" causes a non-partitionable array
716 to be used (though since Linux 2.6.28, these array devices are in fact
717 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
718 later) to be used. "yes" requires the named md device to have
719 a 'standard' format, and the type and minor number will be determined
720 from this. With mdadm 3.0, device creation is normally left up to
722 so this option is unlikely to be needed.
723 See DEVICE NAMES below.
725 The argument can also come immediately after
730 is not given on the command line or in the config file, then
736 is also given, then any
738 entries in the config file will override the
740 instruction given on the command line.
742 For partitionable arrays,
744 will create the device file for the whole array and for the first 4
745 partitions. A different number of partitions can be specified at the
746 end of this option (e.g.
748 If the device name ends with a digit, the partition names add a 'p',
750 .IR /dev/md/home1p3 .
751 If there is no trailing digit, then the partition names just have a
753 .IR /dev/md/scratch3 .
755 If the md device name is in a 'standard' format as described in DEVICE
756 NAMES, then it will be created, if necessary, with the appropriate
757 device number based on that name. If the device name is not in one of these
758 formats, then a unused device number will be allocated. The device
759 number will be considered unused if there is no active array for that
760 number, and there is no entry in /dev for that number and with a
761 non-standard name. Names that are not in 'standard' format are only
762 allowed in "/dev/md/".
766 .\".BR \-\-symlink = no
771 .\"to create devices in
773 .\"it will also create symlinks from
775 .\"with names starting with
781 .\"to suppress this, or
782 .\".B \-\-symlink=yes
783 .\"to enforce this even if it is suppressing
791 .BR \-u ", " \-\-uuid=
792 uuid of array to assemble. Devices which don't have this uuid are
796 .BR \-m ", " \-\-super\-minor=
797 Minor number of device that array was created for. Devices which
798 don't have this minor number are excluded. If you create an array as
799 /dev/md1, then all superblocks will contain the minor number 1, even if
800 the array is later assembled as /dev/md2.
802 Giving the literal word "dev" for
806 to use the minor number of the md device that is being assembled.
809 .B \-\-super\-minor=dev
810 will look for super blocks with a minor number of 0.
813 is only relevant for v0.90 metadata, and should not normally be used.
819 .BR \-N ", " \-\-name=
820 Specify the name of the array to assemble. This must be the name
821 that was specified when creating the array. It must either match
822 the name stored in the superblock exactly, or it must match
825 prefixed to the start of the given name.
828 .BR \-f ", " \-\-force
829 Assemble the array even if the metadata on some devices appears to be
832 cannot find enough working devices to start the array, but can find
833 some devices that are recorded as having failed, then it will mark
834 those devices as working so that the array can be started.
835 An array which requires
837 to be started may contain data corruption. Use it carefully.
841 Attempt to start the array even if fewer drives were given than were
842 present last time the array was active. Normally if not all the
843 expected drives are found and
845 is not used, then the array will be assembled but not started.
848 an attempt will be made to start it anyway.
852 This is the reverse of
854 in that it inhibits the startup of array unless all expected drives
855 are present. This is only needed with
857 and can be used if the physical connections to devices are
858 not as reliable as you would like.
861 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
862 See this option under Create and Build options.
865 .BR \-b ", " \-\-bitmap=
866 Specify the bitmap file that was given when the array was created. If
869 bitmap, there is no need to specify this when assembling the array.
872 .BR \-\-backup\-file=
875 was used to grow the number of raid-devices in a RAID5, and the system
876 crashed during the critical section, then the same
880 to allow possibly corrupted data to be restored.
883 .BR \-U ", " \-\-update=
884 Update the superblock on each device while assembling the array. The
885 argument given to this flag can be one of
899 option will adjust the superblock of an array what was created on a Sparc
900 machine running a patched 2.2 Linux kernel. This kernel got the
901 alignment of part of the superblock wrong. You can use the
902 .B "\-\-examine \-\-sparc2.2"
905 to see what effect this would have.
909 option will update the
911 field on each superblock to match the minor number of the array being
913 This can be useful if
915 reports a different "Preferred Minor" to
917 In some cases this update will be performed automatically
918 by the kernel driver. In particular the update happens automatically
919 at the first write to an array with redundancy (RAID level 1 or
920 greater) on a 2.6 (or later) kernel.
924 option will change the uuid of the array. If a UUID is given with the
926 option that UUID will be used as a new UUID and will
928 be used to help identify the devices in the array.
931 is given, a random UUID is chosen.
935 option will change the
937 of the array as stored in the superblock. This is only supported for
938 version-1 superblocks.
942 option will change the
944 as recorded in the superblock. For version-0 superblocks, this is the
945 same as updating the UUID.
946 For version-1 superblocks, this involves updating the name.
950 option will cause the array to be marked
952 meaning that any redundancy in the array (e.g. parity for RAID5,
953 copies for RAID1) may be incorrect. This will cause the RAID system
954 to perform a "resync" pass to make sure that all redundant information
959 option allows arrays to be moved between machines with different
961 When assembling such an array for the first time after a move, giving
962 .B "\-\-update=byteorder"
965 to expect superblocks to have their byteorder reversed, and will
966 correct that order before assembling the array. This is only valid
967 with original (Version 0.90) superblocks.
