2 .\" Copyright Neil Brown and others.
3 .\" This program is free software; you can redistribute it and/or modify
4 .\" it under the terms of the GNU General Public License as published by
5 .\" the Free Software Foundation; either version 2 of the License, or
6 .\" (at your option) any later version.
7 .\" See file COPYING in distribution for details.
10 mdadm \- manage MD devices
16 .BI mdadm " [mode] <raiddevice> [options] <component-devices>"
19 RAID devices are virtual devices created from two or more
20 real block devices. This allows multiple devices (typically disk
21 drives or partitions thereof) to be combined into a single device to
22 hold (for example) a single filesystem.
23 Some RAID levels include redundancy and so can survive some degree of
26 Linux Software RAID devices are implemented through the md (Multiple
27 Devices) device driver.
29 Currently, Linux supports
46 is not a Software RAID mechanism, but does involve
48 each device is a path to one common physical storage device.
49 New installations should not use md/multipath as it is not well
50 supported and has no ongoing development. Use the Device Mapper based
51 multipath-tools instead.
54 is also not true RAID, and it only involves one device. It
55 provides a layer over a true device that can be used to inject faults.
60 is a collection of devices that are
61 managed as a set. This is similar to the set of devices connected to
62 a hardware RAID controller. The set of devices may contain a number
63 of different RAID arrays each utilising some (or all) of the blocks from a
64 number of the devices in the set. For example, two devices in a 5-device set
65 might form a RAID1 using the whole devices. The remaining three might
66 have a RAID5 over the first half of each device, and a RAID0 over the
71 there is one set of metadata that describes all of
72 the arrays in the container. So when
76 device, the device just represents the metadata. Other normal arrays (RAID1
77 etc) can be created inside the container.
80 mdadm has several major modes of operation:
83 Assemble the components of a previously created
84 array into an active array. Components can be explicitly given
85 or can be searched for.
87 checks that the components
88 do form a bona fide array, and can, on request, fiddle superblock
89 information so as to assemble a faulty array.
93 Build an array that doesn't have per-device metadata (superblocks). For these
96 cannot differentiate between initial creation and subsequent assembly
97 of an array. It also cannot perform any checks that appropriate
98 components have been requested. Because of this, the
100 mode should only be used together with a complete understanding of
105 Create a new array with per-device metadata (superblocks).
106 Appropriate metadata is written to each device, and then the array
107 comprising those devices is activated. A 'resync' process is started
108 to make sure that the array is consistent (e.g. both sides of a mirror
109 contain the same data) but the content of the device is left otherwise
111 The array can be used as soon as it has been created. There is no
112 need to wait for the initial resync to finish.
115 .B "Follow or Monitor"
116 Monitor one or more md devices and act on any state changes. This is
117 only meaningful for RAID1, 4, 5, 6, 10 or multipath arrays, as
118 only these have interesting state. RAID0 or Linear never have
119 missing, spare, or failed drives, so there is nothing to monitor.
123 Grow (or shrink) an array, or otherwise reshape it in some way.
124 Currently supported growth options including changing the active size
125 of component devices and changing the number of active devices in
126 Linear and RAID levels 0/1/4/5/6,
127 changing the RAID level between 0, 1, 5, and 6, and between 0 and 10,
128 changing the chunk size and layout for RAID 0,4,5,6,10 as well as adding or
129 removing a write-intent bitmap.
132 .B "Incremental Assembly"
133 Add a single device to an appropriate array. If the addition of the
134 device makes the array runnable, the array will be started.
135 This provides a convenient interface to a
137 system. As each device is detected,
139 has a chance to include it in some array as appropriate.
142 flag is passed in we will remove the device from any active array
143 instead of adding it.
149 in this mode, then any arrays within that container will be assembled
154 This is for doing things to specific components of an array such as
155 adding new spares and removing faulty devices.
159 This is an 'everything else' mode that supports operations on active
160 arrays, operations on component devices such as erasing old superblocks, and
161 information gathering operations.
162 .\"This mode allows operations on independent devices such as examine MD
163 .\"superblocks, erasing old superblocks and stopping active arrays.
167 This mode does not act on a specific device or array, but rather it
168 requests the Linux Kernel to activate any auto-detected arrays.
171 .SH Options for selecting a mode are:
174 .BR \-A ", " \-\-assemble
175 Assemble a pre-existing array.
178 .BR \-B ", " \-\-build
179 Build a legacy array without superblocks.
182 .BR \-C ", " \-\-create
186 .BR \-F ", " \-\-follow ", " \-\-monitor
192 .BR \-G ", " \-\-grow
193 Change the size or shape of an active array.
196 .BR \-I ", " \-\-incremental
197 Add/remove a single device to/from an appropriate array, and possibly start the array.
201 Request that the kernel starts any auto-detected arrays. This can only
204 is compiled into the kernel \(em not if it is a module.
205 Arrays can be auto-detected by the kernel if all the components are in
206 primary MS-DOS partitions with partition type
208 and all use v0.90 metadata.
209 In-kernel autodetect is not recommended for new installations. Using
211 to detect and assemble arrays \(em possibly in an
213 \(em is substantially more flexible and should be preferred.
216 If a device is given before any options, or if the first option is
225 then the MANAGE mode is assumed.
226 Anything other than these will cause the
230 .SH Options that are not mode-specific are:
233 .BR \-h ", " \-\-help
234 Display general help message or, after one of the above options, a
235 mode-specific help message.
239 Display more detailed help about command line parsing and some commonly
243 .BR \-V ", " \-\-version
244 Print version information for mdadm.
247 .BR \-v ", " \-\-verbose
248 Be more verbose about what is happening. This can be used twice to be
250 The extra verbosity currently only affects
251 .B \-\-detail \-\-scan
253 .BR "\-\-examine \-\-scan" .
256 .BR \-q ", " \-\-quiet
257 Avoid printing purely informative messages. With this,
259 will be silent unless there is something really important to report.
263 .BR \-f ", " \-\-force
264 Be more forceful about certain operations. See the various modes for
265 the exact meaning of this option in different contexts.
268 .BR \-c ", " \-\-config=
269 Specify the config file or directory. Default is to use
272 .BR /etc/mdadm.conf.d ,
273 or if those are missing then
274 .B /etc/mdadm/mdadm.conf
276 .BR /etc/mdadm/mdadm.conf.d .
277 If the config file given is
279 then nothing will be read, but
281 will act as though the config file contained exactly
283 .B " DEVICE partitions containers"
287 to find a list of devices to scan, and
289 to find a list of containers to examine.
292 is given for the config file, then
294 will act as though the config file were empty.
296 If the name given is of a directory, then
298 will collect all the files contained in the directory with a name ending
301 sort them lexically, and process all of those files as config files.
304 .BR \-s ", " \-\-scan
307 for missing information.
308 In general, this option gives
310 permission to get any missing information (like component devices,
311 array devices, array identities, and alert destination) from the
312 configuration file (see previous option);
313 one exception is MISC mode when using
319 says to get a list of array devices from
323 .BR \-e ", " \-\-metadata=
324 Declare the style of RAID metadata (superblock) to be used. The
325 default is {DEFAULT_METADATA} for
327 and to guess for other operations.
328 The default can be overridden by setting the
337 .ie '{DEFAULT_METADATA}'0.90'
338 .IP "0, 0.90, default"
341 Use the original 0.90 format superblock. This format limits arrays to
342 28 component devices and limits component devices of levels 1 and
343 greater to 2 terabytes. It is also possible for there to be confusion
344 about whether the superblock applies to a whole device or just the
345 last partition, if that partition starts on a 64K boundary.
346 .ie '{DEFAULT_METADATA}'0.90'
347 .IP "1, 1.0, 1.1, 1.2"
349 .IP "1, 1.0, 1.1, 1.2 default"
350 Use the new version-1 format superblock. This has fewer restrictions.
351 It can easily be moved between hosts with different endian-ness, and a
352 recovery operation can be checkpointed and restarted. The different
353 sub-versions store the superblock at different locations on the
354 device, either at the end (for 1.0), at the start (for 1.1) or 4K from
355 the start (for 1.2). "1" is equivalent to "1.2" (the commonly
356 preferred 1.x format).
357 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
359 Use the "Industry Standard" DDF (Disk Data Format) format defined by
361 When creating a DDF array a
363 will be created, and normal arrays can be created in that container.
365 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
367 which is managed in a similar manner to DDF, and is supported by an
368 option-rom on some platforms:
370 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
376 This will override any
378 setting in the config file and provides the identity of the host which
379 should be considered the home for any arrays.
381 When creating an array, the
383 will be recorded in the metadata. For version-1 superblocks, it will
384 be prefixed to the array name. For version-0.90 superblocks, part of
385 the SHA1 hash of the hostname will be stored in the later half of the
388 When reporting information about an array, any array which is tagged
389 for the given homehost will be reported as such.
391 When using Auto-Assemble, only arrays tagged for the given homehost
392 will be allowed to use 'local' names (i.e. not ending in '_' followed
393 by a digit string). See below under
394 .BR "Auto Assembly" .
400 needs to print the name for a device it normally finds the name in
402 which refers to the device and is shortest. When a path component is
406 will prefer a longer name if it contains that component. For example
407 .B \-\-prefer=by-uuid
408 will prefer a name in a subdirectory of
413 This functionality is currently only provided by
419 .B \-\-home\-cluster=
420 specifies the cluster name for the md device. The md device can be assembled
421 only on the cluster which matches the name specified. If this option is not
422 provided, mdadm tries to detect the cluster name automatically.
