TODO

   1
   2
   3 ?? Allow -S /dev/md? - current complains subsequent not a/d/r
   4
   5 * new "Query" mode to subsume --detail and --examine.
   6    --query or -Q, takes a device and tells if it is an MD device,
   7    and also tells in a raid superblock is found.
   8
   9 * write proc.c to parse /proc/mdstat file, and maybe /proc/partitions too.
  10    Build list of arrays:  name, rebuild-percent
  11
  12 * --detail --scan to read mdadm.conf, and then iterate over these,
  13     but assume --brief.  --verbose can override
  14     check each subdevice to see if it is in conf_get_devs.
  15     Warn if not.
  16
  17 * Support multipath ... maybe...
  18
  19 * --follow to syslog
  20
  21 * --follow to move spares around
  22
  23 * --follow to notice other events:
  24      rebuild started
  25      spare activated
  26      spare removed
  27      spare added
  28
  29 ------------------------------------
  30 - --examine --scan scans all drives and build an mdadm.conf file DONE
  31
  32 - check superblock checksum in examine DONE
  33 - report "chunk" or "rounding" depending on raid level DONE
  34 - report "linear" instead of "-1" for raid level DONE
  35 - decode ayout depending on raid level DONE
  36 - --verbose and --force flags. DONE
  37
  38 - set md_minor, *_disks for Create  - DONE
  39 - for create raid5, how to choose between
  40    all working, but not insync
  41    one missing, one spare, insync  DONE (--force)
  42 - and for raid1 - some failed drives...  (missing)
  43
  44 - when RUN_ARRAY, make sure *_disks counts are right
  45
  46 - get --detail to extract extra stuff from superblock,
  47    like uuid  DONE
  48 - --detail --brief to give a config file line DONE
  49 - parse config file. DONE
  50 - test...
  51
  52 - when --assemble --scan, if an underlying device is an md device,
  53   then try to assemble that device first.
  54
  55
  56 - mdadm -S /dev/md0 /dev/md1 gives internal error FIXED
  57
  58 - mdadm --detail --scan print summary of what it can find?
  59
  60
  61 ---------
  62 Assemble doesn't add spares. - DONE
  63 Create to allow "missing" name for devices.
  64 Create to accept "--force" for do exactly what is requested
  65 - get Assemble to upgrade devices if force flag.
  66 ARRAY lines in config file to have super_minor=n
  67 ARRAY lines in config file to have device=pattern, and only accept
  68    those devices
  69    If UUID given, insist on that
  70    If not, but super_minor given, require all found with that minor
  71     to have same uuid
  72    If only device given, all valid supers on those devices must have
  73     same uuid
  74 allow /dev/mdX as first argument before any options
  75 Possible --dry-run option for create and assemble--force
  76
  77 Assemble to check that all devices mentioned in superblock
  78   are present.
  79
  80 New mode: --Monitor (or --Follow)
  81   Periodically check status of all arrays (listed in config file).
  82   Log every event and apparent cause - or differences
  83   Email and alert - or run a program - for important events
  84   Move spares around if necessary.
  85
  86   An Array line can have a spare-group= field that indicates that
  87    the array shares spares with other arrays with the same
  88    spare-group name.
  89    If an array has a failed and no spares, then check all other
  90      arrays in the spare group.  If one has no failures and a spare,
  91      then consider that spare.
  92     Choose the smallest considered spare that is large enough.
  93     If there is one, then hot-remove it from it's home, and
  94     hot-add it to the array in question.
  95
  96   --mail-to address
  97   --alert-handler program
  98
  99   Will also extract information from /proc/mdstat if present,
 100   and consider 20% marks in rebuild as events.
 101
 102   Events are:
 103      drive fails  - causes mail to be sent
 104      rebuild started
 105      spare activated
 106      spare removed
 107      spare added