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.
326 .ie '{DEFAULT_METADATA}'0.90'
327 .IP "1, 1.0, 1.1, 1.2"
329 .IP "1, 1.0, 1.1, 1.2 default"
331 Use the new version-1 format superblock. This has few restrictions.
332 The different sub-versions store the superblock at different locations
333 on the device, either at the end (for 1.0), at the start (for 1.1) or
334 4K from the start (for 1.2). "1" is equivalent to "1.0".
335 'if '{DEFAULT_METADATA}'1.2' "default" is equivalent to "1.2".
337 Use the "Industry Standard" DDF (Disk Data Format) format defined by
339 When creating a DDF array a
341 will be created, and normal arrays can be created in that container.
343 Use the Intel(R) Matrix Storage Manager metadata format. This creates a
345 which is managed in a similar manner to DDF, and is supported by an
346 option-rom on some platforms:
348 .B http://www.intel.com/design/chipsets/matrixstorage_sb.htm
354 This will override any
356 setting in the config file and provides the identity of the host which
357 should be considered the home for any arrays.
359 When creating an array, the
361 will be recorded in the metadata. For version-1 superblocks, it will
362 be prefixed to the array name. For version-0.90 superblocks, part of
363 the SHA1 hash of the hostname will be stored in the later half of the
366 When reporting information about an array, any array which is tagged
367 for the given homehost will be reported as such.
369 When using Auto-Assemble, only arrays tagged for the given homehost
370 will be allowed to use 'local' names (i.e. not ending in '_' followed
371 by a digit string). See below under
372 .BR "Auto Assembly" .
374 .SH For create, build, or grow:
377 .BR \-n ", " \-\-raid\-devices=
378 Specify the number of active devices in the array. This, plus the
379 number of spare devices (see below) must equal the number of
381 (including "\fBmissing\fP" devices)
382 that are listed on the command line for
384 Setting a value of 1 is probably
385 a mistake and so requires that
387 be specified first. A value of 1 will then be allowed for linear,
388 multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6.
390 This number can only be changed using
392 for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide
393 the necessary support.
396 .BR \-x ", " \-\-spare\-devices=
397 Specify the number of spare (eXtra) devices in the initial array.
398 Spares can also be added
399 and removed later. The number of component devices listed
400 on the command line must equal the number of RAID devices plus the
401 number of spare devices.
404 .BR \-z ", " \-\-size=
405 Amount (in Kibibytes) of space to use from each drive in RAID levels 1/4/5/6.
406 This must be a multiple of the chunk size, and must leave about 128Kb
407 of space at the end of the drive for the RAID superblock.
408 If this is not specified
409 (as it normally is not) the smallest drive (or partition) sets the
410 size, though if there is a variance among the drives of greater than 1%, a warning is
413 A suffix of 'M' or 'G' can be given to indicate Megabytes or
414 Gigabytes respectively.
416 This value can be set with
418 for RAID level 1/4/5/6. If the array was created with a size smaller
419 than the currently active drives, the extra space can be accessed
422 The size can be given as
424 which means to choose the largest size that fits on all current drives.
426 This value can not be used with
428 metadata such as DDF and IMSM.
431 .BR \-Z ", " \-\-array-size=
432 This is only meaningful with
434 and its effect is not persistent: when the array is stopped and
435 restarted the default array size will be restored.
437 Setting the array-size causes the array to appear smaller to programs
438 that access the data. This is particularly needed before reshaping an
439 array so that it will be smaller. As the reshape is not reversible,
440 but setting the size with
442 is, it is required that the array size is reduced as appropriate
443 before the number of devices in the array is reduced.
445 A suffix of 'M' or 'G' can be given to indicate Megabytes or
446 Gigabytes respectively.
449 restores the apparent size of the array to be whatever the real
450 amount of available space is.
453 .BR \-c ", " \-\-chunk=
454 Specify chunk size of kibibytes. The default when creating an
455 array is 512KB. To ensure compatibility with earlier versions, the
456 default when Building and array with no persistent metadata is 64KB.
457 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
459 A suffix of 'M' or 'G' can be given to indicate Megabytes or
460 Gigabytes respectively.
464 Specify rounding factor for a Linear array. The size of each
465 component will be rounded down to a multiple of this size.
466 This is a synonym for
468 but highlights the different meaning for Linear as compared to other
469 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
470 use, and is 0K (i.e. no rounding) in later kernels.
473 .BR \-l ", " \-\-level=
474 Set RAID level. When used with
476 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
477 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
478 Obviously some of these are synonymous.
482 metadata type is requested, only the
484 level is permitted, and it does not need to be explicitly given.
488 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
492 to change the RAID level in some cases. See LEVEL CHANGES below.
495 .BR \-p ", " \-\-layout=
496 This option configures the fine details of data layout for RAID5, RAID6,
497 and RAID10 arrays, and controls the failure modes for
500 The layout of the RAID5 parity block can be one of
501 .BR left\-asymmetric ,
502 .BR left\-symmetric ,
503 .BR right\-asymmetric ,
504 .BR right\-symmetric ,
505 .BR la ", " ra ", " ls ", " rs .
507 .BR left\-symmetric .
509 It is also possibly to cause RAID5 to use a RAID4-like layout by
515 Finally for RAID5 there are DDF\-compatible layouts,
516 .BR ddf\-zero\-restart ,
517 .BR ddf\-N\-restart ,
519 .BR ddf\-N\-continue .
521 These same layouts are available for RAID6. There are also 4 layouts
522 that will provide an intermediate stage for converting between RAID5
523 and RAID6. These provide a layout which is identical to the
524 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
525 syndrome (the second 'parity' block used by RAID6) on the last device.
527 .BR left\-symmetric\-6 ,
528 .BR right\-symmetric\-6 ,
529 .BR left\-asymmetric\-6 ,
530 .BR right\-asymmetric\-6 ,
532 .BR parity\-first\-6 .
534 When setting the failure mode for level
537 .BR write\-transient ", " wt ,
538 .BR read\-transient ", " rt ,
539 .BR write\-persistent ", " wp ,
540 .BR read\-persistent ", " rp ,
542 .BR read\-fixable ", " rf ,
543 .BR clear ", " flush ", " none .
545 Each failure mode can be followed by a number, which is used as a period
546 between fault generation. Without a number, the fault is generated
547 once on the first relevant request. With a number, the fault will be
548 generated after that many requests, and will continue to be generated
549 every time the period elapses.
551 Multiple failure modes can be current simultaneously by using the
553 option to set subsequent failure modes.
555 "clear" or "none" will remove any pending or periodic failure modes,
556 and "flush" will clear any persistent faults.
558 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
559 by a small number. The default is 'n2'. The supported options are:
562 signals 'near' copies. Multiple copies of one data block are at
563 similar offsets in different devices.
566 signals 'offset' copies. Rather than the chunks being duplicated
567 within a stripe, whole stripes are duplicated but are rotated by one
568 device so duplicate blocks are on different devices. Thus subsequent
569 copies of a block are in the next drive, and are one chunk further
574 (multiple copies have very different offsets).
575 See md(4) for more detail about 'near', 'offset', and 'far'.
577 The number is the number of copies of each datablock. 2 is normal, 3
578 can be useful. This number can be at most equal to the number of
579 devices in the array. It does not need to divide evenly into that
580 number (e.g. it is perfectly legal to have an 'n2' layout for an array
581 with an odd number of devices).
583 When an array is converted between RAID5 and RAID6 an intermediate
584 RAID6 layout is used in which the second parity block (Q) is always on
585 the last device. To convert a RAID5 to RAID6 and leave it in this new
586 layout (which does not require re-striping) use
587 .BR \-\-layout=preserve .
588 This will try to avoid any restriping.
590 The converse of this is
591 .B \-\-layout=normalise
592 which will change a non-standard RAID6 layout into a more standard
599 (thus explaining the p of
603 .BR \-b ", " \-\-bitmap=
604 Specify a file to store a write-intent bitmap in. The file should not
607 is also given. The same file should be provided
608 when assembling the array. If the word
610 is given, then the bitmap is stored with the metadata on the array,
611 and so is replicated on all devices. If the word
615 mode, then any bitmap that is present is removed.
617 To help catch typing errors, the filename must contain at least one
618 slash ('/') if it is a real file (not 'internal' or 'none').
620 Note: external bitmaps are only known to work on ext2 and ext3.
621 Storing bitmap files on other filesystems may result in serious problems.
624 .BR \-\-bitmap\-chunk=
625 Set the chunksize of the bitmap. Each bit corresponds to that many
626 Kilobytes of storage.
627 When using a file based bitmap, the default is to use the smallest
628 size that is at-least 4 and requires no more than 2^21 chunks.
631 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
632 fit the bitmap into the available space.
634 A suffix of 'M' or 'G' can be given to indicate Megabytes or
635 Gigabytes respectively.
638 .BR \-W ", " \-\-write\-mostly
639 subsequent devices listed in a
644 command will be flagged as 'write-mostly'. This is valid for RAID1
645 only and means that the 'md' driver will avoid reading from these
646 devices if at all possible. This can be useful if mirroring over a
650 .BR \-\-write\-behind=
651 Specify that write-behind mode should be enabled (valid for RAID1
652 only). If an argument is specified, it will set the maximum number
653 of outstanding writes allowed. The default value is 256.
654 A write-intent bitmap is required in order to use write-behind
655 mode, and write-behind is only attempted on drives marked as
659 .BR \-\-assume\-clean
662 that the array pre-existed and is known to be clean. It can be useful
663 when trying to recover from a major failure as you can be sure that no
664 data will be affected unless you actually write to the array. It can
665 also be used when creating a RAID1 or RAID10 if you want to avoid the
666 initial resync, however this practice \(em while normally safe \(em is not
667 recommended. Use this only if you really know what you are doing.
669 When the devices that will be part of a new array were filled
670 with zeros before creation the operator knows the array is
671 actually clean. If that is the case, such as after running
672 badblocks, this argument can be used to tell mdadm the
673 facts the operator knows.
676 .BR \-\-backup\-file=
679 is used to increase the number of
680 raid-devices in a RAID5 if there are no spare devices available.
681 See the GROW MODE section below on RAID\-DEVICES CHANGES. The file
682 should be stored on a separate device, not on the RAID array being
686 .BR \-N ", " \-\-name=
689 for the array. This is currently only effective when creating an
690 array with a version-1 superblock, or an array in a DDF container.
691 The name is a simple textual string that can be used to identify array
692 components when assembling. If name is needed but not specified, it
693 is taken from the basename of the device that is being created.
705 run the array, even if some of the components
706 appear to be active in another array or filesystem. Normally
708 will ask for confirmation before including such components in an
709 array. This option causes that question to be suppressed.
712 .BR \-f ", " \-\-force
715 accept the geometry and layout specified without question. Normally
717 will not allow creation of an array with only one device, and will try
718 to create a RAID5 array with one missing drive (as this makes the
719 initial resync work faster). With
722 will not try to be so clever.
725 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
726 Instruct mdadm how to create the device file if needed, possibly allocating
727 an unused minor number. "md" causes a non-partitionable array
728 to be used (though since Linux 2.6.28, these array devices are in fact
729 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
730 later) to be used. "yes" requires the named md device to have
731 a 'standard' format, and the type and minor number will be determined
732 from this. With mdadm 3.0, device creation is normally left up to
734 so this option is unlikely to be needed.
735 See DEVICE NAMES below.
737 The argument can also come immediately after
742 is not given on the command line or in the config file, then
748 is also given, then any
750 entries in the config file will override the
752 instruction given on the command line.
754 For partitionable arrays,
756 will create the device file for the whole array and for the first 4
757 partitions. A different number of partitions can be specified at the
758 end of this option (e.g.
760 If the device name ends with a digit, the partition names add a 'p',
762 .IR /dev/md/home1p3 .
763 If there is no trailing digit, then the partition names just have a
765 .IR /dev/md/scratch3 .
767 If the md device name is in a 'standard' format as described in DEVICE
768 NAMES, then it will be created, if necessary, with the appropriate
769 device number based on that name. If the device name is not in one of these
770 formats, then a unused device number will be allocated. The device
771 number will be considered unused if there is no active array for that
772 number, and there is no entry in /dev for that number and with a
773 non-standard name. Names that are not in 'standard' format are only
774 allowed in "/dev/md/".
778 .\".BR \-\-symlink = no
783 .\"to create devices in
785 .\"it will also create symlinks from
787 .\"with names starting with
793 .\"to suppress this, or
794 .\".B \-\-symlink=yes
795 .\"to enforce this even if it is suppressing
803 .BR \-u ", " \-\-uuid=
804 uuid of array to assemble. Devices which don't have this uuid are
808 .BR \-m ", " \-\-super\-minor=
809 Minor number of device that array was created for. Devices which
810 don't have this minor number are excluded. If you create an array as
811 /dev/md1, then all superblocks will contain the minor number 1, even if
812 the array is later assembled as /dev/md2.
814 Giving the literal word "dev" for
818 to use the minor number of the md device that is being assembled.
821 .B \-\-super\-minor=dev
822 will look for super blocks with a minor number of 0.
825 is only relevant for v0.90 metadata, and should not normally be used.
831 .BR \-N ", " \-\-name=
832 Specify the name of the array to assemble. This must be the name
833 that was specified when creating the array. It must either match
834 the name stored in the superblock exactly, or it must match
837 prefixed to the start of the given name.
