gpg: failed to create temporary file '/var/lib/lurker/.#lk0x58009100.hantslug.org.uk.8573': Permission denied
gpg: keyblock resource '/var/lib/lurker/pubring.gpg': Permission denied
gpg: Signature made Tue Apr 22 17:43:16 2008 BST
gpg: using DSA key 20ACB3BE515C238D
gpg: Can't check signature: No public key
On Tue, Apr 22, 2008 at 05:01:41PM +0100, Richard Brown wrote:
> On Tue, Apr 22, 2008 at 4:58 PM, Hugo Mills <hugo@???> wrote:
> > Does anyone actually have the reference to this paper? What journal
> > was it published in, for example?
>
> http://research.google.com/archive/disk_failures.pdf
So, in summary:
- There's a fairly high incidence of "infant mortality": new (<3
months old) drives fail more often than older ones.
- High utilisation of drives only has an effect at the very young and
the very old ends of the spectrum.
- High temperature is only a factor in drive failure for old drives
(>3 years old)
- Low temperatures (<25degC) are correlated with significantly higher
failure rates.
- After receiving the first SMART error for scan errors,
reallocations, offline reallocations and probational sector counts,
the probability of the drive failing within 60 days is much higher.
However, it's still not a good predictor for drive failure.
I've probably missed a few bits, but those are the ones that stood
out.
Hugo.
--
=== Hugo Mills: hugo@... carfax.org.uk | darksatanic.net | lug.org.uk ===
PGP key: 515C238D from wwwkeys.eu.pgp.net or http://www.carfax.org.uk
--- Summoning his Cosmic Powers, and glowing slightly ---
from his toes...