971 option will correct the summaries in the superblock. That is the
972 counts of total, working, active, failed, and spare devices.
976 will rarely be of use. It applies to version 1.1 and 1.2 metadata
977 only (where the metadata is at the start of the device) and is only
978 useful when the component device has changed size (typically become
979 larger). The version 1 metadata records the amount of the device that
980 can be used to store data, so if a device in a version 1.1 or 1.2
981 array becomes larger, the metadata will still be visible, but the
982 extra space will not. In this case it might be useful to assemble the
984 .BR \-\-update=devicesize .
987 to determine the maximum usable amount of space on each device and
988 update the relevant field in the metadata.
992 .B \-\-auto\-update\-homehost
993 This flag is only meaningful with auto-assembly (see discussion below).
994 In that situation, if no suitable arrays are found for this homehost,
996 will rescan for any arrays at all and will assemble them and update the
997 homehost to match the current host.
1000 .SH For Manage mode:
1003 .BR \-a ", " \-\-add
1004 hot-add listed devices. For arrays with redundancy, the listed
1005 devices become available as spares. If the array is degraded, it will
1006 immediately start recovering data on to one of these spares.
1010 re-add a device that was recently removed from an array. This is only
1011 needed for arrays that have be built (i.e. with
1013 For created arrays, devices are always re-added if that is possible.
1014 When re-adding a device, if nothing has changed on the array since the
1015 device was removed, no recovery is performed. Also, if the array has
1016 a write-intent bitmap, then the recovery performed after a re-add will
1017 be limited to those blocks which, according to the bitmap, might have
1018 changed since the device was removed.
1021 .BR \-r ", " \-\-remove
1022 remove listed devices. They must not be active. i.e. they should
1023 be failed or spare devices. As well as the name of a device file
1032 The first causes all failed device to be removed. The second causes
1033 any device which is no longer connected to the system (i.e an 'open'
1036 to be removed. This will only succeed for devices that are spares or
1037 have already been marked as failed.
1040 .BR \-f ", " \-\-fail
1041 mark listed devices as faulty.
1042 As well as the name of a device file, the word
1044 can be given. This will cause any device that has been detached from
1045 the system to be marked as failed. It can then be removed.
1053 .BR \-\-write\-mostly
1054 Subsequent devices that are added or re-added will have the 'write-mostly'
1055 flag set. This is only valid for RAID1 and means that the 'md' driver
1056 will avoid reading from these devices if possible.
1059 Subsequent devices that are added or re-added will have the 'write-mostly'
1063 Each of these options requires that the first device listed is the array
1064 to be acted upon, and the remainder are component devices to be added,
1065 removed, marked as faulty, etc. Several different operations can be
1066 specified for different devices, e.g.
1068 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1070 Each operation applies to all devices listed until the next
1073 If an array is using a write-intent bitmap, then devices which have
1074 been removed can be re-added in a way that avoids a full
1075 reconstruction but instead just updates the blocks that have changed
1076 since the device was removed. For arrays with persistent metadata
1077 (superblocks) this is done automatically. For arrays created with
1079 mdadm needs to be told that this device we removed recently with
1082 Devices can only be removed from an array if they are not in active
1083 use, i.e. that must be spares or failed devices. To remove an active
1084 device, it must first be marked as
1090 .BR \-Q ", " \-\-query
1091 Examine a device to see
1092 (1) if it is an md device and (2) if it is a component of an md
1094 Information about what is discovered is presented.
1097 .BR \-D ", " \-\-detail
1098 Print details of one or more md devices.
1101 .BR \-\-detail\-platform
1102 Print details of the platform's RAID capabilities (firmware / hardware
1103 topology) for a given metadata format.
1106 .BR \-Y ", " \-\-export
1111 output will be formatted as
1113 pairs for easy import into the environment.
1116 .BR \-E ", " \-\-examine
1117 Print contents of the metadata stored on the named device(s).
1118 Note the contrast between
1123 applies to devices which are components of an array, while
1125 applies to a whole array which is currently active.
1128 If an array was created on a SPARC machine with a 2.2 Linux kernel
1129 patched with RAID support, the superblock will have been created
1130 incorrectly, or at least incompatibly with 2.4 and later kernels.
1135 will fix the superblock before displaying it. If this appears to do
1136 the right thing, then the array can be successfully assembled using
1137 .BR "\-\-assemble \-\-update=sparc2.2" .
1140 .BR \-X ", " \-\-examine\-bitmap
1141 Report information about a bitmap file.
1142 The argument is either an external bitmap file or an array component
1143 in case of an internal bitmap. Note that running this on an array
1146 does not report the bitmap for that array.
1149 .BR \-R ", " \-\-run
1150 start a partially assembled array. If
1152 did not find enough devices to fully start the array, it might leaving
1153 it partially assembled. If you wish, you can then use
1155 to start the array in degraded mode.
1158 .BR \-S ", " \-\-stop
1159 deactivate array, releasing all resources.