840 .BR \-f ", " \-\-force
841 Assemble the array even if the metadata on some devices appears to be
844 cannot find enough working devices to start the array, but can find
845 some devices that are recorded as having failed, then it will mark
846 those devices as working so that the array can be started.
847 An array which requires
849 to be started may contain data corruption. Use it carefully.
853 Attempt to start the array even if fewer drives were given than were
854 present last time the array was active. Normally if not all the
855 expected drives are found and
857 is not used, then the array will be assembled but not started.
860 an attempt will be made to start it anyway.
864 This is the reverse of
866 in that it inhibits the startup of array unless all expected drives
867 are present. This is only needed with
869 and can be used if the physical connections to devices are
870 not as reliable as you would like.
873 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
874 See this option under Create and Build options.
877 .BR \-b ", " \-\-bitmap=
878 Specify the bitmap file that was given when the array was created. If
881 bitmap, there is no need to specify this when assembling the array.
884 .BR \-\-backup\-file=
887 was used to grow the number of raid-devices in a RAID5, and the system
888 crashed during the critical section, then the same
892 to allow possibly corrupted data to be restored.
895 .BR \-U ", " \-\-update=
896 Update the superblock on each device while assembling the array. The
897 argument given to this flag can be one of
912 option will adjust the superblock of an array what was created on a Sparc
913 machine running a patched 2.2 Linux kernel. This kernel got the
914 alignment of part of the superblock wrong. You can use the
915 .B "\-\-examine \-\-sparc2.2"
918 to see what effect this would have.
922 option will update the
924 field on each superblock to match the minor number of the array being
926 This can be useful if
928 reports a different "Preferred Minor" to
930 In some cases this update will be performed automatically
931 by the kernel driver. In particular the update happens automatically
932 at the first write to an array with redundancy (RAID level 1 or
933 greater) on a 2.6 (or later) kernel.
937 option will change the uuid of the array. If a UUID is given with the
939 option that UUID will be used as a new UUID and will
941 be used to help identify the devices in the array.
944 is given, a random UUID is chosen.
948 option will change the
950 of the array as stored in the superblock. This is only supported for
951 version-1 superblocks.
955 option will change the
957 as recorded in the superblock. For version-0 superblocks, this is the
958 same as updating the UUID.
959 For version-1 superblocks, this involves updating the name.
963 option will cause the array to be marked
965 meaning that any redundancy in the array (e.g. parity for RAID5,
966 copies for RAID1) may be incorrect. This will cause the RAID system
967 to perform a "resync" pass to make sure that all redundant information
972 option allows arrays to be moved between machines with different
974 When assembling such an array for the first time after a move, giving
975 .B "\-\-update=byteorder"
978 to expect superblocks to have their byteorder reversed, and will
979 correct that order before assembling the array. This is only valid
980 with original (Version 0.90) superblocks.
984 option will correct the summaries in the superblock. That is the
985 counts of total, working, active, failed, and spare devices.
989 option will rarely be of use. It applies to version 1.1 and 1.2 metadata
990 only (where the metadata is at the start of the device) and is only
991 useful when the component device has changed size (typically become
992 larger). The version 1 metadata records the amount of the device that
993 can be used to store data, so if a device in a version 1.1 or 1.2
994 array becomes larger, the metadata will still be visible, but the
995 extra space will not. In this case it might be useful to assemble the
997 .BR \-\-update=devicesize .
1000 to determine the maximum usable amount of space on each device and
1001 update the relevant field in the metadata.
1005 option can be used when an array has an internal bitmap which is
1006 corrupt in some way so that assembling the array normally fails. It
1007 will cause any internal bitmap to be ignored.
1011 .B \-\-auto\-update\-homehost
1012 This flag is only meaningful with auto-assembly (see discussion below).
1013 In that situation, if no suitable arrays are found for this homehost,
1015 will rescan for any arrays at all and will assemble them and update the
1016 homehost to match the current host.
1019 .SH For Manage mode:
1022 .BR \-t ", " \-\-test
1023 Unless a more serious error occurred,
1025 will exit with a status of 2 if no changes were made to the array and
1026 0 if at least one change was made.
1027 This can be useful when an indirect specifier such as
1032 is used in requesting an operation on the array.
1034 will report failure if these specifiers didn't find any match.
1037 .BR \-a ", " \-\-add
1038 hot-add listed devices.
1039 If a device appears to have recently been part of the array
1040 (possibly it failed or was removed) the device is re-added as describe
1042 If that fails or the device was never part of the array, the device is
1043 added as a hot-spare.
1044 If the array is degraded, it will immediately start to rebuild data
1047 Note that this and the following options are only meaningful on array
1048 with redundancy. They don't apply to RAID0 or Linear.
1052 re\-add a device that was previous removed from an array.
1053 If the metadata on the device reports that it is a member of the
1054 array, and the slot that it used is still vacant, then the device will
1055 be added back to the array in the same position. This will normally
1056 cause the data for that device to be recovered. However based on the
1057 event count on the device, the recovery may only require sections that
1058 are flagged a write-intent bitmap to be recovered or may not require
1059 any recovery at all.
1061 When used on an array that has no metadata (i.e. it was built with
1063 it will be assumed that bitmap-based recovery is enough to make the
1064 device fully consistent with the array.
1066 If the device name given is
1068 then mdadm will try to find any device that looks like it should be
1069 part of the array but isn't and will try to re\-add all such devices.
1072 .BR \-r ", " \-\-remove
1073 remove listed devices. They must not be active. i.e. they should
1074 be failed or spare devices. As well as the name of a device file
1083 The first causes all failed device to be removed. The second causes
1084 any device which is no longer connected to the system (i.e an 'open'
1087 to be removed. This will only succeed for devices that are spares or
1088 have already been marked as failed.
1091 .BR \-f ", " \-\-fail
1092 mark listed devices as faulty.
1093 As well as the name of a device file, the word
1095 can be given. This will cause any device that has been detached from
1096 the system to be marked as failed. It can then be removed.
1104 .BR \-\-write\-mostly
1105 Subsequent devices that are added or re\-added will have the 'write-mostly'
1106 flag set. This is only valid for RAID1 and means that the 'md' driver
1107 will avoid reading from these devices if possible.
1110 Subsequent devices that are added or re\-added will have the 'write-mostly'
1114 Each of these options requires that the first device listed is the array
1115 to be acted upon, and the remainder are component devices to be added,
1116 removed, marked as faulty, etc. Several different operations can be
1117 specified for different devices, e.g.