424 .SH For create, build, or grow:
427 .BR \-n ", " \-\-raid\-devices=
428 Specify the number of active devices in the array. This, plus the
429 number of spare devices (see below) must equal the number of
431 (including "\fBmissing\fP" devices)
432 that are listed on the command line for
434 Setting a value of 1 is probably
435 a mistake and so requires that
437 be specified first. A value of 1 will then be allowed for linear,
438 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
440 This number can only be changed using
442 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
443 the necessary support.
446 .BR \-x ", " \-\-spare\-devices=
447 Specify the number of spare (eXtra) devices in the initial array.
448 Spares can also be added
449 and removed later. The number of component devices listed
450 on the command line must equal the number of RAID devices plus the
451 number of spare devices.
454 .BR \-z ", " \-\-size=
455 Amount (in Kibibytes) of space to use from each drive in RAID levels 1/4/5/6.
456 This must be a multiple of the chunk size, and must leave about 128Kb
457 of space at the end of the drive for the RAID superblock.
458 If this is not specified
459 (as it normally is not) the smallest drive (or partition) sets the
460 size, though if there is a variance among the drives of greater than 1%, a warning is
463 A suffix of 'M' or 'G' can be given to indicate Megabytes or
464 Gigabytes respectively.
466 Sometimes a replacement drive can be a little smaller than the
467 original drives though this should be minimised by IDEMA standards.
468 Such a replacement drive will be rejected by
470 To guard against this it can be useful to set the initial size
471 slightly smaller than the smaller device with the aim that it will
472 still be larger than any replacement.
474 This value can be set with
476 for RAID level 1/4/5/6 though
478 based arrays such as those with IMSM metadata may not be able to
480 If the array was created with a size smaller than the currently
481 active drives, the extra space can be accessed using
483 The size can be given as
485 which means to choose the largest size that fits on all current drives.
487 Before reducing the size of the array (with
488 .BR "\-\-grow \-\-size=" )
489 you should make sure that space isn't needed. If the device holds a
490 filesystem, you would need to resize the filesystem to use less space.
492 After reducing the array size you should check that the data stored in
493 the device is still available. If the device holds a filesystem, then
494 an 'fsck' of the filesystem is a minimum requirement. If there are
495 problems the array can be made bigger again with no loss with another
496 .B "\-\-grow \-\-size="
499 This value cannot be used when creating a
501 such as with DDF and IMSM metadata, though it perfectly valid when
502 creating an array inside a container.
505 .BR \-Z ", " \-\-array\-size=
506 This is only meaningful with
508 and its effect is not persistent: when the array is stopped and
509 restarted the default array size will be restored.
511 Setting the array-size causes the array to appear smaller to programs
512 that access the data. This is particularly needed before reshaping an
513 array so that it will be smaller. As the reshape is not reversible,
514 but setting the size with
516 is, it is required that the array size is reduced as appropriate
517 before the number of devices in the array is reduced.
519 Before reducing the size of the array you should make sure that space
520 isn't needed. If the device holds a filesystem, you would need to
521 resize the filesystem to use less space.
523 After reducing the array size you should check that the data stored in
524 the device is still available. If the device holds a filesystem, then
525 an 'fsck' of the filesystem is a minimum requirement. If there are
526 problems the array can be made bigger again with no loss with another
527 .B "\-\-grow \-\-array\-size="
530 A suffix of 'M' or 'G' can be given to indicate Megabytes or
531 Gigabytes respectively.
534 restores the apparent size of the array to be whatever the real
535 amount of available space is.
538 .BR \-c ", " \-\-chunk=
539 Specify chunk size of kibibytes. The default when creating an
540 array is 512KB. To ensure compatibility with earlier versions, the
541 default when building an array with no persistent metadata is 64KB.
542 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
544 RAID4, RAID5, RAID6, and RAID10 require the chunk size to be a power
545 of 2. In any case it must be a multiple of 4KB.
547 A suffix of 'M' or 'G' can be given to indicate Megabytes or
548 Gigabytes respectively.
552 Specify rounding factor for a Linear array. The size of each
553 component will be rounded down to a multiple of this size.
554 This is a synonym for
556 but highlights the different meaning for Linear as compared to other
557 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
558 use, and is 0K (i.e. no rounding) in later kernels.
561 .BR \-l ", " \-\-level=
562 Set RAID level. When used with
564 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
565 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
566 Obviously some of these are synonymous.
570 metadata type is requested, only the
572 level is permitted, and it does not need to be explicitly given.
576 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
580 to change the RAID level in some cases. See LEVEL CHANGES below.
583 .BR \-p ", " \-\-layout=
584 This option configures the fine details of data layout for RAID5, RAID6,
585 and RAID10 arrays, and controls the failure modes for
588 The layout of the RAID5 parity block can be one of
589 .BR left\-asymmetric ,
590 .BR left\-symmetric ,
591 .BR right\-asymmetric ,
592 .BR right\-symmetric ,
593 .BR la ", " ra ", " ls ", " rs .
595 .BR left\-symmetric .
597 It is also possible to cause RAID5 to use a RAID4-like layout by
603 Finally for RAID5 there are DDF\-compatible layouts,
604 .BR ddf\-zero\-restart ,
605 .BR ddf\-N\-restart ,
607 .BR ddf\-N\-continue .
609 These same layouts are available for RAID6. There are also 4 layouts
610 that will provide an intermediate stage for converting between RAID5
611 and RAID6. These provide a layout which is identical to the
612 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
613 syndrome (the second 'parity' block used by RAID6) on the last device.
615 .BR left\-symmetric\-6 ,
616 .BR right\-symmetric\-6 ,
617 .BR left\-asymmetric\-6 ,
618 .BR right\-asymmetric\-6 ,
620 .BR parity\-first\-6 .
622 When setting the failure mode for level
625 .BR write\-transient ", " wt ,
626 .BR read\-transient ", " rt ,
627 .BR write\-persistent ", " wp ,
628 .BR read\-persistent ", " rp ,
630 .BR read\-fixable ", " rf ,
631 .BR clear ", " flush ", " none .
633 Each failure mode can be followed by a number, which is used as a period
634 between fault generation. Without a number, the fault is generated
635 once on the first relevant request. With a number, the fault will be
636 generated after that many requests, and will continue to be generated
637 every time the period elapses.
639 Multiple failure modes can be current simultaneously by using the
641 option to set subsequent failure modes.
643 "clear" or "none" will remove any pending or periodic failure modes,
644 and "flush" will clear any persistent faults.
646 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
647 by a small number. The default is 'n2'. The supported options are:
650 signals 'near' copies. Multiple copies of one data block are at
651 similar offsets in different devices.
654 signals 'offset' copies. Rather than the chunks being duplicated
655 within a stripe, whole stripes are duplicated but are rotated by one
656 device so duplicate blocks are on different devices. Thus subsequent
657 copies of a block are in the next drive, and are one chunk further
662 (multiple copies have very different offsets).
663 See md(4) for more detail about 'near', 'offset', and 'far'.
665 The number is the number of copies of each datablock. 2 is normal, 3
666 can be useful. This number can be at most equal to the number of
667 devices in the array. It does not need to divide evenly into that
668 number (e.g. it is perfectly legal to have an 'n2' layout for an array
669 with an odd number of devices).
671 When an array is converted between RAID5 and RAID6 an intermediate
672 RAID6 layout is used in which the second parity block (Q) is always on
673 the last device. To convert a RAID5 to RAID6 and leave it in this new
674 layout (which does not require re-striping) use
675 .BR \-\-layout=preserve .
676 This will try to avoid any restriping.
678 The converse of this is
679 .B \-\-layout=normalise
680 which will change a non-standard RAID6 layout into a more standard
687 (thus explaining the p of
691 .BR \-b ", " \-\-bitmap=
692 Specify a file to store a write-intent bitmap in. The file should not
695 is also given. The same file should be provided
696 when assembling the array. If the word
698 is given, then the bitmap is stored with the metadata on the array,
699 and so is replicated on all devices. If the word
703 mode, then any bitmap that is present is removed. If the word
705 is given, the array is created for a clustered environment. One bitmap
706 is created for each node as defined by the
708 parameter and are stored internally.
710 To help catch typing errors, the filename must contain at least one
711 slash ('/') if it is a real file (not 'internal' or 'none').
713 Note: external bitmaps are only known to work on ext2 and ext3.
714 Storing bitmap files on other filesystems may result in serious problems.
716 When creating an array on devices which are 100G or larger,
718 automatically adds an internal bitmap as it will usually be
719 beneficial. This can be suppressed with
720 .B "\-\-bitmap=none".
723 .BR \-\-bitmap\-chunk=
724 Set the chunksize of the bitmap. Each bit corresponds to that many
725 Kilobytes of storage.
726 When using a file based bitmap, the default is to use the smallest
727 size that is at-least 4 and requires no more than 2^21 chunks.
730 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
731 fit the bitmap into the available space.
733 A suffix of 'M' or 'G' can be given to indicate Megabytes or
734 Gigabytes respectively.
737 .BR \-W ", " \-\-write\-mostly
738 subsequent devices listed in a
743 command will be flagged as 'write-mostly'. This is valid for RAID1
744 only and means that the 'md' driver will avoid reading from these
745 devices if at all possible. This can be useful if mirroring over a
749 .BR \-\-write\-behind=
750 Specify that write-behind mode should be enabled (valid for RAID1
751 only). If an argument is specified, it will set the maximum number
752 of outstanding writes allowed. The default value is 256.
