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 RAID
126 levels 1/4/5/6, changing the RAID level between 1, 5, and 6, changing
127 the chunk size and layout for RAID5 and RAID5, as well as adding or
128 removing a write-intent bitmap.
131 .B "Incremental Assembly"
132 Add a single device to an appropriate array. If the addition of the
133 device makes the array runnable, the array will be started.
134 This provides a convenient interface to a
136 system. As each device is detected,
138 has a chance to include it in some array as appropriate.
141 flag is passed in we will remove the device from any active array
142 instead of adding it.
148 in this mode, then any arrays within that container will be assembled
153 This is for doing things to specific components of an array such as
154 adding new spares and removing faulty devices.
158 This is an 'everything else' mode that supports operations on active
159 arrays, operations on component devices such as erasing old superblocks, and
160 information gathering operations.
161 .\"This mode allows operations on independent devices such as examine MD
162 .\"superblocks, erasing old superblocks and stopping active arrays.
166 This mode does not act on a specific device or array, but rather it
167 requests the Linux Kernel to activate any auto-detected arrays.
170 .SH Options for selecting a mode are:
173 .BR \-A ", " \-\-assemble
174 Assemble a pre-existing array.
177 .BR \-B ", " \-\-build
178 Build a legacy array without superblocks.
181 .BR \-C ", " \-\-create
185 .BR \-F ", " \-\-follow ", " \-\-monitor
191 .BR \-G ", " \-\-grow
192 Change the size or shape of an active array.
195 .BR \-I ", " \-\-incremental
196 Add/remove a single device to/from an appropriate array, and possibly start the array.
200 Request that the kernel starts any auto-detected arrays. This can only
203 is compiled into the kernel \(em not if it is a module.
204 Arrays can be auto-detected by the kernel if all the components are in
205 primary MS-DOS partitions with partition type
207 and all use v0.90 metadata.
208 In-kernel autodetect is not recommended for new installations. Using
210 to detect and assemble arrays \(em possibly in an
212 \(em is substantially more flexible and should be preferred.
215 If a device is given before any options, or if the first option is
220 then the MANAGE mode is assumed.
221 Anything other than these will cause the
225 .SH Options that are not mode-specific are:
228 .BR \-h ", " \-\-help
229 Display general help message or, after one of the above options, a
230 mode-specific help message.
234 Display more detailed help about command line parsing and some commonly
238 .BR \-V ", " \-\-version
239 Print version information for mdadm.
242 .BR \-v ", " \-\-verbose
243 Be more verbose about what is happening. This can be used twice to be
245 The extra verbosity currently only affects
246 .B \-\-detail \-\-scan
248 .BR "\-\-examine \-\-scan" .
251 .BR \-q ", " \-\-quiet
252 Avoid printing purely informative messages. With this,
254 will be silent unless there is something really important to report.
257 .BR \-f ", " \-\-force
258 Be more forceful about certain operations. See the various modes for
259 the exact meaning of this option in different contexts.
262 .BR \-c ", " \-\-config=
263 Specify the config file. Default is to use
264 .BR /etc/mdadm.conf ,
265 or if that is missing then
266 .BR /etc/mdadm/mdadm.conf .
267 If the config file given is
269 then nothing will be read, but
271 will act as though the config file contained exactly
272 .B "DEVICE partitions containers"
275 to find a list of devices to scan, and
277 to find a list of containers to examine.
280 is given for the config file, then
282 will act as though the config file were empty.
285 .BR \-s ", " \-\-scan
288 for missing information.
289 In general, this option gives
291 permission to get any missing information (like component devices,
292 array devices, array identities, and alert destination) from the
293 configuration file (see previous option);
294 one exception is MISC mode when using
300 says to get a list of array devices from
304 .BR \-e ", " \-\-metadata=
305 Declare the style of RAID metadata (superblock) to be used. The
306 default is {DEFAULT_METADATA} for
308 and to guess for other operations.
309 The default can be overridden by setting the
318 .ie '{DEFAULT_METADATA}'0.90'
319 .IP "0, 0.90, default"
323 Use the original 0.90 format superblock. This format limits arrays to
324 28 component devices and limits component devices of levels 1 and
325 greater to 2 terabytes. It is also possible for there to be confusion
326 about whether the superblock applies to a whole device or just the
327 last partition, if that partition starts on a 64K boundary.
328 .ie '{DEFAULT_METADATA}'0.90'
329 .IP "1, 1.0, 1.1, 1.2"
331 .IP "1, 1.0, 1.1, 1.2 default"
333 Use the new version-1 format superblock. This has fewer restrictions.
334 It can easily be moved between hosts with different endian-ness, and a
335 recovery operation can be checkpointed and restarted. The different
336 sub-versions store the superblock at different locations on the
337 device, either at the end (for 1.0), at the start (for 1.1) or 4K from
338 the start (for 1.2). "1" is equivalent to "1.0".
339 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
341 Use the "Industry Standard" DDF (Disk Data Format) format defined by
343 When creating a DDF array a
345 will be created, and normal arrays can be created in that container.
347 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
349 which is managed in a similar manner to DDF, and is supported by an
350 option-rom on some platforms:
352 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
358 This will override any
360 setting in the config file and provides the identity of the host which
361 should be considered the home for any arrays.
363 When creating an array, the
365 will be recorded in the metadata. For version-1 superblocks, it will
366 be prefixed to the array name. For version-0.90 superblocks, part of
367 the SHA1 hash of the hostname will be stored in the later half of the
370 When reporting information about an array, any array which is tagged
371 for the given homehost will be reported as such.
373 When using Auto-Assemble, only arrays tagged for the given homehost
374 will be allowed to use 'local' names (i.e. not ending in '_' followed
375 by a digit string). See below under
376 .BR "Auto Assembly" .
378 .SH For create, build, or grow:
381 .BR \-n ", " \-\-raid\-devices=
382 Specify the number of active devices in the array. This, plus the
383 number of spare devices (see below) must equal the number of
385 (including "\fBmissing\fP" devices)
386 that are listed on the command line for
388 Setting a value of 1 is probably
389 a mistake and so requires that
391 be specified first. A value of 1 will then be allowed for linear,
392 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
394 This number can only be changed using
396 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
397 the necessary support.
400 .BR \-x ", " \-\-spare\-devices=
401 Specify the number of spare (eXtra) devices in the initial array.
402 Spares can also be added
403 and removed later. The number of component devices listed
404 on the command line must equal the number of RAID devices plus the
405 number of spare devices.
408 .BR \-z ", " \-\-size=
409 Amount (in Kibibytes) of space to use from each drive in RAID levels 1/4/5/6.
410 This must be a multiple of the chunk size, and must leave about 128Kb
411 of space at the end of the drive for the RAID superblock.
412 If this is not specified
413 (as it normally is not) the smallest drive (or partition) sets the
414 size, though if there is a variance among the drives of greater than 1%, a warning is
417 This value can be set with
419 for RAID level 1/4/5/6. If the array was created with a size smaller
420 than the currently active drives, the extra space can be accessed
423 The size can be given as
425 which means to choose the largest size that fits on all current drives.
427 Before reducing the size of the array (with
428 .BR "\-\-grow \-\-size=" )
429 you should make sure that space isn't needed. If the device holds a
430 filesystem, you would need to resize the filesystem to use less space.
432 After reducing the array size you should check that the data stored in
433 the device is still available. If the device holds a filesystem, then
434 an 'fsck' of the filesystem is a minimum requirement. If there are
435 problems the array can be made bigger again with no loss with another
436 .B "\-\-grow \-\-size="
439 This value can not be used with
441 metadata such as DDF and IMSM.
