Re: Running HGST's DFT utility from a flash drive
Celejar wrote:
> On Wed, 21 Oct 2020 13:00:11 -0400
> Dan Ritter <dsr@randomstring.org> wrote:
>
> > Celejar wrote:
> > > On Wed, 21 Oct 2020 11:15:40 -0400
> > > https://wiki.archlinux.org/index.php/Flashing_BIOS_from_Linux#Using_a_FreeDOS-provided_Disk_Image_+_USB_stick_on_Linux
> > >
> > > and made a FreeDos installer USB stick and put DFT on it. It actually
> > > runs (although it skips the opening menu that comes up when booting the
> > > disk image itself (via PXE or the UBCD image)), but it fails to find
> > > any disks on the system :(. I don't know if that's a problem with
> > > the way I'm running it, or some problem with my system configuration
> > > (UEFI / BIOS, etc.) I tried changing the SATA controller mode from RAID
> > > to IDE, but it didn't help.
> >
> > A disk behind a weird SATA controller that sometimes thinks it's
> > a RAID controller? That's going to be problematic.
>
> This is an HP Z440 workstation. I don't know much about these things,
> but this is apparently standard HP design:
>
> https://support.hp.com/us-en/document/c02961221#AbT1
> http://h10032.www1.hp.com/ctg/Manual/c04823811 (p. 108)
OK, it's an Intel FakeRaid. This will work optimally with Debian
in AHCI mode.
But I suspect that we now know why the DFT utility can't see it:
because DOS can't see it, because DOS doesn't have any SATA
support. The best that can be done is providing a virtualized
interface, and that won't let you interrogate the disk.
> > Let's go back to: what are you trying to accomplish? Do you have
> > a new disk that you're trying to qualify in some way, or an old
> > disk that may have failed and you want to send in for warranty
> > replacement, or what?
>
> A "refurbished" HGST Ultrastar, almost certainly a server pull,
> purchased on eBay. Originally, I just wanted to verify the health of
> the disk, and to do so "by the book" (I can always just run badblocks
> and call it a day), but now I'm intrigued, I suppose, by the
> challenge ;/
I think you've come to the end of this road. Do run badblocks
and check it out that way. And have smartctl run a long test.
-dsr-
Reply to: