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 general help message or, after one of the above options, a
235 mode-specific help message.
239 Display more detailed help about command line parsing and some commonly
243 .BR \-V ", " \-\-version
244 Print version information for mdadm.
247 .BR \-v ", " \-\-verbose
248 Be more verbose about what is happening. This can be used twice to be
250 The extra verbosity currently only affects
251 .B \-\-detail \-\-scan
253 .BR "\-\-examine \-\-scan" .
256 .BR \-q ", " \-\-quiet
257 Avoid printing purely informative messages. With this,
259 will be silent unless there is something really important to report.
263 .BR \-f ", " \-\-force
264 Be more forceful about certain operations. See the various modes for
265 the exact meaning of this option in different contexts.
268 .BR \-c ", " \-\-config=
269 Specify the config file or directory. Default is to use
273 or if those are missing then
277 If the config file given is
279 then nothing will be read, but
281 will act as though the config file contained exactly
283 .B " DEVICE partitions containers"
287 to find a list of devices to scan, and
289 to find a list of containers to examine.
292 is given for the config file, then
294 will act as though the config file were empty.
296 If the name given is of a directory, then
298 will collect all the files contained in the directory with a name ending
301 sort them lexically, and process all of those files as config files.
304 .BR \-s ", " \-\-scan
307 for missing information.
308 In general, this option gives
310 permission to get any missing information (like component devices,
311 array devices, array identities, and alert destination) from the
312 configuration file (see previous option);
313 one exception is MISC mode when using
319 says to get a list of array devices from
323 .BR \-e ", " \-\-metadata=
324 Declare the style of RAID metadata (superblock) to be used. The
325 default is {DEFAULT_METADATA} for
327 and to guess for other operations.
328 The default can be overridden by setting the
337 .ie '{DEFAULT_METADATA}'0.90'
338 .IP "0, 0.90, default"
341 Use the original 0.90 format superblock. This format limits arrays to
342 28 component devices and limits component devices of levels 1 and
343 greater to 2 terabytes. It is also possible for there to be confusion
344 about whether the superblock applies to a whole device or just the
345 last partition, if that partition starts on a 64K boundary.
346 .ie '{DEFAULT_METADATA}'0.90'
347 .IP "1, 1.0, 1.1, 1.2"
349 .IP "1, 1.0, 1.1, 1.2 default"
350 Use the new version-1 format superblock. This has fewer restrictions.
351 It can easily be moved between hosts with different endian-ness, and a
352 recovery operation can be checkpointed and restarted. The different
353 sub-versions store the superblock at different locations on the
354 device, either at the end (for 1.0), at the start (for 1.1) or 4K from
355 the start (for 1.2). "1" is equivalent to "1.2" (the commonly
356 preferred 1.x format).
357 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
359 Use the "Industry Standard" DDF (Disk Data Format) format defined by
361 When creating a DDF array a
363 will be created, and normal arrays can be created in that container.
365 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
367 which is managed in a similar manner to DDF, and is supported by an
368 option-rom on some platforms:
370 .B https://www.intel.com/content/www/us/en/support/products/122484/memory-and-storage/ssd-software/intel-virtual-raid-on-cpu-intel-vroc.html
376 This will override any
378 setting in the config file and provides the identity of the host which
379 should be considered the home for any arrays.
381 When creating an array, the
383 will be recorded in the metadata. For version-1 superblocks, it will
384 be prefixed to the array name. For version-0.90 superblocks, part of
385 the SHA1 hash of the hostname will be stored in the later half of the
388 When reporting information about an array, any array which is tagged
389 for the given homehost will be reported as such.
391 When using Auto-Assemble, only arrays tagged for the given homehost
392 will be allowed to use 'local' names (i.e. not ending in '_' followed
393 by a digit string). See below under
394 .BR "Auto Assembly" .
396 The special name "\fBany\fP" can be used as a wild card. If an array
399 then the name "\fBany\fP" will be stored in the array and it can be
400 assembled in the same way on any host. If an array is assembled with
401 this option, then the homehost recorded on the array will be ignored.
407 needs to print the name for a device it normally finds the name in
409 which refers to the device and is shortest. When a path component is
413 will prefer a longer name if it contains that component. For example
414 .B \-\-prefer=by-uuid
415 will prefer a name in a subdirectory of
420 This functionality is currently only provided by
426 .B \-\-home\-cluster=
427 specifies the cluster name for the md device. The md device can be assembled
428 only on the cluster which matches the name specified. If this option is not
429 provided, mdadm tries to detect the cluster name automatically.
431 .SH For create, build, or grow:
434 .BR \-n ", " \-\-raid\-devices=
435 Specify the number of active devices in the array. This, plus the
436 number of spare devices (see below) must equal the number of
438 (including "\fBmissing\fP" devices)
439 that are listed on the command line for
441 Setting a value of 1 is probably
442 a mistake and so requires that
444 be specified first. A value of 1 will then be allowed for linear,
445 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
447 This number can only be changed using
449 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
450 the necessary support.
453 .BR \-x ", " \-\-spare\-devices=
454 Specify the number of spare (eXtra) devices in the initial array.
455 Spares can also be added
456 and removed later. The number of component devices listed
457 on the command line must equal the number of RAID devices plus the
458 number of spare devices.
461 .BR \-z ", " \-\-size=
462 Amount (in Kilobytes) of space to use from each drive in RAID levels 1/4/5/6/10
463 and for RAID 0 on external metadata.
464 This must be a multiple of the chunk size, and must leave about 128Kb
465 of space at the end of the drive for the RAID superblock.
466 If this is not specified
467 (as it normally is not) the smallest drive (or partition) sets the
468 size, though if there is a variance among the drives of greater than 1%, a warning is
471 A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes,
472 Megabytes, Gigabytes or Terabytes respectively.
474 Sometimes a replacement drive can be a little smaller than the
475 original drives though this should be minimised by IDEMA standards.
476 Such a replacement drive will be rejected by
478 To guard against this it can be useful to set the initial size
479 slightly smaller than the smaller device with the aim that it will
480 still be larger than any replacement.
482 This option can be used with
484 for determining initial size of an array. For external metadata,
485 it can be used on a volume, but not on a container itself.
486 Setting initial size of
488 array is only valid for external metadata.
490 This value can be set with
492 for RAID level 1/4/5/6/10 though
493 DDF arrays may not be able to support this.
494 RAID 0 array size cannot be changed.
495 If the array was created with a size smaller than the currently
496 active drives, the extra space can be accessed using
498 The size can be given as
500 which means to choose the largest size that fits on all current drives.
502 Before reducing the size of the array (with
503 .BR "\-\-grow \-\-size=" )
504 you should make sure that space isn't needed. If the device holds a
505 filesystem, you would need to resize the filesystem to use less space.
507 After reducing the array size you should check that the data stored in
508 the device is still available. If the device holds a filesystem, then
509 an 'fsck' of the filesystem is a minimum requirement. If there are
510 problems the array can be made bigger again with no loss with another
511 .B "\-\-grow \-\-size="
515 .BR \-Z ", " \-\-array\-size=
516 This is only meaningful with
518 and its effect is not persistent: when the array is stopped and
519 restarted the default array size will be restored.
521 Setting the array-size causes the array to appear smaller to programs
522 that access the data. This is particularly needed before reshaping an
523 array so that it will be smaller. As the reshape is not reversible,
524 but setting the size with
526 is, it is required that the array size is reduced as appropriate
527 before the number of devices in the array is reduced.
529 Before reducing the size of the array you should make sure that space
530 isn't needed. If the device holds a filesystem, you would need to
531 resize the filesystem to use less space.
533 After reducing the array size you should check that the data stored in
534 the device is still available. If the device holds a filesystem, then
535 an 'fsck' of the filesystem is a minimum requirement. If there are
536 problems the array can be made bigger again with no loss with another
537 .B "\-\-grow \-\-array\-size="
540 A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes,
541 Megabytes, Gigabytes or Terabytes respectively.
544 restores the apparent size of the array to be whatever the real
545 amount of available space is.
547 Clustered arrays do not support this parameter yet.
550 .BR \-c ", " \-\-chunk=
551 Specify chunk size of kilobytes. The default when creating an
552 array is 512KB. To ensure compatibility with earlier versions, the
553 default when building an array with no persistent metadata is 64KB.
554 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
556 RAID4, RAID5, RAID6, and RAID10 require the chunk size to be a power
557 of 2. In any case it must be a multiple of 4KB.
559 A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes,
560 Megabytes, Gigabytes or Terabytes respectively.
564 Specify rounding factor for a Linear array. The size of each
565 component will be rounded down to a multiple of this size.
566 This is a synonym for
568 but highlights the different meaning for Linear as compared to other
569 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
570 use, and is 0K (i.e. no rounding) in later kernels.
573 .BR \-l ", " \-\-level=
574 Set RAID level. When used with
576 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
577 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
578 Obviously some of these are synonymous.
582 metadata type is requested, only the
584 level is permitted, and it does not need to be explicitly given.
588 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
592 to change the RAID level in some cases. See LEVEL CHANGES below.
595 .BR \-p ", " \-\-layout=
596 This option configures the fine details of data layout for RAID5, RAID6,
597 and RAID10 arrays, and controls the failure modes for
599 It can also be used for working around a kernel bug with RAID0, but generally
600 doesn't need to be used explicitly.
602 The layout of the RAID5 parity block can be one of
603 .BR left\-asymmetric ,
604 .BR left\-symmetric ,
605 .BR right\-asymmetric ,
606 .BR right\-symmetric ,
607 .BR la ", " ra ", " ls ", " rs .
609 .BR left\-symmetric .
