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 there-of) 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
45 is not a Software RAID mechanism, but does involve
48 each device is a path to one common physical storage device.
51 is also not true RAID, and it only involves one device. It
52 provides a layer over a true device that can be used to inject faults.
55 '''is a program that can be used to create, manage, and monitor
57 '''such it provides a similar set of functionality to the
60 '''The key differences between
67 '''is a single program and not a collection of programs.
70 '''can perform (almost) all of its functions without having a
71 '''configuration file and does not use one by default. Also
73 '''helps with management of the configuration
77 '''can provide information about your arrays (through Query, Detail, and Examine)
87 '''configuration file, at all. It has a different configuration file
88 '''with a different format and a different purpose.
91 mdadm has several major modes of operation:
94 Assemble the parts of a previously created
95 array into an active array. Components can be explicitly given
96 or can be searched for.
98 checks that the components
99 do form a bona fide array, and can, on request, fiddle superblock
100 information so as to assemble a faulty array.
104 Build an array that doesn't have per-device superblocks. For these
107 cannot differentiate between initial creation and subsequent assembly
108 of an array. It also cannot perform any checks that appropriate
109 devices have been requested. Because of this, the
111 mode should only be used together with a complete understanding of
116 Create a new array with per-device superblocks.
118 '''in several step create-add-add-run or it can all happen with one command.
121 .B "Follow or Monitor"
122 Monitor one or more md devices and act on any state changes. This is
123 only meaningful for raid1, 4, 5, 6, 10 or multipath arrays as
124 only these have interesting state. raid0 or linear never have
125 missing, spare, or failed drives, so there is nothing to monitor.
129 Grow (or shrink) an array, or otherwise reshape it in some way.
130 Currently supported growth options including changing the active size
131 of component devices in RAID level 1/4/5/6 and changing the number of
132 active devices in RAID1.
135 .B "Incremental Assembly"
136 Add a single device to an appropriate array. If the addition of the
137 device makes the array runnable, the array will be started.
138 This provides a convenient interface to a
140 system. As each device is detected,
142 has a chance to include it in some array as appropriate.
146 This is for doing things to specific components of an array such as
147 adding new spares and removing faulty devices.
151 This is an 'everything else' mode that supports operations on active
152 arrays, operations on component devices such as erasing old superblocks, and
153 information gathering operations.
154 '''This mode allows operations on independent devices such as examine MD
155 '''superblocks, erasing old superblocks and stopping active arrays.
159 .SH Options for selecting a mode are:
162 .BR -A ", " --assemble
163 Assemble a pre-existing array.
167 Build a legacy array without superblocks.
174 .BR -F ", " --follow ", " --monitor
181 Change the size or shape of an active array.
184 .BE -I ", " --incremental
185 Add a single device into an appropriate array, and possibly start the array.
188 If a device is given before any options, or if the first option is
193 then the MANAGE mode is assume.
194 Anything other than these will cause the
198 .SH Options that are not mode-specific are:
202 Display general help message or, after one of the above options, a
203 mode specific help message.
207 Display more detailed help about command line parsing and some commonly
211 .BR -V ", " --version
212 Print version information for mdadm.
215 .BR -v ", " --verbose
216 Be more verbose about what is happening. This can be used twice to be
218 The extra verbosity currently only affects
221 .BR "--examine --scan" .
225 Avoid printing purely informative messages. With this,
227 will be silent unless there is something really important to report.
231 Be less verbose. This is used with
239 gives an intermediate level of verbosity.
243 Be more forceful about certain operations. See the various modes of
244 the exact meaning of this option in different contexts.
247 .BR -c ", " --config=
248 Specify the config file. Default is to use
249 .BR /etc/mdadm.conf ,
250 or if that is missing, then
251 .BR /etc/mdadm/mdadm.conf .
252 If the config file given is
254 then nothing will be read, but
256 will act as though the config file contained exactly
257 .B "DEVICE partitions"
260 to find a list of devices to scan.
263 is given for the config file, then
265 will act as though the config file were empty.
271 for missing information.
272 In general, this option gives
274 permission to get any missing information, like component devices,
275 array devices, array identities, and alert destination from the
277 .BR /etc/mdadm.conf .
278 One exception is MISC mode when using
284 says to get a list of array devices from
288 .B -e ", " --metadata=
289 Declare the style of superblock (raid metadata) to be used. The
290 default is 0.90 for --create, and to guess for other operations.
291 The default can be overridden by setting the
300 .IP "0, 0.90, default"
301 Use the original 0.90 format superblock. This format limits arrays to
302 28 componenet devices and limits component devices of levels 1 and
303 greater to 2 terabytes.
304 .IP "1, 1.0, 1.1, 1.2"
305 Use the new version-1 format superblock. This has few restrictions.
306 The different subversion store the superblock at different locations
307 on the device, either at the end (for 1.0), at the start (for 1.1) or
308 4K from the start (for 1.2).
313 This will over-ride any
315 setting in the config file and provides the identify of the host which
316 should be considered the home for any arrays.
318 When creating an array, the
320 will be recorded in the superblock. For version-1 superblocks, it will
321 be prefixed to the array name. For version-0.90 superblocks part of
322 the SHA1 hash of the hostname will be stored in the later half of the
325 When reporting information about an array, any array which is tagged
326 for the given homehost will be reported as such.
