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.
144 in this mode, then any arrays within that container will be assembled
149 This is for doing things to specific components of an array such as
150 adding new spares and removing faulty devices.
154 This is an 'everything else' mode that supports operations on active
155 arrays, operations on component devices such as erasing old superblocks, and
156 information gathering operations.
157 .\"This mode allows operations on independent devices such as examine MD
158 .\"superblocks, erasing old superblocks and stopping active arrays.
162 This mode does not act on a specific device or array, but rather it
163 requests the Linux Kernel to activate any auto-detected arrays.
166 .SH Options for selecting a mode are:
169 .BR \-A ", " \-\-assemble
170 Assemble a pre-existing array.
173 .BR \-B ", " \-\-build
174 Build a legacy array without superblocks.
177 .BR \-C ", " \-\-create
181 .BR \-F ", " \-\-follow ", " \-\-monitor
187 .BR \-G ", " \-\-grow
188 Change the size or shape of an active array.
191 .BR \-I ", " \-\-incremental
192 Add a single device into an appropriate array, and possibly start the array.
196 Request that the kernel starts any auto-detected arrays. This can only
199 is compiled into the kernel \(em not if it is a module.
200 Arrays can be auto-detected by the kernel if all the components are in
201 primary MS-DOS partitions with partition type
203 and all use v0.90 metadata.
204 In-kernel autodetect is not recommended for new installations. Using
206 to detect and assemble arrays \(em possibly in an
208 \(em is substantially more flexible and should be preferred.
211 If a device is given before any options, or if the first option is
216 then the MANAGE mode is assumed.
217 Anything other than these will cause the
221 .SH Options that are not mode-specific are:
224 .BR \-h ", " \-\-help
225 Display general help message or, after one of the above options, a
226 mode-specific help message.
230 Display more detailed help about command line parsing and some commonly
234 .BR \-V ", " \-\-version
235 Print version information for mdadm.
238 .BR \-v ", " \-\-verbose
239 Be more verbose about what is happening. This can be used twice to be
241 The extra verbosity currently only affects
242 .B \-\-detail \-\-scan
244 .BR "\-\-examine \-\-scan" .
247 .BR \-q ", " \-\-quiet
248 Avoid printing purely informative messages. With this,
250 will be silent unless there is something really important to report.
253 .BR \-b ", " \-\-brief
254 Be less verbose. This is used with
262 gives an intermediate level of verbosity.
265 .BR \-f ", " \-\-force
266 Be more forceful about certain operations. See the various modes for
267 the exact meaning of this option in different contexts.
270 .BR \-c ", " \-\-config=
271 Specify the config file. Default is to use
272 .BR /etc/mdadm.conf ,
273 or if that is missing then
274 .BR /etc/mdadm/mdadm.conf .
275 If the config file given is
277 then nothing will be read, but
279 will act as though the config file contained exactly
280 .B "DEVICE partitions containers"
283 to find a list of devices to scan, and
285 to find a list of containers to examine.
288 is given for the config file, then
290 will act as though the config file were empty.
293 .BR \-s ", " \-\-scan
296 for missing information.
297 In general, this option gives
299 permission to get any missing information (like component devices,
300 array devices, array identities, and alert destination) from the
301 configuration file (see previous option);
302 one exception is MISC mode when using
308 says to get a list of array devices from
312 .BR \-e ", " \-\-metadata=
313 Declare the style of RAID metadata (superblock) to be used. The
316 and to guess for other operations.
317 The default can be overridden by setting the
327 Use the original 0.90 format superblock. This format limits arrays to
328 28 component devices and limits component devices of levels 1 and
329 greater to 2 terabytes.
330 .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', 'default' is
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 This value can be set with
415 for RAID level 1/4/5/6. If the array was created with a size smaller
416 than the currently active drives, the extra space can be accessed
419 The size can be given as
421 which means to choose the largest size that fits on all current drives.
423 This value can not be used with
425 metadata such as DDF and IMSM.
428 .BR \-Z ", " \-\-array-size=
429 This is only meaningful with
431 and its effect is not persistent: when the array is stopped an
432 restarted the default array size will be restored.
434 Setting the array-size causes the array to appear smaller to programs
435 that access the data. This is particularly needed before reshaping an
436 array so that it will be smaller. As the reshape is not reversible,
437 but setting the size with
439 is, it is required that the array size is reduced as appropriate
440 before the number of devices in the array is reduced.
443 .BR \-c ", " \-\-chunk=
444 Specify chunk size of kibibytes. The default when creating an
445 array is 512KB. To ensure compatibility with earlier versions, the
446 default when Building and array with no persistent metadata is 64KB.
447 This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10.
451 Specify rounding factor for a Linear array. The size of each
452 component will be rounded down to a multiple of this size.
453 This is a synonym for
455 but highlights the different meaning for Linear as compared to other
456 RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in
457 use, and is 0K (i.e. no rounding) in later kernels.
460 .BR \-l ", " \-\-level=
461 Set RAID level. When used with
463 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
464 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container.
465 Obviously some of these are synonymous.
469 metadata type is requested, only the
471 level is permitted, and it does not need to be explicitly given.
475 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
477 Not yet supported with
481 .BR \-p ", " \-\-layout=
482 This option configures the fine details of data layout for RAID5, RAID6,
483 and RAID10 arrays, and controls the failure modes for
486 The layout of the RAID5 parity block can be one of
487 .BR left\-asymmetric ,
488 .BR left\-symmetric ,
489 .BR right\-asymmetric ,
490 .BR right\-symmetric ,
491 .BR la ", " ra ", " ls ", " rs .
493 .BR left\-symmetric .
495 It is also possibly to cause RAID5 to use a RAID4-like layout by
501 Finally for RAID5 there are DDF\-compatible layouts,
502 .BR ddf\-zero\-restart ,
503 .BR ddf\-N\-restart ,
505 .BR ddf\-N\-continue .
507 These same layouts are available for RAID6. There are also 4 layouts
508 that will provide an intermediate stage for converting between RAID5
509 and RAID6. These provide a layout which is identical to the
510 corresponding RAID5 layout on the first N\-1 devices, and has the 'Q'
511 syndrome (the second 'parity' block used by RAID6) on the last device.
513 .BR left\-symmetric\-6 ,
514 .BR right\-symmetric\-6 ,
515 .BR left\-asymmetric\-6 ,
516 .BR right\-asymmetric\-6 ,
518 .BR pairty\-first\-6 .
520 When setting the failure mode for level
523 .BR write\-transient ", " wt ,
524 .BR read\-transient ", " rt ,
525 .BR write\-persistent ", " wp ,
526 .BR read\-persistent ", " rp ,
528 .BR read\-fixable ", " rf ,
529 .BR clear ", " flush ", " none .
531 Each failure mode can be followed by a number, which is used as a period
532 between fault generation. Without a number, the fault is generated
533 once on the first relevant request. With a number, the fault will be
534 generated after that many requests, and will continue to be generated
535 every time the period elapses.
