{ "mode": "man", "parameter": "mdadm", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/mdadm/8/json", "generated": "2026-10-08T23:49:55Z", "synopsis": "mdadm [mode] [options] ", "sections": { "NAME": { "content": "mdadm - manage MD devices aka Linux Software RAID\n\n", "subsections": [] }, "SYNOPSIS": { "content": "mdadm [mode] [options] \n\n", "subsections": [] }, "DESCRIPTION": { "content": "RAID devices are virtual devices created from two or more real block devices. This allows\nmultiple devices (typically disk drives or partitions thereof) to be combined into a single\ndevice to hold (for example) a single filesystem. Some RAID levels include redundancy and so\ncan survive some degree of device failure.\n\nLinux Software RAID devices are implemented through the md (Multiple Devices) device driver.\n\nCurrently, Linux supports LINEAR md devices, RAID0 (striping), RAID1 (mirroring), RAID4,\nRAID5, RAID6, RAID10, MULTIPATH, FAULTY, and CONTAINER.\n\nMULTIPATH is not a Software RAID mechanism, but does involve multiple devices: each device is\na path to one common physical storage device. New installations should not use md/multipath\nas it is not well supported and has no ongoing development. Use the Device Mapper based mul‐\ntipath-tools instead.\n\nFAULTY is also not true RAID, and it only involves one device. It provides a layer over a\ntrue device that can be used to inject faults.\n\nCONTAINER is different again. A CONTAINER is a collection of devices that are managed as a\nset. This is similar to the set of devices connected to a hardware RAID controller. The set\nof devices may contain a number of different RAID arrays each utilising some (or all) of the\nblocks from a number of the devices in the set. For example, two devices in a 5-device set\nmight form a RAID1 using the whole devices. The remaining three might have a RAID5 over the\nfirst half of each device, and a RAID0 over the second half.\n\nWith a CONTAINER, there is one set of metadata that describes all of the arrays in the con‐\ntainer. So when mdadm creates a CONTAINER device, the device just represents the metadata.\nOther normal arrays (RAID1 etc) can be created inside the container.\n\n", "subsections": [] }, "MODES": { "content": "mdadm has several major modes of operation:\n", "subsections": [ { "name": "Assemble", "content": "Assemble the components of a previously created array into an active array. Compo‐\nnents can be explicitly given or can be searched for. mdadm checks that the compo‐\nnents do form a bona fide array, and can, on request, fiddle superblock information so\nas to assemble a faulty array.\n\n\nBuild Build an array that doesn't have per-device metadata (superblocks). For these sorts\nof arrays, mdadm cannot differentiate between initial creation and subsequent assembly\nof an array. It also cannot perform any checks that appropriate components have been\nrequested. Because of this, the Build mode should only be used together with a com‐\nplete understanding of what you are doing.\n\n\nCreate Create a new array with per-device metadata (superblocks). Appropriate metadata is\nwritten to each device, and then the array comprising those devices is activated. A\n'resync' process is started to make sure that the array is consistent (e.g. both sides\nof a mirror contain the same data) but the content of the device is left otherwise un‐\ntouched. The array can be used as soon as it has been created. There is no need to\nwait for the initial resync to finish.\n\n" }, { "name": "Follow or Monitor", "content": "Monitor one or more md devices and act on any state changes. This is only meaningful\nfor RAID1, 4, 5, 6, 10 or multipath arrays, as only these have interesting state.\nRAID0 or Linear never have missing, spare, or failed drives, so there is nothing to\nmonitor.\n\n\nGrow Grow (or shrink) an array, or otherwise reshape it in some way. Currently supported\ngrowth options including changing the active size of component devices and changing\nthe number of active devices in Linear and RAID levels 0/1/4/5/6, changing the RAID\nlevel between 0, 1, 5, and 6, and between 0 and 10, changing the chunk size and layout\nfor RAID 0,4,5,6,10 as well as adding or removing a write-intent bitmap and changing\nthe array's consistency policy.\n\n" }, { "name": "Incremental Assembly", "content": "Add a single device to an appropriate array. If the addition of the device makes the\narray runnable, the array will be started. This provides a convenient interface to a\nhot-plug system. As each device is detected, mdadm has a chance to include it in some\narray as appropriate. Optionally, when the --fail flag is passed in we will remove\nthe device from any active array instead of adding it.\n\nIf a CONTAINER is passed to mdadm in this mode, then any arrays within that container\nwill be assembled and started.\n\n\nManage This is for doing things to specific components of an array such as adding new spares\nand removing faulty devices.\n\n\nMisc This is an 'everything else' mode that supports operations on active arrays, opera‐\ntions on component devices such as erasing old superblocks, and information-gathering\noperations.\n\n" }, { "name": "Auto-detect", "content": "This mode does not act on a specific device or array, but rather it requests the Linux\nKernel to activate any auto-detected arrays.\n" } ] }, "OPTIONS": { "content": "Options for selecting a mode are:", "subsections": [ { "name": "-A --assemble", "content": "Assemble a pre-existing array.\n\n", "flag": "-A", "long": "--assemble" }, { "name": "-B --build", "content": "Build a legacy array without superblocks.\n\n", "flag": "-B", "long": "--build" }, { "name": "-C --create", "content": "Create a new array.\n\n", "flag": "-C", "long": "--create" }, { "name": "-F --follow --monitor", "content": "Select Monitor mode.\n\n", "flag": "-F", "long": "--monitor" }, { "name": "-G --grow", "content": "Change the size or shape of an active array.\n\n", "flag": "-G", "long": "--grow" }, { "name": "-I --incremental", "content": "Add/remove a single device to/from an appropriate array, and possibly start the array.\n\n", "flag": "-I", "long": "--incremental" }, { "name": "--auto-detect", "content": "Request that the kernel starts any auto-detected arrays. This can only work if md is\ncompiled into the kernel — not if it is a module. Arrays can be auto-detected by the\nkernel if all the components are in primary MS-DOS partitions with partition type FD,\nand all use v0.90 metadata. In-kernel autodetect is not recommended for new installa‐\ntions. Using mdadm to detect and assemble arrays — possibly in an initrd — is sub‐\nstantially more flexible and should be preferred.\n\n\nIf a device is given before any options, or if the first option is one of --add, --re-add,\n--add-spare, --fail, --remove, or --replace, then the MANAGE mode is assumed. Anything other\nthan these will cause the Misc mode to be assumed.\n\n\nOptions that are not mode-specific are:", "long": "--auto-detect" }, { "name": "-h --help", "content": "Display a general help message or, after one of the above options, a mode-specific\nhelp message.\n\n", "flag": "-h", "long": "--help" }, { "name": "--help-options", "content": "Display more detailed help about command-line parsing and some commonly used options.\n\n", "long": "--help-options" }, { "name": "-V --version", "content": "Print version information for mdadm.\n\n", "flag": "-V", "long": "--version" }, { "name": "-v --verbose", "content": "Be more verbose about what is happening. This can be used twice to be extra-verbose.\nThe extra verbosity currently only affects --detail --scan and --examine --scan.\n\n", "flag": "-v", "long": "--verbose" }, { "name": "-q --quiet", "content": "Avoid printing purely informative messages. With this, mdadm will be silent unless\nthere is something really important to report.\n\n\n", "flag": "-q", "long": "--quiet" }, { "name": "-f --force", "content": "Be more forceful about certain operations. See the various modes for the exact mean‐\ning of this option in different contexts.\n\n", "flag": "-f", "long": "--force" }, { "name": "-c --config=", "content": "Specify the config file or directory. If not specified, the default config file and\ndefault conf.d directory will be used. See mdadm.conf(5) for more details.\n\nIf the config file given is partitions then nothing will be read, but mdadm will act\nas though the config file contained exactly\nDEVICE partitions containers\nand will read /proc/partitions to find a list of devices to scan, and /proc/mdstat to\nfind a list of containers to examine. If the word none is given for the config file,\nthen mdadm will act as though the config file were empty.\n\nIf the name given is of a directory, then mdadm will collect all the files contained\nin the directory with a name ending in .conf, sort them lexically, and process all of\nthose files as config files.\n\n", "flag": "-c" }, { "name": "-s --scan", "content": "Scan config file or /proc/mdstat for missing information. In general, this option\ngives mdadm permission to get any missing information (like component devices, array\ndevices, array identities, and alert destination) from the configuration file (see\nprevious option); one exception is MISC mode when using --detail or --stop, in which\ncase --scan says to get a list of array devices from /proc/mdstat.\n\n", "flag": "-s", "long": "--scan" }, { "name": "-e --metadata=", "content": "Declare the style of RAID metadata (superblock) to be used. The default is 1.2 for\n--create, and to guess for other operations. The default can be overridden by setting\nthe metadata value for the CREATE keyword in mdadm.conf.\n\nOptions are:\n\n\n0, 0.90\nUse the original 0.90 format superblock. This format limits arrays to 28 com‐\nponent devices and limits component devices of levels 1 and greater to 2 ter‐\nabytes. It is also possible for there to be confusion about whether the su‐\nperblock applies to a whole device or just the last partition, if that parti‐\ntion starts on a 64K boundary.\n\n\n1, 1.0, 1.1, 1.2 default\nUse the new version-1 format superblock. This has fewer restrictions. It can\neasily be moved between hosts with different endian-ness, and a recovery opera‐\ntion can be checkpointed and restarted. The different sub-versions store the\nsuperblock at different locations on the device, either at the end (for 1.0),\nat the start (for 1.1) or 4K from the start (for 1.2). \"1\" is equivalent to\n\"1.2\" (the commonly preferred 1.x format). \"default\" is equivalent to \"1.2\".\n\nddf Use the \"Industry Standard\" DDF (Disk Data Format) format defined by SNIA.\nWhen creating a DDF array a CONTAINER will be created, and normal arrays can be\ncreated in that container.\n\nimsm Use the Intel(R) Matrix Storage Manager metadata format. This creates a CON‐\nTAINER which is managed in a similar manner to DDF, and is supported by an op‐\ntion-rom on some platforms:\n\nhttps://www.intel.com/content/www/us/en/support/products/122484\n", "flag": "-e" }, { "name": "--homehost=", "content": "This will override any HOMEHOST setting in the config file and provides the identity\nof the host which should be considered the home for any arrays.\n\nWhen creating an array, the homehost will be recorded in the metadata. For version-1\nsuperblocks, it will be prefixed to the array name. For version-0.90 superblocks,\npart of the SHA1 hash of the hostname will be stored in the latter half of the UUID.\n\nWhen reporting information about an array, any array which is tagged for the given\nhomehost will be reported as such.\n\nWhen using Auto-Assemble, only arrays tagged for the given homehost will be allowed to\nuse 'local' names (i.e. not ending in '' followed by a digit string). See below un‐\nder Auto-Assembly.\n\nThe special name \"any\" can be used as a wild card. If an array is created with\n--homehost=any then the name \"any\" will be stored in the array and it can be assembled\nin the same way on any host. If an array is assembled with this option, then the\nhomehost recorded on the array will be ignored.\n\n" }, { "name": "--prefer=", "content": "When mdadm needs to print the name for a device it normally finds the name in /dev\nwhich refers to the device and is the shortest. When a path component is given with\n--prefer mdadm will prefer a longer name if it contains that component. For example\n--prefer=by-uuid will prefer a name in a subdirectory of /dev called by-uuid.\n\nThis functionality is currently only provided by --detail and --monitor.\n\n" }, { "name": "--home-cluster=", "content": "specifies the cluster name for the md device. The md device can be assembled only on\nthe cluster which matches the name specified. If this option is not provided, mdadm\ntries to detect the cluster name automatically.\n\n\nFor create, build, or grow:" }, { "name": "-n --raid-devices=", "content": "Specify the number of active devices in the array. This, plus the number of spare de‐\nvices (see below) must equal the number of component-devices (including \"missing\" de‐\nvices) that are listed on the command line for --create. Setting a value of 1 is\nprobably a mistake and so requires that --force be specified first. A value of 1 will\nthen be allowed for linear, multipath, RAID0 and RAID1. It is never allowed for\nRAID4, RAID5 or RAID6.\nThis number can only be changed using --grow for RAID1, RAID4, RAID5 and RAID6 arrays,\nand only on kernels which provide the necessary support.\n\n", "flag": "-n" }, { "name": "-x --spare-devices=", "content": "Specify the number of spare (eXtra) devices in the initial array. Spares can also be\nadded and removed later. The number of component devices listed on the command line\nmust equal the number of RAID devices plus the number of spare devices.\n\n", "flag": "-x" }, { "name": "-z --size=", "content": "Amount (in Kilobytes) of space to use from each drive in RAID levels 1/4/5/6/10 and\nfor RAID 0 on external metadata. This must be a multiple of the chunk size, and must\nleave about 128Kb of space at the end of the drive for the RAID superblock. When spec‐\nified as ¸max¸ (as it often is) the smallest drive (or partition) sets the size. In\nthat case, a warning will follow if the drives, as a group, have sizes that differ by\nmore than one percent.\n\nA suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes, Megabytes, Giga‐\nbytes or Terabytes respectively.\n\nSometimes a replacement drive can be a little smaller than the original drives though\nthis should be minimised by IDEMA standards. Such a replacement drive will be re‐\njected by md. To guard against this it can be useful to set the initial size slightly\nsmaller than the smaller device with the aim that it will still be larger than any re‐\nplacement.\n\nThis option can be used with --create for determining the initial size of an array.\nFor external metadata, it can be used on a volume, but not on a container itself.\nSetting the initial size of RAID 0 array is only valid for external metadata.\n\nThis value can be set with --grow for RAID level 1/4/5/6/10 though DDF arrays may not\nbe able to support this. RAID 0 array size cannot be changed. If the array was cre‐\nated with a size smaller than the currently active drives, the extra space can be ac‐\ncessed using --grow. The size can be given as max which means to choose the largest\nsize that fits on all current drives.\n\nBefore reducing the size of the array (with --grow --size=) you should make sure that\nspace isn't needed. If the device holds a filesystem, you would need to resize the\nfilesystem to use less space.\n\nAfter reducing the array size you should check that the data stored in the device is\nstill available. If the device holds a filesystem, then an 'fsck' of the filesystem\nis a minimum requirement. If there are problems the array can be made bigger again\nwith no loss with another --grow --size= command.\n\n", "flag": "-z" }, { "name": "-Z --array-size=", "content": "This is only meaningful with --grow and its effect is not persistent: when the array\nis stopped and restarted the default array size will be restored.\n\nSetting the array-size causes the array to appear smaller to programs that access the\ndata. This is particularly needed before reshaping an array so that it will be\nsmaller. As the reshape is not reversible, but setting the size with --array-size is,\nit is required that the array size is reduced as appropriate before the number of de‐\nvices in the array is reduced.\n\nBefore reducing the size of the array you should make sure that space isn't needed.\nIf the device holds a filesystem, you would need to resize the filesystem to use less\nspace.\n\nAfter reducing the array size you should check that the data stored in the device is\nstill available. If the device holds a filesystem, then an 'fsck' of the filesystem\nis a minimum requirement. If there are problems the array can be made bigger again\nwith no loss with another --grow --array-size= command.\n\nA suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes, Megabytes, Giga‐\nbytes or Terabytes respectively. A value of max restores the apparent size of the ar‐\nray to be whatever the real amount of available space is.\n\nClustered arrays do not support this parameter yet.\n\n", "flag": "-Z" }, { "name": "-c --chunk=", "content": "Specify chunk size in kilobytes. The default when creating an array is 512KB. To en‐\nsure compatibility with earlier versions, the default when building an array with no\npersistent metadata is 64KB. This is only meaningful for RAID0, RAID4, RAID5, RAID6,\nand RAID10.\n\nRAID4, RAID5, RAID6, and RAID10 require the chunk size to be a power of 2, with mini‐\nmal chunk size being 4KB.\n\nA suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes, Megabytes, Giga‐\nbytes or Terabytes respectively.\n\n", "flag": "-c" }, { "name": "--rounding=", "content": "Specify the rounding factor for a Linear array. The size of each component will be\nrounded down to a multiple of this size. This is a synonym for --chunk but highlights\nthe different meaning for Linear as compared to other RAID levels. The default is 64K\nif a kernel earlier than 2.6.16 is in use, and is 0K (i.e. no rounding) in later ker‐\nnels.\n\n" }, { "name": "-l --level=", "content": "Set RAID level. When used with --create, options are: linear, raid0, 0, stripe,\nraid1, 1, mirror, raid4, 4, raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty,\ncontainer. Obviously some of these are synonymous.\n\nWhen a CONTAINER metadata type is requested, only the container level is permitted,\nand it does not need to be explicitly given.\n\nWhen used with --build, only linear, stripe, raid0, 0, raid1, multipath, mp, and\nfaulty are valid.\n\nCan be used with --grow to change the RAID level in some cases. See LEVEL CHANGES be‐\nlow.\n\n", "flag": "-l" }, { "name": "-p --layout=", "content": "This option configures the fine details of data layout for RAID5, RAID6, and RAID10\narrays, and controls the failure modes for faulty. It can also be used for working\naround a kernel bug with RAID0, but generally doesn't need to be used explicitly.\n\nThe layout of the RAID5 parity block can be one of left-asymmetric, left-symmetric,\nright-asymmetric, right-symmetric, la, ra, ls, rs. The default is left-symmetric.\n\nIt is also possible to cause RAID5 to use a RAID4-like layout by choosing par‐\nity-first, or parity-last.\n\nFinally for RAID5 there are DDF-compatible layouts, ddf-zero-restart, ddf-N-restart,\nand ddf-N-continue.\n\nThese same layouts are available for RAID6. There are also 4 layouts that will pro‐\nvide an intermediate stage for converting between RAID5 and RAID6. These provide a\nlayout which is identical to the corresponding RAID5 layout on the first N-1 devices,\nand has the 'Q' syndrome (the second 'parity' block used by RAID6) on the last device.\nThese layouts are: left-symmetric-6, right-symmetric-6, left-asymmetric-6, right-asym‐\nmetric-6, and parity-first-6.\n\nWhen setting the failure mode for level faulty, the options are: write-transient, wt,\nread-transient, rt, write-persistent, wp, read-persistent, rp, write-all, read-fix‐\nable, rf, clear, flush, none.\n\nEach failure mode can be followed by a number, which is used as a period between fault\ngeneration. Without a number, the fault is generated once on the first relevant re‐\nquest. With a number, the fault will be generated after that many requests, and will\ncontinue to be generated every time the period elapses.\n\nMultiple failure modes can be current simultaneously by using the --grow option to set\nsubsequent failure modes.\n\n\"clear\" or \"none\" will remove any pending or periodic failure modes, and \"flush\" will\nclear any persistent faults.\n\nThe layout options for RAID10 are one of 'n', 'o' or 'f' followed by a small number\nsignifying the number of copies of each datablock. The default is 'n2'. The sup‐\nported options are:\n\n'n' signals 'near' copies. Multiple copies of one data block are at similar offsets\nin different devices.