328 When using Auto-Assemble, only arrays tagged for the given homehost
331 .SH For create, build, or grow:
334 .BR -n ", " --raid-devices=
335 Specify the number of active devices in the array. This, plus the
336 number of spare devices (see below) must equal the number of
338 (including "\fBmissing\fP" devices)
339 that are listed on the command line for
341 Setting a value of 1 is probably
342 a mistake and so requires that
344 be specified first. A value of 1 will then be allowed for linear,
345 multipath, raid0 and raid1. It is never allowed for raid4 or raid5.
347 This number can only be changed using
349 for RAID1 arrays, and only on kernels which provide necessary support.
352 .BR -x ", " --spare-devices=
353 Specify the number of spare (eXtra) devices in the initial array.
354 Spares can also be added
355 and removed later. The number of component devices listed
356 on the command line must equal the number of raid devices plus the
357 number of spare devices.
362 Amount (in Kibibytes) of space to use from each drive in RAID1/4/5/6.
363 This must be a multiple of the chunk size, and must leave about 128Kb
364 of space at the end of the drive for the RAID superblock.
365 If this is not specified
366 (as it normally is not) the smallest drive (or partition) sets the
367 size, though if there is a variance among the drives of greater than 1%, a warning is
370 This value can be set with
372 for RAID level 1/4/5/6. If the array was created with a size smaller
373 than the currently active drives, the extra space can be accessed
376 The size can be given as
378 which means to choose the largest size that fits on all current drives.
382 Specify chunk size of kibibytes. The default is 64.
386 Specify rounding factor for linear array (==chunk size)
390 Set raid level. When used with
392 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
393 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty. Obviously some of these are synonymous.
397 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
399 Not yet supported with
403 .BR -p ", " --layout=
404 This option configures the fine details of data layout for raid5,
405 and raid10 arrays, and controls the failure modes for
408 The layout of the raid5 parity block can be one of
413 la, ra, ls, rs. The default is left-symmetric.
415 When setting the failure mode for
433 Each mode can be followed by a number which is used as a period
434 between fault generation. Without a number, the fault is generated
435 once on the first relevant request. With a number, the fault will be
436 generated after that many request, and will continue to be generated
437 every time the period elapses.
439 Multiple failure modes can be current simultaneously by using the
440 "--grow" option to set subsequent failure modes.
442 "clear" or "none" will remove any pending or periodic failure modes,
443 and "flush" will clear any persistent faults.
445 To set the parity with "--grow", the level of the array ("faulty")
446 must be specified before the fault mode is specified.
448 Finally, the layout options for RAID10 are one of 'n', 'o' or 'p' followed
449 by a small number. The default is 'n2'.
452 signals 'near' copies. Multiple copies of one data block are at
453 similar offsets in different devices.
456 signals 'offset' copies. Rather than the chunks being duplicated
457 within a stripe, whole stripes are duplicated but are rotated by one
458 device so duplicate blocks are on different devices. Thus subsequent
459 copies of a block are in the next drive, and are one chunk further
464 (multiple copies have very different offsets). See md(4) for more
465 detail about 'near' and 'far'.
467 The number is the number of copies of each datablock. 2 is normal, 3
468 can be useful. This number can be at most equal to the number of
469 devices in the array. It does not need to divide evenly into that
470 number (e.g. it is perfectly legal to have an 'n2' layout for an array
471 with an odd number of devices).
475 same as --layout (thus explaining the p of
479 .BR -b ", " --bitmap=
480 Specify a file to store a write-intent bitmap in. The file should not
481 exist unless --force is also given. The same file should be provided
482 when assembling the array. If the word
484 is given, then the bitmap is stored with the metadata on the array,
485 and so is replicated on all devices. If the word
489 mode, then any bitmap that is present is removed.
491 To help catch typing errors, the filename must contain at least one
492 slash ('/') if it is a real file (not 'internal' or 'none').
494 Note: external bitmaps are only known to work on ext2 and ext3.
495 Storing bitmap files on other filesystems may result in serious problems.
499 Set the chunksize of the bitmap. Each bit corresponds to that many
500 Kilobytes of storage.
501 When using a file based bitmap, the default is to use the smallest
502 size that is atleast 4 and requires no more than 2^21 chunks.
505 bitmap, the chunksize is automatically determined to make best use of
510 .BR -W ", " --write-mostly
511 subsequent devices lists in a
516 command will be flagged as 'write-mostly'. This is valid for RAID1
517 only and means that the 'md' driver will avoid reading from these
518 devices if at all possible. This can be useful if mirroring over a
523 Specify that write-behind mode should be enabled (valid for RAID1
524 only). If an argument is specified, it will set the maximum number
525 of outstanding writes allowed. The default value is 256.
526 A write-intent bitmap is required in order to use write-behind
527 mode, and write-behind is only attempted on drives marked as
534 that the array pre-existed and is known to be clean. It can be useful
535 when trying to recover from a major failure as you can be sure that no
536 data will be affected unless you actually write to the array. It can
537 also be used when creating a RAID1 or RAID10 if you want to avoid the
538 initial resync, however this practice - while normally safe - is not
539 recommended. Use this ony if you really know what you are doing.
543 This is needed when --grow is used to increase the number of
544 raid-devices in a RAID5 if there are no spare devices available.
545 See the section below on RAID_DEVICE CHANGES. The file should be
546 stored on a separate device, not on the raid array being reshaped.
552 for the array. This is currently only effective when creating an
553 array with a version-1 superblock. The name is a simple textual
554 string that can be used to identify array components when assembling.
560 run the array, even if some of the components
561 appear to be active in another array or filesystem. Normally
563 will ask for confirmation before including such components in an
564 array. This option causes that question to be suppressed.
570 accept the geometry and layout specified without question. Normally
572 will not allow creation of an array with only one device, and will try
573 to create a raid5 array with one missing drive (as this makes the
574 initial resync work faster). With
577 will not try to be so clever.
580 .BR -a ", " "--auto{=no,yes,md,mdp,part,p}{NN}"
581 Instruct mdadm to create the device file if needed, possibly allocating
582 an unused minor number. "md" causes a non-partitionable array
583 to be used. "mdp", "part" or "p" causes a partitionable array (2.6 and
584 later) to be used. "yes" requires the named md device to have
585 a 'standard' format, and the type and minor number will be determined
586 from this. See DEVICE NAMES below.
588 The argument can also come immediately after
591 If --auto is not given on the command line or in the config file, then
597 is also given, then any
599 entries in the config file will over-ride the
601 instruction given on the command line.
603 For partitionable arrays,
605 will create the device file for the whole array and for the first 4
606 partitions. A different number of partitions can be specified at the
607 end of this option (e.g.
609 If the device name ends with a digit, the partition names add a 'p',
610 and a number, e.g. "/dev/home1p3". If there is no
611 trailing digit, then the partition names just have a number added,
612 e.g. "/dev/scratch3".
614 If the md device name is in a 'standard' format as described in DEVICE
615 NAMES, then it will be created, if necessary, with the appropriate
616 number based on that name. If the device name is not in one of these
617 formats, then a unused minor number will be allocated. The minor
618 number will be considered unused if there is no active array for that
619 number, and there is no entry in /dev for that number and with a
630 it will also create symlinks from
632 with names starting with
640 to enforce this even if it is suppressing
648 uuid of array to assemble. Devices which don't have this uuid are
652 .BR -m ", " --super-minor=
653 Minor number of device that array was created for. Devices which
654 don't have this minor number are excluded. If you create an array as
655 /dev/md1, then all superblocks will contain the minor number 1, even if
656 the array is later assembled as /dev/md2.
658 Giving the literal word "dev" for
662 to use the minor number of the md device that is being assembled.
666 will look for super blocks with a minor number of 0.
670 Specify the name of the array to assemble. This must be the name
671 that was specified when creating the array. It must either match
672 then name stored in the superblock exactly, or it must match
675 is added to the start of the given name.
679 Assemble the array even if some superblocks appear out-of-date
683 Attempt to start the array even if fewer drives were given than were
684 present last time the array was active. Normally if not all the
685 expected drives are found and
687 is not used, then the array will be assembled but not started.
690 an attempt will be made to start it anyway.
694 This is the reverse of
696 in that it inhibits the started if array unless all expected drives
697 are present. This is only needed with
699 and can be used if you physical connections to devices are
700 not as reliable as you would like.
703 .BR -a ", " "--auto{=no,yes,md,mdp,part}"
704 See this option under Create and Build options.
707 .BR -b ", " --bitmap=
708 Specify the bitmap file that was given when the array was created. If
711 bitmap, there is no need to specify this when assembling the array.
717 was used to grow the number of raid-devices in a RAID5, and the system
718 crashed during the critical section, then the same
720 must be presented to --assemble to allow possibly corrupted data to be
724 .BR -U ", " --update=
725 Update the superblock on each device while assembling the array. The
726 argument given to this flag can be one of
740 option will adjust the superblock of an array what was created on a Sparc
741 machine running a patched 2.2 Linux kernel. This kernel got the
742 alignment of part of the superblock wrong. You can use the
743 .B "--examine --sparc2.2"
746 to see what effect this would have.
750 option will update the
752 field on each superblock to match the minor number of the array being
754 This can be useful if
756 reports a different "Preferred Minor" to
758 In some cases this update will be performed automatically
759 by the kernel driver. In particular the update happens automatically
760 at the first write to an array with redundancy (RAID level 1 or
761 greater) on a 2.6 (or later) kernel.
765 option will change the uuid of the array. If a UUID is given with the
766 "--uuid" option that UUID will be used as a new UUID and will
768 be used to help identify the devices in the array.
769 If no "--uuid" is given, a random uuid is chosen.
773 option will change the
775 of the array as stored in the superblock. This is only supported for
776 version-1 superblocks.
780 option will change the
782 as recorded in the superblock. For version-0 superblocks, this is the
783 same as updating the UUID.
784 For version-1 superblocks, this involves updating the name.
788 option will cause the array to be marked
790 meaning that any redundancy in the array (e.g. parity for raid5,
791 copies for raid1) may be incorrect. This will cause the raid system
792 to perform a "resync" pass to make sure that all redundant information
797 option allows arrays to be moved between machines with different
799 When assembling such an array for the first time after a move, giving
800 .B "--update=byteorder"
803 to expect superblocks to have their byteorder reversed, and will
804 correct that order before assembling the array. This is only valid
805 with original (Version 0.90) superblocks.
809 option will correct the summaries in the superblock. That is the
810 counts of total, working, active, failed, and spare devices.
