One of my customers ran into the problem of the main VM host not continuing to boot further than the BIOS POST.
I had a hunch that some of the failed disks had been messing with the controller since we have been replacing several disks in that machine this year and the system had always rebooted so far.
However, this time the defective disk was the first one. So my guess is that the RAID controller assumed the first disk to be correct if it was present, thereby rendering all the working disks invalid which then lead to a massively degraded state.
Solution: Pull out the first disk and press