\n\n'o' signals 'offset' copies. Rather than the chunks being duplicated within a stripe,\nwhole stripes are duplicated but are rotated by one device so duplicate blocks are on\ndifferent devices. Thus subsequent copies of a block are in the next drive, and are\none chunk further down.\n\n'f' signals 'far' copies (multiple copies have very different offsets). See md(4) for\nmore detail about 'near', 'offset', and 'far'.\n\nAs for the number of copies of each data block, 2 is normal, 3 can be useful. This\nnumber can be at most equal to the number of devices in the array. It does not need\nto divide evenly into that number (e.g. it is perfectly legal to have an 'n2' layout\nfor an array with an odd number of devices).\n\nA bug introduced in Linux 3.14 means that RAID0 arrays with devices of differing sizes\nstarted using a different layout. This could lead to data corruption. Since Linux\n5.4 (and various stable releases that received backports), the kernel will not accept\nsuch an array unless a layout is explicitly set. It can be set to 'original' or 'al‐\nternate'. When creating a new array, mdadm will select 'original' by default, so the\nlayout does not normally need to be set. An array created for either 'original' or\n'alternate' will not be recognized by an (unpatched) kernel prior to 5.4. To create a\nRAID0 array with devices of differing sizes that can be used on an older kernel, you\ncan set the layout to 'dangerous'. This will use whichever layout the running kernel\nsupports, so the data on the array may become corrupt when changing kernel from\npre-3.14 to a later kernel.\n\nWhen an array is converted between RAID5 and RAID6 an intermediate RAID6 layout is\nused in which the second parity block (Q) is always on the last device. To convert a\nRAID5 to RAID6 and leave it in this new layout (which does not require re-striping)\nuse --layout=preserve. This will try to avoid any restriping.\n\nThe converse of this is --layout=normalise which will change a non-standard RAID6 lay‐\nout into a more standard arrangement.\n\n", "flag": "-p" }, { "name": "--parity=", "content": "same as --layout (thus explaining the p of -p).\n\n" }, { "name": "-b --bitmap=", "content": "Specify a file to store a write-intent bitmap in. The file should not exist unless\n--force is also given. The same file should be provided when assembling the array.\nIf the word internal is given, then the bitmap is stored with the metadata on the ar‐\nray, and so is replicated on all devices. If the word none is given with --grow mode,\nthen any bitmap that is present is removed. If the word clustered is given, the array\nis created for a clustered environment. One bitmap is created for each node as defined\nby the --nodes parameter and are stored internally.\n\nTo help catch typing errors, the filename must contain at least one slash ('/') if it\nis a real file (not 'internal' or 'none').\n\nNote: external bitmaps are only known to work on ext2 and ext3. Storing bitmap files\non other filesystems may result in serious problems.\n\nWhen creating an array on devices which are 100G or larger, mdadm automatically adds\nan internal bitmap as it will usually be beneficial. This can be suppressed with\n--bitmap=none or by selecting a different consistency policy with --consistency-pol‐\nicy.\n\n", "flag": "-b" }, { "name": "--bitmap-chunk=", "content": "Set the chunk size of the bitmap. Each bit corresponds to that many Kilobytes of\nstorage. When using a file-based bitmap, the default is to use the smallest size that\nis at least 4 and requires no more than 2^21 chunks. When using an internal bitmap,\nthe chunk size defaults to 64Meg, or larger if necessary to fit the bitmap into the\navailable space.\n\nA suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes, Megabytes, Giga‐\nbytes or Terabytes respectively.\n\n" }, { "name": "-W --write-mostly", "content": "subsequent devices listed in a --build, --create, or --add command will be flagged as\n'write-mostly'. This is valid for RAID1 only and means that the 'md' driver will\navoid reading from these devices if at all possible. This can be useful if mirroring\nover a slow link.\n\n", "flag": "-W", "long": "--write-mostly" }, { "name": "--write-behind=", "content": "Specify that write-behind mode should be enabled (valid for RAID1 only). If an argu‐\nment is specified, it will set the maximum number of outstanding writes allowed. The\ndefault value is 256. A write-intent bitmap is required in order to use write-behind\nmode, and write-behind is only attempted on drives marked as write-mostly.\n\n" }, { "name": "--failfast", "content": "subsequent devices listed in a --create or --add command will be flagged as 'fail‐\nfast'. This is valid for RAID1 and RAID10 only. IO requests to these devices will be\nencouraged to fail quickly rather than cause long delays due to error handling. Also\nno attempt is made to repair a read error on these devices.\n\nIf an array becomes degraded so that the 'failfast' device is the only usable device,\nthe 'failfast' flag will then be ignored and extended delays will be preferred to com‐\nplete failure.\n\nThe 'failfast' flag is appropriate for storage arrays which have a low probability of\ntrue failure, but which may sometimes cause unacceptable delays due to internal main‐\ntenance functions.\n\n", "long": "--failfast" }, { "name": "--assume-clean", "content": "Tell mdadm that the array pre-existed and is known to be clean. It can be useful when\ntrying to recover from a major failure as you can be sure that no data will be af‐\nfected unless you actually write to the array. It can also be used when creating a\nRAID1 or RAID10 if you want to avoid the initial resync, however this practice — while\nnormally safe — is not recommended. Use this only if you really know what you are do‐\ning.\n\nWhen the devices that will be part of a new array were filled with zeros before cre‐\nation the operator knows the array is actually clean. If that is the case, such as af‐\nter running badblocks, this argument can be used to tell mdadm the facts the operator\nknows.\n\nWhen an array is resized to a larger size with --grow --size= the new space is nor‐\nmally resynced in that same way that the whole array is resynced at creation. From\nLinux version 3.0, --assume-clean can be used with that command to avoid the automatic\nresync.\n\n", "long": "--assume-clean" }, { "name": "--write-zeroes", "content": "When creating an array, send write zeroes requests to all the block devices. This\nshould zero the data area on all disks such that the initial sync is not necessary\nand, if successfull, will behave as if --assume-clean was specified.\n\nThis is intended for use with devices that have hardware offload for zeroing, but de‐\nspite this zeroing can still take several minutes for large disks. Thus a message is\nprinted before and after zeroing and each disk is zeroed in parallel with the others.\n\nThis is only meaningful with --create.\n\n", "long": "--write-zeroes" }, { "name": "--backup-file=", "content": "This is needed when --grow is used to increase the number of raid devices in a RAID5\nor RAID6 if there are no spare devices available, or to shrink, change RAID level or\nlayout. See the GROW MODE section below on RAID-DEVICES CHANGES. The file must be\nstored on a separate device, not on the RAID array being reshaped.\n\n" }, { "name": "--data-offset=", "content": "Arrays with 1.x metadata can leave a gap between the start of the device and the start\nof array data. This gap can be used for various metadata. The start of data is known\nas the data-offset. Normally an appropriate data offset is computed automatically.\nHowever it can be useful to set it explicitly such as when re-creating an array which\nwas originally created using a different version of mdadm which computed a different\noffset.\n\nSetting the offset explicitly over-rides the default. The value given is in Kilobytes\nunless a suffix of 'K', 'M', 'G' or 'T' is used to explicitly indicate Kilobytes,\nMegabytes, Gigabytes or Terabytes respectively.\n\nSince Linux 3.4, --data-offset can also be used with --grow for some RAID levels (ini‐\ntially on RAID10). This allows the data-offset to be changed as part of the reshape\nprocess. When the data offset is changed, no backup file is required as the differ‐\nence in offsets is used to provide the same functionality.\n\nWhen the new offset is earlier than the old offset, the number of devices in the array\ncannot shrink. When it is after the old offset, the number of devices in the array\ncannot increase.\n\nWhen creating an array, --data-offset can be specified as variable. In the case each\nmember device is expected to have an offset appended to the name, separated by a\ncolon. This makes it possible to recreate exactly an array which has varying data\noffsets (as can happen when different versions of mdadm are used to add different de‐\nvices).\n\n" }, { "name": "--continue", "content": "This option is complementary to the --freeze-reshape option for assembly. It is needed\nwhen --grow operation is interrupted and it is not restarted automatically due to\n--freeze-reshape usage during array assembly. This option is used together with -G ,\n( --grow ) command and device for a pending reshape to be continued. All parameters\nrequired for reshape continuation will be read from array metadata. If initial --grow\ncommand had required --backup-file= option to be set, continuation option will require\nto have exactly the same backup file given as well.\n\nAny other parameter passed together with --continue option will be ignored.\n\n", "long": "--continue" }, { "name": "-N --name=", "content": "Set a name for the array. It must be POSIX PORTABLE NAME compatible and cannot be\nlonger than 32 chars. This is effective when creating an array with a v1 metadata, or\nan external array.\n\nIf name is needed but not specified, it is taken from the basename of the device that\nis being created. See DEVICE NAMES\n\n", "flag": "-N" }, { "name": "-R --run", "content": "Insist that mdadm run the array, even if some of the components appear to be active in\nanother array or filesystem. Normally mdadm will ask for confirmation before includ‐\ning such components in an array. This option causes that question to be suppressed.\n\n", "flag": "-R", "long": "--run" }, { "name": "-f --force", "content": "Insist that mdadm accept the geometry and layout specified without question. Normally\nmdadm will not allow the creation of an array with only one device, and will try to\ncreate a RAID5 array with one missing drive (as this makes the initial resync work\nfaster). With --force, mdadm will not try to be so clever.\n\n", "flag": "-f", "long": "--force" }, { "name": "-o --readonly", "content": "Start the array read only rather than read-write as normal. No writes will be allowed\nto the array, and no resync, recovery, or reshape will be started. It works with Cre‐\nate, Assemble, Manage and Misc mode.\n\n", "flag": "-o", "long": "--readonly" }, { "name": "-a --auto{=yes,md,mdp,part,p}{NN}", "content": "Instruct mdadm how to create the device file if needed, possibly allocating an unused\nminor number. \"md\" causes a non-partitionable array to be used (though since Linux\n2.6.28, these array devices are in fact partitionable). \"mdp\", \"part\" or \"p\" causes a\npartitionable array (2.6 and later) to be used. \"yes\" requires the named md device to\nhave a 'standard' format, and the type and minor number will be determined from this.\nWith mdadm 3.0, device creation is normally left up to udev so this option is unlikely\nto be needed. See DEVICE NAMES below.\n\nThe argument can also come immediately after \"-a\". e.g. \"-ap\".\n\nIf --auto is not given on the command line or in the config file, then the default\nwill be --auto=yes.\n\nIf --scan is also given, then any auto= entries in the config file will override the\n--auto instruction given on the command line.\n\nFor partitionable arrays, mdadm will create the device file for the whole array and\nfor the first 4 partitions. A different number of partitions can be specified at the\nend of this option (e.g. --auto=p7). If the device name ends with a digit, the par‐\ntition names add a 'p', and a number, e.g. /dev/md/home1p3. If there is no trailing\ndigit, then the partition names just have a number added, e.g. /dev/md/scratch3.\n\nIf the md device name is in a 'standard' format as described in DEVICE NAMES, then it\nwill be created, if necessary, with the appropriate device number based on that name.\nIf the device name is not in one of these formats, then an unused device number will\nbe allocated. The device number will be considered unused if there is no active array\nfor that number, and there is no entry in /dev for that number and with a non-standard\nname. Names that are not in 'standard' format are only allowed in \"/dev/md/\".\n\nThis is meaningful with --create or --build.\n\n", "flag": "-a" }, { "name": "-a --add", "content": "This option can be used in Grow mode in two cases.\n\nIf the target array is a Linear array, then --add can be used to add one or more de‐\nvices to the array. They are simply catenated on to the end of the array. Once\nadded, the devices cannot be removed.\n\nIf the --raid-disks option is being used to increase the number of devices in an ar‐\nray, then --add can be used to add some extra devices to be included in the array. In\nmost cases this is not needed as the extra devices can be added as spares first, and\nthen the number of raid disks can be changed. However, for RAID0 it is not possible\nto add spares. So to increase the number of devices in a RAID0, it is necessary to\nset the new number of devices, and to add the new devices, in the same command.\n\n", "flag": "-a", "long": "--add" }, { "name": "--nodes", "content": "Only works when the array is created for a clustered environment. It specifies the\nmaximum number of nodes in the cluster that will use this device simultaneously. If\nnot specified, this defaults to 4.\n\n", "long": "--nodes" }, { "name": "--write-journal", "content": "Specify journal device for the RAID-4/5/6 array. The journal device should be an SSD\nwith a reasonable lifetime.\n\n", "long": "--write-journal" }, { "name": "-k --consistency-policy=", "content": "Specify how the array maintains consistency in the case of an unexpected shutdown.\nOnly relevant for RAID levels with redundancy. Currently supported options are:\n\n\nresync Full resync is performed and all redundancy is regenerated when the array is\nstarted after an unclean shutdown.\n\n\nbitmap Resync assisted by a write-intent bitmap. Implicitly selected when using\n--bitmap.\n\n\njournal\nFor RAID levels 4/5/6, the journal device is used to log transactions and re‐\nplay after an unclean shutdown. Implicitly selected when using --write-journal.\n\n\nppl For RAID5 only, Partial Parity Log is used to close the write hole and elimi‐\nnate resync. PPL is stored in the metadata region of RAID member drives, no ad‐\nditional journal drive is needed.\n\n\nCan be used with --grow to change the consistency policy of an active array in some\ncases. See CONSISTENCY POLICY CHANGES below.\n\n\n\nFor assemble:", "flag": "-k" }, { "name": "-u --uuid=", "content": "uuid of array to assemble. Devices which don't have this uuid are excluded\n\n", "flag": "-u" }, { "name": "-m --super-minor=", "content": "Minor number of device that array was created for. Devices which don't have this mi‐\nnor number are excluded. If you create an array as /dev/md1, then all superblocks\nwill contain the minor number 1, even if the array is later assembled as /dev/md2.\n\nGiving the literal word \"dev\" for --super-minor will cause mdadm to use the minor num‐\nber of the md device that is being assembled. e.g. when assembling /dev/md0, --su‐\nper-minor=dev will look for super blocks with a minor number of 0.\n\n--super-minor is only relevant for v0.90 metadata, and should not normally be used.\nUsing --uuid is much safer.\n\n", "flag": "-m" }, { "name": "-N --name=", "content": "Specify the name of the array to assemble. It must be POSIX PORTABLE NAME compatible\nand cannot be longer than 32 chars. This must be the name that was specified when cre‐\nating the array. It must either match the name stored in the superblock exactly, or it\nmust match with the current homehost prefixed to the start of the given name.\n\n", "flag": "-N" }, { "name": "-f --force", "content": "Assemble the array even if the metadata on some devices appears to be out-of-date. If\nmdadm cannot find enough working devices to start the array, but can find some devices\nthat are recorded as having failed, then it will mark those devices as working so that\nthe array can be started. This works only for native. For external metadata it allows\none to start dirty degraded RAID 4, 5, 6. An array which requires --force to be\nstarted may contain data corruption. Use it carefully.\n\n", "flag": "-f", "long": "--force" }, { "name": "-R --run", "content": "Attempt to start the array even if fewer drives were given than were present last time\nthe array was active. Normally if not all the expected drives are found and --scan is\nnot used, then the array will be assembled but not started. With --run an attempt\nwill be made to start it anyway.\n\n", "flag": "-R", "long": "--run" }, { "name": "--no-degraded", "content": "This is the reverse of --run in that it inhibits the startup of array unless all ex‐\npected drives are present. This is only needed with --scan, and can be used if the\nphysical connections to devices are not as reliable as you would like.\n\n", "long": "--no-degraded" }, { "name": "-a --auto{=no,yes,md,mdp,part}", "content": "See this option under Create and Build options.\n\n", "flag": "-a" }, { "name": "-b --bitmap=", "content": "Specify the bitmap file that was given when the array was created. If an array has an\ninternal bitmap, there is no need to specify this when assembling the array.\n\n", "flag": "-b" }, { "name": "--backup-file=", "content": "If --backup-file was used while reshaping an array (e.g. changing number of devices or\nchunk size) and the system crashed during the critical section, then the same\n--backup-file must be presented to --assemble to allow possibly corrupted data to be\nrestored, and the reshape to be completed.\n\n" }, { "name": "--invalid-backup", "content": "If the file needed for the above option is not available for any reason an empty file\ncan be given together with this option to indicate that the backup file is invalid.\nIn this case the data that was being rearranged at the time of the crash could be ir‐\nrecoverably lost, but the rest of the array may still be recoverable. This option\nshould only be used as a last resort if there is no way to recover the backup file.\n\n\n", "long": "--invalid-backup" }, { "name": "-U --update=", "content": "Update the superblock on each device while assembling the array. The argument given\nto this flag can be one of sparc2.2, summaries, uuid, name, nodes, homehost, home-\ncluster, resync, byteorder, devicesize, no-bitmap, bbl, no-bbl, ppl, no-ppl, lay‐\nout-original, layout-alternate, layout-unspecified, metadata, or super-minor.\n\nThe sparc2.2 option will adjust the superblock of an array what was created on a Sparc\nmachine running a patched 2.2 Linux kernel. This kernel got the alignment of part of\nthe superblock wrong. You can use the --examine --sparc2.2 option to mdadm to see\nwhat effect this would have.\n\nThe super-minor option will update the preferred minor field on each superblock to\nmatch the minor number of the array being assembled. This can be useful if --examine\nreports a different \"Preferred Minor\" to --detail. In some cases this update will be\nperformed automatically by the kernel driver. In particular, the update happens auto‐\nmatically at the first write to an array with redundancy (RAID level 1 or greater) on\na 2.6 (or later) kernel.\n\nThe uuid option will change the uuid of the array. If a UUID is given with the --uuid\noption that UUID will be used as a new UUID and will NOT be used to help identify the\ndevices in the array. If no --uuid is given, a random UUID is chosen.\n\nThe name option will change the name of the array as stored in the superblock. This\nis only supported for version-1 superblocks.\n\nThe nodes option will change the nodes of the array as stored in the bitmap su‐\nperblock. This option only works for a clustered environment.\n\nThe homehost option will change the homehost as recorded in the superblock. For ver‐\nsion-0 superblocks, this is the same as updating the UUID. For version-1 superblocks,\nthis involves updating the name.\n\nThe home-cluster option will change the cluster name as recorded in the superblock and\nbitmap. This option only works for a clustered environment.