537 Multiple failure modes can be current simultaneously by using the
539 option to set subsequent failure modes.
541 "clear" or "none" will remove any pending or periodic failure modes,
542 and "flush" will clear any persistent faults.
544 Finally, the layout options for RAID10 are one of 'n', 'o' or 'f' followed
545 by a small number. The default is 'n2'. The supported options are:
548 signals 'near' copies. Multiple copies of one data block are at
549 similar offsets in different devices.
552 signals 'offset' copies. Rather than the chunks being duplicated
553 within a stripe, whole stripes are duplicated but are rotated by one
554 device so duplicate blocks are on different devices. Thus subsequent
555 copies of a block are in the next drive, and are one chunk further
560 (multiple copies have very different offsets).
561 See md(4) for more detail about 'near', 'offset', and 'far'.
563 The number is the number of copies of each datablock. 2 is normal, 3
564 can be useful. This number can be at most equal to the number of
565 devices in the array. It does not need to divide evenly into that
566 number (e.g. it is perfectly legal to have an 'n2' layout for an array
567 with an odd number of devices).
569 When an array is converted between RAID5 and RAID6 an intermediate
570 RAID6 layout is used in which the second parity block (Q) is always on
571 the last device. To convert a RAID5 to RAID6 and leave it in this new
572 layout (which does not require re-striping) use
573 .BR \-\-layout=preserve .
574 This will try to avoid any restriping.
576 The converse of this is
577 .B \-\-layout=normalise
578 which will change a non-standard RAID6 layout into a more standard
585 (thus explaining the p of
589 .BR \-b ", " \-\-bitmap=
590 Specify a file to store a write-intent bitmap in. The file should not
593 is also given. The same file should be provided
594 when assembling the array. If the word
596 is given, then the bitmap is stored with the metadata on the array,
597 and so is replicated on all devices. If the word
601 mode, then any bitmap that is present is removed.
603 To help catch typing errors, the filename must contain at least one
604 slash ('/') if it is a real file (not 'internal' or 'none').
606 Note: external bitmaps are only known to work on ext2 and ext3.
607 Storing bitmap files on other filesystems may result in serious problems.
610 .BR \-\-bitmap\-chunk=
611 Set the chunksize of the bitmap. Each bit corresponds to that many
612 Kilobytes of storage.
613 When using a file based bitmap, the default is to use the smallest
614 size that is at-least 4 and requires no more than 2^21 chunks.
617 bitmap, the chunksize defaults to 64Meg, or larger if necessary to
618 fit the bitmap into the available space.
621 .BR \-W ", " \-\-write\-mostly
622 subsequent devices listed in a
627 command will be flagged as 'write-mostly'. This is valid for RAID1
628 only and means that the 'md' driver will avoid reading from these
629 devices if at all possible. This can be useful if mirroring over a
633 .BR \-\-write\-behind=
634 Specify that write-behind mode should be enabled (valid for RAID1
635 only). If an argument is specified, it will set the maximum number
636 of outstanding writes allowed. The default value is 256.
637 A write-intent bitmap is required in order to use write-behind
638 mode, and write-behind is only attempted on drives marked as
642 .BR \-\-assume\-clean
645 that the array pre-existed and is known to be clean. It can be useful
646 when trying to recover from a major failure as you can be sure that no
647 data will be affected unless you actually write to the array. It can
648 also be used when creating a RAID1 or RAID10 if you want to avoid the
649 initial resync, however this practice \(em while normally safe \(em is not
650 recommended. Use this only if you really know what you are doing.
652 When the devices that will be part of a new array were filled
653 with zeros before creation the operator knows the array is
654 actually clean. If that is the case, such as after running
655 badblocks, this argument can be used to tell mdadm the
656 facts the operator knows.
659 .BR \-\-backup\-file=
662 is used to increase the number of
663 raid-devices in a RAID5 if there are no spare devices available.
664 See the GROW MODE section below on RAID\-DEVICES CHANGES. The file
665 should be stored on a separate device, not on the RAID array being
669 .BR \-\-array-size= ", " \-Z
670 Set the size of the array which is seen by users of the device such as
671 filesystems. This can be less that the real size, but never greater.
672 The size set this way does not persist across restarts of the array.
674 This is most useful when reducing the number of devices in a RAID5 or
675 RAID6. Such arrays require the array-size to be reduced before a
676 reshape can be performed that reduces the real size.
680 restores the apparent size of the array to be whatever the real
681 amount of available space is.
684 .BR \-N ", " \-\-name=
687 for the array. This is currently only effective when creating an
688 array with a version-1 superblock, or an array in a DDF container.
689 The name is a simple textual string that can be used to identify array
690 components when assembling. If name is needed but not specified, it
691 is taken from the basename of the device that is being created.
703 run the array, even if some of the components
704 appear to be active in another array or filesystem. Normally
706 will ask for confirmation before including such components in an
707 array. This option causes that question to be suppressed.
710 .BR \-f ", " \-\-force
713 accept the geometry and layout specified without question. Normally
715 will not allow creation of an array with only one device, and will try
716 to create a RAID5 array with one missing drive (as this makes the
717 initial resync work faster). With
720 will not try to be so clever.
723 .BR \-a ", " "\-\-auto{=yes,md,mdp,part,p}{NN}"
724 Instruct mdadm how to create the device file if needed, possibly allocating
725 an unused minor number. "md" causes a non-partitionable array
726 to be used (though since Linux 2.6.28, these array devices are in fact
727 partitionable). "mdp", "part" or "p" causes a partitionable array (2.6 and
728 later) to be used. "yes" requires the named md device to have
729 a 'standard' format, and the type and minor number will be determined
730 from this. With mdadm 3.0, device creation is normally left up to
732 so this option is unlikely to be needed.
733 See DEVICE NAMES below.
735 The argument can also come immediately after
740 is not given on the command line or in the config file, then
746 is also given, then any
748 entries in the config file will override the
750 instruction given on the command line.
752 For partitionable arrays,
754 will create the device file for the whole array and for the first 4
755 partitions. A different number of partitions can be specified at the
756 end of this option (e.g.
758 If the device name ends with a digit, the partition names add a 'p',
760 .IR /dev/md/home1p3 .
761 If there is no trailing digit, then the partition names just have a
763 .IR /dev/md/scratch3 .
765 If the md device name is in a 'standard' format as described in DEVICE
766 NAMES, then it will be created, if necessary, with the appropriate
767 device number based on that name. If the device name is not in one of these
768 formats, then a unused device number will be allocated. The device
769 number will be considered unused if there is no active array for that
770 number, and there is no entry in /dev for that number and with a
771 non-standard name. Names that are not in 'standard' format are only
772 allowed in "/dev/md/".
