Re: [gentoo-user] WARNING: Crucial MX300 drives SUUUUUCK!!!!

[Poison BL.]

On Sun, Feb 12, 2017 at 5:40 PM, Alan Grimes <ALONZOTG@×××××××.net> wrote:

> Dear god, I think I have come in contact with one of the suckiest things
> in the universe!
>
> I mean first there are supermassive black holes... OK... Then there's
> Crucial MX300 SSDs, and in a distant third there's Justin Beiber.
>
> I mean the absolute suckyness of MX300 SSDs defy human comprehension. I
> mean you could connect one of these:
> http://www.zmescience.com/science/biggest-most-poweful-engine-world/ to
> a suction pump and it couldn't possibly suck one quintillionth as much
> as this SSD...
>
> I mean if the power goes out, and you want to do some vacuuming, just
> put your MX300 behind the bag in your vacuum and it'll work better than
> normal.
>
> Seriously, what could possibly suck harder than a SSD which dies stone
> cold dead after only 20 days?!?!?! Thank god I had done nothing worse
> than store my rusty old Velociraptor on a shelf, and by good fortune it
> only took about two days to get it updated... I really hadn't intended
> to ever use it again. =\
>
> I'm not sure what lesson I should take away from this except that
> Crucial does not have any business selling SSDs. =\
>
> I'm not sure where to go from here. My 'raptor is very close to
> exceeding it's reliable lifespan, by some standards it already has...
> But now the QC of these SSDs has been shown to be outrageously bad. =(
>
> So does anyone have any evidence of a current generation SSD lasting
> more than 20 days?
>
> --
> Strange Game.
> The only winning move is not to play.
>
> Powers are not rights.
>


I've had all manner of drives fall on the leading edge of the failure rate
'bathtub curve', both SSDs and spinning rust (including hybrid drives), and
many more of each last far past what should typically be the tail end
'spike' that occurs on that curve (including a few of WD's raptor drives of
various vintage). A single drive failure is an anecdote, not an indicator
of a systemic failure of the entire production line (let alone brand),
especially if it lasted 20 days past install... which is well outside
anything a quick, every Nth drive, QA test on the production line is going
to pick up unless they happened to grab that single, specific, drive. As 20
days is also well inside the warranty they give on that drive, a
replacement's not likely to be difficult to get from the manufacturer (and
if you purchased it within the past 30 days, from typical vendors if you
prefer).

I don't *think* I have any of the MX series, but I have had good luck with
the one BX200 I have in my work desktop, and I've heard good things on the
handfull of BX series drives my boss's deployed in various desktops and
laptops. I haven't done any deliberate performance testing on them, but I
can attest to much better speeds out of my BX200 than the spinning 500GB
sitting under it in the same machine.

As for the failure *mode* of your drive, simply, completely, dead... that's
been my experience on every failed SSD I've seen, be it a samsung 850 pro,
an early crucial drive, corsair, and even intels. I suspect either the
controller itself is the point of failure on them, or it's simply incapable
of working around a failure of some other component.

Lastly, while drive failures do, very much, suck... they happen. That's why
backups are essential, and also exactly why raid levels outside of raid0
exist (one is not a replacement for the other, 'course).

-- 
Poison [BLX]
Joshua M. Murphy