\n\nThe resync option will cause the array to be marked dirty meaning that any redundancy\nin the array (e.g. parity for RAID5, copies for RAID1) may be incorrect. This will\ncause the RAID system to perform a \"resync\" pass to make sure that all redundant in‐\nformation is correct.\n\nThe byteorder option allows arrays to be moved between machines with different byte-\norder, such as from a big-endian machine like a Sparc or some MIPS machines, to a lit‐\ntle-endian x8664 machine. When assembling such an array for the first time after a\nmove, giving --update=byteorder will cause mdadm to expect superblocks to have their\nbyteorder reversed, and will correct that order before assembling the array. This is\nonly valid with original (Version 0.90) superblocks.\n\nThe summaries option will correct the summaries in the superblock. That is the counts\nof total, working, active, failed, and spare devices.\n\nThe devicesize option will rarely be of use. It applies to version 1.1 and 1.2 meta‐\ndata only (where the metadata is at the start of the device) and is only useful when\nthe component device has changed size (typically become larger). The version 1 meta‐\ndata records the amount of the device that can be used to store data, so if a device\nin a version 1.1 or 1.2 array becomes larger, the metadata will still be visible, but\nthe extra space will not. In this case it might be useful to assemble the array with\n--update=devicesize. This will cause mdadm to determine the maximum usable amount of\nspace on each device and update the relevant field in the metadata.\n\nThe metadata option only works on v0.90 metadata arrays and will convert them to v1.0\nmetadata. The array must not be dirty (i.e. it must not need a sync) and it must not\nhave a write-intent bitmap.\n\nThe old metadata will remain on the devices, but will appear older than the new meta‐\ndata and so will usually be ignored. The old metadata (or indeed the new metadata) can\nbe removed by giving the appropriate --metadata= option to --zero-superblock.\n\nThe no-bitmap option can be used when an array has an internal bitmap which is corrupt\nin some way so that assembling the array normally fails. It will cause any internal\nbitmap to be ignored.\n\nThe bbl option will reserve space in each device for a bad block list. This will be\n4K in size and positioned near the end of any free space between the superblock and\nthe data.\n\nThe no-bbl option will cause any reservation of space for a bad block list to be re‐\nmoved. If the bad block list contains entries, this will fail, as removing the list\ncould cause data corruption.\n\nThe ppl option will enable PPL for a RAID5 array and reserve space for PPL on each de‐\nvice. There must be enough free space between the data and superblock and a write-in‐\ntent bitmap or journal must not be used.\n\nThe no-ppl option will disable PPL in the superblock.\n\nThe layout-original and layout-alternate options are for RAID0 arrays with non-uniform\ndevices size that were in use before Linux 5.4. If the array was being used with\nLinux 3.13 or earlier, then to assemble the array on a new kernel, --update=lay‐\nout-original must be given. If the array was created and used with a kernel from\nLinux 3.14 to Linux 5.3, then --update=layout-alternate must be given. This only\nneeds to be given once. Subsequent assembly of the array will happen normally. For\nmore information, see md(4).\n\nThe layout-unspecified option reverts the effect of layout-orignal or layout-alternate\nand allows the array to be again used on a kernel prior to Linux 5.3. This option\nshould be used with great caution.\n\n", "flag": "-U" }, { "name": "--freeze-reshape", "content": "This option is intended to be used in start-up scripts during the initrd boot phase.\nWhen the array under reshape is assembled during the initrd phase, this option stops\nthe reshape after the reshape-critical section has been restored. This happens before\nthe file system pivot operation and avoids loss of filesystem context. Losing file\nsystem context would cause reshape to be broken.\n\nReshape can be continued later using the --continue option for the grow command.\n\n\nFor Manage mode:", "long": "--freeze-reshape" }, { "name": "-t --test", "content": "Unless a more serious error occurred, mdadm will exit with a status of 2 if no changes\nwere made to the array and 0 if at least one change was made. This can be useful when\nan indirect specifier such as missing, detached or faulty is used in requesting an op‐\neration on the array. --test will report failure if these specifiers didn't find any\nmatch.\n\n", "flag": "-t", "long": "--test" }, { "name": "-a --add", "content": "hot-add listed devices. If a device appears to have recently been part of the array\n(possibly it failed or was removed) the device is re-added as described in the next\npoint. If that fails or the device was never part of the array, the device is added\nas a hot-spare. If the array is degraded, it will immediately start to rebuild data\nonto that spare.\n\nNote that this and the following options are only meaningful on array with redundancy.\nThey don't apply to RAID0 or Linear.\n\n", "flag": "-a", "long": "--add" }, { "name": "--re-add", "content": "re-add a device that was previously removed from an array. If the metadata on the de‐\nvice reports that it is a member of the array, and the slot that it used is still va‐\ncant, then the device will be added back to the array in the same position. This will\nnormally cause the data for that device to be recovered. However, based on the event\ncount on the device, the recovery may only require sections that are flagged by a\nwrite-intent bitmap to be recovered or may not require any recovery at all.\n\nWhen used on an array that has no metadata (i.e. it was built with --build) it will be\nassumed that bitmap-based recovery is enough to make the device fully consistent with\nthe array.\n\n--re-add can also be accompanied by --update=devicesize, --update=bbl, or --up‐\ndate=no-bbl. See descriptions of these options when used in Assemble mode for an ex‐\nplanation of their use.\n\nIf the device name given is missing then mdadm will try to find any device that looks\nlike it should be part of the array but isn't and will try to re-add all such devices.\n\nIf the device name given is faulty then mdadm will find all devices in the array that\nare marked faulty, remove them and attempt to immediately re-add them. This can be\nuseful if you are certain that the reason for failure has been resolved.\n\n", "long": "--re-add" }, { "name": "--add-spare", "content": "Add a device as a spare. This is similar to --add except that it does not attempt\n--re-add first. The device will be added as a spare even if it looks like it could be\na recent member of the array.\n\n", "long": "--add-spare" }, { "name": "-r --remove", "content": "remove listed devices. They must not be active. i.e. they should be failed or spare\ndevices.\n\nAs well as the name of a device file (e.g. /dev/sda1) the words failed, detached and\nnames like set-A can be given to --remove. The first causes all failed devices to be\nremoved. The second causes any device which is no longer connected to the system (i.e\nan 'open' returns ENXIO) to be removed. The third will remove a set as described be‐\nlow under --fail.\n\n", "flag": "-r", "long": "--remove" }, { "name": "-f --fail", "content": "Mark listed devices as faulty. As well as the name of a device file, the word de‐\ntached or a set name like set-A can be given. The former will cause any device that\nhas been detached from the system to be marked as failed. It can then be removed.\n\nFor RAID10 arrays where the number of copies evenly divides the number of devices, the\ndevices can be conceptually divided into sets where each set contains a single com‐\nplete copy of the data on the array. Sometimes a RAID10 array will be configured so\nthat these sets are on separate controllers. In this case, all the devices in one set\ncan be failed by giving a name like set-A or set-B to --fail. The appropriate set\nnames are reported by --detail.\n\n", "flag": "-f", "long": "--fail" }, { "name": "--set-faulty", "content": "same as --fail.\n\n", "long": "--set-faulty" }, { "name": "--replace", "content": "Mark listed devices as requiring replacement. As soon as a spare is available, it\nwill be rebuilt and will replace the marked device. This is similar to marking a de‐\nvice as faulty, but the device remains in service during the recovery process to in‐\ncrease resilience against multiple failures. When the replacement process finishes,\nthe replaced device will be marked as faulty.\n\n\n--with This can follow a list of --replace devices. The devices listed after --with will\npreferentially be used to replace the devices listed after --replace. These devices\nmust already be spare devices in the array.\n\n", "long": "--replace" }, { "name": "--write-mostly", "content": "Subsequent devices that are added or re-added will have the 'write-mostly' flag set.\nThis is only valid for RAID1 and means that the 'md' driver will avoid reading from\nthese devices if possible.\n", "long": "--write-mostly" }, { "name": "--readwrite", "content": "Subsequent devices that are added or re-added will have the 'write-mostly' flag\ncleared.\n", "long": "--readwrite" }, { "name": "--cluster-confirm", "content": "Confirm the existence of the device. This is issued in response to an --add request by\na node in a cluster. When a node adds a device it sends a message to all nodes in the\ncluster to look for a device with a UUID. This translates to a udev notification with\nthe UUID of the device to be added and the slot number. The receiving node must ac‐\nknowledge this message with --cluster-confirm. Valid arguments are :\nin case the device is found or :missing in case the device is not found.\n\n", "long": "--cluster-confirm" }, { "name": "--add-journal", "content": "Add a journal to an existing array, or recreate journal for a RAID-4/5/6 array that\nlost a journal device. To avoid interrupting ongoing write operations, --add-journal\nonly works for array in Read-Only state.\n\n", "long": "--add-journal" }, { "name": "--failfast", "content": "Subsequent devices that are added or re-added will have the 'failfast' flag set. This\nis only valid for RAID1 and RAID10 and means that the 'md' driver will avoid long\ntimeouts on error handling where possible.\n", "long": "--failfast" }, { "name": "--nofailfast", "content": "Subsequent devices that are re-added will be re-added without the 'failfast' flag set.\n\n\nEach of these options requires that the first device listed is the array to be acted upon,\nand the remainder are component devices to be added, removed, marked as faulty, etc. Several\ndifferent operations can be specified for different devices, e.g.\nmdadm /dev/md0 --add /dev/sda1 --fail /dev/sdb1 --remove /dev/sdb1\nEach operation applies to all devices listed until the next operation.\n\nIf an array is using a write-intent bitmap, then devices which have been removed can be\nre-added in a way that avoids a full reconstruction but instead just updates the blocks that\nhave changed since the device was removed. For arrays with persistent metadata (superblocks)\nthis is done automatically. For arrays created with --build mdadm needs to be told that this\ndevice we removed recently with --re-add.\n\nDevices can only be removed from an array if they are not in active use, i.e. that must be\nspares or failed devices. To remove an active device, it must first be marked as faulty.\n\n\nFor Misc mode:", "long": "--nofailfast" }, { "name": "-Q --query", "content": "Examine a device to see (1) if it is an md device and (2) if it is a component of an\nmd array. Information about what is discovered is presented.\n\n", "flag": "-Q", "long": "--query" }, { "name": "-D --detail", "content": "Print details of one or more md devices.\n\n", "flag": "-D", "long": "--detail" }, { "name": "--detail-platform", "content": "Print details of the platform's RAID capabilities (firmware / hardware topology) for a\ngiven metadata format. If used without an argument, mdadm will scan all controllers\nlooking for their capabilities. Otherwise, mdadm will only look at the controller\nspecified by the argument in the form of an absolute filepath or a link, e.g.\n/sys/devices/pci0000:00/0000:00:1f.2.\n\n", "long": "--detail-platform" }, { "name": "-Y --export", "content": "When used with --detail, --detail-platform, --examine, or --incremental output will be\nformatted as key=value pairs for easy import into the environment.\n\nWith --incremental The value MDSTARTED indicates whether an array was started (yes)\nor not, which may include a reason (unsafe, nothing, no). Also the value MDFOREIGN\nindicates if the array is expected on this host (no), or seems to be from elsewhere\n(yes).\n\n", "flag": "-Y", "long": "--export" }, { "name": "-E --examine", "content": "Print contents of the metadata stored on the named device(s). Note the contrast be‐\ntween --examine and --detail. --examine applies to devices which are components of an\narray, while --detail applies to a whole array which is currently active.\n", "flag": "-E", "long": "--examine" }, { "name": "--sparc2.2", "content": "If an array was created on a SPARC machine with a 2.2 Linux kernel patched with RAID\nsupport, the superblock will have been created incorrectly, or at least incompatibly\nwith 2.4 and later kernels. Using the --sparc2.2 flag with --examine will fix the su‐\nperblock before displaying it. If this appears to do the right thing, then the array\ncan be successfully assembled using --assemble --update=sparc2.2.\n\n", "long": "--sparc2.2" }, { "name": "-X --examine-bitmap", "content": "Report information about a bitmap file. The argument is either an external bitmap\nfile or an array component in case of an internal bitmap. Note that running this on\nan array device (e.g. /dev/md0) does not report the bitmap for that array.\n\n", "flag": "-X", "long": "--examine-bitmap" }, { "name": "--examine-badblocks", "content": "List the bad-blocks recorded for the device, if a bad-blocks list has been configured.\nCurrently only 1.x and IMSM metadata support bad-blocks lists.\n\n\n--dump=directory\n\n--restore=directory\nSave metadata from lists devices, or restore metadata to listed devices.\n\n", "long": "--examine-badblocks" }, { "name": "-R --run", "content": "start a partially assembled array. If --assemble did not find enough devices to fully\nstart the array, it might leaving it partially assembled. If you wish, you can then\nuse --run to start the array in degraded mode.\n\n", "flag": "-R", "long": "--run" }, { "name": "-S --stop", "content": "deactivate array, releasing all resources.\n\n", "flag": "-S", "long": "--stop" }, { "name": "-o --readonly", "content": "mark array as readonly.\n\n", "flag": "-o", "long": "--readonly" }, { "name": "-w --readwrite", "content": "mark array as readwrite.\n\n", "flag": "-w", "long": "--readwrite" }, { "name": "--zero-superblock", "content": "If the device contains a valid md superblock, the block is overwritten with zeros.\nWith --force the block where the superblock would be is overwritten even if it doesn't\nappear to be valid.\n\nNote: Be careful when calling --zero-superblock with clustered raid. Make sure the ar‐\nray isn't used or assembled in another cluster node before executing it.\n\n", "long": "--zero-superblock" }, { "name": "--kill-subarray=", "content": "If the device is a container and the argument to --kill-subarray specifies an inactive\nsubarray in the container, then the subarray is deleted. Deleting all subarrays will\nleave an 'empty-container' or spare superblock on the drives. See --zero-superblock\nfor completely removing a superblock. Note that some formats depend on the subarray\nindex for generating a UUID, this command will fail if it would change the UUID of an\nactive subarray.\n\n" }, { "name": "--update-subarray=", "content": "If the device is a container and the argument to --update-subarray specifies a subar‐\nray in the container, then attempt to update the given superblock field in the subar‐\nray. See below in MISC MODE for details.\n\n" }, { "name": "-t --test", "content": "When used with --detail, the exit status of mdadm is set to reflect the status of the\ndevice. See below in MISC MODE for details.\n\n", "flag": "-t", "long": "--test" }, { "name": "-W --wait", "content": "For each md device given, wait for any resync, recovery, or reshape activity to finish\nbefore returning. mdadm will return with success if it actually waited for every de‐\nvice listed, otherwise it will return failure.\n\n", "flag": "-W", "long": "--wait" }, { "name": "--wait-clean", "content": "For each md device given, or each device in /proc/mdstat if --scan is given, arrange\nfor the array to be marked clean as soon as possible. mdadm will return with success\nif the array uses external metadata and we successfully waited. For native arrays,\nthis returns immediately as the kernel handles dirty-clean transitions at shutdown.\nNo action is taken if safe-mode handling is disabled.\n\n", "long": "--wait-clean" }, { "name": "--action=", "content": "Set the \"syncaction\" for all md devices given to one of idle, frozen, check, repair.\nSetting to idle will abort any currently running action though some actions will auto‐\nmatically restart. Setting to frozen will abort any current action and ensure no\nother action starts automatically.\n\nDetails of check and repair can be found in md(4) under SCRUBBING AND MISMATCHES.\n\n\nFor Incremental Assembly mode:\n--rebuild-map, -r\nRebuild the map file (/run/mdadm/map) that mdadm uses to help track which arrays are\ncurrently being assembled.\n\n\n--run, -R\nRun any array assembled as soon as a minimal number of devices is available, rather\nthan waiting until all expected devices are present.\n\n\n--scan, -s\nOnly meaningful with -R this will scan the map file for arrays that are being incre‐\nmentally assembled and will try to start any that are not already started. If any\nsuch array is listed in mdadm.conf as requiring an external bitmap, that bitmap will\nbe attached first.\n\n\n--fail, -f\nThis allows the hot-plug system to remove devices that have fully disappeared from the\nkernel. It will first fail and then remove the device from any array it belongs to.\nThe device name given should be a kernel device name such as \"sda\", not a name in\n/dev.\n\n" }, { "name": "--path=", "content": "Only used with --fail. The 'path' given will be recorded so that if a new device ap‐\npears at the same location it can be automatically added to the same array. This al‐\nlows the failed device to be automatically replaced by a new device without metadata\nif it appears at specified path. This option is normally only set by an udev script.\n\n\nFor Monitor mode:" }, { "name": "-m --mail", "content": "Give a mail address to send alerts to.\n\n", "flag": "-m", "long": "--mail" }, { "name": "-p --program --alert", "content": "Give a program to be run whenever an event is detected.\n\n", "flag": "-p", "long": "--alert" }, { "name": "-y --syslog", "content": "Cause all events to be reported through 'syslog'. The messages have facility of 'dae‐\nmon' and varying priorities.\n\n", "flag": "-y", "long": "--syslog" }, { "name": "-d --delay", "content": "Give a delay in seconds. mdadm polls the md arrays and then waits this many seconds\nbefore polling again. The default is 60 seconds. Since 2.6.16, there is no need to\nreduce this as the kernel alerts mdadm immediately when there is any change.\n\n", "flag": "-d", "long": "--delay" }, { "name": "-r --increment", "content": "Give a percentage increment. mdadm will generate RebuildNN events with the given per‐\ncentage increment.\n\n", "flag": "-r", "long": "--increment" }, { "name": "-f --daemonise", "content": "Tell mdadm to run as a background daemon if it decides to monitor anything. This\ncauses it to fork and run in the child, and to disconnect from the terminal. The\nprocess id of the child is written to stdout. This is useful with --scan which will\nonly continue monitoring if a mail address or alert program is found in the config\nfile.\n\n", "flag": "-f", "long": "--daemonise" }, { "name": "-i --pid-file", "content": "When mdadm is running in daemon mode, write the pid of the daemon process to the spec‐\nified file, instead of printing it on standard output.\n\n", "flag": "-i", "long": "--pid-file" }, { "name": "-1 --oneshot", "content": "Check arrays only once. This will generate NewArray events and more significantly De‐\ngradedArray and SparesMissing events. Running\nmdadm --monitor --scan -1\nfrom a cron script will ensure regular notification of any degraded arrays.\n\n", "flag": "-1", "long": "--oneshot" }, { "name": "-t --test", "content": "Generate a TestMessage alert for every array found at startup. This alert gets mailed\nand passed to the alert program. This can be used for testing that alert message do\nget through successfully.