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 "1.2" (the commonly
340 preferred 1.x format).
341 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
343 Use the "Industry Standard" DDF (Disk Data Format) format defined by
345 When creating a DDF array a
347 will be created, and normal arrays can be created in that container.
349 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
351 which is managed in a similar manner to DDF, and is supported by an
352 option-rom on some platforms:
354 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
360 This will override any
362 setting in the config file and provides the identity of the host which
363 should be considered the home for any arrays.
365 When creating an array, the
367 will be recorded in the metadata. For version-1 superblocks, it will
368 be prefixed to the array name. For version-0.90 superblocks, part of
369 the SHA1 hash of the hostname will be stored in the later half of the
372 When reporting information about an array, any array which is tagged
373 for the given homehost will be reported as such.
375 When using Auto-Assemble, only arrays tagged for the given homehost
376 will be allowed to use 'local' names (i.e. not ending in '_' followed
377 by a digit string). See below under
378 .BR "Auto Assembly" .
380 .SH For create, build, or grow:
383 .BR \-n ", " \-\-raid\-devices=
384 Specify the number of active devices in the array. This, plus the
385 number of spare devices (see below) must equal the number of
387 (including "\fBmissing\fP" devices)
388 that are listed on the command line for
390 Setting a value of 1 is probably
391 a mistake and so requires that
393 be specified first. A value of 1 will then be allowed for linear,
394 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
396 This number can only be changed using
398 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
399 the necessary support.
402 .BR \-x ", " \-\-spare\-devices=
403 Specify the number of spare (eXtra) devices in the initial array.
404 Spares can also be added
405 and removed later. The number of component devices listed
406 on the command line must equal the number of RAID devices plus the
407 number of spare devices.
410 .BR \-z ", " \-\-size=
411 Amount (in Kibibytes) of space to use from each drive in RAID levels 1/4/5/6.
412 This must be a multiple of the chunk size, and must leave about 128Kb
413 of space at the end of the drive for the RAID superblock.
414 If this is not specified
415 (as it normally is not) the smallest drive (or partition) sets the
416 size, though if there is a variance among the drives of greater than 1%, a warning is
419 A suffix of 'M' or 'G' can be given to indicate Megabytes or
420 Gigabytes respectively.
422 Sometimes a replacement drive can be a little smaller than the
423 original drives though this should be minimised by IDEMA standards.
424 Such a replacement drive will be rejected by
426 To guard against this it can be useful to set the initial size
427 slightly smaller than the smaller device with the aim that it will
428 still be larger than any replacement.
430 This value can be set with
432 for RAID level 1/4/5/6 though
434 based arrays such as those with IMSM metadata may not be able to
436 If the array was created with a size smaller than the currently
437 active drives, the extra space can be accessed using
439 The size can be given as
441 which means to choose the largest size that fits on all current drives.
443 Before reducing the size of the array (with
444 .BR "\-\-grow \-\-size=" )
445 you should make sure that space isn't needed. If the device holds a
446 filesystem, you would need to resize the filesystem to use less space.
448 After reducing the array size you should check that the data stored in
449 the device is still available. If the device holds a filesystem, then
450 an 'fsck' of the filesystem is a minimum requirement. If there are
451 problems the array can be made bigger again with no loss with another
452 .B "\-\-grow \-\-size="
455 This value cannot be used when creating a
457 such as with DDF and IMSM metadata, though it perfectly valid when
458 creating an array inside a container.
461 .BR \-Z ", " \-\-array\-size=
462 This is only meaningful with
464 and its effect is not persistent: when the array is stopped and
465 restarted the default array size will be restored.
467 Setting the array-size causes the array to appear smaller to programs
468 that access the data. This is particularly needed before reshaping an
469 array so that it will be smaller. As the reshape is not reversible,
470 but setting the size with
472 is, it is required that the array size is reduced as appropriate
473 before the number of devices in the array is reduced.
475 Before reducing the size of the array you should make sure that space
476 isn't needed. If the device holds a filesystem, you would need to
477 resize the filesystem to use less space.
479 After reducing the array size you should check that the data stored in
480 the device is still available. If the device holds a filesystem, then
481 an 'fsck' of the filesystem is a minimum requirement. If there are
482 problems the array can be made bigger again with no loss with another
483 .B "\-\-grow \-\-array\-size="
486 A suffix of 'M' or 'G' can be given to indicate Megabytes or
487 Gigabytes respectively.
490 restores the apparent size of the array to be whatever the real
491 amount of available space is.
494 .BR \-c ", " \-\-chunk=
495 Specify chunk size of kibibytes. The default when creating an
496 array is 512KB. To ensure compatibility with earlier versions, the
497 default when Building and array with no persistent metadata is 64KB.
498 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
500 RAID4, RAID5, RAID6, and RAID10 require the chunk size to be a power
501 of 2. In any case it must be a multiple of 4KB.
503 A suffix of 'M' or 'G' can be given to indicate Megabytes or
504 Gigabytes respectively.
508 Specify rounding factor for a Linear array. The size of each
509 component will be rounded down to a multiple of this size.
510 This is a synonym for
512 but highlights the different meaning for Linear as compared to other
513 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
514 use, and is 0K (i.e. no rounding) in later kernels.
517 .BR \-l ", " \-\-level=
518 Set RAID level. When used with
520 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
521 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
522 Obviously some of these are synonymous.
526 metadata type is requested, only the
528 level is permitted, and it does not need to be explicitly given.
532 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
536 to change the RAID level in some cases. See LEVEL CHANGES below.
539 .BR \-p ", " \-\-layout=
540 This option configures the fine details of data layout for RAID5, RAID6,
541 and RAID10 arrays, and controls the failure modes for
544 The layout of the RAID5 parity block can be one of
545 .BR left\-asymmetric ,
546 .BR left\-symmetric ,
547 .BR right\-asymmetric ,
548 .BR right\-symmetric ,
549 .BR la ", " ra ", " ls ", " rs .
551 .BR left\-symmetric .
553 It is also possible to cause RAID5 to use a RAID4-like layout by
559 Finally for RAID5 there are DDF\-compatible layouts,
560 .BR ddf\-zero\-restart ,
561 .BR ddf\-N\-restart ,
563 .BR ddf\-N\-continue .
565 These same layouts are available for RAID6. There are also 4 layouts
566 that will provide an intermediate stage for converting between RAID5
567 and RAID6. These provide a layout which is identical to the
568 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
569 syndrome (the second 'parity' block used by RAID6) on the last device.
571 .BR left\-symmetric\-6 ,
572 .BR right\-symmetric\-6 ,
573 .BR left\-asymmetric\-6 ,
574 .BR right\-asymmetric\-6 ,
576 .BR parity\-first\-6 .
578 When setting the failure mode for level
581 .BR write\-transient ", " wt ,
582 .BR read\-transient ", " rt ,
583 .BR write\-persistent ", " wp ,
584 .BR read\-persistent ", " rp ,
586 .BR read\-fixable ", " rf ,
587 .BR clear ", " flush ", " none .
589 Each failure mode can be followed by a number, which is used as a period
590 between fault generation. Without a number, the fault is generated
591 once on the first relevant request. With a number, the fault will be
592 generated after that many requests, and will continue to be generated
593 every time the period elapses.
595 Multiple failure modes can be current simultaneously by using the
597 option to set subsequent failure modes.
599 "clear" or "none" will remove any pending or periodic failure modes,
600 and "flush" will clear any persistent faults.
602 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
603 by a small number. The default is 'n2'. The supported options are:
606 signals 'near' copies. Multiple copies of one data block are at
607 similar offsets in different devices.
