disk kaput?

[bad sector]

2.5" wd spinner plugged-in raid tray on Legacy desktop
(might have been sliced on the EFI laptop, don't remember)
=======================================================

Disk /dev/sdd: 931.51 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: WDC WD10JPVT-00A
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos
Disk identifier: 0x8efc6e65

Device Boot Start End Sectors Size Id Type
/dev/sdd1 256 244190645 244190390 116.4G 83 Linux



resliced to 1 full size primary partition on Legacy desktop:
-------------------------------------------------------------
Disk /dev/sdd: 931.51 GiB, 1000204886016 bytes, 1953525168 sectors
Disk model: WDC WD10JPVT-00A
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos
Disk identifier: 0x8efc6e65

Device Boot Start End Sectors Size Id Type
/dev/sdd1 2048 1953525167 1953523120 931.5G 83 Linux


Then plugged-in to t480 EFI laptop via sata/usb adapter:
.........................................................
Disk /dev/sdb: 931.51 GiB, 1000204886016 bytes, 244190646 sectors
Disk model: JPVT-00A1YT0
Units: sectors of 1 * 4096 = 4096 bytes
Sector size (logical/physical): 4096 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos
Disk identifier: 0x8efc6e65

Device Boot Start End Sectors Size Id Type
/dev/sdb1 2048 1953525167 1953523120 7.3T 83 Linu


is this device toast?

[David W. Hodgins]

According to https://www.storagereview.com/review/western-digital-scorpio-blue-1tb-review-wd10jpvt
the JPVT-00A1YT0 drive does use a 4096 byte physical sector size.

The question is why one system is seeing a 4096 byte logical sector size while
the other is seeing a 512 byte logical sector size. I'd check to see if the
t480 has a firmware update available to handle the larger logical sector size.

Regards, Dave Hodgins

[Carlos E.R.]

Why?

If you refer to these:

Sector size (logical/physical): 512 bytes / 4096 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes

Sector size (logical/physical): 4096 bytes / 4096 bytes

That's because the USB chipset has done that transformation. Also the
disk model changes. The computer does no longer see directly the disk,
but the view the USB chipset offers.


Garbage that usb-sata thing, it has even modified the apparent size of
the partition. If you use the disk that way, result will be corruption.

4096÷512 = 8.

Maybe fresh partition the disk on the usb thing would work, but then you
can never plug it directly again. Tape a notice about this on the disk.

--
Cheers, Carlos.

[bad sector]

I had separated the thread but it looks like the're both
about the same problem




With the 2tb extra data drive directly in raid rack:


# fdisk -l
_______________________________________________________
...
GPT PMBR size mismatch (488378645 != 3907029167) will be corrected by write.

Disk /dev/sdc: 1.82 TiB, 2000398934016 bytes, 3907029168 sectors
Disk model: WDC WD20EZBX-00A

Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos

Disk identifier: 0x00000000

Device Boot Start End Sectors Size Id Type

/dev/sdc1 1 3907029167 3907029167 1.8T ee GPT

Partition 1 does not start on physical sector boundary.
----------------------------------------------------------




One curve above is that the GPT mismatch' blurb at the top
is printed directly below the previous disk data making me
think that it applied to THAT disk, wrooooong.


The same 2tb extra data drive plugged into desktop via
the sata/usb adapter:


# fdisk -l
_______________________________________________________
Disk /dev/sdc: 1.82 TiB, 2000398934016 bytes, 488378646 sectors
Disk model: EZBX-00AYRA0

Units: sectors of 1 * 4096 = 4096 bytes
Sector size (logical/physical): 4096 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes

Disklabel type: gpt
Disk identifier: 35C27F01-CE38-43A0-8A2E-CB4EDC5B9918

Device Start End Sectors Size Type
/dev/sdc1 256 488378623 488378368 1.8T Linux filesystem
--------------------------------------------------------

THIS last connection WITH the sata/usb adapter gives the
same result on BOTH machines because I had sliced the disk
while it was connected to the laptop in this manner. I had
booted the TW live or rescue frisby to make the backup. I
can mount it and get at its guts which happens to be a 1tb
image made of the entire original bundled windows whorehouse:

# mount /dev/sdc1 /0/sc01
/home/u3 # ls /0/sc01
Screenshot_20240207_013830.png
lenovo-t480-2024-02-06.dd
lost+found

Thanks, you at least pointed me in the right direction, though
I will have to keep the crappy adapter if I wanna ever recover
the laptop from that dd image file, or make another copy of it
to another disk that is NOT hen tconnected via a sata/usb adapter.


