RAID5 failed on Dell server

Have a Dell Poweredge 2800 server with 3 x 72Gb UW scsi hard drives built around an on-board RAID5 array running SBS2003. Recently, on re-booting, it started produced a message along the lines of 'RAID degraded - press Alt-something to enter RAID config or any other key to continue'. Pressing any other key allowed it to boot & run ok. Today however, it refuses to boot. Looking for inspiration on whether the issue most likely lies with the drives or the RAID controller and, in each case, possible remedies. I understand a 3-disk RAID5 will run ok on 2 disks so one theory is that one disk failed some weeks back, hence the 'degraded' error described above and another disk failed today, leaving only one good disk. If this is the case, I've taken advice to effect that I'm screwed as RAID5 needs 2 out of 3 drives to function. Another possibility, and I can't rule this out, is that somebody took it upon themselves to go into the RAID config and mess with some settings.... in which case, does anyone know what the correct settings should be so I can check them? Final possibility is that the RAID controller is wrecked, in which case are we talking new motherboard? In this scenario, as a new board is likely to be expensive and may take some time to get hold of, what are chances of sticking all 3 drives in a new server & getting them to work?
laurencoullAsked:
Who is Participating?
 
jamietonerCommented:
I think someone has raid 5 and concatinated disks mixed. With a raid 5 on this controller the data(including mbr) is not on a single drive but spread out. With a concatinated arrary the mbr would be located on the first drive but not in a true hardware based raid 5. If you know which drive was the last to go offline run diags on that drive and if it passes force it back online, be careful if you force the wrong one online or force both it can corrupt the data to an unrecoverable state. Do you have a backup?
0
 
jamietonerCommented:
You're most likely boned. Action should have been taken when the array first became degraded. It maybe possible to recover the array, however it also possible by trying to recover the array it makes the data impossible to recover. I doubt ot the controller given the events bad hdds is more likely
0
 
shadowmantxCommented:
One thing you can try.  Power down the server and pull out the middle drive and reboot the server without the middle drive in the array.  See if it boots.  If it doesn't, but back in the middle drive and pull out the last drive and reboot.  Usually the MBR for OS is usually on the first drive and you have a 33% chance it is one of the other drives failed which is causing the whole array to fail.

So if it does boot, just slap a new drive in the bad slot and RAID will rebuild it.

Good luck.
0
 
laurencoullAuthor Commented:
Ah! Worth a shot! Ok let's be pessimistic & say the drive with the MBR is the failed one:
Would there be any mileage in building an identical OS on a new clean drive & sticking it in as the bootable drive?
0
 
shadowmantxCommented:
It's unorthodoxed it could work.  You would have to setup the whole array with all three new drives and replace the first drive in your old array and see what happens.  Make sure the new drives are the same size as your original drives.

Wish you luck. :-)
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.