610 signals 'offset' copies. Rather than the chunks being duplicated
611 within a stripe, whole stripes are duplicated but are rotated by one
612 device so duplicate blocks are on different devices. Thus subsequent
613 copies of a block are in the next drive, and are one chunk further
618 (multiple copies have very different offsets).
619 See md(4) for more detail about 'near', 'offset', and 'far'.
621 The number is the number of copies of each datablock. 2 is normal, 3
622 can be useful. This number can be at most equal to the number of
623 devices in the array. It does not need to divide evenly into that
624 number (e.g. it is perfectly legal to have an 'n2' layout for an array
625 with an odd number of devices).
627 When an array is converted between RAID5 and RAID6 an intermediate
628 RAID6 layout is used in which the second parity block (Q) is always on
629 the last device. To convert a RAID5 to RAID6 and leave it in this new
630 layout (which does not require re-striping) use
631 .BR \-\-layout=preserve .
632 This will try to avoid any restriping.
634 The converse of this is
635 .B \-\-layout=normalise
636 which will change a non-standard RAID6 layout into a more standard
643 (thus explaining the p of
647 .BR \-b ", " \-\-bitmap=
648 Specify a file to store a write-intent bitmap in. The file should not
651 is also given. The same file should be provided
652 when assembling the array. If the word
654 is given, then the bitmap is stored with the metadata on the array,
655 and so is replicated on all devices. If the word
659 mode, then any bitmap that is present is removed.
661 To help catch typing errors, the filename must contain at least one
662 slash ('/') if it is a real file (not 'internal' or 'none').
664 Note: external bitmaps are only known to work on ext2 and ext3.
665 Storing bitmap files on other filesystems may result in serious problems.
668 .BR \-\-bitmap\-chunk=
669 Set the chunksize of the bitmap. Each bit corresponds to that many
670 Kilobytes of storage.
671 When using a file based bitmap, the default is to use the smallest
672 size that is at-least 4 and requires no more than 2^21 chunks.
675 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
676 fit the bitmap into the available space.
678 A suffix of 'M' or 'G' can be given to indicate Megabytes or
679 Gigabytes respectively.
682 .BR \-W ", " \-\-write\-mostly
683 subsequent devices listed in a
688 command will be flagged as 'write-mostly'. This is valid for RAID1
689 only and means that the 'md' driver will avoid reading from these
690 devices if at all possible. This can be useful if mirroring over a
694 .BR \-\-write\-behind=
695 Specify that write-behind mode should be enabled (valid for RAID1
696 only). If an argument is specified, it will set the maximum number
697 of outstanding writes allowed. The default value is 256.
698 A write-intent bitmap is required in order to use write-behind
699 mode, and write-behind is only attempted on drives marked as
703 .BR \-\-assume\-clean
706 that the array pre-existed and is known to be clean. It can be useful
707 when trying to recover from a major failure as you can be sure that no
708 data will be affected unless you actually write to the array. It can
709 also be used when creating a RAID1 or RAID10 if you want to avoid the
710 initial resync, however this practice \(em while normally safe \(em is not
711 recommended. Use this only if you really know what you are doing.
713 When the devices that will be part of a new array were filled
714 with zeros before creation the operator knows the array is
715 actually clean. If that is the case, such as after running
716 badblocks, this argument can be used to tell mdadm the
717 facts the operator knows.
719 When an array is resized to a larger size with
720 .B "\-\-grow \-\-size="
721 the new space is normally resynced in that same way that the whole
722 array is resynced at creation. From Linux version 3.0,
724 can be used with that command to avoid the automatic resync.
727 .BR \-\-backup\-file=
730 is used to increase the number of raid-devices in a RAID5 or RAID6 if
731 there are no spare devices available, or to shrink, change RAID level
732 or layout. See the GROW MODE section below on RAID\-DEVICES CHANGES.
733 The file must be stored on a separate device, not on the RAID array
738 This option is complementary to the
739 .B \-\-freeze-reshape
740 option for assembly. It is needed when
742 operation is interrupted and it is not restarted automatically due to
743 .B \-\-freeze-reshape
744 usage during array assembly. This option is used together with
748 ) command and device for a pending reshape to be continued.
749 All parameters required for reshape continuation will be read from array metadata.
753 .BR \-\-backup\-file=
754 option to be set, continuation option will require to have exactly the same
755 backup file given as well.
757 Any other parameter passed together with
759 option will be ignored.
762 .BR \-N ", " \-\-name=
765 for the array. This is currently only effective when creating an
766 array with a version-1 superblock, or an array in a DDF container.
767 The name is a simple textual string that can be used to identify array
768 components when assembling. If name is needed but not specified, it
769 is taken from the basename of the device that is being created.
781 run the array, even if some of the components
782 appear to be active in another array or filesystem. Normally
784 will ask for confirmation before including such components in an
785 array. This option causes that question to be suppressed.
788 .BR \-f ", " \-\-force
791 accept the geometry and layout specified without question. Normally
793 will not allow creation of an array with only one device, and will try
794 to create a RAID5 array with one missing drive (as this makes the
795 initial resync work faster). With
798 will not try to be so clever.
801 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
802 Instruct mdadm how to create the device file if needed, possibly allocating
803 an unused minor number. "md" causes a non-partitionable array
804 to be used (though since Linux 2.6.28, these array devices are in fact
805 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
806 later) to be used. "yes" requires the named md device to have
807 a 'standard' format, and the type and minor number will be determined
808 from this. With mdadm 3.0, device creation is normally left up to
810 so this option is unlikely to be needed.
811 See DEVICE NAMES below.
813 The argument can also come immediately after
818 is not given on the command line or in the config file, then
824 is also given, then any
826 entries in the config file will override the
828 instruction given on the command line.
830 For partitionable arrays,
832 will create the device file for the whole array and for the first 4
833 partitions. A different number of partitions can be specified at the
834 end of this option (e.g.
836 If the device name ends with a digit, the partition names add a 'p',
838 .IR /dev/md/home1p3 .
839 If there is no trailing digit, then the partition names just have a
841 .IR /dev/md/scratch3 .
843 If the md device name is in a 'standard' format as described in DEVICE
844 NAMES, then it will be created, if necessary, with the appropriate
845 device number based on that name. If the device name is not in one of these
846 formats, then a unused device number will be allocated. The device
847 number will be considered unused if there is no active array for that
848 number, and there is no entry in /dev for that number and with a
849 non-standard name. Names that are not in 'standard' format are only
850 allowed in "/dev/md/".
852 This is meaningful with
859 .\".BR \-\-symlink = no
864 .\"to create devices in
866 .\"it will also create symlinks from
868 .\"with names starting with
874 .\"to suppress this, or
875 .\".B \-\-symlink=yes
876 .\"to enforce this even if it is suppressing
882 .BR \-a ", " "\-\-add"
883 This option can be used in Grow mode in two cases.
885 If the target array is a Linear array, then
887 can be used to add one or more devices to the array. They
888 are simply catenated on to the end of the array. Once added, the
889 devices cannot be removed.
893 option is being used to increase the number of devices in an array,
896 can be used to add some extra devices to be included in the array.
897 In most cases this is not needed as the extra devices can be added as
898 spares first, and then the number of raid-disks can be changed.
899 However for RAID0, it is not possible to add spares. So to increase
900 the number of devices in a RAID0, it is necessary to set the new
901 number of devices, and to add the new devices, in the same command.
