Documentation/md.txt

   1 Tools that manage md devices can be found at
   2    http://www.<country>.kernel.org/pub/linux/utils/raid/....
   3
   4
   5 Boot time assembly of RAID arrays
   6 ---------------------------------
   7
   8 You can boot with your md device with the following kernel command
   9 lines:
  10
  11 for old raid arrays without persistent superblocks:
  12   md=<md device no.>,<raid level>,<chunk size factor>,<fault level>,dev0,dev1,...,devn
  13
  14 for raid arrays with persistent superblocks
  15   md=<md device no.>,dev0,dev1,...,devn
  16 or, to assemble a partitionable array:
  17   md=d<md device no.>,dev0,dev1,...,devn
  18
  19 md device no. = the number of the md device ...
  20               0 means md0,
  21               1 md1,
  22               2 md2,
  23               3 md3,
  24               4 md4
  25
  26 raid level = -1 linear mode
  27               0 striped mode
  28               other modes are only supported with persistent super blocks
  29
  30 chunk size factor = (raid-0 and raid-1 only)
  31               Set  the chunk size as 4k << n.
  32
  33 fault level = totally ignored
  34
  35 dev0-devn: e.g. /dev/hda1,/dev/hdc1,/dev/sda1,/dev/sdb1
  36
  37 A possible loadlin line (Harald Hoyer <HarryH@Royal.Net>)  looks like this:
  38
  39 e:\loadlin\loadlin e:\zimage root=/dev/md0 md=0,0,4,0,/dev/hdb2,/dev/hdc3 ro
  40
  41
  42 Boot time autodetection of RAID arrays
  43 --------------------------------------
  44
  45 When md is compiled into the kernel (not as module), partitions of
  46 type 0xfd are scanned and automatically assembled into RAID arrays.
  47 This autodetection may be suppressed with the kernel parameter
  48 "raid=noautodetect".  As of kernel 2.6.9, only drives with a type 0
  49 superblock can be autodetected and run at boot time.
  50
  51 The kernel parameter "raid=partitionable" (or "raid=part") means
  52 that all auto-detected arrays are assembled as partitionable.
  53
  54 Boot time assembly of degraded/dirty arrays
  55 -------------------------------------------
  56
  57 If a raid5 or raid6 array is both dirty and degraded, it could have
  58 undetectable data corruption.  This is because the fact that it is
  59 'dirty' means that the parity cannot be trusted, and the fact that it
  60 is degraded means that some datablocks are missing and cannot reliably
  61 be reconstructed (due to no parity).
  62
  63 For this reason, md will normally refuse to start such an array.  This
  64 requires the sysadmin to take action to explicitly start the array
  65 desipite possible corruption.  This is normally done with
  66    mdadm --assemble --force ....
  67
  68 This option is not really available if the array has the root
  69 filesystem on it.  In order to support this booting from such an
  70 array, md supports a module parameter "start_dirty_degraded" which,
  71 when set to 1, bypassed the checks and will allows dirty degraded
  72 arrays to be started.
  73
  74 So, to boot with a root filesystem of a dirty degraded raid[56], use
  75
  76    md-mod.start_dirty_degraded=1
  77
  78
  79 Superblock formats
  80 ------------------
  81
  82 The md driver can support a variety of different superblock formats.
  83 Currently, it supports superblock formats "0.90.0" and the "md-1" format
  84 introduced in the 2.5 development series.
  85
  86 The kernel will autodetect which format superblock is being used.
  87
  88 Superblock format '0' is treated differently to others for legacy
  89 reasons - it is the original superblock format.
  90
  91
  92 General Rules - apply for all superblock formats
  93 ------------------------------------------------
  94
  95 An array is 'created' by writing appropriate superblocks to all
  96 devices.
  97
  98 It is 'assembled' by associating each of these devices with an
  99 particular md virtual device.  Once it is completely assembled, it can
 100 be accessed.
