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 7 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.
136 This is for doing things to specific components of an array such as
137 adding new spares and removing faulty devices.
141 This is an 'everything else' mode that supports operations on active
142 arrays, operations on component devices such as erasing old superblocks, and
143 information gathering operations.
144 '''This mode allows operations on independent devices such as examine MD
145 '''superblocks, erasing old superblocks and stopping active arrays.
149 .SH Options for selecting a mode are:
152 .BR -A ", " --assemble
153 Assemble a pre-existing array.
157 Build a legacy array without superblocks.
164 .BR -F ", " --follow ", " --monitor
171 Change the size or shape of an active array.
173 If a device is given before any options, or if the first option is
178 then the MANAGE mode is assume.
179 Anything other than these will cause the
183 .SH Options that are not mode-specific are:
187 Display general help message or, after one of the above options, a
188 mode specific help message.
192 Display more detailed help about command line parsing and some commonly
196 .BR -V ", " --version
197 Print version information for mdadm.
200 .BR -v ", " --verbose
201 Be more verbose about what is happening. This can be used twice to be
203 The extra verbosity currently only affects
206 .BR "--examine --scan" .
210 Avoid printing purely informative messages. With this,
212 will be silent unless there is something really important to report.
216 Be less verbose. This is used with
224 gives an intermediate level of verbosity.
228 Be more forceful about certain operations. See the various modes of
229 the exact meaning of this option in different contexts.
232 .BR -c ", " --config=
233 Specify the config file. Default is to use
234 .BR /etc/mdadm.conf ,
235 or if that is missing, then
236 .BR /etc/mdadm/mdadm.conf .
237 If the config file given is
239 then nothing will be read, but
241 will act as though the config file contained exactly
242 .B "DEVICE partitions"
245 to find a list of devices to scan.
248 is given for the config file, then
250 will act as though the config file were empty.
256 for missing information.
257 In general, this option gives
259 permission to get any missing information, like component devices,
260 array devices, array identities, and alert destination from the
262 .BR /etc/mdadm.conf .
263 One exception is MISC mode when using
269 says to get a list of array devices from
273 .B -e ", " --metadata=
274 Declare the style of superblock (raid metadata) to be used. The
275 default is 0.90 for --create, and to guess for other operations.
279 .IP "0, 0.90, default"
280 Use the original 0.90 format superblock. This format limits arrays to
281 28 componenet devices and limits component devices of levels 1 and
282 greater to 2 terabytes.
283 .IP "1, 1.0, 1.1, 1.2"
284 Use the new version-1 format superblock. This has few restrictions.
285 The different subversion store the superblock at different locations
286 on the device, either at the end (for 1.0), at the start (for 1.1) or
287 4K from the start (for 1.2).
292 This will over-ride any
294 setting in the config file and provides the identify of the host which
295 should be considered the home for any arrays.
297 When creating an array, the
299 will be recorded in the superblock. For version-1 superblocks, it will
300 be prefixed to the array name. For version-0.90 superblocks part of
301 the SHA1 hash of the hostname will be stored in the later half of the
304 When reporting information about an array, any array which is tagged
305 for the given homehost will be reported as such.
307 When using Auto-Assemble, only arrays tagged for the given homehost
310 .SH For create, build, or grow:
313 .BR -n ", " --raid-devices=
314 Specify the number of active devices in the array. This, plus the
315 number of spare devices (see below) must equal the number of
317 (including "\fBmissing\fP" devices)
318 that are listed on the command line for
320 Setting a value of 1 is probably
321 a mistake and so requires that
323 be specified first. A value of 1 will then be allowed for linear,
324 multipath, raid0 and raid1. It is never allowed for raid4 or raid5.
326 This number can only be changed using
328 for RAID1 arrays, and only on kernels which provide necessary support.
331 .BR -x ", " --spare-devices=
332 Specify the number of spare (eXtra) devices in the initial array.
333 Spares can also be added
334 and removed later. The number of component devices listed
335 on the command line must equal the number of raid devices plus the
336 number of spare devices.
341 Amount (in Kibibytes) of space to use from each drive in RAID1/4/5/6.
342 This must be a multiple of the chunk size, and must leave about 128Kb
343 of space at the end of the drive for the RAID superblock.
344 If this is not specified
345 (as it normally is not) the smallest drive (or partition) sets the
346 size, though if there is a variance among the drives of greater than 1%, a warning is
349 This value can be set with
351 for RAID level 1/4/5/6. If the array was created with a size smaller
352 than the currently active drives, the extra space can be accessed
355 The size can be given as
357 which means to choose the largest size that fits on all current drives.