1119 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1121 Each operation applies to all devices listed until the next
1124 If an array is using a write-intent bitmap, then devices which have
1125 been removed can be re\-added in a way that avoids a full
1126 reconstruction but instead just updates the blocks that have changed
1127 since the device was removed. For arrays with persistent metadata
1128 (superblocks) this is done automatically. For arrays created with
1130 mdadm needs to be told that this device we removed recently with
1133 Devices can only be removed from an array if they are not in active
1134 use, i.e. that must be spares or failed devices. To remove an active
1135 device, it must first be marked as
1141 .BR \-Q ", " \-\-query
1142 Examine a device to see
1143 (1) if it is an md device and (2) if it is a component of an md
1145 Information about what is discovered is presented.
1148 .BR \-D ", " \-\-detail
1149 Print details of one or more md devices.
1152 .BR \-\-detail\-platform
1153 Print details of the platform's RAID capabilities (firmware / hardware
1154 topology) for a given metadata format.
1157 .BR \-Y ", " \-\-export
1162 output will be formatted as
1164 pairs for easy import into the environment.
1167 .BR \-E ", " \-\-examine
1168 Print contents of the metadata stored on the named device(s).
1169 Note the contrast between
1174 applies to devices which are components of an array, while
1176 applies to a whole array which is currently active.
1179 If an array was created on a SPARC machine with a 2.2 Linux kernel
1180 patched with RAID support, the superblock will have been created
1181 incorrectly, or at least incompatibly with 2.4 and later kernels.
1186 will fix the superblock before displaying it. If this appears to do
1187 the right thing, then the array can be successfully assembled using
1188 .BR "\-\-assemble \-\-update=sparc2.2" .
1191 .BR \-X ", " \-\-examine\-bitmap
1192 Report information about a bitmap file.
1193 The argument is either an external bitmap file or an array component
1194 in case of an internal bitmap. Note that running this on an array
1197 does not report the bitmap for that array.
1200 .BR \-R ", " \-\-run
1201 start a partially assembled array. If
1203 did not find enough devices to fully start the array, it might leaving
1204 it partially assembled. If you wish, you can then use
1206 to start the array in degraded mode.
1209 .BR \-S ", " \-\-stop
1210 deactivate array, releasing all resources.
1213 .BR \-o ", " \-\-readonly
1214 mark array as readonly.
1217 .BR \-w ", " \-\-readwrite
1218 mark array as readwrite.
1221 .B \-\-zero\-superblock
1222 If the device contains a valid md superblock, the block is
1223 overwritten with zeros. With
1225 the block where the superblock would be is overwritten even if it
1226 doesn't appear to be valid.
1229 .B \-\-kill\-subarray=
1230 If the device is a container and the argument to \-\-kill\-subarray
1231 specifies an inactive subarray in the container, then the subarray is
1232 deleted. Deleting all subarrays will leave an 'empty-container' or
1233 spare superblock on the drives. See \-\-zero\-superblock for completely
1234 removing a superblock. Note that some formats depend on the subarray
1235 index for generating a UUID, this command will fail if it would change
1236 the UUID of an active subarray.
1239 .B \-\-update\-subarray=
1240 If the device is a container and the argument to \-\-update\-subarray
1241 specifies a subarray in the container, then attempt to update the given
1242 superblock field in the subarray. See below in
1247 .BR \-t ", " \-\-test
1252 is set to reflect the status of the device. See below in
1257 .BR \-W ", " \-\-wait
1258 For each md device given, wait for any resync, recovery, or reshape
1259 activity to finish before returning.
1261 will return with success if it actually waited for every device
1262 listed, otherwise it will return failure.
1266 For each md device given, or each device in /proc/mdstat if
1268 is given, arrange for the array to be marked clean as soon as possible.
1270 will return with success if the array uses external metadata and we
1271 successfully waited. For native arrays this returns immediately as the
1272 kernel handles dirty-clean transitions at shutdown. No action is taken
1273 if safe-mode handling is disabled.
1275 .SH For Incremental Assembly mode:
1277 .BR \-\-rebuild\-map ", " \-r
1278 Rebuild the map file
1279 .RB ( /var/run/mdadm/map )
1282 uses to help track which arrays are currently being assembled.
1285 .BR \-\-run ", " \-R
1286 Run any array assembled as soon as a minimal number of devices are
1287 available, rather than waiting until all expected devices are present.
1290 .BR \-\-scan ", " \-s
1291 Only meaningful with
1295 file for arrays that are being incrementally assembled and will try to
1296 start any that are not already started. If any such array is listed
1299 as requiring an external bitmap, that bitmap will be attached first.
1302 .BR \-\-fail ", " \-f
1303 This allows the hot-plug system to remove devices that have fully disappeared
1304 from the kernel. It will first fail and then remove the device from any
1305 array it belongs to.
1306 The device name given should be a kernel device name such as "sda",
1310 .SH For Monitor mode:
1312 .BR \-m ", " \-\-mail
1313 Give a mail address to send alerts to.
1316 .BR \-p ", " \-\-program ", " \-\-alert
1317 Give a program to be run whenever an event is detected.
1320 .BR \-y ", " \-\-syslog
1321 Cause all events to be reported through 'syslog'. The messages have
1322 facility of 'daemon' and varying priorities.
1325 .BR \-d ", " \-\-delay
1326 Give a delay in seconds.
1328 polls the md arrays and then waits this many seconds before polling
1329 again. The default is 60 seconds. Since 2.6.16, there is no need to
1330 reduce this as the kernel alerts
1332 immediately when there is any change.
1335 .BR \-r ", " \-\-increment
1336 Give a percentage increment.
1338 will generate RebuildNN events with the given percentage increment.
1341 .BR \-f ", " \-\-daemonise
1344 to run as a background daemon if it decides to monitor anything. This
1345 causes it to fork and run in the child, and to disconnect from the
1346 terminal. The process id of the child is written to stdout.
1349 which will only continue monitoring if a mail address or alert program
1350 is found in the config file.
1353 .BR \-i ", " \-\-pid\-file
1356 is running in daemon mode, write the pid of the daemon process to
1357 the specified file, instead of printing it on standard output.
1360 .BR \-1 ", " \-\-oneshot
1361 Check arrays only once. This will generate
1363 events and more significantly
1369 .B " mdadm \-\-monitor \-\-scan \-1"
1371 from a cron script will ensure regular notification of any degraded arrays.
1374 .BR \-t ", " \-\-test
1377 alert for every array found at startup. This alert gets mailed and
1378 passed to the alert program. This can be used for testing that alert
1379 message do get through successfully.
1385 .B mdadm \-\-assemble
1386 .I md-device options-and-component-devices...
1389 .B mdadm \-\-assemble \-\-scan
1390 .I md-devices-and-options...
1393 .B mdadm \-\-assemble \-\-scan
1397 This usage assembles one or more RAID arrays from pre-existing components.
1398 For each array, mdadm needs to know the md device, the identity of the
1399 array, and a number of component-devices. These can be found in a number of ways.
1401 In the first usage example (without the
1403 the first device given is the md device.
1404 In the second usage example, all devices listed are treated as md
1405 devices and assembly is attempted.
