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, as well as adding or
129 removing a write-intent bitmap.
132 .B "Incremental Assembly"
133 Add a single device to an appropriate array. If the addition of the
134 device makes the array runnable, the array will be started.
135 This provides a convenient interface to a
137 system. As each device is detected,
139 has a chance to include it in some array as appropriate.
142 flag is passed in we will remove the device from any active array
143 instead of adding it.
149 in this mode, then any arrays within that container will be assembled
154 This is for doing things to specific components of an array such as
155 adding new spares and removing faulty devices.
159 This is an 'everything else' mode that supports operations on active
160 arrays, operations on component devices such as erasing old superblocks, and
161 information gathering operations.
162 .\"This mode allows operations on independent devices such as examine MD
163 .\"superblocks, erasing old superblocks and stopping active arrays.
167 This mode does not act on a specific device or array, but rather it
168 requests the Linux Kernel to activate any auto-detected arrays.
171 .SH Options for selecting a mode are:
174 .BR \-A ", " \-\-assemble
175 Assemble a pre-existing array.
178 .BR \-B ", " \-\-build
179 Build a legacy array without superblocks.
182 .BR \-C ", " \-\-create
186 .BR \-F ", " \-\-follow ", " \-\-monitor
192 .BR \-G ", " \-\-grow
193 Change the size or shape of an active array.
196 .BR \-I ", " \-\-incremental
197 Add/remove a single device to/from an appropriate array, and possibly start the array.
201 Request that the kernel starts any auto-detected arrays. This can only
204 is compiled into the kernel \(em not if it is a module.
205 Arrays can be auto-detected by the kernel if all the components are in
206 primary MS-DOS partitions with partition type
208 and all use v0.90 metadata.
209 In-kernel autodetect is not recommended for new installations. Using
211 to detect and assemble arrays \(em possibly in an
213 \(em is substantially more flexible and should be preferred.
216 If a device is given before any options, or if the first option is
221 then the MANAGE mode is assumed.
222 Anything other than these will cause the
226 .SH Options that are not mode-specific are:
229 .BR \-h ", " \-\-help
230 Display general help message or, after one of the above options, a
231 mode-specific help message.
235 Display more detailed help about command line parsing and some commonly
239 .BR \-V ", " \-\-version
240 Print version information for mdadm.
243 .BR \-v ", " \-\-verbose
244 Be more verbose about what is happening. This can be used twice to be
246 The extra verbosity currently only affects
247 .B \-\-detail \-\-scan
249 .BR "\-\-examine \-\-scan" .
252 .BR \-q ", " \-\-quiet
253 Avoid printing purely informative messages. With this,
255 will be silent unless there is something really important to report.
258 .BR \-f ", " \-\-force
259 Be more forceful about certain operations. See the various modes for
260 the exact meaning of this option in different contexts.
263 .BR \-c ", " \-\-config=
264 Specify the config file. Default is to use
265 .BR /etc/mdadm.conf ,
266 or if that is missing then
267 .BR /etc/mdadm/mdadm.conf .
268 If the config file given is
270 then nothing will be read, but
272 will act as though the config file contained exactly
273 .B "DEVICE partitions containers"
276 to find a list of devices to scan, and
278 to find a list of containers to examine.
281 is given for the config file, then
283 will act as though the config file were empty.
286 .BR \-s ", " \-\-scan
289 for missing information.
290 In general, this option gives
292 permission to get any missing information (like component devices,
293 array devices, array identities, and alert destination) from the
294 configuration file (see previous option);
295 one exception is MISC mode when using
301 says to get a list of array devices from
305 .BR \-e ", " \-\-metadata=
306 Declare the style of RAID metadata (superblock) to be used. The
307 default is {DEFAULT_METADATA} for
309 and to guess for other operations.
310 The default can be overridden by setting the
319 .ie '{DEFAULT_METADATA}'0.90'
320 .IP "0, 0.90, default"
324 Use the original 0.90 format superblock. This format limits arrays to
325 28 component devices and limits component devices of levels 1 and
326 greater to 2 terabytes. It is also possible for there to be confusion
327 about whether the superblock applies to a whole device or just the
328 last partition, if that partition starts on a 64K boundary.
329 .ie '{DEFAULT_METADATA}'0.90'
330 .IP "1, 1.0, 1.1, 1.2"
332 .IP "1, 1.0, 1.1, 1.2 default"
334 Use the new version-1 format superblock. This has fewer restrictions.
335 It can easily be moved between hosts with different endian-ness, and a
336 recovery operation can be checkpointed and restarted. The different
337 sub-versions store the superblock at different locations on the
338 device, either at the end (for 1.0), at the start (for 1.1) or 4K from
339 the start (for 1.2). "1" is equivalent to "1.0".
340 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
342 Use the "Industry Standard" DDF (Disk Data Format) format defined by
344 When creating a DDF array a
346 will be created, and normal arrays can be created in that container.
348 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
350 which is managed in a similar manner to DDF, and is supported by an
351 option-rom on some platforms:
353 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
359 This will override any
361 setting in the config file and provides the identity of the host which
362 should be considered the home for any arrays.
364 When creating an array, the
366 will be recorded in the metadata. For version-1 superblocks, it will
367 be prefixed to the array name. For version-0.90 superblocks, part of
368 the SHA1 hash of the hostname will be stored in the later half of the
371 When reporting information about an array, any array which is tagged
372 for the given homehost will be reported as such.
374 When using Auto-Assemble, only arrays tagged for the given homehost
375 will be allowed to use 'local' names (i.e. not ending in '_' followed
376 by a digit string). See below under
377 .BR "Auto Assembly" .
379 .SH For create, build, or grow:
382 .BR \-n ", " \-\-raid\-devices=
383 Specify the number of active devices in the array. This, plus the
384 number of spare devices (see below) must equal the number of
386 (including "\fBmissing\fP" devices)
387 that are listed on the command line for
389 Setting a value of 1 is probably
390 a mistake and so requires that
392 be specified first. A value of 1 will then be allowed for linear,
393 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
395 This number can only be changed using
397 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
398 the necessary support.
401 .BR \-x ", " \-\-spare\-devices=
402 Specify the number of spare (eXtra) devices in the initial array.
403 Spares can also be added
404 and removed later. The number of component devices listed
405 on the command line must equal the number of RAID devices plus the
406 number of spare devices.
409 .BR \-z ", " \-\-size=
410 Amount (in Kibibytes) of space to use from each drive in RAID levels 1/4/5/6.
411 This must be a multiple of the chunk size, and must leave about 128Kb
412 of space at the end of the drive for the RAID superblock.
413 If this is not specified
414 (as it normally is not) the smallest drive (or partition) sets the
415 size, though if there is a variance among the drives of greater than 1%, a warning is
418 A suffix of 'M' or 'G' can be given to indicate Megabytes or
419 Gigabytes respectively.
421 Sometimes a replacement drive can be a little smaller than the
422 original drives though this should be minimised by IDEMA standards.
423 Such a replacement drive will be rejected by
425 To guard against this it can be useful to set the initial size
426 slightly smaller than the smaller device with the aim that it will
427 still be larger than any replacement.
429 This value can be set with
431 for RAID level 1/4/5/6 though
433 based arrays such as those with IMSM metadata may not be able to
435 If the array was created with a size smaller than the currently
436 active drives, the extra space can be accessed using
438 The size can be given as
440 which means to choose the largest size that fits on all current drives.
442 Before reducing the size of the array (with
443 .BR "\-\-grow \-\-size=" )
444 you should make sure that space isn't needed. If the device holds a
445 filesystem, you would need to resize the filesystem to use less space.
447 After reducing the array size you should check that the data stored in
448 the device is still available. If the device holds a filesystem, then
449 an 'fsck' of the filesystem is a minimum requirement. If there are
450 problems the array can be made bigger again with no loss with another
451 .B "\-\-grow \-\-size="
454 This value cannot be used when creating a
456 such as with DDF and IMSM metadata, though it perfectly valid when
457 creating an array inside a container.
