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 and changing the array's consistency policy.
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 a 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. If not specified, the default config file
270 and default conf.d directory will be used. See
274 If the config file given is
276 then nothing will be read, but
278 will act as though the config file contained exactly
280 .B " DEVICE partitions containers"
284 to find a list of devices to scan, and
286 to find a list of containers to examine.
289 is given for the config file, then
291 will act as though the config file were empty.
293 If the name given is of a directory, then
295 will collect all the files contained in the directory with a name ending
298 sort them lexically, and process all of those files as config files.
301 .BR \-s ", " \-\-scan
304 for missing information.
305 In general, this option gives
307 permission to get any missing information (like component devices,
308 array devices, array identities, and alert destination) from the
309 configuration file (see previous option);
310 one exception is MISC mode when using
316 says to get a list of array devices from
320 .BR \-e ", " \-\-metadata=
321 Declare the style of RAID metadata (superblock) to be used. The
322 default is {DEFAULT_METADATA} for
324 and to guess for other operations.
325 The default can be overridden by setting the
334 .ie '{DEFAULT_METADATA}'0.90'
335 .IP "0, 0.90, default"
338 Use the original 0.90 format superblock. This format limits arrays to
339 28 component devices and limits component devices of levels 1 and
340 greater to 2 terabytes. It is also possible for there to be confusion
341 about whether the superblock applies to a whole device or just the
342 last partition, if that partition starts on a 64K boundary.
343 .ie '{DEFAULT_METADATA}'0.90'
344 .IP "1, 1.0, 1.1, 1.2"
346 .IP "1, 1.0, 1.1, 1.2 default"
347 Use the new version-1 format superblock. This has fewer restrictions.
348 It can easily be moved between hosts with different endian-ness, and a
349 recovery operation can be checkpointed and restarted. The different
350 sub-versions store the superblock at different locations on the
351 device, either at the end (for 1.0), at the start (for 1.1) or 4K from
352 the start (for 1.2). "1" is equivalent to "1.2" (the commonly
353 preferred 1.x format).
354 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
356 Use the "Industry Standard" DDF (Disk Data Format) format defined by
358 When creating a DDF array a
360 will be created, and normal arrays can be created in that container.
362 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
364 which is managed in a similar manner to DDF, and is supported by an
365 option-rom on some platforms:
367 .B https://www.intel.com/content/www/us/en/support/products/122484/memory-and-storage/ssd-software/intel-virtual-raid-on-cpu-intel-vroc.html
373 This will override any
375 setting in the config file and provides the identity of the host which
376 should be considered the home for any arrays.
378 When creating an array, the
380 will be recorded in the metadata. For version-1 superblocks, it will
381 be prefixed to the array name. For version-0.90 superblocks, part of
382 the SHA1 hash of the hostname will be stored in the latter half of the
385 When reporting information about an array, any array which is tagged
386 for the given homehost will be reported as such.
388 When using Auto-Assemble, only arrays tagged for the given homehost
389 will be allowed to use 'local' names (i.e. not ending in '_' followed
390 by a digit string). See below under
391 .BR "Auto-Assembly" .
393 The special name "\fBany\fP" can be used as a wild card. If an array
396 then the name "\fBany\fP" will be stored in the array and it can be
397 assembled in the same way on any host. If an array is assembled with
398 this option, then the homehost recorded on the array will be ignored.
404 needs to print the name for a device it normally finds the name in
406 which refers to the device and is the shortest. When a path component is
410 will prefer a longer name if it contains that component. For example
411 .B \-\-prefer=by-uuid
412 will prefer a name in a subdirectory of
417 This functionality is currently only provided by
423 .B \-\-home\-cluster=
424 specifies the cluster name for the md device. The md device can be assembled
425 only on the cluster which matches the name specified. If this option is not
426 provided, mdadm tries to detect the cluster name automatically.
428 .SH For create, build, or grow:
431 .BR \-n ", " \-\-raid\-devices=
432 Specify the number of active devices in the array. This, plus the
433 number of spare devices (see below) must equal the number of
435 (including "\fBmissing\fP" devices)
436 that are listed on the command line for
438 Setting a value of 1 is probably
439 a mistake and so requires that
441 be specified first. A value of 1 will then be allowed for linear,
442 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
444 This number can only be changed using
446 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
447 the necessary support.
450 .BR \-x ", " \-\-spare\-devices=
451 Specify the number of spare (eXtra) devices in the initial array.
452 Spares can also be added
453 and removed later. The number of component devices listed
454 on the command line must equal the number of RAID devices plus the
455 number of spare devices.
458 .BR \-z ", " \-\-size=
459 Amount (in Kilobytes) of space to use from each drive in RAID levels 1/4/5/6/10
460 and for RAID 0 on external metadata.
461 This must be a multiple of the chunk size, and must leave about 128Kb
462 of space at the end of the drive for the RAID superblock.
463 If this is not specified
464 (as it normally is not) the smallest drive (or partition) sets the
465 size, though if there is a variance among the drives of greater than 1%, a warning is
468 A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes,
469 Megabytes, Gigabytes or Terabytes respectively.
471 Sometimes a replacement drive can be a little smaller than the
472 original drives though this should be minimised by IDEMA standards.
473 Such a replacement drive will be rejected by
475 To guard against this it can be useful to set the initial size
476 slightly smaller than the smaller device with the aim that it will
477 still be larger than any replacement.
479 This option can be used with
481 for determining the initial size of an array. For external metadata,
482 it can be used on a volume, but not on a container itself.
483 Setting the initial size of
485 array is only valid for external metadata.
487 This value can be set with
489 for RAID level 1/4/5/6/10 though
490 DDF arrays may not be able to support this.
491 RAID 0 array size cannot be changed.
492 If the array was created with a size smaller than the currently
493 active drives, the extra space can be accessed using
495 The size can be given as
497 which means to choose the largest size that fits on all current drives.
499 Before reducing the size of the array (with
500 .BR "\-\-grow \-\-size=" )
501 you should make sure that space isn't needed. If the device holds a
502 filesystem, you would need to resize the filesystem to use less space.
504 After reducing the array size you should check that the data stored in
505 the device is still available. If the device holds a filesystem, then
506 an 'fsck' of the filesystem is a minimum requirement. If there are
507 problems the array can be made bigger again with no loss with another
508 .B "\-\-grow \-\-size="
512 .BR \-Z ", " \-\-array\-size=
513 This is only meaningful with
515 and its effect is not persistent: when the array is stopped and
516 restarted the default array size will be restored.
518 Setting the array-size causes the array to appear smaller to programs
519 that access the data. This is particularly needed before reshaping an
520 array so that it will be smaller. As the reshape is not reversible,
521 but setting the size with
523 is, it is required that the array size is reduced as appropriate
524 before the number of devices in the array is reduced.
526 Before reducing the size of the array you should make sure that space
527 isn't needed. If the device holds a filesystem, you would need to
528 resize the filesystem to use less space.
530 After reducing the array size you should check that the data stored in
531 the device is still available. If the device holds a filesystem, then
532 an 'fsck' of the filesystem is a minimum requirement. If there are
533 problems the array can be made bigger again with no loss with another
534 .B "\-\-grow \-\-array\-size="
537 A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes,
538 Megabytes, Gigabytes or Terabytes respectively.
541 restores the apparent size of the array to be whatever the real
542 amount of available space is.
544 Clustered arrays do not support this parameter yet.
547 .BR \-c ", " \-\-chunk=
548 Specify chunk size in kilobytes. The default when creating an
549 array is 512KB. To ensure compatibility with earlier versions, the
550 default when building an array with no persistent metadata is 64KB.
551 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
553 RAID4, RAID5, RAID6, and RAID10 require the chunk size to be a power
554 of 2, with minimal chunk size being 4KB.
556 A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes,
557 Megabytes, Gigabytes or Terabytes respectively.
561 Specify the rounding factor for a Linear array. The size of each
562 component will be rounded down to a multiple of this size.
563 This is a synonym for
565 but highlights the different meaning for Linear as compared to other
566 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
567 use, and is 0K (i.e. no rounding) in later kernels.
570 .BR \-l ", " \-\-level=
571 Set RAID level. When used with
573 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
574 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
575 Obviously some of these are synonymous.
579 metadata type is requested, only the
581 level is permitted, and it does not need to be explicitly given.
585 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
589 to change the RAID level in some cases. See LEVEL CHANGES below.
592 .BR \-p ", " \-\-layout=
593 This option configures the fine details of data layout for RAID5, RAID6,
594 and RAID10 arrays, and controls the failure modes for
596 It can also be used for working around a kernel bug with RAID0, but generally
597 doesn't need to be used explicitly.
599 The layout of the RAID5 parity block can be one of
600 .BR left\-asymmetric ,
601 .BR left\-symmetric ,
602 .BR right\-asymmetric ,
603 .BR right\-symmetric ,
604 .BR la ", " ra ", " ls ", " rs .
606 .BR left\-symmetric .
608 It is also possible to cause RAID5 to use a RAID4-like layout by
614 Finally for RAID5 there are DDF\-compatible layouts,
615 .BR ddf\-zero\-restart ,
616 .BR ddf\-N\-restart ,
618 .BR ddf\-N\-continue .
620 These same layouts are available for RAID6. There are also 4 layouts
621 that will provide an intermediate stage for converting between RAID5
622 and RAID6. These provide a layout which is identical to the
623 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
624 syndrome (the second 'parity' block used by RAID6) on the last device.
626 .BR left\-symmetric\-6 ,
627 .BR right\-symmetric\-6 ,
628 .BR left\-asymmetric\-6 ,
629 .BR right\-asymmetric\-6 ,
631 .BR parity\-first\-6 .