611 It is also possible to cause RAID5 to use a RAID4-like layout by
617 Finally for RAID5 there are DDF\-compatible layouts,
618 .BR ddf\-zero\-restart ,
619 .BR ddf\-N\-restart ,
621 .BR ddf\-N\-continue .
623 These same layouts are available for RAID6. There are also 4 layouts
624 that will provide an intermediate stage for converting between RAID5
625 and RAID6. These provide a layout which is identical to the
626 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
627 syndrome (the second 'parity' block used by RAID6) on the last device.
629 .BR left\-symmetric\-6 ,
630 .BR right\-symmetric\-6 ,
631 .BR left\-asymmetric\-6 ,
632 .BR right\-asymmetric\-6 ,
634 .BR parity\-first\-6 .
636 When setting the failure mode for level
639 .BR write\-transient ", " wt ,
640 .BR read\-transient ", " rt ,
641 .BR write\-persistent ", " wp ,
642 .BR read\-persistent ", " rp ,
644 .BR read\-fixable ", " rf ,
645 .BR clear ", " flush ", " none .
647 Each failure mode can be followed by a number, which is used as a period
648 between fault generation. Without a number, the fault is generated
649 once on the first relevant request. With a number, the fault will be
650 generated after that many requests, and will continue to be generated
651 every time the period elapses.
653 Multiple failure modes can be current simultaneously by using the
655 option to set subsequent failure modes.
657 "clear" or "none" will remove any pending or periodic failure modes,
658 and "flush" will clear any persistent faults.
660 The layout options for RAID10 are one of 'n', 'o' or 'f' followed
661 by a small number. The default is 'n2'. The supported options are:
664 signals 'near' copies. Multiple copies of one data block are at
665 similar offsets in different devices.
668 signals 'offset' copies. Rather than the chunks being duplicated
669 within a stripe, whole stripes are duplicated but are rotated by one
670 device so duplicate blocks are on different devices. Thus subsequent
671 copies of a block are in the next drive, and are one chunk further
676 (multiple copies have very different offsets).
677 See md(4) for more detail about 'near', 'offset', and 'far'.
679 The number is the number of copies of each datablock. 2 is normal, 3
680 can be useful. This number can be at most equal to the number of
681 devices in the array. It does not need to divide evenly into that
682 number (e.g. it is perfectly legal to have an 'n2' layout for an array
683 with an odd number of devices).
685 A bug introduced in Linux 3.14 means that RAID0 arrays
686 .B "with devices of differing sizes"
687 started using a different layout. This could lead to
688 data corruption. Since Linux 5.4 (and various stable releases that received
689 backports), the kernel will not accept such an array unless
690 a layout is explictly set. It can be set to
694 When creating a new array,
698 by default, so the layout does not normally need to be set.
699 An array created for either
703 will not be recognized by an (unpatched) kernel prior to 5.4. To create
704 a RAID0 array with devices of differing sizes that can be used on an
705 older kernel, you can set the layout to
707 This will use whichever layout the running kernel supports, so the data
708 on the array may become corrupt when changing kernel from pre-3.14 to a
711 When an array is converted between RAID5 and RAID6 an intermediate
712 RAID6 layout is used in which the second parity block (Q) is always on
713 the last device. To convert a RAID5 to RAID6 and leave it in this new
714 layout (which does not require re-striping) use
715 .BR \-\-layout=preserve .
716 This will try to avoid any restriping.
718 The converse of this is
719 .B \-\-layout=normalise
720 which will change a non-standard RAID6 layout into a more standard
727 (thus explaining the p of
731 .BR \-b ", " \-\-bitmap=
732 Specify a file to store a write-intent bitmap in. The file should not
735 is also given. The same file should be provided
736 when assembling the array. If the word
738 is given, then the bitmap is stored with the metadata on the array,
739 and so is replicated on all devices. If the word
743 mode, then any bitmap that is present is removed. If the word
745 is given, the array is created for a clustered environment. One bitmap
746 is created for each node as defined by the
748 parameter and are stored internally.
750 To help catch typing errors, the filename must contain at least one
751 slash ('/') if it is a real file (not 'internal' or 'none').
753 Note: external bitmaps are only known to work on ext2 and ext3.
754 Storing bitmap files on other filesystems may result in serious problems.
756 When creating an array on devices which are 100G or larger,
758 automatically adds an internal bitmap as it will usually be
759 beneficial. This can be suppressed with
761 or by selecting a different consistency policy with
762 .BR \-\-consistency\-policy .
765 .BR \-\-bitmap\-chunk=
766 Set the chunksize of the bitmap. Each bit corresponds to that many
767 Kilobytes of storage.
768 When using a file based bitmap, the default is to use the smallest
769 size that is at-least 4 and requires no more than 2^21 chunks.
772 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
773 fit the bitmap into the available space.
775 A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes,
776 Megabytes, Gigabytes or Terabytes respectively.
779 .BR \-W ", " \-\-write\-mostly
780 subsequent devices listed in a
785 command will be flagged as 'write\-mostly'. This is valid for RAID1
786 only and means that the 'md' driver will avoid reading from these
787 devices if at all possible. This can be useful if mirroring over a
791 .BR \-\-write\-behind=
792 Specify that write-behind mode should be enabled (valid for RAID1
793 only). If an argument is specified, it will set the maximum number
794 of outstanding writes allowed. The default value is 256.
795 A write-intent bitmap is required in order to use write-behind
796 mode, and write-behind is only attempted on drives marked as
801 subsequent devices listed in a
805 command will be flagged as 'failfast'. This is valid for RAID1 and
806 RAID10 only. IO requests to these devices will be encouraged to fail
807 quickly rather than cause long delays due to error handling. Also no
808 attempt is made to repair a read error on these devices.
810 If an array becomes degraded so that the 'failfast' device is the only
811 usable device, the 'failfast' flag will then be ignored and extended
812 delays will be preferred to complete failure.
814 The 'failfast' flag is appropriate for storage arrays which have a
815 low probability of true failure, but which may sometimes
816 cause unacceptable delays due to internal maintenance functions.
819 .BR \-\-assume\-clean
822 that the array pre-existed and is known to be clean. It can be useful
823 when trying to recover from a major failure as you can be sure that no
824 data will be affected unless you actually write to the array. It can
825 also be used when creating a RAID1 or RAID10 if you want to avoid the
826 initial resync, however this practice \(em while normally safe \(em is not
827 recommended. Use this only if you really know what you are doing.
829 When the devices that will be part of a new array were filled
830 with zeros before creation the operator knows the array is
831 actually clean. If that is the case, such as after running
832 badblocks, this argument can be used to tell mdadm the
833 facts the operator knows.
835 When an array is resized to a larger size with
836 .B "\-\-grow \-\-size="
837 the new space is normally resynced in that same way that the whole
838 array is resynced at creation. From Linux version 3.0,
840 can be used with that command to avoid the automatic resync.
843 .BR \-\-backup\-file=
846 is used to increase the number of raid-devices in a RAID5 or RAID6 if
847 there are no spare devices available, or to shrink, change RAID level
848 or layout. See the GROW MODE section below on RAID\-DEVICES CHANGES.
849 The file must be stored on a separate device, not on the RAID array
854 Arrays with 1.x metadata can leave a gap between the start of the
855 device and the start of array data. This gap can be used for various
856 metadata. The start of data is known as the
858 Normally an appropriate data offset is computed automatically.
859 However it can be useful to set it explicitly such as when re-creating
860 an array which was originally created using a different version of
862 which computed a different offset.
864 Setting the offset explicitly over-rides the default. The value given
865 is in Kilobytes unless a suffix of 'K', 'M', 'G' or 'T' is used to explicitly
866 indicate Kilobytes, Megabytes, Gigabytes or Terabytes respectively.
870 can also be used with
872 for some RAID levels (initially on RAID10). This allows the
873 data\-offset to be changed as part of the reshape process. When the
874 data offset is changed, no backup file is required as the difference
875 in offsets is used to provide the same functionality.
877 When the new offset is earlier than the old offset, the number of
878 devices in the array cannot shrink. When it is after the old offset,
879 the number of devices in the array cannot increase.
881 When creating an array,
885 In the case each member device is expected to have a offset appended
886 to the name, separated by a colon. This makes it possible to recreate
887 exactly an array which has varying data offsets (as can happen when
888 different versions of
890 are used to add different devices).
894 This option is complementary to the
895 .B \-\-freeze-reshape
896 option for assembly. It is needed when
898 operation is interrupted and it is not restarted automatically due to
899 .B \-\-freeze-reshape
900 usage during array assembly. This option is used together with
904 ) command and device for a pending reshape to be continued.
905 All parameters required for reshape continuation will be read from array metadata.
909 .BR \-\-backup\-file=
910 option to be set, continuation option will require to have exactly the same
911 backup file given as well.
913 Any other parameter passed together with
915 option will be ignored.
918 .BR \-N ", " \-\-name=
921 for the array. This is currently only effective when creating an
922 array with a version-1 superblock, or an array in a DDF container.
923 The name is a simple textual string that can be used to identify array
924 components when assembling. If name is needed but not specified, it
925 is taken from the basename of the device that is being created.
937 run the array, even if some of the components
938 appear to be active in another array or filesystem. Normally
940 will ask for confirmation before including such components in an
941 array. This option causes that question to be suppressed.
944 .BR \-f ", " \-\-force
947 accept the geometry and layout specified without question. Normally
949 will not allow creation of an array with only one device, and will try
950 to create a RAID5 array with one missing drive (as this makes the
951 initial resync work faster). With
954 will not try to be so clever.