\n\n", "flag": "-t", "long": "--test" }, { "name": "--no-sharing", "content": "This inhibits the functionality for moving spares between arrays. Only one monitoring\nprocess started with --scan but without this flag is allowed, otherwise the two could\ninterfere with each other.\n\n", "long": "--no-sharing" } ] }, "ASSEMBLE MODE": { "content": "Usage: mdadm --assemble md-device options-and-component-devices...\n\nUsage: mdadm --assemble --scan md-devices-and-options...\n\nUsage: mdadm --assemble --scan options...\n\n\nThis usage assembles one or more RAID arrays from pre-existing components. For each array,\nmdadm needs to know the md device, the identity of the array, and the number of component de‐\nvices. These can be found in a number of ways.\n\nIn the first usage example (without the --scan) the first device given is the md device. In\nthe second usage example, all devices listed are treated as md devices and assembly is at‐\ntempted. In the third (where no devices are listed) all md devices that are listed in the\nconfiguration file are assembled. If no arrays are described by the configuration file, then\nany arrays that can be found on unused devices will be assembled.\n\nIf precisely one device is listed, but --scan is not given, then mdadm acts as though --scan\nwas given and identity information is extracted from the configuration file.\n\nThe identity can be given with the --uuid option, the --name option, or the --super-minor op‐\ntion, will be taken from the md-device record in the config file, or will be taken from the\nsuper block of the first component-device listed on the command line.\n\nDevices can be given on the --assemble command line or in the config file. Only devices\nwhich have an md superblock which contains the right identity will be considered for any ar‐\nray.\n\nThe config file is only used if explicitly named with --config or requested with (a possibly\nimplicit) --scan. In the latter case, the default config file is used. See mdadm.conf(5)\nfor more details.\n\nIf --scan is not given, then the config file will only be used to find the identity of md ar‐\nrays.\n\nNormally the array will be started after it is assembled. However if --scan is not given and\nnot all expected drives were listed, then the array is not started (to guard against usage\nerrors). To insist that the array be started in this case (as may work for RAID1, 4, 5, 6,\nor 10), give the --run flag.\n\nIf udev is active, mdadm does not create any entries in /dev but leaves that to udev. It\ndoes record information in /run/mdadm/map which will allow udev to choose the correct name.\n\nIf mdadm detects that udev is not configured, it will create the devices in /dev itself.\n\nIn Linux kernels prior to version 2.6.28 there were two distinct types of md devices that\ncould be created: one that could be partitioned using standard partitioning tools and one\nthat could not. Since 2.6.28 that distinction is no longer relevant as both types of devices\ncan be partitioned. mdadm will normally create the type that originally could not be parti‐\ntioned as it has a well-defined major number (9).\n\nPrior to 2.6.28, it is important that mdadm chooses the correct type of array device to use.\nThis can be controlled with the --auto option. In particular, a value of \"mdp\" or \"part\" or\n\"p\" tells mdadm to use a partitionable device rather than the default.\n\nIn the no-udev case, the value given to --auto can be suffixed by a number. This tells mdadm\nto create that number of partition devices rather than the default of 4.\n\nThe value given to --auto can also be given in the configuration file as a word starting\nauto= on the ARRAY line for the relevant array.\n\n", "subsections": [ { "name": "Auto-Assembly", "content": "When --assemble is used with --scan and no devices are listed, mdadm will first attempt to\nassemble all the arrays listed in the config file.\n\nIf no arrays are listed in the config (other than those marked ) it will look through\nthe available devices for possible arrays and will try to assemble anything that it finds.\nArrays which are tagged as belonging to the given homehost will be assembled and started nor‐\nmally. Arrays which do not obviously belong to this host are given names that are expected\nnot to conflict with anything local, and are started \"read-auto\" so that nothing is written\nto any device until the array is written to. i.e. automatic resync etc is delayed.\n\nIf mdadm finds a consistent set of devices that look like they should comprise an array, and\nif the superblock is tagged as belonging to the given home host, it will automatically choose\na device name and try to assemble the array. If the array uses version-0.90 metadata, then\nthe minor number as recorded in the superblock is used to create a name in /dev/md/ so for\nexample /dev/md/3. If the array uses version-1 metadata, then the name from the superblock\nis used to similarly create a name in /dev/md/ (the name will have any 'host' prefix stripped\nfirst).\n\nThis behaviour can be modified by the AUTO line in the mdadm.conf configuration file. This\nline can indicate that specific metadata type should, or should not, be automatically assem‐\nbled. If an array is found which is not listed in mdadm.conf and has a metadata format that\nis denied by the AUTO line, then it will not be assembled. The AUTO line can also request\nthat all arrays identified as being for this homehost should be assembled regardless of their\nmetadata type. See mdadm.conf(5) for further details.\n\nNote: Auto-assembly cannot be used for assembling and activating some arrays which are under‐\ngoing reshape. In particular as the backup-file cannot be given, any reshape which requires\na backup file to continue cannot be started by auto-assembly. An array which is growing to\nmore devices and has passed the critical section can be assembled using auto-assembly.\n\n" } ] }, "BUILD MODE": { "content": "Usage: mdadm --build md-device --chunk=X --level=Y --raid-devices=Z devices\n\n\nThis usage is similar to --create. The difference is that it creates an array without a su‐\nperblock. With these arrays there is no difference between initially creating the array and\nsubsequently assembling the array, except that hopefully there is useful data there in the\nsecond case.\n\nThe level may raid0, linear, raid1, raid10, multipath, or faulty, or one of their synonyms.\nAll devices must be listed and the array will be started once complete. It will often be ap‐\npropriate to use --assume-clean with levels raid1 or raid10.\n\n", "subsections": [] }, "CREATE MODE": { "content": "Usage: mdadm --create md-device --chunk=X --level=Y --raid-devices=Z devices\n\n\nThis usage will initialize a new md array, associate some devices with it, and activate the\narray.\n\nmd-device is a new device. This could be standard name or chosen name. For details see: DE‐\nVICE NAMES\n\nThe named device will normally not exist when mdadm --create is run, but will be created by\nudev once the array becomes active.\n\nThe max length md-device name is limited to 32 characters. Different metadata types have\nmore strict limitation (like IMSM where only 16 characters are allowed). For that reason,\nlong name could be truncated or rejected, it depends on metadata policy.\n\nAs devices are added, they are checked to see if they contain RAID superblocks or filesys‐\ntems. They are also checked to see if the variance in device size exceeds 1%.\n\nIf any discrepancy is found, the array will not automatically be run, though the presence of\na --run can override this caution.\n\nTo create a \"degraded\" array in which some devices are missing, simply give the word \"miss‐\ning\" in place of a device name. This will cause mdadm to leave the corresponding slot in the\narray empty. For a RAID4 or RAID5 array at most one slot can be \"missing\"; for a RAID6 array\nat most two slots. For a RAID1 array, only one real device needs to be given. All of the\nothers can be \"missing\".\n\nWhen creating a RAID5 array, mdadm will automatically create a degraded array with an extra\nspare drive. This is because building the spare into a degraded array is in general faster\nthan resyncing the parity on a non-degraded, but not clean, array. This feature can be over‐\nridden with the --force option.\n\nWhen creating a partition based array, using mdadm with version-1.x metadata, the partition\ntype should be set to 0xDA (non fs-data). This type of selection allows for greater preci‐\nsion since using any other [RAID auto-detect (0xFD) or a GNU/Linux partition (0x83)], might\ncreate problems in the event of array recovery through a live cdrom.\n\nA new array will normally get a randomly assigned 128bit UUID which is very likely to be\nunique. If you have a specific need, you can choose a UUID for the array by giving the\n--uuid= option. Be warned that creating two arrays with the same UUID is a recipe for disas‐\nter. Also, using --uuid= when creating a v0.90 array will silently override any --homehost=\nsetting.\n\nIf the array type supports a write-intent bitmap, and if the devices in the array exceed 100G\nis size, an internal write-intent bitmap will automatically be added unless some other option\nis explicitly requested with the --bitmap option or a different consistency policy is se‐\nlected with the --consistency-policy option. In any case, space for a bitmap will be reserved\nso that one can be added later with --grow --bitmap=internal.\n\nIf the metadata type supports it (currently only 1.x and IMSM metadata), space will be allo‐\ncated to store a bad block list. This allows a modest number of bad blocks to be recorded,\nallowing the drive to remain in service while only partially functional.\n\nWhen creating an array within a CONTAINER mdadm can be given either the list of devices to\nuse, or simply the name of the container. The former case gives control over which devices\nin the container will be used for the array. The latter case allows mdadm to automatically\nchoose which devices to use based on how much spare space is available.\n\nThe General Management options that are valid with --create are:\n\n--run insist on running the array even if some devices look like they might be in use.\n\n", "subsections": [ { "name": "--readonly", "content": "start the array in readonly mode.\n\n", "long": "--readonly" } ] }, "MANAGE MODE": { "content": "Usage: mdadm device options... devices...\n\nThis usage will allow individual devices in an array to be failed, removed or added. It is\npossible to perform multiple operations with on command. For example:\nmdadm /dev/md0 -f /dev/hda1 -r /dev/hda1 -a /dev/hda1\nwill firstly mark /dev/hda1 as faulty in /dev/md0 and will then remove it from the array and\nfinally add it back in as a spare. However, only one md array can be affected by a single\ncommand.\n\nWhen a device is added to an active array, mdadm checks to see if it has metadata on it which\nsuggests that it was recently a member of the array. If it does, it tries to \"re-add\" the\ndevice. If there have been no changes since the device was removed, or if the array has a\nwrite-intent bitmap which has recorded whatever changes there were, then the device will im‐\nmediately become a full member of the array and those differences recorded in the bitmap will\nbe resolved.\n\n", "subsections": [] }, "MISC MODE": { "content": "Usage: mdadm options ... devices ...\n\nMISC mode includes a number of distinct operations that operate on distinct devices. The op‐\nerations are:\n", "subsections": [ { "name": "--query", "content": "The device is examined to see if it is (1) an active md array, or (2) a component of\nan md array. The information discovered is reported.\n\n", "long": "--query" }, { "name": "--detail", "content": "The device should be an active md device. mdadm will display a detailed description\nof the array. --brief or --scan will cause the output to be less detailed and the\nformat to be suitable for inclusion in mdadm.conf. The exit status of mdadm will nor‐\nmally be 0 unless mdadm failed to get useful information about the device(s); however,\nif the --test option is given, then the exit status will be:\n\n0 The array is functioning normally.\n\n1 The array has at least one failed device.\n\n2 The array has multiple failed devices such that it is unusable.\n\n4 There was an error while trying to get information about the device.\n\n", "long": "--detail" }, { "name": "--detail-platform", "content": "Print detail of the platform's RAID capabilities (firmware / hardware topology). If\nthe metadata is specified with -e or --metadata= then the return status will be:\n\n0 metadata successfully enumerated its platform components on this system\n\n1 metadata is platform independent\n\n2 metadata failed to find its platform components on this system\n\n", "long": "--detail-platform" }, { "name": "--update-subarray=", "content": "If the device is a container and the argument to --update-subarray specifies a subar‐\nray in the container, then attempt to update the given superblock field in the subar‐\nray. Similar to updating an array in \"assemble\" mode, the field to update is selected\nby -U or --update= option. The supported options are name, ppl, no-ppl, bitmap and\nno-bitmap.\n\nThe name option updates the subarray name in the metadata. It must be POSIX PORTABLE\nNAME compatible and cannot be longer than 32 chars. If successes, new value will be\nrespected after next assembly.\n\nThe ppl and no-ppl options enable and disable PPL in the metadata. Currently supported\nonly for IMSM subarrays.\n\nThe bitmap and no-bitmap options enable and disable write-intent bitmap in the meta‐\ndata. Currently supported only for IMSM subarrays.\n\n" }, { "name": "--examine", "content": "The device should be a component of an md array. mdadm will read the md superblock of\nthe device and display the contents. If --brief or --scan is given, then multiple de‐\nvices that are components of the one array are grouped together and reported in a sin‐\ngle entry suitable for inclusion in mdadm.conf.\n\nHaving --scan without listing any devices will cause all devices listed in the config\nfile to be examined.\n\n\n--dump=directory\nIf the device contains RAID metadata, a file will be created in the directory and the\nmetadata will be written to it. The file will be the same size as the device and will\nhave the metadata written at the same location as it exists in the device. However,\nthe file will be \"sparse\" so that only those blocks containing metadata will be allo‐\ncated. The total space used will be small.\n\nThe filename used in the directory will be the base name of the device. Further, if\nany links appear in /dev/disk/by-id which point to the device, then hard links to the\nfile will be created in directory based on these by-id names.\n\nMultiple devices can be listed and their metadata will all be stored in the one direc‐\ntory.\n\n\n--restore=directory\nThis is the reverse of --dump. mdadm will locate a file in the directory that has a\nname appropriate for the given device and will restore metadata from it. Names that\nmatch /dev/disk/by-id names are preferred, however if two of those refer to different\nfiles, mdadm will not choose between them but will abort the operation.\n\nIf a file name is given instead of a directory then mdadm will restore from that file\nto a single device, always provided the size of the file matches that of the device,\nand the file contains valid metadata.\n\n--stop The devices should be active md arrays which will be deactivated, as long as they are\nnot currently in use.\n\n\n--run This will fully activate a partially assembled md array.\n\n", "long": "--examine" }, { "name": "--readonly", "content": "This will mark an active array as read-only, providing that it is not currently being\nused.\n\n", "long": "--readonly" }, { "name": "--readwrite", "content": "This will change a readonly array back to being read/write.\n\n\n--scan For all operations except --examine, --scan will cause the operation to be applied to\nall arrays listed in /proc/mdstat. For --examine, --scan causes all devices listed in\nthe config file to be examined.\n\n", "long": "--readwrite" }, { "name": "-b --brief", "content": "Be less verbose. This is used with --detail and --examine. Using --brief with --ver‐\nbose gives an intermediate level of verbosity.\n\n", "flag": "-b", "long": "--brief" } ] }, "MONITOR MODE": { "content": "Usage: mdadm --monitor options... devices...\n\n\nMonitor option can work in two modes:\n\n• system wide mode, follow all md devices based on /proc/mdstat,\n\n• follow only specified MD devices in command line.\n\n--scan - indicates system wide mode. Option causes the monitor to track all md devices that\nappear in /proc/mdstat. If it is not set, then at least one device must be specified.\n\nMonitor usage causes mdadm to periodically poll a number of md arrays and to report on any\nevents noticed.\n\nIn both modes, monitor will work as long as there is an active array with redundancy and it\nis defined to follow (for --scan every array is followed).\n\nAs well as reporting events, mdadm may move a spare drive from one array to another if they\nare in the same spare-group or domain and if the destination array has a failed drive but no\nspares.\n\nThe result of monitoring the arrays is the generation of events. These events are passed to\na separate program (if specified) and may be mailed to a given E-mail address.\n\nWhen passing events to a program, the program is run once for each event, and is given 2 or 3\ncommand-line arguments: the first is the name of the event (see below), the second is the\nname of the md device which is affected, and the third is the name of a related device if\nrelevant (such as a component device that has failed).\n\nIf --scan is given, then a program or an e-mail address must be specified on the command line\nor in the config file. If neither are available, then mdadm will not monitor anything. For\ndevices given directly in command line, without program or email specified, each event is re‐\nported to stdout.\n\nNote: For systems where is configured via systemd, mdmonitor(mdmonitor.service) should be\nconfigured. The service is designed to be primary solution for array monitoring, it is con‐\nfigured to work in system wide mode. It is automatically started and stopped according to\ncurrent state and types of MD arrays in system. The service may require additional configu‐\nration, like e-mail or delay. That should be done in mdadm.conf.\n\nThe different events are:\n\n\nDeviceDisappeared\nAn md array which previously was configured appears to no longer be configured.\n(syslog priority: Critical)\n\nIf mdadm was told to monitor an array which is RAID0 or Linear, then it will re‐\nport DeviceDisappeared with the extra information Wrong-Level. This is because\nRAID0 and Linear do not support the device-failed, hot-spare and resync operations\nwhich are monitored.\n\n\nRebuildStarted\nAn md array started reconstruction (e.g. recovery, resync, reshape, check, re‐\npair). (syslog priority: Warning)\n\n\nRebuildNN\nWhere NN is a two-digit number (eg. 05, 48). This indicates that the rebuild has\nreached that percentage of the total. The events are generated at a fixed incre‐\nment from 0. The increment size may be specified with a command-line option (the\ndefault is 20). (syslog priority: Warning)\n\n\nRebuildFinished\nAn md array that was rebuilding, isn't any more, either because it finished nor‐\nmally or was aborted. (syslog priority: Warning)\n\n\nFail An active component device of an array has been marked as faulty. (syslog prior‐\nity: Critical)\n\n\nFailSpare\nA spare component device which was being rebuilt to replace a faulty device has\nfailed. (syslog priority: Critical)\n\n\nSpareActive\nA spare component device which was being rebuilt to replace a faulty device has\nbeen successfully rebuilt and has been made active. (syslog priority: Info)\n\n\nNewArray\nA new md array has been detected in the /proc/mdstat file. (syslog priority:\nInfo)\n\n\nDegradedArray\nA newly noticed array appears to be degraded. This message is not generated when\nmdadm notices a drive failure which causes degradation, but only when mdadm no‐\ntices that an array is degraded when it first sees the array. (syslog priority:\nCritical)\n\n\nMoveSpare\nA spare drive has been moved from one array in a spare-group or domain to another\nto allow a failed drive to be replaced. (syslog priority: Info)\n\n\nSparesMissing\nIf mdadm has been told, via the config file, that an array should have a certain\nnumber of spare devices, and mdadm detects that it has fewer than this number when\nit first sees the array, it will report a SparesMissing message. (syslog prior‐\nity: Warning)\n\n\nTestMessage\nAn array was found at startup, and the --test flag was given. (syslog priority:\nInfo)\n\nOnly Fail, FailSpare, DegradedArray, SparesMissing and TestMessage cause Email to be sent.