633 When setting the failure mode for level
636 .BR write\-transient ", " wt ,
637 .BR read\-transient ", " rt ,
638 .BR write\-persistent ", " wp ,
639 .BR read\-persistent ", " rp ,
641 .BR read\-fixable ", " rf ,
642 .BR clear ", " flush ", " none .
644 Each failure mode can be followed by a number, which is used as a period
645 between fault generation. Without a number, the fault is generated
646 once on the first relevant request. With a number, the fault will be
647 generated after that many requests, and will continue to be generated
648 every time the period elapses.
650 Multiple failure modes can be current simultaneously by using the
652 option to set subsequent failure modes.
654 "clear" or "none" will remove any pending or periodic failure modes,
655 and "flush" will clear any persistent faults.
657 The layout options for RAID10 are one of 'n', 'o' or 'f' followed
658 by a small number signifying the number of copies of each datablock.
659 The default is 'n2'. The supported options are:
662 signals 'near' copies. Multiple copies of one data block are at
663 similar offsets in different devices.
666 signals 'offset' copies. Rather than the chunks being duplicated
667 within a stripe, whole stripes are duplicated but are rotated by one
668 device so duplicate blocks are on different devices. Thus subsequent
669 copies of a block are in the next drive, and are one chunk further
674 (multiple copies have very different offsets).
675 See md(4) for more detail about 'near', 'offset', and 'far'.
677 As for the number of copies of each data block, 2 is normal, 3
678 can be useful. This number can be at most equal to the number of
679 devices in the array. It does not need to divide evenly into that
680 number (e.g. it is perfectly legal to have an 'n2' layout for an array
681 with an odd number of devices).
683 A bug introduced in Linux 3.14 means that RAID0 arrays
684 .B "with devices of differing sizes"
685 started using a different layout. This could lead to
686 data corruption. Since Linux 5.4 (and various stable releases that received
687 backports), the kernel will not accept such an array unless
688 a layout is explicitly set. It can be set to
692 When creating a new array,
696 by default, so the layout does not normally need to be set.
697 An array created for either
701 will not be recognized by an (unpatched) kernel prior to 5.4. To create
702 a RAID0 array with devices of differing sizes that can be used on an
703 older kernel, you can set the layout to
705 This will use whichever layout the running kernel supports, so the data
706 on the array may become corrupt when changing kernel from pre-3.14 to a
709 When an array is converted between RAID5 and RAID6 an intermediate
710 RAID6 layout is used in which the second parity block (Q) is always on
711 the last device. To convert a RAID5 to RAID6 and leave it in this new
712 layout (which does not require re-striping) use
713 .BR \-\-layout=preserve .
714 This will try to avoid any restriping.
716 The converse of this is
717 .B \-\-layout=normalise
718 which will change a non-standard RAID6 layout into a more standard
725 (thus explaining the p of
729 .BR \-b ", " \-\-bitmap=
730 Specify a file to store a write-intent bitmap in. The file should not
733 is also given. The same file should be provided
734 when assembling the array. If the word
736 is given, then the bitmap is stored with the metadata on the array,
737 and so is replicated on all devices. If the word
741 mode, then any bitmap that is present is removed. If the word
743 is given, the array is created for a clustered environment. One bitmap
744 is created for each node as defined by the
746 parameter and are stored internally.
748 To help catch typing errors, the filename must contain at least one
749 slash ('/') if it is a real file (not 'internal' or 'none').
751 Note: external bitmaps are only known to work on ext2 and ext3.
752 Storing bitmap files on other filesystems may result in serious problems.
754 When creating an array on devices which are 100G or larger,
756 automatically adds an internal bitmap as it will usually be
757 beneficial. This can be suppressed with
759 or by selecting a different consistency policy with
760 .BR \-\-consistency\-policy .
763 .BR \-\-bitmap\-chunk=
764 Set the chunk size of the bitmap. Each bit corresponds to that many
765 Kilobytes of storage.
766 When using a file-based bitmap, the default is to use the smallest
767 size that is at least 4 and requires no more than 2^21 chunks.
770 bitmap, the chunk size defaults to 64Meg, or larger if necessary to
771 fit the bitmap into the available space.
773 A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes,
774 Megabytes, Gigabytes or Terabytes respectively.
777 .BR \-W ", " \-\-write\-mostly
778 subsequent devices listed in a
783 command will be flagged as 'write\-mostly'. This is valid for RAID1
784 only and means that the 'md' driver will avoid reading from these
785 devices if at all possible. This can be useful if mirroring over a
789 .BR \-\-write\-behind=
790 Specify that write-behind mode should be enabled (valid for RAID1
791 only). If an argument is specified, it will set the maximum number
792 of outstanding writes allowed. The default value is 256.
793 A write-intent bitmap is required in order to use write-behind
794 mode, and write-behind is only attempted on drives marked as
799 subsequent devices listed in a
803 command will be flagged as 'failfast'. This is valid for RAID1 and
804 RAID10 only. IO requests to these devices will be encouraged to fail
805 quickly rather than cause long delays due to error handling. Also no
806 attempt is made to repair a read error on these devices.
808 If an array becomes degraded so that the 'failfast' device is the only
809 usable device, the 'failfast' flag will then be ignored and extended
810 delays will be preferred to complete failure.
812 The 'failfast' flag is appropriate for storage arrays which have a
813 low probability of true failure, but which may sometimes
814 cause unacceptable delays due to internal maintenance functions.
817 .BR \-\-assume\-clean
820 that the array pre-existed and is known to be clean. It can be useful
821 when trying to recover from a major failure as you can be sure that no
822 data will be affected unless you actually write to the array. It can
823 also be used when creating a RAID1 or RAID10 if you want to avoid the
824 initial resync, however this practice \(em while normally safe \(em is not
825 recommended. Use this only if you really know what you are doing.
827 When the devices that will be part of a new array were filled
828 with zeros before creation the operator knows the array is
829 actually clean. If that is the case, such as after running
830 badblocks, this argument can be used to tell mdadm the
831 facts the operator knows.
833 When an array is resized to a larger size with
834 .B "\-\-grow \-\-size="
835 the new space is normally resynced in that same way that the whole
836 array is resynced at creation. From Linux version 3.0,
838 can be used with that command to avoid the automatic resync.
842 When creating an array, send write zeroes requests to all the block
843 devices. This should zero the data area on all disks such that the
844 initial sync is not necessary and, if successfull, will behave
849 This is intended for use with devices that have hardware offload for
850 zeroing, but despite this zeroing can still take several minutes for
851 large disks. Thus a message is printed before and after zeroing and
852 each disk is zeroed in parallel with the others.
854 This is only meaningful with --create.
857 .BR \-\-backup\-file=
860 is used to increase the number of raid devices in a RAID5 or RAID6 if
861 there are no spare devices available, or to shrink, change RAID level
862 or layout. See the GROW MODE section below on RAID\-DEVICES CHANGES.
863 The file must be stored on a separate device, not on the RAID array
868 Arrays with 1.x metadata can leave a gap between the start of the
869 device and the start of array data. This gap can be used for various
870 metadata. The start of data is known as the
872 Normally an appropriate data offset is computed automatically.
873 However it can be useful to set it explicitly such as when re-creating
874 an array which was originally created using a different version of
876 which computed a different offset.
878 Setting the offset explicitly over-rides the default. The value given
879 is in Kilobytes unless a suffix of 'K', 'M', 'G' or 'T' is used to explicitly
880 indicate Kilobytes, Megabytes, Gigabytes or Terabytes respectively.
884 can also be used with
886 for some RAID levels (initially on RAID10). This allows the
887 data\-offset to be changed as part of the reshape process. When the
888 data offset is changed, no backup file is required as the difference
889 in offsets is used to provide the same functionality.
891 When the new offset is earlier than the old offset, the number of
892 devices in the array cannot shrink. When it is after the old offset,
893 the number of devices in the array cannot increase.
895 When creating an array,
899 In the case each member device is expected to have an offset appended
900 to the name, separated by a colon. This makes it possible to recreate
901 exactly an array which has varying data offsets (as can happen when
902 different versions of
904 are used to add different devices).
908 This option is complementary to the
909 .B \-\-freeze-reshape
910 option for assembly. It is needed when
912 operation is interrupted and it is not restarted automatically due to
913 .B \-\-freeze-reshape
914 usage during array assembly. This option is used together with
918 ) command and device for a pending reshape to be continued.
919 All parameters required for reshape continuation will be read from array metadata.
923 .BR \-\-backup\-file=
924 option to be set, continuation option will require to have exactly the same
925 backup file given as well.
927 Any other parameter passed together with
929 option will be ignored.
932 .BR \-N ", " \-\-name=
935 for the array. This is currently only effective when creating an
936 array with a version-1 superblock, or an array in a DDF container.
937 The name is a simple textual string that can be used to identify array
938 components when assembling. If name is needed but not specified, it
939 is taken from the basename of the device that is being created.
951 run the array, even if some of the components
952 appear to be active in another array or filesystem. Normally
954 will ask for confirmation before including such components in an
955 array. This option causes that question to be suppressed.
958 .BR \-f ", " \-\-force
961 accept the geometry and layout specified without question. Normally
963 will not allow the creation of an array with only one device, and will try
964 to create a RAID5 array with one missing drive (as this makes the
965 initial resync work faster). With
968 will not try to be so clever.
971 .BR \-o ", " \-\-readonly
974 rather than read-write as normal. No writes will be allowed to the
975 array, and no resync, recovery, or reshape will be started. It works with
976 Create, Assemble, Manage and Misc mode.
979 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
980 Instruct mdadm how to create the device file if needed, possibly allocating
981 an unused minor number. "md" causes a non-partitionable array
982 to be used (though since Linux 2.6.28, these array devices are in fact
983 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
984 later) to be used. "yes" requires the named md device to have
985 a 'standard' format, and the type and minor number will be determined
986 from this. With mdadm 3.0, device creation is normally left up to
988 so this option is unlikely to be needed.