753 A write-intent bitmap is required in order to use write-behind
754 mode, and write-behind is only attempted on drives marked as
758 .BR \-\-assume\-clean
761 that the array pre-existed and is known to be clean. It can be useful
762 when trying to recover from a major failure as you can be sure that no
763 data will be affected unless you actually write to the array. It can
764 also be used when creating a RAID1 or RAID10 if you want to avoid the
765 initial resync, however this practice \(em while normally safe \(em is not
766 recommended. Use this only if you really know what you are doing.
768 When the devices that will be part of a new array were filled
769 with zeros before creation the operator knows the array is
770 actually clean. If that is the case, such as after running
771 badblocks, this argument can be used to tell mdadm the
772 facts the operator knows.
774 When an array is resized to a larger size with
775 .B "\-\-grow \-\-size="
776 the new space is normally resynced in that same way that the whole
777 array is resynced at creation. From Linux version 3.0,
779 can be used with that command to avoid the automatic resync.
782 .BR \-\-backup\-file=
785 is used to increase the number of raid-devices in a RAID5 or RAID6 if
786 there are no spare devices available, or to shrink, change RAID level
787 or layout. See the GROW MODE section below on RAID\-DEVICES CHANGES.
788 The file must be stored on a separate device, not on the RAID array
793 Arrays with 1.x metadata can leave a gap between the start of the
794 device and the start of array data. This gap can be used for various
795 metadata. The start of data is known as the
797 Normally an appropriate data offset is computed automatically.
798 However it can be useful to set it explicitly such as when re-creating
799 an array which was originally created using a different version of
801 which computed a different offset.
803 Setting the offset explicitly over-rides the default. The value given
804 is in Kilobytes unless an 'M' or 'G' suffix is given.
808 can also be used with
810 for some RAID levels (initially on RAID10). This allows the
811 data\-offset to be changed as part of the reshape process. When the
812 data offset is changed, no backup file is required as the difference
813 in offsets is used to provide the same functionality.
815 When the new offset is earlier than the old offset, the number of
816 devices in the array cannot shrink. When it is after the old offset,
817 the number of devices in the array cannot increase.
819 When creating an array,
823 In the case each member device is expected to have a offset appended
824 to the name, separated by a colon. This makes it possible to recreate
825 exactly an array which has varying data offsets (as can happen when
826 different versions of
828 are used to add different devices).
832 This option is complementary to the
833 .B \-\-freeze-reshape
834 option for assembly. It is needed when
836 operation is interrupted and it is not restarted automatically due to
837 .B \-\-freeze-reshape
838 usage during array assembly. This option is used together with
842 ) command and device for a pending reshape to be continued.
843 All parameters required for reshape continuation will be read from array metadata.
847 .BR \-\-backup\-file=
848 option to be set, continuation option will require to have exactly the same
849 backup file given as well.
851 Any other parameter passed together with
853 option will be ignored.
856 .BR \-N ", " \-\-name=
859 for the array. This is currently only effective when creating an
860 array with a version-1 superblock, or an array in a DDF container.
861 The name is a simple textual string that can be used to identify array
862 components when assembling. If name is needed but not specified, it
863 is taken from the basename of the device that is being created.
875 run the array, even if some of the components
876 appear to be active in another array or filesystem. Normally
878 will ask for confirmation before including such components in an
879 array. This option causes that question to be suppressed.
882 .BR \-f ", " \-\-force
885 accept the geometry and layout specified without question. Normally
887 will not allow creation of an array with only one device, and will try
888 to create a RAID5 array with one missing drive (as this makes the
889 initial resync work faster). With
892 will not try to be so clever.
895 .BR \-o ", " \-\-readonly
898 rather than read-write as normal. No writes will be allowed to the
899 array, and no resync, recovery, or reshape will be started.
902 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
903 Instruct mdadm how to create the device file if needed, possibly allocating
904 an unused minor number. "md" causes a non-partitionable array
905 to be used (though since Linux 2.6.28, these array devices are in fact
906 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
907 later) to be used. "yes" requires the named md device to have
908 a 'standard' format, and the type and minor number will be determined
909 from this. With mdadm 3.0, device creation is normally left up to
911 so this option is unlikely to be needed.
912 See DEVICE NAMES below.
914 The argument can also come immediately after
919 is not given on the command line or in the config file, then
925 is also given, then any
927 entries in the config file will override the
929 instruction given on the command line.
931 For partitionable arrays,
933 will create the device file for the whole array and for the first 4
934 partitions. A different number of partitions can be specified at the
935 end of this option (e.g.
937 If the device name ends with a digit, the partition names add a 'p',
939 .IR /dev/md/home1p3 .
940 If there is no trailing digit, then the partition names just have a
942 .IR /dev/md/scratch3 .
944 If the md device name is in a 'standard' format as described in DEVICE
945 NAMES, then it will be created, if necessary, with the appropriate
946 device number based on that name. If the device name is not in one of these
947 formats, then a unused device number will be allocated. The device
948 number will be considered unused if there is no active array for that
949 number, and there is no entry in /dev for that number and with a
950 non-standard name. Names that are not in 'standard' format are only
951 allowed in "/dev/md/".
953 This is meaningful with
959 .BR \-a ", " "\-\-add"
960 This option can be used in Grow mode in two cases.
962 If the target array is a Linear array, then
964 can be used to add one or more devices to the array. They
965 are simply catenated on to the end of the array. Once added, the
966 devices cannot be removed.
970 option is being used to increase the number of devices in an array,
973 can be used to add some extra devices to be included in the array.
974 In most cases this is not needed as the extra devices can be added as
975 spares first, and then the number of raid-disks can be changed.
976 However for RAID0, it is not possible to add spares. So to increase
977 the number of devices in a RAID0, it is necessary to set the new
978 number of devices, and to add the new devices, in the same command.
982 Only works when the array is for clustered environment. It specifies
983 the maximum number of nodes in the cluster that will use this device
984 simultaneously. If not specified, this defaults to 4.
989 .BR \-u ", " \-\-uuid=
990 uuid of array to assemble. Devices which don't have this uuid are
994 .BR \-m ", " \-\-super\-minor=
995 Minor number of device that array was created for. Devices which
996 don't have this minor number are excluded. If you create an array as
997 /dev/md1, then all superblocks will contain the minor number 1, even if
998 the array is later assembled as /dev/md2.
1000 Giving the literal word "dev" for
1004 to use the minor number of the md device that is being assembled.
1005 e.g. when assembling
1007 .B \-\-super\-minor=dev
1008 will look for super blocks with a minor number of 0.
1011 is only relevant for v0.90 metadata, and should not normally be used.
1017 .BR \-N ", " \-\-name=
1018 Specify the name of the array to assemble. This must be the name
1019 that was specified when creating the array. It must either match
1020 the name stored in the superblock exactly, or it must match
1023 prefixed to the start of the given name.
1026 .BR \-f ", " \-\-force
1027 Assemble the array even if the metadata on some devices appears to be
1030 cannot find enough working devices to start the array, but can find
1031 some devices that are recorded as having failed, then it will mark
1032 those devices as working so that the array can be started.
1033 An array which requires
1035 to be started may contain data corruption. Use it carefully.
1038 .BR \-R ", " \-\-run
1039 Attempt to start the array even if fewer drives were given than were
1040 present last time the array was active. Normally if not all the
1041 expected drives are found and
1043 is not used, then the array will be assembled but not started.
1046 an attempt will be made to start it anyway.
1050 This is the reverse of
1052 in that it inhibits the startup of array unless all expected drives
1053 are present. This is only needed with
1055 and can be used if the physical connections to devices are
1056 not as reliable as you would like.
1059 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
1060 See this option under Create and Build options.
1063 .BR \-b ", " \-\-bitmap=
1064 Specify the bitmap file that was given when the array was created. If
1067 bitmap, there is no need to specify this when assembling the array.
1070 .BR \-\-backup\-file=
1073 was used while reshaping an array (e.g. changing number of devices or
1074 chunk size) and the system crashed during the critical section, then the same
1076 must be presented to
1078 to allow possibly corrupted data to be restored, and the reshape
1082 .BR \-\-invalid\-backup
1083 If the file needed for the above option is not available for any
1084 reason an empty file can be given together with this option to
1085 indicate that the backup file is invalid. In this case the data that
1086 was being rearranged at the time of the crash could be irrecoverably
1087 lost, but the rest of the array may still be recoverable. This option
1088 should only be used as a last resort if there is no way to recover the
1093 .BR \-U ", " \-\-update=
1094 Update the superblock on each device while assembling the array. The
1095 argument given to this flag can be one of
1113 option will adjust the superblock of an array what was created on a Sparc
1114 machine running a patched 2.2 Linux kernel. This kernel got the
1115 alignment of part of the superblock wrong. You can use the
1116 .B "\-\-examine \-\-sparc2.2"
1119 to see what effect this would have.
1123 option will update the
1124 .B "preferred minor"
1125 field on each superblock to match the minor number of the array being
1127 This can be useful if
1129 reports a different "Preferred Minor" to
1131 In some cases this update will be performed automatically
1132 by the kernel driver. In particular the update happens automatically
1133 at the first write to an array with redundancy (RAID level 1 or
1134 greater) on a 2.6 (or later) kernel.
1138 option will change the uuid of the array. If a UUID is given with the
1140 option that UUID will be used as a new UUID and will
1142 be used to help identify the devices in the array.
1145 is given, a random UUID is chosen.
1149 option will change the
1151 of the array as stored in the superblock. This is only supported for
1152 version-1 superblocks.
