Link to home
Start Free TrialLog in
Avatar of Bad_Fish
Bad_FishFlag for United States of America

asked on

RAID 10 Problem

Greetings,

Please see the image located at...

http://www.screencast.com/users/TPreston/folders/Jing/media/f1358857-d346-4d8b-ace9-8ccba04cc9a2

Array_0000 is a RAID 10 array. The fourth disk in the array (SN: 0708) has a status of "Missing," yet the drive is also displayed as an independent drive (Internal Disk) with a status of "Normal." On boot up, the BIOS indicates (in bright red letters) that drive 0708 is an "Offline Member."

After receiving this message, I downed the server and replaced drive 0708 with a new drive. But when I attempted to start the server, I received the "Missing Operating System" error and the server would not boot. I then removed the new drive and returned drive 0708 back to its rightful place...so we are up and running. But I would very much like to resolve this problem.

I should also mention that when I click on the "Internal Disk" (drive 0708 with the "Normal" status), there is an option to "Clear and Reset" the disk. But online Help indicates that doing so will "permanently delete existing metadata" on the disk. And since I can't even boot the server without that disk, I'm hesitant to do that.

Has anyone here encountered this problem and, if so, how can I safely resolve this?

Thanks,
Tom
Avatar of Rob Wesley
Rob Wesley
Flag of Canada image

Does your Intel RAID controller have the Rebuild Checkpoint Failure feature?

Is there an option to rebuild anywhere?

What's the RAID Controller product number?


Rob
Avatar of Bad_Fish

ASKER

To the best of my knowledge, it has no Rebuild Checkpoint Failure feature. And no, the only "option" I see is "Clear and Reset Disk" (on the Internal Drive).

The RAID controller is the Intel ESB2 SATA RAID Controller.
For the sake of clarity, I'm including two more screen shots. This one shows the status of the errored drive...

http://www.screencast.com/users/TPreston/folders/Jing/media/497387c1-4bfb-4945-88cd-2161e86a187b

And this one shows the status of the "Normal" drive...

http://www.screencast.com/users/TPreston/folders/Jing/media/bab615e2-89fb-4760-a04f-afb6275bf3bf

Thanks for your assistance...hope this helps.
I don't believe the Clear and reset Disk will make a difference, because that from what I've read will only return the disk to a normal state, which it is in already.

Is the model number of your new drive and failed drive identical? I'm just thinking that if the drive is of a slightly different model then that could be the source of your problem. A potential solution might be to upgrade the firmware on your RAID controller BUT call Intel to confirm. My first thought was to attempt the Clear and Reset disk option which I would do but not without first having a BACKUP in place.

Does Array_0000 have the OS on it?

I'd of course strongly suggest while in an operative state to ensure full backups are carried out.

Ok I don't like this RAID controller!!

This is what I would do if say I inherited this server and if certain resources were in place. I'd buy another server that has a better RAID solution and storage capacity required. The RAID would be RAID 1 with hotspare and perhaps another RAID1 with hotspare on another controller. I'd transfer all necessary server items and user files. Then glady take that server out of production.

Rob
Yes, the OS is on this disk...but I thought the whole purpose of RAID was redundency...and the ability to salvage things...like an OS.

Really? There's no way to "fix" this other than building a new server? What about the other three drives in the array? Shouldn't they also have the OS? There must be a solution.
ASKER CERTIFIED SOLUTION
Avatar of Bad_Fish
Bad_Fish
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
After researching the problem, I found the solution at the Intel website.

http://www.intel.com/support/chipsets/imsm/sb/CS-021234.htm.