989 See DEVICE NAMES below.
991 The argument can also come immediately after
996 is not given on the command line or in the config file, then
1002 is also given, then any
1004 entries in the config file will override the
1006 instruction given on the command line.
1008 For partitionable arrays,
1010 will create the device file for the whole array and for the first 4
1011 partitions. A different number of partitions can be specified at the
1012 end of this option (e.g.
1014 If the device name ends with a digit, the partition names add a 'p',
1016 .IR /dev/md/home1p3 .
1017 If there is no trailing digit, then the partition names just have a
1019 .IR /dev/md/scratch3 .
1021 If the md device name is in a 'standard' format as described in DEVICE
1022 NAMES, then it will be created, if necessary, with the appropriate
1023 device number based on that name. If the device name is not in one of these
1024 formats, then an unused device number will be allocated. The device
1025 number will be considered unused if there is no active array for that
1026 number, and there is no entry in /dev for that number and with a
1027 non-standard name. Names that are not in 'standard' format are only
1028 allowed in "/dev/md/".
1030 This is meaningful with
1036 .BR \-a ", " "\-\-add"
1037 This option can be used in Grow mode in two cases.
1039 If the target array is a Linear array, then
1041 can be used to add one or more devices to the array. They
1042 are simply catenated on to the end of the array. Once added, the
1043 devices cannot be removed.
1047 option is being used to increase the number of devices in an array,
1050 can be used to add some extra devices to be included in the array.
1051 In most cases this is not needed as the extra devices can be added as
1052 spares first, and then the number of raid disks can be changed.
1053 However, for RAID0 it is not possible to add spares. So to increase
1054 the number of devices in a RAID0, it is necessary to set the new
1055 number of devices, and to add the new devices, in the same command.
1059 Only works when the array is created for a clustered environment. It specifies
1060 the maximum number of nodes in the cluster that will use this device
1061 simultaneously. If not specified, this defaults to 4.
1064 .BR \-\-write-journal
1065 Specify journal device for the RAID-4/5/6 array. The journal device
1066 should be an SSD with a reasonable lifetime.
1069 .BR \-k ", " \-\-consistency\-policy=
1070 Specify how the array maintains consistency in the case of an unexpected shutdown.
1071 Only relevant for RAID levels with redundancy.
1072 Currently supported options are:
1077 Full resync is performed and all redundancy is regenerated when the array is
1078 started after an unclean shutdown.
1082 Resync assisted by a write-intent bitmap. Implicitly selected when using
1087 For RAID levels 4/5/6, the journal device is used to log transactions and replay
1088 after an unclean shutdown. Implicitly selected when using
1089 .BR \-\-write\-journal .
1093 For RAID5 only, Partial Parity Log is used to close the write hole and
1094 eliminate resync. PPL is stored in the metadata region of RAID member drives,
1095 no additional journal drive is needed.
1098 Can be used with \-\-grow to change the consistency policy of an active array
1099 in some cases. See CONSISTENCY POLICY CHANGES below.
1106 .BR \-u ", " \-\-uuid=
1107 uuid of array to assemble. Devices which don't have this uuid are
1111 .BR \-m ", " \-\-super\-minor=
1112 Minor number of device that array was created for. Devices which
1113 don't have this minor number are excluded. If you create an array as
1114 /dev/md1, then all superblocks will contain the minor number 1, even if
1115 the array is later assembled as /dev/md2.
1117 Giving the literal word "dev" for
1121 to use the minor number of the md device that is being assembled.
1122 e.g. when assembling
1124 .B \-\-super\-minor=dev
1125 will look for super blocks with a minor number of 0.
1128 is only relevant for v0.90 metadata, and should not normally be used.
1134 .BR \-N ", " \-\-name=
1135 Specify the name of the array to assemble. This must be the name
1136 that was specified when creating the array. It must either match
1137 the name stored in the superblock exactly, or it must match
1140 prefixed to the start of the given name.
1143 .BR \-f ", " \-\-force
1144 Assemble the array even if the metadata on some devices appears to be
1147 cannot find enough working devices to start the array, but can find
1148 some devices that are recorded as having failed, then it will mark
1149 those devices as working so that the array can be started. This works only for
1150 native. For external metadata it allows to start dirty degraded RAID 4, 5, 6.
1151 An array which requires
1153 to be started may contain data corruption. Use it carefully.
1156 .BR \-R ", " \-\-run
1157 Attempt to start the array even if fewer drives were given than were
1158 present last time the array was active. Normally if not all the
1159 expected drives are found and
1161 is not used, then the array will be assembled but not started.
1164 an attempt will be made to start it anyway.
1168 This is the reverse of
1170 in that it inhibits the startup of array unless all expected drives
1171 are present. This is only needed with
1173 and can be used if the physical connections to devices are
1174 not as reliable as you would like.
1177 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
1178 See this option under Create and Build options.
1181 .BR \-b ", " \-\-bitmap=
1182 Specify the bitmap file that was given when the array was created. If
1185 bitmap, there is no need to specify this when assembling the array.
1188 .BR \-\-backup\-file=
1191 was used while reshaping an array (e.g. changing number of devices or
1192 chunk size) and the system crashed during the critical section, then the same
1194 must be presented to
1196 to allow possibly corrupted data to be restored, and the reshape
1200 .BR \-\-invalid\-backup
1201 If the file needed for the above option is not available for any
1202 reason an empty file can be given together with this option to
1203 indicate that the backup file is invalid. In this case the data that
1204 was being rearranged at the time of the crash could be irrecoverably
1205 lost, but the rest of the array may still be recoverable. This option
1206 should only be used as a last resort if there is no way to recover the
1211 .BR \-U ", " \-\-update=
1212 Update the superblock on each device while assembling the array. The
1213 argument given to this flag can be one of
1229 .BR layout\-original ,
1230 .BR layout\-alternate ,
1231 .BR layout\-unspecified ,
1238 option will adjust the superblock of an array what was created on a Sparc
1239 machine running a patched 2.2 Linux kernel. This kernel got the
1240 alignment of part of the superblock wrong. You can use the
1241 .B "\-\-examine \-\-sparc2.2"
1244 to see what effect this would have.
1248 option will update the
1249 .B "preferred minor"
1250 field on each superblock to match the minor number of the array being
1252 This can be useful if
1254 reports a different "Preferred Minor" to
1256 In some cases this update will be performed automatically
1257 by the kernel driver. In particular, the update happens automatically
1258 at the first write to an array with redundancy (RAID level 1 or
1259 greater) on a 2.6 (or later) kernel.
1263 option will change the uuid of the array. If a UUID is given with the
1265 option that UUID will be used as a new UUID and will
1267 be used to help identify the devices in the array.
1270 is given, a random UUID is chosen.
1274 option will change the
1276 of the array as stored in the superblock. This is only supported for
1277 version-1 superblocks.
1281 option will change the
1283 of the array as stored in the bitmap superblock. This option only
1284 works for a clustered environment.
1288 option will change the
1290 as recorded in the superblock. For version-0 superblocks, this is the
1291 same as updating the UUID.
1292 For version-1 superblocks, this involves updating the name.
1296 option will change the cluster name as recorded in the superblock and
1297 bitmap. This option only works for a clustered environment.
1301 option will cause the array to be marked
1303 meaning that any redundancy in the array (e.g. parity for RAID5,
1304 copies for RAID1) may be incorrect. This will cause the RAID system
1305 to perform a "resync" pass to make sure that all redundant information
1310 option allows arrays to be moved between machines with different
1311 byte-order, such as from a big-endian machine like a Sparc or some
1312 MIPS machines, to a little-endian x86_64 machine.
1313 When assembling such an array for the first time after a move, giving
1314 .B "\-\-update=byteorder"
1317 to expect superblocks to have their byteorder reversed, and will
1318 correct that order before assembling the array. This is only valid
1319 with original (Version 0.90) superblocks.
1323 option will correct the summaries in the superblock. That is the
1324 counts of total, working, active, failed, and spare devices.
1328 option will rarely be of use. It applies to version 1.1 and 1.2 metadata
1329 only (where the metadata is at the start of the device) and is only
1330 useful when the component device has changed size (typically become
1331 larger). The version 1 metadata records the amount of the device that
1332 can be used to store data, so if a device in a version 1.1 or 1.2
1333 array becomes larger, the metadata will still be visible, but the
1334 extra space will not. In this case it might be useful to assemble the
1336 .BR \-\-update=devicesize .
1339 to determine the maximum usable amount of space on each device and
1340 update the relevant field in the metadata.
1344 option only works on v0.90 metadata arrays and will convert them to
1345 v1.0 metadata. The array must not be dirty (i.e. it must not need a
1346 sync) and it must not have a write-intent bitmap.
1348 The old metadata will remain on the devices, but will appear older
1349 than the new metadata and so will usually be ignored. The old metadata
1350 (or indeed the new metadata) can be removed by giving the appropriate
1353 .BR \-\-zero\-superblock .
1357 option can be used when an array has an internal bitmap which is
1358 corrupt in some way so that assembling the array normally fails. It
1359 will cause any internal bitmap to be ignored.
1363 option will reserve space in each device for a bad block list. This
1364 will be 4K in size and positioned near the end of any free space
1365 between the superblock and the data.
1369 option will cause any reservation of space for a bad block list to be
1370 removed. If the bad block list contains entries, this will fail, as
1371 removing the list could cause data corruption.
1375 option will enable PPL for a RAID5 array and reserve space for PPL on each
1376 device. There must be enough free space between the data and superblock and a
1377 write-intent bitmap or journal must not be used.