906 .BR \-u ", " \-\-uuid=
907 uuid of array to assemble. Devices which don't have this uuid are
911 .BR \-m ", " \-\-super\-minor=
912 Minor number of device that array was created for. Devices which
913 don't have this minor number are excluded. If you create an array as
914 /dev/md1, then all superblocks will contain the minor number 1, even if
915 the array is later assembled as /dev/md2.
917 Giving the literal word "dev" for
921 to use the minor number of the md device that is being assembled.
924 .B \-\-super\-minor=dev
925 will look for super blocks with a minor number of 0.
928 is only relevant for v0.90 metadata, and should not normally be used.
934 .BR \-N ", " \-\-name=
935 Specify the name of the array to assemble. This must be the name
936 that was specified when creating the array. It must either match
937 the name stored in the superblock exactly, or it must match
940 prefixed to the start of the given name.
943 .BR \-f ", " \-\-force
944 Assemble the array even if the metadata on some devices appears to be
947 cannot find enough working devices to start the array, but can find
948 some devices that are recorded as having failed, then it will mark
949 those devices as working so that the array can be started.
950 An array which requires
952 to be started may contain data corruption. Use it carefully.
956 Attempt to start the array even if fewer drives were given than were
957 present last time the array was active. Normally if not all the
958 expected drives are found and
960 is not used, then the array will be assembled but not started.
963 an attempt will be made to start it anyway.
967 This is the reverse of
969 in that it inhibits the startup of array unless all expected drives
970 are present. This is only needed with
972 and can be used if the physical connections to devices are
973 not as reliable as you would like.
976 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
977 See this option under Create and Build options.
980 .BR \-b ", " \-\-bitmap=
981 Specify the bitmap file that was given when the array was created. If
984 bitmap, there is no need to specify this when assembling the array.
987 .BR \-\-backup\-file=
990 was used while reshaping an array (e.g. changing number of devices or
991 chunk size) and the system crashed during the critical section, then the same
995 to allow possibly corrupted data to be restored, and the reshape
999 .BR \-\-invalid\-backup
1000 If the file needed for the above option is not available for any
1001 reason an empty file can be given together with this option to
1002 indicate that the backup file is invalid. In this case the data that
1003 was being rearranged at the time of the crash could be irrecoverably
1004 lost, but the rest of the array may still be recoverable. This option
1005 should only be used as a last resort if there is no way to recover the
1010 .BR \-U ", " \-\-update=
1011 Update the superblock on each device while assembling the array. The
1012 argument given to this flag can be one of
1027 option will adjust the superblock of an array what was created on a Sparc
1028 machine running a patched 2.2 Linux kernel. This kernel got the
1029 alignment of part of the superblock wrong. You can use the
1030 .B "\-\-examine \-\-sparc2.2"
1033 to see what effect this would have.
1037 option will update the
1038 .B "preferred minor"
1039 field on each superblock to match the minor number of the array being
1041 This can be useful if
1043 reports a different "Preferred Minor" to
1045 In some cases this update will be performed automatically
1046 by the kernel driver. In particular the update happens automatically
1047 at the first write to an array with redundancy (RAID level 1 or
1048 greater) on a 2.6 (or later) kernel.
1052 option will change the uuid of the array. If a UUID is given with the
1054 option that UUID will be used as a new UUID and will
1056 be used to help identify the devices in the array.
1059 is given, a random UUID is chosen.
1063 option will change the
1065 of the array as stored in the superblock. This is only supported for
1066 version-1 superblocks.
1070 option will change the
1072 as recorded in the superblock. For version-0 superblocks, this is the
1073 same as updating the UUID.
1074 For version-1 superblocks, this involves updating the name.
1078 option will cause the array to be marked
1080 meaning that any redundancy in the array (e.g. parity for RAID5,
1081 copies for RAID1) may be incorrect. This will cause the RAID system
1082 to perform a "resync" pass to make sure that all redundant information
1087 option allows arrays to be moved between machines with different
1089 When assembling such an array for the first time after a move, giving
1090 .B "\-\-update=byteorder"
1093 to expect superblocks to have their byteorder reversed, and will
1094 correct that order before assembling the array. This is only valid
1095 with original (Version 0.90) superblocks.
1099 option will correct the summaries in the superblock. That is the
1100 counts of total, working, active, failed, and spare devices.
1104 option will rarely be of use. It applies to version 1.1 and 1.2 metadata
1105 only (where the metadata is at the start of the device) and is only
1106 useful when the component device has changed size (typically become
1107 larger). The version 1 metadata records the amount of the device that
1108 can be used to store data, so if a device in a version 1.1 or 1.2
1109 array becomes larger, the metadata will still be visible, but the
1110 extra space will not. In this case it might be useful to assemble the
1112 .BR \-\-update=devicesize .
1115 to determine the maximum usable amount of space on each device and
1116 update the relevant field in the metadata.
1120 option can be used when an array has an internal bitmap which is
1121 corrupt in some way so that assembling the array normally fails. It
1122 will cause any internal bitmap to be ignored.
1125 .BR \-\-freeze\-reshape
1126 Option is intended to be used in start-up scripts during initrd boot phase.
1127 When array under reshape is assembled during initrd phase, this option
1128 stops reshape after reshape critical section is being restored. This happens
1129 before file system pivot operation and avoids loss of file system context.
1130 Losing file system context would cause reshape to be broken.
1132 Reshape can be continued later using the
1134 option for the grow command.
1136 .SH For Manage mode:
1139 .BR \-t ", " \-\-test
1140 Unless a more serious error occurred,
1142 will exit with a status of 2 if no changes were made to the array and
1143 0 if at least one change was made.
1144 This can be useful when an indirect specifier such as
1149 is used in requesting an operation on the array.
1151 will report failure if these specifiers didn't find any match.
1154 .BR \-a ", " \-\-add
1155 hot-add listed devices.
1156 If a device appears to have recently been part of the array
1157 (possibly it failed or was removed) the device is re\-added as describe
1159 If that fails or the device was never part of the array, the device is
1160 added as a hot-spare.
1161 If the array is degraded, it will immediately start to rebuild data
1164 Note that this and the following options are only meaningful on array
1165 with redundancy. They don't apply to RAID0 or Linear.
1169 re\-add a device that was previous removed from an array.
1170 If the metadata on the device reports that it is a member of the
1171 array, and the slot that it used is still vacant, then the device will
1172 be added back to the array in the same position. This will normally
1173 cause the data for that device to be recovered. However based on the
1174 event count on the device, the recovery may only require sections that
1175 are flagged a write-intent bitmap to be recovered or may not require
1176 any recovery at all.
1178 When used on an array that has no metadata (i.e. it was built with
1180 it will be assumed that bitmap-based recovery is enough to make the
1181 device fully consistent with the array.
1185 can be accompanied by
1186 .BR \-\-update=devicesize .
1187 See the description of this option when used in Assemble mode for an
1188 explanation of its use.
1190 If the device name given is
1192 then mdadm will try to find any device that looks like it should be
1193 part of the array but isn't and will try to re\-add all such devices.
1196 .BR \-r ", " \-\-remove
1197 remove listed devices. They must not be active. i.e. they should
1198 be failed or spare devices. As well as the name of a device file
1207 The first causes all failed device to be removed. The second causes
1208 any device which is no longer connected to the system (i.e an 'open'
1211 to be removed. This will only succeed for devices that are spares or
1212 have already been marked as failed.
1215 .BR \-f ", " \-\-fail
1216 mark listed devices as faulty.
1217 As well as the name of a device file, the word
1219 can be given. This will cause any device that has been detached from
1220 the system to be marked as failed. It can then be removed.
