Re: [gentoo-user] smartctrl drive error @60%

[Alan McKinnon]

On 26/06/2014 05:54, Dale wrote:
> Curious.  I hope I don't start a flame war here.  I have had WD, Seagate
> and I think there is a Samsung here somewhere, may be the one that is
> rolling over on its back now.  The one drive that failed a few years ago
> was a WD drive.  That said, all the other WD drives I have had just got
> to small to really use, and slow when SATA came out.  I'm partial to WD
> and Seagate still since I got good long term use out of those.  Based on
> your experience, you tend to be of the same opinion?
>
> Allan, your situation should involve a lot of hard drives.  Any
> thoughts?  Neil, you have a nice big opinion on this?


My experiences aren't worth much in this case, what I had to deal with
was data center setups where

- the power has never gone off for 6 years
- the drives never spin down and just keep on turning year after year
- the servers were the nice big ones Dell makes with awesome cooling
- the data center feels like a fridge and the ambient temp never varies
more than 1 deg
- the server power supplies are seriously high grade, the 5V and 12V out
of them are solid and do not fluctuate at all

Add all this up and it's an almost perfect environment for drives to
last a long time. You don't have that, not even close.

I have only 1 little bit of anecdotal data:

my nas at home has 4 x 3T WD Green drives in it, going on almost 2 years
now. My kids hammer the blazes out of that thing, and ZFS scrubs keep it
real busy when the kids don't. And those drives just keep on turning and
turning and turning, I didn't do anything special. I put it down to
statistics - no-one makes bad drives (or cars) these days and I haven't
pulled the unlucky card yet. I dunno, go figure

-- 
Alan McKinnon
alan.mckinnon@×××××.com