I found some interesting log entries that did show the issue is locking, and I also see a whole bunch of scsi commands being retried and rejected or reset. Unfortunately I'm not a scsi expert so I can't easily decode what's going on here or why. I'm attaching a link that shows the relevent portions of the vmkernel log and the hostd log. The relevant time period starts at about 06:17:25 when I started VM SBS2008. About 9 seconds later is when the lost access message shows in the hostd log, although meanwhile in the vmkernel log are a whole bunch of scsi commands hanging and being reset. Maybe somebody more fluent in scsi protocol can give me some direction as to what's going on. I'm still very skeptical that it's a hardware problem in that a disk is bad because of all the stuff I mentioned before, but maybe there is some sort of scsi command being issued here that causes the controller a problem.