Recovered a dead RAID array with a $20 USB adapter and patience
Last week a small dental office called me in panic mode. Their server with two mirrored drives showed one disk as failed and the other was throwing SMART errors. I tried booting with their backup controller and it kept dropping the array during the rebuild. After 3 tries I pulled both drives and hooked them to my bench PC using a cheap USB SATA adapter I normally use for cloning laptops. Turns out the array metadata was fine on both disks, the issue was the controller firmware locking up on the staggered spin-up. I ran a manual reassembly with mdadm on Linux and copied everything off before the second drive fully died. Client paid me $400 after their last quote was over $2,000 for data recovery. Has anyone else had luck bypassing a bad controller instead of trusting the RAID card's own boot process?
Read a post on a data recovery forum a while back where a guy had the same thing happen with a Dell PERC card. The card kept marking drives as foreign and would not let him import them no matter what he did in the BIOS. He moved the drives to a plain HBA and the array came right up like nothing was ever wrong. The takeaway from that thread was to never trust the card that is already acting up, because the controller can be the actual problem and not the disks. Sounds like you ran into the same firmware bug and the USB adapter just sidestepped all of it. Good save on the copy too, catching the second drive before it gave out is the whole ballgame.