460 .BR \-Z ", " \-\-array\-size=
461 This is only meaningful with
463 and its effect is not persistent: when the array is stopped and
464 restarted the default array size will be restored.
466 Setting the array-size causes the array to appear smaller to programs
467 that access the data. This is particularly needed before reshaping an
468 array so that it will be smaller. As the reshape is not reversible,
469 but setting the size with
471 is, it is required that the array size is reduced as appropriate
472 before the number of devices in the array is reduced.
474 Before reducing the size of the array you should make sure that space
475 isn't needed. If the device holds a filesystem, you would need to
476 resize the filesystem to use less space.
478 After reducing the array size you should check that the data stored in
479 the device is still available. If the device holds a filesystem, then
480 an 'fsck' of the filesystem is a minimum requirement. If there are
481 problems the array can be made bigger again with no loss with another
482 .B "\-\-grow \-\-array\-size="
485 A suffix of 'M' or 'G' can be given to indicate Megabytes or
486 Gigabytes respectively.
489 restores the apparent size of the array to be whatever the real
490 amount of available space is.
493 .BR \-c ", " \-\-chunk=
494 Specify chunk size of kibibytes. The default when creating an
495 array is 512KB. To ensure compatibility with earlier versions, the
496 default when Building and array with no persistent metadata is 64KB.
497 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
499 RAID4, RAID5, RAID6, and RAID10 require the chunk size to be a power
500 of 2. In any case it must be a multiple of 4KB.
502 A suffix of 'M' or 'G' can be given to indicate Megabytes or
503 Gigabytes respectively.
507 Specify rounding factor for a Linear array. The size of each
508 component will be rounded down to a multiple of this size.
509 This is a synonym for
511 but highlights the different meaning for Linear as compared to other
512 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
513 use, and is 0K (i.e. no rounding) in later kernels.
516 .BR \-l ", " \-\-level=
517 Set RAID level. When used with
519 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
520 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
521 Obviously some of these are synonymous.
525 metadata type is requested, only the
527 level is permitted, and it does not need to be explicitly given.
531 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
535 to change the RAID level in some cases. See LEVEL CHANGES below.
538 .BR \-p ", " \-\-layout=
539 This option configures the fine details of data layout for RAID5, RAID6,
540 and RAID10 arrays, and controls the failure modes for
543 The layout of the RAID5 parity block can be one of
544 .BR left\-asymmetric ,
545 .BR left\-symmetric ,
546 .BR right\-asymmetric ,
547 .BR right\-symmetric ,
548 .BR la ", " ra ", " ls ", " rs .
550 .BR left\-symmetric .
552 It is also possible to cause RAID5 to use a RAID4-like layout by
558 Finally for RAID5 there are DDF\-compatible layouts,
559 .BR ddf\-zero\-restart ,
560 .BR ddf\-N\-restart ,
562 .BR ddf\-N\-continue .
564 These same layouts are available for RAID6. There are also 4 layouts
565 that will provide an intermediate stage for converting between RAID5
566 and RAID6. These provide a layout which is identical to the
567 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
568 syndrome (the second 'parity' block used by RAID6) on the last device.
570 .BR left\-symmetric\-6 ,
571 .BR right\-symmetric\-6 ,
572 .BR left\-asymmetric\-6 ,
573 .BR right\-asymmetric\-6 ,
575 .BR parity\-first\-6 .
577 When setting the failure mode for level
580 .BR write\-transient ", " wt ,
581 .BR read\-transient ", " rt ,
582 .BR write\-persistent ", " wp ,
583 .BR read\-persistent ", " rp ,
585 .BR read\-fixable ", " rf ,
586 .BR clear ", " flush ", " none .
588 Each failure mode can be followed by a number, which is used as a period
589 between fault generation. Without a number, the fault is generated
590 once on the first relevant request. With a number, the fault will be
591 generated after that many requests, and will continue to be generated
592 every time the period elapses.
594 Multiple failure modes can be current simultaneously by using the
596 option to set subsequent failure modes.
598 "clear" or "none" will remove any pending or periodic failure modes,
599 and "flush" will clear any persistent faults.
601 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
602 by a small number. The default is 'n2'. The supported options are:
605 signals 'near' copies. Multiple copies of one data block are at
606 similar offsets in different devices.
609 signals 'offset' copies. Rather than the chunks being duplicated
610 within a stripe, whole stripes are duplicated but are rotated by one
611 device so duplicate blocks are on different devices. Thus subsequent
612 copies of a block are in the next drive, and are one chunk further
617 (multiple copies have very different offsets).
618 See md(4) for more detail about 'near', 'offset', and 'far'.
620 The number is the number of copies of each datablock. 2 is normal, 3
621 can be useful. This number can be at most equal to the number of
622 devices in the array. It does not need to divide evenly into that
623 number (e.g. it is perfectly legal to have an 'n2' layout for an array
624 with an odd number of devices).
626 When an array is converted between RAID5 and RAID6 an intermediate
627 RAID6 layout is used in which the second parity block (Q) is always on
628 the last device. To convert a RAID5 to RAID6 and leave it in this new
629 layout (which does not require re-striping) use
630 .BR \-\-layout=preserve .
631 This will try to avoid any restriping.
633 The converse of this is
634 .B \-\-layout=normalise
635 which will change a non-standard RAID6 layout into a more standard
642 (thus explaining the p of
646 .BR \-b ", " \-\-bitmap=
647 Specify a file to store a write-intent bitmap in. The file should not
650 is also given. The same file should be provided
651 when assembling the array. If the word
653 is given, then the bitmap is stored with the metadata on the array,
654 and so is replicated on all devices. If the word
658 mode, then any bitmap that is present is removed.
660 To help catch typing errors, the filename must contain at least one
661 slash ('/') if it is a real file (not 'internal' or 'none').
663 Note: external bitmaps are only known to work on ext2 and ext3.
664 Storing bitmap files on other filesystems may result in serious problems.
667 .BR \-\-bitmap\-chunk=
668 Set the chunksize of the bitmap. Each bit corresponds to that many
669 Kilobytes of storage.
670 When using a file based bitmap, the default is to use the smallest
671 size that is at-least 4 and requires no more than 2^21 chunks.
674 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
675 fit the bitmap into the available space.
677 A suffix of 'M' or 'G' can be given to indicate Megabytes or
678 Gigabytes respectively.
681 .BR \-W ", " \-\-write\-mostly
682 subsequent devices listed in a
687 command will be flagged as 'write-mostly'. This is valid for RAID1
688 only and means that the 'md' driver will avoid reading from these
689 devices if at all possible. This can be useful if mirroring over a
693 .BR \-\-write\-behind=
694 Specify that write-behind mode should be enabled (valid for RAID1
695 only). If an argument is specified, it will set the maximum number
696 of outstanding writes allowed. The default value is 256.
697 A write-intent bitmap is required in order to use write-behind
698 mode, and write-behind is only attempted on drives marked as
702 .BR \-\-assume\-clean
705 that the array pre-existed and is known to be clean. It can be useful
706 when trying to recover from a major failure as you can be sure that no
707 data will be affected unless you actually write to the array. It can
708 also be used when creating a RAID1 or RAID10 if you want to avoid the
709 initial resync, however this practice \(em while normally safe \(em is not
710 recommended. Use this only if you really know what you are doing.
712 When the devices that will be part of a new array were filled
713 with zeros before creation the operator knows the array is
714 actually clean. If that is the case, such as after running
715 badblocks, this argument can be used to tell mdadm the
716 facts the operator knows.
718 When an array is resized to a larger size with
719 .B "\-\-grow \-\-size="
720 the new space is normally resynced in that same way that the whole
721 array is resynced at creation. From Linux version 3.0,
723 can be used with that command to avoid the automatic resync.
726 .BR \-\-backup\-file=
729 is used to increase the number of raid-devices in a RAID5 or RAID6 if
730 there are no spare devices available, or to shrink, change RAID level
731 or layout. See the GROW MODE section below on RAID\-DEVICES CHANGES.
732 The file must be stored on a separate device, not on the RAID array
736 .BR \-N ", " \-\-name=
739 for the array. This is currently only effective when creating an
740 array with a version-1 superblock, or an array in a DDF container.
741 The name is a simple textual string that can be used to identify array
742 components when assembling. If name is needed but not specified, it
743 is taken from the basename of the device that is being created.
755 run the array, even if some of the components
756 appear to be active in another array or filesystem. Normally
758 will ask for confirmation before including such components in an
759 array. This option causes that question to be suppressed.
762 .BR \-f ", " \-\-force
765 accept the geometry and layout specified without question. Normally
767 will not allow creation of an array with only one device, and will try
768 to create a RAID5 array with one missing drive (as this makes the
769 initial resync work faster). With
772 will not try to be so clever.
775 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
776 Instruct mdadm how to create the device file if needed, possibly allocating
777 an unused minor number. "md" causes a non-partitionable array
778 to be used (though since Linux 2.6.28, these array devices are in fact
779 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
780 later) to be used. "yes" requires the named md device to have
781 a 'standard' format, and the type and minor number will be determined
782 from this. With mdadm 3.0, device creation is normally left up to
784 so this option is unlikely to be needed.
785 See DEVICE NAMES below.
787 The argument can also come immediately after
792 is not given on the command line or in the config file, then
798 is also given, then any
800 entries in the config file will override the
802 instruction given on the command line.
804 For partitionable arrays,
806 will create the device file for the whole array and for the first 4
807 partitions. A different number of partitions can be specified at the
808 end of this option (e.g.
810 If the device name ends with a digit, the partition names add a 'p',
812 .IR /dev/md/home1p3 .
813 If there is no trailing digit, then the partition names just have a
815 .IR /dev/md/scratch3 .
817 If the md device name is in a 'standard' format as described in DEVICE
818 NAMES, then it will be created, if necessary, with the appropriate
819 device number based on that name. If the device name is not in one of these
820 formats, then a unused device number will be allocated. The device
821 number will be considered unused if there is no active array for that
822 number, and there is no entry in /dev for that number and with a
823 non-standard name. Names that are not in 'standard' format are only
824 allowed in "/dev/md/".
826 This is meaningful with
833 .\".BR \-\-symlink = no
838 .\"to create devices in
840 .\"it will also create symlinks from
842 .\"with names starting with
848 .\"to suppress this, or
849 .\".B \-\-symlink=yes
850 .\"to enforce this even if it is suppressing
856 .BR \-a ", " "\-\-add"
857 This option can be used in Grow mode in two cases.
859 If the target array is a Linear array, then
861 can be used to add one or more devices to the array. They
862 are simply catenated on to the end of the array. Once added, the
863 devices cannot be removed.
867 option is being used to increase the number of devices in an array,
870 can be used to add some extra devices to be included in the array.
871 In most cases this is not needed as the extra devices can be added as
872 spares first, and then the number of raid-disks can be changed.
873 However for RAID0, it is not possible to add spares. So to increase
874 the number of devices in a RAID0, it is necessary to set the new
875 number of devices, and to add the new devices, in the same command.
880 .BR \-u ", " \-\-uuid=
881 uuid of array to assemble. Devices which don't have this uuid are
885 .BR \-m ", " \-\-super\-minor=
886 Minor number of device that array was created for. Devices which
887 don't have this minor number are excluded. If you create an array as
888 /dev/md1, then all superblocks will contain the minor number 1, even if
889 the array is later assembled as /dev/md2.
891 Giving the literal word "dev" for
895 to use the minor number of the md device that is being assembled.
898 .B \-\-super\-minor=dev
899 will look for super blocks with a minor number of 0.
902 is only relevant for v0.90 metadata, and should not normally be used.
908 .BR \-N ", " \-\-name=
909 Specify the name of the array to assemble. This must be the name
910 that was specified when creating the array. It must either match
911 the name stored in the superblock exactly, or it must match
914 prefixed to the start of the given name.
917 .BR \-f ", " \-\-force
918 Assemble the array even if the metadata on some devices appears to be
921 cannot find enough working devices to start the array, but can find
922 some devices that are recorded as having failed, then it will mark
923 those devices as working so that the array can be started.
924 An array which requires
926 to be started may contain data corruption. Use it carefully.
930 Attempt to start the array even if fewer drives were given than were
931 present last time the array was active. Normally if not all the
932 expected drives are found and
934 is not used, then the array will be assembled but not started.
937 an attempt will be made to start it anyway.
941 This is the reverse of
943 in that it inhibits the startup of array unless all expected drives
944 are present. This is only needed with
946 and can be used if the physical connections to devices are
947 not as reliable as you would like.
950 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
951 See this option under Create and Build options.
954 .BR \-b ", " \-\-bitmap=
955 Specify the bitmap file that was given when the array was created. If
958 bitmap, there is no need to specify this when assembling the array.
961 .BR \-\-backup\-file=
964 was used while reshaping an array (e.g. changing number of devices or
965 chunk size) and the system crashed during the critical section, then the same
969 to allow possibly corrupted data to be restored, and the reshape
973 .BR \-\-invalid\-backup
974 If the file needed for the above option is not available for any
975 reason an empty file can be given together with this option to
976 indicate that the backup file is invalid. In this case the data that
977 was being rearranged at the time of the crash could be irrecoverably
978 lost, but the rest of the array may still be recoverable. This option
979 should only be used as a last resort if there is no way to recover the
984 .BR \-U ", " \-\-update=
985 Update the superblock on each device while assembling the array. The
986 argument given to this flag can be one of
1001 option will adjust the superblock of an array what was created on a Sparc
1002 machine running a patched 2.2 Linux kernel. This kernel got the
1003 alignment of part of the superblock wrong. You can use the
1004 .B "\-\-examine \-\-sparc2.2"
1007 to see what effect this would have.
1011 option will update the
1012 .B "preferred minor"
1013 field on each superblock to match the minor number of the array being
1015 This can be useful if
1017 reports a different "Preferred Minor" to
1019 In some cases this update will be performed automatically
1020 by the kernel driver. In particular the update happens automatically
1021 at the first write to an array with redundancy (RAID level 1 or
1022 greater) on a 2.6 (or later) kernel.
1026 option will change the uuid of the array. If a UUID is given with the
1028 option that UUID will be used as a new UUID and will
1030 be used to help identify the devices in the array.
1033 is given, a random UUID is chosen.
1037 option will change the
1039 of the array as stored in the superblock. This is only supported for
1040 version-1 superblocks.
1044 option will change the
1046 as recorded in the superblock. For version-0 superblocks, this is the
1047 same as updating the UUID.
1048 For version-1 superblocks, this involves updating the name.
1052 option will cause the array to be marked
1054 meaning that any redundancy in the array (e.g. parity for RAID5,
1055 copies for RAID1) may be incorrect. This will cause the RAID system
1056 to perform a "resync" pass to make sure that all redundant information
1061 option allows arrays to be moved between machines with different
1063 When assembling such an array for the first time after a move, giving
1064 .B "\-\-update=byteorder"
1067 to expect superblocks to have their byteorder reversed, and will
1068 correct that order before assembling the array. This is only valid
1069 with original (Version 0.90) superblocks.
1073 option will correct the summaries in the superblock. That is the
1074 counts of total, working, active, failed, and spare devices.
1078 option will rarely be of use. It applies to version 1.1 and 1.2 metadata
1079 only (where the metadata is at the start of the device) and is only
1080 useful when the component device has changed size (typically become
1081 larger). The version 1 metadata records the amount of the device that
1082 can be used to store data, so if a device in a version 1.1 or 1.2
1083 array becomes larger, the metadata will still be visible, but the
1084 extra space will not. In this case it might be useful to assemble the
1086 .BR \-\-update=devicesize .
1089 to determine the maximum usable amount of space on each device and
1090 update the relevant field in the metadata.
1094 option can be used when an array has an internal bitmap which is
1095 corrupt in some way so that assembling the array normally fails. It
1096 will cause any internal bitmap to be ignored.
1098 .SH For Manage mode:
1101 .BR \-t ", " \-\-test
1102 Unless a more serious error occurred,
1104 will exit with a status of 2 if no changes were made to the array and
1105 0 if at least one change was made.
1106 This can be useful when an indirect specifier such as
1111 is used in requesting an operation on the array.
1113 will report failure if these specifiers didn't find any match.
1116 .BR \-a ", " \-\-add
1117 hot-add listed devices.
1118 If a device appears to have recently been part of the array
1119 (possibly it failed or was removed) the device is re\-added as describe
1121 If that fails or the device was never part of the array, the device is
1122 added as a hot-spare.
1123 If the array is degraded, it will immediately start to rebuild data
1126 Note that this and the following options are only meaningful on array
1127 with redundancy. They don't apply to RAID0 or Linear.
1131 re\-add a device that was previous removed from an array.
1132 If the metadata on the device reports that it is a member of the
1133 array, and the slot that it used is still vacant, then the device will
1134 be added back to the array in the same position. This will normally
1135 cause the data for that device to be recovered. However based on the
1136 event count on the device, the recovery may only require sections that
1137 are flagged a write-intent bitmap to be recovered or may not require
1138 any recovery at all.
1140 When used on an array that has no metadata (i.e. it was built with
1142 it will be assumed that bitmap-based recovery is enough to make the
1143 device fully consistent with the array.
1147 can be accompanied by
1148 .BR \-\-update=devicesize .
1149 See the description of this option when used in Assemble mode for an
1150 explanation of its use.
1152 If the device name given is
1154 then mdadm will try to find any device that looks like it should be
1155 part of the array but isn't and will try to re\-add all such devices.
1158 .BR \-r ", " \-\-remove
1159 remove listed devices. They must not be active. i.e. they should
1160 be failed or spare devices. As well as the name of a device file
1169 The first causes all failed device to be removed. The second causes
1170 any device which is no longer connected to the system (i.e an 'open'
1173 to be removed. This will only succeed for devices that are spares or
1174 have already been marked as failed.
1177 .BR \-f ", " \-\-fail
1178 mark listed devices as faulty.
1179 As well as the name of a device file, the word
1181 can be given. This will cause any device that has been detached from
1182 the system to be marked as failed. It can then be removed.
1190 .BR \-\-write\-mostly
1191 Subsequent devices that are added or re\-added will have the 'write-mostly'
1192 flag set. This is only valid for RAID1 and means that the 'md' driver
1193 will avoid reading from these devices if possible.
1196 Subsequent devices that are added or re\-added will have the 'write-mostly'
1200 Each of these options requires that the first device listed is the array
1201 to be acted upon, and the remainder are component devices to be added,
1202 removed, marked as faulty, etc. Several different operations can be
1203 specified for different devices, e.g.
1205 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1207 Each operation applies to all devices listed until the next
1210 If an array is using a write-intent bitmap, then devices which have
1211 been removed can be re\-added in a way that avoids a full
1212 reconstruction but instead just updates the blocks that have changed
1213 since the device was removed. For arrays with persistent metadata
1214 (superblocks) this is done automatically. For arrays created with
1216 mdadm needs to be told that this device we removed recently with
1219 Devices can only be removed from an array if they are not in active
1220 use, i.e. that must be spares or failed devices. To remove an active
1221 device, it must first be marked as
1227 .BR \-Q ", " \-\-query
1228 Examine a device to see
1229 (1) if it is an md device and (2) if it is a component of an md
1231 Information about what is discovered is presented.
1234 .BR \-D ", " \-\-detail
1235 Print details of one or more md devices.
1238 .BR \-\-detail\-platform
1239 Print details of the platform's RAID capabilities (firmware / hardware
1240 topology) for a given metadata format.
1243 .BR \-Y ", " \-\-export
1248 output will be formatted as
1250 pairs for easy import into the environment.
1253 .BR \-E ", " \-\-examine
1254 Print contents of the metadata stored on the named device(s).
1255 Note the contrast between
1260 applies to devices which are components of an array, while
1262 applies to a whole array which is currently active.
1265 If an array was created on a SPARC machine with a 2.2 Linux kernel
1266 patched with RAID support, the superblock will have been created
1267 incorrectly, or at least incompatibly with 2.4 and later kernels.
1272 will fix the superblock before displaying it. If this appears to do
1273 the right thing, then the array can be successfully assembled using
1274 .BR "\-\-assemble \-\-update=sparc2.2" .
1277 .BR \-X ", " \-\-examine\-bitmap
1278 Report information about a bitmap file.
1279 The argument is either an external bitmap file or an array component
1280 in case of an internal bitmap. Note that running this on an array
1283 does not report the bitmap for that array.
1286 .BR \-R ", " \-\-run
1287 start a partially assembled array. If
1289 did not find enough devices to fully start the array, it might leaving
1290 it partially assembled. If you wish, you can then use
1292 to start the array in degraded mode.
1295 .BR \-S ", " \-\-stop
1296 deactivate array, releasing all resources.
1299 .BR \-o ", " \-\-readonly
1300 mark array as readonly.
1303 .BR \-w ", " \-\-readwrite
1304 mark array as readwrite.
1307 .B \-\-zero\-superblock
1308 If the device contains a valid md superblock, the block is
1309 overwritten with zeros. With
1311 the block where the superblock would be is overwritten even if it
1312 doesn't appear to be valid.
1315 .B \-\-kill\-subarray=
1316 If the device is a container and the argument to \-\-kill\-subarray
1317 specifies an inactive subarray in the container, then the subarray is
1318 deleted. Deleting all subarrays will leave an 'empty-container' or
1319 spare superblock on the drives. See \-\-zero\-superblock for completely
1320 removing a superblock. Note that some formats depend on the subarray
1321 index for generating a UUID, this command will fail if it would change
1322 the UUID of an active subarray.
1325 .B \-\-update\-subarray=
1326 If the device is a container and the argument to \-\-update\-subarray
1327 specifies a subarray in the container, then attempt to update the given
1328 superblock field in the subarray. See below in
1333 .BR \-t ", " \-\-test
1338 is set to reflect the status of the device. See below in
1343 .BR \-W ", " \-\-wait
1344 For each md device given, wait for any resync, recovery, or reshape
1345 activity to finish before returning.
1347 will return with success if it actually waited for every device
1348 listed, otherwise it will return failure.
1352 For each md device given, or each device in /proc/mdstat if
1354 is given, arrange for the array to be marked clean as soon as possible.
1356 will return with success if the array uses external metadata and we
1357 successfully waited. For native arrays this returns immediately as the
1358 kernel handles dirty-clean transitions at shutdown. No action is taken
1359 if safe-mode handling is disabled.
1361 .SH For Incremental Assembly mode:
1363 .BR \-\-rebuild\-map ", " \-r
1364 Rebuild the map file
1365 .RB ( /var/run/mdadm/map )
1368 uses to help track which arrays are currently being assembled.
1371 .BR \-\-run ", " \-R
1372 Run any array assembled as soon as a minimal number of devices are
1373 available, rather than waiting until all expected devices are present.
1376 .BR \-\-scan ", " \-s
1377 Only meaningful with
1381 file for arrays that are being incrementally assembled and will try to
1382 start any that are not already started. If any such array is listed
1385 as requiring an external bitmap, that bitmap will be attached first.
1388 .BR \-\-fail ", " \-f
1389 This allows the hot-plug system to remove devices that have fully disappeared
1390 from the kernel. It will first fail and then remove the device from any
1391 array it belongs to.