814 will rarely be of use. It applies to version 1.1 and 1.2 metadata
815 only (where the metadata is at the start of the device) and is only
816 useful when the component device has changed size (typically become
817 larger). The version 1 metadata records the amount of the device that
818 can be used to store data, so if a device in a version 1.1 or 1.2
819 array becomes larger, the metadata will still be visible, but the
820 extra space will not. In this case it might be useful to assemble the
822 .BR --update=devicesize .
825 to determine the maximum usable amount of space on each device and
826 update the relevant field in the metadata.
829 .B --auto-update-homehost
830 This flag is only meaning with auto-assembly (see discussion below).
831 In that situation, if no suitable arrays are found for this homehost,
833 will recan for any arrays at all and will assemble them and update the
834 homehost to match the current host.
840 hot-add listed devices.
844 re-add a device that was recently removed from an array.
848 remove listed devices. They must not be active. i.e. they should
849 be failed or spare devices.
853 mark listed devices as faulty.
860 Each of these options require that the first device list is the array
861 to be acted upon and the remainder are component devices to be added,
862 removed, or marked as fault. Several different operations can be
863 specified for different devices, e.g.
865 mdadm /dev/md0 --add /dev/sda1 --fail /dev/sdb1 --remove /dev/sdb1
867 Each operation applies to all devices listed until the next
870 If an array is using a write-intent bitmap, then devices which have
871 been removed can be re-added in a way that avoids a full
872 reconstruction but instead just updated the blocks that have changed
873 since the device was removed. For arrays with persistent metadata
874 (superblocks) this is done automatically. For arrays created with
876 mdadm needs to be told that this device we removed recently with
879 Devices can only be removed from an array if they are not in active
880 use. i.e. that must be spares or failed devices. To remove an active
881 device, it must be marked as
889 Examine a device to see
890 (1) if it is an md device and (2) if it is a component of an md
892 Information about what is discovered is presented.
896 Print detail of one or more md devices.
899 .BR -E ", " --examine
900 Print content of md superblock on device(s).
903 If an array was created on a 2.2 Linux kernel patched with RAID
904 support, the superblock will have been created incorrectly, or at
905 least incompatibly with 2.4 and later kernels. Using the
909 will fix the superblock before displaying it. If this appears to do
910 the right thing, then the array can be successfully assembled using
911 .BR "--assemble --update=sparc2.2" .
914 .BR -X ", " --examine-bitmap
915 Report information about a bitmap file.
919 start a partially built array.
923 deactivate array, releasing all resources.
926 .BR -o ", " --readonly
927 mark array as readonly.
930 .BR -w ", " --readwrite
931 mark array as readwrite.
935 If the device contains a valid md superblock, the block is
936 over-written with zeros. With
938 the block where the superblock would be is over-written even if it
939 doesn't appear to be valid.
947 is set to reflect the status of the device.
951 For each md device given, wait for any resync, recovery, or reshape
952 activity to finish before returning.
954 will return with success if it actually waited for every device
955 listed, otherwise it will return failure.
957 .SH For Incremental Assembly mode:
959 .BR --rebuild-map ", " -r
961 .RB ( /var/run/mdadm/map )
964 uses to help track which arrays are currently being assembled.
968 Run any array assembled as soon as a minimal number of devices are
969 available, rather than waiting until all expected devices are present.
977 file for arrays that are being incrementally assembled and will try to
978 start any that are not already started. If any such array is listed
981 as requiring an external bitmap, that bitmap will be attached first.
983 .SH For Monitor mode:
986 Give a mail address to send alerts to.
989 .BR -p ", " --program ", " --alert
990 Give a program to be run whenever an event is detected.
994 Cause all events to be reported through 'syslog'. The messages have
995 facility of 'daemon' and varying priorities.
999 Give a delay in seconds.
1001 polls the md arrays and then waits this many seconds before polling
1002 again. The default is 60 seconds.
1005 .BR -f ", " --daemonise
1008 to run as a background daemon if it decides to monitor anything. This
1009 causes it to fork and run in the child, and to disconnect form the
1010 terminal. The process id of the child is written to stdout.
1013 which will only continue monitoring if a mail address or alert program
1014 is found in the config file.
1017 .BR -i ", " --pid-file
1020 is running in daemon mode, write the pid of the daemon process to
1021 the specified file, instead of printing it on standard output.
1024 .BR -1 ", " --oneshot
1025 Check arrays only once. This will generate
1027 events and more significantly
1033 .B " mdadm --monitor --scan -1"
1035 from a cron script will ensure regular notification of any degraded arrays.
1041 alert for every array found at startup. This alert gets mailed and
1042 passed to the alert program. This can be used for testing that alert
1043 message do get through successfully.
1050 .I md-device options-and-component-devices...
1053 .B mdadm --assemble --scan
1054 .I md-devices-and-options...
1057 .B mdadm --assemble --scan
1061 This usage assembles one or more raid arrays from pre-existing components.
1062 For each array, mdadm needs to know the md device, the identity of the
1063 array, and a number of component-devices. These can be found in a number of ways.
1065 In the first usage example (without the
1067 the first device given is the md device.
1068 In the second usage example, all devices listed are treated as md
1069 devices and assembly is attempted.
1070 In the third (where no devices are listed) all md devices that are
1071 listed in the configuration file are assembled.
1073 If precisely one device is listed, but
1079 was given and identify information is extracted from the configuration file.
1081 The identity can be given with the
1085 option, can be found in the config file, or will be taken from the
1086 super block on the first component-device listed on the command line.
