[stgt] [BUG] Tgt-1.0.8 exited unexpectedly

[Hirokazu Takahashi]

Hello,

> > > > > > > I'm not sure the modern SATA disk can detect such failure.
> > > > > > 
> > > > > > I think the modern SATA disk has this feature while the IDE disk doesn't
> > > > > > have.
> > > > > 
> > > > > Do you have any pointer?
> > > > 
> > > > This may help you:
> > > > http://www.seagate.com/content/pdf/whitepaper/SerialATA_comparison_UATA_Technology.pdf
> > > > It says serial ATA adds 32-bit CRC error correction for all bits transmitted,
> > > > as opposed to only data packets in Ultra ATA.
> > > > And it is known that each sector of modern disks has extra bits for ECCs to
> > > > correct errors.
> > > 
> > > Hmm, this isn't same as what SCSI DIF (and enterprise storage) does to
> > > prevent silient data corruption. This handles only transmission
> > > corruption. So there is still a good chance that silient data
> > > corruption could happen.
> > > 
> > > SCSI DIF and enterprise storage maintain extra bytes per sector for
> > > checksumming to prevent silient data corruption.
> > 
> > I guess SCSI DIF is a feature which adds extra bytes to each sector
> > and its protocol allows software to control them.
> 
> Kinda.
> 
> 
> > I think each sector has another extra field which SCSI or SATA drives
> > internally use to correct errors.
> 
> Can you provide a pointer to the info that SATA drives internally keep
> extra bytes per sector for error correction?
> 
> That's what I asked you in the previous mail.

How about this document?
http://www.serialata.org/documents/S2Ext_1_2_Gold.pdf
This isn't the final one but section 4.7.2 will be informative.

The following page is also helpful. It tells why they want to introduce
4kb sector.
http://www.zdnet.com/blog/storage/are-you-ready-for-4k-sector-drives/731

I guess the way of correcting error is out of the specification of SATA
because the specification just define the interface.

Thanks,
Hirokazu Takahashi.
--
To unsubscribe from this list: send the line "unsubscribe stgt" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html