Are some adapters better than others?

[bad sector]

On 2/10/24 13:24, Carlos E.R. wrote:

I had separated the thread but it looks like the're both
about the same problem




With the 2tb extra data drive directly in raid rack:


# fdisk -l
_______________________________________________________
...
GPT PMBR size mismatch (488378645 != 3907029167) will be corrected by write.

Disk /dev/sdc: 1.82 TiB, 2000398934016 bytes, 3907029168 sectors
Disk model: WDC WD20EZBX-00A

Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos

Disk identifier: 0x00000000

Device Boot Start End Sectors Size Id Type

/dev/sdc1 1 3907029167 3907029167 1.8T ee GPT

Partition 1 does not start on physical sector boundary.
----------------------------------------------------------




One curve above is that the GPT mismatch' blurb at the top
is printed directly below the previous disk data making me
think that it applied to THAT disk, wrooooong.


The same 2tb extra data drive plugged into desktop via
the sata/usb adapter:


# fdisk -l
_______________________________________________________
Disk /dev/sdc: 1.82 TiB, 2000398934016 bytes, 488378646 sectors
Disk model: EZBX-00AYRA0

Units: sectors of 1 * 4096 = 4096 bytes
Sector size (logical/physical): 4096 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes

[Paul]

That's an adapter trick to get around the 2TB limit with MBR.

https://superuser.com/questions/1271871/4k-emulation-sata-usb-controllers

"Summary:

512e drives ("1st emulation") allow you to use old drivers with new disks (since 2009!).

"4Ke controllers" ("2nd emulation") allow you to use legacy MBR with huge disks (cf. Update 2 in question)."

*******

"4Ke" - logical/physical = 4096/4096 - identical to 4Kn :(

Intenso (e.g. Memory Center)
LogiLink QP0020 (thanks Thomas Weller)

"Non-4Ke" - logical/physical = 512/4096 -

AS-Media (e.g. Anker)
inateck FD2002 (thanks Thomas Weller)
"

I don't think I can adequately explain this.

*******

The first case of this I ever saw, was in an
Areca RAID user manual I happened to be reading.
It allowed a 16TB array to be used in WinXP or something.
And it's because the sectors were 4K instead of 512.
The 32-bit MBR numbers, then allowed reference to a
16TB span (32-bit number times 4Kbyte sector).

It seems since the Areca RAID card, some other designers
have tried their hand at this "trick". The CDB sent to the
controller, must be "munged" and translated, to support
the trick. The command sent by the OS, is adulterated before
it hits the disk drive controller board.

I do not know what the name of these bodged adapters is.

The article here hasn't been updated in a while, so
it's not going to be useful to a practitioner.
In some cases, even when a company declares their
stuff is ready for 4K sectors, in fact, it isn't.
It takes a patch, if you are lucky.

https://en.wikipedia.org/wiki/Advanced_Format

Paul

[Carlos E.R.]

On 2024-02-10 16:46, bad sector wrote:

...

> Are some adapters better than others?

Yes, but there is no way to know in advance.

Some list a maximum disk size.

--
Cheers, Carlos.

[Paul]

On 2/11/2024 9:30 AM, Carlos E.R. wrote:
> On 2024-02-10 16:46, bad sector wrote:
>
> ...
>
>> Are some adapters better than others?
>
> Yes, but there is no way to know in advance.
>
> Some list a maximum disk size.
>

In many cases, that is the "maximum size we have tested with".

I have seen adverts, that one year claim the adapter worked
with a 2TB drive, and then a few years later, it advertises
that it handles 8TB drives. There are several CDB formats,
and there is room in 48-bit LBAs to handle some pretty big disks.
The plumbing is big enough for good sized disks.

It's certainly true, that there have been quirks in the past.
Like the NAS that could not handle more than a 200GB drive
(a nice round number that has nothing to do with hardware standards).
That was a software issue (as that thing used a processor
and some sort of OS).

And you would expect with UASP, that it would be using a CDB
for the LBA address purpose. The protocol does not have to be
UASP, but the stack might still use SCSI CDB anyway. Both
Windows and Linux, use SCSI for glue. Check the very last
line in this article, for inspiration, as to where a SCSI CDB
might be hiding.

https://www.kernel.org/doc/html/v4.17/driver-api/scsi.html

Paul

[bad sector]

I can use the adapter with extra-caution and in a limted way to move
stuff around but I'm really looking for a better way. If push comes to
shove, a usb3/thunderbolt external drive. My desktop box has an external
sata port, that's what's missing on the laptop!