1228 .BR \-\-write\-mostly
1229 Subsequent devices that are added or re\-added will have the 'write-mostly'
1230 flag set. This is only valid for RAID1 and means that the 'md' driver
1231 will avoid reading from these devices if possible.
1234 Subsequent devices that are added or re\-added will have the 'write-mostly'
1238 Each of these options requires that the first device listed is the array
1239 to be acted upon, and the remainder are component devices to be added,
1240 removed, marked as faulty, etc. Several different operations can be
1241 specified for different devices, e.g.
1243 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1245 Each operation applies to all devices listed until the next
1248 If an array is using a write-intent bitmap, then devices which have
1249 been removed can be re\-added in a way that avoids a full
1250 reconstruction but instead just updates the blocks that have changed
1251 since the device was removed. For arrays with persistent metadata
1252 (superblocks) this is done automatically. For arrays created with
1254 mdadm needs to be told that this device we removed recently with
1257 Devices can only be removed from an array if they are not in active
1258 use, i.e. that must be spares or failed devices. To remove an active
1259 device, it must first be marked as
1265 .BR \-Q ", " \-\-query
1266 Examine a device to see
1267 (1) if it is an md device and (2) if it is a component of an md
1269 Information about what is discovered is presented.
1272 .BR \-D ", " \-\-detail
1273 Print details of one or more md devices.
1276 .BR \-\-detail\-platform
1277 Print details of the platform's RAID capabilities (firmware / hardware
1278 topology) for a given metadata format.
1281 .BR \-Y ", " \-\-export
1286 output will be formatted as
1288 pairs for easy import into the environment.
1291 .BR \-E ", " \-\-examine
1292 Print contents of the metadata stored on the named device(s).
1293 Note the contrast between
1298 applies to devices which are components of an array, while
1300 applies to a whole array which is currently active.
1303 If an array was created on a SPARC machine with a 2.2 Linux kernel
1304 patched with RAID support, the superblock will have been created
1305 incorrectly, or at least incompatibly with 2.4 and later kernels.
1310 will fix the superblock before displaying it. If this appears to do
1311 the right thing, then the array can be successfully assembled using
1312 .BR "\-\-assemble \-\-update=sparc2.2" .
1315 .BR \-X ", " \-\-examine\-bitmap
1316 Report information about a bitmap file.
1317 The argument is either an external bitmap file or an array component
1318 in case of an internal bitmap. Note that running this on an array
1321 does not report the bitmap for that array.
1324 .BR \-R ", " \-\-run
1325 start a partially assembled array. If
1327 did not find enough devices to fully start the array, it might leaving
1328 it partially assembled. If you wish, you can then use
1330 to start the array in degraded mode.
1333 .BR \-S ", " \-\-stop
1334 deactivate array, releasing all resources.
1337 .BR \-o ", " \-\-readonly
1338 mark array as readonly.
1341 .BR \-w ", " \-\-readwrite
1342 mark array as readwrite.
1345 .B \-\-zero\-superblock
1346 If the device contains a valid md superblock, the block is
1347 overwritten with zeros. With
1349 the block where the superblock would be is overwritten even if it
1350 doesn't appear to be valid.
1353 .B \-\-kill\-subarray=
1354 If the device is a container and the argument to \-\-kill\-subarray
1355 specifies an inactive subarray in the container, then the subarray is
1356 deleted. Deleting all subarrays will leave an 'empty-container' or
1357 spare superblock on the drives. See \-\-zero\-superblock for completely
1358 removing a superblock. Note that some formats depend on the subarray
1359 index for generating a UUID, this command will fail if it would change
1360 the UUID of an active subarray.
1363 .B \-\-update\-subarray=
1364 If the device is a container and the argument to \-\-update\-subarray
1365 specifies a subarray in the container, then attempt to update the given
1366 superblock field in the subarray. See below in
1371 .BR \-t ", " \-\-test
1376 is set to reflect the status of the device. See below in
1381 .BR \-W ", " \-\-wait
1382 For each md device given, wait for any resync, recovery, or reshape
1383 activity to finish before returning.
1385 will return with success if it actually waited for every device
1386 listed, otherwise it will return failure.
1390 For each md device given, or each device in /proc/mdstat if
1392 is given, arrange for the array to be marked clean as soon as possible.
1394 will return with success if the array uses external metadata and we
1395 successfully waited. For native arrays this returns immediately as the
1396 kernel handles dirty-clean transitions at shutdown. No action is taken
1397 if safe-mode handling is disabled.
1399 .SH For Incremental Assembly mode:
1401 .BR \-\-rebuild\-map ", " \-r
1402 Rebuild the map file
1403 .RB ( /var/run/mdadm/map )
1406 uses to help track which arrays are currently being assembled.
1409 .BR \-\-run ", " \-R
1410 Run any array assembled as soon as a minimal number of devices are
1411 available, rather than waiting until all expected devices are present.
1414 .BR \-\-scan ", " \-s
1415 Only meaningful with
1419 file for arrays that are being incrementally assembled and will try to
1420 start any that are not already started. If any such array is listed
1423 as requiring an external bitmap, that bitmap will be attached first.
1426 .BR \-\-fail ", " \-f
1427 This allows the hot-plug system to remove devices that have fully disappeared
1428 from the kernel. It will first fail and then remove the device from any
1429 array it belongs to.
1430 The device name given should be a kernel device name such as "sda",
1436 Only used with \-\-fail. The 'path' given will be recorded so that if
1437 a new device appears at the same location it can be automatically
1438 added to the same array. This allows the failed device to be
1439 automatically replaced by a new device without metadata if it appears
1440 at specified path. This option is normally only set by a
1444 .SH For Monitor mode:
1446 .BR \-m ", " \-\-mail
1447 Give a mail address to send alerts to.
1450 .BR \-p ", " \-\-program ", " \-\-alert
1451 Give a program to be run whenever an event is detected.
1454 .BR \-y ", " \-\-syslog
1455 Cause all events to be reported through 'syslog'. The messages have
1456 facility of 'daemon' and varying priorities.
1459 .BR \-d ", " \-\-delay
1460 Give a delay in seconds.
1462 polls the md arrays and then waits this many seconds before polling
1463 again. The default is 60 seconds. Since 2.6.16, there is no need to
1464 reduce this as the kernel alerts
1466 immediately when there is any change.
1469 .BR \-r ", " \-\-increment
1470 Give a percentage increment.
1472 will generate RebuildNN events with the given percentage increment.
1475 .BR \-f ", " \-\-daemonise
1478 to run as a background daemon if it decides to monitor anything. This
1479 causes it to fork and run in the child, and to disconnect from the
1480 terminal. The process id of the child is written to stdout.
1483 which will only continue monitoring if a mail address or alert program
1484 is found in the config file.
1487 .BR \-i ", " \-\-pid\-file
1490 is running in daemon mode, write the pid of the daemon process to
1491 the specified file, instead of printing it on standard output.
1494 .BR \-1 ", " \-\-oneshot
1495 Check arrays only once. This will generate
1497 events and more significantly
1503 .B " mdadm \-\-monitor \-\-scan \-1"
1505 from a cron script will ensure regular notification of any degraded arrays.
1508 .BR \-t ", " \-\-test
1511 alert for every array found at startup. This alert gets mailed and
1512 passed to the alert program. This can be used for testing that alert
1513 message do get through successfully.