1406 In the third (where no devices are listed) all md devices that are
1407 listed in the configuration file are assembled. If not arrays are
1408 described by the configuration file, then any arrays that
1409 can be found on unused devices will be assembled.
1411 If precisely one device is listed, but
1417 was given and identity information is extracted from the configuration file.
1419 The identity can be given with the
1425 option, will be taken from the md-device record in the config file, or
1426 will be taken from the super block of the first component-device
1427 listed on the command line.
1429 Devices can be given on the
1431 command line or in the config file. Only devices which have an md
1432 superblock which contains the right identity will be considered for
1435 The config file is only used if explicitly named with
1437 or requested with (a possibly implicit)
1442 .B /etc/mdadm/mdadm.conf
1447 is not given, then the config file will only be used to find the
1448 identity of md arrays.
1450 Normally the array will be started after it is assembled. However if
1452 is not given and not all expected drives were listed, then the array
1453 is not started (to guard against usage errors). To insist that the
1454 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1463 does not create any entries in
1467 It does record information in
1468 .B /var/run/mdadm/map
1471 to choose the correct name.
1475 detects that udev is not configured, it will create the devices in
1479 In Linux kernels prior to version 2.6.28 there were two distinctly
1480 different types of md devices that could be created: one that could be
1481 partitioned using standard partitioning tools and one that could not.
1482 Since 2.6.28 that distinction is no longer relevant as both type of
1483 devices can be partitioned.
1485 will normally create the type that originally could not be partitioned
1486 as it has a well defined major number (9).
1488 Prior to 2.6.28, it is important that mdadm chooses the correct type
1489 of array device to use. This can be controlled with the
1491 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1492 to use a partitionable device rather than the default.
1494 In the no-udev case, the value given to
1496 can be suffixed by a number. This tells
1498 to create that number of partition devices rather than the default of 4.
1502 can also be given in the configuration file as a word starting
1504 on the ARRAY line for the relevant array.
1511 and no devices are listed,
1513 will first attempt to assemble all the arrays listed in the config
1516 In no array at listed in the config (other than those marked
1518 it will look through the available devices for possible arrays and
1519 will try to assemble anything that it finds. Arrays which are tagged
1520 as belonging to the given homehost will be assembled and started
1521 normally. Arrays which do not obviously belong to this host are given
1522 names that are expected not to conflict with anything local, and are
1523 started "read-auto" so that nothing is written to any device until the
1524 array is written to. i.e. automatic resync etc is delayed.
1528 finds a consistent set of devices that look like they should comprise
1529 an array, and if the superblock is tagged as belonging to the given
1530 home host, it will automatically choose a device name and try to
1531 assemble the array. If the array uses version-0.90 metadata, then the
1533 number as recorded in the superblock is used to create a name in
1537 If the array uses version-1 metadata, then the
1539 from the superblock is used to similarly create a name in
1541 (the name will have any 'host' prefix stripped first).
1543 This behaviour can be modified by the
1547 configuration file. This line can indicate that specific metadata
1548 type should, or should not, be automatically assembled. If an array
1549 is found which is not listed in
1551 and has a metadata format that is denied by the
1553 line, then it will not be assembled.
1556 line can also request that all arrays identified as being for this
1557 homehost should be assembled regardless of their metadata type.
1560 for further details.
1565 cannot find any array for the given host at all, and if
1566 .B \-\-auto\-update\-homehost
1569 will search again for any array (not just an array created for this
1570 host) and will assemble each assuming
1571 .BR \-\-update=homehost .
1572 This will change the host tag in the superblock so that on the next run,
1573 these arrays will be found without the second pass. The intention of
1574 this feature is to support transitioning a set of md arrays to using
1577 The reason for requiring arrays to be tagged with the homehost for
1578 auto assembly is to guard against problems that can arise when moving
1579 devices from one host to another.
1590 .BI \-\-raid\-devices= Z
1594 This usage is similar to
1596 The difference is that it creates an array without a superblock. With
1597 these arrays there is no difference between initially creating the array and
1598 subsequently assembling the array, except that hopefully there is useful
1599 data there in the second case.
1601 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1602 one of their synonyms. All devices must be listed and the array will
1603 be started once complete. It will often be appropriate to use
1604 .B \-\-assume\-clean
1605 with levels raid1 or raid10.
1616 .BI \-\-raid\-devices= Z
1620 This usage will initialise a new md array, associate some devices with
1621 it, and activate the array.
1623 The named device will normally not exist when
1624 .I "mdadm \-\-create"
1625 is run, but will be created by
1627 once the array becomes active.
1629 As devices are added, they are checked to see if they contain RAID
1630 superblocks or filesystems. They are also checked to see if the variance in
1631 device size exceeds 1%.
1633 If any discrepancy is found, the array will not automatically be run, though
1636 can override this caution.
1638 To create a "degraded" array in which some devices are missing, simply
1639 give the word "\fBmissing\fP"
1640 in place of a device name. This will cause
1642 to leave the corresponding slot in the array empty.
1643 For a RAID4 or RAID5 array at most one slot can be
1644 "\fBmissing\fP"; for a RAID6 array at most two slots.
1645 For a RAID1 array, only one real device needs to be given. All of the
1649 When creating a RAID5 array,
1651 will automatically create a degraded array with an extra spare drive.
1652 This is because building the spare into a degraded array is in general
1653 faster than resyncing the parity on a non-degraded, but not clean,
1654 array. This feature can be overridden with the
1658 When creating an array with version-1 metadata a name for the array is
1660 If this is not given with the
1664 will choose a name based on the last component of the name of the
1665 device being created. So if
1667 is being created, then the name
1672 is being created, then the name
1676 When creating a partition based array, using
1678 with version-1.x metadata, the partition type should be set to
1680 (non fs-data). This type selection allows for greater precision since
1681 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
1682 might create problems in the event of array recovery through a live cdrom.
1684 A new array will normally get a randomly assigned 128bit UUID which is
1685 very likely to be unique. If you have a specific need, you can choose
1686 a UUID for the array by giving the
1688 option. Be warned that creating two arrays with the same UUID is a
1689 recipe for disaster. Also, using
1691 when creating a v0.90 array will silently override any
1696 .\"option is given, it is not necessary to list any component-devices in this command.
1697 .\"They can be added later, before a
1701 .\"is given, the apparent size of the smallest drive given is used.
1703 When creating an array within a
1706 can be given either the list of devices to use, or simply the name of
1707 the container. The former case gives control over which devices in
1708 the container will be used for the array. The latter case allows
1710 to automatically choose which devices to use based on how much spare
1713 The General Management options that are valid with
1718 insist on running the array even if some devices look like they might
1723 start the array readonly \(em not supported yet.
1730 .I options... devices...