361 Specify chunk size of kibibytes. The default is 64.
365 Specify rounding factor for linear array (==chunk size)
369 Set raid level. When used with
371 options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4,
372 raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty. Obviously some of these are synonymous.
376 only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid.
378 Not yet supported with
382 .BR -p ", " --layout=
383 This option configures the fine details of data layout for raid5,
384 and raid10 arrays, and controls the failure modes for
387 The layout of the raid5 parity block can be one of
392 la, ra, ls, rs. The default is left-symmetric.
394 When setting the failure mode for
412 Each mode can be followed by a number which is used as a period
413 between fault generation. Without a number, the fault is generated
414 once on the first relevant request. With a number, the fault will be
415 generated after that many request, and will continue to be generated
416 every time the period elapses.
418 Multiple failure modes can be current simultaneously by using the
419 "--grow" option to set subsequent failure modes.
421 "clear" or "none" will remove any pending or periodic failure modes,
422 and "flush" will clear any persistent faults.
424 To set the parity with "--grow", the level of the array ("faulty")
425 must be specified before the fault mode is specified.
427 Finally, the layout options for RAID10 are one of 'n', 'o' or 'p' followed
428 by a small number. The default is 'n2'.
431 signals 'near' copies. Multiple copies of one data block are at
432 similar offsets in different devices.
435 signals 'offset' copies. Rather than the chunks being duplicated
436 within a stripe, whole stripes are duplicated but are rotated by one
437 device so duplicate blocks are on different devices. Thus subsequent
438 copies of a block are in the next drive, and are one chunk further
443 (multiple copies have very different offsets). See md(4) for more
444 detail about 'near' and 'far'.
446 The number is the number of copies of each datablock. 2 is normal, 3
447 can be useful. This number can be at most equal to the number of
448 devices in the array. It does not need to divide evenly into that
449 number (e.g. it is perfectly legal to have an 'n2' layout for an array
450 with an odd number of devices).
454 same as --layout (thus explaining the p of
458 .BR -b ", " --bitmap=
459 Specify a file to store a write-intent bitmap in. The file should not
460 exist unless --force is also given. The same file should be provided
461 when assembling the array. If the word
463 is given, then the bitmap is stored with the metadata on the array,
464 and so is replicated on all devices. If the word
468 mode, then any bitmap that is present is removed.
470 To help catch typing errors, the filename must contain at least one
471 slash ('/') if it is a real file (not 'internal' or 'none').
473 Note: external bitmaps are only known to work on ext2 and ext3.
474 Storing bitmap files on other filesystems may result in serious problems.
478 Set the chunksize of the bitmap. Each bit corresponds to that many
479 Kilobytes of storage.
480 When using a file based bitmap, the default is to use the smallest
481 size that is atleast 4 and requires no more than 2^21 chunks.
484 bitmap, the chunksize is automatically determined to make best use of
489 .BR -W ", " --write-mostly
490 subsequent devices lists in a
495 command will be flagged as 'write-mostly'. This is valid for RAID1
496 only and means that the 'md' driver will avoid reading from these
497 devices if at all possible. This can be useful if mirroring over a
502 Specify that write-behind mode should be enabled (valid for RAID1
503 only). If an argument is specified, it will set the maximum number
504 of outstanding writes allowed. The default value is 256.
505 A write-intent bitmap is required in order to use write-behind
506 mode, and write-behind is only attempted on drives marked as
513 that the array pre-existed and is known to be clean. It can be useful
514 when trying to recover from a major failure as you can be sure that no
515 data will be affected unless you actually write to the array. It can
516 also be used when creating a RAID1 or RAID10 if you want to avoid the
517 initial resync, however this practice - while normally safe - is not
518 recommended. Use this ony if you really know what you are doing.
522 This is needed when --grow is used to increase the number of
523 raid-devices in a RAID5 if there are no spare devices available.
524 See the section below on RAID_DEVICE CHANGES. The file should be
525 stored on a separate device, not on the raid array being reshaped.
531 for the array. This is currently only effective when creating an
532 array with a version-1 superblock. The name is a simple textual
533 string that can be used to identify array components when assembling.
539 run the array, even if some of the components
540 appear to be active in another array or filesystem. Normally
542 will ask for confirmation before including such components in an
543 array. This option causes that question to be suppressed.