1381 option will disable PPL in the superblock.
1386 .B layout\-alternate
1387 options are for RAID0 arrays with non-uniform devices size that were in
1388 use before Linux 5.4. If the array was being used with Linux 3.13 or
1389 earlier, then to assemble the array on a new kernel,
1390 .B \-\-update=layout\-original
1391 must be given. If the array was created and used with a kernel from Linux 3.14 to
1393 .B \-\-update=layout\-alternate
1394 must be given. This only needs to be given once. Subsequent assembly of the array
1395 will happen normally.
1396 For more information, see
1400 .B layout\-unspecified
1401 option reverts the effect of
1404 .B layout\-alternate
1405 and allows the array to be again used on a kernel prior to Linux 5.3.
1406 This option should be used with great caution.
1409 .BR \-\-freeze\-reshape
1410 This option is intended to be used in start-up scripts during the initrd boot phase.
1411 When the array under reshape is assembled during the initrd phase, this option
1412 stops the reshape after the reshape-critical section has been restored. This happens
1413 before the file system pivot operation and avoids loss of filesystem context.
1414 Losing file system context would cause reshape to be broken.
1416 Reshape can be continued later using the
1418 option for the grow command.
1420 .SH For Manage mode:
1423 .BR \-t ", " \-\-test
1424 Unless a more serious error occurred,
1426 will exit with a status of 2 if no changes were made to the array and
1427 0 if at least one change was made.
1428 This can be useful when an indirect specifier such as
1433 is used in requesting an operation on the array.
1435 will report failure if these specifiers didn't find any match.
1438 .BR \-a ", " \-\-add
1439 hot-add listed devices.
1440 If a device appears to have recently been part of the array
1441 (possibly it failed or was removed) the device is re\-added as described
1443 If that fails or the device was never part of the array, the device is
1444 added as a hot-spare.
1445 If the array is degraded, it will immediately start to rebuild data
1448 Note that this and the following options are only meaningful on array
1449 with redundancy. They don't apply to RAID0 or Linear.
1453 re\-add a device that was previously removed from an array.
1454 If the metadata on the device reports that it is a member of the
1455 array, and the slot that it used is still vacant, then the device will
1456 be added back to the array in the same position. This will normally
1457 cause the data for that device to be recovered. However, based on the
1458 event count on the device, the recovery may only require sections that
1459 are flagged by a write-intent bitmap to be recovered or may not require
1460 any recovery at all.
1462 When used on an array that has no metadata (i.e. it was built with
1464 it will be assumed that bitmap-based recovery is enough to make the
1465 device fully consistent with the array.
1468 can also be accompanied by
1469 .BR \-\-update=devicesize ,
1470 .BR \-\-update=bbl ", or"
1471 .BR \-\-update=no\-bbl .
1472 See descriptions of these options when used in Assemble mode for an
1473 explanation of their use.
1475 If the device name given is
1479 will try to find any device that looks like it should be
1480 part of the array but isn't and will try to re\-add all such devices.
1482 If the device name given is
1486 will find all devices in the array that are marked
1488 remove them and attempt to immediately re\-add them. This can be
1489 useful if you are certain that the reason for failure has been
1494 Add a device as a spare. This is similar to
1496 except that it does not attempt
1498 first. The device will be added as a spare even if it looks like it
1499 could be a recent member of the array.
1502 .BR \-r ", " \-\-remove
1503 remove listed devices. They must not be active. i.e. they should
1504 be failed or spare devices.
1506 As well as the name of a device file
1516 The first causes all failed devices to be removed. The second causes
1517 any device which is no longer connected to the system (i.e an 'open'
1521 The third will remove a set as described below under
1525 .BR \-f ", " \-\-fail
1526 Mark listed devices as faulty.
1527 As well as the name of a device file, the word
1531 can be given. The former will cause any device that has been detached from
1532 the system to be marked as failed. It can then be removed.
1534 For RAID10 arrays where the number of copies evenly divides the number
1535 of devices, the devices can be conceptually divided into sets where
1536 each set contains a single complete copy of the data on the array.
1537 Sometimes a RAID10 array will be configured so that these sets are on
1538 separate controllers. In this case, all the devices in one set can be
1539 failed by giving a name like
1545 The appropriate set names are reported by
1555 Mark listed devices as requiring replacement. As soon as a spare is
1556 available, it will be rebuilt and will replace the marked device.
1557 This is similar to marking a device as faulty, but the device remains
1558 in service during the recovery process to increase resilience against
1559 multiple failures. When the replacement process finishes, the
1560 replaced device will be marked as faulty.
1564 This can follow a list of
1566 devices. The devices listed after
1568 will preferentially be used to replace the devices listed after
1570 These devices must already be spare devices in the array.
1573 .BR \-\-write\-mostly
1574 Subsequent devices that are added or re\-added will have the 'write-mostly'
1575 flag set. This is only valid for RAID1 and means that the 'md' driver
1576 will avoid reading from these devices if possible.
1579 Subsequent devices that are added or re\-added will have the 'write-mostly'
1582 .BR \-\-cluster\-confirm
1583 Confirm the existence of the device. This is issued in response to an \-\-add
1584 request by a node in a cluster. When a node adds a device it sends a message
1585 to all nodes in the cluster to look for a device with a UUID. This translates
1586 to a udev notification with the UUID of the device to be added and the slot
1587 number. The receiving node must acknowledge this message
1588 with \-\-cluster\-confirm. Valid arguments are <slot>:<devicename> in case
1589 the device is found or <slot>:missing in case the device is not found.
1593 Add a journal to an existing array, or recreate journal for a RAID-4/5/6 array
1594 that lost a journal device. To avoid interrupting ongoing write operations,
1596 only works for array in Read-Only state.
1600 Subsequent devices that are added or re\-added will have
1601 the 'failfast' flag set. This is only valid for RAID1 and RAID10 and
1602 means that the 'md' driver will avoid long timeouts on error handling
1606 Subsequent devices that are re\-added will be re\-added without
1607 the 'failfast' flag set.
1610 Each of these options requires that the first device listed is the array
1611 to be acted upon, and the remainder are component devices to be added,
1612 removed, marked as faulty, etc. Several different operations can be
1613 specified for different devices, e.g.
1615 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1617 Each operation applies to all devices listed until the next
1620 If an array is using a write-intent bitmap, then devices which have
1621 been removed can be re\-added in a way that avoids a full
1622 reconstruction but instead just updates the blocks that have changed
1623 since the device was removed. For arrays with persistent metadata
1624 (superblocks) this is done automatically. For arrays created with
1626 mdadm needs to be told that this device we removed recently with
1629 Devices can only be removed from an array if they are not in active
1630 use, i.e. that must be spares or failed devices. To remove an active
1631 device, it must first be marked as
1637 .BR \-Q ", " \-\-query
1638 Examine a device to see
1639 (1) if it is an md device and (2) if it is a component of an md
1641 Information about what is discovered is presented.
1644 .BR \-D ", " \-\-detail
1645 Print details of one or more md devices.
1648 .BR \-\-detail\-platform
1649 Print details of the platform's RAID capabilities (firmware / hardware
1650 topology) for a given metadata format. If used without an argument, mdadm
1651 will scan all controllers looking for their capabilities. Otherwise, mdadm
1652 will only look at the controller specified by the argument in the form of an
1653 absolute filepath or a link, e.g.
1654 .IR /sys/devices/pci0000:00/0000:00:1f.2 .
1657 .BR \-Y ", " \-\-export
1660 .BR \-\-detail-platform ,
1664 output will be formatted as
1666 pairs for easy import into the environment.
1672 indicates whether an array was started
1674 or not, which may include a reason
1675 .RB ( unsafe ", " nothing ", " no ).
1678 indicates if the array is expected on this host
1680 or seems to be from elsewhere
1684 .BR \-E ", " \-\-examine
1685 Print contents of the metadata stored on the named device(s).
1686 Note the contrast between
1691 applies to devices which are components of an array, while
1693 applies to a whole array which is currently active.
1696 If an array was created on a SPARC machine with a 2.2 Linux kernel
1697 patched with RAID support, the superblock will have been created
1698 incorrectly, or at least incompatibly with 2.4 and later kernels.
1703 will fix the superblock before displaying it. If this appears to do
1704 the right thing, then the array can be successfully assembled using
1705 .BR "\-\-assemble \-\-update=sparc2.2" .
1708 .BR \-X ", " \-\-examine\-bitmap
1709 Report information about a bitmap file.
1710 The argument is either an external bitmap file or an array component
1711 in case of an internal bitmap. Note that running this on an array
1714 does not report the bitmap for that array.
1717 .B \-\-examine\-badblocks
1718 List the bad-blocks recorded for the device, if a bad-blocks list has
1719 been configured. Currently only
1723 metadata support bad-blocks lists.
1726 .BI \-\-dump= directory
1728 .BI \-\-restore= directory
1729 Save metadata from lists devices, or restore metadata to listed devices.
1732 .BR \-R ", " \-\-run
1733 start a partially assembled array. If
1735 did not find enough devices to fully start the array, it might leaving
1736 it partially assembled. If you wish, you can then use
1738 to start the array in degraded mode.
1741 .BR \-S ", " \-\-stop
1742 deactivate array, releasing all resources.
1745 .BR \-o ", " \-\-readonly
1746 mark array as readonly.
1749 .BR \-w ", " \-\-readwrite
1750 mark array as readwrite.
1753 .B \-\-zero\-superblock
1754 If the device contains a valid md superblock, the block is
1755 overwritten with zeros. With
1757 the block where the superblock would be is overwritten even if it
1758 doesn't appear to be valid.
