why won't this RAID5 device start?

[David T-G <davidtg-robot@justpickone.org>]

Hi, all --

I recently migrated our disk farm to a new box with a new OS build
(openSuSE from KNOPPIX).  Aside from the usual challenges of setting
up the world again, I have a 3-device RAID5 volume that won't start.
The other metadevice is fine, though; I think we can say that the md
system is running.  Soooooo ...  Where do I start?

  diskfarm:~ # cat /proc/mdstat 
  Personalities : [raid6] [raid5] [raid4] 
  md0 : active raid5 sdc1[3] sdd1[4] sdb1[0]
        11720265216 blocks super 1.2 level 5, 512k chunk, algorithm 2 [4/3] [U_UU]
        
  md127 : inactive sdl2[0](S) sdj2[3](S) sdf2[1](S)
        2196934199 blocks super 1.2
         
  unused devices: <none>

[No, I don't know why md127 was first a moment ago!]  I tried reassembling
the device, but mdadm doesn't like it.

  diskfarm:~ # mdadm --stop /dev/md127
  mdadm: stopped /dev/md127
  diskfarm:~ # mdadm --assemble --scan
  mdadm: /dev/md/750Graid5md assembled from 1 drive - not enough to start the array.
  mdadm: Found some drive for an array that is already active: /dev/md/0
  mdadm: giving up.
  mdadm: No arrays found in config file or automatically

But ... what's with just 1 drive?

  diskfarm:~ # for D in /dev/sd[fjl] ; do parted $D print ; done
  Model: ATA WDC WD7500BPKX-7 (scsi)
  Disk /dev/sdf: 750GB
  Sector size (logical/physical): 512B/4096B
  Partition Table: gpt
  Disk Flags: 

  Number  Start   End    Size   File system  Name              Flags
   2      1049kB  750GB  750GB  ntfs         Linux RAID        raid
   3      750GB   750GB  134MB  ext3         Linux filesystem

  Model: ATA WDC WD7500BPKX-7 (scsi)
  Disk /dev/sdj: 750GB
  Sector size (logical/physical): 512B/4096B
  Partition Table: gpt
  Disk Flags: 

  Number  Start   End    Size   File system  Name              Flags
   2      1049kB  750GB  750GB  ntfs         Linux RAID        raid
   3      750GB   750GB  134MB  xfs          Linux filesystem

  Model: ATA Hitachi HDE72101 (scsi)
  Disk /dev/sdl: 1000GB
  Sector size (logical/physical): 512B/512B
  Partition Table: msdos
  Disk Flags: 

  Number  Start   End     Size    Type     File system  Flags
   1      1049kB  4227MB  4226MB  primary  ntfs         diag, type=27
   2      4227MB  754GB   750GB   primary  reiserfs     raid, type=fd
   3      754GB   754GB   134MB   primary  reiserfs     type=83
   4      754GB   1000GB  246GB   primary  reiserfs     type=83

Slice 2 on each is the RAID partition, slice 3 on each is a little 
filesystem for bare-bones info, and slice 4 on sdl is a normal basic
filesystem for scratch content.

  diskfarm:~ # mdadm --examine /dev/sd[fjl]2 | egrep '/dev|Name|Role|State|Checksum|Events|UUID'
  /dev/sdf2:
       Array UUID : 88575f01:592167fd:bd9f9ba1:a61fafc4
             Name : diskfarm:750Graid5md  (local to host diskfarm)
            State : clean
      Device UUID : e916fc67:b8b7fc59:51440134:fa431d02
         Checksum : 43f9e7a4 - correct
           Events : 720
     Device Role : Active device 1
     Array State : AAA ('A' == active, '.' == missing, 'R' == replacing)
  /dev/sdj2:
       Array UUID : 88575f01:592167fd:bd9f9ba1:a61fafc4
             Name : diskfarm:750Graid5md  (local to host diskfarm)
            State : clean
      Device UUID : 0b847f84:83e80a3d:a0dc11e7:60bffc9f
         Checksum : 9522782b - correct
           Events : 177792
     Device Role : Active device 2
     Array State : A.A ('A' == active, '.' == missing, 'R' == replacing)
  /dev/sdl2:
       Array UUID : 88575f01:592167fd:bd9f9ba1:a61fafc4
             Name : diskfarm:750Graid5md  (local to host diskfarm)
            State : clean
      Device UUID : cc53440e:cb9180e4:be4c38d4:88a676eb
         Checksum : fef95256 - correct
           Events : 177794
     Device Role : Active device 0
     Array State : A.. ('A' == active, '.' == missing, 'R' == replacing)

Slice f2 looks great, but slices j2 & l2 seem to be missing -- even though
they are present.  Worse, the Events counter on sdf2 is frighteningly
small.  Where did it go?!?  So maybe I consider sdf2 failed and reassemble
from the other two [only] and then put f2 back in?

Definitely time to stop, take a deep breath, and ask for help :-)


TIA & HAND

:-D
-- 
David T-G
See http://justpickone.org/davidtg/email/
See http://justpickone.org/davidtg/tofu.txt