08-21-2008, 12:11 AM
|
 Re: Seagate 7200.11 High Failure Rate
wrote in message
news:rvrh94hf451ggitqvhqj49b3b4fekifk15@4ax.com...
>I bought a 2.5" 160GB for my laptop.
> Had a problem. Ran Seatools, showed nothing, called Seagate.
> They couldn't figure it out, said drive must have a problem.
> Tried running Spinrite - it showed the drive making constant seek
> errors. Got a replacement. this one is making lots - not constant -
> seek errors. Seagate just blows it off and says we don't support
> anything but Seatools and it doesn't tell what the seek errors are. if
> it fails u have 5 years to replace it. I said - Ya, but that
> translates into a slow drive - he couldn't understand that so I gave
> up. SO seek errors are not important to Seagate.
>
> THEREFORE, Seagate sucks. Don't know if others are better - the orig
> drive had NO seek errors and one ECC.
I learned something this week about S.M.A.R.T. As well-meaning as "the
specification" may have been at it's conception, the way it is implemented
today is highly proprietary to individual manufacturers. It is very hard for
other parties (read, software developers of hard drive monitoring software)
to decipher the meaning of these values unless they figure it out some way
on their own. Drive manufactures HIDE their S.M.A.R.T. implementation
details. I too listed "seek errors" as one of the things I noted on failing
drives, but I got that info from a software program that didn't come from
the manufacturer. That program cannot assess the data across all drives
because they don't have the S.M.A.R.T. value meanings worked out. I tend to
believe that the "reallocated sector count" is valid (just a hunch), but
think that "seek errors" probably isn't, because I can watch on of my drives
accumulate seek errors as I watch the program, but the drive continues to
run fine (for over a year).
S.M.A.R.T. is a good start in theory only. In practice, being proprietarily
implemented by manufacturers, it sucks! I have read that S.M.A.R.T. can
detect or predict 70% of hard drive failures. But drive manufacturers don't
want anyone analyzing the operation of their drives so they HIDE their
implementation details of S.M.A.R.T. and what could be something of greater
value becomes a source of confusion.
The above posters concerns may very well be invalid then, but drive
manufacturers are, IMO, schmucks for not divulging or for purposely
obfuscating their S.M.A.R.T. implementations.
Tony
 |