1162 .BR \-o ", " \-\-readonly
1163 mark array as readonly.
1166 .BR \-w ", " \-\-readwrite
1167 mark array as readwrite.
1170 .B \-\-zero\-superblock
1171 If the device contains a valid md superblock, the block is
1172 overwritten with zeros. With
1174 the block where the superblock would be is overwritten even if it
1175 doesn't appear to be valid.
1178 .BR \-t ", " \-\-test
1183 is set to reflect the status of the device. See below in
1188 .BR \-W ", " \-\-wait
1189 For each md device given, wait for any resync, recovery, or reshape
1190 activity to finish before returning.
1192 will return with success if it actually waited for every device
1193 listed, otherwise it will return failure.
1197 For each md device given, or each device in /proc/mdstat if
1199 is given, arrange for the array to be marked clean as soon as possible.
1200 Also, quiesce resync so that the monitor for external metadata arrays
1201 (mdmon) has an opportunity to checkpoint the resync position.
1203 will return with success if the array uses external metadata and we
1204 successfully waited. For native arrays this returns immediately as the
1205 kernel handles both dirty-clean transitions and resync checkpointing in
1206 the kernel at shutdown. No action is taken if safe-mode handling is
1209 .SH For Incremental Assembly mode:
1211 .BR \-\-rebuild\-map ", " \-r
1212 Rebuild the map file
1213 .RB ( /var/run/mdadm/map )
1216 uses to help track which arrays are currently being assembled.
1219 .BR \-\-run ", " \-R
1220 Run any array assembled as soon as a minimal number of devices are
1221 available, rather than waiting until all expected devices are present.
1225 This allows the hot-plug system to prevent arrays from running when it knows
1226 that more disks may arrive later in the discovery process.
1229 .BR \-\-scan ", " \-s
1230 Only meaningful with
1234 file for arrays that are being incrementally assembled and will try to
1235 start any that are not already started. If any such array is listed
1238 as requiring an external bitmap, that bitmap will be attached first.
1240 .SH For Monitor mode:
1242 .BR \-m ", " \-\-mail
1243 Give a mail address to send alerts to.
1246 .BR \-p ", " \-\-program ", " \-\-alert
1247 Give a program to be run whenever an event is detected.
1250 .BR \-y ", " \-\-syslog
1251 Cause all events to be reported through 'syslog'. The messages have
1252 facility of 'daemon' and varying priorities.
1255 .BR \-d ", " \-\-delay
1256 Give a delay in seconds.
1258 polls the md arrays and then waits this many seconds before polling
1259 again. The default is 60 seconds. Since 2.6.16, there is no need to
1260 reduce this as the kernel alerts
1262 immediately when there is any change.
1265 .BR \-f ", " \-\-daemonise
1268 to run as a background daemon if it decides to monitor anything. This
1269 causes it to fork and run in the child, and to disconnect from the
1270 terminal. The process id of the child is written to stdout.
1273 which will only continue monitoring if a mail address or alert program
1274 is found in the config file.
1277 .BR \-i ", " \-\-pid\-file
1280 is running in daemon mode, write the pid of the daemon process to
1281 the specified file, instead of printing it on standard output.
1284 .BR \-1 ", " \-\-oneshot
1285 Check arrays only once. This will generate
1287 events and more significantly
1293 .B " mdadm \-\-monitor \-\-scan \-1"
1295 from a cron script will ensure regular notification of any degraded arrays.
1298 .BR \-t ", " \-\-test
1301 alert for every array found at startup. This alert gets mailed and
1302 passed to the alert program. This can be used for testing that alert
1303 message do get through successfully.
1309 .B mdadm \-\-assemble
1310 .I md-device options-and-component-devices...
1313 .B mdadm \-\-assemble \-\-scan
1314 .I md-devices-and-options...
1317 .B mdadm \-\-assemble \-\-scan
1321 This usage assembles one or more RAID arrays from pre-existing components.
1322 For each array, mdadm needs to know the md device, the identity of the
1323 array, and a number of component-devices. These can be found in a number of ways.
1325 In the first usage example (without the
1327 the first device given is the md device.
1328 In the second usage example, all devices listed are treated as md
1329 devices and assembly is attempted.
1330 In the third (where no devices are listed) all md devices that are
1331 listed in the configuration file are assembled. If not arrays are
1332 described by the configuration file, then any arrays that
1333 can be found on unused devices will be assembled.
1335 If precisely one device is listed, but
1341 was given and identity information is extracted from the configuration file.
1343 The identity can be given with the
1349 option, will be taken from the md-device record in the config file, or
1350 will be taken from the super block of the first component-device
1351 listed on the command line.
1353 Devices can be given on the
1355 command line or in the config file. Only devices which have an md
1356 superblock which contains the right identity will be considered for
1359 The config file is only used if explicitly named with
1361 or requested with (a possibly implicit)
1366 .B /etc/mdadm/mdadm.conf
1371 is not given, then the config file will only be used to find the
1372 identity of md arrays.