1392 The device name given should be a kernel device name such as "sda",
1398 Only used with \-\-fail. The 'path' given will be recorded so that if
1399 a new device appears at the same location it can be automatically
1400 added to the same array. This allows the failed device to be
1401 automatically replaced by a new device without metadata if it appears
1402 at specified path. This option is normally only set by a
1406 .SH For Monitor mode:
1408 .BR \-m ", " \-\-mail
1409 Give a mail address to send alerts to.
1412 .BR \-p ", " \-\-program ", " \-\-alert
1413 Give a program to be run whenever an event is detected.
1416 .BR \-y ", " \-\-syslog
1417 Cause all events to be reported through 'syslog'. The messages have
1418 facility of 'daemon' and varying priorities.
1421 .BR \-d ", " \-\-delay
1422 Give a delay in seconds.
1424 polls the md arrays and then waits this many seconds before polling
1425 again. The default is 60 seconds. Since 2.6.16, there is no need to
1426 reduce this as the kernel alerts
1428 immediately when there is any change.
1431 .BR \-r ", " \-\-increment
1432 Give a percentage increment.
1434 will generate RebuildNN events with the given percentage increment.
1437 .BR \-f ", " \-\-daemonise
1440 to run as a background daemon if it decides to monitor anything. This
1441 causes it to fork and run in the child, and to disconnect from the
1442 terminal. The process id of the child is written to stdout.
1445 which will only continue monitoring if a mail address or alert program
1446 is found in the config file.
1449 .BR \-i ", " \-\-pid\-file
1452 is running in daemon mode, write the pid of the daemon process to
1453 the specified file, instead of printing it on standard output.
1456 .BR \-1 ", " \-\-oneshot
1457 Check arrays only once. This will generate
1459 events and more significantly
1465 .B " mdadm \-\-monitor \-\-scan \-1"
1467 from a cron script will ensure regular notification of any degraded arrays.
1470 .BR \-t ", " \-\-test
1473 alert for every array found at startup. This alert gets mailed and
1474 passed to the alert program. This can be used for testing that alert
1475 message do get through successfully.
1479 This inhibits the functionality for moving spares between arrays.
1480 Only one monitoring process started with
1482 but without this flag is allowed, otherwise the two could interfere
1489 .B mdadm \-\-assemble
1490 .I md-device options-and-component-devices...
1493 .B mdadm \-\-assemble \-\-scan
1494 .I md-devices-and-options...
1497 .B mdadm \-\-assemble \-\-scan
1501 This usage assembles one or more RAID arrays from pre-existing components.
1502 For each array, mdadm needs to know the md device, the identity of the
1503 array, and a number of component-devices. These can be found in a number of ways.
1505 In the first usage example (without the
1507 the first device given is the md device.
1508 In the second usage example, all devices listed are treated as md
1509 devices and assembly is attempted.
1510 In the third (where no devices are listed) all md devices that are
1511 listed in the configuration file are assembled. If no arrays are
1512 described by the configuration file, then any arrays that
1513 can be found on unused devices will be assembled.
1515 If precisely one device is listed, but
1521 was given and identity information is extracted from the configuration file.
1523 The identity can be given with the
1529 option, will be taken from the md-device record in the config file, or
1530 will be taken from the super block of the first component-device
1531 listed on the command line.
1533 Devices can be given on the
1535 command line or in the config file. Only devices which have an md
1536 superblock which contains the right identity will be considered for
1539 The config file is only used if explicitly named with
1541 or requested with (a possibly implicit)
1546 .B /etc/mdadm/mdadm.conf
1551 is not given, then the config file will only be used to find the
1552 identity of md arrays.
1554 Normally the array will be started after it is assembled. However if
1556 is not given and not all expected drives were listed, then the array
1557 is not started (to guard against usage errors). To insist that the
1558 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1567 does not create any entries in
1571 It does record information in
1572 .B /var/run/mdadm/map
1575 to choose the correct name.
1579 detects that udev is not configured, it will create the devices in
1583 In Linux kernels prior to version 2.6.28 there were two distinctly
1584 different types of md devices that could be created: one that could be
1585 partitioned using standard partitioning tools and one that could not.
1586 Since 2.6.28 that distinction is no longer relevant as both type of
1587 devices can be partitioned.
1589 will normally create the type that originally could not be partitioned
1590 as it has a well defined major number (9).
1592 Prior to 2.6.28, it is important that mdadm chooses the correct type
1593 of array device to use. This can be controlled with the
1595 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1596 to use a partitionable device rather than the default.
1598 In the no-udev case, the value given to
1600 can be suffixed by a number. This tells
1602 to create that number of partition devices rather than the default of 4.
1606 can also be given in the configuration file as a word starting
1608 on the ARRAY line for the relevant array.
1615 and no devices are listed,
1617 will first attempt to assemble all the arrays listed in the config
1620 If no arrays are listed in the config (other than those marked
1622 it will look through the available devices for possible arrays and
1623 will try to assemble anything that it finds. Arrays which are tagged
1624 as belonging to the given homehost will be assembled and started
1625 normally. Arrays which do not obviously belong to this host are given
1626 names that are expected not to conflict with anything local, and are
1627 started "read-auto" so that nothing is written to any device until the
1628 array is written to. i.e. automatic resync etc is delayed.
1632 finds a consistent set of devices that look like they should comprise
1633 an array, and if the superblock is tagged as belonging to the given
1634 home host, it will automatically choose a device name and try to
1635 assemble the array. If the array uses version-0.90 metadata, then the
1637 number as recorded in the superblock is used to create a name in
1641 If the array uses version-1 metadata, then the
1643 from the superblock is used to similarly create a name in
1645 (the name will have any 'host' prefix stripped first).
1647 This behaviour can be modified by the
1651 configuration file. This line can indicate that specific metadata
1652 type should, or should not, be automatically assembled. If an array
1653 is found which is not listed in
1655 and has a metadata format that is denied by the
1657 line, then it will not be assembled.
1660 line can also request that all arrays identified as being for this
1661 homehost should be assembled regardless of their metadata type.
1664 for further details.
1666 Note: Auto assembly cannot be used for assembling and activating some
1667 arrays which are undergoing reshape. In particular as the
1669 cannot be given, any reshape which requires a backup-file to continue
1670 cannot be started by auto assembly. An array which is growing to more
1671 devices and has passed the critical section can be assembled using
1682 .BI \-\-raid\-devices= Z
1686 This usage is similar to
1688 The difference is that it creates an array without a superblock. With
1689 these arrays there is no difference between initially creating the array and
1690 subsequently assembling the array, except that hopefully there is useful
1691 data there in the second case.
1693 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1694 one of their synonyms. All devices must be listed and the array will
1695 be started once complete. It will often be appropriate to use
1696 .B \-\-assume\-clean
1697 with levels raid1 or raid10.
1708 .BI \-\-raid\-devices= Z
1712 This usage will initialise a new md array, associate some devices with
1713 it, and activate the array.
1715 The named device will normally not exist when
1716 .I "mdadm \-\-create"
1717 is run, but will be created by
1719 once the array becomes active.
1721 As devices are added, they are checked to see if they contain RAID
1722 superblocks or filesystems. They are also checked to see if the variance in
1723 device size exceeds 1%.
1725 If any discrepancy is found, the array will not automatically be run, though
1728 can override this caution.
1730 To create a "degraded" array in which some devices are missing, simply
1731 give the word "\fBmissing\fP"
1732 in place of a device name. This will cause
1734 to leave the corresponding slot in the array empty.
1735 For a RAID4 or RAID5 array at most one slot can be
1736 "\fBmissing\fP"; for a RAID6 array at most two slots.
1737 For a RAID1 array, only one real device needs to be given. All of the
1741 When creating a RAID5 array,
1743 will automatically create a degraded array with an extra spare drive.
1744 This is because building the spare into a degraded array is in general
1745 faster than resyncing the parity on a non-degraded, but not clean,
1746 array. This feature can be overridden with the
1750 When creating an array with version-1 metadata a name for the array is
1752 If this is not given with the
1756 will choose a name based on the last component of the name of the
1757 device being created. So if
1759 is being created, then the name
1764 is being created, then the name
1768 When creating a partition based array, using
1770 with version-1.x metadata, the partition type should be set to
1772 (non fs-data). This type selection allows for greater precision since
1773 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
1774 might create problems in the event of array recovery through a live cdrom.