549 accept the geometry and layout specified without question. Normally
551 will not allow creation of an array with only one device, and will try
552 to create a raid5 array with one missing drive (as this makes the
553 initial resync work faster). With
556 will not try to be so clever.
559 .BR -a ", " "--auto{=no,yes,md,mdp,part,p}{NN}"
560 Instruct mdadm to create the device file if needed, possibly allocating
561 an unused minor number. "md" causes a non-partitionable array
562 to be used. "mdp", "part" or "p" causes a partitionable array (2.6 and
563 later) to be used. "yes" requires the named md device to have
564 a 'standard' format, and the type and minor number will be determined
565 from this. See DEVICE NAMES below.
567 The argument can also come immediately after
572 is also given, then any
574 entries in the config file will over-ride the
576 instruction given on the command line.
578 For partitionable arrays,
580 will create the device file for the whole array and for the first 4
581 partitions. A different number of partitions can be specified at the
582 end of this option (e.g.
584 If the device name ends with a digit, the partition names add a 'p',
585 and a number, e.g. "/dev/home1p3". If there is no
586 trailing digit, then the partition names just have a number added,
587 e.g. "/dev/scratch3".
589 If the md device name is in a 'standard' format as described in DEVICE
590 NAMES, then it will be created, if necessary, with the appropriate
591 number based on that name. If the device name is not in one of these
592 formats, then a unused minor number will be allocated. The minor
593 number will be considered unused if there is no active array for that
594 number, and there is no entry in /dev for that number and with a
605 it will also create symlinks from
607 with names starting with
615 to enforce this even if it is suppressing
623 uuid of array to assemble. Devices which don't have this uuid are
627 .BR -m ", " --super-minor=
628 Minor number of device that array was created for. Devices which
629 don't have this minor number are excluded. If you create an array as
630 /dev/md1, then all superblocks will contain the minor number 1, even if
631 the array is later assembled as /dev/md2.
633 Giving the literal word "dev" for
637 to use the minor number of the md device that is being assembled.
641 will look for super blocks with a minor number of 0.
645 Specify the name of the array to assemble. This must be the name
646 that was specified when creating the array. It must either match
647 then name stored in the superblock exactly, or it must match
650 is added to the start of the given name.
654 Assemble the array even if some superblocks appear out-of-date
658 Attempt to start the array even if fewer drives were given than were
659 present last time the array was active. Normally if not all the
660 expected drives are found and
662 is not used, then the array will be assembled but not started.
665 an attempt will be made to start it anyway.
669 This is the reverse of
671 in that it inhibits the started if array unless all expected drives
672 are present. This is only needed with
674 and can be used if you physical connections to devices are
675 not as reliable as you would like.
678 .BR -a ", " "--auto{=no,yes,md,mdp,part}"
679 See this option under Create and Build options.
682 .BR -b ", " --bitmap=
683 Specify the bitmap file that was given when the array was created. If
686 bitmap, there is no need to specify this when assembling the array.
692 was used to grow the number of raid-devices in a RAID5, and the system
693 crashed during the critical section, then the same
695 must be presented to --assemble to allow possibly corrupted data to be
699 .BR -U ", " --update=
700 Update the superblock on each device while assembling the array. The
701 argument given to this flag can be one of
714 option will adjust the superblock of an array what was created on a Sparc
715 machine running a patched 2.2 Linux kernel. This kernel got the
716 alignment of part of the superblock wrong. You can use the
717 .B "--examine --sparc2.2"
720 to see what effect this would have.
724 option will update the
726 field on each superblock to match the minor number of the array being
728 This can be useful if
730 reports a different "Preferred Minor" to
732 In some cases this update will be performed automatically
733 by the kernel driver. In particular the update happens automatically
734 at the first write to an array with redundancy (RAID level 1 or
735 greater) on a 2.6 (or later) kernel.
739 option will change the uuid of the array. If a UUID is given with the
740 "--uuid" option that UUID will be used as a new UUID and will
742 be used to help identify the devices in the array.
743 If no "--uuid" is given, a random uuid is chosen.
747 option will change the
749 of the array as stored in the superblock. This is only supported for
750 version-1 superblocks.
754 option will change the
756 as recorded in the superblock. For version-0 superblocks, this is the
757 same as updating the UUID.
758 For version-1 superblocks, this involves updating the name.
762 option will cause the array to be marked
764 meaning that any redundancy in the array (e.g. parity for raid5,
765 copies for raid1) may be incorrect. This will cause the raid system
766 to perform a "resync" pass to make sure that all redundant information
771 option allows arrays to be moved between machines with different
773 When assembling such an array for the first time after a move, giving
774 .B "--update=byteorder"
777 to expect superblocks to have their byteorder reversed, and will
778 correct that order before assembling the array. This is only valid
779 with original (Version 0.90) superblocks.