1156 option will change the
1158 as recorded in the superblock. For version-0 superblocks, this is the
1159 same as updating the UUID.
1160 For version-1 superblocks, this involves updating the name.
1164 option will cause the array to be marked
1166 meaning that any redundancy in the array (e.g. parity for RAID5,
1167 copies for RAID1) may be incorrect. This will cause the RAID system
1168 to perform a "resync" pass to make sure that all redundant information
1173 option allows arrays to be moved between machines with different
1175 When assembling such an array for the first time after a move, giving
1176 .B "\-\-update=byteorder"
1179 to expect superblocks to have their byteorder reversed, and will
1180 correct that order before assembling the array. This is only valid
1181 with original (Version 0.90) superblocks.
1185 option will correct the summaries in the superblock. That is the
1186 counts of total, working, active, failed, and spare devices.
1190 option will rarely be of use. It applies to version 1.1 and 1.2 metadata
1191 only (where the metadata is at the start of the device) and is only
1192 useful when the component device has changed size (typically become
1193 larger). The version 1 metadata records the amount of the device that
1194 can be used to store data, so if a device in a version 1.1 or 1.2
1195 array becomes larger, the metadata will still be visible, but the
1196 extra space will not. In this case it might be useful to assemble the
1198 .BR \-\-update=devicesize .
1201 to determine the maximum usable amount of space on each device and
1202 update the relevant field in the metadata.
1206 option only works on v0.90 metadata arrays and will convert them to
1207 v1.0 metadata. The array must not be dirty (i.e. it must not need a
1208 sync) and it must not have a write-intent bitmap.
1210 The old metadata will remain on the devices, but will appear older
1211 than the new metadata and so will usually be ignored. The old metadata
1212 (or indeed the new metadata) can be removed by giving the appropriate
1215 .BR \-\-zero\-superblock .
1219 option can be used when an array has an internal bitmap which is
1220 corrupt in some way so that assembling the array normally fails. It
1221 will cause any internal bitmap to be ignored.
1225 option will reserve space in each device for a bad block list. This
1226 will be 4K in size and positioned near the end of any free space
1227 between the superblock and the data.
1231 option will cause any reservation of space for a bad block list to be
1232 removed. If the bad block list contains entries, this will fail, as
1233 removing the list could cause data corruption.
1236 .BR \-\-freeze\-reshape
1237 Option is intended to be used in start-up scripts during initrd boot phase.
1238 When array under reshape is assembled during initrd phase, this option
1239 stops reshape after reshape critical section is being restored. This happens
1240 before file system pivot operation and avoids loss of file system context.
1241 Losing file system context would cause reshape to be broken.
1243 Reshape can be continued later using the
1245 option for the grow command.
1247 .SH For Manage mode:
1250 .BR \-t ", " \-\-test
1251 Unless a more serious error occurred,
1253 will exit with a status of 2 if no changes were made to the array and
1254 0 if at least one change was made.
1255 This can be useful when an indirect specifier such as
1260 is used in requesting an operation on the array.
1262 will report failure if these specifiers didn't find any match.
1265 .BR \-a ", " \-\-add
1266 hot-add listed devices.
1267 If a device appears to have recently been part of the array
1268 (possibly it failed or was removed) the device is re\-added as described
1270 If that fails or the device was never part of the array, the device is
1271 added as a hot-spare.
1272 If the array is degraded, it will immediately start to rebuild data
1275 Note that this and the following options are only meaningful on array
1276 with redundancy. They don't apply to RAID0 or Linear.
1280 re\-add a device that was previously removed from an array.
1281 If the metadata on the device reports that it is a member of the
1282 array, and the slot that it used is still vacant, then the device will
1283 be added back to the array in the same position. This will normally
1284 cause the data for that device to be recovered. However based on the
1285 event count on the device, the recovery may only require sections that
1286 are flagged a write-intent bitmap to be recovered or may not require
1287 any recovery at all.
1289 When used on an array that has no metadata (i.e. it was built with
1291 it will be assumed that bitmap-based recovery is enough to make the
1292 device fully consistent with the array.
1294 When used with v1.x metadata,
1296 can be accompanied by
1297 .BR \-\-update=devicesize ,
1298 .BR \-\-update=bbl ", or"
1299 .BR \-\-update=no\-bbl .
1300 See the description of these option when used in Assemble mode for an
1301 explanation of their use.
1303 If the device name given is
1307 will try to find any device that looks like it should be
1308 part of the array but isn't and will try to re\-add all such devices.
1310 If the device name given is
1314 will find all devices in the array that are marked
1316 remove them and attempt to immediately re\-add them. This can be
1317 useful if you are certain that the reason for failure has been
1322 Add a device as a spare. This is similar to
1324 except that it does not attempt
1326 first. The device will be added as a spare even if it looks like it
1327 could be an recent member of the array.
1330 .BR \-r ", " \-\-remove
1331 remove listed devices. They must not be active. i.e. they should
1332 be failed or spare devices.
1334 As well as the name of a device file
1344 The first causes all failed device to be removed. The second causes
1345 any device which is no longer connected to the system (i.e an 'open'
1349 The third will remove a set as describe below under
1353 .BR \-f ", " \-\-fail
1354 Mark listed devices as faulty.
1355 As well as the name of a device file, the word
1359 can be given. The former will cause any device that has been detached from
1360 the system to be marked as failed. It can then be removed.
1362 For RAID10 arrays where the number of copies evenly divides the number
1363 of devices, the devices can be conceptually divided into sets where
1364 each set contains a single complete copy of the data on the array.
1365 Sometimes a RAID10 array will be configured so that these sets are on
1366 separate controllers. In this case all the devices in one set can be
1367 failed by giving a name like
1373 The appropriate set names are reported by
1383 Mark listed devices as requiring replacement. As soon as a spare is
1384 available, it will be rebuilt and will replace the marked device.
1385 This is similar to marking a device as faulty, but the device remains
1386 in service during the recovery process to increase resilience against
1387 multiple failures. When the replacement process finishes, the
1388 replaced device will be marked as faulty.
1392 This can follow a list of
1394 devices. The devices listed after
1396 will be preferentially used to replace the devices listed after
1398 These device must already be spare devices in the array.
1401 .BR \-\-write\-mostly
1402 Subsequent devices that are added or re\-added will have the 'write-mostly'
1403 flag set. This is only valid for RAID1 and means that the 'md' driver
1404 will avoid reading from these devices if possible.
1407 Subsequent devices that are added or re\-added will have the 'write-mostly'
1411 Each of these options requires that the first device listed is the array
1412 to be acted upon, and the remainder are component devices to be added,
1413 removed, marked as faulty, etc. Several different operations can be
1414 specified for different devices, e.g.
1416 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1418 Each operation applies to all devices listed until the next
1421 If an array is using a write-intent bitmap, then devices which have
1422 been removed can be re\-added in a way that avoids a full
1423 reconstruction but instead just updates the blocks that have changed
1424 since the device was removed. For arrays with persistent metadata
1425 (superblocks) this is done automatically. For arrays created with
1427 mdadm needs to be told that this device we removed recently with
1430 Devices can only be removed from an array if they are not in active
1431 use, i.e. that must be spares or failed devices. To remove an active
1432 device, it must first be marked as
1438 .BR \-Q ", " \-\-query
1439 Examine a device to see
1440 (1) if it is an md device and (2) if it is a component of an md
1442 Information about what is discovered is presented.
1445 .BR \-D ", " \-\-detail
1446 Print details of one or more md devices.
1449 .BR \-\-detail\-platform
1450 Print details of the platform's RAID capabilities (firmware / hardware
1451 topology) for a given metadata format. If used without argument, mdadm
1452 will scan all controllers looking for their capabilities. Otherwise, mdadm
1453 will only look at the controller specified by the argument in form of an
1454 absolute filepath or a link, e.g.
1455 .IR /sys/devices/pci0000:00/0000:00:1f.2 .
1458 .BR \-Y ", " \-\-export
1461 .BR \-\-detail-platform ,
1465 output will be formatted as
1467 pairs for easy import into the environment.
1473 indicates whether an array was started
1475 or not, which may include a reason
1476 .RB ( unsafe ", " nothing ", " no ).
1479 indicates if the array is expected on this host
1481 or seems to be from elsewhere
1485 .BR \-E ", " \-\-examine
1486 Print contents of the metadata stored on the named device(s).
1487 Note the contrast between
1492 applies to devices which are components of an array, while
1494 applies to a whole array which is currently active.
1497 If an array was created on a SPARC machine with a 2.2 Linux kernel
1498 patched with RAID support, the superblock will have been created
1499 incorrectly, or at least incompatibly with 2.4 and later kernels.
1504 will fix the superblock before displaying it. If this appears to do
1505 the right thing, then the array can be successfully assembled using
1506 .BR "\-\-assemble \-\-update=sparc2.2" .
1509 .BR \-X ", " \-\-examine\-bitmap
1510 Report information about a bitmap file.
1511 The argument is either an external bitmap file or an array component
1512 in case of an internal bitmap. Note that running this on an array
1515 does not report the bitmap for that array.
1518 .B \-\-examine\-badblocks
1519 List the bad-blocks recorded for the device, if a bad-blocks list has
1520 been configured. Currently only
1522 metadata supports bad-blocks lists.
1525 .BI \-\-dump= directory
1527 .BI \-\-restore= directory
1528 Save metadata from lists devices, or restore metadata to listed devices.
1531 .BR \-R ", " \-\-run
1532 start a partially assembled array. If
1534 did not find enough devices to fully start the array, it might leaving
1535 it partially assembled. If you wish, you can then use
1537 to start the array in degraded mode.