1776 A new array will normally get a randomly assigned 128bit UUID which is
1777 very likely to be unique. If you have a specific need, you can choose
1778 a UUID for the array by giving the
1780 option. Be warned that creating two arrays with the same UUID is a
1781 recipe for disaster. Also, using
1783 when creating a v0.90 array will silently override any
1788 .\"option is given, it is not necessary to list any component-devices in this command.
1789 .\"They can be added later, before a
1793 .\"is given, the apparent size of the smallest drive given is used.
1795 When creating an array within a
1798 can be given either the list of devices to use, or simply the name of
1799 the container. The former case gives control over which devices in
1800 the container will be used for the array. The latter case allows
1802 to automatically choose which devices to use based on how much spare
1805 The General Management options that are valid with
1810 insist on running the array even if some devices look like they might
1815 start the array readonly \(em not supported yet.
1822 .I options... devices...
1825 This usage will allow individual devices in an array to be failed,
1826 removed or added. It is possible to perform multiple operations with
1827 on command. For example:
1829 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
1835 and will then remove it from the array and finally add it back
1836 in as a spare. However only one md array can be affected by a single
1839 When a device is added to an active array, mdadm checks to see if it
1840 has metadata on it which suggests that it was recently a member of the
1841 array. If it does, it tries to "re\-add" the device. If there have
1842 been no changes since the device was removed, or if the array has a
1843 write-intent bitmap which has recorded whatever changes there were,
1844 then the device will immediately become a full member of the array and
1845 those differences recorded in the bitmap will be resolved.
1855 MISC mode includes a number of distinct operations that
1856 operate on distinct devices. The operations are:
1859 The device is examined to see if it is
1860 (1) an active md array, or
1861 (2) a component of an md array.
1862 The information discovered is reported.
1866 The device should be an active md device.
1868 will display a detailed description of the array.
1872 will cause the output to be less detailed and the format to be
1873 suitable for inclusion in
1877 will normally be 0 unless
1879 failed to get useful information about the device(s); however, if the
1881 option is given, then the exit status will be:
1885 The array is functioning normally.
1888 The array has at least one failed device.
1891 The array has multiple failed devices such that it is unusable.
1894 There was an error while trying to get information about the device.
1898 .B \-\-detail\-platform
1899 Print detail of the platform's RAID capabilities (firmware / hardware
1900 topology). If the metadata is specified with
1904 then the return status will be:
1908 metadata successfully enumerated its platform components on this system
1911 metadata is platform independent
1914 metadata failed to find its platform components on this system
1918 .B \-\-update\-subarray=
1919 If the device is a container and the argument to \-\-update\-subarray
1920 specifies a subarray in the container, then attempt to update the given
1921 superblock field in the subarray. Similar to updating an array in
1922 "assemble" mode, the field to update is selected by
1926 option. Currently only
1932 option updates the subarray name in the metadata, it may not affect the
1933 device node name or the device node symlink until the subarray is
1934 re\-assembled. If updating
1936 would change the UUID of an active subarray this operation is blocked,
1937 and the command will end in an error.
1941 The device should be a component of an md array.
1943 will read the md superblock of the device and display the contents.
1948 is given, then multiple devices that are components of the one array
1949 are grouped together and reported in a single entry suitable
1955 without listing any devices will cause all devices listed in the
1956 config file to be examined.
1960 The devices should be active md arrays which will be deactivated, as
1961 long as they are not currently in use.
1965 This will fully activate a partially assembled md array.
1969 This will mark an active array as read-only, providing that it is
1970 not currently being used.
1976 array back to being read/write.
1980 For all operations except
1983 will cause the operation to be applied to all arrays listed in
1988 causes all devices listed in the config file to be examined.
1991 .BR \-b ", " \-\-brief
1992 Be less verbose. This is used with
2000 gives an intermediate level of verbosity.
2006 .B mdadm \-\-monitor
2007 .I options... devices...
2012 to periodically poll a number of md arrays and to report on any events
2015 will never exit once it decides that there are arrays to be checked,
2016 so it should normally be run in the background.
2018 As well as reporting events,
2020 may move a spare drive from one array to another if they are in the
2025 and if the destination array has a failed drive but no spares.
2027 If any devices are listed on the command line,
2029 will only monitor those devices. Otherwise all arrays listed in the
2030 configuration file will be monitored. Further, if
2032 is given, then any other md devices that appear in
2034 will also be monitored.
2036 The result of monitoring the arrays is the generation of events.
2037 These events are passed to a separate program (if specified) and may
2038 be mailed to a given E-mail address.
2040 When passing events to a program, the program is run once for each event,
2041 and is given 2 or 3 command-line arguments: the first is the
2042 name of the event (see below), the second is the name of the
2043 md device which is affected, and the third is the name of a related
2044 device if relevant (such as a component device that has failed).
2048 is given, then a program or an E-mail address must be specified on the
2049 command line or in the config file. If neither are available, then
2051 will not monitor anything.
2055 will continue monitoring as long as something was found to monitor. If
2056 no program or email is given, then each event is reported to
2059 The different events are:
2063 .B DeviceDisappeared
2064 An md array which previously was configured appears to no longer be
2065 configured. (syslog priority: Critical)
2069 was told to monitor an array which is RAID0 or Linear, then it will
2071 .B DeviceDisappeared
2072 with the extra information
2074 This is because RAID0 and Linear do not support the device-failed,
2075 hot-spare and resync operations which are monitored.
2079 An md array started reconstruction. (syslog priority: Warning)
2085 is a two-digit number (ie. 05, 48). This indicates that rebuild
2086 has passed that many percent of the total. The events are generated
2087 with fixed increment since 0. Increment size may be specified with
2088 a commandline option (default is 20). (syslog priority: Warning)
2092 An md array that was rebuilding, isn't any more, either because it
2093 finished normally or was aborted. (syslog priority: Warning)
2097 An active component device of an array has been marked as
2098 faulty. (syslog priority: Critical)
2102 A spare component device which was being rebuilt to replace a faulty
2103 device has failed. (syslog priority: Critical)
2107 A spare component device which was being rebuilt to replace a faulty
2108 device has been successfully rebuilt and has been made active.
2109 (syslog priority: Info)
2113 A new md array has been detected in the
2115 file. (syslog priority: Info)
2119 A newly noticed array appears to be degraded. This message is not
2122 notices a drive failure which causes degradation, but only when
2124 notices that an array is degraded when it first sees the array.
2125 (syslog priority: Critical)
2129 A spare drive has been moved from one array in a
2133 to another to allow a failed drive to be replaced.
2134 (syslog priority: Info)
2140 has been told, via the config file, that an array should have a certain
2141 number of spare devices, and
2143 detects that it has fewer than this number when it first sees the
2144 array, it will report a
2147 (syslog priority: Warning)
2151 An array was found at startup, and the
2154 (syslog priority: Info)
2164 cause Email to be sent. All events cause the program to be run.
2165 The program is run with two or three arguments: the event
2166 name, the array device and possibly a second device.
2168 Each event has an associated array device (e.g.
2170 and possibly a second device. For
2175 the second device is the relevant component device.
2178 the second device is the array that the spare was moved from.
2182 to move spares from one array to another, the different arrays need to
2183 be labeled with the same
2185 or the spares must be allowed to migrate through matching POLICY domains
2186 in the configuration file. The
2188 name can be any string; it is only necessary that different spare
2189 groups use different names.
2193 detects that an array in a spare group has fewer active
2194 devices than necessary for the complete array, and has no spare
2195 devices, it will look for another array in the same spare group that
2196 has a full complement of working drive and a spare. It will then
2197 attempt to remove the spare from the second drive and add it to the
2199 If the removal succeeds but the adding fails, then it is added back to
2202 If the spare group for a degraded array is not defined,
2204 will look at the rules of spare migration specified by POLICY lines in
2206 and then follow similar steps as above if a matching spare is found.
