Link to home
Start Free TrialLog in
Avatar of metalfubar
metalfubar

asked on

Raid 5 array puncture

I have an sbs 2008 box running on dell T310 hardware.  Raid 5 with 3 drives with 250gb 7.2k drives, 8 gb ram.

I'm seeing one drive offline in OMSA (running in degraded mode on two drives currently) and puncture errors in the controller logs.  I also cannot get a valid backup using symantec or nt backup.

I contacted Dell and they recommended replacing the drive, backing up and restoring from backup after deleting and recreating the array.

I'm wondering if there are any 3rd party tools, steps t take to either fix the puncture or get a valid backup for restore.

I'm running exchange and db are on c drive but under 10gb of space.

chkdsk does not work but I'm wondering if verify disk media is an option?

thanks much.
ASKER CERTIFIED SOLUTION
Avatar of David
David
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of metalfubar
metalfubar

ASKER

thanks for your reponse Dleth,

My client doesn't have money from any recovery tools...

I think my only opion is to copy as much data over as possible, export psts, system state, files, shares to a clean usb drive.

Install the new drive, create a new array.

Install sbs from scratch and then import.

Can  you elaborate on the steps needed so I can complete asap?

thx again.
Do steps 1-3.  The runtime.org site elaborates in great detail.  But basically you hook the drives into a non-RAID controller and it reconstructs the array into a single disk.  Then if you want, you can replace the bad drive in the RAID controller, build a new empty RAID5 target, and then clone the reconstructed array from the single disk back onto the RAID controller.

Note that the software is free to try.  But do remember, you have lots of damage, so don't expect to get everything back.
Thanks for both posts...

This is a really good article and explains puncturing well.

the fix is for perc 3, 4 controllers however while I have a perc 6i.

Should the process be the same?

another thought was moving the array to a single drive (hot spare) and then moving it back to a rebuilt array.
what about something like this with the puncture scenario?

1. I install 1TB hot space into server. Connect it to SATA port on the RAID controller that my RAID 5 was running on.
2. Boot PC, Run Acronis True Image 2012
3. Run Clone Disk in manual mode. Select source disc. existing raid 5 C/D: drive.
4. Select Target disc as new drive.
5. Hit start button.
6. C: Drive partion is locked and server reboots
7. Acronis TI starts without booting into windows. Clone process starts
8. About 1 hour it's done and press any key to power off server.
9. With PC off unplug raid 5 array but leave new Cloned disc on same RAID controller. (do not go into motherboard and disable raid)
10. Start server. server boots to windows. done.
11. Tuned of server. Plugged in 3 of disks that were part of raid 5 array
12. Reboot, Delete the raid 5 array. Create new raid array.
15. done.
If you can set it to ignore disk errors,it should work.