783 option will correct the summaries in the superblock. That is the
784 counts of total, working, active, failed, and spare devices.
787 .B --auto-update-homehost
788 This flag is only meaning with auto-assembly (see discussion below).
789 In that situation, if no suitable arrays are found for this homehost,
791 will recan for any arrays at all and will assemble them and update the
792 homehost to match the current host.
798 hot-add listed devices.
802 re-add a device that was recently removed from an array.
806 remove listed devices. They must not be active. i.e. they should
807 be failed or spare devices.
811 mark listed devices as faulty.
818 Each of these options require that the first device list is the array
819 to be acted upon and the remainder are component devices to be added,
820 removed, or marked as fault. Several different operations can be
821 specified for different devices, e.g.
823 mdadm /dev/md0 --add /dev/sda1 --fail /dev/sdb1 --remove /dev/sdb1
825 Each operation applies to all devices listed until the next
828 If an array is using a write-intent bitmap, then devices which have
829 been removed can be re-added in a way that avoids a full
830 reconstruction but instead just updated the blocks that have changed
831 since the device was removed. For arrays with persistent metadata
832 (superblocks) this is done automatically. For arrays created with
834 mdadm needs to be told that this device we removed recently with
837 Devices can only be removed from an array if they are not in active
838 use. i.e. that must be spares or failed devices. To remove an active
839 device, it must be marked as
847 Examine a device to see
848 (1) if it is an md device and (2) if it is a component of an md
850 Information about what is discovered is presented.
854 Print detail of one or more md devices.
857 .BR -E ", " --examine
858 Print content of md superblock on device(s).
861 If an array was created on a 2.2 Linux kernel patched with RAID
862 support, the superblock will have been created incorrectly, or at
863 least incompatibly with 2.4 and later kernels. Using the
867 will fix the superblock before displaying it. If this appears to do
868 the right thing, then the array can be successfully assembled using
869 .BR "--assemble --update=sparc2.2" .
872 .BR -X ", " --examine-bitmap
873 Report information about a bitmap file.
877 start a partially built array.
881 deactivate array, releasing all resources.
884 .BR -o ", " --readonly
885 mark array as readonly.
888 .BR -w ", " --readwrite
889 mark array as readwrite.
893 If the device contains a valid md superblock, the block is
894 over-written with zeros. With
896 the block where the superblock would be is over-written even if it
897 doesn't appear to be valid.
905 is set to reflect the status of the device.
907 .SH For Monitor mode:
910 Give a mail address to send alerts to.
913 .BR -p ", " --program ", " --alert
914 Give a program to be run whenever an event is detected.
918 Cause all events to be reported through 'syslog'. The messages have
919 facility of 'daemon' and varying priorities.
923 Give a delay in seconds.
925 polls the md arrays and then waits this many seconds before polling
926 again. The default is 60 seconds.
929 .BR -f ", " --daemonise
932 to run as a background daemon if it decides to monitor anything. This
933 causes it to fork and run in the child, and to disconnect form the
934 terminal. The process id of the child is written to stdout.
937 which will only continue monitoring if a mail address or alert program
938 is found in the config file.
941 .BR -i ", " --pid-file
944 is running in daemon mode, write the pid of the daemon process to
945 the specified file, instead of printing it on standard output.
948 .BR -1 ", " --oneshot
949 Check arrays only once. This will generate
951 events and more significantly
957 .B " mdadm --monitor --scan -1"
959 from a cron script will ensure regular notification of any degraded arrays.
965 alert for every array found at startup. This alert gets mailed and
966 passed to the alert program. This can be used for testing that alert
967 message do get through successfully.
974 .I md-device options-and-component-devices...
977 .B mdadm --assemble --scan
978 .I md-devices-and-options...
981 .B mdadm --assemble --scan
985 This usage assembles one or more raid arrays from pre-existing components.
986 For each array, mdadm needs to know the md device, the identity of the
987 array, and a number of component-devices. These can be found in a number of ways.
989 In the first usage example (without the
991 the first device given is the md device.
992 In the second usage example, all devices listed are treated as md
993 devices and assembly is attempted.
994 In the third (where no devices are listed) all md devices that are
995 listed in the configuration file are assembled.