1761 Be careful when calling \-\-zero\-superblock with clustered raid. Make sure
1762 the array isn't used or assembled in another cluster node before executing it.
1765 .B \-\-kill\-subarray=
1766 If the device is a container and the argument to \-\-kill\-subarray
1767 specifies an inactive subarray in the container, then the subarray is
1768 deleted. Deleting all subarrays will leave an 'empty-container' or
1769 spare superblock on the drives. See
1770 .B \-\-zero\-superblock
1772 removing a superblock. Note that some formats depend on the subarray
1773 index for generating a UUID, this command will fail if it would change
1774 the UUID of an active subarray.
1777 .B \-\-update\-subarray=
1778 If the device is a container and the argument to \-\-update\-subarray
1779 specifies a subarray in the container, then attempt to update the given
1780 superblock field in the subarray. See below in
1785 .BR \-t ", " \-\-test
1790 is set to reflect the status of the device. See below in
1795 .BR \-W ", " \-\-wait
1796 For each md device given, wait for any resync, recovery, or reshape
1797 activity to finish before returning.
1799 will return with success if it actually waited for every device
1800 listed, otherwise it will return failure.
1804 For each md device given, or each device in /proc/mdstat if
1806 is given, arrange for the array to be marked clean as soon as possible.
1808 will return with success if the array uses external metadata and we
1809 successfully waited. For native arrays, this returns immediately as the
1810 kernel handles dirty-clean transitions at shutdown. No action is taken
1811 if safe-mode handling is disabled.
1815 Set the "sync_action" for all md devices given to one of
1822 will abort any currently running action though some actions will
1823 automatically restart.
1826 will abort any current action and ensure no other action starts
1836 .BR "SCRUBBING AND MISMATCHES" .
1838 .SH For Incremental Assembly mode:
1840 .BR \-\-rebuild\-map ", " \-r
1841 Rebuild the map file
1845 uses to help track which arrays are currently being assembled.
1848 .BR \-\-run ", " \-R
1849 Run any array assembled as soon as a minimal number of devices is
1850 available, rather than waiting until all expected devices are present.
1853 .BR \-\-scan ", " \-s
1854 Only meaningful with
1858 file for arrays that are being incrementally assembled and will try to
1859 start any that are not already started. If any such array is listed
1862 as requiring an external bitmap, that bitmap will be attached first.
1865 .BR \-\-fail ", " \-f
1866 This allows the hot-plug system to remove devices that have fully disappeared
1867 from the kernel. It will first fail and then remove the device from any
1868 array it belongs to.
1869 The device name given should be a kernel device name such as "sda",
1875 Only used with \-\-fail. The 'path' given will be recorded so that if
1876 a new device appears at the same location it can be automatically
1877 added to the same array. This allows the failed device to be
1878 automatically replaced by a new device without metadata if it appears
1879 at specified path. This option is normally only set by an
1883 .SH For Monitor mode:
1885 .BR \-m ", " \-\-mail
1886 Give a mail address to send alerts to.
1889 .BR \-p ", " \-\-program ", " \-\-alert
1890 Give a program to be run whenever an event is detected.
1893 .BR \-y ", " \-\-syslog
1894 Cause all events to be reported through 'syslog'. The messages have
1895 facility of 'daemon' and varying priorities.
1898 .BR \-d ", " \-\-delay
1899 Give a delay in seconds.
1901 polls the md arrays and then waits this many seconds before polling
1902 again. The default is 60 seconds. Since 2.6.16, there is no need to
1903 reduce this as the kernel alerts
1905 immediately when there is any change.
1908 .BR \-r ", " \-\-increment
1909 Give a percentage increment.
1911 will generate RebuildNN events with the given percentage increment.
1914 .BR \-f ", " \-\-daemonise
1917 to run as a background daemon if it decides to monitor anything. This
1918 causes it to fork and run in the child, and to disconnect from the
1919 terminal. The process id of the child is written to stdout.
1922 which will only continue monitoring if a mail address or alert program
1923 is found in the config file.
1926 .BR \-i ", " \-\-pid\-file
1929 is running in daemon mode, write the pid of the daemon process to
1930 the specified file, instead of printing it on standard output.
1933 .BR \-1 ", " \-\-oneshot
1934 Check arrays only once. This will generate
1936 events and more significantly
1942 .B " mdadm \-\-monitor \-\-scan \-1"
1944 from a cron script will ensure regular notification of any degraded arrays.
1947 .BR \-t ", " \-\-test
1950 alert for every array found at startup. This alert gets mailed and
1951 passed to the alert program. This can be used for testing that alert
1952 message do get through successfully.
1956 This inhibits the functionality for moving spares between arrays.
1957 Only one monitoring process started with
1959 but without this flag is allowed, otherwise the two could interfere
1966 .B mdadm \-\-assemble
1967 .I md-device options-and-component-devices...
1970 .B mdadm \-\-assemble \-\-scan
1971 .I md-devices-and-options...
1974 .B mdadm \-\-assemble \-\-scan
1978 This usage assembles one or more RAID arrays from pre-existing components.
1979 For each array, mdadm needs to know the md device, the identity of the
1980 array, and the number of component devices. These can be found in a number of ways.
1982 In the first usage example (without the
1984 the first device given is the md device.
1985 In the second usage example, all devices listed are treated as md
1986 devices and assembly is attempted.
1987 In the third (where no devices are listed) all md devices that are
1988 listed in the configuration file are assembled. If no arrays are
1989 described by the configuration file, then any arrays that
1990 can be found on unused devices will be assembled.
1992 If precisely one device is listed, but
1998 was given and identity information is extracted from the configuration file.
2000 The identity can be given with the
2006 option, will be taken from the md-device record in the config file, or
2007 will be taken from the super block of the first component-device
2008 listed on the command line.
2010 Devices can be given on the
2012 command line or in the config file. Only devices which have an md
2013 superblock which contains the right identity will be considered for
2016 The config file is only used if explicitly named with
2018 or requested with (a possibly implicit)
2020 In the latter case, the default config file is used. See
2026 is not given, then the config file will only be used to find the
2027 identity of md arrays.
2029 Normally the array will be started after it is assembled. However if
2031 is not given and not all expected drives were listed, then the array
2032 is not started (to guard against usage errors). To insist that the
2033 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
2042 does not create any entries in
2046 It does record information in
2050 to choose the correct name.
2054 detects that udev is not configured, it will create the devices in
2058 In Linux kernels prior to version 2.6.28 there were two distinct
2059 types of md devices that could be created: one that could be
2060 partitioned using standard partitioning tools and one that could not.
2061 Since 2.6.28 that distinction is no longer relevant as both types of
2062 devices can be partitioned.
2064 will normally create the type that originally could not be partitioned
2065 as it has a well-defined major number (9).
2067 Prior to 2.6.28, it is important that mdadm chooses the correct type
2068 of array device to use. This can be controlled with the
2070 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
2071 to use a partitionable device rather than the default.
2073 In the no-udev case, the value given to
2075 can be suffixed by a number. This tells
2077 to create that number of partition devices rather than the default of 4.
2081 can also be given in the configuration file as a word starting
2083 on the ARRAY line for the relevant array.
2090 and no devices are listed,
2092 will first attempt to assemble all the arrays listed in the config
2095 If no arrays are listed in the config (other than those marked
2097 it will look through the available devices for possible arrays and
2098 will try to assemble anything that it finds. Arrays which are tagged
2099 as belonging to the given homehost will be assembled and started
2100 normally. Arrays which do not obviously belong to this host are given
2101 names that are expected not to conflict with anything local, and are
2102 started "read-auto" so that nothing is written to any device until the
2103 array is written to. i.e. automatic resync etc is delayed.
2107 finds a consistent set of devices that look like they should comprise
2108 an array, and if the superblock is tagged as belonging to the given
2109 home host, it will automatically choose a device name and try to
2110 assemble the array. If the array uses version-0.90 metadata, then the
2112 number as recorded in the superblock is used to create a name in
2116 If the array uses version-1 metadata, then the
2118 from the superblock is used to similarly create a name in
2120 (the name will have any 'host' prefix stripped first).
2122 This behaviour can be modified by the
2126 configuration file. This line can indicate that specific metadata
2127 type should, or should not, be automatically assembled. If an array
2128 is found which is not listed in
2130 and has a metadata format that is denied by the
2132 line, then it will not be assembled.
2135 line can also request that all arrays identified as being for this
2136 homehost should be assembled regardless of their metadata type.
2139 for further details.
2141 Note: Auto-assembly cannot be used for assembling and activating some
2142 arrays which are undergoing reshape. In particular as the
2144 cannot be given, any reshape which requires a backup file to continue
2145 cannot be started by auto-assembly. An array which is growing to more
2146 devices and has passed the critical section can be assembled using
2157 .BI \-\-raid\-devices= Z
2161 This usage is similar to
2163 The difference is that it creates an array without a superblock. With
2164 these arrays there is no difference between initially creating the array and
2165 subsequently assembling the array, except that hopefully there is useful
2166 data there in the second case.
2168 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
2169 one of their synonyms. All devices must be listed and the array will
2170 be started once complete. It will often be appropriate to use
2171 .B \-\-assume\-clean
2172 with levels raid1 or raid10.
2183 .BI \-\-raid\-devices= Z
2187 This usage will initialise a new md array, associate some devices with
2188 it, and activate the array.
2190 The named device will normally not exist when
2191 .I "mdadm \-\-create"
2192 is run, but will be created by
2194 once the array becomes active.
2196 The max length md-device name is limited to 32 characters.
2197 Different metadata types have more strict limitation
2198 (like IMSM where only 16 characters are allowed).