444 .BR \-c ", " \-\-chunk=
445 Specify chunk size of kibibytes. The default when creating an
446 array is 512KB. To ensure compatibility with earlier versions, the
447 default when Building and array with no persistent metadata is 64KB.
448 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
452 Specify rounding factor for a Linear array. The size of each
453 component will be rounded down to a multiple of this size.
454 This is a synonym for
456 but highlights the different meaning for Linear as compared to other
457 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
458 use, and is 0K (i.e. no rounding) in later kernels.
461 .BR \-l ", " \-\-level=
462 Set RAID level. When used with
464 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
465 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
466 Obviously some of these are synonymous.
470 metadata type is requested, only the
472 level is permitted, and it does not need to be explicitly given.
476 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
480 to change the RAID level in some cases. See LEVEL CHANGES below.
483 .BR \-p ", " \-\-layout=
484 This option configures the fine details of data layout for RAID5, RAID6,
485 and RAID10 arrays, and controls the failure modes for
488 The layout of the RAID5 parity block can be one of
489 .BR left\-asymmetric ,
490 .BR left\-symmetric ,
491 .BR right\-asymmetric ,
492 .BR right\-symmetric ,
493 .BR la ", " ra ", " ls ", " rs .
495 .BR left\-symmetric .
497 It is also possible to cause RAID5 to use a RAID4-like layout by
503 Finally for RAID5 there are DDF\-compatible layouts,
504 .BR ddf\-zero\-restart ,
505 .BR ddf\-N\-restart ,
507 .BR ddf\-N\-continue .
509 These same layouts are available for RAID6. There are also 4 layouts
510 that will provide an intermediate stage for converting between RAID5
511 and RAID6. These provide a layout which is identical to the
512 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
513 syndrome (the second 'parity' block used by RAID6) on the last device.
515 .BR left\-symmetric\-6 ,
516 .BR right\-symmetric\-6 ,
517 .BR left\-asymmetric\-6 ,
518 .BR right\-asymmetric\-6 ,
520 .BR parity\-first\-6 .
522 When setting the failure mode for level
525 .BR write\-transient ", " wt ,
526 .BR read\-transient ", " rt ,
527 .BR write\-persistent ", " wp ,
528 .BR read\-persistent ", " rp ,
530 .BR read\-fixable ", " rf ,
531 .BR clear ", " flush ", " none .
533 Each failure mode can be followed by a number, which is used as a period
534 between fault generation. Without a number, the fault is generated
535 once on the first relevant request. With a number, the fault will be
536 generated after that many requests, and will continue to be generated
537 every time the period elapses.
539 Multiple failure modes can be current simultaneously by using the
541 option to set subsequent failure modes.
543 "clear" or "none" will remove any pending or periodic failure modes,
544 and "flush" will clear any persistent faults.
546 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
547 by a small number. The default is 'n2'. The supported options are:
550 signals 'near' copies. Multiple copies of one data block are at
551 similar offsets in different devices.
554 signals 'offset' copies. Rather than the chunks being duplicated
555 within a stripe, whole stripes are duplicated but are rotated by one
556 device so duplicate blocks are on different devices. Thus subsequent
557 copies of a block are in the next drive, and are one chunk further
562 (multiple copies have very different offsets).
563 See md(4) for more detail about 'near', 'offset', and 'far'.
565 The number is the number of copies of each datablock. 2 is normal, 3
566 can be useful. This number can be at most equal to the number of
567 devices in the array. It does not need to divide evenly into that
568 number (e.g. it is perfectly legal to have an 'n2' layout for an array
569 with an odd number of devices).
571 When an array is converted between RAID5 and RAID6 an intermediate
572 RAID6 layout is used in which the second parity block (Q) is always on
573 the last device. To convert a RAID5 to RAID6 and leave it in this new
574 layout (which does not require re-striping) use
575 .BR \-\-layout=preserve .
576 This will try to avoid any restriping.
578 The converse of this is
579 .B \-\-layout=normalise
580 which will change a non-standard RAID6 layout into a more standard
587 (thus explaining the p of
591 .BR \-b ", " \-\-bitmap=
592 Specify a file to store a write-intent bitmap in. The file should not
595 is also given. The same file should be provided
596 when assembling the array. If the word
598 is given, then the bitmap is stored with the metadata on the array,
599 and so is replicated on all devices. If the word
603 mode, then any bitmap that is present is removed.
605 To help catch typing errors, the filename must contain at least one
606 slash ('/') if it is a real file (not 'internal' or 'none').
608 Note: external bitmaps are only known to work on ext2 and ext3.
609 Storing bitmap files on other filesystems may result in serious problems.
612 .BR \-\-bitmap\-chunk=
613 Set the chunksize of the bitmap. Each bit corresponds to that many
614 Kilobytes of storage.
615 When using a file based bitmap, the default is to use the smallest
616 size that is at-least 4 and requires no more than 2^21 chunks.
619 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
620 fit the bitmap into the available space.
623 .BR \-W ", " \-\-write\-mostly
624 subsequent devices listed in a
629 command will be flagged as 'write-mostly'. This is valid for RAID1
630 only and means that the 'md' driver will avoid reading from these
631 devices if at all possible. This can be useful if mirroring over a
635 .BR \-\-write\-behind=
636 Specify that write-behind mode should be enabled (valid for RAID1
637 only). If an argument is specified, it will set the maximum number
638 of outstanding writes allowed. The default value is 256.
639 A write-intent bitmap is required in order to use write-behind
640 mode, and write-behind is only attempted on drives marked as
644 .BR \-\-assume\-clean
647 that the array pre-existed and is known to be clean. It can be useful
648 when trying to recover from a major failure as you can be sure that no
649 data will be affected unless you actually write to the array. It can
650 also be used when creating a RAID1 or RAID10 if you want to avoid the
651 initial resync, however this practice \(em while normally safe \(em is not
652 recommended. Use this only if you really know what you are doing.
654 When the devices that will be part of a new array were filled
655 with zeros before creation the operator knows the array is
656 actually clean. If that is the case, such as after running
657 badblocks, this argument can be used to tell mdadm the
658 facts the operator knows.
661 .BR \-\-backup\-file=
664 is used to increase the number of raid-devices in a RAID5 or RAID6 if
665 there are no spare devices available, or to shrink, change RAID level
666 or layout. See the GROW MODE section below on RAID\-DEVICES CHANGES.
667 The file must be stored on a separate device, not on the RAID array
671 .BR \-\-array-size= ", " \-Z
672 This is only meaningful with
674 and its effect is not persistent: when the array is stopped an
675 restarted the default array size will be restored.
677 Setting the array-size causes the array to appear smaller to programs
678 that access the data. This is particularly needed before reshaping an
679 array so that it will be smaller. As the reshape is not reversible,
680 but setting the size with
682 is, it is required that the array size is reduced as appropriate
683 before the number of devices in the array is reduced.
687 restores the apparent size of the array to be whatever the real
688 amount of available space is.
690 Before reducing the size of the array you should make sure that space
691 isn't needed. If the device holds a filesystem, you would need to
692 resize the filesystem to use less space.
694 After reducing the array size you should check that the data stored in
695 the device is still available. If the device holds a filesystem, then
696 an 'fsck' of the filesystem is a minimum requirement. If there are
697 problems the array can be made bigger again with no loss with another
698 .B "\-\-grow \-\-array\-size="
702 .BR \-N ", " \-\-name=
705 for the array. This is currently only effective when creating an
706 array with a version-1 superblock, or an array in a DDF container.