776 .\".BR \-\-symlink = no
781 .\"to create devices in
783 .\"it will also create symlinks from
785 .\"with names starting with
791 .\"to suppress this, or
792 .\".B \-\-symlink=yes
793 .\"to enforce this even if it is suppressing
801 .BR \-u ", " \-\-uuid=
802 uuid of array to assemble. Devices which don't have this uuid are
806 .BR \-m ", " \-\-super\-minor=
807 Minor number of device that array was created for. Devices which
808 don't have this minor number are excluded. If you create an array as
809 /dev/md1, then all superblocks will contain the minor number 1, even if
810 the array is later assembled as /dev/md2.
812 Giving the literal word "dev" for
816 to use the minor number of the md device that is being assembled.
819 .B \-\-super\-minor=dev
820 will look for super blocks with a minor number of 0.
823 is only relevant for v0.90 metadata, and should not normally be used.
829 .BR \-N ", " \-\-name=
830 Specify the name of the array to assemble. This must be the name
831 that was specified when creating the array. It must either match
832 the name stored in the superblock exactly, or it must match
835 prefixed to the start of the given name.
838 .BR \-f ", " \-\-force
839 Assemble the array even if the metadata on some devices appears to be
842 cannot find enough working devices to start the array, but can find
843 some devices that are recorded as having failed, then it will mark
844 those devices as working so that the array can be started.
845 An array which requires
847 to be started may contain data corruption. Use it carefully.
851 Attempt to start the array even if fewer drives were given than were
852 present last time the array was active. Normally if not all the
853 expected drives are found and
855 is not used, then the array will be assembled but not started.
858 an attempt will be made to start it anyway.
862 This is the reverse of
864 in that it inhibits the startup of array unless all expected drives
865 are present. This is only needed with
867 and can be used if the physical connections to devices are
868 not as reliable as you would like.
871 .BR \-a ", " "\-\-auto{=no,yes,md,mdp,part}"
872 See this option under Create and Build options.
875 .BR \-b ", " \-\-bitmap=
876 Specify the bitmap file that was given when the array was created. If
879 bitmap, there is no need to specify this when assembling the array.
882 .BR \-\-backup\-file=
885 was used to grow the number of raid-devices in a RAID5, and the system
886 crashed during the critical section, then the same
890 to allow possibly corrupted data to be restored.
893 .BR \-U ", " \-\-update=
894 Update the superblock on each device while assembling the array. The
895 argument given to this flag can be one of
909 option will adjust the superblock of an array what was created on a Sparc
910 machine running a patched 2.2 Linux kernel. This kernel got the
911 alignment of part of the superblock wrong. You can use the
912 .B "\-\-examine \-\-sparc2.2"
915 to see what effect this would have.
919 option will update the
921 field on each superblock to match the minor number of the array being
923 This can be useful if
925 reports a different "Preferred Minor" to
927 In some cases this update will be performed automatically
928 by the kernel driver. In particular the update happens automatically
929 at the first write to an array with redundancy (RAID level 1 or
930 greater) on a 2.6 (or later) kernel.
934 option will change the uuid of the array. If a UUID is given with the
936 option that UUID will be used as a new UUID and will
938 be used to help identify the devices in the array.
941 is given, a random UUID is chosen.
945 option will change the
947 of the array as stored in the superblock. This is only supported for
948 version-1 superblocks.
952 option will change the
954 as recorded in the superblock. For version-0 superblocks, this is the
955 same as updating the UUID.
956 For version-1 superblocks, this involves updating the name.
960 option will cause the array to be marked
962 meaning that any redundancy in the array (e.g. parity for RAID5,
963 copies for RAID1) may be incorrect. This will cause the RAID system
964 to perform a "resync" pass to make sure that all redundant information
969 option allows arrays to be moved between machines with different
971 When assembling such an array for the first time after a move, giving
972 .B "\-\-update=byteorder"
975 to expect superblocks to have their byteorder reversed, and will
976 correct that order before assembling the array. This is only valid
977 with original (Version 0.90) superblocks.
981 option will correct the summaries in the superblock. That is the
982 counts of total, working, active, failed, and spare devices.
986 will rarely be of use. It applies to version 1.1 and 1.2 metadata
987 only (where the metadata is at the start of the device) and is only
988 useful when the component device has changed size (typically become
989 larger). The version 1 metadata records the amount of the device that
990 can be used to store data, so if a device in a version 1.1 or 1.2
991 array becomes larger, the metadata will still be visible, but the
992 extra space will not. In this case it might be useful to assemble the
994 .BR \-\-update=devicesize .
997 to determine the maximum usable amount of space on each device and
998 update the relevant field in the metadata.
1002 .B \-\-auto\-update\-homehost
1003 This flag is only meaningful with auto-assembly (see discussion below).
1004 In that situation, if no suitable arrays are found for this homehost,
1006 will rescan for any arrays at all and will assemble them and update the
1007 homehost to match the current host.
1010 .SH For Manage mode:
1013 .BR \-a ", " \-\-add
1014 hot-add listed devices. For arrays with redundancy, the listed
1015 devices become available as spares. If the array is degraded, it will
1016 immediately start recovering data on to one of these spares.
1020 re-add a device that was recently removed from an array. This is only
1021 needed for arrays that have be built (i.e. with
1023 For created arrays, devices are always re-added if that is possible.
1024 When re-adding a device, if nothing has changed on the array since the
1025 device was removed, no recovery is performed. Also, if the array has
1026 a write-intent bitmap, then the recovery performed after a re-add will
1027 be limited to those blocks which, according to the bitmap, might have
1028 changed since the device was removed.
1031 .BR \-r ", " \-\-remove
1032 remove listed devices. They must not be active. i.e. they should
1033 be failed or spare devices. As well as the name of a device file
1042 The first causes all failed device to be removed. The second causes
1043 any device which is no longer connected to the system (i.e an 'open'
1046 to be removed. This will only succeed for devices that are spares or
1047 have already been marked as failed.
1050 .BR \-f ", " \-\-fail
1051 mark listed devices as faulty.
1052 As well as the name of a device file, the word
1054 can be given. This will cause any device that has been detached from
1055 the system to be marked as failed. It can then be removed.
1063 .BR \-\-write\-mostly
1064 Subsequent devices that are added or re-added will have the 'write-mostly'
1065 flag set. This is only valid for RAID1 and means that the 'md' driver
1066 will avoid reading from these devices if possible.
1069 Subsequent devices that are added or re-added will have the 'write-mostly'
1073 Each of these options requires that the first device listed is the array
1074 to be acted upon, and the remainder are component devices to be added,
1075 removed, marked as faulty, etc. Several different operations can be
1076 specified for different devices, e.g.
