I am new to this problem. Sure hope there is help out there :-). I hope I picked the correct forum.....
Server with S3200H M/B running 5 identical SATA 1.5GIG drives in RAID 5 sliced into two volumes. One is 1TB (bootable) and the other is just over 4.4TB. The OS is Server 2008 R2 CORE and the box is a Hyper V host. The OS resides on the 1TB volume.
This RAID 5 array has worked flawlessly for 4 years. Never an issue replacing several failed drives over the years. Recently, I lost a drive on port 0. Swapped in a new identical replacement and accessed the RAID setup using CTRL-I. Selected the new drive and started the rebuild. Booted into the OS, my Hyper V clients loaded and all seemed well. It usually takes 5 to 6 days to rebuild. I am able to check status using Intel's RAID command line utility by RDPing into the box via 2008 R2 Core.
After 30 hours, volume 1 (1TB) was rebuilt and the status changed to "NORMAL" After about 4 days, volume 2 reached 70% of the rebuild ... and that is when the fun started. Sometime after it hit 70% the rebuild percentage changed/jumped back to 0% and we started all over again. Using the command "RAIDCFG32 /ST" I am able to see that all 5 drives are good "members" (and green) ... no drives missing (except now the original bad/missing drive no longer shows up). The second volume continues to say "updating" and volume 1 "normal". I rebooted the box, checked the RAID array using CTRL-I and all looks like it should... except that volume 2 continues to rebuild/update. This has happened twice now... and we are into the 3rd "restart" back to 0%. This is like the movie Groundhog Day!
I find it very odd that volume 1 remains normal, while the rebuild of Volume 2 keeps recycling. Makes me think this is not a hardware issue. Or, do I pull the new drive I put in on port 0, replace it again and see what happens????
The Driver version is 126.96.36.1994, OROM version is 188.8.131.527 and the app version is 184.108.40.2063.
It would be extremely difficult for me to copy everything off and just start over with a new array (one of the Hyper V clients is the PDC and the other is a file share).
I am hoping there is some underlying issue that will be obvious to folks with RAID experience. PLEASE.....:-)
Any thoughts on how to get my volume back to "normal" would be greatly appreciated.
Since you have been able to rebuild the volumes before with no issues, I would consider replacing the new drive to compare results. It is possible that there might be corrupted metadata and this could be the reason why it always restarts at the 70% mark.
If the issue remains after replacing the new drive, I would recommend contacting http://www.intel.com/p/en_US/support/contactsupport Intel Customer Support for proper follow up of your case.
Is it possible to update the driver? Would that help?
Since one of the volumes has already completed a rebuild, I would be concerned that pulling that drive again might cause one or the other volumes to fail. Is that not something I should be concerned with? Any thoughts on that?
It is possible that a driver update might help resolve this situation. However, I noticed you are already running with the most recent version for your controller.
I do understand your concern about swapping the drive as your Volume 1 is optimal. Because of this, I would contact http://www.intel.com/p/en_US/support/contactsupport Intel Customer Support to expedite the resolution of your case.