1088 Devices can be given on the
1090 command line or in the config file. Only devices which have an md
1091 superblock which contains the right identity will be considered for
1094 The config file is only used if explicitly named with
1096 or requested with (a possibly implicit)
1104 is not given, then the config file will only be used to find the
1105 identity of md arrays.
1107 Normally the array will be started after it is assembled. However if
1109 is not given and insufficient drives were listed to start a complete
1110 (non-degraded) array, then the array is not started (to guard against
1111 usage errors). To insist that the array be started in this case (as
1112 may work for RAID1, 4, 5, 6, or 10), give the
1116 If the md device does not exist, then it will be created providing the
1117 intent is clear. i.e. the name must be in a standard form, or the
1119 option must be given to clarify how and whether the device should be
1122 This can be useful for handling partitioned devices (which don't have
1123 a stable device number - it can change after a reboot) and when using
1124 "udev" to manage your
1126 tree (udev cannot handle md devices because of the unusual device
1127 initialisation conventions).
1129 If the option to "auto" is "mdp" or "part" or (on the command line
1130 only) "p", then mdadm will create a partitionable array, using the
1131 first free one that is not in use, and does not already have an entry
1132 in /dev (apart from numeric /dev/md* entries).
1134 If the option to "auto" is "yes" or "md" or (on the command line)
1135 nothing, then mdadm will create a traditional, non-partitionable md
1138 It is expected that the "auto" functionality will be used to create
1139 device entries with meaningful names such as "/dev/md/home" or
1140 "/dev/md/root", rather than names based on the numerical array number.
1142 When using this option to create a partitionable array, the device
1143 files for the first 4 partitions are also created. If a different
1144 number is required it can be simply appended to the auto option.
1145 e.g. "auto=part8". Partition names are created by appending a digit
1146 string to the device name, with an intervening "p" if the device name
1151 option is also available in Build and Create modes. As those modes do
1152 not use a config file, the "auto=" config option does not apply to
1160 and no devices are listed,
1162 will first attempt to assemble all the arrays listed in the config
1167 has been specified (either in the config file or on the command line),
1169 will look further for possible arrays and will try to assemble
1170 anything that it finds which is tagged as belonging to the given
1171 homehost. This is the only situation where
1173 will assemble arrays without being given specific device name or
1174 identify information for the array.
1178 finds a consistent set of devices that look like they should comprise
1179 an array, and if the superblock is tagged as belonging to the given
1180 home host, it will automatically choose a device name and try to
1181 assemble the array. If the array uses version-0.90 metadata, then the
1183 number as recorded in the superblock is used to create a name in
1187 If the array uses version-1 metadata, then the
1189 from the superblock is used to similarly create a name in
1191 The name will have any 'host' prefix stripped first.
1195 cannot find any array for the given host at all, and if
1196 .B --auto-update-homehost
1199 will search again for any array (not just an array created for this
1200 host) and will assemble each assuming
1201 .IR --update=homehost .
1202 This will change the host tag in the superblock so that on the next run,
1203 these arrays will be found without the second pass. The intention of
1204 this feature is to support transitioning a set of md arrays to using
1207 The reason for requiring arrays to be tagged with the homehost for
1208 auto assembly is to guard against problems that can arise when moving
1209 devices from one host to another.
1219 .BI --raid-devices= Z
1223 This usage is similar to
1225 The difference is that it creates an array without a superblock. With
1226 these arrays there is no difference between initially creating the array and
1227 subsequently assembling the array, except that hopefully there is useful
1228 data there in the second case.
1230 The level may raid0, linear, multipath, or faulty, or one of their
1231 synonyms. All devices must be listed and the array will be started
1243 .BI --raid-devices= Z
1247 This usage will initialise a new md array, associate some devices with
1248 it, and activate the array.
1252 option is given (as described in more detail in the section on
1253 Assemble mode), then the md device will be created with a suitable
1254 device number if necessary.
1256 As devices are added, they are checked to see if they contain raid
1257 superblocks or filesystems. They are also checked to see if the variance in
1258 device size exceeds 1%.
1260 If any discrepancy is found, the array will not automatically be run, though
1263 can override this caution.
1265 To create a "degraded" array in which some devices are missing, simply
1266 give the word "\fBmissing\fP"
1267 in place of a device name. This will cause
1269 to leave the corresponding slot in the array empty.
1270 For a RAID4 or RAID5 array at most one slot can be
1271 "\fBmissing\fP"; for a RAID6 array at most two slots.
1272 For a RAID1 array, only one real device needs to be given. All of the
1276 When creating a RAID5 array,
1278 will automatically create a degraded array with an extra spare drive.
1279 This is because building the spare into a degraded array is in general faster than resyncing
1280 the parity on a non-degraded, but not clean, array. This feature can
1281 be over-ridden with the
1285 When creating an array with version-1 metadata a name for the host is
1287 If this is not given with the
1291 will chose a name based on the last component of the name of the
1292 device being created. So if
1294 is being created, then the name
1299 is being created, then the name
1303 A new array will normally get a randomly assigned 128bit UUID which is
1304 very likely to be unique. If you have a specific need, you can choose
1305 a UUID for the array by giving the
1307 option. Be warned that creating two arrays with the same UUID is a
1308 recipe for disaster. Also, using
1310 when creating a v0.90 array will silently override any
1315 '''option is given, it is not necessary to list any component-devices in this command.