1517 This inhibits the functionality for moving spares between arrays.
1518 Only one monitoring process started with
1520 but without this flag is allowed, otherwise the two could interfere
1527 .B mdadm \-\-assemble
1528 .I md-device options-and-component-devices...
1531 .B mdadm \-\-assemble \-\-scan
1532 .I md-devices-and-options...
1535 .B mdadm \-\-assemble \-\-scan
1539 This usage assembles one or more RAID arrays from pre-existing components.
1540 For each array, mdadm needs to know the md device, the identity of the
1541 array, and a number of component-devices. These can be found in a number of ways.
1543 In the first usage example (without the
1545 the first device given is the md device.
1546 In the second usage example, all devices listed are treated as md
1547 devices and assembly is attempted.
1548 In the third (where no devices are listed) all md devices that are
1549 listed in the configuration file are assembled. If no arrays are
1550 described by the configuration file, then any arrays that
1551 can be found on unused devices will be assembled.
1553 If precisely one device is listed, but
1559 was given and identity information is extracted from the configuration file.
1561 The identity can be given with the
1567 option, will be taken from the md-device record in the config file, or
1568 will be taken from the super block of the first component-device
1569 listed on the command line.
1571 Devices can be given on the
1573 command line or in the config file. Only devices which have an md
1574 superblock which contains the right identity will be considered for
1577 The config file is only used if explicitly named with
1579 or requested with (a possibly implicit)
1584 .B /etc/mdadm/mdadm.conf
1589 is not given, then the config file will only be used to find the
1590 identity of md arrays.
1592 Normally the array will be started after it is assembled. However if
1594 is not given and not all expected drives were listed, then the array
1595 is not started (to guard against usage errors). To insist that the
1596 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1605 does not create any entries in
1609 It does record information in
1610 .B /var/run/mdadm/map
1613 to choose the correct name.
1617 detects that udev is not configured, it will create the devices in
1621 In Linux kernels prior to version 2.6.28 there were two distinctly
1622 different types of md devices that could be created: one that could be
1623 partitioned using standard partitioning tools and one that could not.
1624 Since 2.6.28 that distinction is no longer relevant as both type of
1625 devices can be partitioned.
1627 will normally create the type that originally could not be partitioned
1628 as it has a well defined major number (9).
1630 Prior to 2.6.28, it is important that mdadm chooses the correct type
1631 of array device to use. This can be controlled with the
1633 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1634 to use a partitionable device rather than the default.
1636 In the no-udev case, the value given to
1638 can be suffixed by a number. This tells
1640 to create that number of partition devices rather than the default of 4.
1644 can also be given in the configuration file as a word starting
1646 on the ARRAY line for the relevant array.
1653 and no devices are listed,
1655 will first attempt to assemble all the arrays listed in the config
1658 If no arrays are listed in the config (other than those marked
1660 it will look through the available devices for possible arrays and
1661 will try to assemble anything that it finds. Arrays which are tagged
1662 as belonging to the given homehost will be assembled and started
1663 normally. Arrays which do not obviously belong to this host are given
1664 names that are expected not to conflict with anything local, and are
1665 started "read-auto" so that nothing is written to any device until the
1666 array is written to. i.e. automatic resync etc is delayed.
1670 finds a consistent set of devices that look like they should comprise
1671 an array, and if the superblock is tagged as belonging to the given
1672 home host, it will automatically choose a device name and try to
1673 assemble the array. If the array uses version-0.90 metadata, then the
1675 number as recorded in the superblock is used to create a name in
1679 If the array uses version-1 metadata, then the
1681 from the superblock is used to similarly create a name in
1683 (the name will have any 'host' prefix stripped first).
1685 This behaviour can be modified by the
1689 configuration file. This line can indicate that specific metadata
1690 type should, or should not, be automatically assembled. If an array
1691 is found which is not listed in
1693 and has a metadata format that is denied by the
1695 line, then it will not be assembled.
1698 line can also request that all arrays identified as being for this
1699 homehost should be assembled regardless of their metadata type.
1702 for further details.
1704 Note: Auto assembly cannot be used for assembling and activating some
1705 arrays which are undergoing reshape. In particular as the
1707 cannot be given, any reshape which requires a backup-file to continue
1708 cannot be started by auto assembly. An array which is growing to more
1709 devices and has passed the critical section can be assembled using
1720 .BI \-\-raid\-devices= Z
1724 This usage is similar to
1726 The difference is that it creates an array without a superblock. With
1727 these arrays there is no difference between initially creating the array and
1728 subsequently assembling the array, except that hopefully there is useful
1729 data there in the second case.
1731 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1732 one of their synonyms. All devices must be listed and the array will
1733 be started once complete. It will often be appropriate to use
1734 .B \-\-assume\-clean
1735 with levels raid1 or raid10.
1746 .BI \-\-raid\-devices= Z
1750 This usage will initialise a new md array, associate some devices with
1751 it, and activate the array.
1753 The named device will normally not exist when
1754 .I "mdadm \-\-create"
1755 is run, but will be created by
1757 once the array becomes active.
1759 As devices are added, they are checked to see if they contain RAID
1760 superblocks or filesystems. They are also checked to see if the variance in
1761 device size exceeds 1%.
1763 If any discrepancy is found, the array will not automatically be run, though
1766 can override this caution.
1768 To create a "degraded" array in which some devices are missing, simply
1769 give the word "\fBmissing\fP"
1770 in place of a device name. This will cause
1772 to leave the corresponding slot in the array empty.
1773 For a RAID4 or RAID5 array at most one slot can be
1774 "\fBmissing\fP"; for a RAID6 array at most two slots.
1775 For a RAID1 array, only one real device needs to be given. All of the
1779 When creating a RAID5 array,
1781 will automatically create a degraded array with an extra spare drive.
1782 This is because building the spare into a degraded array is in general
1783 faster than resyncing the parity on a non-degraded, but not clean,
1784 array. This feature can be overridden with the
1788 When creating an array with version-1 metadata a name for the array is
1790 If this is not given with the
1794 will choose a name based on the last component of the name of the
1795 device being created. So if
1797 is being created, then the name
1802 is being created, then the name
1806 When creating a partition based array, using
1808 with version-1.x metadata, the partition type should be set to
1810 (non fs-data). This type selection allows for greater precision since
1811 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
1812 might create problems in the event of array recovery through a live cdrom.
1814 A new array will normally get a randomly assigned 128bit UUID which is
1815 very likely to be unique. If you have a specific need, you can choose
1816 a UUID for the array by giving the
1818 option. Be warned that creating two arrays with the same UUID is a
1819 recipe for disaster. Also, using
1821 when creating a v0.90 array will silently override any
1826 .\"option is given, it is not necessary to list any component-devices in this command.
1827 .\"They can be added later, before a
1831 .\"is given, the apparent size of the smallest drive given is used.
1833 When creating an array within a
1836 can be given either the list of devices to use, or simply the name of
1837 the container. The former case gives control over which devices in
1838 the container will be used for the array. The latter case allows
1840 to automatically choose which devices to use based on how much spare
1843 The General Management options that are valid with
1848 insist on running the array even if some devices look like they might
1853 start the array readonly \(em not supported yet.
1860 .I options... devices...
1863 This usage will allow individual devices in an array to be failed,
1864 removed or added. It is possible to perform multiple operations with
1865 on command. For example:
1867 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
1873 and will then remove it from the array and finally add it back
1874 in as a spare. However only one md array can be affected by a single
1877 When a device is added to an active array, mdadm checks to see if it
1878 has metadata on it which suggests that it was recently a member of the
1879 array. If it does, it tries to "re\-add" the device. If there have
1880 been no changes since the device was removed, or if the array has a
1881 write-intent bitmap which has recorded whatever changes there were,
1882 then the device will immediately become a full member of the array and
1883 those differences recorded in the bitmap will be resolved.