997 If precisely one device is listed, but
1003 was given and identify information is extracted from the configuration file.
1005 The identity can be given with the
1009 option, can be found in the config file, or will be taken from the
1010 super block on the first component-device listed on the command line.
1012 Devices can be given on the
1014 command line or in the config file. Only devices which have an md
1015 superblock which contains the right identity will be considered for
1018 The config file is only used if explicitly named with
1020 or requested with (a possibly implicit)
1028 is not given, then the config file will only be used to find the
1029 identity of md arrays.
1031 Normally the array will be started after it is assembled. However if
1033 is not given and insufficient drives were listed to start a complete
1034 (non-degraded) array, then the array is not started (to guard against
1035 usage errors). To insist that the array be started in this case (as
1036 may work for RAID1, 4, 5, 6, or 10), give the
1042 option is given, either on the command line (--auto) or in the
1043 configuration file (e.g. auto=part), then
1045 will create the md device if necessary or will re-create it if it
1046 doesn't look usable as it is.
1048 This can be useful for handling partitioned devices (which don't have
1049 a stable device number - it can change after a reboot) and when using
1050 "udev" to manage your
1052 tree (udev cannot handle md devices because of the unusual device
1053 initialisation conventions).
1055 If the option to "auto" is "mdp" or "part" or (on the command line
1056 only) "p", then mdadm will create a partitionable array, using the
1057 first free one that is not in use, and does not already have an entry
1058 in /dev (apart from numeric /dev/md* entries).
1060 If the option to "auto" is "yes" or "md" or (on the command line)
1061 nothing, then mdadm will create a traditional, non-partitionable md
1064 It is expected that the "auto" functionality will be used to create
1065 device entries with meaningful names such as "/dev/md/home" or
1066 "/dev/md/root", rather than names based on the numerical array number.
1068 When using this option to create a partitionable array, the device
1069 files for the first 4 partitions are also created. If a different
1070 number is required it can be simply appended to the auto option.
1071 e.g. "auto=part8". Partition names are created by appending a digit
1072 string to the device name, with an intervening "p" if the device name
1077 option is also available in Build and Create modes. As those modes do
1078 not use a config file, the "auto=" config option does not apply to
1086 and no devices are listed,
1088 will first attempt to assemble all the arrays listed in the config
1093 has been specified (either in the config file or on the command line),
1095 will look further for possible arrays and will try to assemble
1096 anything that it finds which is tagged as belonging to the given
1097 homehost. This is the only situation where
1099 will assemble arrays without being given specific device name or
1100 identify information for the array.
1104 finds a consistent set of devices that look like they should comprise
1105 an array, and if the superblock is tagged as belonging to the given
1106 home host, it will automatically choose a device name and try to
1107 assemble the array. If the array uses version-0.90 metadata, then the
1109 number as recorded in the superblock is used to create a name in
1113 If the array uses version-1 metadata, then the
1115 from the superblock is used to similarly create a name in
1117 The name will have any 'host' prefix stripped first.
1121 cannot find any array for the given host at all, and if
1122 .B --auto-update-homehost
1125 will search again for any array (not just an array created for this
1126 host) and will assemble each assuming
1127 .IR --update=homehost .
1128 This will change the host tag in the superblock so that on the next run,
1129 these arrays will be found without the second pass. The intention of
1130 this feature is to support transitioning a set of md arrays to using
1133 The reason for requiring arrays to be tagged with the homehost for
1134 auto assembly is to guard against problems that can arise when moving
1135 devices from one host to another.
1145 .BI --raid-devices= Z
1149 This usage is similar to
1151 The difference is that it creates an array without a superblock. With
1152 these arrays there is no difference between initially creating the array and
1153 subsequently assembling the array, except that hopefully there is useful
1154 data there in the second case.
1156 The level may raid0, linear, multipath, or faulty, or one of their
1157 synonyms. All devices must be listed and the array will be started
1169 .BI --raid-devices= Z
1173 This usage will initialise a new md array, associate some devices with
1174 it, and activate the array.
1178 option is given (as described in more detail in the section on
1179 Assemble mode), then the md device will be created with a suitable
1180 device number if necessary.
1182 As devices are added, they are checked to see if they contain raid
1183 superblocks or filesystems. They are also checked to see if the variance in
1184 device size exceeds 1%.
1186 If any discrepancy is found, the array will not automatically be run, though
1189 can override this caution.
1191 To create a "degraded" array in which some devices are missing, simply
1192 give the word "\fBmissing\fP"
1193 in place of a device name. This will cause
1195 to leave the corresponding slot in the array empty.
