[mythtv-users] OT: Seagate to reduce warranties on consumer drives to 3 years from 2009
Michael T. Dean
mtdean at thirdcontact.com
Wed Dec 17 21:38:46 UTC 2008
On 12/17/2008 03:14 PM, Steven Adeff wrote:
> On Wed, Dec 17, 2008 at 1:53 PM, Michael T. Dean wrote:
>> On 12/17/2008 01:46 PM, Steven Adeff wrote:
>>> does in fact make very very good drives, none of which have the
>>> problem that *one hardware revision* in the Deskstars had that gave
>>> them the Deathstar nickname. I owned some Deskstars back before the
>>> fiasco and they were some of the best drives I ever owned.
>> I have a Hitachi Deskstar that started clicking at least a month ago and
>> has a bunch of SMART read errors (bad sectors, IIRC), but hasn't shown
>> any signs of problems. I will admit that I've already bought the
>> replacement drive, but I haven't switched it out. Part of the reason I
>> haven't replaced it is because I'm really curious why it /sounds/ broken
>> and hasn't failed catastrophically, but within a month before my
>> noticing the Hitachi's clicking, I had 2 other HDD's (Seagate and WD,
>> FWIW) that failed catastrophically with no warning (SMART or sound wise).
> Last time something similar happened to me my power supply "died"
> shortly afterwards.
> side note to that: I ended up purchasing a slightly overpowered (I'm
> an EE I live by the 20% DD rule, and now for computers as well), high
> power quality PS and have noticed the system is also much more stable.
> Especially for those with heavy/large HD usage I can't stress how much
> this realization and my efforts since have affected the stability of
> my systems.
Funny enough, I've had 2 other HDD's die in the system (my main
desktop). Normally, that would scream PSU (or possibly mobo power
About 2 years ago, the first of the 3 HDD's died in that system.
However, I've since replaced /both/ the PSU and the mobo, but...
The first of the 3 HDD's that died recently /was/ in the main desktop
(that now has a different PSU). I bought a new 1.5TB HDD, put it in one
of my Myth backends and rolled the 200GB Deskstar down to the system
whose HDD died***. The Hitachi has been in my main desktop for slightly
over a month, now, and seems to have begun clicking shortly after put
into that system.
So, while the PSU is /not/ the same one that was in the system when the
first HDD to fail in that system failed, this is the 2nd HDD to fail in
that system with the current PSU. The system seemed to run well for a
couple of years with the (dirt-cheap, low-quality) PSU, but the PSU may
just be at the end of its life, which is why HDD's have just begun failing.
About a year ago--after getting the PSU that's in my main desktop--I had
a PSU failure on a day when newegg had a great sale on an Antec
Earthwatts PSU. So, I bought one (80 plus certified and all). It was
the first good PSU I ever bought, and I'm never going back to the cheap
So, I had been considering replacing the PSU. Your post has just
cemented the decision. (And by getting an 80 plus certified PSU, the
power savings may even make up for the cost of the new PSU since it's a
system that runs 24/7 when I'm not traveling). The unfortunate part,
though, is that I actually have 3 PSU's that are beginning to fail.
(I'm pretty certain the other 2 PSU's are failing because of the sounds
the fans make--like a large truck backing up (but without the
beeping)--and it's not constant or only on startup (it will just happen
at seemingly random times)).
***Interesting side-note: In the process of rolling the Hitachi down to
the main desktop, I also moved a 750GB HDD from a Myth backend to a Myth
frontend. Just after copying all recordings from the 750GB and while
copying the root partition of the frontend to the 750GB, the 750GB died
(so that one never touched the suspect system).
But the additional failure resulted in another 1.5TB HDD purchase and
rolling a 300GB HDD down to the frontend.
And, now, the clicking Hitachi has resulted in another 1.5TB HDD
purchase and rolling a 330 GB HDD down for the main desktop--but it's
not yet in there (as I was trying to determine if the failures are due
to something in the system).
More information about the mythtv-users