1733 This usage will allow individual devices in an array to be failed,
1734 removed or added. It is possible to perform multiple operations with
1735 on command. For example:
1737 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
1743 and will then remove it from the array and finally add it back
1744 in as a spare. However only one md array can be affected by a single
1747 When a device is added to an active array, mdadm checks to see if it
1748 has metadata on it which suggests that it was recently a member of the
1749 array. If it does, it tries to "re\-add" the device. If there have
1750 been no changes since the device was removed, or if the array has a
1751 write-intent bitmap which has recorded whatever changes there were,
1752 then the device will immediately become a full member of the array and
1753 those differences recorded in the bitmap will be resolved.
1763 MISC mode includes a number of distinct operations that
1764 operate on distinct devices. The operations are:
1767 The device is examined to see if it is
1768 (1) an active md array, or
1769 (2) a component of an md array.
1770 The information discovered is reported.
1774 The device should be an active md device.
1776 will display a detailed description of the array.
1780 will cause the output to be less detailed and the format to be
1781 suitable for inclusion in
1782 .BR /etc/mdadm.conf .
1785 will normally be 0 unless
1787 failed to get useful information about the device(s); however, if the
1789 option is given, then the exit status will be:
1793 The array is functioning normally.
1796 The array has at least one failed device.
1799 The array has multiple failed devices such that it is unusable.
1802 There was an error while trying to get information about the device.
1806 .B \-\-detail\-platform
1807 Print detail of the platform's RAID capabilities (firmware / hardware
1808 topology). If the metadata is specified with
1812 then the return status will be:
1816 metadata successfully enumerated its platform components on this system
1819 metadata is platform independent
1822 metadata failed to find its platform components on this system
1826 .B \-\-update\-subarray=
1827 If the device is a container and the argument to \-\-update\-subarray
1828 specifies a subarray in the container, then attempt to update the given
1829 superblock field in the subarray. Similar to updating an array in
1830 "assemble" mode, the field to update is selected by
1834 option. Currently only
1840 option updates the subarray name in the metadata, it may not affect the
1841 device node name or the device node symlink until the subarray is
1842 re\-assembled. If updating
1844 would change the UUID of an active subarray this operation is blocked,
1845 and the command will end in an error.
1849 The device should be a component of an md array.
1851 will read the md superblock of the device and display the contents.
1856 is given, then multiple devices that are components of the one array
1857 are grouped together and reported in a single entry suitable
1859 .BR /etc/mdadm.conf .
1863 without listing any devices will cause all devices listed in the
1864 config file to be examined.
1868 The devices should be active md arrays which will be deactivated, as
1869 long as they are not currently in use.
1873 This will fully activate a partially assembled md array.
1877 This will mark an active array as read-only, providing that it is
1878 not currently being used.
1884 array back to being read/write.
1888 For all operations except
1891 will cause the operation to be applied to all arrays listed in
1896 causes all devices listed in the config file to be examined.
1899 .BR \-b ", " \-\-brief
1900 Be less verbose. This is used with
1908 gives an intermediate level of verbosity.
1914 .B mdadm \-\-monitor
1915 .I options... devices...
1920 to periodically poll a number of md arrays and to report on any events
1923 will never exit once it decides that there are arrays to be checked,
1924 so it should normally be run in the background.
1926 As well as reporting events,
1928 may move a spare drive from one array to another if they are in the
1931 and if the destination array has a failed drive but no spares.
1933 If any devices are listed on the command line,
1935 will only monitor those devices. Otherwise all arrays listed in the
1936 configuration file will be monitored. Further, if
1938 is given, then any other md devices that appear in
1940 will also be monitored.
1942 The result of monitoring the arrays is the generation of events.
1943 These events are passed to a separate program (if specified) and may
1944 be mailed to a given E-mail address.
1946 When passing events to a program, the program is run once for each event,
1947 and is given 2 or 3 command-line arguments: the first is the
1948 name of the event (see below), the second is the name of the
1949 md device which is affected, and the third is the name of a related
1950 device if relevant (such as a component device that has failed).
1954 is given, then a program or an E-mail address must be specified on the
1955 command line or in the config file. If neither are available, then
1957 will not monitor anything.
1961 will continue monitoring as long as something was found to monitor. If
1962 no program or email is given, then each event is reported to
1965 The different events are:
1969 .B DeviceDisappeared
1970 An md array which previously was configured appears to no longer be
1971 configured. (syslog priority: Critical)
1975 was told to monitor an array which is RAID0 or Linear, then it will
1977 .B DeviceDisappeared
1978 with the extra information
1980 This is because RAID0 and Linear do not support the device-failed,
1981 hot-spare and resync operations which are monitored.
1985 An md array started reconstruction. (syslog priority: Warning)
1991 is a two-digit number (ie. 05, 48). This indicates that rebuild
1992 has passed that many percent of the total. The events are generated
1993 with fixed increment since 0. Increment size may be specified with
1994 a commandline option (default is 20). (syslog priority: Warning)
1998 An md array that was rebuilding, isn't any more, either because it
1999 finished normally or was aborted. (syslog priority: Warning)
2003 An active component device of an array has been marked as
2004 faulty. (syslog priority: Critical)
2008 A spare component device which was being rebuilt to replace a faulty
2009 device has failed. (syslog priority: Critical)
2013 A spare component device which was being rebuilt to replace a faulty
2014 device has been successfully rebuilt and has been made active.
2015 (syslog priority: Info)
2019 A new md array has been detected in the
2021 file. (syslog priority: Info)
2025 A newly noticed array appears to be degraded. This message is not
2028 notices a drive failure which causes degradation, but only when
2030 notices that an array is degraded when it first sees the array.
2031 (syslog priority: Critical)
2035 A spare drive has been moved from one array in a
2037 to another to allow a failed drive to be replaced.
2038 (syslog priority: Info)
2044 has been told, via the config file, that an array should have a certain
2045 number of spare devices, and
2047 detects that it has fewer than this number when it first sees the
2048 array, it will report a
2051 (syslog priority: Warning)
2055 An array was found at startup, and the
2058 (syslog priority: Info)
2068 cause Email to be sent. All events cause the program to be run.
2069 The program is run with two or three arguments: the event
2070 name, the array device and possibly a second device.
2072 Each event has an associated array device (e.g.
2074 and possibly a second device. For
2079 the second device is the relevant component device.
2082 the second device is the array that the spare was moved from.
2086 to move spares from one array to another, the different arrays need to
2087 be labeled with the same
2089 in the configuration file. The
2091 name can be any string; it is only necessary that different spare
2092 groups use different names.
2096 detects that an array in a spare group has fewer active
2097 devices than necessary for the complete array, and has no spare
2098 devices, it will look for another array in the same spare group that
2099 has a full complement of working drive and a spare. It will then
2100 attempt to remove the spare from the second drive and add it to the
2102 If the removal succeeds but the adding fails, then it is added back to
2106 The GROW mode is used for changing the size or shape of an active
2108 For this to work, the kernel must support the necessary change.
2109 Various types of growth are being added during 2.6 development,
2110 including restructuring a RAID5 array to have more active devices.
