Don't take it out on others just because you're imperfect and ignore all of the S.M.A.R.T. and controller warnings... Some of us tech Gurus do religiously tend to our flock of hard-drives and recognize when they are in spiritual, and physical, need of replacement....
I do hardware diagnostics for an HPC storage system vendor, including drive testing, qualification, and failure analysis. SMART has its uses, but, in my experience, if you've tripped SMART, you're already in serious trouble.
It's not helped by the fact that pieces of SMART which are actually in the ATA standards are basically tripped/not-tripped - none of the attribute structures are in the current specs, let alone which attribute IDs mean what. Heck, even getting the thresholds is no longer in the ATA standard!
Fortunately, *most* vendors implement *most* of the SMART structures the same way, and *most* use the same attribute IDs to mean the same things, and *most* still implement the sub-command for getting the thresholds. But all the really interesting data that could be used for more aggressive failure prediction (i.e. beyond the almost-always too late SMART trip) is vendor specific, and getting that information out of some vendors makes pulling teeth look like taking candy from a baby.
</rant-let>
-Ster