 101
 102 An array should be created by a user-space tool.  This will write
 103 superblocks to all devices.  It will usually mark the array as
 104 'unclean', or with some devices missing so that the kernel md driver
 105 can create appropriate redundancy (copying in raid1, parity
 106 calculation in raid4/5).
 107
 108 When an array is assembled, it is first initialized with the
 109 SET_ARRAY_INFO ioctl.  This contains, in particular, a major and minor
 110 version number.  The major version number selects which superblock
 111 format is to be used.  The minor number might be used to tune handling
 112 of the format, such as suggesting where on each device to look for the
 113 superblock.
 114
 115 Then each device is added using the ADD_NEW_DISK ioctl.  This
 116 provides, in particular, a major and minor number identifying the
 117 device to add.
 118
 119 The array is started with the RUN_ARRAY ioctl.
 120
 121 Once started, new devices can be added.  They should have an
 122 appropriate superblock written to them, and then passed be in with
 123 ADD_NEW_DISK.
 124
 125 Devices that have failed or are not yet active can be detached from an
 126 array using HOT_REMOVE_DISK.
 127
 128
 129 Specific Rules that apply to format-0 super block arrays, and
 130        arrays with no superblock (non-persistent).
 131 -------------------------------------------------------------
 132
 133 An array can be 'created' by describing the array (level, chunksize
 134 etc) in a SET_ARRAY_INFO ioctl.  This must has major_version==0 and
 135 raid_disks != 0.
 136
 137 Then uninitialized devices can be added with ADD_NEW_DISK.  The
 138 structure passed to ADD_NEW_DISK must specify the state of the device
 139 and it's role in the array.
 140
 141 Once started with RUN_ARRAY, uninitialized spares can be added with
 142 HOT_ADD_DISK.
 143
 144
 145
 146 MD devices in sysfs
 147 -------------------
 148 md devices appear in sysfs (/sys) as regular block devices,
 149 e.g.
 150    /sys/block/md0
 151
 152 Each 'md' device will contain a subdirectory called 'md' which
 153 contains further md-specific information about the device.
 154
 155 All md devices contain:
 156   level
 157      a text file indicating the 'raid level'.  This may be a standard
 158      numerical level prefixed by "RAID-" - e.g. "RAID-5", or some
 159      other name such as "linear" or "multipath".
 160      If no raid level has been set yet (array is still being
 161      assembled), this file will be empty.
 162
 163   raid_disks
 164      a text file with a simple number indicating the number of devices
 165      in a fully functional array.  If this is not yet known, the file
 166      will be empty.  If an array is being resized (not currently
 167      possible) this will contain the larger of the old and new sizes.
 168      Some raid level (RAID1) allow this value to be set while the
 169      array is active.  This will reconfigure the array.   Otherwise
 170      it can only be set while assembling an array.
 171
 172   chunk_size
 173      This is the size if bytes for 'chunks' and is only relevant to
 174      raid levels that involve striping (1,4,5,6,10). The address space
 175      of the array is conceptually divided into chunks and consecutive
 176      chunks are striped onto neighbouring devices.
 177      The size should be atleast PAGE_SIZE (4k) and should be a power
 178      of 2.  This can only be set while assembling an array
 179
 180   component_size
 181      For arrays with data redundancy (i.e. not raid0, linear, faulty,
 182      multipath), all components must be the same size - or at least
 183      there must a size that they all provide space for.  This is a key
 184      part or the geometry of the array.  It is measured in sectors
 185      and can be read from here.  Writing to this value may resize
 186      the array if the personality supports it (raid1, raid5, raid6),
 187      and if the component drives are large enough.
 188
 189   metadata_version
 190      This indicates the format that is being used to record metadata
 191      about the array.  It can be 0.90 (traditional format), 1.0, 1.1,
 192      1.2 (newer format in varying locations) or "none" indicating that
 193      the kernel isn't managing metadata at all.
 194
 195   level
 196      The raid 'level' for this array.  The name will often (but not
 197      always) be the same as the name of the module that implements the
 198      level.  To be auto-loaded the module must have an alias
 199         md-$LEVEL  e.g. md-raid5
 200      This can be written only while the array is being assembled, not
 201      after it is started.