957 .BR \-o ", " \-\-readonly
960 rather than read-write as normal. No writes will be allowed to the
961 array, and no resync, recovery, or reshape will be started. It works with
962 Create, Assemble, Manage and Misc mode.
965 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
966 Instruct mdadm how to create the device file if needed, possibly allocating
967 an unused minor number. "md" causes a non-partitionable array
968 to be used (though since Linux 2.6.28, these array devices are in fact
969 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
970 later) to be used. "yes" requires the named md device to have
971 a 'standard' format, and the type and minor number will be determined
972 from this. With mdadm 3.0, device creation is normally left up to
974 so this option is unlikely to be needed.
975 See DEVICE NAMES below.
977 The argument can also come immediately after
982 is not given on the command line or in the config file, then
988 is also given, then any
990 entries in the config file will override the
992 instruction given on the command line.
994 For partitionable arrays,
996 will create the device file for the whole array and for the first 4
997 partitions. A different number of partitions can be specified at the
998 end of this option (e.g.
1000 If the device name ends with a digit, the partition names add a 'p',
1002 .IR /dev/md/home1p3 .
1003 If there is no trailing digit, then the partition names just have a
1005 .IR /dev/md/scratch3 .
1007 If the md device name is in a 'standard' format as described in DEVICE
1008 NAMES, then it will be created, if necessary, with the appropriate
1009 device number based on that name. If the device name is not in one of these
1010 formats, then a unused device number will be allocated. The device
1011 number will be considered unused if there is no active array for that
1012 number, and there is no entry in /dev for that number and with a
1013 non-standard name. Names that are not in 'standard' format are only
1014 allowed in "/dev/md/".
1016 This is meaningful with
1022 .BR \-a ", " "\-\-add"
1023 This option can be used in Grow mode in two cases.
1025 If the target array is a Linear array, then
1027 can be used to add one or more devices to the array. They
1028 are simply catenated on to the end of the array. Once added, the
1029 devices cannot be removed.
1033 option is being used to increase the number of devices in an array,
1036 can be used to add some extra devices to be included in the array.
1037 In most cases this is not needed as the extra devices can be added as
1038 spares first, and then the number of raid-disks can be changed.
1039 However for RAID0, it is not possible to add spares. So to increase
1040 the number of devices in a RAID0, it is necessary to set the new
1041 number of devices, and to add the new devices, in the same command.
1045 Only works when the array is for clustered environment. It specifies
1046 the maximum number of nodes in the cluster that will use this device
1047 simultaneously. If not specified, this defaults to 4.
1050 .BR \-\-write-journal
1051 Specify journal device for the RAID-4/5/6 array. The journal device
1052 should be a SSD with reasonable lifetime.
1056 Auto creation of symlinks in /dev to /dev/md, option --symlinks must
1057 be 'no' or 'yes' and work with --create and --build.
1060 .BR \-k ", " \-\-consistency\-policy=
1061 Specify how the array maintains consistency in case of unexpected shutdown.
1062 Only relevant for RAID levels with redundancy.
1063 Currently supported options are:
1068 Full resync is performed and all redundancy is regenerated when the array is
1069 started after unclean shutdown.
1073 Resync assisted by a write-intent bitmap. Implicitly selected when using
1078 For RAID levels 4/5/6, journal device is used to log transactions and replay
1079 after unclean shutdown. Implicitly selected when using
1080 .BR \-\-write\-journal .
1084 For RAID5 only, Partial Parity Log is used to close the write hole and
1085 eliminate resync. PPL is stored in the metadata region of RAID member drives,
1086 no additional journal drive is needed.
1089 Can be used with \-\-grow to change the consistency policy of an active array
1090 in some cases. See CONSISTENCY POLICY CHANGES below.
1097 .BR \-u ", " \-\-uuid=
1098 uuid of array to assemble. Devices which don't have this uuid are
1102 .BR \-m ", " \-\-super\-minor=
1103 Minor number of device that array was created for. Devices which
1104 don't have this minor number are excluded. If you create an array as
1105 /dev/md1, then all superblocks will contain the minor number 1, even if
1106 the array is later assembled as /dev/md2.
1108 Giving the literal word "dev" for
1112 to use the minor number of the md device that is being assembled.
1113 e.g. when assembling
1115 .B \-\-super\-minor=dev
1116 will look for super blocks with a minor number of 0.
1119 is only relevant for v0.90 metadata, and should not normally be used.
1125 .BR \-N ", " \-\-name=
1126 Specify the name of the array to assemble. This must be the name
1127 that was specified when creating the array. It must either match
1128 the name stored in the superblock exactly, or it must match
1131 prefixed to the start of the given name.
1134 .BR \-f ", " \-\-force
1135 Assemble the array even if the metadata on some devices appears to be
1138 cannot find enough working devices to start the array, but can find
1139 some devices that are recorded as having failed, then it will mark
1140 those devices as working so that the array can be started. This works only for
1141 native. For external metadata it allows to start dirty degraded RAID 4, 5, 6.
1142 An array which requires
1144 to be started may contain data corruption. Use it carefully.
1147 .BR \-R ", " \-\-run
1148 Attempt to start the array even if fewer drives were given than were
1149 present last time the array was active. Normally if not all the
1150 expected drives are found and
1152 is not used, then the array will be assembled but not started.
1155 an attempt will be made to start it anyway.
1159 This is the reverse of
1161 in that it inhibits the startup of array unless all expected drives
1162 are present. This is only needed with
1164 and can be used if the physical connections to devices are
1165 not as reliable as you would like.
1168 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
1169 See this option under Create and Build options.
1172 .BR \-b ", " \-\-bitmap=
1173 Specify the bitmap file that was given when the array was created. If
1176 bitmap, there is no need to specify this when assembling the array.
1179 .BR \-\-backup\-file=
1182 was used while reshaping an array (e.g. changing number of devices or
1183 chunk size) and the system crashed during the critical section, then the same
1185 must be presented to
1187 to allow possibly corrupted data to be restored, and the reshape
1191 .BR \-\-invalid\-backup
1192 If the file needed for the above option is not available for any
1193 reason an empty file can be given together with this option to
1194 indicate that the backup file is invalid. In this case the data that
1195 was being rearranged at the time of the crash could be irrecoverably
1196 lost, but the rest of the array may still be recoverable. This option
1197 should only be used as a last resort if there is no way to recover the
1202 .BR \-U ", " \-\-update=
1203 Update the superblock on each device while assembling the array. The
1204 argument given to this flag can be one of
1220 .BR layout\-original ,
1221 .BR layout\-alternate ,
1222 .BR layout\-unspecified ,
1229 option will adjust the superblock of an array what was created on a Sparc
1230 machine running a patched 2.2 Linux kernel. This kernel got the
1231 alignment of part of the superblock wrong. You can use the
1232 .B "\-\-examine \-\-sparc2.2"
1235 to see what effect this would have.
1239 option will update the
1240 .B "preferred minor"
1241 field on each superblock to match the minor number of the array being
1243 This can be useful if
1245 reports a different "Preferred Minor" to
1247 In some cases this update will be performed automatically
1248 by the kernel driver. In particular the update happens automatically
1249 at the first write to an array with redundancy (RAID level 1 or
1250 greater) on a 2.6 (or later) kernel.
1254 option will change the uuid of the array. If a UUID is given with the
1256 option that UUID will be used as a new UUID and will
1258 be used to help identify the devices in the array.
1261 is given, a random UUID is chosen.
1265 option will change the
1267 of the array as stored in the superblock. This is only supported for
1268 version-1 superblocks.
1272 option will change the
1274 of the array as stored in the bitmap superblock. This option only
1275 works for a clustered environment.
1279 option will change the
1281 as recorded in the superblock. For version-0 superblocks, this is the
1282 same as updating the UUID.
1283 For version-1 superblocks, this involves updating the name.
1287 option will change the cluster name as recorded in the superblock and
1288 bitmap. This option only works for clustered environment.
1292 option will cause the array to be marked
1294 meaning that any redundancy in the array (e.g. parity for RAID5,
1295 copies for RAID1) may be incorrect. This will cause the RAID system
1296 to perform a "resync" pass to make sure that all redundant information
1301 option allows arrays to be moved between machines with different
1302 byte-order, such as from a big-endian machine like a Sparc or some
1303 MIPS machines, to a little-endian x86_64 machine.
1304 When assembling such an array for the first time after a move, giving
1305 .B "\-\-update=byteorder"
1308 to expect superblocks to have their byteorder reversed, and will
1309 correct that order before assembling the array. This is only valid
1310 with original (Version 0.90) superblocks.
1314 option will correct the summaries in the superblock. That is the
1315 counts of total, working, active, failed, and spare devices.
1319 option will rarely be of use. It applies to version 1.1 and 1.2 metadata
1320 only (where the metadata is at the start of the device) and is only
1321 useful when the component device has changed size (typically become
1322 larger). The version 1 metadata records the amount of the device that
1323 can be used to store data, so if a device in a version 1.1 or 1.2
1324 array becomes larger, the metadata will still be visible, but the
1325 extra space will not. In this case it might be useful to assemble the
1327 .BR \-\-update=devicesize .
1330 to determine the maximum usable amount of space on each device and
1331 update the relevant field in the metadata.
1335 option only works on v0.90 metadata arrays and will convert them to
1336 v1.0 metadata. The array must not be dirty (i.e. it must not need a
1337 sync) and it must not have a write-intent bitmap.
1339 The old metadata will remain on the devices, but will appear older
1340 than the new metadata and so will usually be ignored. The old metadata
1341 (or indeed the new metadata) can be removed by giving the appropriate
1344 .BR \-\-zero\-superblock .
1348 option can be used when an array has an internal bitmap which is
1349 corrupt in some way so that assembling the array normally fails. It
1350 will cause any internal bitmap to be ignored.
1354 option will reserve space in each device for a bad block list. This
1355 will be 4K in size and positioned near the end of any free space
1356 between the superblock and the data.