1078 mdadm /dev/md0 \-\-add /dev/sda1 \-\-fail /dev/sdb1 \-\-remove /dev/sdb1
1080 Each operation applies to all devices listed until the next
1083 If an array is using a write-intent bitmap, then devices which have
1084 been removed can be re-added in a way that avoids a full
1085 reconstruction but instead just updates the blocks that have changed
1086 since the device was removed. For arrays with persistent metadata
1087 (superblocks) this is done automatically. For arrays created with
1089 mdadm needs to be told that this device we removed recently with
1092 Devices can only be removed from an array if they are not in active
1093 use, i.e. that must be spares or failed devices. To remove an active
1094 device, it must first be marked as
1100 .BR \-Q ", " \-\-query
1101 Examine a device to see
1102 (1) if it is an md device and (2) if it is a component of an md
1104 Information about what is discovered is presented.
1107 .BR \-D ", " \-\-detail
1108 Print details of one or more md devices.
1111 .BR \-\-detail\-platform
1112 Print details of the platform's RAID capabilities (firmware / hardware
1113 topology) for a given metadata format.
1116 .BR \-Y ", " \-\-export
1121 output will be formatted as
1123 pairs for easy import into the environment.
1126 .BR \-E ", " \-\-examine
1127 Print contents of the metadata stored on the named device(s).
1128 Note the contrast between
1133 applies to devices which are components of an array, while
1135 applies to a whole array which is currently active.
1138 If an array was created on a SPARC machine with a 2.2 Linux kernel
1139 patched with RAID support, the superblock will have been created
1140 incorrectly, or at least incompatibly with 2.4 and later kernels.
1145 will fix the superblock before displaying it. If this appears to do
1146 the right thing, then the array can be successfully assembled using
1147 .BR "\-\-assemble \-\-update=sparc2.2" .
1150 .BR \-X ", " \-\-examine\-bitmap
1151 Report information about a bitmap file.
1152 The argument is either an external bitmap file or an array component
1153 in case of an internal bitmap. Note that running this on an array
1156 does not report the bitmap for that array.
1159 .BR \-R ", " \-\-run
1160 start a partially assembled array. If
1162 did not find enough devices to fully start the array, it might leaving
1163 it partially assembled. If you wish, you can then use
1165 to start the array in degraded mode.
1168 .BR \-S ", " \-\-stop
1169 deactivate array, releasing all resources.
1172 .BR \-o ", " \-\-readonly
1173 mark array as readonly.
1176 .BR \-w ", " \-\-readwrite
1177 mark array as readwrite.
1180 .B \-\-zero\-superblock
1181 If the device contains a valid md superblock, the block is
1182 overwritten with zeros. With
1184 the block where the superblock would be is overwritten even if it
1185 doesn't appear to be valid.
1188 .BR \-t ", " \-\-test
1193 is set to reflect the status of the device. See below in
1198 .BR \-W ", " \-\-wait
1199 For each md device given, wait for any resync, recovery, or reshape
1200 activity to finish before returning.
1202 will return with success if it actually waited for every device
1203 listed, otherwise it will return failure.
1207 For each md device given, or each device in /proc/mdstat if
1209 is given, arrange for the array to be marked clean as soon as possible.
1210 Also, quiesce resync so that the monitor for external metadata arrays
1211 (mdmon) has an opportunity to checkpoint the resync position.
1213 will return with success if the array uses external metadata and we
1214 successfully waited. For native arrays this returns immediately as the
1215 kernel handles both dirty-clean transitions and resync checkpointing in
1216 the kernel at shutdown. No action is taken if safe-mode handling is
1219 .SH For Incremental Assembly mode:
1221 .BR \-\-rebuild\-map ", " \-r
1222 Rebuild the map file
1223 .RB ( /var/run/mdadm/map )
1226 uses to help track which arrays are currently being assembled.
1229 .BR \-\-run ", " \-R
1230 Run any array assembled as soon as a minimal number of devices are
1231 available, rather than waiting until all expected devices are present.
1235 This allows the hot-plug system to prevent arrays from running when it knows
1236 that more disks may arrive later in the discovery process.
1239 .BR \-\-scan ", " \-s
1240 Only meaningful with
1244 file for arrays that are being incrementally assembled and will try to
1245 start any that are not already started. If any such array is listed
1248 as requiring an external bitmap, that bitmap will be attached first.
1250 .SH For Monitor mode:
1252 .BR \-m ", " \-\-mail
1253 Give a mail address to send alerts to.
1256 .BR \-p ", " \-\-program ", " \-\-alert
1257 Give a program to be run whenever an event is detected.
1260 .BR \-y ", " \-\-syslog
1261 Cause all events to be reported through 'syslog'. The messages have
1262 facility of 'daemon' and varying priorities.
1265 .BR \-d ", " \-\-delay
1266 Give a delay in seconds.
1268 polls the md arrays and then waits this many seconds before polling
1269 again. The default is 60 seconds. Since 2.6.16, there is no need to
1270 reduce this as the kernel alerts
1272 immediately when there is any change.
1275 .BR \-r ", " \-\-increment
1276 Give a percentage increment.
1278 will generate RebuildNN events with the given percentage increment.
1281 .BR \-f ", " \-\-daemonise
1284 to run as a background daemon if it decides to monitor anything. This
1285 causes it to fork and run in the child, and to disconnect from the
1286 terminal. The process id of the child is written to stdout.
1289 which will only continue monitoring if a mail address or alert program
1290 is found in the config file.
1293 .BR \-i ", " \-\-pid\-file
1296 is running in daemon mode, write the pid of the daemon process to
1297 the specified file, instead of printing it on standard output.
1300 .BR \-1 ", " \-\-oneshot
1301 Check arrays only once. This will generate
1303 events and more significantly
1309 .B " mdadm \-\-monitor \-\-scan \-1"
1311 from a cron script will ensure regular notification of any degraded arrays.
1314 .BR \-t ", " \-\-test
1317 alert for every array found at startup. This alert gets mailed and
1318 passed to the alert program. This can be used for testing that alert
1319 message do get through successfully.
1325 .B mdadm \-\-assemble
1326 .I md-device options-and-component-devices...
1329 .B mdadm \-\-assemble \-\-scan
1330 .I md-devices-and-options...
1333 .B mdadm \-\-assemble \-\-scan
1337 This usage assembles one or more RAID arrays from pre-existing components.
1338 For each array, mdadm needs to know the md device, the identity of the
1339 array, and a number of component-devices. These can be found in a number of ways.
1341 In the first usage example (without the
1343 the first device given is the md device.
1344 In the second usage example, all devices listed are treated as md
1345 devices and assembly is attempted.
1346 In the third (where no devices are listed) all md devices that are
1347 listed in the configuration file are assembled. If not arrays are
1348 described by the configuration file, then any arrays that
1349 can be found on unused devices will be assembled.
1351 If precisely one device is listed, but
1357 was given and identity information is extracted from the configuration file.
1359 The identity can be given with the
1365 option, will be taken from the md-device record in the config file, or
1366 will be taken from the super block of the first component-device
1367 listed on the command line.