2209 The GROW mode is used for changing the size or shape of an active
2211 For this to work, the kernel must support the necessary change.
2212 Various types of growth are being added during 2.6 development.
2214 Currently the supported changes include
2216 change the "size" attribute for RAID1, RAID4, RAID5 and RAID6.
2218 increase or decrease the "raid\-devices" attribute of RAID0, RAID1, RAID4,
2221 change the chunk-size and layout of RAID0, RAID4, RAID5 and RAID6.
2223 convert between RAID1 and RAID5, between RAID5 and RAID6, between
2224 RAID0, RAID4, and RAID5, and between RAID0 and RAID10 (in the near-2 mode).
2226 add a write-intent bitmap to any array which supports these bitmaps, or
2227 remove a write-intent bitmap from such an array.
2230 Using GROW on containers is currently supported only for Intel's IMSM
2231 container format. The number of devices in a container can be
2232 increased - which affects all arrays in the container - or an array
2233 in a container can be converted between levels where those levels are
2234 supported by the container, and the conversion is on of those listed
2235 above. Resizing arrays in an IMSM container with
2237 is not yet supported.
2239 Grow functionality (e.g. expand a number of raid devices) for Intel's
2240 IMSM container format has an experimental status. It is guarded by the
2241 .B MDADM_EXPERIMENTAL
2242 environment variable which must be set to '1' for a GROW command to
2244 This is for the following reasons:
2247 Intel's native IMSM check-pointing is not fully tested yet.
2248 This can causes IMSM incompatibility during the grow process: an array
2249 which is growing cannot roam between Microsoft Windows(R) and Linux
2253 Interrupting a grow operation is not recommended, because it
2254 has not been fully tested for Intel's IMSM container format yet.
2257 Note: Intel's native checkpointing doesn't use
2259 option and it is transparent for assembly feature.
2262 Normally when an array is built the "size" is taken from the smallest
2263 of the drives. If all the small drives in an arrays are, one at a
2264 time, removed and replaced with larger drives, then you could have an
2265 array of large drives with only a small amount used. In this
2266 situation, changing the "size" with "GROW" mode will allow the extra
2267 space to start being used. If the size is increased in this way, a
2268 "resync" process will start to make sure the new parts of the array
2271 Note that when an array changes size, any filesystem that may be
2272 stored in the array will not automatically grow or shrink to use or
2273 vacate the space. The
2274 filesystem will need to be explicitly told to use the extra space
2275 after growing, or to reduce its size
2277 to shrinking the array.
2279 Also the size of an array cannot be changed while it has an active
2280 bitmap. If an array has a bitmap, it must be removed before the size
2281 can be changed. Once the change is complete a new bitmap can be created.
2283 .SS RAID\-DEVICES CHANGES
2285 A RAID1 array can work with any number of devices from 1 upwards
2286 (though 1 is not very useful). There may be times which you want to
2287 increase or decrease the number of active devices. Note that this is
2288 different to hot-add or hot-remove which changes the number of
2291 When reducing the number of devices in a RAID1 array, the slots which
2292 are to be removed from the array must already be vacant. That is, the
2293 devices which were in those slots must be failed and removed.
2295 When the number of devices is increased, any hot spares that are
2296 present will be activated immediately.
2298 Changing the number of active devices in a RAID5 or RAID6 is much more
2299 effort. Every block in the array will need to be read and written
2300 back to a new location. From 2.6.17, the Linux Kernel is able to
2301 increase the number of devices in a RAID5 safely, including restarting
2302 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2303 increase or decrease the number of devices in a RAID5 or RAID6.
2305 From 2.6.35, the Linux Kernel is able to convert a RAID0 in to a RAID4
2308 uses this functionality and the ability to add
2309 devices to a RAID4 to allow devices to be added to a RAID0. When
2310 requested to do this,
2312 will convert the RAID0 to a RAID4, add the necessary disks and make
2313 the reshape happen, and then convert the RAID4 back to RAID0.
2315 When decreasing the number of devices, the size of the array will also
2316 decrease. If there was data in the array, it could get destroyed and
2317 this is not reversible, so you should firstly shrink the filesystem on
2318 the array to fit within the new size. To help prevent accidents,
2320 requires that the size of the array be decreased first with
2321 .BR "mdadm --grow --array-size" .
2322 This is a reversible change which simply makes the end of the array
2323 inaccessible. The integrity of any data can then be checked before
2324 the non-reversible reduction in the number of devices is request.
2326 When relocating the first few stripes on a RAID5 or RAID6, it is not
2327 possible to keep the data on disk completely consistent and
2328 crash-proof. To provide the required safety, mdadm disables writes to
2329 the array while this "critical section" is reshaped, and takes a
2330 backup of the data that is in that section. For grows, this backup may be
2331 stored in any spare devices that the array has, however it can also be
2332 stored in a separate file specified with the
2334 option, and is required to be specified for shrinks, RAID level
2335 changes and layout changes. If this option is used, and the system
2336 does crash during the critical period, the same file must be passed to
2338 to restore the backup and reassemble the array. When shrinking rather
2339 than growing the array, the reshape is done from the end towards the
2340 beginning, so the "critical section" is at the end of the reshape.
2344 Changing the RAID level of any array happens instantaneously. However
2345 in the RAID5 to RAID6 case this requires a non-standard layout of the
2346 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2347 required before the change can be accomplished. So while the level
2348 change is instant, the accompanying layout change can take quite a
2351 is required. If the array is not simultaneously being grown or
2352 shrunk, so that the array size will remain the same - for example,
2353 reshaping a 3-drive RAID5 into a 4-drive RAID6 - the backup file will
2354 be used not just for a "cricital section" but throughout the reshape
2355 operation, as described below under LAYOUT CHANGES.
2357 .SS CHUNK-SIZE AND LAYOUT CHANGES
2359 Changing the chunk-size of layout without also changing the number of
2360 devices as the same time will involve re-writing all blocks in-place.
2361 To ensure against data loss in the case of a crash, a
2363 must be provided for these changes. Small sections of the array will
2364 be copied to the backup file while they are being rearranged. This
2365 means that all the data is copied twice, once to the backup and once
2366 to the new layout on the array, so this type of reshape will go very
2369 If the reshape is interrupted for any reason, this backup file must be
2371 .B "mdadm --assemble"
2372 so the array can be reassembled. Consequently the file cannot be
2373 stored on the device being reshaped.
2378 A write-intent bitmap can be added to, or removed from, an active
2379 array. Either internal bitmaps, or bitmaps stored in a separate file,
2380 can be added. Note that if you add a bitmap stored in a file which is
2381 in a filesystem that is on the RAID array being affected, the system
2382 will deadlock. The bitmap must be on a separate filesystem.
2384 .SH INCREMENTAL MODE
2388 .B mdadm \-\-incremental
2394 .B mdadm \-\-incremental \-\-fail
2398 .B mdadm \-\-incremental \-\-rebuild\-map
2401 .B mdadm \-\-incremental \-\-run \-\-scan
2404 This mode is designed to be used in conjunction with a device
2405 discovery system. As devices are found in a system, they can be
2407 .B "mdadm \-\-incremental"
2408 to be conditionally added to an appropriate array.
2410 Conversely, it can also be used with the
2412 flag to do just the opposite and find whatever array a particular device
2413 is part of and remove the device from that array.
2415 If the device passed is a
2417 device created by a previous call to
2419 then rather than trying to add that device to an array, all the arrays
2420 described by the metadata of the container will be started.
2423 performs a number of tests to determine if the device is part of an
2424 array, and which array it should be part of. If an appropriate array
2425 is found, or can be created,
2427 adds the device to the array and conditionally starts the array.
2431 will normally only add devices to an array which were previously working
2432 (active or spare) parts of that array. The support for automatic
2433 inclusion of a new drive as a spare in some array requires
2434 a configuration through POLICY in config file.
