Two hours later, another status update.
I plugged in the second RAID drive from the SIL controller to the
nVidia. Same errors are when setup in RAID1. It failed to boot.
Since it was on the nVidia controller, Barts PE could see the drive.
I pulled it up in file explorer and could see the data. Good sign.
I ran chkdsk and scanned the entire drive. Many many lost chains and
inconstancy's. Over a thousand.
I ran HD Tune against it and it reported back no errors. This is a WD
drive, and BE shows 45. HD Tune says "failed", however that is the
temperature return per Wikipedia. The other temperature is 45C so I am
guessing this is a false positive error on HD Tunes part as HD Tune
doesn't know what BE is and is guessing.
HD Tune error checking (slow) showed no errors. The performance tune
was very quick. Not like what you normally have on a failing drive
with redirected sectors. I am posting this post by booting from the
second drive.
So apparently what we have here is a failure of the SIL software RAID.
I've seen this with other software RAID before. With hardware RAID, a
data error on the "mirror" drive would occur it would simply issue a
rebuild and either rebuild, or orphan. If it were the primary drive,
then both drives would have the data error since we are mirroring data.
With software RAID, the software couldn't apparently fix the data
problem.
Now the next step. Hook up the drive that didn't have the data
failure, the primary in the RAID set, and make sure I have it as drive
A (0 or 1 depending on how you count) in the SIL controller. Make
sure it boots. Then reboot and hook up the second drive and rebuild
the array. With any luck I'll be back in business.
What a way to spend 8 hours

As long as I don't loose the drive,
that's fine. I'll be making a complete backup after all of this is
done..