1369 Devices can be given on the
1371 command line or in the config file. Only devices which have an md
1372 superblock which contains the right identity will be considered for
1375 The config file is only used if explicitly named with
1377 or requested with (a possibly implicit)
1382 .B /etc/mdadm/mdadm.conf
1387 is not given, then the config file will only be used to find the
1388 identity of md arrays.
1390 Normally the array will be started after it is assembled. However if
1392 is not given and not all expected drives were listed, then the array
1393 is not started (to guard against usage errors). To insist that the
1394 array be started in this case (as may work for RAID1, 4, 5, 6, or 10),
1403 does not create any entries in
1407 It does record information in
1408 .B /var/run/mdadm/map
1411 to choose the correct name.
1415 detects that udev is not configured, it will create the devices in
1419 In Linux kernels prior to version 2.6.28 there were two distinctly
1420 different types of md devices that could be created: one that could be
1421 partitioned using standard partitioning tools and one that could not.
1422 Since 2.6.28 that distinction is no longer relevant as both type of
1423 devices can be partitioned.
1425 will normally create the type that originally could not be partitioned
1426 as it has a well defined major number (9).
1428 Prior to 2.6.28, it is important that mdadm chooses the correct type
1429 of array device to use. This can be controlled with the
1431 option. In particular, a value of "mdp" or "part" or "p" tells mdadm
1432 to use a partitionable device rather than the default.
1434 In the no-udev case, the value given to
1436 can be suffixed by a number. This tells
1438 to create that number of partition devices rather than the default of 4.
1442 can also be given in the configuration file as a word starting
1444 on the ARRAY line for the relevant array.
1451 and no devices are listed,
1453 will first attempt to assemble all the arrays listed in the config
1456 In no array at listed in the config (other than those marked
1458 it will look through the available devices for possible arrays and
1459 will try to assemble anything that it finds. Arrays which are tagged
1460 as belonging to the given homehost will be assembled and started
1461 normally. Arrays which do not obviously belong to this host are given
1462 names that are expected not to conflict with anything local, and are
1463 started "read-auto" so that nothing is written to any device until the
1464 array is written to. i.e. automatic resync etc is delayed.
1468 finds a consistent set of devices that look like they should comprise
1469 an array, and if the superblock is tagged as belonging to the given
1470 home host, it will automatically choose a device name and try to
1471 assemble the array. If the array uses version-0.90 metadata, then the
1473 number as recorded in the superblock is used to create a name in
1477 If the array uses version-1 metadata, then the
1479 from the superblock is used to similarly create a name in
1481 (the name will have any 'host' prefix stripped first).
1486 cannot find any array for the given host at all, and if
1487 .B \-\-auto\-update\-homehost
1490 will search again for any array (not just an array created for this
1491 host) and will assemble each assuming
1492 .BR \-\-update=homehost .
1493 This will change the host tag in the superblock so that on the next run,
1494 these arrays will be found without the second pass. The intention of
1495 this feature is to support transitioning a set of md arrays to using
1498 The reason for requiring arrays to be tagged with the homehost for
1499 auto assembly is to guard against problems that can arise when moving
1500 devices from one host to another.
1511 .BI \-\-raid\-devices= Z
1515 This usage is similar to
1517 The difference is that it creates an array without a superblock. With
1518 these arrays there is no difference between initially creating the array and
1519 subsequently assembling the array, except that hopefully there is useful
1520 data there in the second case.
1522 The level may raid0, linear, raid1, raid10, multipath, or faulty, or
1523 one of their synonyms. All devices must be listed and the array will
1524 be started once complete. It will often be appropriate to use
1525 .B \-\-assume\-clean
1526 with levels raid1 or raid10.
1537 .BI \-\-raid\-devices= Z
1541 This usage will initialise a new md array, associate some devices with
1542 it, and activate the array.
1544 The named device will normally not exist when
1545 .I "mdadm \-\-create"
1546 is run, but will be created by
1548 once the array becomes active.
1550 As devices are added, they are checked to see if they contain RAID
1551 superblocks or filesystems. They are also checked to see if the variance in
1552 device size exceeds 1%.
1554 If any discrepancy is found, the array will not automatically be run, though
1557 can override this caution.
1559 To create a "degraded" array in which some devices are missing, simply
1560 give the word "\fBmissing\fP"
1561 in place of a device name. This will cause
1563 to leave the corresponding slot in the array empty.
1564 For a RAID4 or RAID5 array at most one slot can be
1565 "\fBmissing\fP"; for a RAID6 array at most two slots.
1566 For a RAID1 array, only one real device needs to be given. All of the
1570 When creating a RAID5 array,
1572 will automatically create a degraded array with an extra spare drive.
1573 This is because building the spare into a degraded array is in general
1574 faster than resyncing the parity on a non-degraded, but not clean,
1575 array. This feature can be overridden with the
1579 When creating an array with version-1 metadata a name for the array is
1581 If this is not given with the
1585 will choose a name based on the last component of the name of the
1586 device being created. So if
1588 is being created, then the name
1593 is being created, then the name
1597 When creating a partition based array, using
1599 with version-1.x metadata, the partition type should be set to
1601 (non fs-data). This type selection allows for greater precision since
1602 using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)],
1603 might create problems in the event of array recovery through a live cdrom.
1605 A new array will normally get a randomly assigned 128bit UUID which is
1606 very likely to be unique. If you have a specific need, you can choose
1607 a UUID for the array by giving the
1609 option. Be warned that creating two arrays with the same UUID is a
1610 recipe for disaster. Also, using
1612 when creating a v0.90 array will silently override any
1617 .\"option is given, it is not necessary to list any component-devices in this command.
1618 .\"They can be added later, before a
1622 .\"is given, the apparent size of the smallest drive given is used.
1624 When creating an array within a
1627 can be given either the list of devices to use, or simply the name of
1628 the container. The former case gives control over which devices in
1629 the container will be used for the array. The latter case allows
1631 to automatically choose which devices to use based on how much spare
1634 The General Management options that are valid with
1639 insist on running the array even if some devices look like they might
1644 start the array readonly \(em not supported yet.
1651 .I options... devices...
1654 This usage will allow individual devices in an array to be failed,
1655 removed or added. It is possible to perform multiple operations with
1656 on command. For example:
1658 .B " mdadm /dev/md0 \-f /dev/hda1 \-r /dev/hda1 \-a /dev/hda1"
1664 and will then remove it from the array and finally add it back
1665 in as a spare. However only one md array can be affected by a single
1668 When a device is added to an active array, mdadm checks to see if it
1669 has metadata on it which suggests that it was recently a member of the
1670 array. If it does, it tried to "re-add" the device. If there have
1671 been no changes since the device was removed, or if the array has a
1672 write-intent bitmap which has recorded whatever changes there were,
1673 then the device will immediately become a full member of the array and
1674 those differences recorded in the bitmap will be resolved.