1316 '''They can be added later, before a
1320 '''is given, the apparent size of the smallest drive given is used.
1322 The General Management options that are valid with --create are:
1325 insist on running the array even if some devices look like they might
1330 start the array readonly - not supported yet.
1338 .I options... devices...
1341 This usage will allow individual devices in an array to be failed,
1342 removed or added. It is possible to perform multiple operations with
1343 on command. For example:
1345 .B " mdadm /dev/md0 -f /dev/hda1 -r /dev/hda1 -a /dev/hda1"
1351 and will then remove it from the array and finally add it back
1352 in as a spare. However only one md array can be affected by a single
1363 MISC mode includes a number of distinct operations that
1364 operate on distinct devices. The operations are:
1367 The device is examined to see if it is
1368 (1) an active md array, or
1369 (2) a component of an md array.
1370 The information discovered is reported.
1374 The device should be an active md device.
1376 will display a detailed description of the array.
1380 will cause the output to be less detailed and the format to be
1381 suitable for inclusion in
1382 .BR /etc/mdadm.conf .
1385 will normally be 0 unless
1387 failed to get useful information about the device(s). However if the
1389 option is given, then the exit status will be:
1393 The array is functioning normally.
1396 The array has at least one failed device.
1399 The array has multiple failed devices and hence is unusable (raid4 or
1403 There was an error while trying to get information about the device.
1408 The device should be a component of an md array.
1410 will read the md superblock of the device and display the contents.
1415 then multiple devices that are components of the one array
1416 are grouped together and reported in a single entry suitable
1418 .BR /etc/mdadm.conf .
1422 without listing any devices will cause all devices listed in the
1423 config file to be examined.
1427 The devices should be active md arrays which will be deactivated, as
1428 long as they are not currently in use.
1432 This will fully activate a partially assembled md array.
1436 This will mark an active array as read-only, providing that it is
1437 not currently being used.
1443 array back to being read/write.
1447 For all operations except
1450 will cause the operation to be applied to all arrays listed in
1455 causes all devices listed in the config file to be examined.
1463 .I options... devices...
1468 to periodically poll a number of md arrays and to report on any events
1471 will never exit once it decides that there are arrays to be checked,
1472 so it should normally be run in the background.
1474 As well as reporting events,
1476 may move a spare drive from one array to another if they are in the
1479 and if the destination array has a failed drive but no spares.
1481 If any devices are listed on the command line,
1483 will only monitor those devices. Otherwise all arrays listed in the
1484 configuration file will be monitored. Further, if
1486 is given, then any other md devices that appear in
1488 will also be monitored.
1490 The result of monitoring the arrays is the generation of events.
1491 These events are passed to a separate program (if specified) and may
1492 be mailed to a given E-mail address.
1494 When passing event to program, the program is run once for each event
1495 and is given 2 or 3 command-line arguments. The first is the
1496 name of the event (see below). The second is the name of the
1497 md device which is affected, and the third is the name of a related
1498 device if relevant, such as a component device that has failed.
1502 is given, then a program or an E-mail address must be specified on the
1503 command line or in the config file. If neither are available, then
1505 will not monitor anything.
1509 will continue monitoring as long as something was found to monitor. If
1510 no program or email is given, then each event is reported to
1513 The different events are:
1517 .B DeviceDisappeared
1518 An md array which previously was configured appears to no longer be
1519 configured. (syslog priority: Critical)
1523 was told to monitor an array which is RAID0 or Linear, then it will
1525 .B DeviceDisappeared
1526 with the extra information
1528 This is because RAID0 and Linear do not support the device-failed,
1529 hot-spare and resync operations which are monitored.
1533 An md array started reconstruction. (syslog priority: Warning)
1539 is 20, 40, 60, or 80, this indicates that rebuild has passed that many
1540 percentage of the total. (syslog priority: Warning)
1544 An md array that was rebuilding, isn't any more, either because it
1545 finished normally or was aborted. (syslog priority: Warning)
1549 An active component device of an array has been marked as
1550 faulty. (syslog priority: Critical)
1554 A spare component device which was being rebuilt to replace a faulty
1555 device has failed. (syslog priority: Critial)
1559 A spare component device which was being rebuilt to replace a faulty
1560 device has been successfully rebuilt and has been made active.
1561 (syslog priority: Info)
1565 A new md array has been detected in the
1567 file. (syslog priority: Info)
1571 A newly noticed array appears to be degraded. This message is not
1574 notices a drive failure which causes degradation, but only when
1576 notices that an array is degraded when it first sees the array.
1577 (syslog priority: Critial)
1581 A spare drive has been moved from one array in a
1583 to another to allow a failed drive to be replaced.
1584 (syslog priority: Info)
1590 has been told, via the config file, that an array should have a certain
1591 number of spare devices, and
1593 detects that it has fewer that this number when it first sees the
1594 array, it will report a
1597 (syslog priority: Warning)
1601 An array was found at startup, and the
1604 (syslog priority: Info)
1614 cause Email to be sent. All events cause the program to be run.
1615 The program is run with two or three arguments, they being the event
1616 name, the array device and possibly a second device.
1618 Each event has an associated array device (e.g.
1620 and possibly a second device. For
1625 the second device is the relevant component device.
1628 the second device is the array that the spare was moved from.
1632 to move spares from one array to another, the different arrays need to
1633 be labelled with the same
1635 in the configuration file. The
1637 name can be any string. It is only necessary that different spare
1638 groups use different names.