2438 makes are as follow:
2440 Is the device permitted by
2442 That is, is it listed in a
2444 line in that file. If
2446 is absent then the default it to allow any device. Similar if
2448 contains the special word
2450 then any device is allowed. Otherwise the device name given to
2452 must match one of the names or patterns in a
2457 Does the device have a valid md superblock? If a specific metadata
2458 version is requested with
2462 then only that style of metadata is accepted, otherwise
2464 finds any known version of metadata. If no
2466 metadata is found, the device may be still added to an array
2467 as a spare if POLICY allows.
2471 Does the metadata match an expected array?
2472 The metadata can match in two ways. Either there is an array listed
2475 which identifies the array (either by UUID, by name, by device list,
2476 or by minor-number), or the array was created with a
2482 or on the command line.
2485 is not able to positively identify the array as belonging to the
2486 current host, the device will be rejected.
2491 keeps a list of arrays that it has partially assembled in
2492 .B /var/run/mdadm/map
2494 .B /var/run/mdadm.map
2495 if the directory doesn't exist. Or maybe even
2496 .BR /dev/.mdadm.map ).
2497 If no array exists which matches
2498 the metadata on the new device,
2500 must choose a device name and unit number. It does this based on any
2503 or any name information stored in the metadata. If this name
2504 suggests a unit number, that number will be used, otherwise a free
2505 unit number will be chosen. Normally
2507 will prefer to create a partitionable array, however if the
2511 suggests that a non-partitionable array is preferred, that will be
2514 If the array is not found in the config file and its metadata does not
2515 identify it as belonging to the "homehost", then
2517 will choose a name for the array which is certain not to conflict with
2518 any array which does belong to this host. It does this be adding an
2519 underscore and a small number to the name preferred by the metadata.
2521 Once an appropriate array is found or created and the device is added,
2523 must decide if the array is ready to be started. It will
2524 normally compare the number of available (non-spare) devices to the
2525 number of devices that the metadata suggests need to be active. If
2526 there are at least that many, the array will be started. This means
2527 that if any devices are missing the array will not be restarted.
2533 in which case the array will be run as soon as there are enough
2534 devices present for the data to be accessible. For a RAID1, that
2535 means one device will start the array. For a clean RAID5, the array
2536 will be started as soon as all but one drive is present.
2538 Note that neither of these approaches is really ideal. If it can
2539 be known that all device discovery has completed, then
2543 can be run which will try to start all arrays that are being
2544 incrementally assembled. They are started in "read-auto" mode in
2545 which they are read-only until the first write request. This means
2546 that no metadata updates are made and no attempt at resync or recovery
2547 happens. Further devices that are found before the first write can
2548 still be added safely.
2551 This section describes environment variables that affect how mdadm
2556 Setting this value to 1 will prevent mdadm from automatically launching
2557 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2563 does not create any device nodes in /dev, but leaves that task to
2567 appears not to be configured, or if this environment variable is set
2570 will create and devices that are needed.
2574 .B " mdadm \-\-query /dev/name-of-device"
2576 This will find out if a given device is a RAID array, or is part of
2577 one, and will provide brief information about the device.
2579 .B " mdadm \-\-assemble \-\-scan"
2581 This will assemble and start all arrays listed in the standard config
2582 file. This command will typically go in a system startup file.
2584 .B " mdadm \-\-stop \-\-scan"
2586 This will shut down all arrays that can be shut down (i.e. are not
2587 currently in use). This will typically go in a system shutdown script.
2589 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2591 If (and only if) there is an Email address or program given in the
2592 standard config file, then
2593 monitor the status of all arrays listed in that file by
2594 polling them ever 2 minutes.
2596 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2598 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2601 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2603 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2605 This will create a prototype config file that describes currently
2606 active arrays that are known to be made from partitions of IDE or SCSI drives.
2607 This file should be reviewed before being used as it may
2608 contain unwanted detail.
2610 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2612 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
2614 This will find arrays which could be assembled from existing IDE and
2615 SCSI whole drives (not partitions), and store the information in the
2616 format of a config file.
2617 This file is very likely to contain unwanted detail, particularly
2620 entries. It should be reviewed and edited before being used as an
2623 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
2625 .B " mdadm \-Ebsc partitions"
2627 Create a list of devices by reading
2628 .BR /proc/partitions ,
2629 scan these for RAID superblocks, and printout a brief listing of all
2632 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
2634 Scan all partitions and devices listed in
2635 .BR /proc/partitions
2638 out of all such devices with a RAID superblock with a minor number of 0.
2640 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /var/run/mdadm"
2642 If config file contains a mail address or alert program, run mdadm in
2643 the background in monitor mode monitoring all md devices. Also write
2644 pid of mdadm daemon to
2645 .BR /var/run/mdadm .
2647 .B " mdadm \-Iq /dev/somedevice"
2649 Try to incorporate newly discovered device into some array as
2652 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
2654 Rebuild the array map from any current arrays, and then start any that
2657 .B " mdadm /dev/md4 --fail detached --remove detached"
2659 Any devices which are components of /dev/md4 will be marked as faulty
2660 and then remove from the array.
2662 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4"
2666 which is currently a RAID5 array will be converted to RAID6. There
2667 should normally already be a spare drive attached to the array as a
2668 RAID6 needs one more drive than a matching RAID5.
2670 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
2672 Create a DDF array over 6 devices.
2674 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
2676 Create a RAID5 array over any 3 devices in the given DDF set. Use
2677 only 30 gigabytes of each device.
2679 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
2681 Assemble a pre-exist ddf array.
2683 .B " mdadm -I /dev/md/ddf1"
2685 Assemble all arrays contained in the ddf array, assigning names as
2688 .B " mdadm \-\-create \-\-help"
2690 Provide help about the Create mode.
2692 .B " mdadm \-\-config \-\-help"
2694 Provide help about the format of the config file.
2696 .B " mdadm \-\-help"
2698 Provide general help.
2708 lists all active md devices with information about them.
2710 uses this to find arrays when
2712 is given in Misc mode, and to monitor array reconstruction
2717 The config file lists which devices may be scanned to see if
2718 they contain MD super block, and gives identifying information
2719 (e.g. UUID) about known MD arrays. See
2723 .SS /var/run/mdadm/map
2726 mode is used, this file gets a list of arrays currently being created.
2729 does not exist as a directory, then
2730 .B /var/run/mdadm.map
2733 is not available (as may be the case during early boot),
2735 is used on the basis that
2737 is usually available very early in boot.
2742 understand two sorts of names for array devices.
2744 The first is the so-called 'standard' format name, which matches the
2745 names used by the kernel and which appear in
2748 The second sort can be freely chosen, but must reside in
2750 When giving a device name to
2752 to create or assemble an array, either full path name such as
2756 can be given, or just the suffix of the second sort of name, such as
2762 chooses device names during auto-assembly or incremental assembly, it
2763 will sometimes add a small sequence number to the end of the name to
2764 avoid conflicted between multiple arrays that have the same name. If
2766 can reasonably determine that the array really is meant for this host,
2767 either by a hostname in the metadata, or by the presence of the array
2770 then it will leave off the suffix if possible.
2771 Also if the homehost is specified as
2774 will only use a suffix if a different array of the same name already
2775 exists or is listed in the config file.
2777 The standard names for non-partitioned arrays (the only sort of md
2778 array available in 2.4 and earlier) are of the form
2782 where NN is a number.
2783 The standard names for partitionable arrays (as available from 2.6
2784 onwards) are of the form
2788 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2790 From kernel version, 2.6.28 the "non-partitioned array" can actually
2791 be partitioned. So the "md_dNN" names are no longer needed, and
2792 partitions such as "/dev/mdNNpXX" are possible.
2796 was previously known as
2800 is completely separate from the
2802 package, and does not use the
2804 configuration file at all.
2807 For further information on mdadm usage, MD and the various levels of
2810 .B http://raid.wiki.kernel.org/
2812 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
2814 .\"for new releases of the RAID driver check out:
2817 .\".UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2818 .\"ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2823 .\".UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2824 .\"http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2827 The latest version of
2829 should always be available from
2831 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/