1893 MISC mode includes a number of distinct operations that
1894 operate on distinct devices. The operations are:
1897 The device is examined to see if it is
1898 (1) an active md array, or
1899 (2) a component of an md array.
1900 The information discovered is reported.
1904 The device should be an active md device.
1906 will display a detailed description of the array.
1910 will cause the output to be less detailed and the format to be
1911 suitable for inclusion in
1915 will normally be 0 unless
1917 failed to get useful information about the device(s); however, if the
1919 option is given, then the exit status will be:
1923 The array is functioning normally.
1926 The array has at least one failed device.
1929 The array has multiple failed devices such that it is unusable.
1932 There was an error while trying to get information about the device.
1936 .B \-\-detail\-platform
1937 Print detail of the platform's RAID capabilities (firmware / hardware
1938 topology). If the metadata is specified with
1942 then the return status will be:
1946 metadata successfully enumerated its platform components on this system
1949 metadata is platform independent
1952 metadata failed to find its platform components on this system
1956 .B \-\-update\-subarray=
1957 If the device is a container and the argument to \-\-update\-subarray
1958 specifies a subarray in the container, then attempt to update the given
1959 superblock field in the subarray. Similar to updating an array in
1960 "assemble" mode, the field to update is selected by
1964 option. Currently only
1970 option updates the subarray name in the metadata, it may not affect the
1971 device node name or the device node symlink until the subarray is
1972 re\-assembled. If updating
1974 would change the UUID of an active subarray this operation is blocked,
1975 and the command will end in an error.
1979 The device should be a component of an md array.
1981 will read the md superblock of the device and display the contents.
1986 is given, then multiple devices that are components of the one array
1987 are grouped together and reported in a single entry suitable
1993 without listing any devices will cause all devices listed in the
1994 config file to be examined.
1998 The devices should be active md arrays which will be deactivated, as
1999 long as they are not currently in use.
2003 This will fully activate a partially assembled md array.
2007 This will mark an active array as read-only, providing that it is
2008 not currently being used.
2014 array back to being read/write.
2018 For all operations except
2021 will cause the operation to be applied to all arrays listed in
2026 causes all devices listed in the config file to be examined.
2029 .BR \-b ", " \-\-brief
2030 Be less verbose. This is used with
2038 gives an intermediate level of verbosity.
2044 .B mdadm \-\-monitor
2045 .I options... devices...
2050 to periodically poll a number of md arrays and to report on any events
2053 will never exit once it decides that there are arrays to be checked,
2054 so it should normally be run in the background.
2056 As well as reporting events,
2058 may move a spare drive from one array to another if they are in the
2063 and if the destination array has a failed drive but no spares.
2065 If any devices are listed on the command line,
2067 will only monitor those devices. Otherwise all arrays listed in the
2068 configuration file will be monitored. Further, if
2070 is given, then any other md devices that appear in
2072 will also be monitored.
2074 The result of monitoring the arrays is the generation of events.
2075 These events are passed to a separate program (if specified) and may
2076 be mailed to a given E-mail address.
2078 When passing events to a program, the program is run once for each event,
2079 and is given 2 or 3 command-line arguments: the first is the
2080 name of the event (see below), the second is the name of the
2081 md device which is affected, and the third is the name of a related
2082 device if relevant (such as a component device that has failed).
2086 is given, then a program or an E-mail address must be specified on the
2087 command line or in the config file. If neither are available, then
2089 will not monitor anything.
2093 will continue monitoring as long as something was found to monitor. If
2094 no program or email is given, then each event is reported to
2097 The different events are:
2101 .B DeviceDisappeared
2102 An md array which previously was configured appears to no longer be
2103 configured. (syslog priority: Critical)
2107 was told to monitor an array which is RAID0 or Linear, then it will
2109 .B DeviceDisappeared
2110 with the extra information
2112 This is because RAID0 and Linear do not support the device-failed,
2113 hot-spare and resync operations which are monitored.
2117 An md array started reconstruction. (syslog priority: Warning)
2123 is a two-digit number (ie. 05, 48). This indicates that rebuild
2124 has passed that many percent of the total. The events are generated
2125 with fixed increment since 0. Increment size may be specified with
2126 a commandline option (default is 20). (syslog priority: Warning)
2130 An md array that was rebuilding, isn't any more, either because it
2131 finished normally or was aborted. (syslog priority: Warning)
2135 An active component device of an array has been marked as
2136 faulty. (syslog priority: Critical)
2140 A spare component device which was being rebuilt to replace a faulty
2141 device has failed. (syslog priority: Critical)
2145 A spare component device which was being rebuilt to replace a faulty
2146 device has been successfully rebuilt and has been made active.
2147 (syslog priority: Info)
2151 A new md array has been detected in the
2153 file. (syslog priority: Info)
2157 A newly noticed array appears to be degraded. This message is not
2160 notices a drive failure which causes degradation, but only when
2162 notices that an array is degraded when it first sees the array.
2163 (syslog priority: Critical)
2167 A spare drive has been moved from one array in a
2171 to another to allow a failed drive to be replaced.
2172 (syslog priority: Info)
2178 has been told, via the config file, that an array should have a certain
2179 number of spare devices, and
2181 detects that it has fewer than this number when it first sees the
2182 array, it will report a
2185 (syslog priority: Warning)
2189 An array was found at startup, and the
2192 (syslog priority: Info)
2202 cause Email to be sent. All events cause the program to be run.
2203 The program is run with two or three arguments: the event
2204 name, the array device and possibly a second device.
2206 Each event has an associated array device (e.g.
2208 and possibly a second device. For
2213 the second device is the relevant component device.
2216 the second device is the array that the spare was moved from.
2220 to move spares from one array to another, the different arrays need to
2221 be labeled with the same
2223 or the spares must be allowed to migrate through matching POLICY domains
2224 in the configuration file. The
2226 name can be any string; it is only necessary that different spare
2227 groups use different names.
2231 detects that an array in a spare group has fewer active
2232 devices than necessary for the complete array, and has no spare
2233 devices, it will look for another array in the same spare group that
2234 has a full complement of working drive and a spare. It will then
2235 attempt to remove the spare from the second drive and add it to the
2237 If the removal succeeds but the adding fails, then it is added back to
2240 If the spare group for a degraded array is not defined,
2242 will look at the rules of spare migration specified by POLICY lines in
2244 and then follow similar steps as above if a matching spare is found.
2247 The GROW mode is used for changing the size or shape of an active
2249 For this to work, the kernel must support the necessary change.
2250 Various types of growth are being added during 2.6 development.
2252 Currently the supported changes include
2254 change the "size" attribute for RAID1, RAID4, RAID5 and RAID6.
2256 increase or decrease the "raid\-devices" attribute of RAID0, RAID1, RAID4,
2259 change the chunk-size and layout of RAID0, RAID4, RAID5 and RAID6.
2261 convert between RAID1 and RAID5, between RAID5 and RAID6, between
2262 RAID0, RAID4, and RAID5, and between RAID0 and RAID10 (in the near-2 mode).
2264 add a write-intent bitmap to any array which supports these bitmaps, or
2265 remove a write-intent bitmap from such an array.