\nAll events cause the program to be run. The program is run with two or three arguments: the\nevent name, the array device and possibly a second device.\n\nEach event has an associated array device (e.g. /dev/md1) and possibly a second device. For\nFail, FailSpare, and SpareActive the second device is the relevant component device. For\nMoveSpare the second device is the array that the spare was moved from.\n\nFor mdadm to move spares from one array to another, the different arrays need to be labeled\nwith the same spare-group or the spares must be allowed to migrate through matching POLICY\ndomains in the configuration file. The spare-group name can be any string; it is only neces‐\nsary that different spare groups use different names.\n\nWhen mdadm detects that an array in a spare group has fewer active devices than necessary for\nthe complete array, and has no spare devices, it will look for another array in the same\nspare group that has a full complement of working drives and a spare. It will then attempt\nto remove the spare from the second array and add it to the first. If the removal succeeds\nbut the adding fails, then it is added back to the original array.\n\nIf the spare group for a degraded array is not defined, mdadm will look at the rules of spare\nmigration specified by POLICY lines in mdadm.conf and then follow similar steps as above if a\nmatching spare is found.\n\n", "subsections": [] }, "GROW MODE": { "content": "The GROW mode is used for changing the size or shape of an active array.\n\nDuring the kernel 2.6 era the following changes were added:\n\n• change the \"size\" attribute for RAID1, RAID4, RAID5 and RAID6.\n\n• increase or decrease the \"raid-devices\" attribute of RAID0, RAID1, RAID4, RAID5, and\nRAID6.\n\n• change the chunk-size and layout of RAID0, RAID4, RAID5, RAID6 and RAID10.\n\n• convert between RAID1 and RAID5, between RAID5 and RAID6, between RAID0, RAID4, and\nRAID5, and between RAID0 and RAID10 (in the near-2 mode).\n\n• add a write-intent bitmap to any array which supports these bitmaps, or remove a write-\nintent bitmap from such an array.\n\n• change the array's consistency policy.\n\nUsing GROW on containers is currently supported only for Intel's IMSM container format. The\nnumber of devices in a container can be increased - which affects all arrays in the container\n- or an array in a container can be converted between levels where those levels are supported\nby the container, and the conversion is on of those listed above.\n\n\nNotes:\n\n• Intel's native checkpointing doesn't use --backup-file option and it is transparent for\nassembly feature.\n\n• Roaming between Windows(R) and Linux systems for IMSM metadata is not supported during\ngrow process.\n\n• When growing a raid0 device, the new component disk size (or external backup size) should\nbe larger than LCM(old, new) * chunk-size * 2, where LCM() is the least common multiple\nof the old and new count of component disks, and \"* 2\" comes from the fact that mdadm re‐\nfuses to use more than half of a spare device for backup space.\n\n\nSIZE CHANGES\nNormally when an array is built the \"size\" is taken from the smallest of the drives. If all\nthe small drives in an arrays are, over time, removed and replaced with larger drives, then\nyou could have an array of large drives with only a small amount used. In this situation,\nchanging the \"size\" with \"GROW\" mode will allow the extra space to start being used. If the\nsize is increased in this way, a \"resync\" process will start to make sure the new parts of\nthe array are synchronised.\n\nNote that when an array changes size, any filesystem that may be stored in the array will not\nautomatically grow or shrink to use or vacate the space. The filesystem will need to be ex‐\nplicitly told to use the extra space after growing, or to reduce its size prior to shrinking\nthe array.\n\nAlso, the size of an array cannot be changed while it has an active bitmap. If an array has\na bitmap, it must be removed before the size can be changed. Once the change is complete a\nnew bitmap can be created.\n\n\nNote: --grow --size is not yet supported for external file bitmap.\n\n\nRAID-DEVICES CHANGES\nA RAID1 array can work with any number of devices from 1 upwards (though 1 is not very use‐\nful). There may be times which you want to increase or decrease the number of active de‐\nvices. Note that this is different to hot-add or hot-remove which changes the number of in‐\nactive devices.\n\nWhen reducing the number of devices in a RAID1 array, the slots which are to be removed from\nthe array must already be vacant. That is, the devices which were in those slots must be\nfailed and removed.\n\nWhen the number of devices is increased, any hot spares that are present will be activated\nimmediately.\n\nChanging the number of active devices in a RAID5 or RAID6 is much more effort. Every block\nin the array will need to be read and written back to a new location. From 2.6.17, the Linux\nKernel is able to increase the number of devices in a RAID5 safely, including restarting an\ninterrupted \"reshape\". From 2.6.31, the Linux Kernel is able to increase or decrease the\nnumber of devices in a RAID5 or RAID6.\n\nFrom 2.6.35, the Linux Kernel is able to convert a RAID0 in to a RAID4 or RAID5. mdadm uses\nthis functionality and the ability to add devices to a RAID4 to allow devices to be added to\na RAID0. When requested to do this, mdadm will convert the RAID0 to a RAID4, add the neces‐\nsary disks and make the reshape happen, and then convert the RAID4 back to RAID0.\n\nWhen decreasing the number of devices, the size of the array will also decrease. If there\nwas data in the array, it could get destroyed and this is not reversible, so you should\nfirstly shrink the filesystem on the array to fit within the new size. To help prevent acci‐\ndents, mdadm requires that the size of the array be decreased first with mdadm --grow --ar‐\nray-size. This is a reversible change which simply makes the end of the array inaccessible.\nThe integrity of any data can then be checked before the non-reversible reduction in the num‐\nber of devices is request.\n\nWhen relocating the first few stripes on a RAID5 or RAID6, it is not possible to keep the\ndata on disk completely consistent and crash-proof. To provide the required safety, mdadm\ndisables writes to the array while this \"critical section\" is reshaped, and takes a backup of\nthe data that is in that section. For grows, this backup may be stored in any spare devices\nthat the array has, however it can also be stored in a separate file specified with the\n--backup-file option, and is required to be specified for shrinks, RAID level changes and\nlayout changes. If this option is used, and the system does crash during the critical pe‐\nriod, the same file must be passed to --assemble to restore the backup and reassemble the ar‐\nray. When shrinking rather than growing the array, the reshape is done from the end towards\nthe beginning, so the \"critical section\" is at the end of the reshape.\n\n\nLEVEL CHANGES\nChanging the RAID level of any array happens instantaneously. However in the RAID5 to RAID6\ncase this requires a non-standard layout of the RAID6 data, and in the RAID6 to RAID5 case\nthat non-standard layout is required before the change can be accomplished. So while the\nlevel change is instant, the accompanying layout change can take quite a long time. A\n--backup-file is required. If the array is not simultaneously being grown or shrunk, so that\nthe array size will remain the same - for example, reshaping a 3-drive RAID5 into a 4-drive\nRAID6 - the backup file will be used not just for a \"critical section\" but throughout the re‐\nshape operation, as described below under LAYOUT CHANGES.\n\n\nCHUNK-SIZE AND LAYOUT CHANGES\nChanging the chunk-size or layout without also changing the number of devices as the same\ntime will involve re-writing all blocks in-place. To ensure against data loss in the case of\na crash, a --backup-file must be provided for these changes. Small sections of the array\nwill be copied to the backup file while they are being rearranged. This means that all the\ndata is copied twice, once to the backup and once to the new layout on the array, so this\ntype of reshape will go very slowly.\n\nIf the reshape is interrupted for any reason, this backup file must be made available to\nmdadm --assemble so the array can be reassembled. Consequently, the file cannot be stored on\nthe device being reshaped.\n\n\n\nBITMAP CHANGES\nA write-intent bitmap can be added to, or removed from, an active array. Either internal\nbitmaps, or bitmaps stored in a separate file, can be added. Note that if you add a bitmap\nstored in a file which is in a filesystem that is on the RAID array being affected, the sys‐\ntem will deadlock. The bitmap must be on a separate filesystem.\n\n\nCONSISTENCY POLICY CHANGES\nThe consistency policy of an active array can be changed by using the --consistency-policy\noption in Grow mode. Currently this works only for the ppl and resync policies and allows one\nto enable or disable the RAID5 Partial Parity Log (PPL).\n\n", "subsections": [] }, "INCREMENTAL MODE": { "content": "Usage: mdadm --incremental [--run] [--quiet] component-device [optional-aliases-for-device]\n\nUsage: mdadm --incremental --fail component-device\n\nUsage: mdadm --incremental --rebuild-map\n\nUsage: mdadm --incremental --run --scan\n\n\nThis mode is designed to be used in conjunction with a device discovery system. As devices\nare found in a system, they can be passed to mdadm --incremental to be conditionally added to\nan appropriate array.\n\nConversely, it can also be used with the --fail flag to do just the opposite and find what‐\never array a particular device is part of and remove the device from that array.\n\nIf the device passed is a CONTAINER device created by a previous call to mdadm, then rather\nthan trying to add that device to an array, all the arrays described by the metadata of the\ncontainer will be started.\n\nmdadm performs a number of tests to determine if the device is part of an array, and which\narray it should be part of. If an appropriate array is found, or can be created, mdadm adds\nthe device to the array and conditionally starts the array.\n\nNote that mdadm will normally only add devices to an array which were previously working (ac‐\ntive or spare) parts of that array. The support for automatic inclusion of a new drive as a\nspare in some array requires a configuration through POLICY in config file.\n\nThe tests that mdadm makes are as follow:\n\n+ Is the device permitted by mdadm.conf? That is, is it listed in a DEVICES line in\nthat file. If DEVICES is absent then the default it to allow any device. Similarly\nif DEVICES contains the special word partitions then any device is allowed. Otherwise\nthe device name given to mdadm, or one of the aliases given, or an alias found in the\nfilesystem, must match one of the names or patterns in a DEVICES line.\n\nThis is the only context where the aliases are used. They are usually provided by a\nudev rules mentioning $env{DEVLINKS}.\n\n\n+ Does the device have a valid md superblock? If a specific metadata version is re‐\nquested with --metadata or -e then only that style of metadata is accepted, otherwise\nmdadm finds any known version of metadata. If no md metadata is found, the device may\nbe still added to an array as a spare if POLICY allows.\n\n\n\nmdadm keeps a list of arrays that it has partially assembled in /run/mdadm/map. If no array\nexists which matches the metadata on the new device, mdadm must choose a device name and unit\nnumber. It does this based on any name given in mdadm.conf or any name information stored in\nthe metadata. If this name suggests a unit number, that number will be used, otherwise a\nfree unit number will be chosen. Normally mdadm will prefer to create a partitionable array,\nhowever if the CREATE line in mdadm.conf suggests that a non-partitionable array is pre‐\nferred, that will be honoured.\n\nIf the array is not found in the config file and its metadata does not identify it as belong‐\ning to the \"homehost\", then mdadm will choose a name for the array which is certain not to\nconflict with any array which does belong to this host. It does this be adding an underscore\nand a small number to the name preferred by the metadata.\n\nOnce an appropriate array is found or created and the device is added, mdadm must decide if\nthe array is ready to be started. It will normally compare the number of available (non-\nspare) devices to the number of devices that the metadata suggests need to be active. If\nthere are at least that many, the array will be started. This means that if any devices are\nmissing the array will not be restarted.\n\nAs an alternative, --run may be passed to mdadm in which case the array will be run as soon\nas there are enough devices present for the data to be accessible. For a RAID1, that means\none device will start the array. For a clean RAID5, the array will be started as soon as all\nbut one drive is present.\n\nNote that neither of these approaches is really ideal. If it can be known that all device\ndiscovery has completed, then\nmdadm -IRs\ncan be run which will try to start all arrays that are being incrementally assembled. They\nare started in \"read-auto\" mode in which they are read-only until the first write request.\nThis means that no metadata updates are made and no attempt at resync or recovery happens.\nFurther devices that are found before the first write can still be added safely.\n\n", "subsections": [] }, "ENVIRONMENT": { "content": "This section describes environment variables that affect how mdadm operates.\n\n\nMDADMNOMDMON\nSetting this value to 1 will prevent mdadm from automatically launching mdmon. This\nvariable is intended primarily for debugging mdadm/mdmon.\n\n\nMDADMNOUDEV\nNormally, mdadm does not create any device nodes in /dev, but leaves that task to\nudev. If udev appears not to be configured, or if this environment variable is set to\n'1', the mdadm will create and devices that are needed.\n\n\nMDADMNOSYSTEMCTL\nIf mdadm detects that systemd is in use it will normally request systemd to start var‐\nious background tasks (particularly mdmon) rather than forking and running them in the\nbackground. This can be suppressed by setting MDADMNOSYSTEMCTL=1.\n\n\nIMSMNOPLATFORM\nA key value of IMSM metadata is that it allows interoperability with boot ROMs on In‐\ntel platforms, and with other major operating systems. Consequently, mdadm will only\nallow an IMSM array to be created or modified if detects that it is running on an In‐\ntel platform which supports IMSM, and supports the particular configuration of IMSM\nthat is being requested (some functionality requires newer OROM support).\n\nThese checks can be suppressed by setting IMSMNOPLATFORM=1 in the environment. This\ncan be useful for testing or for disaster recovery. You should be aware that interop‐\nerability may be compromised by setting this value.\n\nThese change can also be suppressed by adding mdadm.imsm.test=1 to the kernel command\nline. This makes it easy to test IMSM code in a virtual machine that doesn't have IMSM\nvirtual hardware.\n\n\nMDADMGROWALLOWOLD\nIf an array is stopped while it is performing a reshape and that reshape was making\nuse of a backup file, then when the array is re-assembled mdadm will sometimes com‐\nplain that the backup file is too old. If this happens and you are certain it is the\nright backup file, you can over-ride this check by setting MDADMGROWALLOWOLD=1 in\nthe environment.\n\n\nMDADMCONFAUTO\nAny string given in this variable is added to the start of the AUTO line in the config\nfile, or treated as the whole AUTO line if none is given. It can be used to disable\ncertain metadata types when mdadm is called from a boot script. For example\nexport MDADMCONFAUTO='-ddf -imsm'\nwill make sure that mdadm does not automatically assemble any DDF or IMSM arrays that\nare found. This can be useful on systems configured to manage such arrays with dm‐\nraid.\n\n\n", "subsections": [] }, "EXAMPLES": { "content": "mdadm --query /dev/name-of-device\nThis will find out if a given device is a RAID array, or is part of one, and will provide\nbrief information about the device.\n\nmdadm --assemble --scan\nThis will assemble and start all arrays listed in the standard config file. This command\nwill typically go in a system startup file.\n\nmdadm --stop --scan\nThis will shut down all arrays that can be shut down (i.e. are not currently in use). This\nwill typically go in a system shutdown script.\n\nmdadm --follow --scan --delay=120\nIf (and only if) there is an Email address or program given in the standard config file, then\nmonitor the status of all arrays listed in that file by polling them ever 2 minutes.\n\nmdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/hd[ac]1\nCreate /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.\n\necho 'DEVICE /dev/hd*[0-9] /dev/sd*[0-9]' > mdadm.conf\nmdadm --detail --scan >> mdadm.conf\nThis will create a prototype config file that describes currently active arrays that are\nknown to be made from partitions of IDE or SCSI drives. This file should be reviewed before\nbeing used as it may contain unwanted detail.\n\necho 'DEVICE /dev/hd[a-z] /dev/sd*[a-z]' > mdadm.conf\nmdadm --examine --scan --config=mdadm.conf >> mdadm.conf\nThis will find arrays which could be assembled from existing IDE and SCSI whole drives (not\npartitions), and store the information in the format of a config file. This file is very\nlikely to contain unwanted detail, particularly the devices= entries. It should be reviewed\nand edited before being used as an actual config file.\n\nmdadm --examine --brief --scan --config=partitions\nmdadm -Ebsc partitions\nCreate a list of devices by reading /proc/partitions, scan these for RAID superblocks, and\nprintout a brief listing of all that were found.\n\nmdadm -Ac partitions -m 0 /dev/md0\nScan all partitions and devices listed in /proc/partitions and assemble /dev/md0 out of all\nsuch devices with a RAID superblock with a minor number of 0.\n\nmdadm --monitor --scan --daemonise > /run/mdadm/mon.pid\nIf config file contains a mail address or alert program, run mdadm in the background in moni‐\ntor mode monitoring all md devices. Also write pid of mdadm daemon to /run/mdadm/mon.pid.\n\nmdadm -Iq /dev/somedevice\nTry to incorporate newly discovered device into some array as appropriate.\n\nmdadm --incremental --rebuild-map --run --scan\nRebuild the array map from any current arrays, and then start any that can be started.\n\nmdadm /dev/md4 --fail detached --remove detached\nAny devices which are components of /dev/md4 will be marked as faulty and then remove from\nthe array.\n\nmdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4\nThe array /dev/md4 which is currently a RAID5 array will be converted to RAID6. There should\nnormally already be a spare drive attached to the array as a RAID6 needs one more drive than\na matching RAID5.\n\nmdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]\nCreate a DDF array over 6 devices.\n\nmdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf\nCreate a RAID5 array over any 3 devices in the given DDF set. Use only 30 gigabytes of each\ndevice.\n\nmdadm -A /dev/md/ddf1 /dev/sd[a-f]\nAssemble a pre-exist ddf array.\n\nmdadm -I /dev/md/ddf1\nAssemble all arrays contained in the ddf array, assigning names as appropriate.\n\nmdadm --create --help\nProvide help about the Create mode.\n\nmdadm --config --help\nProvide help about the format of the config file.\n\nmdadm --help\nProvide general help.\n\n", "subsections": [] }, "FILES": { "content": "", "subsections": [ { "name": "/proc/mdstat", "content": "If you're using the /proc filesystem, /proc/mdstat lists all active md devices with informa‐\ntion about them. mdadm uses this to find arrays when --scan is given in Misc mode, and to\nmonitor array reconstruction on Monitor mode.\n\n" }, { "name": "/etc/mdadm.conf (or /etc/mdadm/mdadm.conf)", "content": "Default config file. See mdadm.conf(5) for more details.\n\n" }, { "name": "/etc/mdadm.conf.d (or /etc/mdadm/mdadm.conf.d)", "content": "Default directory containing configuration files. See mdadm.conf(5) for more details.\n\n" }, { "name": "/run/mdadm/map", "content": "When --incremental mode is used, this file gets a list of arrays currently being created.\n\n" } ] }, "POSIX PORTABLE NAME": { "content": "A valid name can only consist of characters \"A-Za-z0-9.-\". The name cannot start with a\nleading \"-\" and cannot exceed 255 chars.\n\n", "subsections": [] }, "DEVICE NAMES": { "content": "mdadm understand two sorts of names for array devices.\n\nThe first is the so-called 'standard' format name, which matches the names used by the kernel\nand which appear in /proc/mdstat.