2199 For that reason, long name could be truncated or rejected, it depends on metadata policy.
2201 As devices are added, they are checked to see if they contain RAID
2202 superblocks or filesystems. They are also checked to see if the variance in
2203 device size exceeds 1%.
2205 If any discrepancy is found, the array will not automatically be run, though
2208 can override this caution.
2210 To create a "degraded" array in which some devices are missing, simply
2211 give the word "\fBmissing\fP"
2212 in place of a device name. This will cause
2214 to leave the corresponding slot in the array empty.
2215 For a RAID4 or RAID5 array at most one slot can be
2216 "\fBmissing\fP"; for a RAID6 array at most two slots.
2217 For a RAID1 array, only one real device needs to be given. All of the
2221 When creating a RAID5 array,
2223 will automatically create a degraded array with an extra spare drive.
2224 This is because building the spare into a degraded array is in general
2225 faster than resyncing the parity on a non-degraded, but not clean,
2226 array. This feature can be overridden with the
2230 When creating an array with version-1 metadata a name for the array is
2232 If this is not given with the
2236 will choose a name based on the last component of the name of the
2237 device being created. So if
2239 is being created, then the name
2244 is being created, then the name
2248 When creating a partition based array, using
2250 with version-1.x metadata, the partition type should be set to
2252 (non fs-data). This type of selection allows for greater precision since
2253 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
2254 might create problems in the event of array recovery through a live cdrom.
2256 A new array will normally get a randomly assigned 128bit UUID which is
2257 very likely to be unique. If you have a specific need, you can choose
2258 a UUID for the array by giving the
2260 option. Be warned that creating two arrays with the same UUID is a
2261 recipe for disaster. Also, using
2263 when creating a v0.90 array will silently override any
2268 .\"option is given, it is not necessary to list any component devices in this command.
2269 .\"They can be added later, before a
2273 .\"is given, the apparent size of the smallest drive given is used.
2275 If the array type supports a write-intent bitmap, and if the devices
2276 in the array exceed 100G is size, an internal write-intent bitmap
2277 will automatically be added unless some other option is explicitly
2280 option or a different consistency policy is selected with the
2281 .B \-\-consistency\-policy
2282 option. In any case, space for a bitmap will be reserved so that one
2283 can be added later with
2284 .BR "\-\-grow \-\-bitmap=internal" .
2286 If the metadata type supports it (currently only 1.x and IMSM metadata),
2287 space will be allocated to store a bad block list. This allows a modest
2288 number of bad blocks to be recorded, allowing the drive to remain in
2289 service while only partially functional.
2291 When creating an array within a
2294 can be given either the list of devices to use, or simply the name of
2295 the container. The former case gives control over which devices in
2296 the container will be used for the array. The latter case allows
2298 to automatically choose which devices to use based on how much spare
2301 The General Management options that are valid with
2306 insist on running the array even if some devices look like they might
2311 start the array in readonly mode.
2318 .I options... devices...
2321 This usage will allow individual devices in an array to be failed,
2322 removed or added. It is possible to perform multiple operations with
2323 on command. For example:
2325 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
2331 and will then remove it from the array and finally add it back
2332 in as a spare. However, only one md array can be affected by a single
2335 When a device is added to an active array, mdadm checks to see if it
2336 has metadata on it which suggests that it was recently a member of the
2337 array. If it does, it tries to "re\-add" the device. If there have
2338 been no changes since the device was removed, or if the array has a
2339 write-intent bitmap which has recorded whatever changes there were,
2340 then the device will immediately become a full member of the array and
2341 those differences recorded in the bitmap will be resolved.
2351 MISC mode includes a number of distinct operations that
2352 operate on distinct devices. The operations are:
2355 The device is examined to see if it is
2356 (1) an active md array, or
2357 (2) a component of an md array.
2358 The information discovered is reported.
2362 The device should be an active md device.
2364 will display a detailed description of the array.
2368 will cause the output to be less detailed and the format to be
2369 suitable for inclusion in
2373 will normally be 0 unless
2375 failed to get useful information about the device(s); however, if the
2377 option is given, then the exit status will be:
2381 The array is functioning normally.
2384 The array has at least one failed device.
2387 The array has multiple failed devices such that it is unusable.
2390 There was an error while trying to get information about the device.
2394 .B \-\-detail\-platform
2395 Print detail of the platform's RAID capabilities (firmware / hardware
2396 topology). If the metadata is specified with
2400 then the return status will be:
2404 metadata successfully enumerated its platform components on this system
2407 metadata is platform independent
2410 metadata failed to find its platform components on this system
2414 .B \-\-update\-subarray=
2415 If the device is a container and the argument to \-\-update\-subarray
2416 specifies a subarray in the container, then attempt to update the given
2417 superblock field in the subarray. Similar to updating an array in
2418 "assemble" mode, the field to update is selected by
2422 option. The supported options are
2432 option updates the subarray name in the metadata, it may not affect the
2433 device node name or the device node symlink until the subarray is
2434 re\-assembled. If updating
2436 would change the UUID of an active subarray this operation is blocked,
2437 and the command will end in an error.
2443 options enable and disable PPL in the metadata. Currently supported only for
2450 options enable and disable write-intent bitmap in the metadata. Currently supported only for
2455 The device should be a component of an md array.
2457 will read the md superblock of the device and display the contents.
2462 is given, then multiple devices that are components of the one array
2463 are grouped together and reported in a single entry suitable
2469 without listing any devices will cause all devices listed in the
2470 config file to be examined.
2473 .BI \-\-dump= directory
2474 If the device contains RAID metadata, a file will be created in the
2476 and the metadata will be written to it. The file will be the same
2477 size as the device and will have the metadata written at the
2478 same location as it exists in the device. However, the file will be "sparse" so
2479 that only those blocks containing metadata will be allocated. The
2480 total space used will be small.
2482 The filename used in the
2484 will be the base name of the device. Further, if any links appear in
2486 which point to the device, then hard links to the file will be created
2493 Multiple devices can be listed and their metadata will all be stored
2494 in the one directory.
2497 .BI \-\-restore= directory
2498 This is the reverse of
2501 will locate a file in the directory that has a name appropriate for
2502 the given device and will restore metadata from it. Names that match
2504 names are preferred, however if two of those refer to different files,
2506 will not choose between them but will abort the operation.
2508 If a file name is given instead of a
2512 will restore from that file to a single device, always provided the
2513 size of the file matches that of the device, and the file contains
2517 The devices should be active md arrays which will be deactivated, as
2518 long as they are not currently in use.
2522 This will fully activate a partially assembled md array.
2526 This will mark an active array as read-only, providing that it is
2527 not currently being used.
2533 array back to being read/write.
2537 For all operations except
2540 will cause the operation to be applied to all arrays listed in
2545 causes all devices listed in the config file to be examined.
2548 .BR \-b ", " \-\-brief
2549 Be less verbose. This is used with
2557 gives an intermediate level of verbosity.
2563 .B mdadm \-\-monitor
2564 .I options... devices...
2567 Monitor option can work in two modes:
2569 system wide mode, follow all md devices based on
2572 follow only specified MD devices in command line.
2576 indicates system wide mode. Option causes the
2578 to track all md devices that appear in
2580 If it is not set, then at least one
2584 Monitor usage causes
2586 to periodically poll a number of md arrays and to report on any events
2591 will work as long as there is an active array with redundancy and it is defined to follow (for
2593 every array is followed).
2595 As well as reporting events,
2597 may move a spare drive from one array to another if they are in the
2602 and if the destination array has a failed drive but no spares.
2604 The result of monitoring the arrays is the generation of events.
2605 These events are passed to a separate program (if specified) and may
2606 be mailed to a given E-mail address.
2608 When passing events to a program, the program is run once for each event,
2609 and is given 2 or 3 command-line arguments: the first is the
2610 name of the event (see below), the second is the name of the
2611 md device which is affected, and the third is the name of a related
2612 device if relevant (such as a component device that has failed).
2620 address must be specified on the
2621 command line or in the config file. If neither are available, then
2623 will not monitor anything.
2624 For devices given directly in command line, without
2628 specified, each event is reported to
2631 Note: For systems where
2633 is configured via systemd,
2634 .B mdmonitor(mdmonitor.service)
2635 should be configured. The service is designed to be primary solution for array monitoring,
2636 it is configured to work in system wide mode.
2637 It is automatically started and stopped according to current state and types of MD arrays in system.
2638 The service may require additional configuration, like
2642 That should be done in
2645 The different events are:
2649 .B DeviceDisappeared
2650 An md array which previously was configured appears to no longer be
2651 configured. (syslog priority: Critical)
2655 was told to monitor an array which is RAID0 or Linear, then it will
2657 .B DeviceDisappeared
2658 with the extra information
2660 This is because RAID0 and Linear do not support the device-failed,
2661 hot-spare and resync operations which are monitored.
2665 An md array started reconstruction (e.g. recovery, resync, reshape,
2666 check, repair). (syslog priority: Warning)
2672 is a two-digit number (eg. 05, 48). This indicates that the rebuild
2673 has reached that percentage of the total. The events are generated
2674 at a fixed increment from 0. The increment size may be specified with
2675 a command-line option (the default is 20). (syslog priority: Warning)
2679 An md array that was rebuilding, isn't any more, either because it
2680 finished normally or was aborted. (syslog priority: Warning)
2684 An active component device of an array has been marked as
2685 faulty. (syslog priority: Critical)
2689 A spare component device which was being rebuilt to replace a faulty
2690 device has failed. (syslog priority: Critical)
2694 A spare component device which was being rebuilt to replace a faulty
2695 device has been successfully rebuilt and has been made active.
2696 (syslog priority: Info)
2700 A new md array has been detected in the
2702 file. (syslog priority: Info)
2706 A newly noticed array appears to be degraded. This message is not
2709 notices a drive failure which causes degradation, but only when
2711 notices that an array is degraded when it first sees the array.
