Re: [gentoo-user] Is my hard drive sick?

[Dan Farrell]

On Tue, 20 Nov 2007 23:25:06 +0100
"b.n." <brullonulla@×××××.com> wrote:

> Billy Holmes ha scritto:
> > Quoting "b.n." <brullonulla@×××××.com>:
> >
> >> I found your exact same symptoms, on a SATA hard disk. I frankly
> >> have troubles in understanding the smartctl outputs (googled of
> >> course, but
> >
> > while I think the seek time metric is mostly informational and not
> > really indicative of harddrive failure, my experience was different
> > and I'll explain why.
> >
> > * my harddrive made strange noises at times.
> > * it's seek time metric kept getting lower
> >
> > So, I guess what I'm trying to say is that you have to take all this
> > data in context.
>
> Thanks.
>
> My latest errors are:
> Nov 20 06:52:23 voynich smartd[7451]: Device: /dev/sda, SMART
> Prefailure Attribute: 8 Seek_Time_Performance changed from 247 to 246
> Nov 20 09:52:23 voynich smartd[7451]: Device: /dev/sda, SMART
> Prefailure Attribute: 8 Seek_Time_Performance changed from 246 to 245
> Nov 20 11:22:23 voynich smartd[7451]: Device: /dev/sda, SMART
> Prefailure Attribute: 8 Seek_Time_Performance changed from 245 to 248
> Nov 20 12:22:23 voynich smartd[7451]: Device: /dev/sda, SMART
> Prefailure Attribute: 8 Seek_Time_Performance changed from 248 to 247
> Nov 20 15:22:23 voynich smartd[7451]: Device: /dev/sda, SMART
> Prefailure Attribute: 8 Seek_Time_Performance changed from 247 to 246
> Nov 20 19:22:23 voynich smartd[7451]: Device: /dev/sda, SMART
> Prefailure Attribute: 8 Seek_Time_Performance changed from 246 to 245
> Nov 20 20:22:23 voynich smartd[7451]: Device: /dev/sda, SMART Usage
> Attribute: 195 Hardware_ECC_Recovered changed from 252 to 253
> Nov 20 22:22:31 voynich smartd[7451]: Device: /dev/sda, SMART
> Prefailure Attribute: 8 Seek_Time_Performance changed from 245 to 247
> Nov 20 22:52:23 voynich smartd[7451]: Device: /dev/sda, SMART
> Prefailure Attribute: 8 Seek_Time_Performance changed from 247 to 248
>
> so the Seek_Time_Performarce is kinda oscillating.
>
> I could buy another one and replace, but I fear early mortality. So,
> what should I do (apart from frequent backups)? Use this one as main
> until its errors becomes more serious and use the new one with
> non-essential data stuff to see if it works?
>
> By the way, a dd if=/old/disk of=/new/disk is enough to transfer the
> whole thing, isn't it?
>
> m.
it is; however if the partition is live the data could be messed up if
you read half of an overwritten file or something.  in other words, it
works really well on partitions that aren't live, and if the partition
is live, you could potentially have a bad file.  You might even
potentially have a bad partition.  This all is from my own
consideration; I wouldn't bet the house on it.  
-- 
gentoo-user@g.o mailing list