Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Discuss] Is this bad?



On Fri, Feb 24, 2012 at 2:46 PM, Edward Ned Harvey <blu at nedharvey.com>wrote:
>
> ECC is a different story.  If you have ECC ram, and you're getting errors,
> then yes I suggest investigating.
>
>
Ed I think Derek is referring to ECC corrections on the hard drive.

Modern drives have the data crammed so tightly and in order to get the most
speed out of them there is a DSP chip that is reconstructing the signal
that the heads are reading.  This reconstruction doesn't always work and
that's when the ECC kicks in.  It's perfectly normal for even healthy
drives to have a certain amount of ECC happening.  In fact that number
being reported by SMART is a RATE and not the number of ECC events that the
drive has seen in it's lifetime.  The only time to worry about this value
is if the ECC correction rate starts to rapidly go up.

I agree with Ed here smart data, at least if you're just looking at a
snapshot, is pretty meaningless.  It becomes more meaningful if you're
looking at these counters over time and tracking the tends.  But even then
they are hard to interpret and the data reported varies by manufacture.
 The the only smart counters that I tend to pay attention to
are "reallocated sector" or "reallocation event" counters.
--
David



BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org