1360 option will cause any reservation of space for a bad block list to be
1361 removed. If the bad block list contains entries, this will fail, as
1362 removing the list could cause data corruption.
1366 option will enable PPL for a RAID5 array and reserve space for PPL on each
1367 device. There must be enough free space between the data and superblock and a
1368 write-intent bitmap or journal must not be used.
1372 option will disable PPL in the superblock.
1377 .B layout\-alternate
1378 options are for RAID0 arrays with non-uniform devices size that were in
1379 use before Linux 5.4. If the array was being used with Linux 3.13 or
1380 earlier, then to assemble the array on a new kernel,
1381 .B \-\-update=layout\-original
1382 must be given. If the array was created and used with a kernel from Linux 3.14 to
1384 .B \-\-update=layout\-alternate
1385 must be given. This only needs to be given once. Subsequent assembly of the array
1386 will happen normally.
1387 For more information, see
1391 .B layout\-unspecified
1392 option reverts the effect of
1395 .B layout\-alternate
1396 and allows the array to be again used on a kernel prior to Linux 5.3.
1397 This option should be used with great caution.
1400 .BR \-\-freeze\-reshape
1401 Option is intended to be used in start-up scripts during initrd boot phase.
1402 When array under reshape is assembled during initrd phase, this option
1403 stops reshape after reshape critical section is being restored. This happens
1404 before file system pivot operation and avoids loss of file system context.
1405 Losing file system context would cause reshape to be broken.
1407 Reshape can be continued later using the
1409 option for the grow command.
1413 See this option under Create and Build options.
1415 .SH For Manage mode:
1418 .BR \-t ", " \-\-test
1419 Unless a more serious error occurred,
1421 will exit with a status of 2 if no changes were made to the array and
1422 0 if at least one change was made.
1423 This can be useful when an indirect specifier such as
1428 is used in requesting an operation on the array.
1430 will report failure if these specifiers didn't find any match.
1433 .BR \-a ", " \-\-add
1434 hot-add listed devices.
1435 If a device appears to have recently been part of the array
1436 (possibly it failed or was removed) the device is re\-added as described
1438 If that fails or the device was never part of the array, the device is
1439 added as a hot-spare.
1440 If the array is degraded, it will immediately start to rebuild data
1443 Note that this and the following options are only meaningful on array
1444 with redundancy. They don't apply to RAID0 or Linear.
1448 re\-add a device that was previously removed from an array.
1449 If the metadata on the device reports that it is a member of the
1450 array, and the slot that it used is still vacant, then the device will
1451 be added back to the array in the same position. This will normally
1452 cause the data for that device to be recovered. However based on the
1453 event count on the device, the recovery may only require sections that
1454 are flagged a write-intent bitmap to be recovered or may not require
1455 any recovery at all.
1457 When used on an array that has no metadata (i.e. it was built with
1459 it will be assumed that bitmap-based recovery is enough to make the
1460 device fully consistent with the array.
1462 When used with v1.x metadata,
1464 can be accompanied by
1465 .BR \-\-update=devicesize ,
1466 .BR \-\-update=bbl ", or"
1467 .BR \-\-update=no\-bbl .
1468 See the description of these option when used in Assemble mode for an
1469 explanation of their use.
1471 If the device name given is
1475 will try to find any device that looks like it should be
1476 part of the array but isn't and will try to re\-add all such devices.
1478 If the device name given is
1482 will find all devices in the array that are marked
1484 remove them and attempt to immediately re\-add them. This can be
1485 useful if you are certain that the reason for failure has been
1490 Add a device as a spare. This is similar to
1492 except that it does not attempt
1494 first. The device will be added as a spare even if it looks like it
1495 could be an recent member of the array.
1498 .BR \-r ", " \-\-remove
1499 remove listed devices. They must not be active. i.e. they should
1500 be failed or spare devices.
1502 As well as the name of a device file
1512 The first causes all failed device to be removed. The second causes
1513 any device which is no longer connected to the system (i.e an 'open'
1517 The third will remove a set as describe below under
1521 .BR \-f ", " \-\-fail
1522 Mark listed devices as faulty.
1523 As well as the name of a device file, the word
1527 can be given. The former will cause any device that has been detached from
1528 the system to be marked as failed. It can then be removed.
1530 For RAID10 arrays where the number of copies evenly divides the number
1531 of devices, the devices can be conceptually divided into sets where
1532 each set contains a single complete copy of the data on the array.
1533 Sometimes a RAID10 array will be configured so that these sets are on
1534 separate controllers. In this case all the devices in one set can be
1535 failed by giving a name like
1541 The appropriate set names are reported by
1551 Mark listed devices as requiring replacement. As soon as a spare is
1552 available, it will be rebuilt and will replace the marked device.
1553 This is similar to marking a device as faulty, but the device remains
1554 in service during the recovery process to increase resilience against
1555 multiple failures. When the replacement process finishes, the
1556 replaced device will be marked as faulty.
1560 This can follow a list of
1562 devices. The devices listed after
1564 will be preferentially used to replace the devices listed after
1566 These device must already be spare devices in the array.
1569 .BR \-\-write\-mostly
1570 Subsequent devices that are added or re\-added will have the 'write-mostly'
1571 flag set. This is only valid for RAID1 and means that the 'md' driver
1572 will avoid reading from these devices if possible.
1575 Subsequent devices that are added or re\-added will have the 'write-mostly'
1578 .BR \-\-cluster\-confirm
1579 Confirm the existence of the device. This is issued in response to an \-\-add
1580 request by a node in a cluster. When a node adds a device it sends a message
1581 to all nodes in the cluster to look for a device with a UUID. This translates
1582 to a udev notification with the UUID of the device to be added and the slot
1583 number. The receiving node must acknowledge this message
1584 with \-\-cluster\-confirm. Valid arguments are <slot>:<devicename> in case
1585 the device is found or <slot>:missing in case the device is not found.
1589 Add journal to an existing array, or recreate journal for RAID-4/5/6 array
1590 that lost a journal device. To avoid interrupting on-going write opertions,
1592 only works for array in Read-Only state.
1596 Subsequent devices that are added or re\-added will have
1597 the 'failfast' flag set. This is only valid for RAID1 and RAID10 and
1598 means that the 'md' driver will avoid long timeouts on error handling
1602 Subsequent devices that are re\-added will be re\-added without
1603 the 'failfast' flag set.
1606 Each of these options requires that the first device listed is the array
1607 to be acted upon, and the remainder are component devices to be added,
1608 removed, marked as faulty, etc. Several different operations can be
1609 specified for different devices, e.g.
1611 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1613 Each operation applies to all devices listed until the next
1616 If an array is using a write-intent bitmap, then devices which have
1617 been removed can be re\-added in a way that avoids a full
1618 reconstruction but instead just updates the blocks that have changed
1619 since the device was removed. For arrays with persistent metadata
1620 (superblocks) this is done automatically. For arrays created with
1622 mdadm needs to be told that this device we removed recently with
1625 Devices can only be removed from an array if they are not in active
1626 use, i.e. that must be spares or failed devices. To remove an active
1627 device, it must first be marked as
1633 .BR \-Q ", " \-\-query
1634 Examine a device to see
1635 (1) if it is an md device and (2) if it is a component of an md
1637 Information about what is discovered is presented.
1640 .BR \-D ", " \-\-detail
1641 Print details of one or more md devices.
1644 .BR \-\-detail\-platform
1645 Print details of the platform's RAID capabilities (firmware / hardware
1646 topology) for a given metadata format. If used without argument, mdadm
1647 will scan all controllers looking for their capabilities. Otherwise, mdadm
1648 will only look at the controller specified by the argument in form of an
1649 absolute filepath or a link, e.g.
1650 .IR /sys/devices/pci0000:00/0000:00:1f.2 .
1653 .BR \-Y ", " \-\-export
1656 .BR \-\-detail-platform ,
1660 output will be formatted as
1662 pairs for easy import into the environment.
1668 indicates whether an array was started
1670 or not, which may include a reason
1671 .RB ( unsafe ", " nothing ", " no ).
1674 indicates if the array is expected on this host
1676 or seems to be from elsewhere
1680 .BR \-E ", " \-\-examine
1681 Print contents of the metadata stored on the named device(s).
1682 Note the contrast between
1687 applies to devices which are components of an array, while
1689 applies to a whole array which is currently active.
1692 If an array was created on a SPARC machine with a 2.2 Linux kernel
1693 patched with RAID support, the superblock will have been created
1694 incorrectly, or at least incompatibly with 2.4 and later kernels.
1699 will fix the superblock before displaying it. If this appears to do
1700 the right thing, then the array can be successfully assembled using
1701 .BR "\-\-assemble \-\-update=sparc2.2" .
1704 .BR \-X ", " \-\-examine\-bitmap
1705 Report information about a bitmap file.
1706 The argument is either an external bitmap file or an array component
1707 in case of an internal bitmap. Note that running this on an array
1710 does not report the bitmap for that array.
1713 .B \-\-examine\-badblocks
1714 List the bad-blocks recorded for the device, if a bad-blocks list has
1715 been configured. Currently only
1719 metadata support bad-blocks lists.
1722 .BI \-\-dump= directory
1724 .BI \-\-restore= directory
1725 Save metadata from lists devices, or restore metadata to listed devices.
1728 .BR \-R ", " \-\-run
1729 start a partially assembled array. If
1731 did not find enough devices to fully start the array, it might leaving
1732 it partially assembled. If you wish, you can then use
1734 to start the array in degraded mode.
1737 .BR \-S ", " \-\-stop
1738 deactivate array, releasing all resources.
