ANNOUNCE-3.0-devel1

   1 Subject:  ANNOUNCE: mdadm 3.0-devel1 - A tool for managing Soft RAID under Linux
   2
   3 I am pleased to announce the availability of
   4    mdadm version 3.0-devel1
   5
   6 It is available at the usual places:
   7    countrycode=xx.
   8    http://www.${countrycode}kernel.org/pub/linux/utils/raid/mdadm/
   9 and via git at
  10    git://neil.brown.name/mdadm
  11    http://neil.brown.name/git?p=mdadm
  12
  13 Note that this is a "devel" release.  It is not intended for
  14 production use yet, but rather for testing and ongoing development.
  15
  16 The significant change which justifies the new major version number is
  17 that mdadm can now handle metadata updates entirely in userspace.
  18 This allows mdadm to support metadata formats that the kernel knows
  19 nothing about.
  20
  21 Currently two such metadata formats are supported:
  22   - DDF  - The SNIA standard format
  23   - Intel Matrix - The metadata used by recent Intel ICH controlers.
  24
  25 The manual pages have not yet been updated, but here is a brief outline.
  26
  27 Externally managed metadata introduces the concept of a 'container'.
  28 A container is a collection of (normally) physical devices which have
  29 a common set of metadata.  A container is assembled as an md array, but
  30 is left 'inactive'.
  31
  32 A container can contain one or more data arrays.  These are composed from
  33 slices (partitions?) of various devices in the container.
  34
  35 For example, a 5 devices DDF set can container a RAID1 using the first
  36 half of two devices, a RAID0 using the first half of the remain 3 devices,
  37 and a RAID5 over thte second half of all 5 devices.
  38
  39 A container can be created with
  40
  41    mdadm --create /dev/md0 -e ddf -n5 /dev/sd[abcde]
  42
  43 or "-e imsm" to use the Intel Matrix Storage Manager.
  44
  45 An array can be created within a container either by giving the
  46 container name and the only member:
  47
  48    mdadm -C /dev/md1 --level raid1 -n 2 /dev/md0
  49
  50 or by listing the component devices
  51
  52    mdadm -C /dev/md2 --level raid0 -n 3 /dev/sd[cde]
  53
  54 The assemble a container, it is easiest just to pass each device in turn to
  55 mdadm -I
  56
  57   for i in /dev/sd[abcde]
  58   do mdadm -I $i
  59   done
  60
  61 This will assemble the container and the components.
  62
  63 Alternately the container can be assembled explicitly
  64
  65    mdadm -A /dev/md0 /dev/sd[abcde]
  66
  67 Then the components can all be assembled with
  68
  69    mdadm -I /dev/md0
  70
  71 For each container, mdadm will start a program called "mdmon" which will
  72 monitor the array and effect any metadata updates needed.  The array is
  73 initially assembled readonly. It is up to "mdmon" to mark the metadata
  74 as 'dirty' and which the array to 'read-write'.
  75
  76 The version 0.90 and 1.x metadata formats supported by previous
  77 versions for mdadm are still supported and the kernel still performs
  78 the same updates it use to.  The new 'mdmon' approach is only used for
  79 newly introduced metadata types.
  80
  81 Any testing and feedback will be greatly appreciated.
  82
  83 NeilBrown  18th September 2008
  84