707 The name is a simple textual string that can be used to identify array
708 components when assembling. If name is needed but not specified, it
709 is taken from the basename of the device that is being created.
721 run the array, even if some of the components
722 appear to be active in another array or filesystem. Normally
724 will ask for confirmation before including such components in an
725 array. This option causes that question to be suppressed.
728 .BR \-f ", " \-\-force
731 accept the geometry and layout specified without question. Normally
733 will not allow creation of an array with only one device, and will try
734 to create a RAID5 array with one missing drive (as this makes the
735 initial resync work faster). With
738 will not try to be so clever.
741 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
742 Instruct mdadm how to create the device file if needed, possibly allocating
743 an unused minor number. "md" causes a non-partitionable array
744 to be used (though since Linux 2.6.28, these array devices are in fact
745 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
746 later) to be used. "yes" requires the named md device to have
747 a 'standard' format, and the type and minor number will be determined
748 from this. With mdadm 3.0, device creation is normally left up to
750 so this option is unlikely to be needed.
751 See DEVICE NAMES below.
753 The argument can also come immediately after
758 is not given on the command line or in the config file, then
764 is also given, then any
766 entries in the config file will override the
768 instruction given on the command line.
770 For partitionable arrays,
772 will create the device file for the whole array and for the first 4
773 partitions. A different number of partitions can be specified at the
774 end of this option (e.g.
776 If the device name ends with a digit, the partition names add a 'p',
778 .IR /dev/md/home1p3 .
779 If there is no trailing digit, then the partition names just have a
781 .IR /dev/md/scratch3 .
783 If the md device name is in a 'standard' format as described in DEVICE
784 NAMES, then it will be created, if necessary, with the appropriate
785 device number based on that name. If the device name is not in one of these
786 formats, then a unused device number will be allocated. The device
787 number will be considered unused if there is no active array for that
788 number, and there is no entry in /dev for that number and with a
789 non-standard name. Names that are not in 'standard' format are only
790 allowed in "/dev/md/".
794 .\".BR \-\-symlink = no
799 .\"to create devices in
801 .\"it will also create symlinks from
803 .\"with names starting with
809 .\"to suppress this, or
810 .\".B \-\-symlink=yes
811 .\"to enforce this even if it is suppressing
819 .BR \-u ", " \-\-uuid=
820 uuid of array to assemble. Devices which don't have this uuid are
824 .BR \-m ", " \-\-super\-minor=
825 Minor number of device that array was created for. Devices which
826 don't have this minor number are excluded. If you create an array as
827 /dev/md1, then all superblocks will contain the minor number 1, even if
828 the array is later assembled as /dev/md2.
830 Giving the literal word "dev" for
834 to use the minor number of the md device that is being assembled.
837 .B \-\-super\-minor=dev
838 will look for super blocks with a minor number of 0.
841 is only relevant for v0.90 metadata, and should not normally be used.
847 .BR \-N ", " \-\-name=
848 Specify the name of the array to assemble. This must be the name
849 that was specified when creating the array. It must either match
850 the name stored in the superblock exactly, or it must match
853 prefixed to the start of the given name.
856 .BR \-f ", " \-\-force
857 Assemble the array even if the metadata on some devices appears to be
860 cannot find enough working devices to start the array, but can find
861 some devices that are recorded as having failed, then it will mark
862 those devices as working so that the array can be started.
863 An array which requires
865 to be started may contain data corruption. Use it carefully.
869 Attempt to start the array even if fewer drives were given than were
870 present last time the array was active. Normally if not all the
871 expected drives are found and
873 is not used, then the array will be assembled but not started.
876 an attempt will be made to start it anyway.
880 This is the reverse of
882 in that it inhibits the startup of array unless all expected drives
883 are present. This is only needed with
885 and can be used if the physical connections to devices are
886 not as reliable as you would like.
889 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
890 See this option under Create and Build options.
893 .BR \-b ", " \-\-bitmap=
894 Specify the bitmap file that was given when the array was created. If
897 bitmap, there is no need to specify this when assembling the array.
900 .BR \-\-backup\-file=
903 was used when requesting a grow, shrink, RAID level change or other
904 reshape, and the system crashed during the critical section, then the
909 to allow possibly corrupted data to be restored, and the reshape
913 .BR \-U ", " \-\-update=
914 Update the superblock on each device while assembling the array. The
915 argument given to this flag can be one of
930 option will adjust the superblock of an array what was created on a Sparc
931 machine running a patched 2.2 Linux kernel. This kernel got the
932 alignment of part of the superblock wrong. You can use the
933 .B "\-\-examine \-\-sparc2.2"
936 to see what effect this would have.
940 option will update the
942 field on each superblock to match the minor number of the array being
944 This can be useful if
946 reports a different "Preferred Minor" to
948 In some cases this update will be performed automatically
949 by the kernel driver. In particular the update happens automatically
950 at the first write to an array with redundancy (RAID level 1 or
951 greater) on a 2.6 (or later) kernel.
955 option will change the uuid of the array. If a UUID is given with the
957 option that UUID will be used as a new UUID and will
959 be used to help identify the devices in the array.
962 is given, a random UUID is chosen.
966 option will change the
968 of the array as stored in the superblock. This is only supported for
969 version-1 superblocks.
973 option will change the
975 as recorded in the superblock. For version-0 superblocks, this is the
976 same as updating the UUID.
977 For version-1 superblocks, this involves updating the name.
981 option will cause the array to be marked
983 meaning that any redundancy in the array (e.g. parity for RAID5,
984 copies for RAID1) may be incorrect. This will cause the RAID system
985 to perform a "resync" pass to make sure that all redundant information
990 option allows arrays to be moved between machines with different
992 When assembling such an array for the first time after a move, giving
993 .B "\-\-update=byteorder"
996 to expect superblocks to have their byteorder reversed, and will
997 correct that order before assembling the array. This is only valid
998 with original (Version 0.90) superblocks.
1002 option will correct the summaries in the superblock. That is the
1003 counts of total, working, active, failed, and spare devices.
1007 option will rarely be of use. It applies to version 1.1 and 1.2 metadata
1008 only (where the metadata is at the start of the device) and is only
1009 useful when the component device has changed size (typically become
1010 larger). The version 1 metadata records the amount of the device that
1011 can be used to store data, so if a device in a version 1.1 or 1.2
1012 array becomes larger, the metadata will still be visible, but the
1013 extra space will not. In this case it might be useful to assemble the
1015 .BR \-\-update=devicesize .
1018 to determine the maximum usable amount of space on each device and
1019 update the relevant field in the metadata.
1023 option can be used when an array has an internal bitmap which is
1024 corrupt in some way so that assembling the array normally fails. It
1025 will cause any internal bitmap to be ignored.
1029 .B \-\-auto\-update\-homehost
1030 This flag is only meaningful with auto-assembly (see discussion below).
1031 In that situation, if no suitable arrays are found for this homehost,
1033 will rescan for any arrays at all and will assemble them and update the
1034 homehost to match the current host.
1037 .SH For Manage mode:
1040 .BR \-t ", " \-\-test
1041 Unless a more serious error occurred,
1043 will exit with a status of 2 if no changes were made to the array and
1044 0 if at least one change was made.
1045 This can be useful when an indirect specifier such as
1050 is used in requesting an operation on the array.