1741 .BR \-o ", " \-\-readonly
1742 mark array as readonly.
1745 .BR \-w ", " \-\-readwrite
1746 mark array as readwrite.
1749 .B \-\-zero\-superblock
1750 If the device contains a valid md superblock, the block is
1751 overwritten with zeros. With
1753 the block where the superblock would be is overwritten even if it
1754 doesn't appear to be valid.
1757 Be careful to call \-\-zero\-superblock with clustered raid, make sure
1758 array isn't used or assembled in other cluster node before execute it.
1761 .B \-\-kill\-subarray=
1762 If the device is a container and the argument to \-\-kill\-subarray
1763 specifies an inactive subarray in the container, then the subarray is
1764 deleted. Deleting all subarrays will leave an 'empty-container' or
1765 spare superblock on the drives. See
1766 .B \-\-zero\-superblock
1768 removing a superblock. Note that some formats depend on the subarray
1769 index for generating a UUID, this command will fail if it would change
1770 the UUID of an active subarray.
1773 .B \-\-update\-subarray=
1774 If the device is a container and the argument to \-\-update\-subarray
1775 specifies a subarray in the container, then attempt to update the given
1776 superblock field in the subarray. See below in
1781 .BR \-t ", " \-\-test
1786 is set to reflect the status of the device. See below in
1791 .BR \-W ", " \-\-wait
1792 For each md device given, wait for any resync, recovery, or reshape
1793 activity to finish before returning.
1795 will return with success if it actually waited for every device
1796 listed, otherwise it will return failure.
1800 For each md device given, or each device in /proc/mdstat if
1802 is given, arrange for the array to be marked clean as soon as possible.
1804 will return with success if the array uses external metadata and we
1805 successfully waited. For native arrays this returns immediately as the
1806 kernel handles dirty-clean transitions at shutdown. No action is taken
1807 if safe-mode handling is disabled.
1811 Set the "sync_action" for all md devices given to one of
1818 will abort any currently running action though some actions will
1819 automatically restart.
1822 will abort any current action and ensure no other action starts
1832 .BR "SCRUBBING AND MISMATCHES" .
1834 .SH For Incremental Assembly mode:
1836 .BR \-\-rebuild\-map ", " \-r
1837 Rebuild the map file
1841 uses to help track which arrays are currently being assembled.
1844 .BR \-\-run ", " \-R
1845 Run any array assembled as soon as a minimal number of devices are
1846 available, rather than waiting until all expected devices are present.
1849 .BR \-\-scan ", " \-s
1850 Only meaningful with
1854 file for arrays that are being incrementally assembled and will try to
1855 start any that are not already started. If any such array is listed
1858 as requiring an external bitmap, that bitmap will be attached first.
1861 .BR \-\-fail ", " \-f
1862 This allows the hot-plug system to remove devices that have fully disappeared
1863 from the kernel. It will first fail and then remove the device from any
1864 array it belongs to.
1865 The device name given should be a kernel device name such as "sda",
1871 Only used with \-\-fail. The 'path' given will be recorded so that if
1872 a new device appears at the same location it can be automatically
1873 added to the same array. This allows the failed device to be
1874 automatically replaced by a new device without metadata if it appears
1875 at specified path. This option is normally only set by a
1879 .SH For Monitor mode:
1881 .BR \-m ", " \-\-mail
1882 Give a mail address to send alerts to.
1885 .BR \-p ", " \-\-program ", " \-\-alert
1886 Give a program to be run whenever an event is detected.
1889 .BR \-y ", " \-\-syslog
1890 Cause all events to be reported through 'syslog'. The messages have
1891 facility of 'daemon' and varying priorities.
1894 .BR \-d ", " \-\-delay
1895 Give a delay in seconds.
1897 polls the md arrays and then waits this many seconds before polling
1898 again. The default is 60 seconds. Since 2.6.16, there is no need to
1899 reduce this as the kernel alerts
1901 immediately when there is any change.
1904 .BR \-r ", " \-\-increment
1905 Give a percentage increment.
1907 will generate RebuildNN events with the given percentage increment.
1910 .BR \-f ", " \-\-daemonise
1913 to run as a background daemon if it decides to monitor anything. This
1914 causes it to fork and run in the child, and to disconnect from the
1915 terminal. The process id of the child is written to stdout.
1918 which will only continue monitoring if a mail address or alert program
1919 is found in the config file.
1922 .BR \-i ", " \-\-pid\-file
1925 is running in daemon mode, write the pid of the daemon process to
1926 the specified file, instead of printing it on standard output.
1929 .BR \-1 ", " \-\-oneshot
1930 Check arrays only once. This will generate
1932 events and more significantly
1938 .B " mdadm \-\-monitor \-\-scan \-1"
1940 from a cron script will ensure regular notification of any degraded arrays.
1943 .BR \-t ", " \-\-test
1946 alert for every array found at startup. This alert gets mailed and
1947 passed to the alert program. This can be used for testing that alert
1948 message do get through successfully.
1952 This inhibits the functionality for moving spares between arrays.
1953 Only one monitoring process started with
1955 but without this flag is allowed, otherwise the two could interfere
1962 .B mdadm \-\-assemble
1963 .I md-device options-and-component-devices...
1966 .B mdadm \-\-assemble \-\-scan
1967 .I md-devices-and-options...
1970 .B mdadm \-\-assemble \-\-scan
1974 This usage assembles one or more RAID arrays from pre-existing components.
1975 For each array, mdadm needs to know the md device, the identity of the
1976 array, and a number of component-devices. These can be found in a number of ways.
1978 In the first usage example (without the
1980 the first device given is the md device.
1981 In the second usage example, all devices listed are treated as md
1982 devices and assembly is attempted.
1983 In the third (where no devices are listed) all md devices that are
1984 listed in the configuration file are assembled. If no arrays are
1985 described by the configuration file, then any arrays that
1986 can be found on unused devices will be assembled.
1988 If precisely one device is listed, but
1994 was given and identity information is extracted from the configuration file.
1996 The identity can be given with the
2002 option, will be taken from the md-device record in the config file, or
2003 will be taken from the super block of the first component-device
2004 listed on the command line.
2006 Devices can be given on the
2008 command line or in the config file. Only devices which have an md
2009 superblock which contains the right identity will be considered for
2012 The config file is only used if explicitly named with
2014 or requested with (a possibly implicit)
2024 is not given, then the config file will only be used to find the
2025 identity of md arrays.
2027 Normally the array will be started after it is assembled. However if
2029 is not given and not all expected drives were listed, then the array
2030 is not started (to guard against usage errors). To insist that the
2031 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
2040 does not create any entries in
2044 It does record information in
2048 to choose the correct name.
2052 detects that udev is not configured, it will create the devices in
2056 In Linux kernels prior to version 2.6.28 there were two distinctly
2057 different types of md devices that could be created: one that could be
2058 partitioned using standard partitioning tools and one that could not.
2059 Since 2.6.28 that distinction is no longer relevant as both type of
2060 devices can be partitioned.
2062 will normally create the type that originally could not be partitioned
2063 as it has a well defined major number (9).
2065 Prior to 2.6.28, it is important that mdadm chooses the correct type
2066 of array device to use. This can be controlled with the
2068 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
2069 to use a partitionable device rather than the default.
2071 In the no-udev case, the value given to
2073 can be suffixed by a number. This tells
2075 to create that number of partition devices rather than the default of 4.
2079 can also be given in the configuration file as a word starting
2081 on the ARRAY line for the relevant array.
2088 and no devices are listed,
2090 will first attempt to assemble all the arrays listed in the config
2093 If no arrays are listed in the config (other than those marked
2095 it will look through the available devices for possible arrays and
2096 will try to assemble anything that it finds. Arrays which are tagged
2097 as belonging to the given homehost will be assembled and started
2098 normally. Arrays which do not obviously belong to this host are given
2099 names that are expected not to conflict with anything local, and are
2100 started "read-auto" so that nothing is written to any device until the
2101 array is written to. i.e. automatic resync etc is delayed.
2105 finds a consistent set of devices that look like they should comprise
2106 an array, and if the superblock is tagged as belonging to the given
2107 home host, it will automatically choose a device name and try to
2108 assemble the array. If the array uses version-0.90 metadata, then the
2110 number as recorded in the superblock is used to create a name in
2114 If the array uses version-1 metadata, then the
2116 from the superblock is used to similarly create a name in
2118 (the name will have any 'host' prefix stripped first).
2120 This behaviour can be modified by the
2124 configuration file. This line can indicate that specific metadata
2125 type should, or should not, be automatically assembled. If an array
2126 is found which is not listed in
2128 and has a metadata format that is denied by the
2130 line, then it will not be assembled.
2133 line can also request that all arrays identified as being for this
2134 homehost should be assembled regardless of their metadata type.
2137 for further details.
2139 Note: Auto assembly cannot be used for assembling and activating some
2140 arrays which are undergoing reshape. In particular as the
2142 cannot be given, any reshape which requires a backup-file to continue
2143 cannot be started by auto assembly. An array which is growing to more
2144 devices and has passed the critical section can be assembled using
2155 .BI \-\-raid\-devices= Z
2159 This usage is similar to
2161 The difference is that it creates an array without a superblock. With
2162 these arrays there is no difference between initially creating the array and
2163 subsequently assembling the array, except that hopefully there is useful
2164 data there in the second case.
2166 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
2167 one of their synonyms. All devices must be listed and the array will
2168 be started once complete. It will often be appropriate to use
2169 .B \-\-assume\-clean
2170 with levels raid1 or raid10.
2181 .BI \-\-raid\-devices= Z
2185 This usage will initialise a new md array, associate some devices with
2186 it, and activate the array.