1684 MISC mode includes a number of distinct operations that
1685 operate on distinct devices. The operations are:
1688 The device is examined to see if it is
1689 (1) an active md array, or
1690 (2) a component of an md array.
1691 The information discovered is reported.
1695 The device should be an active md device.
1697 will display a detailed description of the array.
1701 will cause the output to be less detailed and the format to be
1702 suitable for inclusion in
1703 .BR /etc/mdadm.conf .
1706 will normally be 0 unless
1708 failed to get useful information about the device(s); however, if the
1710 option is given, then the exit status will be:
1714 The array is functioning normally.
1717 The array has at least one failed device.
1720 The array has multiple failed devices such that it is unusable.
1723 There was an error while trying to get information about the device.
1727 .B \-\-detail\-platform
1728 Print detail of the platform's RAID capabilities (firmware / hardware
1729 topology). If the metadata is specified with
1733 then the return status will be:
1737 metadata successfully enumerated its platform components on this system
1740 metadata is platform independent
1743 metadata failed to find its platform components on this system
1748 The device should be a component of an md array.
1750 will read the md superblock of the device and display the contents.
1755 is given, then multiple devices that are components of the one array
1756 are grouped together and reported in a single entry suitable
1758 .BR /etc/mdadm.conf .
1762 without listing any devices will cause all devices listed in the
1763 config file to be examined.
1767 The devices should be active md arrays which will be deactivated, as
1768 long as they are not currently in use.
1772 This will fully activate a partially assembled md array.
1776 This will mark an active array as read-only, providing that it is
1777 not currently being used.
1783 array back to being read/write.
1787 For all operations except
1790 will cause the operation to be applied to all arrays listed in
1795 causes all devices listed in the config file to be examined.
1801 .B mdadm \-\-monitor
1802 .I options... devices...
1807 to periodically poll a number of md arrays and to report on any events
1810 will never exit once it decides that there are arrays to be checked,
1811 so it should normally be run in the background.
1813 As well as reporting events,
1815 may move a spare drive from one array to another if they are in the
1818 and if the destination array has a failed drive but no spares.
1820 If any devices are listed on the command line,
1822 will only monitor those devices. Otherwise all arrays listed in the
1823 configuration file will be monitored. Further, if
1825 is given, then any other md devices that appear in
1827 will also be monitored.
1829 The result of monitoring the arrays is the generation of events.
1830 These events are passed to a separate program (if specified) and may
1831 be mailed to a given E-mail address.
1833 When passing events to a program, the program is run once for each event,
1834 and is given 2 or 3 command-line arguments: the first is the
1835 name of the event (see below), the second is the name of the
1836 md device which is affected, and the third is the name of a related
1837 device if relevant (such as a component device that has failed).
1841 is given, then a program or an E-mail address must be specified on the
1842 command line or in the config file. If neither are available, then
1844 will not monitor anything.
1848 will continue monitoring as long as something was found to monitor. If
1849 no program or email is given, then each event is reported to
1852 The different events are:
1856 .B DeviceDisappeared
1857 An md array which previously was configured appears to no longer be
1858 configured. (syslog priority: Critical)
1862 was told to monitor an array which is RAID0 or Linear, then it will
1864 .B DeviceDisappeared
1865 with the extra information
1867 This is because RAID0 and Linear do not support the device-failed,
1868 hot-spare and resync operations which are monitored.
1872 An md array started reconstruction. (syslog priority: Warning)
1878 is a two-digit number (ie. 05, 48). This indicates that rebuild
1879 has passed that many percent of the total. The events are generated
1880 with fixed increment since 0. Increment size may be specified with
1881 a commandline option (default is 20). (syslog priority: Warning)
1885 An md array that was rebuilding, isn't any more, either because it
1886 finished normally or was aborted. (syslog priority: Warning)
1890 An active component device of an array has been marked as
1891 faulty. (syslog priority: Critical)
1895 A spare component device which was being rebuilt to replace a faulty
1896 device has failed. (syslog priority: Critical)
1900 A spare component device which was being rebuilt to replace a faulty
1901 device has been successfully rebuilt and has been made active.
1902 (syslog priority: Info)
1906 A new md array has been detected in the
1908 file. (syslog priority: Info)
1912 A newly noticed array appears to be degraded. This message is not
1915 notices a drive failure which causes degradation, but only when
1917 notices that an array is degraded when it first sees the array.
1918 (syslog priority: Critical)
1922 A spare drive has been moved from one array in a
1924 to another to allow a failed drive to be replaced.
1925 (syslog priority: Info)
1931 has been told, via the config file, that an array should have a certain
1932 number of spare devices, and
1934 detects that it has fewer than this number when it first sees the
1935 array, it will report a
1938 (syslog priority: Warning)
1942 An array was found at startup, and the
1945 (syslog priority: Info)
1955 cause Email to be sent. All events cause the program to be run.
1956 The program is run with two or three arguments: the event
1957 name, the array device and possibly a second device.
1959 Each event has an associated array device (e.g.
1961 and possibly a second device. For
1966 the second device is the relevant component device.
1969 the second device is the array that the spare was moved from.
1973 to move spares from one array to another, the different arrays need to
1974 be labeled with the same
1976 in the configuration file. The
1978 name can be any string; it is only necessary that different spare
1979 groups use different names.
1983 detects that an array in a spare group has fewer active
1984 devices than necessary for the complete array, and has no spare
1985 devices, it will look for another array in the same spare group that
1986 has a full complement of working drive and a spare. It will then
1987 attempt to remove the spare from the second drive and add it to the
1989 If the removal succeeds but the adding fails, then it is added back to
1993 The GROW mode is used for changing the size or shape of an active
1995 For this to work, the kernel must support the necessary change.
1996 Various types of growth are being added during 2.6 development,
1997 including restructuring a RAID5 array to have more active devices.
1999 Currently the only support available is to
2001 change the "size" attribute
2002 for RAID1, RAID5 and RAID6.
2004 increase or decrease the "raid\-devices" attribute of RAID1, RAID5,
2007 change the chunk-size and layout of RAID5 and RAID6.
2009 convert between RAID1 and RAID5, and between RAID5 and RAID6.
2011 add a write-intent bitmap to any array which supports these bitmaps, or
2012 remove a write-intent bitmap from such an array.
2015 GROW mode is not currently supported for
2017 or arrays inside containers.
2020 Normally when an array is built the "size" it taken from the smallest
2021 of the drives. If all the small drives in an arrays are, one at a
2022 time, removed and replaced with larger drives, then you could have an
2023 array of large drives with only a small amount used. In this
2024 situation, changing the "size" with "GROW" mode will allow the extra
2025 space to start being used. If the size is increased in this way, a
2026 "resync" process will start to make sure the new parts of the array
2029 Note that when an array changes size, any filesystem that may be
2030 stored in the array will not automatically grow to use the space. The
2031 filesystem will need to be explicitly told to use the extra space.