1374 Normally the array will be started after it is assembled. However if
1376 is not given and not all expected drives were listed, then the array
1377 is not started (to guard against usage errors). To insist that the
1378 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1387 does not create any entries in
1391 It does record information in
1392 .B /var/run/mdadm/map
1395 to choose the correct name.
1399 detects that udev is not configured, it will create the devices in
1403 In Linux kernels prior to version 2.6.28 there were two distinctly
1404 different types of md devices that could be created: one that could be
1405 partitioned using standard partitioning tools and one that could not.
1406 Since 2.6.28 that distinction is no longer relevant as both type of
1407 devices can be partitioned.
1409 will normally create the type that originally could not be partitioned
1410 as it has a well defined major number (9).
1412 Prior to 2.6.28, it is important that mdadm chooses the correct type
1413 of array device to use. This can be controlled with the
1415 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1416 to use a partitionable device rather than the default.
1418 In the no-udev case, the value given to
1420 can be suffixed by a number. This tells
1422 to create that number of partition devices rather than the default of 4.
1426 can also be given in the configuration file as a word starting
1428 on the ARRAY line for the relevant array.
1435 and no devices are listed,
1437 will first attempt to assemble all the arrays listed in the config
1440 In no array at listed in the config (other than those marked
1442 it will look through the available devices for possible arrays and
1443 will try to assemble anything that it finds. Arrays which are tagged
1444 as belonging to the given homehost will be assembled and started
1445 normally. Arrays which do not obviously belong to this host are given
1446 names that are expected not to conflict with anything local, and are
1447 started "read-auto" so that nothing is written to any device until the
1448 array is written to. i.e. automatic resync etc is delayed.
1452 finds a consistent set of devices that look like they should comprise
1453 an array, and if the superblock is tagged as belonging to the given
1454 home host, it will automatically choose a device name and try to
1455 assemble the array. If the array uses version-0.90 metadata, then the
1457 number as recorded in the superblock is used to create a name in
1461 If the array uses version-1 metadata, then the
1463 from the superblock is used to similarly create a name in
1465 (the name will have any 'host' prefix stripped first).
1470 cannot find any array for the given host at all, and if
1471 .B \-\-auto\-update\-homehost
1474 will search again for any array (not just an array created for this
1475 host) and will assemble each assuming
1476 .BR \-\-update=homehost .
1477 This will change the host tag in the superblock so that on the next run,
1478 these arrays will be found without the second pass. The intention of
1479 this feature is to support transitioning a set of md arrays to using
1482 The reason for requiring arrays to be tagged with the homehost for
1483 auto assembly is to guard against problems that can arise when moving
1484 devices from one host to another.
1495 .BI \-\-raid\-devices= Z
1499 This usage is similar to
1501 The difference is that it creates an array without a superblock. With
1502 these arrays there is no difference between initially creating the array and
1503 subsequently assembling the array, except that hopefully there is useful
1504 data there in the second case.
1506 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1507 one of their synonyms. All devices must be listed and the array will
1508 be started once complete. It will often be appropriate to use
1509 .B \-\-assume\-clean
1510 with levels raid1 or raid10.
1521 .BI \-\-raid\-devices= Z
1525 This usage will initialise a new md array, associate some devices with
1526 it, and activate the array.
1528 The named device will normally not exist when
1529 .I "mdadm \-\-create"
1530 is run, but will be created by
1532 once the array becomes active.
1534 As devices are added, they are checked to see if they contain RAID
1535 superblocks or filesystems. They are also checked to see if the variance in
1536 device size exceeds 1%.
1538 If any discrepancy is found, the array will not automatically be run, though
1541 can override this caution.
1543 To create a "degraded" array in which some devices are missing, simply
1544 give the word "\fBmissing\fP"
1545 in place of a device name. This will cause
1547 to leave the corresponding slot in the array empty.
1548 For a RAID4 or RAID5 array at most one slot can be
1549 "\fBmissing\fP"; for a RAID6 array at most two slots.
1550 For a RAID1 array, only one real device needs to be given. All of the
1554 When creating a RAID5 array,
1556 will automatically create a degraded array with an extra spare drive.
1557 This is because building the spare into a degraded array is in general
1558 faster than resyncing the parity on a non-degraded, but not clean,
1559 array. This feature can be overridden with the
1563 When creating an array with version-1 metadata a name for the array is
1565 If this is not given with the
1569 will choose a name based on the last component of the name of the
1570 device being created. So if
1572 is being created, then the name
1577 is being created, then the name
1581 When creating a partition based array, using
1583 with version-1.x metadata, the partition type should be set to
1585 (non fs-data). This type selection allows for greater precision since
1586 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
1587 might create problems in the event of array recovery through a live cdrom.
1589 A new array will normally get a randomly assigned 128bit UUID which is
1590 very likely to be unique. If you have a specific need, you can choose
1591 a UUID for the array by giving the
1593 option. Be warned that creating two arrays with the same UUID is a
1594 recipe for disaster. Also, using
1596 when creating a v0.90 array will silently override any
1601 .\"option is given, it is not necessary to list any component-devices in this command.