1540 .BR \-S ", " \-\-stop
1541 deactivate array, releasing all resources.
1544 .BR \-o ", " \-\-readonly
1545 mark array as readonly.
1548 .BR \-w ", " \-\-readwrite
1549 mark array as readwrite.
1552 .B \-\-zero\-superblock
1553 If the device contains a valid md superblock, the block is
1554 overwritten with zeros. With
1556 the block where the superblock would be is overwritten even if it
1557 doesn't appear to be valid.
1560 .B \-\-kill\-subarray=
1561 If the device is a container and the argument to \-\-kill\-subarray
1562 specifies an inactive subarray in the container, then the subarray is
1563 deleted. Deleting all subarrays will leave an 'empty-container' or
1564 spare superblock on the drives. See
1565 .B \-\-zero\-superblock
1567 removing a superblock. Note that some formats depend on the subarray
1568 index for generating a UUID, this command will fail if it would change
1569 the UUID of an active subarray.
1572 .B \-\-update\-subarray=
1573 If the device is a container and the argument to \-\-update\-subarray
1574 specifies a subarray in the container, then attempt to update the given
1575 superblock field in the subarray. See below in
1580 .BR \-t ", " \-\-test
1585 is set to reflect the status of the device. See below in
1590 .BR \-W ", " \-\-wait
1591 For each md device given, wait for any resync, recovery, or reshape
1592 activity to finish before returning.
1594 will return with success if it actually waited for every device
1595 listed, otherwise it will return failure.
1599 For each md device given, or each device in /proc/mdstat if
1601 is given, arrange for the array to be marked clean as soon as possible.
1603 will return with success if the array uses external metadata and we
1604 successfully waited. For native arrays this returns immediately as the
1605 kernel handles dirty-clean transitions at shutdown. No action is taken
1606 if safe-mode handling is disabled.
1610 Set the "sync_action" for all md devices given to one of
1617 will abort any currently running action though some actions will
1618 automatically restart.
1621 will abort any current action and ensure no other action starts
1631 .BR "SCRUBBING AND MISMATCHES" .
1633 .SH For Incremental Assembly mode:
1635 .BR \-\-rebuild\-map ", " \-r
1636 Rebuild the map file
1640 uses to help track which arrays are currently being assembled.
1643 .BR \-\-run ", " \-R
1644 Run any array assembled as soon as a minimal number of devices are
1645 available, rather than waiting until all expected devices are present.
1648 .BR \-\-scan ", " \-s
1649 Only meaningful with
1653 file for arrays that are being incrementally assembled and will try to
1654 start any that are not already started. If any such array is listed
1657 as requiring an external bitmap, that bitmap will be attached first.
1660 .BR \-\-fail ", " \-f
1661 This allows the hot-plug system to remove devices that have fully disappeared
1662 from the kernel. It will first fail and then remove the device from any
1663 array it belongs to.
1664 The device name given should be a kernel device name such as "sda",
1670 Only used with \-\-fail. The 'path' given will be recorded so that if
1671 a new device appears at the same location it can be automatically
1672 added to the same array. This allows the failed device to be
1673 automatically replaced by a new device without metadata if it appears
1674 at specified path. This option is normally only set by a
1678 .SH For Monitor mode:
1680 .BR \-m ", " \-\-mail
1681 Give a mail address to send alerts to.
1684 .BR \-p ", " \-\-program ", " \-\-alert
1685 Give a program to be run whenever an event is detected.
1688 .BR \-y ", " \-\-syslog
1689 Cause all events to be reported through 'syslog'. The messages have
1690 facility of 'daemon' and varying priorities.
1693 .BR \-d ", " \-\-delay
1694 Give a delay in seconds.
1696 polls the md arrays and then waits this many seconds before polling
1697 again. The default is 60 seconds. Since 2.6.16, there is no need to
1698 reduce this as the kernel alerts
1700 immediately when there is any change.
1703 .BR \-r ", " \-\-increment
1704 Give a percentage increment.
1706 will generate RebuildNN events with the given percentage increment.
1709 .BR \-f ", " \-\-daemonise
1712 to run as a background daemon if it decides to monitor anything. This
1713 causes it to fork and run in the child, and to disconnect from the
1714 terminal. The process id of the child is written to stdout.
1717 which will only continue monitoring if a mail address or alert program
1718 is found in the config file.
1721 .BR \-i ", " \-\-pid\-file
1724 is running in daemon mode, write the pid of the daemon process to
1725 the specified file, instead of printing it on standard output.
1728 .BR \-1 ", " \-\-oneshot
1729 Check arrays only once. This will generate
1731 events and more significantly
1737 .B " mdadm \-\-monitor \-\-scan \-1"
1739 from a cron script will ensure regular notification of any degraded arrays.
1742 .BR \-t ", " \-\-test
1745 alert for every array found at startup. This alert gets mailed and
1746 passed to the alert program. This can be used for testing that alert
1747 message do get through successfully.
1751 This inhibits the functionality for moving spares between arrays.
1752 Only one monitoring process started with
1754 but without this flag is allowed, otherwise the two could interfere
1761 .B mdadm \-\-assemble
1762 .I md-device options-and-component-devices...
1765 .B mdadm \-\-assemble \-\-scan
1766 .I md-devices-and-options...
1769 .B mdadm \-\-assemble \-\-scan
1773 This usage assembles one or more RAID arrays from pre-existing components.
1774 For each array, mdadm needs to know the md device, the identity of the
1775 array, and a number of component-devices. These can be found in a number of ways.
1777 In the first usage example (without the
1779 the first device given is the md device.
1780 In the second usage example, all devices listed are treated as md
1781 devices and assembly is attempted.
1782 In the third (where no devices are listed) all md devices that are
1783 listed in the configuration file are assembled. If no arrays are
1784 described by the configuration file, then any arrays that
1785 can be found on unused devices will be assembled.
1787 If precisely one device is listed, but
1793 was given and identity information is extracted from the configuration file.
1795 The identity can be given with the
1801 option, will be taken from the md-device record in the config file, or
1802 will be taken from the super block of the first component-device
1803 listed on the command line.
1805 Devices can be given on the
1807 command line or in the config file. Only devices which have an md
1808 superblock which contains the right identity will be considered for
1811 The config file is only used if explicitly named with
1813 or requested with (a possibly implicit)
1818 .B /etc/mdadm/mdadm.conf
1823 is not given, then the config file will only be used to find the
1824 identity of md arrays.
1826 Normally the array will be started after it is assembled. However if
1828 is not given and not all expected drives were listed, then the array
1829 is not started (to guard against usage errors). To insist that the
1830 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1839 does not create any entries in
1843 It does record information in
1847 to choose the correct name.
1851 detects that udev is not configured, it will create the devices in
1855 In Linux kernels prior to version 2.6.28 there were two distinctly
1856 different types of md devices that could be created: one that could be
1857 partitioned using standard partitioning tools and one that could not.
1858 Since 2.6.28 that distinction is no longer relevant as both type of
1859 devices can be partitioned.
1861 will normally create the type that originally could not be partitioned
1862 as it has a well defined major number (9).
1864 Prior to 2.6.28, it is important that mdadm chooses the correct type
1865 of array device to use. This can be controlled with the
1867 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1868 to use a partitionable device rather than the default.
1870 In the no-udev case, the value given to
1872 can be suffixed by a number. This tells
1874 to create that number of partition devices rather than the default of 4.
1878 can also be given in the configuration file as a word starting
1880 on the ARRAY line for the relevant array.
1887 and no devices are listed,
1889 will first attempt to assemble all the arrays listed in the config
1892 If no arrays are listed in the config (other than those marked
1894 it will look through the available devices for possible arrays and
1895 will try to assemble anything that it finds. Arrays which are tagged
1896 as belonging to the given homehost will be assembled and started
1897 normally. Arrays which do not obviously belong to this host are given
1898 names that are expected not to conflict with anything local, and are
1899 started "read-auto" so that nothing is written to any device until the
1900 array is written to. i.e. automatic resync etc is delayed.
1904 finds a consistent set of devices that look like they should comprise
1905 an array, and if the superblock is tagged as belonging to the given
1906 home host, it will automatically choose a device name and try to
1907 assemble the array. If the array uses version-0.90 metadata, then the
1909 number as recorded in the superblock is used to create a name in
1913 If the array uses version-1 metadata, then the
1915 from the superblock is used to similarly create a name in
1917 (the name will have any 'host' prefix stripped first).
1919 This behaviour can be modified by the
1923 configuration file. This line can indicate that specific metadata
1924 type should, or should not, be automatically assembled. If an array
1925 is found which is not listed in
1927 and has a metadata format that is denied by the
1929 line, then it will not be assembled.
1932 line can also request that all arrays identified as being for this
1933 homehost should be assembled regardless of their metadata type.
1936 for further details.
1938 Note: Auto assembly cannot be used for assembling and activating some
1939 arrays which are undergoing reshape. In particular as the
1941 cannot be given, any reshape which requires a backup-file to continue
1942 cannot be started by auto assembly. An array which is growing to more
1943 devices and has passed the critical section can be assembled using
1954 .BI \-\-raid\-devices= Z
1958 This usage is similar to
1960 The difference is that it creates an array without a superblock. With
1961 these arrays there is no difference between initially creating the array and
1962 subsequently assembling the array, except that hopefully there is useful
1963 data there in the second case.