1196 For a RAID4 or RAID5 array at most one slot can be
1197 "\fBmissing\fP"; for a RAID6 array at most two slots.
1198 For a RAID1 array, only one real device needs to be given. All of the
1202 When creating a RAID5 array,
1204 will automatically create a degraded array with an extra spare drive.
1205 This is because building the spare into a degraded array is in general faster than resyncing
1206 the parity on a non-degraded, but not clean, array. This feature can
1207 be over-ridden with the
1211 When creating an array with version-1 metadata a name for the host is
1213 If this is not given with the
1217 will chose a name based on the last component of the name of the
1218 device being created. So if
1220 is being created, then the name
1225 is being created, then the name
1231 '''option is given, it is not necessary to list any component-devices in this command.
1232 '''They can be added later, before a
1236 '''is given, the apparent size of the smallest drive given is used.
1238 The General Management options that are valid with --create are:
1241 insist on running the array even if some devices look like they might
1246 start the array readonly - not supported yet.
1254 .I options... devices...
1257 This usage will allow individual devices in an array to be failed,
1258 removed or added. It is possible to perform multiple operations with
1259 on command. For example:
1261 .B " mdadm /dev/md0 -f /dev/hda1 -r /dev/hda1 -a /dev/hda1"
1267 and will then remove it from the array and finally add it back
1268 in as a spare. However only one md array can be affected by a single
1279 MISC mode includes a number of distinct operations that
1280 operate on distinct devices. The operations are:
1283 The device is examined to see if it is
1284 (1) an active md array, or
1285 (2) a component of an md array.
1286 The information discovered is reported.
1290 The device should be an active md device.
1292 will display a detailed description of the array.
1296 will cause the output to be less detailed and the format to be
1297 suitable for inclusion in
1298 .BR /etc/mdadm.conf .
1301 will normally be 0 unless
1303 failed to get useful information about the device(s). However if the
1305 option is given, then the exit status will be:
1309 The array is functioning normally.
1312 The array has at least one failed device.
1315 The array has multiple failed devices and hence is unusable (raid4 or
1319 There was an error while trying to get information about the device.
1324 The device should be a component of an md array.
1326 will read the md superblock of the device and display the contents.
1331 then multiple devices that are components of the one array
1332 are grouped together and reported in a single entry suitable
1334 .BR /etc/mdadm.conf .
1338 without listing any devices will cause all devices listed in the
1339 config file to be examined.
1343 The devices should be active md arrays which will be deactivated, as
1344 long as they are not currently in use.
1348 This will fully activate a partially assembled md array.
1352 This will mark an active array as read-only, providing that it is
1353 not currently being used.
1359 array back to being read/write.
1363 For all operations except
1366 will cause the operation to be applied to all arrays listed in
1371 causes all devices listed in the config file to be examined.
1379 .I options... devices...
1384 to periodically poll a number of md arrays and to report on any events
1387 will never exit once it decides that there are arrays to be checked,
1388 so it should normally be run in the background.
1390 As well as reporting events,
1392 may move a spare drive from one array to another if they are in the
1395 and if the destination array has a failed drive but no spares.
1397 If any devices are listed on the command line,
1399 will only monitor those devices. Otherwise all arrays listed in the
1400 configuration file will be monitored. Further, if
1402 is given, then any other md devices that appear in
1404 will also be monitored.
1406 The result of monitoring the arrays is the generation of events.
1407 These events are passed to a separate program (if specified) and may
1408 be mailed to a given E-mail address.
1410 When passing event to program, the program is run once for each event
1411 and is given 2 or 3 command-line arguments. The first is the
1412 name of the event (see below). The second is the name of the
1413 md device which is affected, and the third is the name of a related
1414 device if relevant, such as a component device that has failed.
1418 is given, then a program or an E-mail address must be specified on the
1419 command line or in the config file. If neither are available, then
1421 will not monitor anything.
1425 will continue monitoring as long as something was found to monitor. If
1426 no program or email is given, then each event is reported to
1429 The different events are:
1433 .B DeviceDisappeared
1434 An md array which previously was configured appears to no longer be
1435 configured. (syslog priority: Critical)
1439 was told to monitor an array which is RAID0 or Linear, then it will
1441 .B DeviceDisappeared
1442 with the extra information
1444 This is because RAID0 and Linear do not support the device-failed,
1445 hot-spare and resync operations which are monitored.