2112 Currently the only support available is to
2114 change the "size" attribute
2115 for RAID1, RAID5 and RAID6.
2117 increase or decrease the "raid\-devices" attribute of RAID1, RAID5,
2120 change the chunk-size and layout of RAID5 and RAID6.
2122 convert between RAID1 and RAID5, and between RAID5 and RAID6.
2124 add a write-intent bitmap to any array which supports these bitmaps, or
2125 remove a write-intent bitmap from such an array.
2128 GROW mode is not currently supported for
2130 or arrays inside containers.
2133 Normally when an array is built the "size" it taken from the smallest
2134 of the drives. If all the small drives in an arrays are, one at a
2135 time, removed and replaced with larger drives, then you could have an
2136 array of large drives with only a small amount used. In this
2137 situation, changing the "size" with "GROW" mode will allow the extra
2138 space to start being used. If the size is increased in this way, a
2139 "resync" process will start to make sure the new parts of the array
2142 Note that when an array changes size, any filesystem that may be
2143 stored in the array will not automatically grow to use the space. The
2144 filesystem will need to be explicitly told to use the extra space.
2146 Also the size of an array cannot be changed while it has an active
2147 bitmap. If an array has a bitmap, it must be removed before the size
2148 can be changed. Once the change it complete a new bitmap can be created.
2150 .SS RAID\-DEVICES CHANGES
2152 A RAID1 array can work with any number of devices from 1 upwards
2153 (though 1 is not very useful). There may be times which you want to
2154 increase or decrease the number of active devices. Note that this is
2155 different to hot-add or hot-remove which changes the number of
2158 When reducing the number of devices in a RAID1 array, the slots which
2159 are to be removed from the array must already be vacant. That is, the
2160 devices which were in those slots must be failed and removed.
2162 When the number of devices is increased, any hot spares that are
2163 present will be activated immediately.
2165 Changing the number of active devices in a RAID5 or RAID6 is much more
2166 effort. Every block in the array will need to be read and written
2167 back to a new location. From 2.6.17, the Linux Kernel is able to
2168 increase the number of devices in a RAID5 safely, including restarting
2169 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2170 increase or decrease the number of devices in a RAID5 or RAID6.
2172 When decreasing the number of devices, the size of the array will also
2173 decrease. If there was data in the array, it could get destroyed and
2174 this is not reversible. To help prevent accidents,
2176 requires that the size of the array be decreased first with
2177 .BR "mdadm --grow --array-size" .
2178 This is a reversible change which simply makes the end of the array
2179 inaccessible. The integrity of any data can then be checked before
2180 the non-reversible reduction in the number of devices is request.
2182 When relocating the first few stripes on a RAID5, it is not possible
2183 to keep the data on disk completely consistent and crash-proof. To
2184 provide the required safety, mdadm disables writes to the array while
2185 this "critical section" is reshaped, and takes a backup of the data
2186 that is in that section. This backup is normally stored in any spare
2187 devices that the array has, however it can also be stored in a
2188 separate file specified with the
2190 option. If this option is used, and the system does crash during the
2191 critical period, the same file must be passed to
2193 to restore the backup and reassemble the array.
2197 Changing the RAID level of any array happens instantaneously. However
2198 in the RAID to RAID6 case this requires a non-standard layout of the
2199 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2200 required before the change can be accomplish. So while the level
2201 change is instant, the accompanying layout change can take quite a
2204 .SS CHUNK-SIZE AND LAYOUT CHANGES
2206 Changing the chunk-size of layout without also changing the number of
2207 devices as the same time will involve re-writing all blocks in-place.
2208 To ensure against data loss in the case of a crash, a
2210 must be provided for these changes. Small sections of the array will
2211 be copied to the backup file while they are being rearranged.
2213 If the reshape is interrupted for any reason, this backup file must be
2215 .B "mdadm --assemble"
2216 so the array can be reassembled. Consequently the file cannot be
2217 stored on the device being reshaped.
2222 A write-intent bitmap can be added to, or removed from, an active
2223 array. Either internal bitmaps, or bitmaps stored in a separate file,
2224 can be added. Note that if you add a bitmap stored in a file which is
2225 in a filesystem that is on the RAID array being affected, the system
2226 will deadlock. The bitmap must be on a separate filesystem.
2228 .SH INCREMENTAL MODE
2232 .B mdadm \-\-incremental
2238 .B mdadm \-\-incremental \-\-fail
2242 .B mdadm \-\-incremental \-\-rebuild\-map
2245 .B mdadm \-\-incremental \-\-run \-\-scan
2248 This mode is designed to be used in conjunction with a device
2249 discovery system. As devices are found in a system, they can be
2251 .B "mdadm \-\-incremental"
2252 to be conditionally added to an appropriate array.
2254 Conversely, it can also be used with the
2256 flag to do just the opposite and find whatever array a particular device
2257 is part of and remove the device from that array.
2259 If the device passed is a
2261 device created by a previous call to
2263 then rather than trying to add that device to an array, all the arrays
2264 described by the metadata of the container will be started.
2267 performs a number of tests to determine if the device is part of an
2268 array, and which array it should be part of. If an appropriate array
2269 is found, or can be created,
2271 adds the device to the array and conditionally starts the array.
2275 will only add devices to an array which were previously working
2276 (active or spare) parts of that array. It does not currently support
2277 automatic inclusion of a new drive as a spare in some array.
2281 makes are as follow:
2283 Is the device permitted by
2285 That is, is it listed in a
2287 line in that file. If
2289 is absent then the default it to allow any device. Similar if
2291 contains the special word
2293 then any device is allowed. Otherwise the device name given to
2295 must match one of the names or patterns in a
2300 Does the device have a valid md superblock. If a specific metadata
2301 version is request with
2305 then only that style of metadata is accepted, otherwise
2307 finds any known version of metadata. If no
2309 metadata is found, the device is rejected.
2313 Does the metadata match an expected array?
2314 The metadata can match in two ways. Either there is an array listed
2317 which identifies the array (either by UUID, by name, by device list,
2318 or by minor-number), or the array was created with a
2324 or on the command line.
2327 is not able to positively identify the array as belonging to the
2328 current host, the device will be rejected.
2332 keeps a list of arrays that it has partially assembled in
2333 .B /var/run/mdadm/map
2335 .B /var/run/mdadm.map
2336 if the directory doesn't exist. Or maybe even
2337 .BR /dev/.mdadm.map ).
2338 If no array exists which matches
2339 the metadata on the new device,
2341 must choose a device name and unit number. It does this based on any
2344 or any name information stored in the metadata. If this name
2345 suggests a unit number, that number will be used, otherwise a free
2346 unit number will be chosen. Normally
2348 will prefer to create a partitionable array, however if the
2352 suggests that a non-partitionable array is preferred, that will be
2355 If the array is not found in the config file and its metadata does not
2356 identify it as belonging to the "homehost", then
2358 will choose a name for the array which is certain not to conflict with
2359 any array which does belong to this host. It does this be adding an
2360 underscore and a small number to the name preferred by the metadata.