 202
 203    new_dev
 204      This file can be written but not read.  The value written should
 205      be a block device number as major:minor.  e.g. 8:0
 206      This will cause that device to be attached to the array, if it is
 207      available.  It will then appear at md/dev-XXX (depending on the
 208      name of the device) and further configuration is then possible.
 209
 210    safe_mode_delay
 211      When an md array has seen no write requests for a certain period
 212      of time, it will be marked as 'clean'.  When another write
 213      request arrive, the array is marked as 'dirty' before the write
 214      commenses.  This is known as 'safe_mode'.
 215      The 'certain period' is controlled by this file which stores the
 216      period as a number of seconds.  The default is 200msec (0.200).
 217      Writing a value of 0 disables safemode.
 218
 219    array_state
 220      This file contains a single word which describes the current
 221      state of the array.  In many cases, the state can be set by
 222      writing the word for the desired state, however some states
 223      cannot be explicitly set, and some transitions are not allowed.
 224
 225      clear
 226          No devices, no size, no level
 227          Writing is equivalent to STOP_ARRAY ioctl
 228      inactive
 229          May have some settings, but array is not active
 230             all IO results in error
 231          When written, doesn't tear down array, but just stops it
 232      suspended (not supported yet)
 233          All IO requests will block. The array can be reconfigured.
 234          Writing this, if accepted, will block until array is quiessent
 235      readonly
 236          no resync can happen.  no superblocks get written.
 237          write requests fail
 238      read-auto
 239          like readonly, but behaves like 'clean' on a write request.
 240
 241      clean - no pending writes, but otherwise active.
 242          When written to inactive array, starts without resync
 243          If a write request arrives then
 244            if metadata is known, mark 'dirty' and switch to 'active'.
 245            if not known, block and switch to write-pending
 246          If written to an active array that has pending writes, then fails.
 247      active
 248          fully active: IO and resync can be happening.
 249          When written to inactive array, starts with resync
 250
 251      write-pending
 252          clean, but writes are blocked waiting for 'active' to be written.
 253
 254      active-idle
 255          like active, but no writes have been seen for a while (safe_mode_delay).
 256
 257
 258    sync_speed_min
 259    sync_speed_max
 260      This are similar to /proc/sys/dev/raid/speed_limit_{min,max}
 261      however they only apply to the particular array.
 262      If no value has been written to these, of if the word 'system'
 263      is written, then the system-wide value is used.  If a value,
 264      in kibibytes-per-second is written, then it is used.
 265      When the files are read, they show the currently active value
 266      followed by "(local)" or "(system)" depending on whether it is
 267      a locally set or system-wide value.
 268
 269    sync_completed
 270      This shows the number of sectors that have been completed of
 271      whatever the current sync_action is, followed by the number of
 272      sectors in total that could need to be processed.  The two
 273      numbers are separated by a '/'  thus effectively showing one
 274      value, a fraction of the process that is complete.
 275
 276    sync_speed
 277      This shows the current actual speed, in K/sec, of the current
 278      sync_action.  It is averaged over the last 30 seconds.
 279
 280
 281 As component devices are added to an md array, they appear in the 'md'
 282 directory as new directories named
 283       dev-XXX
 284 where XXX is a name that the kernel knows for the device, e.g. hdb1.
 285 Each directory contains:
 286
 287       block
 288         a symlink to the block device in /sys/block, e.g.
 289              /sys/block/md0/md/dev-hdb1/block -> ../../../../block/hdb/hdb1
 290
 291       super
 292         A file containing an image of the superblock read from, or
 293         written to, that device.
 294
 295       state
 296         A file recording the current state of the device in the array
 297         which can be a comma separated list of
 298               faulty   - device has been kicked from active use due to
 299                          a detected fault
 300               in_sync  - device is a fully in-sync member of the array
 301               spare    - device is working, but not a full member.