1449 An md array started reconstruction. (syslog priority: Warning)
1455 is 20, 40, 60, or 80, this indicates that rebuild has passed that many
1456 percentage of the total. (syslog priority: Warning)
1460 An md array that was rebuilding, isn't any more, either because it
1461 finished normally or was aborted. (syslog priority: Warning)
1465 An active component device of an array has been marked as
1466 faulty. (syslog priority: Critical)
1470 A spare component device which was being rebuilt to replace a faulty
1471 device has failed. (syslog priority: Critial)
1475 A spare component device which was being rebuilt to replace a faulty
1476 device has been successfully rebuilt and has been made active.
1477 (syslog priority: Info)
1481 A new md array has been detected in the
1483 file. (syslog priority: Info)
1487 A newly noticed array appears to be degraded. This message is not
1490 notices a drive failure which causes degradation, but only when
1492 notices that an array is degraded when it first sees the array.
1493 (syslog priority: Critial)
1497 A spare drive has been moved from one array in a
1499 to another to allow a failed drive to be replaced.
1500 (syslog priority: Info)
1506 has been told, via the config file, that an array should have a certain
1507 number of spare devices, and
1509 detects that it has fewer that this number when it first sees the
1510 array, it will report a
1513 (syslog priority: Warning)
1517 An array was found at startup, and the
1520 (syslog priority: Info)
1530 cause Email to be sent. All events cause the program to be run.
1531 The program is run with two or three arguments, they being the event
1532 name, the array device and possibly a second device.
1534 Each event has an associated array device (e.g.
1536 and possibly a second device. For
1541 the second device is the relevant component device.
1544 the second device is the array that the spare was moved from.
1548 to move spares from one array to another, the different arrays need to
1549 be labelled with the same
1551 in the configuration file. The
1553 name can be any string. It is only necessary that different spare
1554 groups use different names.
1558 detects that an array which is in a spare group has fewer active
1559 devices than necessary for the complete array, and has no spare
1560 devices, it will look for another array in the same spare group that
1561 has a full complement of working drive and a spare. It will then
1562 attempt to remove the spare from the second drive and add it to the
1564 If the removal succeeds but the adding fails, then it is added back to
1568 The GROW mode is used for changing the size or shape of an active
1570 For this to work, the kernel must support the necessary change.
1571 Various types of growth are being added during 2.6 development,
1572 including restructuring a raid5 array to have more active devices.
1574 Currently the only support available is to
1576 change the "size" attribute
1577 for RAID1, RAID5 and RAID6.
1579 increase the "raid-disks" attribute of RAID1 and RAID5.
1581 add a write-intent bitmap to any array which support these bitmaps, or
1582 remove a write-intent bitmap from such an array.
1586 Normally when an array is built the "size" it taken from the smallest
1587 of the drives. If all the small drives in an arrays are, one at a
1588 time, removed and replaced with larger drives, then you could have an
1589 array of large drives with only a small amount used. In this
1590 situation, changing the "size" with "GROW" mode will allow the extra
1591 space to start being used. If the size is increased in this way, a
1592 "resync" process will start to make sure the new parts of the array
1595 Note that when an array changes size, any filesystem that may be
1596 stored in the array will not automatically grow to use the space. The
1597 filesystem will need to be explicitly told to use the extra space.
1599 .SS RAID-DEVICES CHANGES
1601 A RAID1 array can work with any number of devices from 1 upwards
1602 (though 1 is not very useful). There may be times which you want to
1603 increase or decrease the number of active devices. Note that this is
1604 different to hot-add or hot-remove which changes the number of
1607 When reducing the number of devices in a RAID1 array, the slots which
1608 are to be removed from the array must already be vacant. That is, the
1609 devices that which were in those slots must be failed and removed.
1611 When the number of devices is increased, any hot spares that are
1612 present will be activated immediately.
1614 Increasing the number of active devices in a RAID5 is much more
1615 effort. Every block in the array will need to be read and written
1616 back to a new location. From 2.6.17, the Linux Kernel is able to do
1617 this safely, including restart and interrupted "reshape".
1619 When relocating the first few stripes on a raid5, it is not possible
1620 to keep the data on disk completely consistent and crash-proof. To
1621 provide the required safety, mdadm disables writes to the array while
1622 this "critical section" is reshaped, and takes a backup of the data
1623 that is in that section. This backup is normally stored in any spare
1624 devices that the array has, however it can also be stored in a
1625 separate file specified with the
1627 option. If this option is used, and the system does crash during the
1628 critical period, the same file must be passed to
1630 to restore the backup and reassemble the array.