2712 (syslog priority: Critical)
2716 A spare drive has been moved from one array in a
2720 to another to allow a failed drive to be replaced.
2721 (syslog priority: Info)
2727 has been told, via the config file, that an array should have a certain
2728 number of spare devices, and
2730 detects that it has fewer than this number when it first sees the
2731 array, it will report a
2734 (syslog priority: Warning)
2738 An array was found at startup, and the
2741 (syslog priority: Info)
2751 cause Email to be sent. All events cause the program to be run.
2752 The program is run with two or three arguments: the event
2753 name, the array device and possibly a second device.
2755 Each event has an associated array device (e.g.
2757 and possibly a second device. For
2762 the second device is the relevant component device.
2765 the second device is the array that the spare was moved from.
2769 to move spares from one array to another, the different arrays need to
2770 be labeled with the same
2772 or the spares must be allowed to migrate through matching POLICY domains
2773 in the configuration file. The
2775 name can be any string; it is only necessary that different spare
2776 groups use different names.
2780 detects that an array in a spare group has fewer active
2781 devices than necessary for the complete array, and has no spare
2782 devices, it will look for another array in the same spare group that
2783 has a full complement of working drives and a spare. It will then
2784 attempt to remove the spare from the second array and add it to the
2786 If the removal succeeds but the adding fails, then it is added back to
2789 If the spare group for a degraded array is not defined,
2791 will look at the rules of spare migration specified by POLICY lines in
2793 and then follow similar steps as above if a matching spare is found.
2796 The GROW mode is used for changing the size or shape of an active
2799 During the kernel 2.6 era the following changes were added:
2801 change the "size" attribute for RAID1, RAID4, RAID5 and RAID6.
2803 increase or decrease the "raid\-devices" attribute of RAID0, RAID1, RAID4,
2806 change the chunk-size and layout of RAID0, RAID4, RAID5, RAID6 and RAID10.
2808 convert between RAID1 and RAID5, between RAID5 and RAID6, between
2809 RAID0, RAID4, and RAID5, and between RAID0 and RAID10 (in the near-2 mode).
2811 add a write-intent bitmap to any array which supports these bitmaps, or
2812 remove a write-intent bitmap from such an array.
2814 change the array's consistency policy.
2817 Using GROW on containers is currently supported only for Intel's IMSM
2818 container format. The number of devices in a container can be
2819 increased - which affects all arrays in the container - or an array
2820 in a container can be converted between levels where those levels are
2821 supported by the container, and the conversion is on of those listed
2827 Intel's native checkpointing doesn't use
2829 option and it is transparent for assembly feature.
2831 Roaming between Windows(R) and Linux systems for IMSM metadata is not
2832 supported during grow process.
2834 When growing a raid0 device, the new component disk size (or external
2835 backup size) should be larger than LCM(old, new) * chunk-size * 2,
2836 where LCM() is the least common multiple of the old and new count of
2837 component disks, and "* 2" comes from the fact that mdadm refuses to
2838 use more than half of a spare device for backup space.
2841 Normally when an array is built the "size" is taken from the smallest
2842 of the drives. If all the small drives in an arrays are, over time,
2843 removed and replaced with larger drives, then you could have an
2844 array of large drives with only a small amount used. In this
2845 situation, changing the "size" with "GROW" mode will allow the extra
2846 space to start being used. If the size is increased in this way, a
2847 "resync" process will start to make sure the new parts of the array
2850 Note that when an array changes size, any filesystem that may be
2851 stored in the array will not automatically grow or shrink to use or
2852 vacate the space. The
2853 filesystem will need to be explicitly told to use the extra space
2854 after growing, or to reduce its size
2856 to shrinking the array.
2858 Also, the size of an array cannot be changed while it has an active
2859 bitmap. If an array has a bitmap, it must be removed before the size
2860 can be changed. Once the change is complete a new bitmap can be created.
2865 is not yet supported for external file bitmap.
2867 .SS RAID\-DEVICES CHANGES
2869 A RAID1 array can work with any number of devices from 1 upwards
2870 (though 1 is not very useful). There may be times which you want to
2871 increase or decrease the number of active devices. Note that this is
2872 different to hot-add or hot-remove which changes the number of
2875 When reducing the number of devices in a RAID1 array, the slots which
2876 are to be removed from the array must already be vacant. That is, the
2877 devices which were in those slots must be failed and removed.
2879 When the number of devices is increased, any hot spares that are
2880 present will be activated immediately.
2882 Changing the number of active devices in a RAID5 or RAID6 is much more
2883 effort. Every block in the array will need to be read and written
2884 back to a new location. From 2.6.17, the Linux Kernel is able to
2885 increase the number of devices in a RAID5 safely, including restarting
2886 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2887 increase or decrease the number of devices in a RAID5 or RAID6.
2889 From 2.6.35, the Linux Kernel is able to convert a RAID0 in to a RAID4
2892 uses this functionality and the ability to add
2893 devices to a RAID4 to allow devices to be added to a RAID0. When
2894 requested to do this,
2896 will convert the RAID0 to a RAID4, add the necessary disks and make
2897 the reshape happen, and then convert the RAID4 back to RAID0.
2899 When decreasing the number of devices, the size of the array will also
2900 decrease. If there was data in the array, it could get destroyed and
2901 this is not reversible, so you should firstly shrink the filesystem on
2902 the array to fit within the new size. To help prevent accidents,
2904 requires that the size of the array be decreased first with
2905 .BR "mdadm --grow --array-size" .
2906 This is a reversible change which simply makes the end of the array
2907 inaccessible. The integrity of any data can then be checked before
2908 the non-reversible reduction in the number of devices is request.
2910 When relocating the first few stripes on a RAID5 or RAID6, it is not
2911 possible to keep the data on disk completely consistent and
2912 crash-proof. To provide the required safety, mdadm disables writes to
2913 the array while this "critical section" is reshaped, and takes a
2914 backup of the data that is in that section. For grows, this backup may be
2915 stored in any spare devices that the array has, however it can also be
2916 stored in a separate file specified with the
2918 option, and is required to be specified for shrinks, RAID level
2919 changes and layout changes. If this option is used, and the system
2920 does crash during the critical period, the same file must be passed to
2922 to restore the backup and reassemble the array. When shrinking rather
2923 than growing the array, the reshape is done from the end towards the
2924 beginning, so the "critical section" is at the end of the reshape.
2928 Changing the RAID level of any array happens instantaneously. However
2929 in the RAID5 to RAID6 case this requires a non-standard layout of the
2930 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2931 required before the change can be accomplished. So while the level
2932 change is instant, the accompanying layout change can take quite a
2935 is required. If the array is not simultaneously being grown or
2936 shrunk, so that the array size will remain the same - for example,
2937 reshaping a 3-drive RAID5 into a 4-drive RAID6 - the backup file will
2938 be used not just for a "critical section" but throughout the reshape
2939 operation, as described below under LAYOUT CHANGES.
2941 .SS CHUNK-SIZE AND LAYOUT CHANGES
2943 Changing the chunk-size or layout without also changing the number of
2944 devices as the same time will involve re-writing all blocks in-place.
2945 To ensure against data loss in the case of a crash, a
2947 must be provided for these changes. Small sections of the array will
2948 be copied to the backup file while they are being rearranged. This
2949 means that all the data is copied twice, once to the backup and once
2950 to the new layout on the array, so this type of reshape will go very
2953 If the reshape is interrupted for any reason, this backup file must be
2955 .B "mdadm --assemble"
2956 so the array can be reassembled. Consequently, the file cannot be
2957 stored on the device being reshaped.
2962 A write-intent bitmap can be added to, or removed from, an active
2963 array. Either internal bitmaps, or bitmaps stored in a separate file,
2964 can be added. Note that if you add a bitmap stored in a file which is
2965 in a filesystem that is on the RAID array being affected, the system
2966 will deadlock. The bitmap must be on a separate filesystem.
2968 .SS CONSISTENCY POLICY CHANGES
2970 The consistency policy of an active array can be changed by using the
2971 .B \-\-consistency\-policy
2972 option in Grow mode. Currently this works only for the
2976 policies and allows to enable or disable the RAID5 Partial Parity Log (PPL).
2978 .SH INCREMENTAL MODE
2982 .B mdadm \-\-incremental
2986 .RI [ optional-aliases-for-device ]
2989 .B mdadm \-\-incremental \-\-fail
2993 .B mdadm \-\-incremental \-\-rebuild\-map
2996 .B mdadm \-\-incremental \-\-run \-\-scan
2999 This mode is designed to be used in conjunction with a device
3000 discovery system. As devices are found in a system, they can be
3002 .B "mdadm \-\-incremental"
3003 to be conditionally added to an appropriate array.
3005 Conversely, it can also be used with the
3007 flag to do just the opposite and find whatever array a particular device
3008 is part of and remove the device from that array.
3010 If the device passed is a
3012 device created by a previous call to
3014 then rather than trying to add that device to an array, all the arrays
3015 described by the metadata of the container will be started.
3018 performs a number of tests to determine if the device is part of an
3019 array, and which array it should be part of. If an appropriate array
3020 is found, or can be created,
3022 adds the device to the array and conditionally starts the array.
3026 will normally only add devices to an array which were previously working
3027 (active or spare) parts of that array. The support for automatic
3028 inclusion of a new drive as a spare in some array requires
3029 a configuration through POLICY in config file.
