On Sat, Aug 10, 2019 at 07:09:36PM -0400, fsmithred wrote:
> On 8/10/19 4:32 PM, Hendrik Boom wrote:
> > On Sat, Aug 10, 2019 at 03:11:03PM -0400, fsmithred via Dng wrote:
> > >
> > > >
> > > > Perhaps that disk is defective and I have to replace it.
> > > >
> > >
> > > If you have smartmontools installed, you can check:
> > >

> > >    smartctl -a /dev/whatever

> >
> > Im not familiar with these statistice, but it Looks bad:
> >

> >    7 Seek_Error_Rate         0x000f   084   060   030    Pre-fail  Always       -       324736547

> >
> > and
> >

> > 195 Hardware_ECC_Recovered  0x001a   071   056   000    Old_age   Always       -       204030548

> >
> > Is that a lot of errors/
> >
> > -- hendrik
> >
>
> I think it's normal to see high numbers of errors. I have more Seek errors
> and fewer Recovered than you on a disk with 75000 hours.
>
> Check to make sure the Reallocated Sector count is zero. If it's not, you
> should probably start moving data off the disk.

  5 Reallocated_Sector_Ct   0x0033   098   098   036    Pre-fail  Always       -       104

Count not zero. Time to replace the disk.

And to reconfigure whatever makes grub-mkconfig or grub-probe or
grub-install think that's the disk to boot from. I have no memory of
what I did to configure that in the first place. /dev/sdc shoud be a
nearly identical BIOS disk I could use instead. Not to mention the two
GPT disks.

As for moving data off, the crucial partitions are part of a RAID, so
the data are already duplicated elsewhere. But it's time to review the
contents to see what I realy need. It contains (at least) some home
directories from now-obsolete versions of Debian.

And its main function nowadays is to be the disk I boot from.

>
> Not sure what else would cause the input/output error you got. Bad cable
> maybe?

I don't know either. But I have a vague memory of problems like this
once before when I did not have time or energy to investigate properly.

-- hendrik