2033 Also the size of an array cannot be changed while it has an active
2034 bitmap. If an array has a bitmap, it must be removed before the size
2035 can be changed. Once the change it complete a new bitmap can be created.
2037 .SS RAID\-DEVICES CHANGES
2039 A RAID1 array can work with any number of devices from 1 upwards
2040 (though 1 is not very useful). There may be times which you want to
2041 increase or decrease the number of active devices. Note that this is
2042 different to hot-add or hot-remove which changes the number of
2045 When reducing the number of devices in a RAID1 array, the slots which
2046 are to be removed from the array must already be vacant. That is, the
2047 devices which were in those slots must be failed and removed.
2049 When the number of devices is increased, any hot spares that are
2050 present will be activated immediately.
2052 Changing the number of active devices in a RAID5 or RAID6 is much more
2053 effort. Every block in the array will need to be read and written
2054 back to a new location. From 2.6.17, the Linux Kernel is able to
2055 increase the number of devices in a RAID5 safely, including restarting
2056 an interrupted "reshape". From 2.6.31, the Linux Kernel is able to
2057 increase or decrease the number of devices in a RAID5 or RAID6.
2059 When decreasing the number of devices, the size of the array will also
2060 decrease. If there was data in the array, it could get destroyed and
2061 this is not reversible. To help prevent accidents,
2063 requires that the size of the array be decreased first with
2064 .BR "mdadm --grow --array-size" .
2065 This is a reversible change which simply makes the end of the array
2066 inaccessible. The integrity of any data can then be checked before
2067 the non-reversible reduction in the number of devices is request.
2069 When relocating the first few stripes on a RAID5, it is not possible
2070 to keep the data on disk completely consistent and crash-proof. To
2071 provide the required safety, mdadm disables writes to the array while
2072 this "critical section" is reshaped, and takes a backup of the data
2073 that is in that section. This backup is normally stored in any spare
2074 devices that the array has, however it can also be stored in a
2075 separate file specified with the
2077 option. If this option is used, and the system does crash during the
2078 critical period, the same file must be passed to
2080 to restore the backup and reassemble the array.
2084 Changing the RAID level of any array happens instantaneously. However
2085 in the RAID to RAID6 case this requires a non-standard layout of the
2086 RAID6 data, and in the RAID6 to RAID5 case that non-standard layout is
2087 required before the change can be accomplish. So while the level
2088 change is instant, the accompanying layout change can take quite a
2091 .SS CHUNK-SIZE AND LAYOUT CHANGES
2093 Changing the chunk-size of layout without also changing the number of
2094 devices as the same time will involve re-writing all blocks in-place.
2095 To ensure against data loss in the case of a crash, a
2097 must be provided for these changes. Small sections of the array will
2098 be copied to the backup file while they are being rearranged.
2100 If the reshape is interrupted for any reason, this backup file must be
2102 .B "mdadm --assemble"
2103 so the array can be reassembled. Consequently the file cannot be
2104 stored on the device being reshaped.
2109 A write-intent bitmap can be added to, or removed from, an active
2110 array. Either internal bitmaps, or bitmaps stored in a separate file,
2111 can be added. Note that if you add a bitmap stored in a file which is
2112 in a filesystem that is on the RAID array being affected, the system
2113 will deadlock. The bitmap must be on a separate filesystem.
2115 .SH INCREMENTAL MODE
2119 .B mdadm \-\-incremental
2125 .B mdadm \-\-incremental \-\-rebuild
2128 .B mdadm \-\-incremental \-\-run \-\-scan
2131 This mode is designed to be used in conjunction with a device
2132 discovery system. As devices are found in a system, they can be
2134 .B "mdadm \-\-incremental"
2135 to be conditionally added to an appropriate array.
2137 If the device passed is a
2139 device created by a previous call to
2141 then rather than trying to add that device to an array, all the arrays
2142 described by the metadata of the container will be started.
2145 performs a number of tests to determine if the device is part of an
2146 array, and which array it should be part of. If an appropriate array
2147 is found, or can be created,
2149 adds the device to the array and conditionally starts the array.
2153 will only add devices to an array which were previously working
2154 (active or spare) parts of that array. It does not currently support
2155 automatic inclusion of a new drive as a spare in some array.
2159 makes are as follow:
2161 Is the device permitted by
2163 That is, is it listed in a
2165 line in that file. If
2167 is absent then the default it to allow any device. Similar if
2169 contains the special word
2171 then any device is allowed. Otherwise the device name given to
2173 must match one of the names or patterns in a
2178 Does the device have a valid md superblock. If a specific metadata
2179 version is request with
2183 then only that style of metadata is accepted, otherwise
2185 finds any known version of metadata. If no
2187 metadata is found, the device is rejected.
2191 Does the metadata match an expected array?
2192 The metadata can match in two ways. Either there is an array listed
2195 which identifies the array (either by UUID, by name, by device list,
2196 or by minor-number), or the array was created with a
2202 or on the command line.
2205 is not able to positively identify the array as belonging to the
2206 current host, the device will be rejected.
2210 keeps a list of arrays that it has partially assembled in
2211 .B /var/run/mdadm/map
2213 .B /var/run/mdadm.map
2214 if the directory doesn't exist. Or maybe even
2215 .BR /dev/.mdadm.map ).
2216 If no array exists which matches
2217 the metadata on the new device,
2219 must choose a device name and unit number. It does this based on any
2222 or any name information stored in the metadata. If this name
2223 suggests a unit number, that number will be used, otherwise a free
2224 unit number will be chosen. Normally
2226 will prefer to create a partitionable array, however if the
2230 suggests that a non-partitionable array is preferred, that will be
2233 If the array is not found in the config file and its metadata does not
2234 identify it as belonging to the "homehost", then
2236 will choose a name for the array which is certain not to conflict with
2237 any array which does belong to this host. It does this be adding an
2238 underscore and a small number to the name preferred by the metadata.
2240 Once an appropriate array is found or created and the device is added,
2242 must decide if the array is ready to be started. It will
2243 normally compare the number of available (non-spare) devices to the
2244 number of devices that the metadata suggests need to be active. If
2245 there are at least that many, the array will be started. This means
2246 that if any devices are missing the array will not be restarted.
2252 in which case the array will be run as soon as there are enough
2253 devices present for the data to be accessible. For a RAID1, that
2254 means one device will start the array. For a clean RAID5, the array
2255 will be started as soon as all but one drive is present.
2257 Note that neither of these approaches is really ideal. If it can
2258 be known that all device discovery has completed, then
2262 can be run which will try to start all arrays that are being
2263 incrementally assembled. They are started in "read-auto" mode in
2264 which they are read-only until the first write request. This means
2265 that no metadata updates are made and no attempt at resync or recovery
2266 happens. Further devices that are found before the first write can
2267 still be added safely.