1634 A write-intent bitmap can be added to, or removed from, an active
1635 array. Either internal bitmaps, or bitmaps stored in a separate file
1636 can be added. Note that if you add a bitmap stored in a file which is
1637 in a filesystem that is on the raid array being affected, the system
1638 will deadlock. The bitmap must be on a separate filesystem.
1642 .B " mdadm --query /dev/name-of-device"
1644 This will find out if a given device is a raid array, or is part of
1645 one, and will provide brief information about the device.
1647 .B " mdadm --assemble --scan"
1649 This will assemble and start all arrays listed in the standard config file
1650 file. This command will typically go in a system startup file.
1652 .B " mdadm --stop --scan"
1654 This will shut down all array that can be shut down (i.e. are not
1655 currently in use). This will typically go in a system shutdown script.
1657 .B " mdadm --follow --scan --delay=120"
1659 If (and only if) there is an Email address or program given in the
1660 standard config file, then
1661 monitor the status of all arrays listed in that file by
1662 polling them ever 2 minutes.
1664 .B " mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/hd[ac]1"
1666 Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.
1669 .B " echo 'DEVICE /dev/hd*[0-9] /dev/sd*[0-9]' > mdadm.conf"
1671 .B " mdadm --detail --scan >> mdadm.conf"
1673 This will create a prototype config file that describes currently
1674 active arrays that are known to be made from partitions of IDE or SCSI drives.
1675 This file should be reviewed before being used as it may
1676 contain unwanted detail.
1678 .B " echo 'DEVICE /dev/hd[a-z] /dev/sd*[a-z]' > mdadm.conf"
1680 .B " mdadm --examine --scan --config=mdadm.conf >> mdadm.conf"
1682 This will find what arrays could be assembled from existing IDE and
1683 SCSI whole drives (not partitions) and store the information is the
1684 format of a config file.
1685 This file is very likely to contain unwanted detail, particularly
1688 entries. It should be reviewed and edited before being used as an
1691 .B " mdadm --examine --brief --scan --config=partitions"
1693 .B " mdadm -Ebsc partitions"
1695 Create a list of devices by reading
1696 .BR /proc/partitions ,
1697 scan these for RAID superblocks, and printout a brief listing of all
1700 .B " mdadm -Ac partitions -m 0 /dev/md0"
1702 Scan all partitions and devices listed in
1703 .BR /proc/partitions
1706 out of all such devices with a RAID superblock with a minor number of 0.
1708 .B " mdadm --monitor --scan --daemonise > /var/run/mdadm"
1710 If config file contains a mail address or alert program, run mdadm in
1711 the background in monitor mode monitoring all md devices. Also write
1712 pid of mdadm daemon to
1713 .BR /var/run/mdadm .
1715 .B " mdadm --create --help"
1717 Provide help about the Create mode.
1719 .B " mdadm --config --help"
1721 Provide help about the format of the config file.
1725 Provide general help.
1736 lists all active md devices with information about them.
1738 uses this to find arrays when
1740 is given in Misc mode, and to monitor array reconstruction
1746 The config file lists which devices may be scanned to see if
1747 they contain MD super block, and gives identifying information
1748 (e.g. UUID) about known MD arrays. See
1754 While entries in the /dev directory can have any format you like,
1756 has an understanding of 'standard' formats which it uses to guide its
1757 behaviour when creating device files via the
1761 The standard names for non-partitioned arrays (the only sort of md
1762 array available in 2.4 and earlier) either of
1768 where NN is a number.
1769 The standard names for partitionable arrays (as available from 2.6
1776 Partition numbers should be indicated by added "pMM" to these, thus "/dev/md/d1p2".
1780 was previously known as
1784 is completely separate from the
1786 package, and does not use the
1788 configuration file at all.
1791 For information on the various levels of
1795 .UR http://ostenfeld.dk/~jakob/Software-RAID.HOWTO/
1796 http://ostenfeld.dk/~jakob/Software-RAID.HOWTO/
1799 '''for new releases of the RAID driver check out:
1802 '''.UR ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
1803 '''ftp://ftp.kernel.org/pub/linux/kernel/people/mingo/raid-patches
1808 '''.UR http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
1809 '''http://www.cse.unsw.edu.au/~neilb/patches/linux-stable/
1812 The latest version of
1814 should always be available from
1816 .UR http://www.kernel.org/pub/linux/utils/raid/mdadm/
1817 http://www.kernel.org/pub/linux/utils/raid/mdadm/