1602 .\"They can be added later, before a
1606 .\"is given, the apparent size of the smallest drive given is used.
1608 When creating an array within a
1611 can be given either the list of devices to use, or simply the name of
1612 the container. The former case gives control over which devices in
1613 the container will be used for the array. The latter case allows
1615 to automatically choose which devices to use based on how much spare
1618 The General Management options that are valid with
1623 insist on running the array even if some devices look like they might
1628 start the array readonly \(em not supported yet.
1635 .I options... devices...
1638 This usage will allow individual devices in an array to be failed,
1639 removed or added. It is possible to perform multiple operations with
1640 on command. For example:
1642 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
1648 and will then remove it from the array and finally add it back
1649 in as a spare. However only one md array can be affected by a single
1652 When a device is added to an active array, mdadm checks to see if it
1653 has metadata on it which suggests that it was recently a member of the
1654 array. If it does, it tried to "re-add" the device. If there have
1655 been no changes since the device was removed, or if the array has a
1656 write-intent bitmap which has recorded whatever changes there were,
1657 then the device will immediately become a full member of the array and
1658 those differences recorded in the bitmap will be resolved.
1668 MISC mode includes a number of distinct operations that
1669 operate on distinct devices. The operations are:
1672 The device is examined to see if it is
1673 (1) an active md array, or
1674 (2) a component of an md array.
1675 The information discovered is reported.
1679 The device should be an active md device.
1681 will display a detailed description of the array.
1685 will cause the output to be less detailed and the format to be
1686 suitable for inclusion in
1687 .BR /etc/mdadm.conf .
1690 will normally be 0 unless
1692 failed to get useful information about the device(s); however, if the
1694 option is given, then the exit status will be:
1698 The array is functioning normally.
1701 The array has at least one failed device.
1704 The array has multiple failed devices such that it is unusable.
1707 There was an error while trying to get information about the device.
1711 .B \-\-detail\-platform
1712 Print detail of the platform's RAID capabilities (firmware / hardware
1713 topology). If the metadata is specified with
1717 then the return status will be:
1721 metadata successfully enumerated its platform components on this system
1724 metadata is platform independent
1727 metadata failed to find its platform components on this system
1732 The device should be a component of an md array.
1734 will read the md superblock of the device and display the contents.
1739 is given, then multiple devices that are components of the one array
1740 are grouped together and reported in a single entry suitable
1742 .BR /etc/mdadm.conf .
1746 without listing any devices will cause all devices listed in the
1747 config file to be examined.
1751 The devices should be active md arrays which will be deactivated, as
1752 long as they are not currently in use.
1756 This will fully activate a partially assembled md array.
1760 This will mark an active array as read-only, providing that it is
1761 not currently being used.
1767 array back to being read/write.
1771 For all operations except
1774 will cause the operation to be applied to all arrays listed in
1779 causes all devices listed in the config file to be examined.
1785 .B mdadm \-\-monitor
1786 .I options... devices...
1791 to periodically poll a number of md arrays and to report on any events
1794 will never exit once it decides that there are arrays to be checked,
1795 so it should normally be run in the background.
1797 As well as reporting events,
1799 may move a spare drive from one array to another if they are in the
1802 and if the destination array has a failed drive but no spares.
1804 If any devices are listed on the command line,
1806 will only monitor those devices. Otherwise all arrays listed in the
1807 configuration file will be monitored. Further, if
1809 is given, then any other md devices that appear in
1811 will also be monitored.
1813 The result of monitoring the arrays is the generation of events.
1814 These events are passed to a separate program (if specified) and may
1815 be mailed to a given E-mail address.
1817 When passing events to a program, the program is run once for each event,
1818 and is given 2 or 3 command-line arguments: the first is the
1819 name of the event (see below), the second is the name of the
1820 md device which is affected, and the third is the name of a related
1821 device if relevant (such as a component device that has failed).
1825 is given, then a program or an E-mail address must be specified on the
1826 command line or in the config file. If neither are available, then
1828 will not monitor anything.
1832 will continue monitoring as long as something was found to monitor. If
1833 no program or email is given, then each event is reported to
1836 The different events are:
1840 .B DeviceDisappeared
1841 An md array which previously was configured appears to no longer be
1842 configured. (syslog priority: Critical)
1846 was told to monitor an array which is RAID0 or Linear, then it will
1848 .B DeviceDisappeared
1849 with the extra information
1851 This is because RAID0 and Linear do not support the device-failed,
1852 hot-spare and resync operations which are monitored.
1856 An md array started reconstruction. (syslog priority: Warning)
1862 is 20, 40, 60, or 80, this indicates that rebuild has passed that many
1863 percentage of the total. (syslog priority: Warning)
1867 An md array that was rebuilding, isn't any more, either because it
1868 finished normally or was aborted. (syslog priority: Warning)
1872 An active component device of an array has been marked as
1873 faulty. (syslog priority: Critical)
1877 A spare component device which was being rebuilt to replace a faulty
1878 device has failed. (syslog priority: Critical)
1882 A spare component device which was being rebuilt to replace a faulty
1883 device has been successfully rebuilt and has been made active.