2362 Once an appropriate array is found or created and the device is added,
2364 must decide if the array is ready to be started. It will
2365 normally compare the number of available (non-spare) devices to the
2366 number of devices that the metadata suggests need to be active. If
2367 there are at least that many, the array will be started. This means
2368 that if any devices are missing the array will not be restarted.
2374 in which case the array will be run as soon as there are enough
2375 devices present for the data to be accessible. For a RAID1, that
2376 means one device will start the array. For a clean RAID5, the array
2377 will be started as soon as all but one drive is present.
2379 Note that neither of these approaches is really ideal. If it can
2380 be known that all device discovery has completed, then
2384 can be run which will try to start all arrays that are being
2385 incrementally assembled. They are started in "read-auto" mode in
2386 which they are read-only until the first write request. This means
2387 that no metadata updates are made and no attempt at resync or recovery
2388 happens. Further devices that are found before the first write can
2389 still be added safely.
2392 This section describes environment variables that affect how mdadm
2397 Setting this value to 1 will prevent mdadm from automatically launching
2398 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2404 does not create any device nodes in /dev, but leaves that task to
2408 appears not to be configured, or if this environment variable is set
2411 will create and devices that are needed.
2415 .B " mdadm \-\-query /dev/name-of-device"
2417 This will find out if a given device is a RAID array, or is part of
2418 one, and will provide brief information about the device.
2420 .B " mdadm \-\-assemble \-\-scan"
2422 This will assemble and start all arrays listed in the standard config
2423 file. This command will typically go in a system startup file.
2425 .B " mdadm \-\-stop \-\-scan"
2427 This will shut down all arrays that can be shut down (i.e. are not
2428 currently in use). This will typically go in a system shutdown script.
2430 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2432 If (and only if) there is an Email address or program given in the
2433 standard config file, then
2434 monitor the status of all arrays listed in that file by
2435 polling them ever 2 minutes.
2437 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2439 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2442 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2444 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2446 This will create a prototype config file that describes currently
2447 active arrays that are known to be made from partitions of IDE or SCSI drives.
2448 This file should be reviewed before being used as it may
2449 contain unwanted detail.
2451 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2453 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
2455 This will find arrays which could be assembled from existing IDE and
2456 SCSI whole drives (not partitions), and store the information in the
2457 format of a config file.
2458 This file is very likely to contain unwanted detail, particularly
2461 entries. It should be reviewed and edited before being used as an
2464 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
2466 .B " mdadm \-Ebsc partitions"
2468 Create a list of devices by reading
2469 .BR /proc/partitions ,
2470 scan these for RAID superblocks, and printout a brief listing of all
2473 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
2475 Scan all partitions and devices listed in
2476 .BR /proc/partitions
2479 out of all such devices with a RAID superblock with a minor number of 0.
2481 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /var/run/mdadm"
2483 If config file contains a mail address or alert program, run mdadm in
2484 the background in monitor mode monitoring all md devices. Also write
2485 pid of mdadm daemon to
2486 .BR /var/run/mdadm .
2488 .B " mdadm \-Iq /dev/somedevice"
2490 Try to incorporate newly discovered device into some array as
2493 .B " mdadm \-\-incremental \-\-rebuild\-map \-\-run \-\-scan"
2495 Rebuild the array map from any current arrays, and then start any that
2498 .B " mdadm /dev/md4 --fail detached --remove detached"
2500 Any devices which are components of /dev/md4 will be marked as faulty
2501 and then remove from the array.
2503 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4
2507 which is currently a RAID5 array will be converted to RAID6. There
2508 should normally already be a spare drive attached to the array as a
2509 RAID6 needs one more drive than a matching RAID5.
2511 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
2513 Create a DDF array over 6 devices.
2515 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
2517 Create a RAID5 array over any 3 devices in the given DDF set. Use
2518 only 30 gigabytes of each device.
2520 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
2522 Assemble a pre-exist ddf array.
2524 .B " mdadm -I /dev/md/ddf1"
2526 Assemble all arrays contained in the ddf array, assigning names as
2529 .B " mdadm \-\-create \-\-help"
2531 Provide help about the Create mode.
2533 .B " mdadm \-\-config \-\-help"
2535 Provide help about the format of the config file.
2537 .B " mdadm \-\-help"
2539 Provide general help.
2549 lists all active md devices with information about them.
2551 uses this to find arrays when
2553 is given in Misc mode, and to monitor array reconstruction
2558 The config file lists which devices may be scanned to see if
2559 they contain MD super block, and gives identifying information
2560 (e.g. UUID) about known MD arrays. See
2564 .SS /var/run/mdadm/map
2567 mode is used, this file gets a list of arrays currently being created.
2570 does not exist as a directory, then
2571 .B /var/run/mdadm.map
2574 is not available (as may be the case during early boot),
2576 is used on the basis that
2578 is usually available very early in boot.
2583 understand two sorts of names for array devices.
2585 The first is the so-called 'standard' format name, which matches the
2586 names used by the kernel and which appear in
2589 The second sort can be freely chosen, but must reside in
2591 When giving a device name to
2593 to create or assemble an array, either full path name such as
2597 can be given, or just the suffix of the second sort of name, such as
2603 chooses device names during auto-assembly or incremental assembly, it
2604 will sometimes add a small sequence number to the end of the name to
2605 avoid conflicted between multiple arrays that have the same name. If
2607 can reasonably determine that the array really is meant for this host,
2608 either by a hostname in the metadata, or by the presence of the array
2609 in /etc/mdadm.conf, then it will leave off the suffix if possible.
2610 Also if the homehost is specified as
2613 will only use a suffix if a different array of the same name already
2614 exists or is listed in the config file.
2616 The standard names for non-partitioned arrays (the only sort of md
2617 array available in 2.4 and earlier) are of the form
2621 where NN is a number.
2622 The standard names for partitionable arrays (as available from 2.6
2623 onwards) are of the form
2627 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2629 From kernel version, 2.6.28 the "non-partitioned array" can actually
2630 be partitioned. So the "md_dNN" names are no longer needed, and
2631 partitions such as "/dev/mdNNpXX" are possible.
2635 was previously known as
2639 is completely separate from the
2641 package, and does not use the
2643 configuration file at all.
2646 For further information on mdadm usage, MD and the various levels of
2649 .B http://linux\-raid.osdl.org/
2651 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
2653 .\"for new releases of the RAID driver check out:
2656 .\".UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2657 .\"ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2662 .\".UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2663 .\"http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2666 The latest version of
2668 should always be available from
2670 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/