 302                          This includes spares that are in the process
 303                          of being recoverred to
 304         This list make grow in future.
 305
 306       errors
 307         An approximate count of read errors that have been detected on
 308         this device but have not caused the device to be evicted from
 309         the array (either because they were corrected or because they
 310         happened while the array was read-only).  When using version-1
 311         metadata, this value persists across restarts of the array.
 312
 313         This value can be written while assembling an array thus
 314         providing an ongoing count for arrays with metadata managed by
 315         userspace.
 316
 317       slot
 318         This gives the role that the device has in the array.  It will
 319         either be 'none' if the device is not active in the array
 320         (i.e. is a spare or has failed) or an integer less than the
 321         'raid_disks' number for the array indicating which possition
 322         it currently fills.  This can only be set while assembling an
 323         array.  A device for which this is set is assumed to be working.
 324
 325       offset
 326         This gives the location in the device (in sectors from the
 327         start) where data from the array will be stored.  Any part of
 328         the device before this offset us not touched, unless it is
 329         used for storing metadata (Formats 1.1 and 1.2).
 330
 331       size
 332         The amount of the device, after the offset, that can be used
 333         for storage of data.  This will normally be the same as the
 334         component_size.  This can be written while assembling an
 335         array.  If a value less than the current component_size is
 336         written, component_size will be reduced to this value.
 337
 338
 339 An active md device will also contain and entry for each active device
 340 in the array.  These are named
 341
 342     rdNN
 343
 344 where 'NN' is the possition in the array, starting from 0.
 345 So for a 3 drive array there will be rd0, rd1, rd2.
 346 These are symbolic links to the appropriate 'dev-XXX' entry.
 347 Thus, for example,
 348        cat /sys/block/md*/md/rd*/state
 349 will show 'in_sync' on every line.
 350
 351
 352
 353 Active md devices for levels that support data redundancy (1,4,5,6)
 354 also have
 355
 356    sync_action
 357      a text file that can be used to monitor and control the rebuild
 358      process.  It contains one word which can be one of:
 359        resync        - redundancy is being recalculated after unclean
 360                        shutdown or creation
 361        recover       - a hot spare is being built to replace a
 362                        failed/missing device
 363        idle          - nothing is happening
 364        check         - A full check of redundancy was requested and is
 365                        happening.  This reads all block and checks
 366                        them. A repair may also happen for some raid
 367                        levels.
 368        repair        - A full check and repair is happening.  This is
 369                        similar to 'resync', but was requested by the
 370                        user, and the write-intent bitmap is NOT used to
 371                        optimise the process.
 372
 373       This file is writable, and each of the strings that could be
 374       read are meaningful for writing.
 375
 376        'idle' will stop an active resync/recovery etc.  There is no
 377            guarantee that another resync/recovery may not be automatically
 378            started again, though some event will be needed to trigger
 379            this.
 380         'resync' or 'recovery' can be used to restart the
 381            corresponding operation if it was stopped with 'idle'.
 382         'check' and 'repair' will start the appropriate process
 383            providing the current state is 'idle'.
 384
 385    mismatch_count
 386       When performing 'check' and 'repair', and possibly when
 387       performing 'resync', md will count the number of errors that are
 388       found.  The count in 'mismatch_cnt' is the number of sectors
 389       that were re-written, or (for 'check') would have been
 390       re-written.  As most raid levels work in units of pages rather
 391       than sectors, this my be larger than the number of actual errors
 392       by a factor of the number of sectors in a page.
 393
 394 Each active md device may also have attributes specific to the
 395 personality module that manages it.
 396 These are specific to the implementation of the module and could
 397 change substantially if the implementation changes.
 398
 399 These currently include
 400
 401   stripe_cache_size  (currently raid5 only)
 402       number of entries in the stripe cache.  This is writable, but
 403       there are upper and lower limits (32768, 16).  Default is 128.
 404   strip_cache_active (currently raid5 only)
 405       number of active entries in the stripe cache