1052 will report failure if these specifiers didn't find any match.
1055 .BR \-a ", " \-\-add
1056 hot-add listed devices.
1057 If a device appears to have recently been part of the array
1058 (possibly it failed or was removed) the device is re-added as describe
1060 If that fails or the device was never part of the array, the device is
1061 added as a hot-spare.
1062 If the array is degraded, it will immediately start to rebuild data
1065 Note that this and the following options are only meaningful on array
1066 with redundancy. They don't apply to RAID0 or Linear.
1070 re\-add a device that was previous removed from an array.
1071 If the metadata on the device reports that it is a member of the
1072 array, and the slot that it used is still vacant, then the device will
1073 be added back to the array in the same position. This will normally
1074 cause the data for that device to be recovered. However based on the
1075 event count on the device, the recovery may only require sections that
1076 are flagged a write-intent bitmap to be recovered or may not require
1077 any recovery at all.
1079 When used on an array that has no metadata (i.e. it was built with
1081 it will be assumed that bitmap-based recovery is enough to make the
1082 device fully consistent with the array.
1084 If the device name given is
1086 then mdadm will try to find any device that looks like it should be
1087 part of the array but isn't and will try to re\-add all such devices.
1090 .BR \-r ", " \-\-remove
1091 remove listed devices. They must not be active. i.e. they should
1092 be failed or spare devices. As well as the name of a device file
1101 The first causes all failed device to be removed. The second causes
1102 any device which is no longer connected to the system (i.e an 'open'
1105 to be removed. This will only succeed for devices that are spares or
1106 have already been marked as failed.
1109 .BR \-f ", " \-\-fail
1110 mark listed devices as faulty.
1111 As well as the name of a device file, the word
1113 can be given. This will cause any device that has been detached from
1114 the system to be marked as failed. It can then be removed.
1122 .BR \-\-write\-mostly
1123 Subsequent devices that are added or re\-added will have the 'write-mostly'
1124 flag set. This is only valid for RAID1 and means that the 'md' driver
1125 will avoid reading from these devices if possible.
1128 Subsequent devices that are added or re\-added will have the 'write-mostly'
1132 Each of these options requires that the first device listed is the array
1133 to be acted upon, and the remainder are component devices to be added,
1134 removed, marked as faulty, etc. Several different operations can be
1135 specified for different devices, e.g.
1137 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1139 Each operation applies to all devices listed until the next
1142 If an array is using a write-intent bitmap, then devices which have
1143 been removed can be re\-added in a way that avoids a full
1144 reconstruction but instead just updates the blocks that have changed
1145 since the device was removed. For arrays with persistent metadata
1146 (superblocks) this is done automatically. For arrays created with
1148 mdadm needs to be told that this device we removed recently with
1151 Devices can only be removed from an array if they are not in active
1152 use, i.e. that must be spares or failed devices. To remove an active
1153 device, it must first be marked as
1159 .BR \-Q ", " \-\-query
1160 Examine a device to see
1161 (1) if it is an md device and (2) if it is a component of an md
1163 Information about what is discovered is presented.
1166 .BR \-D ", " \-\-detail
1167 Print details of one or more md devices.
1170 .BR \-\-detail\-platform
1171 Print details of the platform's RAID capabilities (firmware / hardware
1172 topology) for a given metadata format.
1175 .BR \-Y ", " \-\-export
1180 output will be formatted as
1182 pairs for easy import into the environment.
1185 .BR \-E ", " \-\-examine
1186 Print contents of the metadata stored on the named device(s).
1187 Note the contrast between
1192 applies to devices which are components of an array, while
1194 applies to a whole array which is currently active.
1197 If an array was created on a SPARC machine with a 2.2 Linux kernel
1198 patched with RAID support, the superblock will have been created
1199 incorrectly, or at least incompatibly with 2.4 and later kernels.
1204 will fix the superblock before displaying it. If this appears to do
1205 the right thing, then the array can be successfully assembled using
1206 .BR "\-\-assemble \-\-update=sparc2.2" .
1209 .BR \-X ", " \-\-examine\-bitmap
1210 Report information about a bitmap file.
1211 The argument is either an external bitmap file or an array component
1212 in case of an internal bitmap. Note that running this on an array
1215 does not report the bitmap for that array.
1218 .BR \-R ", " \-\-run
1219 start a partially assembled array. If
1221 did not find enough devices to fully start the array, it might leaving
1222 it partially assembled. If you wish, you can then use
1224 to start the array in degraded mode.
1227 .BR \-S ", " \-\-stop
1228 deactivate array, releasing all resources.
1231 .BR \-o ", " \-\-readonly
1232 mark array as readonly.
1235 .BR \-w ", " \-\-readwrite
1236 mark array as readwrite.
1239 .B \-\-zero\-superblock
1240 If the device contains a valid md superblock, the block is
1241 overwritten with zeros. With
1243 the block where the superblock would be is overwritten even if it
1244 doesn't appear to be valid.
1247 .B \-\-kill\-subarray=
1248 If the device is a container and the argument to \-\-kill\-subarray
1249 specifies an inactive subarray in the container, then the subarray is
1250 deleted. Deleting all subarrays will leave an 'empty-container' or
1251 spare superblock on the drives. See \-\-zero\-superblock for completely
1252 removing a superblock. Note that some formats depend on the subarray
1253 index for generating a UUID, this command will fail if it would change
1254 the UUID of an active subarray.
1257 .B \-\-update\-subarray=
1258 If the device is a container and the argument to \-\-update\-subarray
1259 specifies a subarray in the container, then attempt to update the given
1260 superblock field in the subarray. See below in
1265 .BR \-t ", " \-\-test
1270 is set to reflect the status of the device. See below in
1275 .BR \-W ", " \-\-wait
1276 For each md device given, wait for any resync, recovery, or reshape
1277 activity to finish before returning.
1279 will return with success if it actually waited for every device
1280 listed, otherwise it will return failure.
1284 For each md device given, or each device in /proc/mdstat if
1286 is given, arrange for the array to be marked clean as soon as possible.
1288 will return with success if the array uses external metadata and we
1289 successfully waited. For native arrays this returns immediately as the
1290 kernel handles dirty-clean transitions at shutdown. No action is taken
1291 if safe-mode handling is disabled.
1293 .SH For Incremental Assembly mode:
1295 .BR \-\-rebuild\-map ", " \-r
1296 Rebuild the map file
1297 .RB ( /var/run/mdadm/map )
1300 uses to help track which arrays are currently being assembled.
1303 .BR \-\-run ", " \-R
1304 Run any array assembled as soon as a minimal number of devices are
1305 available, rather than waiting until all expected devices are present.
1308 .BR \-\-scan ", " \-s
1309 Only meaningful with
1313 file for arrays that are being incrementally assembled and will try to
1314 start any that are not already started. If any such array is listed
1317 as requiring an external bitmap, that bitmap will be attached first.
1320 .BR \-\-fail ", " \-f
1321 This allows the hot-plug system to remove devices that have fully disappeared
1322 from the kernel. It will first fail and then remove the device from any
1323 array it belongs to.
1324 The device name given should be a kernel device name such as "sda",
1328 .SH For Monitor mode:
1330 .BR \-m ", " \-\-mail
1331 Give a mail address to send alerts to.
1334 .BR \-p ", " \-\-program ", " \-\-alert
1335 Give a program to be run whenever an event is detected.
1338 .BR \-y ", " \-\-syslog
1339 Cause all events to be reported through 'syslog'. The messages have
1340 facility of 'daemon' and varying priorities.