2270 This section describes environment variables that affect how mdadm
2275 Setting this value to 1 will prevent mdadm from automatically launching
2276 mdmon. This variable is intended primarily for debugging mdadm/mdmon.
2282 does not create any device nodes in /dev, but leaves that task to
2286 appears not to be configured, or if this environment variable is set
2289 will create and devices that are needed.
2293 .B " mdadm \-\-query /dev/name-of-device"
2295 This will find out if a given device is a RAID array, or is part of
2296 one, and will provide brief information about the device.
2298 .B " mdadm \-\-assemble \-\-scan"
2300 This will assemble and start all arrays listed in the standard config
2301 file. This command will typically go in a system startup file.
2303 .B " mdadm \-\-stop \-\-scan"
2305 This will shut down all arrays that can be shut down (i.e. are not
2306 currently in use). This will typically go in a system shutdown script.
2308 .B " mdadm \-\-follow \-\-scan \-\-delay=120"
2310 If (and only if) there is an Email address or program given in the
2311 standard config file, then
2312 monitor the status of all arrays listed in that file by
2313 polling them ever 2 minutes.
2315 .B " mdadm \-\-create /dev/md0 \-\-level=1 \-\-raid\-devices=2 /dev/hd[ac]1"
2317 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
2320 .B " echo 'DEVICE /dev/hd*[0\-9] /dev/sd*[0\-9]' > mdadm.conf"
2322 .B " mdadm \-\-detail \-\-scan >> mdadm.conf"
2324 This will create a prototype config file that describes currently
2325 active arrays that are known to be made from partitions of IDE or SCSI drives.
2326 This file should be reviewed before being used as it may
2327 contain unwanted detail.
2329 .B " echo 'DEVICE /dev/hd[a\-z] /dev/sd*[a\-z]' > mdadm.conf"
2331 .B " mdadm \-\-examine \-\-scan \-\-config=mdadm.conf >> mdadm.conf"
2333 This will find arrays which could be assembled from existing IDE and
2334 SCSI whole drives (not partitions), and store the information in the
2335 format of a config file.
2336 This file is very likely to contain unwanted detail, particularly
2339 entries. It should be reviewed and edited before being used as an
2342 .B " mdadm \-\-examine \-\-brief \-\-scan \-\-config=partitions"
2344 .B " mdadm \-Ebsc partitions"
2346 Create a list of devices by reading
2347 .BR /proc/partitions ,
2348 scan these for RAID superblocks, and printout a brief listing of all
2351 .B " mdadm \-Ac partitions \-m 0 /dev/md0"
2353 Scan all partitions and devices listed in
2354 .BR /proc/partitions
2357 out of all such devices with a RAID superblock with a minor number of 0.
2359 .B " mdadm \-\-monitor \-\-scan \-\-daemonise > /var/run/mdadm"
2361 If config file contains a mail address or alert program, run mdadm in
2362 the background in monitor mode monitoring all md devices. Also write
2363 pid of mdadm daemon to
2364 .BR /var/run/mdadm .
2366 .B " mdadm \-Iq /dev/somedevice"
2368 Try to incorporate newly discovered device into some array as
2371 .B " mdadm \-\-incremental \-\-rebuild \-\-run \-\-scan"
2373 Rebuild the array map from any current arrays, and then start any that
2376 .B " mdadm /dev/md4 --fail detached --remove detached"
2378 Any devices which are components of /dev/md4 will be marked as faulty
2379 and then remove from the array.
2381 .B " mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4
2385 which is currently a RAID5 array will be converted to RAID6. There
2386 should normally already be a spare drive attached to the array as a
2387 RAID6 needs one more drive than a matching RAID5.
2389 .B " mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]"
2391 Create a DDF array over 6 devices.
2393 .B " mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf"
2395 Create a RAID5 array over any 3 devices in the given DDF set. Use
2396 only 30 gigabytes of each device.
2398 .B " mdadm -A /dev/md/ddf1 /dev/sd[a-f]"
2400 Assemble a pre-exist ddf array.
2402 .B " mdadm -I /dev/md/ddf1"
2404 Assemble all arrays contained in the ddf array, assigning names as
2407 .B " mdadm \-\-create \-\-help"
2409 Provide help about the Create mode.
2411 .B " mdadm \-\-config \-\-help"
2413 Provide help about the format of the config file.
2415 .B " mdadm \-\-help"
2417 Provide general help.
2427 lists all active md devices with information about them.
2429 uses this to find arrays when
2431 is given in Misc mode, and to monitor array reconstruction
2436 The config file lists which devices may be scanned to see if
2437 they contain MD super block, and gives identifying information
2438 (e.g. UUID) about known MD arrays. See
2442 .SS /var/run/mdadm/map
2445 mode is used, this file gets a list of arrays currently being created.
2448 does not exist as a directory, then
2449 .B /var/run/mdadm.map
2452 is not available (as may be the case during early boot),
2454 is used on the basis that
2456 is usually available very early in boot.
2461 understand two sorts of names for array devices.
2463 The first is the so-called 'standard' format name, which matches the
2464 names used by the kernel and which appear in
2467 The second sort can be freely chosen, but must reside in
2469 When giving a device name to
2471 to create or assemble an array, either full path name such as
2475 can be given, or just the suffix of the second sort of name, such as
2481 chooses device names during auto-assembly or incremental assembly, it
2482 will sometimes add a small sequence number to the end of the name to
2483 avoid conflicted between multiple arrays that have the same name. If
2485 can reasonably determine that the array really is meant for this host,
2486 either by a hostname in the metadata, or by the presence of the array
2487 in /etc/mdadm.conf, then it will leave off the suffix if possible.
2488 Also if the homehost is specified as
2491 will only use a suffix if a different array of the same name already
2492 exists or is listed in the config file.
2494 The standard names for non-partitioned arrays (the only sort of md
2495 array available in 2.4 and earlier) are of the form
2499 where NN is a number.
2500 The standard names for partitionable arrays (as available from 2.6
2501 onwards) are of the form
2505 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2507 From kernel version, 2.6.28 the "non-partitioned array" can actually
2508 be partitioned. So the "md_dNN" names are no longer needed, and
2509 partitions such as "/dev/mdNNpXX" are possible.
2513 was previously known as
2517 is completely separate from the
2519 package, and does not use the
2521 configuration file at all.
2524 For further information on mdadm usage, MD and the various levels of
2527 .B http://linux\-raid.osdl.org/
2529 (based upon Jakob \(/Ostergaard's Software\-RAID.HOWTO)
2531 .\"for new releases of the RAID driver check out:
2534 .\".UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2535 .\"ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2540 .\".UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2541 .\"http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2544 The latest version of
2546 should always be available from
2548 .B http://www.kernel.org/pub/linux/utils/raid/mdadm/
projects / thirdparty / mdadm.git / blob