1965 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1966 one of their synonyms. All devices must be listed and the array will
1967 be started once complete. It will often be appropriate to use
1968 .B \-\-assume\-clean
1969 with levels raid1 or raid10.
1980 .BI \-\-raid\-devices= Z
1984 This usage will initialise a new md array, associate some devices with
1985 it, and activate the array.
1987 The named device will normally not exist when
1988 .I "mdadm \-\-create"
1989 is run, but will be created by
1991 once the array becomes active.
1993 As devices are added, they are checked to see if they contain RAID
1994 superblocks or filesystems. They are also checked to see if the variance in
1995 device size exceeds 1%.
1997 If any discrepancy is found, the array will not automatically be run, though
2000 can override this caution.
2002 To create a "degraded" array in which some devices are missing, simply
2003 give the word "\fBmissing\fP"
2004 in place of a device name. This will cause
2006 to leave the corresponding slot in the array empty.
2007 For a RAID4 or RAID5 array at most one slot can be
2008 "\fBmissing\fP"; for a RAID6 array at most two slots.
2009 For a RAID1 array, only one real device needs to be given. All of the
2013 When creating a RAID5 array,
2015 will automatically create a degraded array with an extra spare drive.
2016 This is because building the spare into a degraded array is in general
2017 faster than resyncing the parity on a non-degraded, but not clean,
2018 array. This feature can be overridden with the
2022 When creating an array with version-1 metadata a name for the array is
2024 If this is not given with the
2028 will choose a name based on the last component of the name of the
2029 device being created. So if
2031 is being created, then the name
2036 is being created, then the name
2040 When creating a partition based array, using
2042 with version-1.x metadata, the partition type should be set to
2044 (non fs-data). This type selection allows for greater precision since
2045 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
2046 might create problems in the event of array recovery through a live cdrom.
2048 A new array will normally get a randomly assigned 128bit UUID which is
2049 very likely to be unique. If you have a specific need, you can choose
2050 a UUID for the array by giving the
2052 option. Be warned that creating two arrays with the same UUID is a
2053 recipe for disaster. Also, using
2055 when creating a v0.90 array will silently override any
2060 .\"option is given, it is not necessary to list any component-devices in this command.
2061 .\"They can be added later, before a
2065 .\"is given, the apparent size of the smallest drive given is used.
2067 If the array type supports a write-intent bitmap, and if the devices
2068 in the array exceed 100G is size, an internal write-intent bitmap
2069 will automatically be added unless some other option is explicitly
2072 option. In any case space for a bitmap will be reserved so that one
2073 can be added layer with
2074 .BR "\-\-grow \-\-bitmap=internal" .
2076 If the metadata type supports it (currently only 1.x metadata), space
2077 will be allocated to store a bad block list. This allows a modest
2078 number of bad blocks to be recorded, allowing the drive to remain in
2079 service while only partially functional.
2081 When creating an array within a
2084 can be given either the list of devices to use, or simply the name of
2085 the container. The former case gives control over which devices in
2086 the container will be used for the array. The latter case allows
2088 to automatically choose which devices to use based on how much spare
2091 The General Management options that are valid with
2096 insist on running the array even if some devices look like they might
2101 start the array readonly \(em not supported yet.
2108 .I options... devices...
2111 This usage will allow individual devices in an array to be failed,
2112 removed or added. It is possible to perform multiple operations with
2113 on command. For example:
2115 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
2121 and will then remove it from the array and finally add it back
2122 in as a spare. However only one md array can be affected by a single
2125 When a device is added to an active array, mdadm checks to see if it
2126 has metadata on it which suggests that it was recently a member of the
2127 array. If it does, it tries to "re\-add" the device. If there have
2128 been no changes since the device was removed, or if the array has a
2129 write-intent bitmap which has recorded whatever changes there were,
2130 then the device will immediately become a full member of the array and
2131 those differences recorded in the bitmap will be resolved.
2141 MISC mode includes a number of distinct operations that
2142 operate on distinct devices. The operations are:
2145 The device is examined to see if it is
2146 (1) an active md array, or
2147 (2) a component of an md array.
2148 The information discovered is reported.
2152 The device should be an active md device.
2154 will display a detailed description of the array.
2158 will cause the output to be less detailed and the format to be
2159 suitable for inclusion in
2163 will normally be 0 unless
2165 failed to get useful information about the device(s); however, if the
2167 option is given, then the exit status will be:
2171 The array is functioning normally.
2174 The array has at least one failed device.
2177 The array has multiple failed devices such that it is unusable.
2180 There was an error while trying to get information about the device.
2184 .B \-\-detail\-platform
2185 Print detail of the platform's RAID capabilities (firmware / hardware
2186 topology). If the metadata is specified with
2190 then the return status will be:
2194 metadata successfully enumerated its platform components on this system
2197 metadata is platform independent
2200 metadata failed to find its platform components on this system
2204 .B \-\-update\-subarray=
2205 If the device is a container and the argument to \-\-update\-subarray
2206 specifies a subarray in the container, then attempt to update the given
2207 superblock field in the subarray. Similar to updating an array in
2208 "assemble" mode, the field to update is selected by
2212 option. Currently only
2218 option updates the subarray name in the metadata, it may not affect the
2219 device node name or the device node symlink until the subarray is
2220 re\-assembled. If updating
2222 would change the UUID of an active subarray this operation is blocked,
2223 and the command will end in an error.
2227 The device should be a component of an md array.
2229 will read the md superblock of the device and display the contents.
2234 is given, then multiple devices that are components of the one array
2235 are grouped together and reported in a single entry suitable
2241 without listing any devices will cause all devices listed in the
2242 config file to be examined.
2245 .BI \-\-dump= directory
2246 If the device contains RAID metadata, a file will be created in the
2248 and the metadata will be written to it. The file will be the same
2249 size as the device and have the metadata written in the file at the
2250 same locate that it exists in the device. However the file will be "sparse" so
2251 that only those blocks containing metadata will be allocated. The
2252 total space used will be small.
2254 The file name used in the
2256 will be the base name of the device. Further if any links appear in
2258 which point to the device, then hard links to the file will be created
2265 Multiple devices can be listed and their metadata will all be stored
2266 in the one directory.
2269 .BI \-\-restore= directory
2270 This is the reverse of
2273 will locate a file in the directory that has a name appropriate for
2274 the given device and will restore metadata from it. Names that match
2276 names are preferred, however if two of those refer to different files,
2278 will not choose between them but will abort the operation.
2280 If a file name is given instead of a
2284 will restore from that file to a single device, always provided the
2285 size of the file matches that of the device, and the file contains
2289 The devices should be active md arrays which will be deactivated, as
2290 long as they are not currently in use.
2294 This will fully activate a partially assembled md array.
2298 This will mark an active array as read-only, providing that it is
2299 not currently being used.
2305 array back to being read/write.
2309 For all operations except
2312 will cause the operation to be applied to all arrays listed in
2317 causes all devices listed in the config file to be examined.
2320 .BR \-b ", " \-\-brief
2321 Be less verbose. This is used with
2329 gives an intermediate level of verbosity.
2335 .B mdadm \-\-monitor
2336 .I options... devices...
2341 to periodically poll a number of md arrays and to report on any events
2344 will never exit once it decides that there are arrays to be checked,
2345 so it should normally be run in the background.
2347 As well as reporting events,
2349 may move a spare drive from one array to another if they are in the
2354 and if the destination array has a failed drive but no spares.
2356 If any devices are listed on the command line,
2358 will only monitor those devices. Otherwise all arrays listed in the
2359 configuration file will be monitored. Further, if
2361 is given, then any other md devices that appear in
2363 will also be monitored.
2365 The result of monitoring the arrays is the generation of events.
2366 These events are passed to a separate program (if specified) and may
2367 be mailed to a given E-mail address.
2369 When passing events to a program, the program is run once for each event,
2370 and is given 2 or 3 command-line arguments: the first is the
2371 name of the event (see below), the second is the name of the
2372 md device which is affected, and the third is the name of a related
2373 device if relevant (such as a component device that has failed).
2377 is given, then a program or an E-mail address must be specified on the
2378 command line or in the config file. If neither are available, then
2380 will not monitor anything.
2384 will continue monitoring as long as something was found to monitor. If
2385 no program or email is given, then each event is reported to
2388 The different events are:
2392 .B DeviceDisappeared
2393 An md array which previously was configured appears to no longer be
2394 configured. (syslog priority: Critical)
2398 was told to monitor an array which is RAID0 or Linear, then it will
2400 .B DeviceDisappeared
2401 with the extra information
2403 This is because RAID0 and Linear do not support the device-failed,
2404 hot-spare and resync operations which are monitored.
2408 An md array started reconstruction (e.g. recovery, resync, reshape,
2409 check, repair). (syslog priority: Warning)
2415 is a two-digit number (ie. 05, 48). This indicates that rebuild
2416 has passed that many percent of the total. The events are generated
2417 with fixed increment since 0. Increment size may be specified with
2418 a commandline option (default is 20). (syslog priority: Warning)
2422 An md array that was rebuilding, isn't any more, either because it
2423 finished normally or was aborted. (syslog priority: Warning)
2427 An active component device of an array has been marked as
2428 faulty. (syslog priority: Critical)
2432 A spare component device which was being rebuilt to replace a faulty
2433 device has failed. (syslog priority: Critical)
2437 A spare component device which was being rebuilt to replace a faulty
2438 device has been successfully rebuilt and has been made active.
2439 (syslog priority: Info)
2443 A new md array has been detected in the
2445 file. (syslog priority: Info)
2449 A newly noticed array appears to be degraded. This message is not
2452 notices a drive failure which causes degradation, but only when
2454 notices that an array is degraded when it first sees the array.