1884 (syslog priority: Info)
1888 A new md array has been detected in the
1890 file. (syslog priority: Info)
1894 A newly noticed array appears to be degraded. This message is not
1897 notices a drive failure which causes degradation, but only when
1899 notices that an array is degraded when it first sees the array.
1900 (syslog priority: Critical)
1904 A spare drive has been moved from one array in a
1906 to another to allow a failed drive to be replaced.
1907 (syslog priority: Info)
1913 has been told, via the config file, that an array should have a certain
1914 number of spare devices, and
1916 detects that it has fewer than this number when it first sees the
1917 array, it will report a
1920 (syslog priority: Warning)
1924 An array was found at startup, and the
1927 (syslog priority: Info)
1937 cause Email to be sent. All events cause the program to be run.
1938 The program is run with two or three arguments: the event
1939 name, the array device and possibly a second device.
1941 Each event has an associated array device (e.g.
1943 and possibly a second device. For
1948 the second device is the relevant component device.
1951 the second device is the array that the spare was moved from.
1955 to move spares from one array to another, the different arrays need to
1956 be labeled with the same
1958 in the configuration file. The
1960 name can be any string; it is only necessary that different spare
1961 groups use different names.
1965 detects that an array in a spare group has fewer active
1966 devices than necessary for the complete array, and has no spare
1967 devices, it will look for another array in the same spare group that
1968 has a full complement of working drive and a spare. It will then
1969 attempt to remove the spare from the second drive and add it to the
1971 If the removal succeeds but the adding fails, then it is added back to
1975 The GROW mode is used for changing the size or shape of an active
1977 For this to work, the kernel must support the necessary change.
1978 Various types of growth are being added during 2.6 development,
1979 including restructuring a RAID5 array to have more active devices.
1981 Currently the only support available is to
1983 change the "size" attribute
1984 for RAID1, RAID5 and RAID6.
1986 increase or decrease the "raid\-devices" attribute of RAID1, RAID5,
1989 change the chunk-size and layout of RAID5 and RAID6.
1991 convert between RAID1 and RAID5, and between RAID5 and RAID6.
1993 add a write-intent bitmap to any array which supports these bitmaps, or
1994 remove a write-intent bitmap from such an array.
1997 GROW mode is not currently supported for
1999 or arrays inside containers.
2002 Normally when an array is built the "size" it taken from the smallest
2003 of the drives. If all the small drives in an arrays are, one at a
2004 time, removed and replaced with larger drives, then you could have an
2005 array of large drives with only a small amount used. In this
2006 situation, changing the "size" with "GROW" mode will allow the extra
2007 space to start being used. If the size is increased in this way, a
2008 "resync" process will start to make sure the new parts of the array
2011 Note that when an array changes size, any filesystem that may be
2012 stored in the array will not automatically grow to use the space. The
2013 filesystem will need to be explicitly told to use the extra space.
2015 Also the size of an array cannot be changed while it has an active
2016 bitmap. If an array has a bitmap, it must be removed before the size
2017 can be changed. Once the change it complete a new bitmap can be created.
2019 .SS RAID\-DEVICES CHANGES
2021 A RAID1 array can work with any number of devices from 1 upwards
2022 (though 1 is not very useful). There may be times which you want to
2023 increase or decrease the number of active devices. Note that this is
2024 different to hot-add or hot-remove which changes the number of
2027 When reducing the number of devices in a RAID1 array, the slots which
2028 are to be removed from the array must already be vacant. That is, the
2029 devices which were in those slots must be failed and removed.
2031 When the number of devices is increased, any hot spares that are
2032 present will be activated immediately.
2034 Changing the number of active devices in a RAID5 or RAID6 is much more
2035 effort. Every block in the array will need to be read and written
2036 back to a new location. From 2.6.17, the Linux Kernel is able to
2037 increase the number of devices in a RAID5 safely, including restarting
2038 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2039 increase or decrease the number of devices in a RAID5 or RAID6.
2041 When decreasing the number of devices, the size of the array will also
2042 decrease. If there was data in the array, it could get destroyed and
2043 this is not reversible. To help prevent accidents,
2045 requires that the size of the array be decreased first with
2046 .BR "mdadm --grow --array-size" .
2047 This is a reversible change which simply makes the end of the array
2048 inaccessible. The integrity of any data can then be checked before
2049 the non-reversible reduction in the number of devices is request.
2051 When relocating the first few stripes on a RAID5, it is not possible
2052 to keep the data on disk completely consistent and crash-proof. To
2053 provide the required safety, mdadm disables writes to the array while
2054 this "critical section" is reshaped, and takes a backup of the data
2055 that is in that section. This backup is normally stored in any spare
2056 devices that the array has, however it can also be stored in a
2057 separate file specified with the
2059 option. If this option is used, and the system does crash during the
2060 critical period, the same file must be passed to
2062 to restore the backup and reassemble the array.