\n\nThe second sort can be freely chosen, but must reside in /dev/md/. When giving a device name\nto mdadm to create or assemble an array, either full path name such as /dev/md0 or\n/dev/md/home can be given, or just the suffix of the second sort of name, such as home can be\ngiven.\n\nIn every style, raw name must be compatible with POSIX PORTABLE NAME and has to be no longer\nthan 32 chars.\n\nWhen mdadm chooses device names during auto-assembly or incremental assembly, it will some‐\ntimes add a small sequence number to the end of the name to avoid conflicted between multiple\narrays that have the same name. If mdadm can reasonably determine that the array really is\nmeant for this host, either by a hostname in the metadata, or by the presence of the array in\nmdadm.conf, then it will leave off the suffix if possible. Also if the homehost is specified\nas mdadm will only use a suffix if a different array of the same name already exists\nor is listed in the config file.\n\nThe standard names for non-partitioned arrays (the only sort of md array available in 2.4 and\nearlier) are of the form\n\n/dev/mdNN\n\nwhere NN is a number. The standard names for partitionable arrays (as available from 2.6 on‐\nwards) are of the form:\n\n/dev/mddNN\n\nPartition numbers should be indicated by adding \"pMM\" to these, thus \"/dev/md/d1p2\".\n\nFrom kernel version 2.6.28 the \"non-partitioned array\" can actually be partitioned. So the\n\"mddNN\" names are no longer needed, and partitions such as \"/dev/mdNNpXX\" are possible.\n\nFrom kernel version 2.6.29 standard names can be non-numeric following the form:\n\n/dev/mdXXX\n\nwhere XXX is any string. These names are supported by mdadm since version 3.3 provided they\nare enabled in mdadm.conf.\n\n", "subsections": [] }, "NOTE": { "content": "mdadm was previously known as mdctl.\n\n", "subsections": [] }, "SEE ALSO": { "content": "For further information on mdadm usage, MD and the various levels of RAID, see:\n\nhttps://raid.wiki.kernel.org/\n\n(based upon Jakob Østergaard's Software-RAID.HOWTO)\n\nThe latest version of mdadm should always be available from\n\nhttps://www.kernel.org/pub/linux/utils/raid/mdadm/\n\nRelated man pages:\n\nmdmon(8), mdadm.conf(5), md(4).\n\nv4.3 MDADM(8)", "subsections": [] } }, "summary": "mdadm - manage MD devices aka Linux Software RAID", "flags": [ { "flag": "-A", "long": "--assemble", "arg": null, "description": "Assemble a pre-existing array." }, { "flag": "-B", "long": "--build", "arg": null, "description": "Build a legacy array without superblocks." }, { "flag": "-C", "long": "--create", "arg": null, "description": "Create a new array." }, { "flag": "-F", "long": "--monitor", "arg": null, "description": "Select Monitor mode." }, { "flag": "-G", "long": "--grow", "arg": null, "description": "Change the size or shape of an active array." }, { "flag": "-I", "long": "--incremental", "arg": null, "description": "Add/remove a single device to/from an appropriate array, and possibly start the array." }, { "flag": "", "long": "--auto-detect", "arg": null, "description": "Request that the kernel starts any auto-detected arrays. This can only work if md is compiled into the kernel — not if it is a module. Arrays can be auto-detected by the kernel if all the components are in primary MS-DOS partitions with partition type FD, and all use v0.90 metadata. In-kernel autodetect is not recommended for new installa‐ tions. Using mdadm to detect and assemble arrays — possibly in an initrd — is sub‐ stantially more flexible and should be preferred. If a device is given before any options, or if the first option is one of --add, --re-add, --add-spare, --fail, --remove, or --replace, then the MANAGE mode is assumed. Anything other than these will cause the Misc mode to be assumed. Options that are not mode-specific are:" }, { "flag": "-h", "long": "--help", "arg": null, "description": "Display a general help message or, after one of the above options, a mode-specific help message." }, { "flag": "", "long": "--help-options", "arg": null, "description": "Display more detailed help about command-line parsing and some commonly used options." }, { "flag": "-V", "long": "--version", "arg": null, "description": "Print version information for mdadm." }, { "flag": "-v", "long": "--verbose", "arg": null, "description": "Be more verbose about what is happening. This can be used twice to be extra-verbose. The extra verbosity currently only affects --detail --scan and --examine --scan." }, { "flag": "-q", "long": "--quiet", "arg": null, "description": "Avoid printing purely informative messages. With this, mdadm will be silent unless there is something really important to report." }, { "flag": "-f", "long": "--force", "arg": null, "description": "Be more forceful about certain operations. See the various modes for the exact mean‐ ing of this option in different contexts." }, { "flag": "-c", "long": null, "arg": null, "description": "Specify the config file or directory. If not specified, the default config file and default conf.d directory will be used. See mdadm.conf(5) for more details. If the config file given is partitions then nothing will be read, but mdadm will act as though the config file contained exactly DEVICE partitions containers and will read /proc/partitions to find a list of devices to scan, and /proc/mdstat to find a list of containers to examine. If the word none is given for the config file, then mdadm will act as though the config file were empty. If the name given is of a directory, then mdadm will collect all the files contained in the directory with a name ending in .conf, sort them lexically, and process all of those files as config files." }, { "flag": "-s", "long": "--scan", "arg": null, "description": "Scan config file or /proc/mdstat for missing information. In general, this option gives mdadm permission to get any missing information (like component devices, array devices, array identities, and alert destination) from the configuration file (see previous option); one exception is MISC mode when using --detail or --stop, in which case --scan says to get a list of array devices from /proc/mdstat." }, { "flag": "-e", "long": null, "arg": null, "description": "Declare the style of RAID metadata (superblock) to be used. The default is 1.2 for --create, and to guess for other operations. The default can be overridden by setting the metadata value for the CREATE keyword in mdadm.conf. Options are: 0, 0.90 Use the original 0.90 format superblock. This format limits arrays to 28 com‐ ponent devices and limits component devices of levels 1 and greater to 2 ter‐ abytes. It is also possible for there to be confusion about whether the su‐ perblock applies to a whole device or just the last partition, if that parti‐ tion starts on a 64K boundary. 1, 1.0, 1.1, 1.2 default Use the new version-1 format superblock. This has fewer restrictions. It can easily be moved between hosts with different endian-ness, and a recovery opera‐ tion can be checkpointed and restarted. The different sub-versions store the superblock at different locations on the device, either at the end (for 1.0), at the start (for 1.1) or 4K from the start (for 1.2). \"1\" is equivalent to \"1.2\" (the commonly preferred 1.x format). \"default\" is equivalent to \"1.2\". ddf Use the \"Industry Standard\" DDF (Disk Data Format) format defined by SNIA. When creating a DDF array a CONTAINER will be created, and normal arrays can be created in that container. imsm Use the Intel(R) Matrix Storage Manager metadata format. This creates a CON‐ TAINER which is managed in a similar manner to DDF, and is supported by an op‐ tion-rom on some platforms: https://www.intel.com/content/www/us/en/support/products/122484" }, { "flag": "", "long": null, "arg": null, "description": "This will override any HOMEHOST setting in the config file and provides the identity of the host which should be considered the home for any arrays. When creating an array, the homehost will be recorded in the metadata. For version-1 superblocks, it will be prefixed to the array name. For version-0.90 superblocks, part of the SHA1 hash of the hostname will be stored in the latter half of the UUID. When reporting information about an array, any array which is tagged for the given homehost will be reported as such. When using Auto-Assemble, only arrays tagged for the given homehost will be allowed to use 'local' names (i.e. not ending in '' followed by a digit string). See below un‐ der Auto-Assembly. The special name \"any\" can be used as a wild card. If an array is created with --homehost=any then the name \"any\" will be stored in the array and it can be assembled in the same way on any host. If an array is assembled with this option, then the homehost recorded on the array will be ignored." }, { "flag": "", "long": null, "arg": null, "description": "When mdadm needs to print the name for a device it normally finds the name in /dev which refers to the device and is the shortest. When a path component is given with --prefer mdadm will prefer a longer name if it contains that component. For example --prefer=by-uuid will prefer a name in a subdirectory of /dev called by-uuid. This functionality is currently only provided by --detail and --monitor." }, { "flag": "", "long": null, "arg": null, "description": "specifies the cluster name for the md device. The md device can be assembled only on the cluster which matches the name specified. If this option is not provided, mdadm tries to detect the cluster name automatically. For create, build, or grow:" }, { "flag": "-n", "long": null, "arg": null, "description": "Specify the number of active devices in the array. This, plus the number of spare de‐ vices (see below) must equal the number of component-devices (including \"missing\" de‐ vices) that are listed on the command line for --create. Setting a value of 1 is probably a mistake and so requires that --force be specified first. A value of 1 will then be allowed for linear, multipath, RAID0 and RAID1. It is never allowed for RAID4, RAID5 or RAID6. This number can only be changed using --grow for RAID1, RAID4, RAID5 and RAID6 arrays, and only on kernels which provide the necessary support." }, { "flag": "-x", "long": null, "arg": null, "description": "Specify the number of spare (eXtra) devices in the initial array. Spares can also be added and removed later. The number of component devices listed on the command line must equal the number of RAID devices plus the number of spare devices." }, { "flag": "-z", "long": null, "arg": null, "description": "Amount (in Kilobytes) of space to use from each drive in RAID levels 1/4/5/6/10 and for RAID 0 on external metadata. This must be a multiple of the chunk size, and must leave about 128Kb of space at the end of the drive for the RAID superblock. When spec‐ ified as ¸max¸ (as it often is) the smallest drive (or partition) sets the size. In that case, a warning will follow if the drives, as a group, have sizes that differ by more than one percent. A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes, Megabytes, Giga‐ bytes or Terabytes respectively. Sometimes a replacement drive can be a little smaller than the original drives though this should be minimised by IDEMA standards. Such a replacement drive will be re‐ jected by md. To guard against this it can be useful to set the initial size slightly smaller than the smaller device with the aim that it will still be larger than any re‐ placement. This option can be used with --create for determining the initial size of an array. For external metadata, it can be used on a volume, but not on a container itself. Setting the initial size of RAID 0 array is only valid for external metadata. This value can be set with --grow for RAID level 1/4/5/6/10 though DDF arrays may not be able to support this. RAID 0 array size cannot be changed. If the array was cre‐ ated with a size smaller than the currently active drives, the extra space can be ac‐ cessed using --grow. The size can be given as max which means to choose the largest size that fits on all current drives. Before reducing the size of the array (with --grow --size=) you should make sure that space isn't needed. If the device holds a filesystem, you would need to resize the filesystem to use less space. After reducing the array size you should check that the data stored in the device is still available. If the device holds a filesystem, then an 'fsck' of the filesystem is a minimum requirement. If there are problems the array can be made bigger again with no loss with another --grow --size= command." }, { "flag": "-Z", "long": null, "arg": null, "description": "This is only meaningful with --grow and its effect is not persistent: when the array is stopped and restarted the default array size will be restored. Setting the array-size causes the array to appear smaller to programs that access the data. This is particularly needed before reshaping an array so that it will be smaller. As the reshape is not reversible, but setting the size with --array-size is, it is required that the array size is reduced as appropriate before the number of de‐ vices in the array is reduced. Before reducing the size of the array you should make sure that space isn't needed. If the device holds a filesystem, you would need to resize the filesystem to use less space. After reducing the array size you should check that the data stored in the device is still available. If the device holds a filesystem, then an 'fsck' of the filesystem is a minimum requirement. If there are problems the array can be made bigger again with no loss with another --grow --array-size= command. A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes, Megabytes, Giga‐ bytes or Terabytes respectively. A value of max restores the apparent size of the ar‐ ray to be whatever the real amount of available space is. Clustered arrays do not support this parameter yet." }, { "flag": "-c", "long": null, "arg": null, "description": "Specify chunk size in kilobytes. The default when creating an array is 512KB. To en‐ sure compatibility with earlier versions, the default when building an array with no persistent metadata is 64KB. This is only meaningful for RAID0, RAID4, RAID5, RAID6, and RAID10. RAID4, RAID5, RAID6, and RAID10 require the chunk size to be a power of 2, with mini‐ mal chunk size being 4KB. A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes, Megabytes, Giga‐ bytes or Terabytes respectively." }, { "flag": "", "long": null, "arg": null, "description": "Specify the rounding factor for a Linear array. The size of each component will be rounded down to a multiple of this size. This is a synonym for --chunk but highlights the different meaning for Linear as compared to other RAID levels. The default is 64K if a kernel earlier than 2.6.16 is in use, and is 0K (i.e. no rounding) in later ker‐ nels." }, { "flag": "-l", "long": null, "arg": null, "description": "Set RAID level. When used with --create, options are: linear, raid0, 0, stripe, raid1, 1, mirror, raid4, 4, raid5, 5, raid6, 6, raid10, 10, multipath, mp, faulty, container. Obviously some of these are synonymous. When a CONTAINER metadata type is requested, only the container level is permitted, and it does not need to be explicitly given. When used with --build, only linear, stripe, raid0, 0, raid1, multipath, mp, and faulty are valid. Can be used with --grow to change the RAID level in some cases. See LEVEL CHANGES be‐ low." }, { "flag": "-p", "long": null, "arg": null, "description": "This option configures the fine details of data layout for RAID5, RAID6, and RAID10 arrays, and controls the failure modes for faulty. It can also be used for working around a kernel bug with RAID0, but generally doesn't need to be used explicitly. The layout of the RAID5 parity block can be one of left-asymmetric, left-symmetric, right-asymmetric, right-symmetric, la, ra, ls, rs. The default is left-symmetric. It is also possible to cause RAID5 to use a RAID4-like layout by choosing par‐ ity-first, or parity-last. Finally for RAID5 there are DDF-compatible layouts, ddf-zero-restart, ddf-N-restart, and ddf-N-continue. These same layouts are available for RAID6. There are also 4 layouts that will pro‐ vide an intermediate stage for converting between RAID5 and RAID6. These provide a layout which is identical to the corresponding RAID5 layout on the first N-1 devices, and has the 'Q' syndrome (the second 'parity' block used by RAID6) on the last device. These layouts are: left-symmetric-6, right-symmetric-6, left-asymmetric-6, right-asym‐ metric-6, and parity-first-6. When setting the failure mode for level faulty, the options are: write-transient, wt, read-transient, rt, write-persistent, wp, read-persistent, rp, write-all, read-fix‐ able, rf, clear, flush, none. Each failure mode can be followed by a number, which is used as a period between fault generation. Without a number, the fault is generated once on the first relevant re‐ quest. With a number, the fault will be generated after that many requests, and will continue to be generated every time the period elapses. Multiple failure modes can be current simultaneously by using the --grow option to set subsequent failure modes. \"clear\" or \"none\" will remove any pending or periodic failure modes, and \"flush\" will clear any persistent faults. The layout options for RAID10 are one of 'n', 'o' or 'f' followed by a small number signifying the number of copies of each datablock. The default is 'n2'. The sup‐ ported options are: 'n' signals 'near' copies. Multiple copies of one data block are at similar offsets in different devices. 'o' signals 'offset' copies. Rather than the chunks being duplicated within a stripe, whole stripes are duplicated but are rotated by one device so duplicate blocks are on different devices. Thus subsequent copies of a block are in the next drive, and are one chunk further down. 