1343 .BR \-d ", " \-\-delay
1344 Give a delay in seconds.
1346 polls the md arrays and then waits this many seconds before polling
1347 again. The default is 60 seconds. Since 2.6.16, there is no need to
1348 reduce this as the kernel alerts
1350 immediately when there is any change.
1353 .BR \-r ", " \-\-increment
1354 Give a percentage increment.
1356 will generate RebuildNN events with the given percentage increment.
1359 .BR \-f ", " \-\-daemonise
1362 to run as a background daemon if it decides to monitor anything. This
1363 causes it to fork and run in the child, and to disconnect from the
1364 terminal. The process id of the child is written to stdout.
1367 which will only continue monitoring if a mail address or alert program
1368 is found in the config file.
1371 .BR \-i ", " \-\-pid\-file
1374 is running in daemon mode, write the pid of the daemon process to
1375 the specified file, instead of printing it on standard output.
1378 .BR \-1 ", " \-\-oneshot
1379 Check arrays only once. This will generate
1381 events and more significantly
1387 .B " mdadm \-\-monitor \-\-scan \-1"
1389 from a cron script will ensure regular notification of any degraded arrays.
1392 .BR \-t ", " \-\-test
1395 alert for every array found at startup. This alert gets mailed and
1396 passed to the alert program. This can be used for testing that alert
1397 message do get through successfully.
1403 .B mdadm \-\-assemble
1404 .I md-device options-and-component-devices...
1407 .B mdadm \-\-assemble \-\-scan
1408 .I md-devices-and-options...
1411 .B mdadm \-\-assemble \-\-scan
1415 This usage assembles one or more RAID arrays from pre-existing components.
1416 For each array, mdadm needs to know the md device, the identity of the
1417 array, and a number of component-devices. These can be found in a number of ways.
1419 In the first usage example (without the
1421 the first device given is the md device.
1422 In the second usage example, all devices listed are treated as md
1423 devices and assembly is attempted.
1424 In the third (where no devices are listed) all md devices that are
1425 listed in the configuration file are assembled. If not arrays are
1426 described by the configuration file, then any arrays that
1427 can be found on unused devices will be assembled.
1429 If precisely one device is listed, but
1435 was given and identity information is extracted from the configuration file.
1437 The identity can be given with the
1443 option, will be taken from the md-device record in the config file, or
1444 will be taken from the super block of the first component-device
1445 listed on the command line.
1447 Devices can be given on the
1449 command line or in the config file. Only devices which have an md
1450 superblock which contains the right identity will be considered for
1453 The config file is only used if explicitly named with
1455 or requested with (a possibly implicit)
1460 .B /etc/mdadm/mdadm.conf
1465 is not given, then the config file will only be used to find the
1466 identity of md arrays.
1468 Normally the array will be started after it is assembled. However if
1470 is not given and not all expected drives were listed, then the array
1471 is not started (to guard against usage errors). To insist that the
1472 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1481 does not create any entries in
1485 It does record information in
1486 .B /var/run/mdadm/map
1489 to choose the correct name.
1493 detects that udev is not configured, it will create the devices in
1497 In Linux kernels prior to version 2.6.28 there were two distinctly
1498 different types of md devices that could be created: one that could be
1499 partitioned using standard partitioning tools and one that could not.
1500 Since 2.6.28 that distinction is no longer relevant as both type of
1501 devices can be partitioned.
1503 will normally create the type that originally could not be partitioned
1504 as it has a well defined major number (9).
1506 Prior to 2.6.28, it is important that mdadm chooses the correct type
1507 of array device to use. This can be controlled with the
1509 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1510 to use a partitionable device rather than the default.
1512 In the no-udev case, the value given to
1514 can be suffixed by a number. This tells
1516 to create that number of partition devices rather than the default of 4.
1520 can also be given in the configuration file as a word starting
1522 on the ARRAY line for the relevant array.
1529 and no devices are listed,
1531 will first attempt to assemble all the arrays listed in the config
1534 In no array at listed in the config (other than those marked
1536 it will look through the available devices for possible arrays and
1537 will try to assemble anything that it finds. Arrays which are tagged
1538 as belonging to the given homehost will be assembled and started
1539 normally. Arrays which do not obviously belong to this host are given
1540 names that are expected not to conflict with anything local, and are
1541 started "read-auto" so that nothing is written to any device until the
1542 array is written to. i.e. automatic resync etc is delayed.
1546 finds a consistent set of devices that look like they should comprise
1547 an array, and if the superblock is tagged as belonging to the given
1548 home host, it will automatically choose a device name and try to
1549 assemble the array. If the array uses version-0.90 metadata, then the
1551 number as recorded in the superblock is used to create a name in
1555 If the array uses version-1 metadata, then the
1557 from the superblock is used to similarly create a name in
1559 (the name will have any 'host' prefix stripped first).
1561 This behaviour can be modified by the
1565 configuration file. This line can indicate that specific metadata
1566 type should, or should not, be automatically assembled. If an array
1567 is found which is not listed in
1569 and has a metadata format that is denied by the
1571 line, then it will not be assembled.
1574 line can also request that all arrays identified as being for this
1575 homehost should be assembled regardless of their metadata type.
1578 for further details.
1583 cannot find any array for the given host at all, and if
1584 .B \-\-auto\-update\-homehost
1587 will search again for any array (not just an array created for this
1588 host) and will assemble each assuming
1589 .BR \-\-update=homehost .
1590 This will change the host tag in the superblock so that on the next run,
1591 these arrays will be found without the second pass. The intention of
1592 this feature is to support transitioning a set of md arrays to using
1595 The reason for requiring arrays to be tagged with the homehost for
1596 auto assembly is to guard against problems that can arise when moving
1597 devices from one host to another.
1608 .BI \-\-raid\-devices= Z
1612 This usage is similar to
1614 The difference is that it creates an array without a superblock. With
1615 these arrays there is no difference between initially creating the array and
1616 subsequently assembling the array, except that hopefully there is useful
1617 data there in the second case.
1619 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1620 one of their synonyms. All devices must be listed and the array will
1621 be started once complete. It will often be appropriate to use
1622 .B \-\-assume\-clean
1623 with levels raid1 or raid10.
1634 .BI \-\-raid\-devices= Z
1638 This usage will initialise a new md array, associate some devices with
1639 it, and activate the array.
1641 The named device will normally not exist when
1642 .I "mdadm \-\-create"
1643 is run, but will be created by
1645 once the array becomes active.
1647 As devices are added, they are checked to see if they contain RAID
1648 superblocks or filesystems. They are also checked to see if the variance in
1649 device size exceeds 1%.
1651 If any discrepancy is found, the array will not automatically be run, though
1654 can override this caution.
1656 To create a "degraded" array in which some devices are missing, simply
1657 give the word "\fBmissing\fP"
1658 in place of a device name. This will cause
1660 to leave the corresponding slot in the array empty.
1661 For a RAID4 or RAID5 array at most one slot can be
1662 "\fBmissing\fP"; for a RAID6 array at most two slots.
1663 For a RAID1 array, only one real device needs to be given. All of the
1667 When creating a RAID5 array,
1669 will automatically create a degraded array with an extra spare drive.