2066 Changing the RAID level of any array happens instantaneously. However
2067 in the RAID to RAID6 case this requires a non-standard layout of the
2068 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2069 required before the change can be accomplish. So while the level
2070 change is instant, the accompanying layout change can take quite a
2073 .SS CHUNK-SIZE AND LAYOUT CHANGES
2075 Changing the chunk-size of layout without also changing the number of
2076 devices as the same time will involve re-writing all blocks in-place.
2077 To ensure against data loss in the case of a crash, a
2079 must be provided for these changes. Small sections of the array will
2080 be copied to the backup file while they are being rearranged.
2082 If the reshape is interrupted for any reason, this backup file must be
2084 .B "mdadm --assemble"
2085 so the array can be reassembled. Consequently the file cannot be
2086 stored on the device being reshaped.
2091 A write-intent bitmap can be added to, or removed from, an active
2092 array. Either internal bitmaps, or bitmaps stored in a separate file,
2093 can be added. Note that if you add a bitmap stored in a file which is
2094 in a filesystem that is on the RAID array being affected, the system
2095 will deadlock. The bitmap must be on a separate filesystem.
2097 .SH INCREMENTAL MODE
2101 .B mdadm \-\-incremental
2107 .B mdadm \-\-incremental \-\-rebuild
2110 .B mdadm \-\-incremental \-\-run \-\-scan
2113 This mode is designed to be used in conjunction with a device
2114 discovery system. As devices are found in a system, they can be
2116 .B "mdadm \-\-incremental"
2117 to be conditionally added to an appropriate array.
2119 If the device passed is a
2121 device created by a previous call to
2123 then rather than trying to add that device to an array, all the arrays
2124 described by the metadata of the container will be started.
2127 performs a number of tests to determine if the device is part of an
2128 array, and which array it should be part of. If an appropriate array
2129 is found, or can be created,
2131 adds the device to the array and conditionally starts the array.
2135 will only add devices to an array which were previously working
2136 (active or spare) parts of that array. It does not currently support
2137 automatic inclusion of a new drive as a spare in some array.
2141 makes are as follow:
2143 Is the device permitted by
2145 That is, is it listed in a
2147 line in that file. If
2149 is absent then the default it to allow any device. Similar if
2151 contains the special word
2153 then any device is allowed. Otherwise the device name given to
2155 must match one of the names or patterns in a
2160 Does the device have a valid md superblock. If a specific metadata
2161 version is request with
2165 then only that style of metadata is accepted, otherwise
2167 finds any known version of metadata. If no
2169 metadata is found, the device is rejected.
2173 Does the metadata match an expected array?
2174 The metadata can match in two ways. Either there is an array listed
2177 which identifies the array (either by UUID, by name, by device list,
2178 or by minor-number), or the array was created with a
2184 or on the command line.
2187 is not able to positively identify the array as belonging to the
2188 current host, the device will be rejected.
2192 keeps a list of arrays that it has partially assembled in
2193 .B /var/run/mdadm/map
2195 .B /var/run/mdadm.map
2196 if the directory doesn't exist. Or maybe even
2197 .BR /dev/.mdadm.map ).
2198 If no array exists which matches
2199 the metadata on the new device,
2201 must choose a device name and unit number. It does this based on any
2204 or any name information stored in the metadata. If this name
2205 suggests a unit number, that number will be used, otherwise a free
2206 unit number will be chosen. Normally
2208 will prefer to create a partitionable array, however if the
2212 suggests that a non-partitionable array is preferred, that will be
2215 If the array is not found in the config file and its metadata does not
2216 identify it as belonging to the "homehost", then
2218 will choose a name for the array which is certain not to conflict with
2219 any array which does belong to this host. It does this be adding an
2220 underscore and a small number to the name preferred by the metadata.
2222 Once an appropriate array is found or created and the device is added,
2224 must decide if the array is ready to be started. It will
2225 normally compare the number of available (non-spare) devices to the
2226 number of devices that the metadata suggests need to be active. If
2227 there are at least that many, the array will be started. This means
2228 that if any devices are missing the array will not be restarted.
2234 in which case the array will be run as soon as there are enough
2235 devices present for the data to be accessible. For a RAID1, that
2236 means one device will start the array. For a clean RAID5, the array
2237 will be started as soon as all but one drive is present.
2239 Note that neither of these approaches is really ideal. If it can
2240 be known that all device discovery has completed, then
2244 can be run which will try to start all arrays that are being
2245 incrementally assembled. They are started in "read-auto" mode in
2246 which they are read-only until the first write request. This means
2247 that no metadata updates are made and no attempt at resync or recovery
2248 happens. Further devices that are found before the first write can
2249 still be added safely.
2252 This section describes environment variables that affect how mdadm
2257 Setting this value to 1 will prevent mdadm from automatically launching
2258 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2264 does not create any device nodes in /dev, but leaves that task to
2268 appears not to be configured, or if this environment variable is set
2271 will create and devices that are needed.
2275 .B " mdadm \-\-query /dev/name-of-device"
2277 This will find out if a given device is a RAID array, or is part of
2278 one, and will provide brief information about the device.
2280 .B " mdadm \-\-assemble \-\-scan"
2282 This will assemble and start all arrays listed in the standard config
2283 file. This command will typically go in a system startup file.
