Re: [smartmontools-support]Is my ICH3M controller broken?

[Volker K. <lis...@pa...>]

Faulty IDE controller / bus hardware can't cause faults inside a hard
disk. Perhaps to the electrical interface of the hard disk, but I would
say this is very unlikely. Data loss of course is highly likely.

> well...I just started to notice that some time something makes an access
> to the disk even when the machine is compeletely idle, the disk light
> comes up for a fraction of second and I can ear a very light noise that
> I couldn't remember before tweaking with the hdparm parameters...

Several possibilities, but I'm not too familiar with all this.

> Sep 18 20:15:43 doroty kernel: hda: drive_cmd: status=0x51 { DriveReady
> SeekComplete Error }
> Sep 18 20:15:43 doroty kernel: hda: drive_cmd: error=0x04 {
> DriveStatusError }

This could indicate problems with data transfer between controller and
disk. Or a defective hard disk.

> 195 Hardware_ECC_Recovered  0x001a   100   100   000    Old_age   Always
>       -       123124

This is normal for many disks. Ignore this number, I don't see any
practical use in it.

> 225 Load_Cycle_Count        0x0012   099   099   000    Old_age   Always
>       -       10520

Does this have to do with waking up the disk from suspend, or similar?
With active power managament you would expect a high number of these.

> The older disk (the samsung one) had less than a year life, I had with
> it a very similar behavior

The lifetime of the previous disk is irrelevant. Disks are expected to
fail any time. Of more interest are reallocated bad blocks, bad blocks
count, offline uncorrectable. The first thing you do when you suspect
the disk might be on its way out is run a short self test. The second
thing you do is run a long self test.

Are you knocking your laptop about too much when the disk is spinning?

Volker

-- 
Volker Kuhlmann			is possibly list0570 with the domain in header
http://volker.dnsalias.net/		Please do not CC list postings to me.