1670 This is because building the spare into a degraded array is in general
1671 faster than resyncing the parity on a non-degraded, but not clean,
1672 array. This feature can be overridden with the
1676 When creating an array with version-1 metadata a name for the array is
1678 If this is not given with the
1682 will choose a name based on the last component of the name of the
1683 device being created. So if
1685 is being created, then the name
1690 is being created, then the name
1694 When creating a partition based array, using
1696 with version-1.x metadata, the partition type should be set to
1698 (non fs-data). This type selection allows for greater precision since
1699 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
1700 might create problems in the event of array recovery through a live cdrom.
1702 A new array will normally get a randomly assigned 128bit UUID which is
1703 very likely to be unique. If you have a specific need, you can choose
1704 a UUID for the array by giving the
1706 option. Be warned that creating two arrays with the same UUID is a
1707 recipe for disaster. Also, using
1709 when creating a v0.90 array will silently override any
1714 .\"option is given, it is not necessary to list any component-devices in this command.
1715 .\"They can be added later, before a
1719 .\"is given, the apparent size of the smallest drive given is used.
1721 When creating an array within a
1724 can be given either the list of devices to use, or simply the name of
1725 the container. The former case gives control over which devices in
1726 the container will be used for the array. The latter case allows
1728 to automatically choose which devices to use based on how much spare
1731 The General Management options that are valid with
1736 insist on running the array even if some devices look like they might
1741 start the array readonly \(em not supported yet.
1748 .I options... devices...
1751 This usage will allow individual devices in an array to be failed,
1752 removed or added. It is possible to perform multiple operations with
1753 on command. For example:
1755 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
1761 and will then remove it from the array and finally add it back
1762 in as a spare. However only one md array can be affected by a single
1765 When a device is added to an active array, mdadm checks to see if it
1766 has metadata on it which suggests that it was recently a member of the
1767 array. If it does, it tries to "re\-add" the device. If there have
1768 been no changes since the device was removed, or if the array has a
1769 write-intent bitmap which has recorded whatever changes there were,
1770 then the device will immediately become a full member of the array and
1771 those differences recorded in the bitmap will be resolved.
1781 MISC mode includes a number of distinct operations that
1782 operate on distinct devices. The operations are:
1785 The device is examined to see if it is
1786 (1) an active md array, or
1787 (2) a component of an md array.
1788 The information discovered is reported.
1792 The device should be an active md device.
1794 will display a detailed description of the array.
1798 will cause the output to be less detailed and the format to be
1799 suitable for inclusion in
1800 .BR /etc/mdadm.conf .
1803 will normally be 0 unless
1805 failed to get useful information about the device(s); however, if the
1807 option is given, then the exit status will be:
1811 The array is functioning normally.
1814 The array has at least one failed device.
1817 The array has multiple failed devices such that it is unusable.
1820 There was an error while trying to get information about the device.
1824 .B \-\-detail\-platform
1825 Print detail of the platform's RAID capabilities (firmware / hardware
1826 topology). If the metadata is specified with
1830 then the return status will be:
1834 metadata successfully enumerated its platform components on this system
1837 metadata is platform independent
1840 metadata failed to find its platform components on this system
1844 .B \-\-update\-subarray=
1845 If the device is a container and the argument to \-\-update\-subarray
1846 specifies a subarray in the container, then attempt to update the given
1847 superblock field in the subarray. Similar to updating an array in
1848 "assemble" mode, the field to update is selected by
1852 option. Currently only
1858 option updates the subarray name in the metadata, it may not affect the
1859 device node name or the device node symlink until the subarray is
1860 re\-assembled. If updating
1862 would change the UUID of an active subarray this operation is blocked,
1863 and the command will end in an error.
1867 The device should be a component of an md array.
1869 will read the md superblock of the device and display the contents.
1874 is given, then multiple devices that are components of the one array
1875 are grouped together and reported in a single entry suitable
1877 .BR /etc/mdadm.conf .
1881 without listing any devices will cause all devices listed in the
1882 config file to be examined.
1886 The devices should be active md arrays which will be deactivated, as
1887 long as they are not currently in use.
1891 This will fully activate a partially assembled md array.
1895 This will mark an active array as read-only, providing that it is
1896 not currently being used.
1902 array back to being read/write.
1906 For all operations except
1909 will cause the operation to be applied to all arrays listed in
1914 causes all devices listed in the config file to be examined.
1917 .BR \-b ", " \-\-brief
1918 Be less verbose. This is used with
1926 gives an intermediate level of verbosity.
1932 .B mdadm \-\-monitor
1933 .I options... devices...
1938 to periodically poll a number of md arrays and to report on any events
1941 will never exit once it decides that there are arrays to be checked,
1942 so it should normally be run in the background.
1944 As well as reporting events,
1946 may move a spare drive from one array to another if they are in the
1949 and if the destination array has a failed drive but no spares.
1951 If any devices are listed on the command line,
1953 will only monitor those devices. Otherwise all arrays listed in the
1954 configuration file will be monitored. Further, if
1956 is given, then any other md devices that appear in
1958 will also be monitored.
1960 The result of monitoring the arrays is the generation of events.
1961 These events are passed to a separate program (if specified) and may
1962 be mailed to a given E-mail address.
1964 When passing events to a program, the program is run once for each event,
1965 and is given 2 or 3 command-line arguments: the first is the
1966 name of the event (see below), the second is the name of the
1967 md device which is affected, and the third is the name of a related
1968 device if relevant (such as a component device that has failed).
1972 is given, then a program or an E-mail address must be specified on the
1973 command line or in the config file. If neither are available, then
1975 will not monitor anything.
1979 will continue monitoring as long as something was found to monitor. If
1980 no program or email is given, then each event is reported to
1983 The different events are:
1987 .B DeviceDisappeared
1988 An md array which previously was configured appears to no longer be
1989 configured. (syslog priority: Critical)
1993 was told to monitor an array which is RAID0 or Linear, then it will
1995 .B DeviceDisappeared
1996 with the extra information
1998 This is because RAID0 and Linear do not support the device-failed,
1999 hot-spare and resync operations which are monitored.
2003 An md array started reconstruction. (syslog priority: Warning)
2009 is a two-digit number (ie. 05, 48). This indicates that rebuild
2010 has passed that many percent of the total. The events are generated
2011 with fixed increment since 0. Increment size may be specified with
2012 a commandline option (default is 20). (syslog priority: Warning)
2016 An md array that was rebuilding, isn't any more, either because it
2017 finished normally or was aborted. (syslog priority: Warning)
2021 An active component device of an array has been marked as
2022 faulty. (syslog priority: Critical)
2026 A spare component device which was being rebuilt to replace a faulty
2027 device has failed. (syslog priority: Critical)
2031 A spare component device which was being rebuilt to replace a faulty
2032 device has been successfully rebuilt and has been made active.
2033 (syslog priority: Info)
2037 A new md array has been detected in the
2039 file. (syslog priority: Info)
2043 A newly noticed array appears to be degraded. This message is not
2046 notices a drive failure which causes degradation, but only when
2048 notices that an array is degraded when it first sees the array.
2049 (syslog priority: Critical)
2053 A spare drive has been moved from one array in a
2055 to another to allow a failed drive to be replaced.
2056 (syslog priority: Info)
2062 has been told, via the config file, that an array should have a certain
2063 number of spare devices, and
2065 detects that it has fewer than this number when it first sees the
2066 array, it will report a
2069 (syslog priority: Warning)
2073 An array was found at startup, and the
2076 (syslog priority: Info)
2086 cause Email to be sent. All events cause the program to be run.
2087 The program is run with two or three arguments: the event
2088 name, the array device and possibly a second device.