2268 Using GROW on containers is currently supported only for Intel's IMSM
2269 container format. The number of devices in a container can be
2270 increased - which affects all arrays in the container - or an array
2271 in a container can be converted between levels where those levels are
2272 supported by the container, and the conversion is on of those listed
2273 above. Resizing arrays in an IMSM container with
2275 is not yet supported.
2277 Grow functionality (e.g. expand a number of raid devices) for Intel's
2278 IMSM container format has an experimental status. It is guarded by the
2279 .B MDADM_EXPERIMENTAL
2280 environment variable which must be set to '1' for a GROW command to
2282 This is for the following reasons:
2285 Intel's native IMSM check-pointing is not fully tested yet.
2286 This can causes IMSM incompatibility during the grow process: an array
2287 which is growing cannot roam between Microsoft Windows(R) and Linux
2291 Interrupting a grow operation is not recommended, because it
2292 has not been fully tested for Intel's IMSM container format yet.
2295 Note: Intel's native checkpointing doesn't use
2297 option and it is transparent for assembly feature.
2300 Normally when an array is built the "size" is taken from the smallest
2301 of the drives. If all the small drives in an arrays are, one at a
2302 time, removed and replaced with larger drives, then you could have an
2303 array of large drives with only a small amount used. In this
2304 situation, changing the "size" with "GROW" mode will allow the extra
2305 space to start being used. If the size is increased in this way, a
2306 "resync" process will start to make sure the new parts of the array
2309 Note that when an array changes size, any filesystem that may be
2310 stored in the array will not automatically grow or shrink to use or
2311 vacate the space. The
2312 filesystem will need to be explicitly told to use the extra space
2313 after growing, or to reduce its size
2315 to shrinking the array.
2317 Also the size of an array cannot be changed while it has an active
2318 bitmap. If an array has a bitmap, it must be removed before the size
2319 can be changed. Once the change is complete a new bitmap can be created.
2321 .SS RAID\-DEVICES CHANGES
2323 A RAID1 array can work with any number of devices from 1 upwards
2324 (though 1 is not very useful). There may be times which you want to
2325 increase or decrease the number of active devices. Note that this is
2326 different to hot-add or hot-remove which changes the number of
2329 When reducing the number of devices in a RAID1 array, the slots which
2330 are to be removed from the array must already be vacant. That is, the
2331 devices which were in those slots must be failed and removed.
2333 When the number of devices is increased, any hot spares that are
2334 present will be activated immediately.
2336 Changing the number of active devices in a RAID5 or RAID6 is much more
2337 effort. Every block in the array will need to be read and written
2338 back to a new location. From 2.6.17, the Linux Kernel is able to
2339 increase the number of devices in a RAID5 safely, including restarting
2340 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2341 increase or decrease the number of devices in a RAID5 or RAID6.
2343 From 2.6.35, the Linux Kernel is able to convert a RAID0 in to a RAID4
2346 uses this functionality and the ability to add
2347 devices to a RAID4 to allow devices to be added to a RAID0. When
2348 requested to do this,
2350 will convert the RAID0 to a RAID4, add the necessary disks and make
2351 the reshape happen, and then convert the RAID4 back to RAID0.
2353 When decreasing the number of devices, the size of the array will also
2354 decrease. If there was data in the array, it could get destroyed and
2355 this is not reversible, so you should firstly shrink the filesystem on
2356 the array to fit within the new size. To help prevent accidents,
2358 requires that the size of the array be decreased first with
2359 .BR "mdadm --grow --array-size" .
2360 This is a reversible change which simply makes the end of the array
2361 inaccessible. The integrity of any data can then be checked before
2362 the non-reversible reduction in the number of devices is request.
2364 When relocating the first few stripes on a RAID5 or RAID6, it is not
2365 possible to keep the data on disk completely consistent and
2366 crash-proof. To provide the required safety, mdadm disables writes to
2367 the array while this "critical section" is reshaped, and takes a
2368 backup of the data that is in that section. For grows, this backup may be
2369 stored in any spare devices that the array has, however it can also be
2370 stored in a separate file specified with the
2372 option, and is required to be specified for shrinks, RAID level
2373 changes and layout changes. If this option is used, and the system
2374 does crash during the critical period, the same file must be passed to
2376 to restore the backup and reassemble the array. When shrinking rather
2377 than growing the array, the reshape is done from the end towards the
2378 beginning, so the "critical section" is at the end of the reshape.
2382 Changing the RAID level of any array happens instantaneously. However
2383 in the RAID5 to RAID6 case this requires a non-standard layout of the
2384 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2385 required before the change can be accomplished. So while the level
2386 change is instant, the accompanying layout change can take quite a
2389 is required. If the array is not simultaneously being grown or
2390 shrunk, so that the array size will remain the same - for example,
2391 reshaping a 3-drive RAID5 into a 4-drive RAID6 - the backup file will
2392 be used not just for a "cricital section" but throughout the reshape
2393 operation, as described below under LAYOUT CHANGES.
2395 .SS CHUNK-SIZE AND LAYOUT CHANGES
2397 Changing the chunk-size of layout without also changing the number of
2398 devices as the same time will involve re-writing all blocks in-place.
2399 To ensure against data loss in the case of a crash, a
2401 must be provided for these changes. Small sections of the array will
2402 be copied to the backup file while they are being rearranged. This
2403 means that all the data is copied twice, once to the backup and once
2404 to the new layout on the array, so this type of reshape will go very
2407 If the reshape is interrupted for any reason, this backup file must be
2409 .B "mdadm --assemble"
2410 so the array can be reassembled. Consequently the file cannot be
2411 stored on the device being reshaped.
2416 A write-intent bitmap can be added to, or removed from, an active
2417 array. Either internal bitmaps, or bitmaps stored in a separate file,
2418 can be added. Note that if you add a bitmap stored in a file which is
2419 in a filesystem that is on the RAID array being affected, the system
2420 will deadlock. The bitmap must be on a separate filesystem.
2422 .SH INCREMENTAL MODE
2426 .B mdadm \-\-incremental
2432 .B mdadm \-\-incremental \-\-fail
2436 .B mdadm \-\-incremental \-\-rebuild\-map
2439 .B mdadm \-\-incremental \-\-run \-\-scan
2442 This mode is designed to be used in conjunction with a device
2443 discovery system. As devices are found in a system, they can be
2445 .B "mdadm \-\-incremental"
2446 to be conditionally added to an appropriate array.
2448 Conversely, it can also be used with the
2450 flag to do just the opposite and find whatever array a particular device
2451 is part of and remove the device from that array.
2453 If the device passed is a
2455 device created by a previous call to
2457 then rather than trying to add that device to an array, all the arrays
2458 described by the metadata of the container will be started.
2461 performs a number of tests to determine if the device is part of an
2462 array, and which array it should be part of. If an appropriate array
2463 is found, or can be created,
2465 adds the device to the array and conditionally starts the array.
2469 will normally only add devices to an array which were previously working
2470 (active or spare) parts of that array. The support for automatic
2471 inclusion of a new drive as a spare in some array requires
2472 a configuration through POLICY in config file.
2476 makes are as follow:
2478 Is the device permitted by
2480 That is, is it listed in a
2482 line in that file. If
2484 is absent then the default it to allow any device. Similar if
2486 contains the special word
2488 then any device is allowed. Otherwise the device name given to
2490 must match one of the names or patterns in a
2495 Does the device have a valid md superblock? If a specific metadata
2496 version is requested with
2500 then only that style of metadata is accepted, otherwise
2502 finds any known version of metadata. If no
2504 metadata is found, the device may be still added to an array
2505 as a spare if POLICY allows.