1642 detects that an array which is in a spare group has fewer active
1643 devices than necessary for the complete array, and has no spare
1644 devices, it will look for another array in the same spare group that
1645 has a full complement of working drive and a spare. It will then
1646 attempt to remove the spare from the second drive and add it to the
1648 If the removal succeeds but the adding fails, then it is added back to
1652 The GROW mode is used for changing the size or shape of an active
1654 For this to work, the kernel must support the necessary change.
1655 Various types of growth are being added during 2.6 development,
1656 including restructuring a raid5 array to have more active devices.
1658 Currently the only support available is to
1660 change the "size" attribute
1661 for RAID1, RAID5 and RAID6.
1663 increase the "raid-disks" attribute of RAID1 and RAID5.
1665 add a write-intent bitmap to any array which support these bitmaps, or
1666 remove a write-intent bitmap from such an array.
1670 Normally when an array is built the "size" it taken from the smallest
1671 of the drives. If all the small drives in an arrays are, one at a
1672 time, removed and replaced with larger drives, then you could have an
1673 array of large drives with only a small amount used. In this
1674 situation, changing the "size" with "GROW" mode will allow the extra
1675 space to start being used. If the size is increased in this way, a
1676 "resync" process will start to make sure the new parts of the array
1679 Note that when an array changes size, any filesystem that may be
1680 stored in the array will not automatically grow to use the space. The
1681 filesystem will need to be explicitly told to use the extra space.
1683 .SS RAID-DEVICES CHANGES
1685 A RAID1 array can work with any number of devices from 1 upwards
1686 (though 1 is not very useful). There may be times which you want to
1687 increase or decrease the number of active devices. Note that this is
1688 different to hot-add or hot-remove which changes the number of
1691 When reducing the number of devices in a RAID1 array, the slots which
1692 are to be removed from the array must already be vacant. That is, the
1693 devices that which were in those slots must be failed and removed.
1695 When the number of devices is increased, any hot spares that are
1696 present will be activated immediately.
1698 Increasing the number of active devices in a RAID5 is much more
1699 effort. Every block in the array will need to be read and written
1700 back to a new location. From 2.6.17, the Linux Kernel is able to do
1701 this safely, including restart and interrupted "reshape".
1703 When relocating the first few stripes on a raid5, it is not possible
1704 to keep the data on disk completely consistent and crash-proof. To
1705 provide the required safety, mdadm disables writes to the array while
1706 this "critical section" is reshaped, and takes a backup of the data
1707 that is in that section. This backup is normally stored in any spare
1708 devices that the array has, however it can also be stored in a
1709 separate file specified with the
1711 option. If this option is used, and the system does crash during the
1712 critical period, the same file must be passed to
1714 to restore the backup and reassemble the array.
1718 A write-intent bitmap can be added to, or removed from, an active
1719 array. Either internal bitmaps, or bitmaps stored in a separate file
1720 can be added. Note that if you add a bitmap stored in a file which is
1721 in a filesystem that is on the raid array being affected, the system
1722 will deadlock. The bitmap must be on a separate filesystem.
1724 .SH INCREMENTAL MODE
1728 .B mdadm --incremental
1734 .B mdadm --incremental --rebuild
1737 .B mdadm --incremental --run --scan
1741 This mode is designed to be used in conjunction with a device
1742 discovery system. As devices are found in a system, they can be
1744 .B "mdadm --incremental"
1745 to be conditionally added to an appropriate array.
1748 performs a number of tests to determine if the device is part of an
1749 array, and which array is should be part of. If an appropriate array
1750 is found, or can be created,
1752 adds the device to the array and conditionally starts the array.
1756 will only add devices to an array which were previously working
1757 (active or spare) parts of that array. It does not currently support
1758 automatic inclusion of a new drive as a spare in some array.
1760 .B "mdadm --incremental"
1761 requires a bug present in all kernels through 2.6.19, to be fixed.
1762 Hopefully this will be fixed in 2.6.20. Alternately apply the patch
1763 which is included with the mdadm source distribution. If
1765 detects that this bug is present, it will abort any attempt to use
1770 makes are as follow:
1772 Is the device permitted by
1774 That is, is it listed in a
1776 line in that file. If
1778 is absent then the default it to allow any device. Similar if
1780 contains the special word
1782 then any device is allowed. Otherwise the device name given to
1784 must match one of the names or patterns in a
1789 Does the device have a valid md superblock. If a specific metadata
1790 version is request with
1794 then only that style of metadata is accepted, otherwise
1796 finds any known version of metadata. If no
1798 metadata is found, the device is rejected.
1801 Does the metadata match an expected array?
1802 The metadata can match in two ways. Either there is an array listed
1805 which identifies the array (either by UUID, by name, by device list,
1806 or by minor-number), the array was created with a
1810 matches that which is given in
1812 or on the command line.
1815 is not able to positively identify the array as belonging to the
1816 current host, the device will be rejected.
