We ship RAID1 systems to our customers (two drives, using Win7 dynamic disk)
We thought (perhaps naively) that this would protect a customer's data if one drive fails.
We had a system in the field that would no longer boot - bluescreen, flash, restart, repeat.
No idea of root cause. We pulled both drives, and they are both exactly the same, with the same set of corrupted files (windows fails to copy many of them, chkdsk fails to fix them). I was very surprised by this, I would have thought if "junk" were written to one, it wouldn't propagate, but I guess that was incorrect thinking.
I believe the RAID1 configuration would have "worked" if one drive had physically failed, but I suppose there is a big difference between a physical failure and data integrity. How could I have prevented this scenario where the filesystem on both drives was effectively corrupted at the same time?