Running HP ML330 with Smart Array Controller 641 and 3 Non Hot Swap Disks in RAID 5 Array. Got a warning that there was a predictive failure on 2 of the disks!
The next time I turn the Server on the RAID Controller said one of the disks had failed. I then replaced that disk ensuring all SCSI id settings etc were the same. On boot the RAID card Detected the replacement and I pressed F1 to rebuild. The Server then started to boot into Windows 2000, then I got blue screen! Say boot errors and to run chkdsk.
I rebooted and have gone into the Raid controller where I can see the 3 disks, 1 of them being the replacement, and the other 2 marked as OK. The Status says recovering.
1. Are the disks actually recovering whilst looking at the RAID config screen? Or do I have to come out of this for the disks to continue there recovery? ie if I leave the server for the next approx 8 hours will the disk rebuild or am I wasting 8 hours?!
2. Is it OK to run CHQDSK disk from a command prompt, ie boot up with CD/floppy disk, on a raid 5 array or will this trash it?
It sounds to me like you have some sort of more serious problem, because Windows should be able to boot and run fine on a RAID 5 array even while it's rebuilding. It will be SLOW, but it should work.
CHKDSK won't help - it only works on the file system, not the underlying RAID configuration.
The Smart Array Controllers have their own processors and should rebuild the array as long as the system is turned on and power is applied to array card. It should be easy to confirm this by seeing if the drive lights are flashing.
You could try waiting for the array to complete it's rebuild, but I have a feeling it won't solve the problem...