2188 The named device will normally not exist when
2189 .I "mdadm \-\-create"
2190 is run, but will be created by
2192 once the array becomes active.
2194 The max length md-device name is limited to 32 characters.
2195 Different metadata types have more strict limitation
2196 (like IMSM where only 16 characters are allowed).
2197 For that reason, long name could be truncated or rejected, it depends on metadata policy.
2199 As devices are added, they are checked to see if they contain RAID
2200 superblocks or filesystems. They are also checked to see if the variance in
2201 device size exceeds 1%.
2203 If any discrepancy is found, the array will not automatically be run, though
2206 can override this caution.
2208 To create a "degraded" array in which some devices are missing, simply
2209 give the word "\fBmissing\fP"
2210 in place of a device name. This will cause
2212 to leave the corresponding slot in the array empty.
2213 For a RAID4 or RAID5 array at most one slot can be
2214 "\fBmissing\fP"; for a RAID6 array at most two slots.
2215 For a RAID1 array, only one real device needs to be given. All of the
2219 When creating a RAID5 array,
2221 will automatically create a degraded array with an extra spare drive.
2222 This is because building the spare into a degraded array is in general
2223 faster than resyncing the parity on a non-degraded, but not clean,
2224 array. This feature can be overridden with the
2228 When creating an array with version-1 metadata a name for the array is
2230 If this is not given with the
2234 will choose a name based on the last component of the name of the
2235 device being created. So if
2237 is being created, then the name
2242 is being created, then the name
2246 When creating a partition based array, using
2248 with version-1.x metadata, the partition type should be set to
2250 (non fs-data). This type selection allows for greater precision since
2251 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
2252 might create problems in the event of array recovery through a live cdrom.
2254 A new array will normally get a randomly assigned 128bit UUID which is
2255 very likely to be unique. If you have a specific need, you can choose
2256 a UUID for the array by giving the
2258 option. Be warned that creating two arrays with the same UUID is a
2259 recipe for disaster. Also, using
2261 when creating a v0.90 array will silently override any
2266 .\"option is given, it is not necessary to list any component-devices in this command.
2267 .\"They can be added later, before a
2271 .\"is given, the apparent size of the smallest drive given is used.
2273 If the array type supports a write-intent bitmap, and if the devices
2274 in the array exceed 100G is size, an internal write-intent bitmap
2275 will automatically be added unless some other option is explicitly
2278 option or a different consistency policy is selected with the
2279 .B \-\-consistency\-policy
2280 option. In any case space for a bitmap will be reserved so that one
2281 can be added later with
2282 .BR "\-\-grow \-\-bitmap=internal" .
2284 If the metadata type supports it (currently only 1.x and IMSM metadata),
2285 space will be allocated to store a bad block list. This allows a modest
2286 number of bad blocks to be recorded, allowing the drive to remain in
2287 service while only partially functional.
2289 When creating an array within a
2292 can be given either the list of devices to use, or simply the name of
2293 the container. The former case gives control over which devices in
2294 the container will be used for the array. The latter case allows
2296 to automatically choose which devices to use based on how much spare
2299 The General Management options that are valid with
2304 insist on running the array even if some devices look like they might
2309 start the array in readonly mode.
2316 .I options... devices...
2319 This usage will allow individual devices in an array to be failed,
2320 removed or added. It is possible to perform multiple operations with
2321 on command. For example:
2323 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
2329 and will then remove it from the array and finally add it back
2330 in as a spare. However only one md array can be affected by a single
2333 When a device is added to an active array, mdadm checks to see if it
2334 has metadata on it which suggests that it was recently a member of the
2335 array. If it does, it tries to "re\-add" the device. If there have
2336 been no changes since the device was removed, or if the array has a
2337 write-intent bitmap which has recorded whatever changes there were,
2338 then the device will immediately become a full member of the array and
2339 those differences recorded in the bitmap will be resolved.
2349 MISC mode includes a number of distinct operations that
2350 operate on distinct devices. The operations are:
2353 The device is examined to see if it is
2354 (1) an active md array, or
2355 (2) a component of an md array.
2356 The information discovered is reported.
2360 The device should be an active md device.
2362 will display a detailed description of the array.
2366 will cause the output to be less detailed and the format to be
2367 suitable for inclusion in
2371 will normally be 0 unless
2373 failed to get useful information about the device(s); however, if the
2375 option is given, then the exit status will be:
2379 The array is functioning normally.
2382 The array has at least one failed device.
2385 The array has multiple failed devices such that it is unusable.
2388 There was an error while trying to get information about the device.
2392 .B \-\-detail\-platform
2393 Print detail of the platform's RAID capabilities (firmware / hardware
2394 topology). If the metadata is specified with
2398 then the return status will be:
2402 metadata successfully enumerated its platform components on this system
2405 metadata is platform independent
2408 metadata failed to find its platform components on this system
2412 .B \-\-update\-subarray=
2413 If the device is a container and the argument to \-\-update\-subarray
2414 specifies a subarray in the container, then attempt to update the given
2415 superblock field in the subarray. Similar to updating an array in
2416 "assemble" mode, the field to update is selected by
2420 option. The supported options are
2430 option updates the subarray name in the metadata, it may not affect the
2431 device node name or the device node symlink until the subarray is
2432 re\-assembled. If updating
2434 would change the UUID of an active subarray this operation is blocked,
2435 and the command will end in an error.
2441 options enable and disable PPL in the metadata. Currently supported only for
2448 options enable and disable write-intent bitmap in the metadata. Currently supported only for
2453 The device should be a component of an md array.
2455 will read the md superblock of the device and display the contents.
2460 is given, then multiple devices that are components of the one array
2461 are grouped together and reported in a single entry suitable
2467 without listing any devices will cause all devices listed in the
2468 config file to be examined.
2471 .BI \-\-dump= directory
2472 If the device contains RAID metadata, a file will be created in the
2474 and the metadata will be written to it. The file will be the same
2475 size as the device and have the metadata written in the file at the
2476 same locate that it exists in the device. However the file will be "sparse" so
2477 that only those blocks containing metadata will be allocated. The
2478 total space used will be small.
2480 The file name used in the
2482 will be the base name of the device. Further if any links appear in
2484 which point to the device, then hard links to the file will be created
2491 Multiple devices can be listed and their metadata will all be stored
2492 in the one directory.
2495 .BI \-\-restore= directory
2496 This is the reverse of
2499 will locate a file in the directory that has a name appropriate for
2500 the given device and will restore metadata from it. Names that match
2502 names are preferred, however if two of those refer to different files,
2504 will not choose between them but will abort the operation.
2506 If a file name is given instead of a
2510 will restore from that file to a single device, always provided the
2511 size of the file matches that of the device, and the file contains
2515 The devices should be active md arrays which will be deactivated, as
2516 long as they are not currently in use.
2520 This will fully activate a partially assembled md array.
2524 This will mark an active array as read-only, providing that it is
2525 not currently being used.
2531 array back to being read/write.
2535 For all operations except
2538 will cause the operation to be applied to all arrays listed in
2543 causes all devices listed in the config file to be examined.
2546 .BR \-b ", " \-\-brief
2547 Be less verbose. This is used with
2555 gives an intermediate level of verbosity.
2561 .B mdadm \-\-monitor
2562 .I options... devices...
2567 to periodically poll a number of md arrays and to report on any events
2570 will never exit once it decides that there are arrays to be checked,
2571 so it should normally be run in the background.
2573 As well as reporting events,
2575 may move a spare drive from one array to another if they are in the
2580 and if the destination array has a failed drive but no spares.
2582 If any devices are listed on the command line,
2584 will only monitor those devices. Otherwise all arrays listed in the
2585 configuration file will be monitored. Further, if
2587 is given, then any other md devices that appear in
2589 will also be monitored.
2591 The result of monitoring the arrays is the generation of events.
2592 These events are passed to a separate program (if specified) and may
2593 be mailed to a given E-mail address.
2595 When passing events to a program, the program is run once for each event,
2596 and is given 2 or 3 command-line arguments: the first is the
2597 name of the event (see below), the second is the name of the
2598 md device which is affected, and the third is the name of a related
2599 device if relevant (such as a component device that has failed).
2603 is given, then a program or an E-mail address must be specified on the
2604 command line or in the config file. If neither are available, then
2606 will not monitor anything.
2610 will continue monitoring as long as something was found to monitor. If
2611 no program or email is given, then each event is reported to
2614 The different events are:
2618 .B DeviceDisappeared
2619 An md array which previously was configured appears to no longer be
2620 configured. (syslog priority: Critical)
2624 was told to monitor an array which is RAID0 or Linear, then it will
2626 .B DeviceDisappeared
2627 with the extra information
2629 This is because RAID0 and Linear do not support the device-failed,
2630 hot-spare and resync operations which are monitored.
2634 An md array started reconstruction (e.g. recovery, resync, reshape,
2635 check, repair). (syslog priority: Warning)
2641 is a two-digit number (ie. 05, 48). This indicates that rebuild
2642 has passed that many percent of the total. The events are generated
2643 with fixed increment since 0. Increment size may be specified with
2644 a commandline option (default is 20). (syslog priority: Warning)
2648 An md array that was rebuilding, isn't any more, either because it
2649 finished normally or was aborted. (syslog priority: Warning)
2653 An active component device of an array has been marked as
2654 faulty. (syslog priority: Critical)
2658 A spare component device which was being rebuilt to replace a faulty
2659 device has failed. (syslog priority: Critical)
2663 A spare component device which was being rebuilt to replace a faulty
2664 device has been successfully rebuilt and has been made active.
2665 (syslog priority: Info)
2669 A new md array has been detected in the
2671 file. (syslog priority: Info)
2675 A newly noticed array appears to be degraded. This message is not
2678 notices a drive failure which causes degradation, but only when
2680 notices that an array is degraded when it first sees the array.
