• 1 reply
  • Latest Post - ‏2013-08-14T15:36:08Z by rk1981
27 Posts

Pinned topic Datapower Xi52 RAID Drives

‏2013-05-10T15:41:36Z |

We have a XI52 with the four drives configured in a RAID 10.  One pair of drives is in a down state but the logical drive is still up.  The problem is none of the RAID drive data is available to us. 

I can't find anything on breaking the RAID drive to use only the good pair drives.  i assume that I can remove the bad pair and continue with the logical drive using only one pair.

IBM support is mailing new drives but it seems like we should be able to break this grouping and run with only one pair in the logical drive and see the data as they are delivered.  Anyone else have instructions on modifying the RAID configuration and how it might impact the device?  

I am curious to see if I can do this without restarting the box but there doesn't seem to be easy to find directions in this area.


Updated on 2013-05-10T15:49:55Z at 2013-05-10T15:49:55Z by Kevin005
  • rk1981
    1 Post

    Re: Datapower Xi52 RAID Drives


    In case you did not find the answer yet-

    XI52 boxes are equipped with RAID10 which means that data striping happens first at top level (RAID-0) and then mirroring at the bottom level (RAID-1). If you see in RAID Physical Drive status - There will be 4 disks, HDD0 and HDD1 are part of array 1 and HDD2 and HDD3 are part of array 2. HDD1 is a mirror of HDD0 and HDD3 is a mirror of HDD2.

    When you said that one pair of drives is in down state, I'm assuming they are part of same RAID-1 Array(either HDD0 and 1 or HDD2 and 3). In such a case RAID-0 won't work because it needs at least one in each array as it stripes the data across two groups of RAID-1. 

    If HDD0 fails in one array and HDD2  fails in second array,  you would still be able to access the data since you have the mirrored disks available (HDD1 and HDD3). I'm not sure though whether you can replace those disks without having to restart the box. 

    Updated on 2013-08-14T15:37:13Z at 2013-08-14T15:37:13Z by rk1981