2090 Each event has an associated array device (e.g.
2092 and possibly a second device. For
2097 the second device is the relevant component device.
2100 the second device is the array that the spare was moved from.
2104 to move spares from one array to another, the different arrays need to
2105 be labeled with the same
2107 in the configuration file. The
2109 name can be any string; it is only necessary that different spare
2110 groups use different names.
2114 detects that an array in a spare group has fewer active
2115 devices than necessary for the complete array, and has no spare
2116 devices, it will look for another array in the same spare group that
2117 has a full complement of working drive and a spare. It will then
2118 attempt to remove the spare from the second drive and add it to the
2120 If the removal succeeds but the adding fails, then it is added back to
2124 The GROW mode is used for changing the size or shape of an active
2126 For this to work, the kernel must support the necessary change.
2127 Various types of growth are being added during 2.6 development,
2128 including restructuring a RAID5 array to have more active devices.
2130 Currently the only support available is to
2132 change the "size" attribute
2133 for RAID1, RAID5 and RAID6.
2135 increase or decrease the "raid\-devices" attribute of RAID1, RAID5,
2138 change the chunk-size and layout of RAID5 and RAID6.
2140 convert between RAID1 and RAID5, and between RAID5 and RAID6.
2142 add a write-intent bitmap to any array which supports these bitmaps, or
2143 remove a write-intent bitmap from such an array.
2146 GROW mode is not currently supported for
2148 or arrays inside containers.
2151 Normally when an array is built the "size" it taken from the smallest
2152 of the drives. If all the small drives in an arrays are, one at a
2153 time, removed and replaced with larger drives, then you could have an
2154 array of large drives with only a small amount used. In this
2155 situation, changing the "size" with "GROW" mode will allow the extra
2156 space to start being used. If the size is increased in this way, a
2157 "resync" process will start to make sure the new parts of the array
2160 Note that when an array changes size, any filesystem that may be
2161 stored in the array will not automatically grow to use the space. The
2162 filesystem will need to be explicitly told to use the extra space.
2164 Also the size of an array cannot be changed while it has an active
2165 bitmap. If an array has a bitmap, it must be removed before the size
2166 can be changed. Once the change it complete a new bitmap can be created.
2168 .SS RAID\-DEVICES CHANGES
2170 A RAID1 array can work with any number of devices from 1 upwards
2171 (though 1 is not very useful). There may be times which you want to
2172 increase or decrease the number of active devices. Note that this is
2173 different to hot-add or hot-remove which changes the number of
2176 When reducing the number of devices in a RAID1 array, the slots which
2177 are to be removed from the array must already be vacant. That is, the
2178 devices which were in those slots must be failed and removed.
2180 When the number of devices is increased, any hot spares that are
2181 present will be activated immediately.
2183 Changing the number of active devices in a RAID5 or RAID6 is much more
2184 effort. Every block in the array will need to be read and written
2185 back to a new location. From 2.6.17, the Linux Kernel is able to
2186 increase the number of devices in a RAID5 safely, including restarting
2187 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2188 increase or decrease the number of devices in a RAID5 or RAID6.
2190 When decreasing the number of devices, the size of the array will also
2191 decrease. If there was data in the array, it could get destroyed and
2192 this is not reversible. To help prevent accidents,
2194 requires that the size of the array be decreased first with
2195 .BR "mdadm --grow --array-size" .
2196 This is a reversible change which simply makes the end of the array
2197 inaccessible. The integrity of any data can then be checked before
2198 the non-reversible reduction in the number of devices is request.
2200 When relocating the first few stripes on a RAID5 or RAID6, it is not
2201 possible to keep the data on disk completely consistent and
2202 crash-proof. To provide the required safety, mdadm disables writes to
2203 the array while this "critical section" is reshaped, and takes a
2204 backup of the data that is in that section. For grows, this backup may be
2205 stored in any spare devices that the array has, however it can also be
2206 stored in a separate file specified with the
2208 option, and is required to be specified for shrinks, RAID level
2209 changes and layout changes. If this option is used, and the system
2210 does crash during the critical period, the same file must be passed to
2212 to restore the backup and reassemble the array. When shrinking rather
2213 than growing the array, the reshape is done from the end towards the
2214 beginning, so the "critical section" is at the end of the reshape.
2218 Changing the RAID level of any array happens instantaneously. However
2219 in the RAID5 to RAID6 case this requires a non-standard layout of the
2220 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2221 required before the change can be accomplished. So while the level
2222 change is instant, the accompanying layout change can take quite a
2225 is required. If the array is not simultaneously being grown or
2226 shrunk, so that the array size will remain the same - for example,
2227 reshaping a 3-drive RAID5 into a 4-drive RAID6 - the backup file will
2228 be used not just for a "cricital section" but throughout the reshape
2229 operation, as described below under LAYOUT CHANGES.
2231 .SS CHUNK-SIZE AND LAYOUT CHANGES
2233 Changing the chunk-size of layout without also changing the number of
2234 devices as the same time will involve re-writing all blocks in-place.
2235 To ensure against data loss in the case of a crash, a
2237 must be provided for these changes. Small sections of the array will
2238 be copied to the backup file while they are being rearranged. This
2239 means that all the data is copied twice, once to the backup and once
2240 to the new layout on the array, so this type of reshape will go very
2243 If the reshape is interrupted for any reason, this backup file must be
2245 .B "mdadm --assemble"
2246 so the array can be reassembled. Consequently the file cannot be
2247 stored on the device being reshaped.
2252 A write-intent bitmap can be added to, or removed from, an active
2253 array. Either internal bitmaps, or bitmaps stored in a separate file,
2254 can be added. Note that if you add a bitmap stored in a file which is
2255 in a filesystem that is on the RAID array being affected, the system
2256 will deadlock. The bitmap must be on a separate filesystem.
2258 .SH INCREMENTAL MODE
2262 .B mdadm \-\-incremental
2268 .B mdadm \-\-incremental \-\-fail
2272 .B mdadm \-\-incremental \-\-rebuild\-map
2275 .B mdadm \-\-incremental \-\-run \-\-scan
2278 This mode is designed to be used in conjunction with a device
2279 discovery system. As devices are found in a system, they can be
2281 .B "mdadm \-\-incremental"
2282 to be conditionally added to an appropriate array.
2284 Conversely, it can also be used with the
2286 flag to do just the opposite and find whatever array a particular device
2287 is part of and remove the device from that array.
2289 If the device passed is a
2291 device created by a previous call to
2293 then rather than trying to add that device to an array, all the arrays
2294 described by the metadata of the container will be started.
2297 performs a number of tests to determine if the device is part of an
2298 array, and which array it should be part of. If an appropriate array
2299 is found, or can be created,
2301 adds the device to the array and conditionally starts the array.
2305 will only add devices to an array which were previously working
2306 (active or spare) parts of that array. It does not currently support
2307 automatic inclusion of a new drive as a spare in some array.
2311 makes are as follow:
2313 Is the device permitted by
2315 That is, is it listed in a
2317 line in that file. If
2319 is absent then the default it to allow any device. Similar if
2321 contains the special word
2323 then any device is allowed. Otherwise the device name given to
2325 must match one of the names or patterns in a
2330 Does the device have a valid md superblock. If a specific metadata
2331 version is request with
2335 then only that style of metadata is accepted, otherwise
2337 finds any known version of metadata. If no
2339 metadata is found, the device is rejected.
2343 Does the metadata match an expected array?
2344 The metadata can match in two ways. Either there is an array listed
2347 which identifies the array (either by UUID, by name, by device list,
2348 or by minor-number), or the array was created with a
2354 or on the command line.