3033 makes are as follow:
3035 Is the device permitted by
3037 That is, is it listed in a
3039 line in that file. If
3041 is absent then the default it to allow any device. Similarly if
3043 contains the special word
3045 then any device is allowed. Otherwise the device name given to
3047 or one of the aliases given, or an alias found in the filesystem,
3048 must match one of the names or patterns in a
3052 This is the only context where the aliases are used. They are
3053 usually provided by a
3056 .BR $env{DEVLINKS} .
3059 Does the device have a valid md superblock? If a specific metadata
3060 version is requested with
3064 then only that style of metadata is accepted, otherwise
3066 finds any known version of metadata. If no
3068 metadata is found, the device may be still added to an array
3069 as a spare if POLICY allows.
3073 Does the metadata match an expected array?
3074 The metadata can match in two ways. Either there is an array listed
3077 which identifies the array (either by UUID, by name, by device list,
3078 or by minor-number), or the array was created with a
3084 or on the command line.
3087 is not able to positively identify the array as belonging to the
3088 current host, the device will be rejected.
3093 keeps a list of arrays that it has partially assembled in
3095 If no array exists which matches
3096 the metadata on the new device,
3098 must choose a device name and unit number. It does this based on any
3101 or any name information stored in the metadata. If this name
3102 suggests a unit number, that number will be used, otherwise a free
3103 unit number will be chosen. Normally
3105 will prefer to create a partitionable array, however if the
3109 suggests that a non-partitionable array is preferred, that will be
3112 If the array is not found in the config file and its metadata does not
3113 identify it as belonging to the "homehost", then
3115 will choose a name for the array which is certain not to conflict with
3116 any array which does belong to this host. It does this be adding an
3117 underscore and a small number to the name preferred by the metadata.
3119 Once an appropriate array is found or created and the device is added,
3121 must decide if the array is ready to be started. It will
3122 normally compare the number of available (non-spare) devices to the
3123 number of devices that the metadata suggests need to be active. If
3124 there are at least that many, the array will be started. This means
3125 that if any devices are missing the array will not be restarted.
3131 in which case the array will be run as soon as there are enough
3132 devices present for the data to be accessible. For a RAID1, that
3133 means one device will start the array. For a clean RAID5, the array
3134 will be started as soon as all but one drive is present.
3136 Note that neither of these approaches is really ideal. If it can
3137 be known that all device discovery has completed, then
3141 can be run which will try to start all arrays that are being
3142 incrementally assembled. They are started in "read-auto" mode in
3143 which they are read-only until the first write request. This means
3144 that no metadata updates are made and no attempt at resync or recovery
3145 happens. Further devices that are found before the first write can
3146 still be added safely.
3149 This section describes environment variables that affect how mdadm
3154 Setting this value to 1 will prevent mdadm from automatically launching
3155 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
3161 does not create any device nodes in /dev, but leaves that task to
3165 appears not to be configured, or if this environment variable is set
3168 will create and devices that are needed.
3171 .B MDADM_NO_SYSTEMCTL
3176 is in use it will normally request
3178 to start various background tasks (particularly
3180 rather than forking and running them in the background. This can be
3181 suppressed by setting
3182 .BR MDADM_NO_SYSTEMCTL=1 .
3186 A key value of IMSM metadata is that it allows interoperability with
3187 boot ROMs on Intel platforms, and with other major operating systems.
3190 will only allow an IMSM array to be created or modified if detects
3191 that it is running on an Intel platform which supports IMSM, and
3192 supports the particular configuration of IMSM that is being requested
3193 (some functionality requires newer OROM support).
3195 These checks can be suppressed by setting IMSM_NO_PLATFORM=1 in the
3196 environment. This can be useful for testing or for disaster
3197 recovery. You should be aware that interoperability may be
3198 compromised by setting this value.
3200 These change can also be suppressed by adding
3201 .B mdadm.imsm.test=1
3202 to the kernel command line. This makes it easy to test IMSM
3203 code in a virtual machine that doesn't have IMSM virtual hardware.
3206 .B MDADM_GROW_ALLOW_OLD
3207 If an array is stopped while it is performing a reshape and that
3208 reshape was making use of a backup file, then when the array is
3211 will sometimes complain that the backup file is too old. If this
3212 happens and you are certain it is the right backup file, you can
3213 over-ride this check by setting
3214 .B MDADM_GROW_ALLOW_OLD=1
3219 Any string given in this variable is added to the start of the
3221 line in the config file, or treated as the whole
3223 line if none is given. It can be used to disable certain metadata
3226 is called from a boot script. For example
3228 .B " export MDADM_CONF_AUTO='-ddf -imsm'
3232 does not automatically assemble any DDF or
3233 IMSM arrays that are found. This can be useful on systems configured
3234 to manage such arrays with
3240 .B " mdadm \-\-query /dev/name-of-device"
3242 This will find out if a given device is a RAID array, or is part of
3243 one, and will provide brief information about the device.
3245 .B " mdadm \-\-assemble \-\-scan"
3247 This will assemble and start all arrays listed in the standard config
3248 file. This command will typically go in a system startup file.
3250 .B " mdadm \-\-stop \-\-scan"
3252 This will shut down all arrays that can be shut down (i.e. are not
3253 currently in use). This will typically go in a system shutdown script.
3255 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
3257 If (and only if) there is an Email address or program given in the
3258 standard config file, then
3259 monitor the status of all arrays listed in that file by
3260 polling them ever 2 minutes.
3262 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
3264 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
3267 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
3269 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
3271 This will create a prototype config file that describes currently
3272 active arrays that are known to be made from partitions of IDE or SCSI drives.
3273 This file should be reviewed before being used as it may
3274 contain unwanted detail.
3276 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
3278 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
3280 This will find arrays which could be assembled from existing IDE and
3281 SCSI whole drives (not partitions), and store the information in the
3282 format of a config file.
3283 This file is very likely to contain unwanted detail, particularly
3286 entries. It should be reviewed and edited before being used as an
3289 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
3291 .B " mdadm \-Ebsc partitions"
3293 Create a list of devices by reading
3294 .BR /proc/partitions ,
3295 scan these for RAID superblocks, and printout a brief listing of all
3298 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
3300 Scan all partitions and devices listed in
3301 .BR /proc/partitions
3304 out of all such devices with a RAID superblock with a minor number of 0.
3306 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /run/mdadm/mon.pid"
3308 If config file contains a mail address or alert program, run mdadm in
3309 the background in monitor mode monitoring all md devices. Also write
3310 pid of mdadm daemon to
3311 .BR /run/mdadm/mon.pid .
3313 .B " mdadm \-Iq /dev/somedevice"
3315 Try to incorporate newly discovered device into some array as
3318 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
3320 Rebuild the array map from any current arrays, and then start any that
3323 .B " mdadm /dev/md4 --fail detached --remove detached"
3325 Any devices which are components of /dev/md4 will be marked as faulty
3326 and then remove from the array.
3328 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4"
3332 which is currently a RAID5 array will be converted to RAID6. There
3333 should normally already be a spare drive attached to the array as a
3334 RAID6 needs one more drive than a matching RAID5.
3336 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
3338 Create a DDF array over 6 devices.
3340 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
3342 Create a RAID5 array over any 3 devices in the given DDF set. Use
3343 only 30 gigabytes of each device.
3345 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
3347 Assemble a pre-exist ddf array.
3349 .B " mdadm -I /dev/md/ddf1"
3351 Assemble all arrays contained in the ddf array, assigning names as
3354 .B " mdadm \-\-create \-\-help"
3356 Provide help about the Create mode.
3358 .B " mdadm \-\-config \-\-help"
3360 Provide help about the format of the config file.
3362 .B " mdadm \-\-help"
3364 Provide general help.
3374 lists all active md devices with information about them.
3376 uses this to find arrays when
3378 is given in Misc mode, and to monitor array reconstruction
3381 .SS {CONFFILE} (or {CONFFILE2})
3383 Default config file. See
3387 .SS {CONFFILE}.d (or {CONFFILE2}.d)
3389 Default directory containing configuration files. See
3396 mode is used, this file gets a list of arrays currently being created.
3401 understand two sorts of names for array devices.
3403 The first is the so-called 'standard' format name, which matches the
3404 names used by the kernel and which appear in
3407 The second sort can be freely chosen, but must reside in
3409 When giving a device name to
3411 to create or assemble an array, either full path name such as
3415 can be given, or just the suffix of the second sort of name, such as
3421 chooses device names during auto-assembly or incremental assembly, it
3422 will sometimes add a small sequence number to the end of the name to
3423 avoid conflicted between multiple arrays that have the same name. If
3425 can reasonably determine that the array really is meant for this host,
3426 either by a hostname in the metadata, or by the presence of the array
3429 then it will leave off the suffix if possible.
3430 Also if the homehost is specified as
3433 will only use a suffix if a different array of the same name already
3434 exists or is listed in the config file.
3436 The standard names for non-partitioned arrays (the only sort of md
3437 array available in 2.4 and earlier) are of the form
3441 where NN is a number.
3442 The standard names for partitionable arrays (as available from 2.6
3443 onwards) are of the form:
3447 Partition numbers should be indicated by adding "pMM" to these, thus "/dev/md/d1p2".
3449 From kernel version 2.6.28 the "non-partitioned array" can actually
3450 be partitioned. So the "md_d\fBNN\fP"
3451 names are no longer needed, and
3452 partitions such as "/dev/md\fBNN\fPp\fBXX\fP"
3455 From kernel version 2.6.29 standard names can be non-numeric following
3462 is any string. These names are supported by
3464 since version 3.3 provided they are enabled in
3469 was previously known as
3473 For further information on mdadm usage, MD and the various levels of
3476 .B https://raid.wiki.kernel.org/
3478 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
3480 The latest version of
3482 should always be available from
3484 .B https://www.kernel.org/pub/linux/utils/raid/mdadm/
projects / thirdparty / mdadm.git / blob