I see no more appropriate discussion "center" to submit this question; if there is, please let me know how to get there - or (Intel) - please resubmit on my behalf to the appropriate forum.
I'm working on a clone server with an Intel motherboard including a Intel 82801GR Raid Controller, on which one of the SATA drives failed. Upon replacement of the drive, the RAID bios/ post showed it as recognizing the drive and apparently then spawned a CHKDSK process. When that completed, it said it would restart the computer - but did not (after waiting 10 minutes for it to restart).
I powered cycled the machine, and after POST it proceeded to restart Windows - with Win 2003 server splash screen and progress indicator. Then it rebooted, starting a Windows boot process again - lather, rinse, infinitely (presumed) repeat. Attempting to restart in any version of safe mode only resulted in a reboot. Final clue to resolving was stopping auto restart upon boot failure - identifying that the software hive was corrupt -