2357 is not able to positively identify the array as belonging to the
2358 current host, the device will be rejected.
2362 keeps a list of arrays that it has partially assembled in
2363 .B /var/run/mdadm/map
2365 .B /var/run/mdadm.map
2366 if the directory doesn't exist. Or maybe even
2367 .BR /dev/.mdadm.map ).
2368 If no array exists which matches
2369 the metadata on the new device,
2371 must choose a device name and unit number. It does this based on any
2374 or any name information stored in the metadata. If this name
2375 suggests a unit number, that number will be used, otherwise a free
2376 unit number will be chosen. Normally
2378 will prefer to create a partitionable array, however if the
2382 suggests that a non-partitionable array is preferred, that will be
2385 If the array is not found in the config file and its metadata does not
2386 identify it as belonging to the "homehost", then
2388 will choose a name for the array which is certain not to conflict with
2389 any array which does belong to this host. It does this be adding an
2390 underscore and a small number to the name preferred by the metadata.
2392 Once an appropriate array is found or created and the device is added,
2394 must decide if the array is ready to be started. It will
2395 normally compare the number of available (non-spare) devices to the
2396 number of devices that the metadata suggests need to be active. If
2397 there are at least that many, the array will be started. This means
2398 that if any devices are missing the array will not be restarted.
2404 in which case the array will be run as soon as there are enough
2405 devices present for the data to be accessible. For a RAID1, that
2406 means one device will start the array. For a clean RAID5, the array
2407 will be started as soon as all but one drive is present.
2409 Note that neither of these approaches is really ideal. If it can
2410 be known that all device discovery has completed, then
2414 can be run which will try to start all arrays that are being
2415 incrementally assembled. They are started in "read-auto" mode in
2416 which they are read-only until the first write request. This means
2417 that no metadata updates are made and no attempt at resync or recovery
2418 happens. Further devices that are found before the first write can
2419 still be added safely.
2422 This section describes environment variables that affect how mdadm
2427 Setting this value to 1 will prevent mdadm from automatically launching
2428 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2434 does not create any device nodes in /dev, but leaves that task to
2438 appears not to be configured, or if this environment variable is set
2441 will create and devices that are needed.
2445 .B " mdadm \-\-query /dev/name-of-device"
2447 This will find out if a given device is a RAID array, or is part of
2448 one, and will provide brief information about the device.
2450 .B " mdadm \-\-assemble \-\-scan"
2452 This will assemble and start all arrays listed in the standard config
2453 file. This command will typically go in a system startup file.
2455 .B " mdadm \-\-stop \-\-scan"
2457 This will shut down all arrays that can be shut down (i.e. are not
2458 currently in use). This will typically go in a system shutdown script.
2460 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2462 If (and only if) there is an Email address or program given in the
2463 standard config file, then
2464 monitor the status of all arrays listed in that file by
2465 polling them ever 2 minutes.
2467 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2469 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2472 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2474 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2476 This will create a prototype config file that describes currently
2477 active arrays that are known to be made from partitions of IDE or SCSI drives.
2478 This file should be reviewed before being used as it may
2479 contain unwanted detail.
2481 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2483 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
2485 This will find arrays which could be assembled from existing IDE and
2486 SCSI whole drives (not partitions), and store the information in the
2487 format of a config file.
2488 This file is very likely to contain unwanted detail, particularly
2491 entries. It should be reviewed and edited before being used as an
2494 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
2496 .B " mdadm \-Ebsc partitions"
2498 Create a list of devices by reading
2499 .BR /proc/partitions ,
2500 scan these for RAID superblocks, and printout a brief listing of all
2503 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
2505 Scan all partitions and devices listed in
2506 .BR /proc/partitions
2509 out of all such devices with a RAID superblock with a minor number of 0.
2511 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /var/run/mdadm"
2513 If config file contains a mail address or alert program, run mdadm in
2514 the background in monitor mode monitoring all md devices. Also write
2515 pid of mdadm daemon to
2516 .BR /var/run/mdadm .
2518 .B " mdadm \-Iq /dev/somedevice"
2520 Try to incorporate newly discovered device into some array as
2523 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
2525 Rebuild the array map from any current arrays, and then start any that
2528 .B " mdadm /dev/md4 --fail detached --remove detached"
2530 Any devices which are components of /dev/md4 will be marked as faulty
2531 and then remove from the array.
2533 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4
2537 which is currently a RAID5 array will be converted to RAID6. There
2538 should normally already be a spare drive attached to the array as a
2539 RAID6 needs one more drive than a matching RAID5.
2541 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
2543 Create a DDF array over 6 devices.
2545 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
2547 Create a RAID5 array over any 3 devices in the given DDF set. Use
2548 only 30 gigabytes of each device.
2550 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
2552 Assemble a pre-exist ddf array.
2554 .B " mdadm -I /dev/md/ddf1"
2556 Assemble all arrays contained in the ddf array, assigning names as
2559 .B " mdadm \-\-create \-\-help"
2561 Provide help about the Create mode.
2563 .B " mdadm \-\-config \-\-help"
2565 Provide help about the format of the config file.
2567 .B " mdadm \-\-help"
2569 Provide general help.
2579 lists all active md devices with information about them.
2581 uses this to find arrays when
2583 is given in Misc mode, and to monitor array reconstruction
2588 The config file lists which devices may be scanned to see if
2589 they contain MD super block, and gives identifying information
2590 (e.g. UUID) about known MD arrays. See
2594 .SS /var/run/mdadm/map
2597 mode is used, this file gets a list of arrays currently being created.
2600 does not exist as a directory, then
2601 .B /var/run/mdadm.map
2604 is not available (as may be the case during early boot),
2606 is used on the basis that
2608 is usually available very early in boot.
2613 understand two sorts of names for array devices.
2615 The first is the so-called 'standard' format name, which matches the
2616 names used by the kernel and which appear in
2619 The second sort can be freely chosen, but must reside in
2621 When giving a device name to
2623 to create or assemble an array, either full path name such as
2627 can be given, or just the suffix of the second sort of name, such as
2633 chooses device names during auto-assembly or incremental assembly, it
2634 will sometimes add a small sequence number to the end of the name to
2635 avoid conflicted between multiple arrays that have the same name. If
2637 can reasonably determine that the array really is meant for this host,
2638 either by a hostname in the metadata, or by the presence of the array
2639 in /etc/mdadm.conf, then it will leave off the suffix if possible.
2640 Also if the homehost is specified as
2643 will only use a suffix if a different array of the same name already
2644 exists or is listed in the config file.
2646 The standard names for non-partitioned arrays (the only sort of md
2647 array available in 2.4 and earlier) are of the form
2651 where NN is a number.
2652 The standard names for partitionable arrays (as available from 2.6
2653 onwards) are of the form
2657 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2659 From kernel version, 2.6.28 the "non-partitioned array" can actually
2660 be partitioned. So the "md_dNN" names are no longer needed, and
2661 partitions such as "/dev/mdNNpXX" are possible.
2665 was previously known as
2669 is completely separate from the
2671 package, and does not use the
2673 configuration file at all.
2676 For further information on mdadm usage, MD and the various levels of
2679 .B http://linux\-raid.osdl.org/
2681 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
2683 .\"for new releases of the RAID driver check out:
2686 .\".UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2687 .\"ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2692 .\".UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2693 .\"http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2696 The latest version of
2698 should always be available from
2700 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/