2681 (syslog priority: Critical)
2685 A spare drive has been moved from one array in a
2689 to another to allow a failed drive to be replaced.
2690 (syslog priority: Info)
2696 has been told, via the config file, that an array should have a certain
2697 number of spare devices, and
2699 detects that it has fewer than this number when it first sees the
2700 array, it will report a
2703 (syslog priority: Warning)
2707 An array was found at startup, and the
2710 (syslog priority: Info)
2720 cause Email to be sent. All events cause the program to be run.
2721 The program is run with two or three arguments: the event
2722 name, the array device and possibly a second device.
2724 Each event has an associated array device (e.g.
2726 and possibly a second device. For
2731 the second device is the relevant component device.
2734 the second device is the array that the spare was moved from.
2738 to move spares from one array to another, the different arrays need to
2739 be labeled with the same
2741 or the spares must be allowed to migrate through matching POLICY domains
2742 in the configuration file. The
2744 name can be any string; it is only necessary that different spare
2745 groups use different names.
2749 detects that an array in a spare group has fewer active
2750 devices than necessary for the complete array, and has no spare
2751 devices, it will look for another array in the same spare group that
2752 has a full complement of working drive and a spare. It will then
2753 attempt to remove the spare from the second drive and add it to the
2755 If the removal succeeds but the adding fails, then it is added back to
2758 If the spare group for a degraded array is not defined,
2760 will look at the rules of spare migration specified by POLICY lines in
2762 and then follow similar steps as above if a matching spare is found.
2765 The GROW mode is used for changing the size or shape of an active
2767 For this to work, the kernel must support the necessary change.
2768 Various types of growth are being added during 2.6 development.
2770 Currently the supported changes include
2772 change the "size" attribute for RAID1, RAID4, RAID5 and RAID6.
2774 increase or decrease the "raid\-devices" attribute of RAID0, RAID1, RAID4,
2777 change the chunk-size and layout of RAID0, RAID4, RAID5, RAID6 and RAID10.
2779 convert between RAID1 and RAID5, between RAID5 and RAID6, between
2780 RAID0, RAID4, and RAID5, and between RAID0 and RAID10 (in the near-2 mode).
2782 add a write-intent bitmap to any array which supports these bitmaps, or
2783 remove a write-intent bitmap from such an array.
2785 change the array's consistency policy.
2788 Using GROW on containers is currently supported only for Intel's IMSM
2789 container format. The number of devices in a container can be
2790 increased - which affects all arrays in the container - or an array
2791 in a container can be converted between levels where those levels are
2792 supported by the container, and the conversion is on of those listed
2798 Intel's native checkpointing doesn't use
2800 option and it is transparent for assembly feature.
2802 Roaming between Windows(R) and Linux systems for IMSM metadata is not
2803 supported during grow process.
2805 When growing a raid0 device, the new component disk size (or external
2806 backup size) should be larger than LCM(old, new) * chunk-size * 2,
2807 where LCM() is the least common multiple of the old and new count of
2808 component disks, and "* 2" comes from the fact that mdadm refuses to
2809 use more than half of a spare device for backup space.
2812 Normally when an array is built the "size" is taken from the smallest
2813 of the drives. If all the small drives in an arrays are, one at a
2814 time, removed and replaced with larger drives, then you could have an
2815 array of large drives with only a small amount used. In this
2816 situation, changing the "size" with "GROW" mode will allow the extra
2817 space to start being used. If the size is increased in this way, a
2818 "resync" process will start to make sure the new parts of the array
2821 Note that when an array changes size, any filesystem that may be
2822 stored in the array will not automatically grow or shrink to use or
2823 vacate the space. The
2824 filesystem will need to be explicitly told to use the extra space
2825 after growing, or to reduce its size
2827 to shrinking the array.
2829 Also the size of an array cannot be changed while it has an active
2830 bitmap. If an array has a bitmap, it must be removed before the size
2831 can be changed. Once the change is complete a new bitmap can be created.
2836 is not yet supported for external file bitmap.
2838 .SS RAID\-DEVICES CHANGES
2840 A RAID1 array can work with any number of devices from 1 upwards
2841 (though 1 is not very useful). There may be times which you want to
2842 increase or decrease the number of active devices. Note that this is
2843 different to hot-add or hot-remove which changes the number of
2846 When reducing the number of devices in a RAID1 array, the slots which
2847 are to be removed from the array must already be vacant. That is, the
2848 devices which were in those slots must be failed and removed.
2850 When the number of devices is increased, any hot spares that are
2851 present will be activated immediately.
2853 Changing the number of active devices in a RAID5 or RAID6 is much more
2854 effort. Every block in the array will need to be read and written
2855 back to a new location. From 2.6.17, the Linux Kernel is able to
2856 increase the number of devices in a RAID5 safely, including restarting
2857 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2858 increase or decrease the number of devices in a RAID5 or RAID6.
2860 From 2.6.35, the Linux Kernel is able to convert a RAID0 in to a RAID4
2863 uses this functionality and the ability to add
2864 devices to a RAID4 to allow devices to be added to a RAID0. When
2865 requested to do this,
2867 will convert the RAID0 to a RAID4, add the necessary disks and make
2868 the reshape happen, and then convert the RAID4 back to RAID0.
2870 When decreasing the number of devices, the size of the array will also
2871 decrease. If there was data in the array, it could get destroyed and
2872 this is not reversible, so you should firstly shrink the filesystem on
2873 the array to fit within the new size. To help prevent accidents,
2875 requires that the size of the array be decreased first with
2876 .BR "mdadm --grow --array-size" .
2877 This is a reversible change which simply makes the end of the array
2878 inaccessible. The integrity of any data can then be checked before
2879 the non-reversible reduction in the number of devices is request.
2881 When relocating the first few stripes on a RAID5 or RAID6, it is not
2882 possible to keep the data on disk completely consistent and
2883 crash-proof. To provide the required safety, mdadm disables writes to
2884 the array while this "critical section" is reshaped, and takes a
2885 backup of the data that is in that section. For grows, this backup may be
2886 stored in any spare devices that the array has, however it can also be
2887 stored in a separate file specified with the
2889 option, and is required to be specified for shrinks, RAID level
2890 changes and layout changes. If this option is used, and the system
2891 does crash during the critical period, the same file must be passed to
2893 to restore the backup and reassemble the array. When shrinking rather
2894 than growing the array, the reshape is done from the end towards the
2895 beginning, so the "critical section" is at the end of the reshape.
2899 Changing the RAID level of any array happens instantaneously. However
2900 in the RAID5 to RAID6 case this requires a non-standard layout of the
2901 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2902 required before the change can be accomplished. So while the level
2903 change is instant, the accompanying layout change can take quite a
2906 is required. If the array is not simultaneously being grown or
2907 shrunk, so that the array size will remain the same - for example,
2908 reshaping a 3-drive RAID5 into a 4-drive RAID6 - the backup file will
2909 be used not just for a "cricital section" but throughout the reshape
2910 operation, as described below under LAYOUT CHANGES.
2912 .SS CHUNK-SIZE AND LAYOUT CHANGES
2914 Changing the chunk-size or layout without also changing the number of
2915 devices as the same time will involve re-writing all blocks in-place.
2916 To ensure against data loss in the case of a crash, a
2918 must be provided for these changes. Small sections of the array will
2919 be copied to the backup file while they are being rearranged. This
2920 means that all the data is copied twice, once to the backup and once
2921 to the new layout on the array, so this type of reshape will go very
2924 If the reshape is interrupted for any reason, this backup file must be
2926 .B "mdadm --assemble"
2927 so the array can be reassembled. Consequently the file cannot be
2928 stored on the device being reshaped.
2933 A write-intent bitmap can be added to, or removed from, an active
2934 array. Either internal bitmaps, or bitmaps stored in a separate file,
2935 can be added. Note that if you add a bitmap stored in a file which is
2936 in a filesystem that is on the RAID array being affected, the system
2937 will deadlock. The bitmap must be on a separate filesystem.
2939 .SS CONSISTENCY POLICY CHANGES
2941 The consistency policy of an active array can be changed by using the
2942 .B \-\-consistency\-policy
2943 option in Grow mode. Currently this works only for the
2947 policies and allows to enable or disable the RAID5 Partial Parity Log (PPL).
2949 .SH INCREMENTAL MODE
2953 .B mdadm \-\-incremental
2957 .RI [ optional-aliases-for-device ]
2960 .B mdadm \-\-incremental \-\-fail
2964 .B mdadm \-\-incremental \-\-rebuild\-map
2967 .B mdadm \-\-incremental \-\-run \-\-scan
2970 This mode is designed to be used in conjunction with a device
2971 discovery system. As devices are found in a system, they can be
2973 .B "mdadm \-\-incremental"
2974 to be conditionally added to an appropriate array.
2976 Conversely, it can also be used with the
2978 flag to do just the opposite and find whatever array a particular device
2979 is part of and remove the device from that array.
2981 If the device passed is a
2983 device created by a previous call to
2985 then rather than trying to add that device to an array, all the arrays
2986 described by the metadata of the container will be started.
2989 performs a number of tests to determine if the device is part of an
2990 array, and which array it should be part of. If an appropriate array
2991 is found, or can be created,
2993 adds the device to the array and conditionally starts the array.
2997 will normally only add devices to an array which were previously working
2998 (active or spare) parts of that array. The support for automatic
2999 inclusion of a new drive as a spare in some array requires
3000 a configuration through POLICY in config file.