2455 (syslog priority: Critical)
2459 A spare drive has been moved from one array in a
2463 to another to allow a failed drive to be replaced.
2464 (syslog priority: Info)
2470 has been told, via the config file, that an array should have a certain
2471 number of spare devices, and
2473 detects that it has fewer than this number when it first sees the
2474 array, it will report a
2477 (syslog priority: Warning)
2481 An array was found at startup, and the
2484 (syslog priority: Info)
2494 cause Email to be sent. All events cause the program to be run.
2495 The program is run with two or three arguments: the event
2496 name, the array device and possibly a second device.
2498 Each event has an associated array device (e.g.
2500 and possibly a second device. For
2505 the second device is the relevant component device.
2508 the second device is the array that the spare was moved from.
2512 to move spares from one array to another, the different arrays need to
2513 be labeled with the same
2515 or the spares must be allowed to migrate through matching POLICY domains
2516 in the configuration file. The
2518 name can be any string; it is only necessary that different spare
2519 groups use different names.
2523 detects that an array in a spare group has fewer active
2524 devices than necessary for the complete array, and has no spare
2525 devices, it will look for another array in the same spare group that
2526 has a full complement of working drive and a spare. It will then
2527 attempt to remove the spare from the second drive and add it to the
2529 If the removal succeeds but the adding fails, then it is added back to
2532 If the spare group for a degraded array is not defined,
2534 will look at the rules of spare migration specified by POLICY lines in
2536 and then follow similar steps as above if a matching spare is found.
2539 The GROW mode is used for changing the size or shape of an active
2541 For this to work, the kernel must support the necessary change.
2542 Various types of growth are being added during 2.6 development.
2544 Currently the supported changes include
2546 change the "size" attribute for RAID1, RAID4, RAID5 and RAID6.
2548 increase or decrease the "raid\-devices" attribute of RAID0, RAID1, RAID4,
2551 change the chunk-size and layout of RAID0, RAID4, RAID5, RAID6 and RAID10.
2553 convert between RAID1 and RAID5, between RAID5 and RAID6, between
2554 RAID0, RAID4, and RAID5, and between RAID0 and RAID10 (in the near-2 mode).
2556 add a write-intent bitmap to any array which supports these bitmaps, or
2557 remove a write-intent bitmap from such an array.
2560 Using GROW on containers is currently supported only for Intel's IMSM
2561 container format. The number of devices in a container can be
2562 increased - which affects all arrays in the container - or an array
2563 in a container can be converted between levels where those levels are
2564 supported by the container, and the conversion is on of those listed
2565 above. Resizing arrays in an IMSM container with
2567 is not yet supported.
2569 Grow functionality (e.g. expand a number of raid devices) for Intel's
2570 IMSM container format has an experimental status. It is guarded by the
2571 .B MDADM_EXPERIMENTAL
2572 environment variable which must be set to '1' for a GROW command to
2574 This is for the following reasons:
2577 Intel's native IMSM check-pointing is not fully tested yet.
2578 This can causes IMSM incompatibility during the grow process: an array
2579 which is growing cannot roam between Microsoft Windows(R) and Linux
2583 Interrupting a grow operation is not recommended, because it
2584 has not been fully tested for Intel's IMSM container format yet.
2587 Note: Intel's native checkpointing doesn't use
2589 option and it is transparent for assembly feature.
2592 Normally when an array is built the "size" is taken from the smallest
2593 of the drives. If all the small drives in an arrays are, one at a
2594 time, removed and replaced with larger drives, then you could have an
2595 array of large drives with only a small amount used. In this
2596 situation, changing the "size" with "GROW" mode will allow the extra
2597 space to start being used. If the size is increased in this way, a
2598 "resync" process will start to make sure the new parts of the array
2601 Note that when an array changes size, any filesystem that may be
2602 stored in the array will not automatically grow or shrink to use or
2603 vacate the space. The
2604 filesystem will need to be explicitly told to use the extra space
2605 after growing, or to reduce its size
2607 to shrinking the array.
2609 Also the size of an array cannot be changed while it has an active
2610 bitmap. If an array has a bitmap, it must be removed before the size
2611 can be changed. Once the change is complete a new bitmap can be created.
2613 .SS RAID\-DEVICES CHANGES
2615 A RAID1 array can work with any number of devices from 1 upwards
2616 (though 1 is not very useful). There may be times which you want to
2617 increase or decrease the number of active devices. Note that this is
2618 different to hot-add or hot-remove which changes the number of
2621 When reducing the number of devices in a RAID1 array, the slots which
2622 are to be removed from the array must already be vacant. That is, the
2623 devices which were in those slots must be failed and removed.
2625 When the number of devices is increased, any hot spares that are
2626 present will be activated immediately.
2628 Changing the number of active devices in a RAID5 or RAID6 is much more
2629 effort. Every block in the array will need to be read and written
2630 back to a new location. From 2.6.17, the Linux Kernel is able to
2631 increase the number of devices in a RAID5 safely, including restarting
2632 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2633 increase or decrease the number of devices in a RAID5 or RAID6.
2635 From 2.6.35, the Linux Kernel is able to convert a RAID0 in to a RAID4
2638 uses this functionality and the ability to add
2639 devices to a RAID4 to allow devices to be added to a RAID0. When
2640 requested to do this,
2642 will convert the RAID0 to a RAID4, add the necessary disks and make
2643 the reshape happen, and then convert the RAID4 back to RAID0.
2645 When decreasing the number of devices, the size of the array will also
2646 decrease. If there was data in the array, it could get destroyed and
2647 this is not reversible, so you should firstly shrink the filesystem on
2648 the array to fit within the new size. To help prevent accidents,
2650 requires that the size of the array be decreased first with
2651 .BR "mdadm --grow --array-size" .
2652 This is a reversible change which simply makes the end of the array
2653 inaccessible. The integrity of any data can then be checked before
2654 the non-reversible reduction in the number of devices is request.
2656 When relocating the first few stripes on a RAID5 or RAID6, it is not
2657 possible to keep the data on disk completely consistent and
2658 crash-proof. To provide the required safety, mdadm disables writes to
2659 the array while this "critical section" is reshaped, and takes a
2660 backup of the data that is in that section. For grows, this backup may be
2661 stored in any spare devices that the array has, however it can also be
2662 stored in a separate file specified with the
2664 option, and is required to be specified for shrinks, RAID level
2665 changes and layout changes. If this option is used, and the system
2666 does crash during the critical period, the same file must be passed to
2668 to restore the backup and reassemble the array. When shrinking rather
2669 than growing the array, the reshape is done from the end towards the
2670 beginning, so the "critical section" is at the end of the reshape.
2674 Changing the RAID level of any array happens instantaneously. However
2675 in the RAID5 to RAID6 case this requires a non-standard layout of the
2676 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2677 required before the change can be accomplished. So while the level
2678 change is instant, the accompanying layout change can take quite a
2681 is required. If the array is not simultaneously being grown or
2682 shrunk, so that the array size will remain the same - for example,
2683 reshaping a 3-drive RAID5 into a 4-drive RAID6 - the backup file will
2684 be used not just for a "cricital section" but throughout the reshape
2685 operation, as described below under LAYOUT CHANGES.
2687 .SS CHUNK-SIZE AND LAYOUT CHANGES
2689 Changing the chunk-size of layout without also changing the number of
2690 devices as the same time will involve re-writing all blocks in-place.
2691 To ensure against data loss in the case of a crash, a
2693 must be provided for these changes. Small sections of the array will
2694 be copied to the backup file while they are being rearranged. This
2695 means that all the data is copied twice, once to the backup and once
2696 to the new layout on the array, so this type of reshape will go very
2699 If the reshape is interrupted for any reason, this backup file must be
2701 .B "mdadm --assemble"
2702 so the array can be reassembled. Consequently the file cannot be
2703 stored on the device being reshaped.
2708 A write-intent bitmap can be added to, or removed from, an active
2709 array. Either internal bitmaps, or bitmaps stored in a separate file,
2710 can be added. Note that if you add a bitmap stored in a file which is
2711 in a filesystem that is on the RAID array being affected, the system
2712 will deadlock. The bitmap must be on a separate filesystem.
2714 .SH INCREMENTAL MODE
2718 .B mdadm \-\-incremental
2722 .RI [ optional-aliases-for-device ]
2725 .B mdadm \-\-incremental \-\-fail
2729 .B mdadm \-\-incremental \-\-rebuild\-map
2732 .B mdadm \-\-incremental \-\-run \-\-scan
2735 This mode is designed to be used in conjunction with a device
2736 discovery system. As devices are found in a system, they can be
2738 .B "mdadm \-\-incremental"
2739 to be conditionally added to an appropriate array.
2741 Conversely, it can also be used with the
2743 flag to do just the opposite and find whatever array a particular device
2744 is part of and remove the device from that array.
2746 If the device passed is a
2748 device created by a previous call to
2750 then rather than trying to add that device to an array, all the arrays
2751 described by the metadata of the container will be started.
2754 performs a number of tests to determine if the device is part of an
2755 array, and which array it should be part of. If an appropriate array
2756 is found, or can be created,
2758 adds the device to the array and conditionally starts the array.
2762 will normally only add devices to an array which were previously working
2763 (active or spare) parts of that array. The support for automatic
2764 inclusion of a new drive as a spare in some array requires
2765 a configuration through POLICY in config file.