2285 .B " mdadm \-\-stop \-\-scan"
2287 This will shut down all arrays that can be shut down (i.e. are not
2288 currently in use). This will typically go in a system shutdown script.
2290 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2292 If (and only if) there is an Email address or program given in the
2293 standard config file, then
2294 monitor the status of all arrays listed in that file by
2295 polling them ever 2 minutes.
2297 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2299 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2302 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2304 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2306 This will create a prototype config file that describes currently
2307 active arrays that are known to be made from partitions of IDE or SCSI drives.
2308 This file should be reviewed before being used as it may
2309 contain unwanted detail.
2311 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2313 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
2315 This will find arrays which could be assembled from existing IDE and
2316 SCSI whole drives (not partitions), and store the information in the
2317 format of a config file.
2318 This file is very likely to contain unwanted detail, particularly
2321 entries. It should be reviewed and edited before being used as an
2324 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
2326 .B " mdadm \-Ebsc partitions"
2328 Create a list of devices by reading
2329 .BR /proc/partitions ,
2330 scan these for RAID superblocks, and printout a brief listing of all
2333 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
2335 Scan all partitions and devices listed in
2336 .BR /proc/partitions
2339 out of all such devices with a RAID superblock with a minor number of 0.
2341 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /var/run/mdadm"
2343 If config file contains a mail address or alert program, run mdadm in
2344 the background in monitor mode monitoring all md devices. Also write
2345 pid of mdadm daemon to
2346 .BR /var/run/mdadm .
2348 .B " mdadm \-Iq /dev/somedevice"
2350 Try to incorporate newly discovered device into some array as
2353 .B " mdadm \-\-incremental \-\-rebuild \-\-run \-\-scan"
2355 Rebuild the array map from any current arrays, and then start any that
2358 .B " mdadm /dev/md4 --fail detached --remove detached"
2360 Any devices which are components of /dev/md4 will be marked as faulty
2361 and then remove from the array.
2363 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4
2367 which is currently a RAID5 array will be converted to RAID6. There
2368 should normally already be a spare drive attached to the array as a
2369 RAID6 needs one more drive than a matching RAID5.
2371 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
2373 Create a DDF array over 6 devices.
2375 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
2377 Create a RAID5 array over any 3 devices in the given DDF set. Use
2378 only 30 gigabytes of each device.
2380 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
2382 Assemble a pre-exist ddf array.
2384 .B " mdadm -I /dev/md/ddf1"
2386 Assemble all arrays contained in the ddf array, assigning names as
2389 .B " mdadm \-\-create \-\-help"
2391 Provide help about the Create mode.
2393 .B " mdadm \-\-config \-\-help"
2395 Provide help about the format of the config file.
2397 .B " mdadm \-\-help"
2399 Provide general help.
2409 lists all active md devices with information about them.
2411 uses this to find arrays when
2413 is given in Misc mode, and to monitor array reconstruction
2418 The config file lists which devices may be scanned to see if
2419 they contain MD super block, and gives identifying information
2420 (e.g. UUID) about known MD arrays. See
2424 .SS /var/run/mdadm/map
2427 mode is used, this file gets a list of arrays currently being created.
2430 does not exist as a directory, then
2431 .B /var/run/mdadm.map
2434 is not available (as may be the case during early boot),
2436 is used on the basis that
2438 is usually available very early in boot.
2443 understand two sorts of names for array devices.
2445 The first is the so-called 'standard' format name, which matches the
2446 names used by the kernel and which appear in
2449 The second sort can be freely chosen, but must reside in
2451 When giving a device name to
2453 to create or assemble an array, either full path name such as
2457 can be given, or just the suffix of the second sort of name, such as
2463 chooses device names during auto-assembly or incremental assembly, it
2464 will sometimes add a small sequence number to the end of the name to
2465 avoid conflicted between multiple arrays that have the same name. If
2467 can reasonably determine that the array really is meant for this host,
2468 either by a hostname in the metadata, or by the presence of the array
2469 in /etc/mdadm.conf, then it will leave off the suffix if possible.
2470 Also if the homehost is specified as
2473 will only use a suffix if a different array of the same name already
2474 exists or is listed in the config file.
2476 The standard names for non-partitioned arrays (the only sort of md
2477 array available in 2.4 and earlier) are of the form
2481 where NN is a number.
2482 The standard names for partitionable arrays (as available from 2.6
2483 onwards) are of the form
2487 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2489 From kernel version, 2.6.28 the "non-partitioned array" can actually
2490 be partitioned. So the "md_dNN" names are no longer needed, and
2491 partitions such as "/dev/mdNNpXX" are possible.
2495 was previously known as
2499 is completely separate from the
2501 package, and does not use the
2503 configuration file at all.
2506 For further information on mdadm usage, MD and the various levels of
2509 .B http://linux\-raid.osdl.org/
2511 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
2513 .\"for new releases of the RAID driver check out:
2516 .\".UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2517 .\"ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2522 .\".UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2523 .\"http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2526 The latest version of
2528 should always be available from
2530 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/