[svlug] fsck on HD booted from CD???

[Nick Austin]

On Mon, Jan 29, 2007 at 05:08:51AM -0800, Daniel Gimpelevich wrote:
> > badblocks will give you a definitive answer as to the health of your disk.
> 
> There was a day and age when badblocks did something useful. Those days
> are long gone.

Why do you say that?

Badblocks does more then that dd you asked Skip to run. It will read and write
data and make sure the drive works in both directions. Not only that, it's
a great way to have your IDE drive remap a few bad blocks if they exist.

What tool would you use to make sure the drive can read back what it's written?

Why not use badblocks instead?

> 
> > sure that the efficacy of this tool is very implementation dependent.
> 
> I have had the health checks pass right up until the drive failed, but
> they sure didn't pass after it failed.

It would be interesting to see if there is data somewhere about SMARTs ability
to predict failures, or even identify a failed disk. The drive I used it on was
a 120G Western Digital, in which it did neither.


From some quick research I was able to do, it looks like there might be some
value in SMART signals, but not nearly as much as you would expect.

From:

E. Pinheiro, W. D. Weber, and L. A. Barroso.
Failure trends in a large disk drive population.
In Proc. of the FAST '07 Conference on File and Storage Technologies, 2007.

   Our results confirm the findings of previous smaller population studies that
suggest that some of the SMART parameters are well-correlated with higher
failure probabilities. We find, for example, that after their first scan error,
drives are 39 times more likely to fail within 60 days than drives with no such
errors. First errors in reallocations, offline reallocations, and probational
counts are also strongly correlated to higher failure probabilities. 

Despite those strong correlations, we find that failure prediction models based
on SMART parameters alone are likely to be severely limited in their prediction
accuracy, given that a large fraction of our failed drives have shown no SMART
error signals whatsoever. 

This result suggests that SMART models are more useful in predicting trends for
large aggregate populations than for individual components. It also suggests
that powerful predictive models need to make use of signals beyond those
provided by SMART.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://kenny.linuxmafia.com/pipermail/svlug/attachments/20070129/151f8c3a/attachment.pgp