2769 makes are as follow:
2771 Is the device permitted by
2773 That is, is it listed in a
2775 line in that file. If
2777 is absent then the default it to allow any device. Similarly if
2779 contains the special word
2781 then any device is allowed. Otherwise the device name given to
2783 or one of the aliases given, or an alias found in the filesystem,
2784 must match one of the names or patterns in a
2788 This is the only context where the aliases are used. They are
2789 usually provided by a
2795 Does the device have a valid md superblock? If a specific metadata
2796 version is requested with
2800 then only that style of metadata is accepted, otherwise
2802 finds any known version of metadata. If no
2804 metadata is found, the device may be still added to an array
2805 as a spare if POLICY allows.
2809 Does the metadata match an expected array?
2810 The metadata can match in two ways. Either there is an array listed
2813 which identifies the array (either by UUID, by name, by device list,
2814 or by minor-number), or the array was created with a
2820 or on the command line.
2823 is not able to positively identify the array as belonging to the
2824 current host, the device will be rejected.
2829 keeps a list of arrays that it has partially assembled in
2831 If no array exists which matches
2832 the metadata on the new device,
2834 must choose a device name and unit number. It does this based on any
2837 or any name information stored in the metadata. If this name
2838 suggests a unit number, that number will be used, otherwise a free
2839 unit number will be chosen. Normally
2841 will prefer to create a partitionable array, however if the
2845 suggests that a non-partitionable array is preferred, that will be
2848 If the array is not found in the config file and its metadata does not
2849 identify it as belonging to the "homehost", then
2851 will choose a name for the array which is certain not to conflict with
2852 any array which does belong to this host. It does this be adding an
2853 underscore and a small number to the name preferred by the metadata.
2855 Once an appropriate array is found or created and the device is added,
2857 must decide if the array is ready to be started. It will
2858 normally compare the number of available (non-spare) devices to the
2859 number of devices that the metadata suggests need to be active. If
2860 there are at least that many, the array will be started. This means
2861 that if any devices are missing the array will not be restarted.
2867 in which case the array will be run as soon as there are enough
2868 devices present for the data to be accessible. For a RAID1, that
2869 means one device will start the array. For a clean RAID5, the array
2870 will be started as soon as all but one drive is present.
2872 Note that neither of these approaches is really ideal. If it can
2873 be known that all device discovery has completed, then
2877 can be run which will try to start all arrays that are being
2878 incrementally assembled. They are started in "read-auto" mode in
2879 which they are read-only until the first write request. This means
2880 that no metadata updates are made and no attempt at resync or recovery
2881 happens. Further devices that are found before the first write can
2882 still be added safely.
2885 This section describes environment variables that affect how mdadm
2890 Setting this value to 1 will prevent mdadm from automatically launching
2891 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2897 does not create any device nodes in /dev, but leaves that task to
2901 appears not to be configured, or if this environment variable is set
2904 will create and devices that are needed.
2907 .B MDADM_NO_SYSTEMCTL
2912 is in use it will normally request
2914 to start various background tasks (particularly
2916 rather than forking and running them in the background. This can be
2917 suppressed by setting
2918 .BR MDADM_NO_SYSTEMCTL=1 .
2922 A key value of IMSM metadata is that it allows interoperability with
2923 boot ROMs on Intel platforms, and with other major operating systems.
2926 will only allow an IMSM array to be created or modified if detects
2927 that it is running on an Intel platform which supports IMSM, and
2928 supports the particular configuration of IMSM that is being requested
2929 (some functionality requires newer OROM support).
2931 These checks can be suppressed by setting IMSM_NO_PLATFORM=1 in the
2932 environment. This can be useful for testing or for disaster
2933 recovery. You should be aware that interoperability may be
2934 compromised by setting this value.
2937 .B MDADM_GROW_ALLOW_OLD
2938 If an array is stopped while it is performing a reshape and that
2939 reshape was making use of a backup file, then when the array is
2942 will sometimes complain that the backup file is too old. If this
2943 happens and you are certain it is the right backup file, you can
2944 over-ride this check by setting
2945 .B MDADM_GROW_ALLOW_OLD=1
2950 Any string given in this variable is added to the start of the
2952 line in the config file, or treated as the whole
2954 line if none is given. It can be used to disable certain metadata
2957 is called from a boot script. For example
2959 .B " export MDADM_CONF_AUTO='-ddf -imsm'
2963 does not automatically assemble any DDF or
2964 IMSM arrays that are found. This can be useful on systems configured
2965 to manage such arrays with
2971 .B " mdadm \-\-query /dev/name-of-device"
2973 This will find out if a given device is a RAID array, or is part of
2974 one, and will provide brief information about the device.
2976 .B " mdadm \-\-assemble \-\-scan"
2978 This will assemble and start all arrays listed in the standard config
2979 file. This command will typically go in a system startup file.
2981 .B " mdadm \-\-stop \-\-scan"
2983 This will shut down all arrays that can be shut down (i.e. are not
2984 currently in use). This will typically go in a system shutdown script.
2986 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2988 If (and only if) there is an Email address or program given in the
2989 standard config file, then
2990 monitor the status of all arrays listed in that file by
2991 polling them ever 2 minutes.
2993 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2995 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2998 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
3000 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
3002 This will create a prototype config file that describes currently
3003 active arrays that are known to be made from partitions of IDE or SCSI drives.
3004 This file should be reviewed before being used as it may
3005 contain unwanted detail.
3007 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
3009 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
3011 This will find arrays which could be assembled from existing IDE and
3012 SCSI whole drives (not partitions), and store the information in the
3013 format of a config file.
3014 This file is very likely to contain unwanted detail, particularly
3017 entries. It should be reviewed and edited before being used as an
3020 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
3022 .B " mdadm \-Ebsc partitions"
3024 Create a list of devices by reading
3025 .BR /proc/partitions ,
3026 scan these for RAID superblocks, and printout a brief listing of all
3029 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
3031 Scan all partitions and devices listed in
3032 .BR /proc/partitions
3035 out of all such devices with a RAID superblock with a minor number of 0.
3037 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /run/mdadm/mon.pid"
3039 If config file contains a mail address or alert program, run mdadm in
3040 the background in monitor mode monitoring all md devices. Also write
3041 pid of mdadm daemon to
3042 .BR /run/mdadm/mon.pid .
3044 .B " mdadm \-Iq /dev/somedevice"
3046 Try to incorporate newly discovered device into some array as
3049 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
3051 Rebuild the array map from any current arrays, and then start any that
3054 .B " mdadm /dev/md4 --fail detached --remove detached"
3056 Any devices which are components of /dev/md4 will be marked as faulty
3057 and then remove from the array.
3059 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4"
3063 which is currently a RAID5 array will be converted to RAID6. There
3064 should normally already be a spare drive attached to the array as a
3065 RAID6 needs one more drive than a matching RAID5.
3067 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
3069 Create a DDF array over 6 devices.
3071 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
3073 Create a RAID5 array over any 3 devices in the given DDF set. Use
3074 only 30 gigabytes of each device.
3076 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
3078 Assemble a pre-exist ddf array.
3080 .B " mdadm -I /dev/md/ddf1"
3082 Assemble all arrays contained in the ddf array, assigning names as
3085 .B " mdadm \-\-create \-\-help"
3087 Provide help about the Create mode.
3089 .B " mdadm \-\-config \-\-help"
3091 Provide help about the format of the config file.
3093 .B " mdadm \-\-help"
3095 Provide general help.
3105 lists all active md devices with information about them.
3107 uses this to find arrays when
3109 is given in Misc mode, and to monitor array reconstruction
3114 The config file lists which devices may be scanned to see if
3115 they contain MD super block, and gives identifying information
3116 (e.g. UUID) about known MD arrays. See
3120 .SS /etc/mdadm.conf.d
3122 A directory containing configuration files which are read in lexical
3128 mode is used, this file gets a list of arrays currently being created.
3133 understand two sorts of names for array devices.
3135 The first is the so-called 'standard' format name, which matches the
3136 names used by the kernel and which appear in
3139 The second sort can be freely chosen, but must reside in
3141 When giving a device name to
3143 to create or assemble an array, either full path name such as
3147 can be given, or just the suffix of the second sort of name, such as
3153 chooses device names during auto-assembly or incremental assembly, it
3154 will sometimes add a small sequence number to the end of the name to
3155 avoid conflicted between multiple arrays that have the same name. If
3157 can reasonably determine that the array really is meant for this host,
3158 either by a hostname in the metadata, or by the presence of the array
3161 then it will leave off the suffix if possible.
3162 Also if the homehost is specified as
3165 will only use a suffix if a different array of the same name already
3166 exists or is listed in the config file.
3168 The standard names for non-partitioned arrays (the only sort of md
3169 array available in 2.4 and earlier) are of the form
3173 where NN is a number.
3174 The standard names for partitionable arrays (as available from 2.6
3175 onwards) are of the form:
3179 Partition numbers should be indicated by adding "pMM" to these, thus "/dev/md/d1p2".
3181 From kernel version 2.6.28 the "non-partitioned array" can actually
3182 be partitioned. So the "md_d\fBNN\fP"
3183 names are no longer needed, and
3184 partitions such as "/dev/md\fBNN\fPp\fBXX\fP"
3187 From kernel version 2.6.29 standard names can be non-numeric following
3194 is any string. These names are supported by
3196 since version 3.3 provided they are enabled in
3201 was previously known as
3205 For further information on mdadm usage, MD and the various levels of
3208 .B http://raid.wiki.kernel.org/
3210 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
3212 The latest version of
3214 should always be available from
3216 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/