[SOLVED] Seatools S/N not matching S/N on disk's label

[valrama]

I found Seagate Ironwolf 4TB drives on Amazon for $15 less than the WD Reds. But having terrible luck with the quality of the drives being shipped. The first one could not even be formatted (Seatools said cyclic redundancy check failed). The replacement drive could be formatted by my Mac, but the Short Generic failed.

Here's the bizzarre part: both of these drives are showing up as the same S/N in Seatools - and not what is printed on the label on the hard drives. This is why the Seatools log file got appended. Here is the log file; note the failed tests on two different dates - they are on two different hard drives. Can someone help me understand what is going on?

1/17/2020 8:23:17 AM
Model: 008
Serial: B85910000000
Firmware: 0
Short DST - Started 1/17/2020 8:23:16 AM
Short DST - Pass 1/17/2020 8:23:26 AM
Identify - Started 1/17/2020 8:24:01 AM
Long Generic - Started 1/17/2020 8:29:17 AM
Long Generic - FAIL 1/17/2020 8:29:26 AM
SeaTools Test Code: 6C9AC2A4
Short Generic - Started 1/21/2020 10:50:32 AM
Short Generic - FAIL 1/21/2020 10:51:25 AM
SeaTools Test Code: 6C9AC2A4

 

[jojesa]

I think you found out why Seagate disks are cheaper than WD disks, they do have a higher fail rate.
I wouldn't trust them with my data.
SeaTools shows that test code when it cannot detect the serial number of the drive.
So it just spit out some random serial.
It doesn't necessarily means that the disk are bad.

 

[fzabkar]

Both drives are bad.

Can you show us the labels of the drives? Amazon has been known to ship fakes.

Can you retrieve a SMART report with CrystalDiskInfo?

 

[valrama]

Thank you for the response.

Both drives are bad.

Can you show us the labels of the drives? Amazon has been known to ship fakes.

Here they are. First one with the cyclic redundancy check failure.

This one failed the Short Generic

Can you retrieve a SMART report with CrystalDiskInfo?

Here you go. It is bizzarre that SeaTools could not fetch the S/N, while CrystalDiskInfo could...

 

[fzabkar]

The labels appear to be genuine:
https://verify.seagate.com/verify/

As part of the RMA process, a user needs to obtain a SeaTools "test code". This code is an encoded hexadecimal number which is unique to each drive. It is generated from the drive's serial number and the number of the failing test.

http://web.archive.org/web/20120406...toid=573797f40cce3210VgnVCM1000001a48090aRCRD

It appears that the test codes have changed in recent times, but my understanding was (in 2013 ?) that the second last digit identifies the failing test, as follows:

• xxxxxx4x - SMART Fail
  xxxxxx5x - Long Test, long LBA test
  xxxxxx6x - Short Test, short LBA test
  xxxxxxAx - Self-Service SeaTools Test Code
  xxxxxxCx - Short Generic
  xxxxxxDx - Long Generic
  xxxxxxEx - Short DST
  xxxxxxFx - Long DST

When a drive cannot access the firmware area in the reserved section of the platters (System Area = SA), it defaults to identifying itself with a dummy ID stored within the "ROM" on the PCB. I suspect that both drives failed to reach the SA at one time, in which case they would have reported the dummy serial number, model number and firmware version in the ROM. If you want to prove this for yourself, remove the PCB with a Torx screwdriver and power it up on its own. I expect you will see the same dummy serial number.

 

[valrama]

Does the CrystalDiskInfo output show anything alarming?

I am not feeling adventurous enough to unscrew anything on the drive, and will be sending this back to Amazon :-D They seem happy to refund the full amount. I am just glad this happened out of the box and not 2 months in the NAS... Now I will just get a WD Red.... having learnt my lesson.

 

[jojesa]

Does the CrystalDiskInfo output show anything alarming?

I am not feeling adventurous enough to unscrew anything on the drive, and will be sending this back to Amazon :-D They seem happy to refund the full amount. I am just glad this happened out of the box and not 2 months in the NAS... Now I will just get a WD Red.... having learnt my lesson.

It appear to be showing normal parameters and the 3 year warranty is valid.
Do not tampered with the drive, that will void the warranty.
As I sated previously, SeaTools shows that test code when it cannot detect the serial number.
These are new optimized drives and SeaTools might need to be updated to read them properly..