'f' signals 'far' copies (multiple copies have very different offsets). See md(4) for more detail about 'near', 'offset', and 'far'. As for the number of copies of each data block, 2 is normal, 3 can be useful. This number can be at most equal to the number of devices in the array. It does not need to divide evenly into that number (e.g. it is perfectly legal to have an 'n2' layout for an array with an odd number of devices). A bug introduced in Linux 3.14 means that RAID0 arrays with devices of differing sizes started using a different layout. This could lead to data corruption. Since Linux 5.4 (and various stable releases that received backports), the kernel will not accept such an array unless a layout is explicitly set. It can be set to 'original' or 'al‐ ternate'. When creating a new array, mdadm will select 'original' by default, so the layout does not normally need to be set. An array created for either 'original' or 'alternate' will not be recognized by an (unpatched) kernel prior to 5.4. To create a RAID0 array with devices of differing sizes that can be used on an older kernel, you can set the layout to 'dangerous'. This will use whichever layout the running kernel supports, so the data on the array may become corrupt when changing kernel from pre-3.14 to a later kernel. When an array is converted between RAID5 and RAID6 an intermediate RAID6 layout is used in which the second parity block (Q) is always on the last device. To convert a RAID5 to RAID6 and leave it in this new layout (which does not require re-striping) use --layout=preserve. This will try to avoid any restriping. The converse of this is --layout=normalise which will change a non-standard RAID6 lay‐ out into a more standard arrangement." }, { "flag": "", "long": null, "arg": null, "description": "same as --layout (thus explaining the p of -p)." }, { "flag": "-b", "long": null, "arg": null, "description": "Specify a file to store a write-intent bitmap in. The file should not exist unless --force is also given. The same file should be provided when assembling the array. If the word internal is given, then the bitmap is stored with the metadata on the ar‐ ray, and so is replicated on all devices. If the word none is given with --grow mode, then any bitmap that is present is removed. If the word clustered is given, the array is created for a clustered environment. One bitmap is created for each node as defined by the --nodes parameter and are stored internally. To help catch typing errors, the filename must contain at least one slash ('/') if it is a real file (not 'internal' or 'none'). Note: external bitmaps are only known to work on ext2 and ext3. Storing bitmap files on other filesystems may result in serious problems. When creating an array on devices which are 100G or larger, mdadm automatically adds an internal bitmap as it will usually be beneficial. This can be suppressed with --bitmap=none or by selecting a different consistency policy with --consistency-pol‐ icy." }, { "flag": "", "long": null, "arg": null, "description": "Set the chunk size of the bitmap. Each bit corresponds to that many Kilobytes of storage. When using a file-based bitmap, the default is to use the smallest size that is at least 4 and requires no more than 2^21 chunks. When using an internal bitmap, the chunk size defaults to 64Meg, or larger if necessary to fit the bitmap into the available space. A suffix of 'K', 'M', 'G' or 'T' can be given to indicate Kilobytes, Megabytes, Giga‐ bytes or Terabytes respectively." }, { "flag": "-W", "long": "--write-mostly", "arg": null, "description": "subsequent devices listed in a --build, --create, or --add command will be flagged as 'write-mostly'. This is valid for RAID1 only and means that the 'md' driver will avoid reading from these devices if at all possible. This can be useful if mirroring over a slow link." }, { "flag": "", "long": null, "arg": null, "description": "Specify that write-behind mode should be enabled (valid for RAID1 only). If an argu‐ ment is specified, it will set the maximum number of outstanding writes allowed. The default value is 256. A write-intent bitmap is required in order to use write-behind mode, and write-behind is only attempted on drives marked as write-mostly." }, { "flag": "", "long": "--failfast", "arg": null, "description": "subsequent devices listed in a --create or --add command will be flagged as 'fail‐ fast'. This is valid for RAID1 and RAID10 only. IO requests to these devices will be encouraged to fail quickly rather than cause long delays due to error handling. Also no attempt is made to repair a read error on these devices. If an array becomes degraded so that the 'failfast' device is the only usable device, the 'failfast' flag will then be ignored and extended delays will be preferred to com‐ plete failure. The 'failfast' flag is appropriate for storage arrays which have a low probability of true failure, but which may sometimes cause unacceptable delays due to internal main‐ tenance functions." }, { "flag": "", "long": "--assume-clean", "arg": null, "description": "Tell mdadm that the array pre-existed and is known to be clean. It can be useful when trying to recover from a major failure as you can be sure that no data will be af‐ fected unless you actually write to the array. It can also be used when creating a RAID1 or RAID10 if you want to avoid the initial resync, however this practice — while normally safe — is not recommended. Use this only if you really know what you are do‐ ing. When the devices that will be part of a new array were filled with zeros before cre‐ ation the operator knows the array is actually clean. If that is the case, such as af‐ ter running badblocks, this argument can be used to tell mdadm the facts the operator knows. When an array is resized to a larger size with --grow --size= the new space is nor‐ mally resynced in that same way that the whole array is resynced at creation. From Linux version 3.0, --assume-clean can be used with that command to avoid the automatic resync." }, { "flag": "", "long": "--write-zeroes", "arg": null, "description": "When creating an array, send write zeroes requests to all the block devices. This should zero the data area on all disks such that the initial sync is not necessary and, if successfull, will behave as if --assume-clean was specified. This is intended for use with devices that have hardware offload for zeroing, but de‐ spite this zeroing can still take several minutes for large disks. Thus a message is printed before and after zeroing and each disk is zeroed in parallel with the others. This is only meaningful with --create." }, { "flag": "", "long": null, "arg": null, "description": "This is needed when --grow is used to increase the number of raid devices in a RAID5 or RAID6 if there are no spare devices available, or to shrink, change RAID level or layout. See the GROW MODE section below on RAID-DEVICES CHANGES. The file must be stored on a separate device, not on the RAID array being reshaped." }, { "flag": "", "long": null, "arg": null, "description": "Arrays with 1.x metadata can leave a gap between the start of the device and the start of array data. This gap can be used for various metadata. The start of data is known as the data-offset. Normally an appropriate data offset is computed automatically. However it can be useful to set it explicitly such as when re-creating an array which was originally created using a different version of mdadm which computed a different offset. Setting the offset explicitly over-rides the default. The value given is in Kilobytes unless a suffix of 'K', 'M', 'G' or 'T' is used to explicitly indicate Kilobytes, Megabytes, Gigabytes or Terabytes respectively. Since Linux 3.4, --data-offset can also be used with --grow for some RAID levels (ini‐ tially on RAID10). This allows the data-offset to be changed as part of the reshape process. When the data offset is changed, no backup file is required as the differ‐ ence in offsets is used to provide the same functionality. When the new offset is earlier than the old offset, the number of devices in the array cannot shrink. When it is after the old offset, the number of devices in the array cannot increase. When creating an array, --data-offset can be specified as variable. In the case each member device is expected to have an offset appended to the name, separated by a colon. This makes it possible to recreate exactly an array which has varying data offsets (as can happen when different versions of mdadm are used to add different de‐ vices)." }, { "flag": "", "long": "--continue", "arg": null, "description": "This option is complementary to the --freeze-reshape option for assembly. It is needed when --grow operation is interrupted and it is not restarted automatically due to --freeze-reshape usage during array assembly. This option is used together with -G , ( --grow ) command and device for a pending reshape to be continued. All parameters required for reshape continuation will be read from array metadata. If initial --grow command had required --backup-file= option to be set, continuation option will require to have exactly the same backup file given as well. Any other parameter passed together with --continue option will be ignored." }, { "flag": "-N", "long": null, "arg": null, "description": "Set a name for the array. It must be POSIX PORTABLE NAME compatible and cannot be longer than 32 chars. This is effective when creating an array with a v1 metadata, or an external array. If name is needed but not specified, it is taken from the basename of the device that is being created. See DEVICE NAMES" }, { "flag": "-R", "long": "--run", "arg": null, "description": "Insist that mdadm run the array, even if some of the components appear to be active in another array or filesystem. Normally mdadm will ask for confirmation before includ‐ ing such components in an array. This option causes that question to be suppressed." }, { "flag": "-f", "long": "--force", "arg": null, "description": "Insist that mdadm accept the geometry and layout specified without question. Normally mdadm will not allow the creation of an array with only one device, and will try to create a RAID5 array with one missing drive (as this makes the initial resync work faster). With --force, mdadm will not try to be so clever." }, { "flag": "-o", "long": "--readonly", "arg": null, "description": "Start the array read only rather than read-write as normal. No writes will be allowed to the array, and no resync, recovery, or reshape will be started. It works with Cre‐ ate, Assemble, Manage and Misc mode." }, { "flag": "-a", "long": null, "arg": null, "description": "Instruct mdadm how to create the device file if needed, possibly allocating an unused minor number. \"md\" causes a non-partitionable array to be used (though since Linux 2.6.28, these array devices are in fact partitionable). \"mdp\", \"part\" or \"p\" causes a partitionable array (2.6 and later) to be used. \"yes\" requires the named md device to have a 'standard' format, and the type and minor number will be determined from this. With mdadm 3.0, device creation is normally left up to udev so this option is unlikely to be needed. See DEVICE NAMES below. The argument can also come immediately after \"-a\". e.g. \"-ap\". If --auto is not given on the command line or in the config file, then the default will be --auto=yes. If --scan is also given, then any auto= entries in the config file will override the --auto instruction given on the command line. For partitionable arrays, mdadm will create the device file for the whole array and for the first 4 partitions. A different number of partitions can be specified at the end of this option (e.g. --auto=p7). If the device name ends with a digit, the par‐ tition names add a 'p', and a number, e.g. /dev/md/home1p3. If there is no trailing digit, then the partition names just have a number added, e.g. /dev/md/scratch3. If the md device name is in a 'standard' format as described in DEVICE NAMES, then it will be created, if necessary, with the appropriate device number based on that name. If the device name is not in one of these formats, then an unused device number will be allocated. The device number will be considered unused if there is no active array for that number, and there is no entry in /dev for that number and with a non-standard name. Names that are not in 'standard' format are only allowed in \"/dev/md/\". This is meaningful with --create or --build." }, { "flag": "-a", "long": "--add", "arg": null, "description": "This option can be used in Grow mode in two cases. If the target array is a Linear array, then --add can be used to add one or more de‐ vices to the array. They are simply catenated on to the end of the array. Once added, the devices cannot be removed. If the --raid-disks option is being used to increase the number of devices in an ar‐ ray, then --add can be used to add some extra devices to be included in the array. In most cases this is not needed as the extra devices can be added as spares first, and then the number of raid disks can be changed. However, for RAID0 it is not possible to add spares. So to increase the number of devices in a RAID0, it is necessary to set the new number of devices, and to add the new devices, in the same command." }, { "flag": "", "long": "--nodes", "arg": null, "description": "Only works when the array is created for a clustered environment. It specifies the maximum number of nodes in the cluster that will use this device simultaneously. If not specified, this defaults to 4." }, { "flag": "", "long": "--write-journal", "arg": null, "description": "Specify journal device for the RAID-4/5/6 array. The journal device should be an SSD with a reasonable lifetime." }, { "flag": "-k", "long": null, "arg": null, "description": "Specify how the array maintains consistency in the case of an unexpected shutdown. Only relevant for RAID levels with redundancy. Currently supported options are: resync Full resync is performed and all redundancy is regenerated when the array is started after an unclean shutdown. bitmap Resync assisted by a write-intent bitmap. Implicitly selected when using --bitmap. journal For RAID levels 4/5/6, the journal device is used to log transactions and re‐ play after an unclean shutdown. Implicitly selected when using --write-journal. ppl For RAID5 only, Partial Parity Log is used to close the write hole and elimi‐ nate resync. PPL is stored in the metadata region of RAID member drives, no ad‐ ditional journal drive is needed. Can be used with --grow to change the consistency policy of an active array in some cases. See CONSISTENCY POLICY CHANGES below. For assemble:" }, { "flag": "-u", "long": null, "arg": null, "description": "uuid of array to assemble. Devices which don't have this uuid are excluded" }, { "flag": "-m", "long": null, "arg": null, "description": "Minor number of device that array was created for. Devices which don't have this mi‐ nor number are excluded. If you create an array as /dev/md1, then all superblocks will contain the minor number 1, even if the array is later assembled as /dev/md2. Giving the literal word \"dev\" for --super-minor will cause mdadm to use the minor num‐ ber of the md device that is being assembled. e.g. when assembling /dev/md0, --su‐ per-minor=dev will look for super blocks with a minor number of 0. --super-minor is only relevant for v0.90 metadata, and should not normally be used. Using --uuid is much safer." }, { "flag": "-N", "long": null, "arg": null, "description": "Specify the name of the array to assemble. It must be POSIX PORTABLE NAME compatible and cannot be longer than 32 chars. This must be the name that was specified when cre‐ ating the array. It must either match the name stored in the superblock exactly, or it must match with the current homehost prefixed to the start of the given name." }, { "flag": "-f", "long": "--force", "arg": null, "description": "Assemble the array even if the metadata on some devices appears to be out-of-date. If mdadm cannot find enough working devices to start the array, but can find some devices that are recorded as having failed, then it will mark those devices as working so that the array can be started. This works only for native. For external metadata it allows one to start dirty degraded RAID 4, 5, 6. An array which requires --force to be started may contain data corruption. Use it carefully." }, { "flag": "-R", "long": "--run", "arg": null, "description": "Attempt to start the array even if fewer drives were given than were present last time the array was active. Normally if not all the expected drives are found and --scan is not used, then the array will be assembled but not started. With --run an attempt will be made to start it anyway." }, { "flag": "", "long": "--no-degraded", "arg": null, "description": "This is the reverse of --run in that it inhibits the startup of array unless all ex‐ pected drives are present. This is only needed with --scan, and can be used if the physical connections to devices are not as reliable as you would like." }, { "flag": "-a", "long": null, "arg": null, "description": "See this option under Create and Build options." }, { "flag": "-b", "long": null, "arg": null, "description": "Specify the bitmap file that was given when the array was created. If an array has an internal bitmap, there is no need to specify this when assembling the array." }, { "flag": "", "long": null, "arg": null, "description": "If --backup-file was used while reshaping an array (e.g. changing number of devices or chunk size) and the system crashed during the critical section, then the same --backup-file must be presented to --assemble to allow possibly corrupted data to be restored, and the reshape to be completed." }, { "flag": "", "long": "--invalid-backup", "arg": null, "description": "If the file needed for the above option is not available for any reason an empty file can be given together with this option to indicate that the backup file is invalid. In this case the data that was being rearranged at the time of the crash could be ir‐ recoverably lost, but the rest of the array may still be recoverable. This option should only be used as a last resort if there is no way to recover the backup file." }, { "flag": "-U", "long": null, "arg": null, "description": "Update the superblock on each device while assembling the array. The argument given to this flag can be one of sparc2.2, summaries, uuid, name, nodes, homehost, home- cluster, resync, byteorder, devicesize, no-bitmap, bbl, no-bbl, ppl, no-ppl, lay‐ out-original, layout-alternate, layout-unspecified, metadata, or super-minor. The sparc2.2 option will adjust the superblock of an array what was created on a Sparc machine running a patched 2.2 Linux kernel. This kernel got the alignment of part of the superblock wrong. You can use the --examine --sparc2.2 option to mdadm to see what effect this would have. The super-minor option will update the preferred minor field on each superblock to match the minor number of the array being assembled. This can be useful if --examine reports a different \"Preferred Minor\" to --detail. In some cases this update will be performed automatically by the kernel driver. In particular, the update happens auto‐ matically at the first write to an array with redundancy (RAID level 1 or greater) on a 2.6 (or later) kernel. The uuid option will change the uuid of the array. If a UUID is given with the --uuid option that UUID will be used as a new UUID and will NOT be used to help identify the devices in the array. If no --uuid is given, a random UUID is chosen. The name option will change the name of the array as stored in the superblock. This is only supported for version-1 superblocks. The nodes option will change the nodes of the array as stored in the bitmap su‐ perblock. This option only works for a clustered environment. The homehost option will change the homehost as recorded in the superblock. For ver‐ sion-0 superblocks, this is the same as updating the UUID. For version-1 superblocks, this involves updating the name. The home-cluster option will change the cluster name as recorded in the superblock and bitmap. This option only works for a clustered environment. The resync option will cause the array to be marked dirty meaning that any redundancy in the array (e.g. parity for RAID5, copies for RAID1) may be incorrect. This will cause the RAID system to perform a \"resync\" pass to make sure that all redundant in‐ formation is correct. The byteorder option allows arrays to be moved between machines with different byte- order, such as from a big-endian machine like a Sparc or some MIPS machines, to a lit‐ tle-endian x8664 machine. When assembling such an array for the first time after a move, giving --update=byteorder will cause mdadm to expect superblocks to have their byteorder reversed, and will correct that order before assembling the array. This is only valid with original (Version 0.90) superblocks. The summaries option will correct the summaries in the superblock. That is the counts of total, working, active, failed, and spare devices. The devicesize option will rarely be of use. It applies to version 1.1 and 1.2 meta‐ data only (where the metadata is at the start of the device) and is only useful when the component device has changed size (typically become larger). The version 1 meta‐ data records the amount of the device that can be used to store data, so if a device in a version 1.1 or 1.2 array becomes larger, the metadata will still be visible, but the extra space will not. In this case it might be useful to assemble the array with --update=devicesize. This will cause mdadm to determine the maximum usable amount of space on each device and update the relevant field in the metadata. The metadata option only works on v0.90 metadata arrays and will convert them to v1.0 metadata. The array must not be dirty (i.e. it must not need a sync) and it must not have a write-intent bitmap. The old metadata will remain on the devices, but will appear older than the new meta‐ data and so will usually be ignored. The old metadata (or indeed the new metadata) can be removed by giving the appropriate --metadata= option to --zero-superblock. The no-bitmap option can be used when an array has an internal bitmap which is corrupt in some way so that assembling the array normally fails. It will cause any internal bitmap to be ignored. The bbl option will reserve space in each device for a bad block list. This will be 4K in size and positioned near the end of any free space between the superblock and the data. The no-bbl option will cause any reservation of space for a bad block list to be re‐ moved. If the bad block list contains entries, this will fail, as removing the list could cause data corruption. The ppl option will enable PPL for a RAID5 array and reserve space for PPL on each de‐ vice. There must be enough free space between the data and superblock and a write-in‐ tent bitmap or journal must not be used. The no-ppl option will disable PPL in the superblock. The layout-original and layout-alternate options are for RAID0 arrays with non-uniform devices size that were in use before Linux 5.4. If the array was being used with Linux 3.13 or earlier, then to assemble the array on a new kernel, --update=lay‐ out-original must be given. If the array was created and used with a kernel from Linux 3.14 to Linux 5.3, then --update=layout-alternate must be given. This only needs to be given once. Subsequent assembly of the array will happen normally. For more information, see md(4). The layout-unspecified option reverts the effect of layout-orignal or layout-alternate and allows the array to be again used on a kernel prior to Linux 5.3. This option should be used with great caution." }, { "flag": "", "long": "--freeze-reshape", "arg": null, "description": "This option is intended to be used in start-up scripts during the initrd boot phase. When the array under reshape is assembled during the initrd phase, this option stops the reshape after the reshape-critical section