• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 681
  • Last Modified:

Creating a RAID-5 volume

Good morning!

Recently, two disks out of our six RAID-5 disk array failed, and were replaced.  I am now trying to recreate the RAID-5 volume.  The server is split into two arrays, A+B.  A consists of two disks (one logical disk) for OS, and B consists of six disks (one logical disk) for data.  In the disk management console, there were two disks displayed.  Disk 0, which contains the OS, and Disk 1, which is Unallocated space.  I initialized and converted Disk 1 to a dynamic disk and rebooted.  When I try to create the RAID-5 volume on Disk 1, everything but "Simple" volume is greyed out.

Do I need more than one logical disk?  If so, how do I get Windows to view my disks in the array individually?  In ACU?  If you can't tell, this is my first time rebuilding a RAID array.

I have included screenshots of what I am seeing in ACU, as well as the disk management console.

Thanks!

Nick
screenshot3.jpg
screenshot1.jpg
screenshot2.jpg
0
helpdesk_ninja
Asked:
helpdesk_ninja
  • 2
  • 2
1 Solution
 
DavidCommented:
By design the O/S won't see individual disks.  The P400 RAID controller just presents the 2 RAID arrays.   This is what you want.  If you want to drill inside of the array and look at health, then use the app shown in screenshot #3.

Just treat the RAID arrays as single disk drives.   Your config is fine the way it is.  Just add a drive letter to disk1 (SAS array B) and put your data on it.
0
 
helpdesk_ninjaAuthor Commented:
Ahhhh, I get it!  So, if I DIDN'T have a 3rd party RAID utility, I would create my RAID array in the disk management console, but since the array is already configured in ACU, I can just treat it as one disk in Windows, and format that disk as an NTFS primary partition, assign it a drive letter, etc.  

Thanks so much!
0
 
helpdesk_ninjaAuthor Commented:
Thanks again!
0
 
DavidCommented:
So if you didn't have a HARDWARE RAID controller, then you would let the O/S do the RAID, which is called SOFTWARE RAID, and then you would use disk administrator to combine them..

But the light bulb went off, glad to help.   Note, it is vital, you need to run regular consistency checks on the RAID using the HP utility.  The check looks for any bad blocks or data inconsistencies and repairs them.    To make it easy to understand ... say you have RAID1 which is 2 drives with same data.   Block #123 fails on disk A and is unreadable.   The consistency check/repair discovers that, remaps a spare block from disk A, reads block #123 from disk B, and writes that value into block #123 on disk A.  

Reason why you want that, is what if disk B dies and you never did consistency check/repairs?   You would end up with an unreadable block, i.e, unrecoverable data loss.

If you have a free slot, buy a replacement disk NOW, stick it in, and make it a hot spare.  Then if any disk does die, the RAID engine will automatically use the replacement drive and do a rebuild.  Had you been using a hot spare before, then perhaps the RAID controller would have rebuilt the 1st failed drive in the RAID5 before you lost the 2nd disk.

Also finally, your O/S will never know if you have a drive failure and the RAID is degraded, so keep an  eye on ACU, and set it up to give you an alert in case of a drive failure.  All that is documented.
0

Featured Post

What does it mean to be "Always On"?

Is your cloud always on? With an Always On cloud you won't have to worry about downtime for maintenance or software application code updates, ensuring that your bottom line isn't affected.

  • 2
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now