 

[fzabkar]

Seagate's SeaTools documentation is ambiguous.

Self-Service SeaTools Test Codes:
https://www.seagate.com/staticfiles/support/seatools/seatools-test-codes.html

Seagate understands that there are several legitimate reasons why you might not have a SeaTools Test Code. Please review the following suggestions and if you still need a Self-Service test Code, copy the 8-digit code from the one of the choices below and paste it in to the SeaTools Test Code field on the RMA form.

6C9AC2A4 -- Serial number not detected, SeaTools did fail the drive

Now here is a case where SeaTools did correctly detect the serial number, but it still produced the same test code:

https://www.computerforum.com/threads/help-need-for-samsung-hard-drive.213356/

Presumably the reason is that Seagate's serial numbers have 8 characters whereas Samsung's have 14. Seagate's test code generation algorithm probably expects to see 8 characters. AISI, a short generic test failure would be more accurately reported with a test code of 6C9AC2C4, rather than 6C9AC2A4, at least for non-Seagate models.

Here is another case where SeaTools generated a test code for a dummy serial number:

https://forum.hddguru.com/viewtopic.php?t=11403

• Unit Model: Seagate Desktop
  Model: ST_M13FQBL
  Serial Number: QNR_BFW
  Firmware Revision: 1117F393

Short Generic - Started 15/11/2010 10:02:19 PM
Short Generic - FAIL 15/11/2010 10:02:21 PM
SeaTools Test Code: 606060C0
Long Generic - Started 15/11/2010 10:04:13 PM
Bad LBA: 0 Unable to repair
Long Generic - FAIL 15/11/2010 10:04:30 PM
SeaTools Test Code: 606060D0

 

[valrama]

Thank you so much for your time and sharing your expertise. One final question on these blasted disks..

What does it mean for the SMART parameters to be within range but for SeaTools Short Generic test to fail?

 

[fzabkar]

It's hard to know what the drive is doing during its DST (Drive Self Test). In fact I can find no reference to short/long generic tests in the ATA standard.

http://www.t13.org/Documents/UploadedDocuments/docs2007/D1699r4a-ATA8-ACS.pdf (page 275 of PDF)

Here are two proposals that were considered for inclusion in the standard:
http://www.t13.org/Documents/UploadedDocuments/docs2002/e01139r2.pdf
http://www.t13.org/Documents/UploadedDocuments/technical/e01137r0.pdf

ISTM that the definitions and internal implementations of these tests is vendor specific. That is, each HDD manufacturer has its own way of doing things.

So, to answer your question, I don't know exactly what is involved in each test. However, the standard does say this:

The SMART Selective self-test routine is an optional self-test routine. ... This self-test routine shall include the initial tests performed by the Extended self-test routine plus a selectable read scan.

So the short test scans small areas of the surface, while the extended test scans the full surface, among other things.

The "Self-test execution status values" (page 280) would suggest that the electrical elements, read elements and servo are tested. There is even a provision for detecting handling damage.

Unfortunately SeaTools did not report a status value, so it's hard to say why the error occurred. Instead you could use smartmontools (Linux) to examine the "SMART Self-Test Log (Log Address 06h)" (see page 416). Smartmontools may identify the LBA that caused the test failure.

 

[valrama]

To close the chapter on the replacement disk: I plugged it into an empty Synology Bay and ran the Quick and Long SMART tests. Interesting enough both went through without a hitch. So if this was really a SeaTools bug or something could not work well on Windows via VirtualBox, well - too bad for Seagate. The second disk (which was a replacement of my first purchase) is now on its way to Amazon... And I am back in the market for a 4TB NAS disk

 

[jojesa]

I came to trust WD HDD, because of the history I had with their HDD.
I have been using them at work and at home for years.
The only issue I had, happened like 11 years ago when several WD starting spewing errors.
After that I have not had a single WD disk fail me and I have hundreds of them. We have servers, NAS, PCs laptops, with black, red and blue disk running 24/7.. Also a couple of HGST disk. Usually we keep PCs for 6 years and return them back.
At home I use WD blue, red & purple and none have yet failed on me.
I cannot said the same about Seagate. I have a dozen failed disk pulled from desktops and laptops.

I do have a 2TB portable Seagate which is more than 5 years old and still working fine. It was given to me.