has been restored. This happens before the file system pivot operation and avoids loss of filesystem context. Losing file system context would cause reshape to be broken. Reshape can be continued later using the --continue option for the grow command. For Manage mode:" }, { "flag": "-t", "long": "--test", "arg": null, "description": "Unless a more serious error occurred, mdadm will exit with a status of 2 if no changes were made to the array and 0 if at least one change was made. This can be useful when an indirect specifier such as missing, detached or faulty is used in requesting an op‐ eration on the array. --test will report failure if these specifiers didn't find any match." }, { "flag": "-a", "long": "--add", "arg": null, "description": "hot-add listed devices. If a device appears to have recently been part of the array (possibly it failed or was removed) the device is re-added as described in the next point. If that fails or the device was never part of the array, the device is added as a hot-spare. If the array is degraded, it will immediately start to rebuild data onto that spare. Note that this and the following options are only meaningful on array with redundancy. They don't apply to RAID0 or Linear." }, { "flag": "", "long": "--re-add", "arg": null, "description": "re-add a device that was previously removed from an array. If the metadata on the de‐ vice reports that it is a member of the array, and the slot that it used is still va‐ cant, then the device will be added back to the array in the same position. This will normally cause the data for that device to be recovered. However, based on the event count on the device, the recovery may only require sections that are flagged by a write-intent bitmap to be recovered or may not require any recovery at all. When used on an array that has no metadata (i.e. it was built with --build) it will be assumed that bitmap-based recovery is enough to make the device fully consistent with the array. --re-add can also be accompanied by --update=devicesize, --update=bbl, or --up‐ date=no-bbl. See descriptions of these options when used in Assemble mode for an ex‐ planation of their use. If the device name given is missing then mdadm will try to find any device that looks like it should be part of the array but isn't and will try to re-add all such devices. If the device name given is faulty then mdadm will find all devices in the array that are marked faulty, remove them and attempt to immediately re-add them. This can be useful if you are certain that the reason for failure has been resolved." }, { "flag": "", "long": "--add-spare", "arg": null, "description": "Add a device as a spare. This is similar to --add except that it does not attempt --re-add first. The device will be added as a spare even if it looks like it could be a recent member of the array." }, { "flag": "-r", "long": "--remove", "arg": null, "description": "remove listed devices. They must not be active. i.e. they should be failed or spare devices. As well as the name of a device file (e.g. /dev/sda1) the words failed, detached and names like set-A can be given to --remove. The first causes all failed devices to be removed. The second causes any device which is no longer connected to the system (i.e an 'open' returns ENXIO) to be removed. The third will remove a set as described be‐ low under --fail." }, { "flag": "-f", "long": "--fail", "arg": null, "description": "Mark listed devices as faulty. As well as the name of a device file, the word de‐ tached or a set name like set-A can be given. The former will cause any device that has been detached from the system to be marked as failed. It can then be removed. For RAID10 arrays where the number of copies evenly divides the number of devices, the devices can be conceptually divided into sets where each set contains a single com‐ plete copy of the data on the array. Sometimes a RAID10 array will be configured so that these sets are on separate controllers. In this case, all the devices in one set can be failed by giving a name like set-A or set-B to --fail. The appropriate set names are reported by --detail." }, { "flag": "", "long": "--set-faulty", "arg": null, "description": "same as --fail." }, { "flag": "", "long": "--replace", "arg": null, "description": "Mark listed devices as requiring replacement. As soon as a spare is available, it will be rebuilt and will replace the marked device. This is similar to marking a de‐ vice as faulty, but the device remains in service during the recovery process to in‐ crease resilience against multiple failures. When the replacement process finishes, the replaced device will be marked as faulty. --with This can follow a list of --replace devices. The devices listed after --with will preferentially be used to replace the devices listed after --replace. These devices must already be spare devices in the array." }, { "flag": "", "long": "--write-mostly", "arg": null, "description": "Subsequent devices that are added or re-added will have the 'write-mostly' flag set. This is only valid for RAID1 and means that the 'md' driver will avoid reading from these devices if possible." }, { "flag": "", "long": "--readwrite", "arg": null, "description": "Subsequent devices that are added or re-added will have the 'write-mostly' flag cleared." }, { "flag": "", "long": "--cluster-confirm", "arg": null, "description": "Confirm the existence of the device. This is issued in response to an --add request by a node in a cluster. When a node adds a device it sends a message to all nodes in the cluster to look for a device with a UUID. This translates to a udev notification with the UUID of the device to be added and the slot number. The receiving node must ac‐ knowledge this message with --cluster-confirm. Valid arguments are : in case the device is found or :missing in case the device is not found." }, { "flag": "", "long": "--add-journal", "arg": null, "description": "Add a journal to an existing array, or recreate journal for a RAID-4/5/6 array that lost a journal device. To avoid interrupting ongoing write operations, --add-journal only works for array in Read-Only state." }, { "flag": "", "long": "--failfast", "arg": null, "description": "Subsequent devices that are added or re-added will have the 'failfast' flag set. This is only valid for RAID1 and RAID10 and means that the 'md' driver will avoid long timeouts on error handling where possible." }, { "flag": "", "long": "--nofailfast", "arg": null, "description": "Subsequent devices that are re-added will be re-added without the 'failfast' flag set. Each of these options requires that the first device listed is the array to be acted upon, and the remainder are component devices to be added, removed, marked as faulty, etc. Several different operations can be specified for different devices, e.g. mdadm /dev/md0 --add /dev/sda1 --fail /dev/sdb1 --remove /dev/sdb1 Each operation applies to all devices listed until the next operation. If an array is using a write-intent bitmap, then devices which have been removed can be re-added in a way that avoids a full reconstruction but instead just updates the blocks that have changed since the device was removed. For arrays with persistent metadata (superblocks) this is done automatically. For arrays created with --build mdadm needs to be told that this device we removed recently with --re-add. Devices can only be removed from an array if they are not in active use, i.e. that must be spares or failed devices. To remove an active device, it must first be marked as faulty. For Misc mode:" }, { "flag": "-Q", "long": "--query", "arg": null, "description": "Examine a device to see (1) if it is an md device and (2) if it is a component of an md array. Information about what is discovered is presented." }, { "flag": "-D", "long": "--detail", "arg": null, "description": "Print details of one or more md devices." }, { "flag": "", "long": "--detail-platform", "arg": null, "description": "Print details of the platform's RAID capabilities (firmware / hardware topology) for a given metadata format. If used without an argument, mdadm will scan all controllers looking for their capabilities. Otherwise, mdadm will only look at the controller specified by the argument in the form of an absolute filepath or a link, e.g. /sys/devices/pci0000:00/0000:00:1f.2." }, { "flag": "-Y", "long": "--export", "arg": null, "description": "When used with --detail, --detail-platform, --examine, or --incremental output will be formatted as key=value pairs for easy import into the environment. With --incremental The value MDSTARTED indicates whether an array was started (yes) or not, which may include a reason (unsafe, nothing, no). Also the value MDFOREIGN indicates if the array is expected on this host (no), or seems to be from elsewhere (yes)." }, { "flag": "-E", "long": "--examine", "arg": null, "description": "Print contents of the metadata stored on the named device(s). Note the contrast be‐ tween --examine and --detail. --examine applies to devices which are components of an array, while --detail applies to a whole array which is currently active." }, { "flag": "", "long": "--sparc2.2", "arg": null, "description": "If an array was created on a SPARC machine with a 2.2 Linux kernel patched with RAID support, the superblock will have been created incorrectly, or at least incompatibly with 2.4 and later kernels. Using the --sparc2.2 flag with --examine will fix the su‐ perblock before displaying it. If this appears to do the right thing, then the array can be successfully assembled using --assemble --update=sparc2.2." }, { "flag": "-X", "long": "--examine-bitmap", "arg": null, "description": "Report information about a bitmap file. The argument is either an external bitmap file or an array component in case of an internal bitmap. Note that running this on an array device (e.g. /dev/md0) does not report the bitmap for that array." }, { "flag": "", "long": "--examine-badblocks", "arg": null, "description": "List the bad-blocks recorded for the device, if a bad-blocks list has been configured. Currently only 1.x and IMSM metadata support bad-blocks lists. --dump=directory --restore=directory Save metadata from lists devices, or restore metadata to listed devices." }, { "flag": "-R", "long": "--run", "arg": null, "description": "start a partially assembled array. If --assemble did not find enough devices to fully start the array, it might leaving it partially assembled. If you wish, you can then use --run to start the array in degraded mode." }, { "flag": "-S", "long": "--stop", "arg": null, "description": "deactivate array, releasing all resources." }, { "flag": "-o", "long": "--readonly", "arg": null, "description": "mark array as readonly." }, { "flag": "-w", "long": "--readwrite", "arg": null, "description": "mark array as readwrite." }, { "flag": "", "long": "--zero-superblock", "arg": null, "description": "If the device contains a valid md superblock, the block is overwritten with zeros. With --force the block where the superblock would be is overwritten even if it doesn't appear to be valid. Note: Be careful when calling --zero-superblock with clustered raid. Make sure the ar‐ ray isn't used or assembled in another cluster node before executing it." }, { "flag": "", "long": null, "arg": null, "description": "If the device is a container and the argument to --kill-subarray specifies an inactive subarray in the container, then the subarray is deleted. Deleting all subarrays will leave an 'empty-container' or spare superblock on the drives. See --zero-superblock for completely removing a superblock. Note that some formats depend on the subarray index for generating a UUID, this command will fail if it would change the UUID of an active subarray." }, { "flag": "", "long": null, "arg": null, "description": "If the device is a container and the argument to --update-subarray specifies a subar‐ ray in the container, then attempt to update the given superblock field in the subar‐ ray. See below in MISC MODE for details." }, { "flag": "-t", "long": "--test", "arg": null, "description": "When used with --detail, the exit status of mdadm is set to reflect the status of the device. See below in MISC MODE for details." }, { "flag": "-W", "long": "--wait", "arg": null, "description": "For each md device given, wait for any resync, recovery, or reshape activity to finish before returning. mdadm will return with success if it actually waited for every de‐ vice listed, otherwise it will return failure." }, { "flag": "", "long": "--wait-clean", "arg": null, "description": "For each md device given, or each device in /proc/mdstat if --scan is given, arrange for the array to be marked clean as soon as possible. mdadm will return with success if the array uses external metadata and we successfully waited. For native arrays, this returns immediately as the kernel handles dirty-clean transitions at shutdown. No action is taken if safe-mode handling is disabled." }, { "flag": "", "long": null, "arg": null, "description": "Set the \"syncaction\" for all md devices given to one of idle, frozen, check, repair. Setting to idle will abort any currently running action though some actions will auto‐ matically restart. Setting to frozen will abort any current action and ensure no other action starts automatically. Details of check and repair can be found in md(4) under SCRUBBING AND MISMATCHES. For Incremental Assembly mode: --rebuild-map, -r Rebuild the map file (/run/mdadm/map) that mdadm uses to help track which arrays are currently being assembled. --run, -R Run any array assembled as soon as a minimal number of devices is available, rather than waiting until all expected devices are present. --scan, -s Only meaningful with -R this will scan the map file for arrays that are being incre‐ mentally assembled and will try to start any that are not already started. If any such array is listed in mdadm.conf as requiring an external bitmap, that bitmap will be attached first. --fail, -f This allows the hot-plug system to remove devices that have fully disappeared from the kernel. It will first fail and then remove the device from any array it belongs to. The device name given should be a kernel device name such as \"sda\", not a name in /dev." }, { "flag": "", "long": null, "arg": null, "description": "Only used with --fail. The 'path' given will be recorded so that if a new device ap‐ pears at the same location it can be automatically added to the same array. This al‐ lows the failed device to be automatically replaced by a new device without metadata if it appears at specified path. This option is normally only set by an udev script. For Monitor mode:" }, { "flag": "-m", "long": "--mail", "arg": null, "description": "Give a mail address to send alerts to." }, { "flag": "-p", "long": "--alert", "arg": null, "description": "Give a program to be run whenever an event is detected." }, { "flag": "-y", "long": "--syslog", "arg": null, "description": "Cause all events to be reported through 'syslog'. The messages have facility of 'dae‐ mon' and varying priorities." }, { "flag": "-d", "long": "--delay", "arg": null, "description": "Give a delay in seconds. mdadm polls the md arrays and then waits this many seconds before polling again. The default is 60 seconds. Since 2.6.16, there is no need to reduce this as the kernel alerts mdadm immediately when there is any change." }, { "flag": "-r", "long": "--increment", "arg": null, "description": "Give a percentage increment. mdadm will generate RebuildNN events with the given per‐ centage increment." }, { "flag": "-f", "long": "--daemonise", "arg": null, "description": "Tell mdadm to run as a background daemon if it decides to monitor anything. This causes it to fork and run in the child, and to disconnect from the terminal. The process id of the child is written to stdout. This is useful with --scan which will only continue monitoring if a mail address or alert program is found in the config file." }, { "flag": "-i", "long": "--pid-file", "arg": null, "description": "When mdadm is running in daemon mode, write the pid of the daemon process to the spec‐ ified file, instead of printing it on standard output." }, { "flag": "-1", "long": "--oneshot", "arg": null, "description": "Check arrays only once. This will generate NewArray events and more significantly De‐ gradedArray and SparesMissing events. Running mdadm --monitor --scan -1 from a cron script will ensure regular notification of any degraded arrays." }, { "flag": "-t", "long": "--test", "arg": null, "description": "Generate a TestMessage alert for every array found at startup. This alert gets mailed and passed to the alert program. This can be used for testing that alert message do get through successfully." }, { "flag": "", "long": "--no-sharing", "arg": null, "description": "This inhibits the functionality for moving spares between arrays. Only one monitoring process started with --scan but without this flag is allowed, otherwise the two could interfere with each other." } ], "examples": [ "mdadm --query /dev/name-of-device", "This will find out if a given device is a RAID array, or is part of one, and will provide", "brief information about the device.", "mdadm --assemble --scan", "This will assemble and start all arrays listed in the standard config file. This command", "will typically go in a system startup file.", "mdadm --stop --scan", "This will shut down all arrays that can be shut down (i.e. are not currently in use). This", "will typically go in a system shutdown script.", "mdadm --follow --scan --delay=120", "If (and only if) there is an Email address or program given in the standard config file, then", "monitor the status of all arrays listed in that file by polling them ever 2 minutes.", "mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/hd[ac]1", "Create /dev/md0 as a RAID1 array consisting of /dev/hda1 and /dev/hdc1.", "echo 'DEVICE /dev/hd*[0-9] /dev/sd*[0-9]' > mdadm.conf", "mdadm --detail --scan >> mdadm.conf", "This will create a prototype config file that describes currently active arrays that are", "known to be made from partitions of IDE or SCSI drives. This file should be reviewed before", "being used as it may contain unwanted detail.", "echo 'DEVICE /dev/hd[a-z] /dev/sd*[a-z]' > mdadm.conf", "mdadm --examine --scan --config=mdadm.conf >> mdadm.conf", "This will find arrays which could be assembled from existing IDE and SCSI whole drives (not", "partitions), and store the information in the format of a config file. This file is very", "likely to contain unwanted detail, particularly the devices= entries. It should be reviewed", "and edited before being used as an actual config file.", "mdadm --examine --brief --scan --config=partitions", "mdadm -Ebsc partitions", "Create a list of devices by reading /proc/partitions, scan these for RAID superblocks, and", "printout a brief listing of all that were found.", "mdadm -Ac partitions -m 0 /dev/md0", "Scan all partitions and devices listed in /proc/partitions and assemble /dev/md0 out of all", "such devices with a RAID superblock with a minor number of 0.", "mdadm --monitor --scan --daemonise > /run/mdadm/mon.pid", "If config file contains a mail address or alert program, run mdadm in the background in moni‐", "tor mode monitoring all md devices. Also write pid of mdadm daemon to /run/mdadm/mon.pid.", "mdadm -Iq /dev/somedevice", "Try to incorporate newly discovered device into some array as appropriate.", "mdadm --incremental --rebuild-map --run --scan", "Rebuild the array map from any current arrays, and then start any that can be started.", "mdadm /dev/md4 --fail detached --remove detached", "Any devices which are components of /dev/md4 will be marked as faulty and then remove from", "the array.", "mdadm --grow /dev/md4 --level=6 --backup-file=/root/backup-md4", "The array /dev/md4 which is currently a RAID5 array will be converted to RAID6. There should", "normally already be a spare drive attached to the array as a RAID6 needs one more drive than", "a matching RAID5.", "mdadm --create /dev/md/ddf --metadata=ddf --raid-disks 6 /dev/sd[a-f]", "Create a DDF array over 6 devices.", "mdadm --create /dev/md/home -n3 -l5 -z 30000000 /dev/md/ddf", "Create a RAID5 array over any 3 devices in the given DDF set. Use only 30 gigabytes of each", "device.", "mdadm -A /dev/md/ddf1 /dev/sd[a-f]", "Assemble a pre-exist ddf array.", "mdadm -I /dev/md/ddf1", "Assemble all arrays contained in the ddf array, assigning names as appropriate.", "mdadm --create --help", "Provide help about the Create mode.", "mdadm --config --help", "Provide help about the format of the config file.", "mdadm --help", "Provide general help." ], "see_also": [ { "name": "mdmon", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/mdmon/8/json" }, { "name": "mdadm.conf", "section": "5", "url": "https://www.chedong.com/phpMan.php/man/mdadm.conf/5/json" }, { "name": "md", "section": "4", "url": "https://www.chedong.com/phpMan.php/man/md/4/json" } ] }