2509 Does the metadata match an expected array?
2510 The metadata can match in two ways. Either there is an array listed
2513 which identifies the array (either by UUID, by name, by device list,
2514 or by minor-number), or the array was created with a
2520 or on the command line.
2523 is not able to positively identify the array as belonging to the
2524 current host, the device will be rejected.
2529 keeps a list of arrays that it has partially assembled in
2530 .B /var/run/mdadm/map
2532 .B /var/run/mdadm.map
2533 if the directory doesn't exist. Or maybe even
2534 .BR /dev/.mdadm.map ).
2535 If no array exists which matches
2536 the metadata on the new device,
2538 must choose a device name and unit number. It does this based on any
2541 or any name information stored in the metadata. If this name
2542 suggests a unit number, that number will be used, otherwise a free
2543 unit number will be chosen. Normally
2545 will prefer to create a partitionable array, however if the
2549 suggests that a non-partitionable array is preferred, that will be
2552 If the array is not found in the config file and its metadata does not
2553 identify it as belonging to the "homehost", then
2555 will choose a name for the array which is certain not to conflict with
2556 any array which does belong to this host. It does this be adding an
2557 underscore and a small number to the name preferred by the metadata.
2559 Once an appropriate array is found or created and the device is added,
2561 must decide if the array is ready to be started. It will
2562 normally compare the number of available (non-spare) devices to the
2563 number of devices that the metadata suggests need to be active. If
2564 there are at least that many, the array will be started. This means
2565 that if any devices are missing the array will not be restarted.
2571 in which case the array will be run as soon as there are enough
2572 devices present for the data to be accessible. For a RAID1, that
2573 means one device will start the array. For a clean RAID5, the array
2574 will be started as soon as all but one drive is present.
2576 Note that neither of these approaches is really ideal. If it can
2577 be known that all device discovery has completed, then
2581 can be run which will try to start all arrays that are being
2582 incrementally assembled. They are started in "read-auto" mode in
2583 which they are read-only until the first write request. This means
2584 that no metadata updates are made and no attempt at resync or recovery
2585 happens. Further devices that are found before the first write can
2586 still be added safely.
2589 This section describes environment variables that affect how mdadm
2594 Setting this value to 1 will prevent mdadm from automatically launching
2595 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2601 does not create any device nodes in /dev, but leaves that task to
2605 appears not to be configured, or if this environment variable is set
2608 will create and devices that are needed.
2612 .B " mdadm \-\-query /dev/name-of-device"
2614 This will find out if a given device is a RAID array, or is part of
2615 one, and will provide brief information about the device.
2617 .B " mdadm \-\-assemble \-\-scan"
2619 This will assemble and start all arrays listed in the standard config
2620 file. This command will typically go in a system startup file.
2622 .B " mdadm \-\-stop \-\-scan"
2624 This will shut down all arrays that can be shut down (i.e. are not
2625 currently in use). This will typically go in a system shutdown script.
2627 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2629 If (and only if) there is an Email address or program given in the
2630 standard config file, then
2631 monitor the status of all arrays listed in that file by
2632 polling them ever 2 minutes.
2634 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2636 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2639 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2641 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2643 This will create a prototype config file that describes currently
2644 active arrays that are known to be made from partitions of IDE or SCSI drives.
2645 This file should be reviewed before being used as it may
2646 contain unwanted detail.
2648 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2650 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
2652 This will find arrays which could be assembled from existing IDE and
2653 SCSI whole drives (not partitions), and store the information in the
2654 format of a config file.
2655 This file is very likely to contain unwanted detail, particularly
2658 entries. It should be reviewed and edited before being used as an
2661 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
2663 .B " mdadm \-Ebsc partitions"
2665 Create a list of devices by reading
2666 .BR /proc/partitions ,
2667 scan these for RAID superblocks, and printout a brief listing of all
2670 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
2672 Scan all partitions and devices listed in
2673 .BR /proc/partitions
2676 out of all such devices with a RAID superblock with a minor number of 0.
2678 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /var/run/mdadm"
2680 If config file contains a mail address or alert program, run mdadm in
2681 the background in monitor mode monitoring all md devices. Also write
2682 pid of mdadm daemon to
2683 .BR /var/run/mdadm .
2685 .B " mdadm \-Iq /dev/somedevice"
2687 Try to incorporate newly discovered device into some array as
2690 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
2692 Rebuild the array map from any current arrays, and then start any that
2695 .B " mdadm /dev/md4 --fail detached --remove detached"
2697 Any devices which are components of /dev/md4 will be marked as faulty
2698 and then remove from the array.
2700 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4"
2704 which is currently a RAID5 array will be converted to RAID6. There
2705 should normally already be a spare drive attached to the array as a
2706 RAID6 needs one more drive than a matching RAID5.
2708 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
2710 Create a DDF array over 6 devices.
2712 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
2714 Create a RAID5 array over any 3 devices in the given DDF set. Use
2715 only 30 gigabytes of each device.
2717 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
2719 Assemble a pre-exist ddf array.
2721 .B " mdadm -I /dev/md/ddf1"
2723 Assemble all arrays contained in the ddf array, assigning names as
2726 .B " mdadm \-\-create \-\-help"
2728 Provide help about the Create mode.
2730 .B " mdadm \-\-config \-\-help"
2732 Provide help about the format of the config file.
2734 .B " mdadm \-\-help"
2736 Provide general help.
2746 lists all active md devices with information about them.
2748 uses this to find arrays when
2750 is given in Misc mode, and to monitor array reconstruction
2755 The config file lists which devices may be scanned to see if
2756 they contain MD super block, and gives identifying information
2757 (e.g. UUID) about known MD arrays. See
2761 .SS /var/run/mdadm/map
2764 mode is used, this file gets a list of arrays currently being created.
2767 does not exist as a directory, then
2768 .B /var/run/mdadm.map
2771 is not available (as may be the case during early boot),
2773 is used on the basis that
2775 is usually available very early in boot.
2780 understand two sorts of names for array devices.
2782 The first is the so-called 'standard' format name, which matches the
2783 names used by the kernel and which appear in
2786 The second sort can be freely chosen, but must reside in
2788 When giving a device name to
2790 to create or assemble an array, either full path name such as
2794 can be given, or just the suffix of the second sort of name, such as
2800 chooses device names during auto-assembly or incremental assembly, it
2801 will sometimes add a small sequence number to the end of the name to
2802 avoid conflicted between multiple arrays that have the same name. If
2804 can reasonably determine that the array really is meant for this host,
2805 either by a hostname in the metadata, or by the presence of the array
2808 then it will leave off the suffix if possible.
2809 Also if the homehost is specified as
2812 will only use a suffix if a different array of the same name already
2813 exists or is listed in the config file.
2815 The standard names for non-partitioned arrays (the only sort of md
2816 array available in 2.4 and earlier) are of the form
2820 where NN is a number.
2821 The standard names for partitionable arrays (as available from 2.6
2822 onwards) are of the form
2826 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2828 From kernel version, 2.6.28 the "non-partitioned array" can actually
2829 be partitioned. So the "md_dNN" names are no longer needed, and
2830 partitions such as "/dev/mdNNpXX" are possible.
2834 was previously known as
2838 is completely separate from the
2840 package, and does not use the
2842 configuration file at all.
2845 For further information on mdadm usage, MD and the various levels of
2848 .B http://raid.wiki.kernel.org/
2850 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
2852 .\"for new releases of the RAID driver check out:
2855 .\".UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2856 .\"ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2861 .\".UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2862 .\"http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2865 The latest version of
2867 should always be available from
2869 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/