1820 keeps a list of arrays that is has partly assembled in
1821 .B /var/run/mdadm/map
1823 .B /var/run/mdadm.map
1824 if the directory doesn't exist). If no array exists which matches
1825 the metadata on the new device,
1827 must choose a device name and unit number. It does this based on any
1830 or any name information stored in the metadata. If this name
1831 suggests a unit number, that number will be used, otherwise a free
1832 unit number will be chosen. Normally
1834 will prefer to create a partitionable array, however if the
1838 suggests that a non-partitionable array is preferred, that will be
1842 Once an appropriate array is found or created and the device is added,
1844 must decide if the array is ready to be started. It will
1845 normally compare the number of available (non-spare) devices to the
1846 number of devices that the metadata suggests need to be active. If
1847 there are at least that many, the array will be started. This means
1848 that if any devices are missing the array will not be restarted.
1854 in which case the array will be run as soon as there are enough
1855 devices present for the data to be accessible. For a raid1, that
1856 means one device will start the array. For a clean raid5, the array
1857 will be started as soon as all but one drive is present.
1859 Note that neither of these approaches is really ideal. If it is can
1860 be known that all device discovery has completed, then
1864 can be run which will try to start all arrays that are being
1865 incrementally assembled. They are started in "read-auto" mode in
1866 which they are read-only until the first write request. This means
1867 that no metadata updates are made and no attempt at resync or recovery
1868 happens. Further devices that are found before the first write can
1869 still be added safely.
1873 .B " mdadm --query /dev/name-of-device"
1875 This will find out if a given device is a raid array, or is part of
1876 one, and will provide brief information about the device.
1878 .B " mdadm --assemble --scan"
1880 This will assemble and start all arrays listed in the standard config file
1881 file. This command will typically go in a system startup file.
1883 .B " mdadm --stop --scan"
1885 This will shut down all array that can be shut down (i.e. are not
1886 currently in use). This will typically go in a system shutdown script.
1888 .B " mdadm --follow --scan --delay=120"
1890 If (and only if) there is an Email address or program given in the
1891 standard config file, then
1892 monitor the status of all arrays listed in that file by
1893 polling them ever 2 minutes.
1895 .B " mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/hd[ac]1"
1897 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
1900 .B " echo 'DEVICE /dev/hd*[0-9] /dev/sd*[0-9]' > mdadm.conf"
1902 .B " mdadm --detail --scan >> mdadm.conf"
1904 This will create a prototype config file that describes currently
1905 active arrays that are known to be made from partitions of IDE or SCSI drives.
1906 This file should be reviewed before being used as it may
1907 contain unwanted detail.
1909 .B " echo 'DEVICE /dev/hd[a-z] /dev/sd*[a-z]' > mdadm.conf"
1911 .B " mdadm --examine --scan --config=mdadm.conf >> mdadm.conf"
1913 This will find what arrays could be assembled from existing IDE and
1914 SCSI whole drives (not partitions) and store the information is the
1915 format of a config file.
1916 This file is very likely to contain unwanted detail, particularly
1919 entries. It should be reviewed and edited before being used as an
1922 .B " mdadm --examine --brief --scan --config=partitions"
1924 .B " mdadm -Ebsc partitions"
1926 Create a list of devices by reading
1927 .BR /proc/partitions ,
1928 scan these for RAID superblocks, and printout a brief listing of all
1931 .B " mdadm -Ac partitions -m 0 /dev/md0"
1933 Scan all partitions and devices listed in
1934 .BR /proc/partitions
1937 out of all such devices with a RAID superblock with a minor number of 0.
1939 .B " mdadm --monitor --scan --daemonise > /var/run/mdadm"
1941 If config file contains a mail address or alert program, run mdadm in
1942 the background in monitor mode monitoring all md devices. Also write
1943 pid of mdadm daemon to
1944 .BR /var/run/mdadm .
1946 .B " mdadm -Iq /dev/somedevice"
1948 Try to incorporate newly discovered device into some array as
1951 .B " mdadm --incremental --rebuild --run --scan"
1953 Rebuild the array map from any current arrays, and then start any that
1956 .B " mdadm --create --help"
1958 Provide help about the Create mode.
1960 .B " mdadm --config --help"
1962 Provide help about the format of the config file.
1966 Provide general help.
1977 lists all active md devices with information about them.
1979 uses this to find arrays when
1981 is given in Misc mode, and to monitor array reconstruction
1987 The config file lists which devices may be scanned to see if
1988 they contain MD super block, and gives identifying information
1989 (e.g. UUID) about known MD arrays. See
1993 .SS /var/run/mdadm/map
1996 mode is used. this file gets a list of arrays currently being created.
1999 does not exist as a directory, then
2000 .B /var/run/mdadm.map
2005 While entries in the /dev directory can have any format you like,
2007 has an understanding of 'standard' formats which it uses to guide its
2008 behaviour when creating device files via the
2012 The standard names for non-partitioned arrays (the only sort of md
2013 array available in 2.4 and earlier) either of
2019 where NN is a number.
2020 The standard names for partitionable arrays (as available from 2.6
2027 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
2031 was previously known as
2035 is completely separate from the
2037 package, and does not use the
2039 configuration file at all.
2042 For information on the various levels of
2046 .UR http://ostenfeld.dk/~jakob/Software-RAID.HOWTO/
2047 http://ostenfeld.dk/~jakob/Software-RAID.HOWTO/
2050 '''for new releases of the RAID driver check out:
2053 '''.UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2054 '''ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
2059 '''.UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2060 '''http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
2063 The latest version of
2065 should always be available from
2067 .UR http://www.kernel.org/pub/linux/utils/raid/mdadm/
2068 http://www.kernel.org/pub/linux/utils/raid/mdadm/
projects / thirdparty / mdadm.git / blob