3004 makes are as follow:
3006 Is the device permitted by
3008 That is, is it listed in a
3010 line in that file. If
3012 is absent then the default it to allow any device. Similarly if
3014 contains the special word
3016 then any device is allowed. Otherwise the device name given to
3018 or one of the aliases given, or an alias found in the filesystem,
3019 must match one of the names or patterns in a
3023 This is the only context where the aliases are used. They are
3024 usually provided by a
3027 .BR $env{DEVLINKS} .
3030 Does the device have a valid md superblock? If a specific metadata
3031 version is requested with
3035 then only that style of metadata is accepted, otherwise
3037 finds any known version of metadata. If no
3039 metadata is found, the device may be still added to an array
3040 as a spare if POLICY allows.
3044 Does the metadata match an expected array?
3045 The metadata can match in two ways. Either there is an array listed
3048 which identifies the array (either by UUID, by name, by device list,
3049 or by minor-number), or the array was created with a
3055 or on the command line.
3058 is not able to positively identify the array as belonging to the
3059 current host, the device will be rejected.
3064 keeps a list of arrays that it has partially assembled in
3066 If no array exists which matches
3067 the metadata on the new device,
3069 must choose a device name and unit number. It does this based on any
3072 or any name information stored in the metadata. If this name
3073 suggests a unit number, that number will be used, otherwise a free
3074 unit number will be chosen. Normally
3076 will prefer to create a partitionable array, however if the
3080 suggests that a non-partitionable array is preferred, that will be
3083 If the array is not found in the config file and its metadata does not
3084 identify it as belonging to the "homehost", then
3086 will choose a name for the array which is certain not to conflict with
3087 any array which does belong to this host. It does this be adding an
3088 underscore and a small number to the name preferred by the metadata.
3090 Once an appropriate array is found or created and the device is added,
3092 must decide if the array is ready to be started. It will
3093 normally compare the number of available (non-spare) devices to the
3094 number of devices that the metadata suggests need to be active. If
3095 there are at least that many, the array will be started. This means
3096 that if any devices are missing the array will not be restarted.
3102 in which case the array will be run as soon as there are enough
3103 devices present for the data to be accessible. For a RAID1, that
3104 means one device will start the array. For a clean RAID5, the array
3105 will be started as soon as all but one drive is present.
3107 Note that neither of these approaches is really ideal. If it can
3108 be known that all device discovery has completed, then
3112 can be run which will try to start all arrays that are being
3113 incrementally assembled. They are started in "read-auto" mode in
3114 which they are read-only until the first write request. This means
3115 that no metadata updates are made and no attempt at resync or recovery
3116 happens. Further devices that are found before the first write can
3117 still be added safely.
3120 This section describes environment variables that affect how mdadm
3125 Setting this value to 1 will prevent mdadm from automatically launching
3126 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
3132 does not create any device nodes in /dev, but leaves that task to
3136 appears not to be configured, or if this environment variable is set
3139 will create and devices that are needed.
3142 .B MDADM_NO_SYSTEMCTL
3147 is in use it will normally request
3149 to start various background tasks (particularly
3151 rather than forking and running them in the background. This can be
3152 suppressed by setting
3153 .BR MDADM_NO_SYSTEMCTL=1 .
3157 A key value of IMSM metadata is that it allows interoperability with
3158 boot ROMs on Intel platforms, and with other major operating systems.
3161 will only allow an IMSM array to be created or modified if detects
3162 that it is running on an Intel platform which supports IMSM, and
3163 supports the particular configuration of IMSM that is being requested
3164 (some functionality requires newer OROM support).
3166 These checks can be suppressed by setting IMSM_NO_PLATFORM=1 in the
3167 environment. This can be useful for testing or for disaster
3168 recovery. You should be aware that interoperability may be
3169 compromised by setting this value.
3172 .B MDADM_GROW_ALLOW_OLD
3173 If an array is stopped while it is performing a reshape and that
3174 reshape was making use of a backup file, then when the array is
3177 will sometimes complain that the backup file is too old. If this
3178 happens and you are certain it is the right backup file, you can
3179 over-ride this check by setting
3180 .B MDADM_GROW_ALLOW_OLD=1
3185 Any string given in this variable is added to the start of the
3187 line in the config file, or treated as the whole
3189 line if none is given. It can be used to disable certain metadata
3192 is called from a boot script. For example
3194 .B " export MDADM_CONF_AUTO='-ddf -imsm'
3198 does not automatically assemble any DDF or
3199 IMSM arrays that are found. This can be useful on systems configured
3200 to manage such arrays with
3206 .B " mdadm \-\-query /dev/name-of-device"
3208 This will find out if a given device is a RAID array, or is part of
3209 one, and will provide brief information about the device.
3211 .B " mdadm \-\-assemble \-\-scan"
3213 This will assemble and start all arrays listed in the standard config
3214 file. This command will typically go in a system startup file.
3216 .B " mdadm \-\-stop \-\-scan"
3218 This will shut down all arrays that can be shut down (i.e. are not
3219 currently in use). This will typically go in a system shutdown script.
3221 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
3223 If (and only if) there is an Email address or program given in the
3224 standard config file, then
3225 monitor the status of all arrays listed in that file by
3226 polling them ever 2 minutes.
3228 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
3230 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
3233 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
3235 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
3237 This will create a prototype config file that describes currently
3238 active arrays that are known to be made from partitions of IDE or SCSI drives.
3239 This file should be reviewed before being used as it may
3240 contain unwanted detail.
3242 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
3244 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
3246 This will find arrays which could be assembled from existing IDE and
3247 SCSI whole drives (not partitions), and store the information in the
3248 format of a config file.
3249 This file is very likely to contain unwanted detail, particularly
3252 entries. It should be reviewed and edited before being used as an
3255 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
3257 .B " mdadm \-Ebsc partitions"
3259 Create a list of devices by reading
3260 .BR /proc/partitions ,
3261 scan these for RAID superblocks, and printout a brief listing of all
3264 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
3266 Scan all partitions and devices listed in
3267 .BR /proc/partitions
3270 out of all such devices with a RAID superblock with a minor number of 0.
3272 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /run/mdadm/mon.pid"
3274 If config file contains a mail address or alert program, run mdadm in
3275 the background in monitor mode monitoring all md devices. Also write
3276 pid of mdadm daemon to
3277 .BR /run/mdadm/mon.pid .
3279 .B " mdadm \-Iq /dev/somedevice"
3281 Try to incorporate newly discovered device into some array as
3284 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
3286 Rebuild the array map from any current arrays, and then start any that
3289 .B " mdadm /dev/md4 --fail detached --remove detached"
3291 Any devices which are components of /dev/md4 will be marked as faulty
3292 and then remove from the array.
3294 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4"
3298 which is currently a RAID5 array will be converted to RAID6. There
3299 should normally already be a spare drive attached to the array as a
3300 RAID6 needs one more drive than a matching RAID5.
3302 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
3304 Create a DDF array over 6 devices.
3306 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
3308 Create a RAID5 array over any 3 devices in the given DDF set. Use
3309 only 30 gigabytes of each device.
3311 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
3313 Assemble a pre-exist ddf array.
3315 .B " mdadm -I /dev/md/ddf1"
3317 Assemble all arrays contained in the ddf array, assigning names as
3320 .B " mdadm \-\-create \-\-help"
3322 Provide help about the Create mode.
3324 .B " mdadm \-\-config \-\-help"
3326 Provide help about the format of the config file.
3328 .B " mdadm \-\-help"
3330 Provide general help.
3340 lists all active md devices with information about them.
3342 uses this to find arrays when
3344 is given in Misc mode, and to monitor array reconstruction
3347 .SS {CONFFILE} (or {CONFFILE2})
3349 The config file lists which devices may be scanned to see if
3350 they contain MD super block, and gives identifying information
3351 (e.g. UUID) about known MD arrays. See
3355 .SS {CONFFILE}.d (or {CONFFILE2}.d)
3357 A directory containing configuration files which are read in lexical
3363 mode is used, this file gets a list of arrays currently being created.
3368 understand two sorts of names for array devices.
3370 The first is the so-called 'standard' format name, which matches the
3371 names used by the kernel and which appear in
3374 The second sort can be freely chosen, but must reside in
3376 When giving a device name to
3378 to create or assemble an array, either full path name such as
3382 can be given, or just the suffix of the second sort of name, such as
3388 chooses device names during auto-assembly or incremental assembly, it
3389 will sometimes add a small sequence number to the end of the name to
3390 avoid conflicted between multiple arrays that have the same name. If
3392 can reasonably determine that the array really is meant for this host,
3393 either by a hostname in the metadata, or by the presence of the array
3396 then it will leave off the suffix if possible.
3397 Also if the homehost is specified as
3400 will only use a suffix if a different array of the same name already
3401 exists or is listed in the config file.
3403 The standard names for non-partitioned arrays (the only sort of md
3404 array available in 2.4 and earlier) are of the form
3408 where NN is a number.
3409 The standard names for partitionable arrays (as available from 2.6
3410 onwards) are of the form:
3414 Partition numbers should be indicated by adding "pMM" to these, thus "/dev/md/d1p2".
3416 From kernel version 2.6.28 the "non-partitioned array" can actually
3417 be partitioned. So the "md_d\fBNN\fP"
3418 names are no longer needed, and
3419 partitions such as "/dev/md\fBNN\fPp\fBXX\fP"
3422 From kernel version 2.6.29 standard names can be non-numeric following
3429 is any string. These names are supported by
3431 since version 3.3 provided they are enabled in
3436 was previously known as
3440 For further information on mdadm usage, MD and the various levels of
3443 .B https://raid.wiki.kernel.org/
3445 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
3447 The latest version of
3449 should always be available from
3451 .B https://www.kernel.org/pub/linux/utils/raid/mdadm/
projects / thirdparty / mdadm.git / blob