Barracuda ES.2 peculiarity

[hnch]

Hi all,

I have been reading through your forum the last days, but somehow the information I found didn't help me, so I'm posting my question here:

I own 2 Seagate ES.2 ST31000340NS I salvaged from the trash (running 24x7 in a server for 2 years until the reallocated sectors became too high). After zero'ing the disks completely, I have been using them sporadically for data not too valuable - until last week when one of the disks didn't spin up correctly, but started clicking periodically. It sounds like loading/unloading (you know, the sound a working disk makes when powered on) once a second (and of course, it's not detected by the OS). I assume it's a lost cause, but I was curious if I could find out a bit more about it.

Since I'm working with microcontrollers as a hobby, I had a USB-TTL converter at hand and hooked it up. But: No output on power-on. But when I wait for approx. 1 minute (until the drive makes a short pause before continuing to click) and hit Ctrl-z repeatedly, I can get a F3 T> prompt. If I wait a bit longer, I can issue commands, e.g. Ctrl-l:

Code:

F3 T>
M13TNLQC TetonST 2.0 SATA Moose Gen3.0(RAP10)QNRdisc(RV)
Product FamilyId: 27, MemberId: 02
HDA SN: 9QJ2YTR4, RPM: 7206, Wedges: 108, Heads: 8, Lbas: 74706DB0, PreampType: 47 28
PCBA SN: 0000C9207WLJ, Controller: TETONST_2(639D)(3-0E-3-5), Channel: AGERE_COPPERHEAD_LITE, PowerAsic: MCKINLEY Rev 51, BufferBytes: 2000000
Package Version: MS031B.SDN1.AL0P00.SN05    , Package P/N: 100512179, Package Builder ID: F1,
Package Build Date: 04/01/2008, Package Build Time: 09:03:23, Package CFW Version: MS03.SDN1.00067102.F100,
Package SFW1 Version: B101, Package SFW2 Version: C0FE, Package SFW3 Version: ----, Package SFW4 Version: ----
Controller FW Rev: 04010903, CustomerRel: SN05, Changelist: 00067102, ProdType: MS03.SDN1, Date: 04/01/2008, Time: 090323, UserId: 00391559
Servo FW Rev: B101
RAP Format Rev: 0092, Contents Rev: 15 09 00 11
Features:
- Quadradic Equation AFH enabled
- VBAR with adjustable zone boundaries enabled
- Volume Based Sparing enabled
- IOEDC enabled
- IOECC enabled
- DERP Read Retries enabled

When I wait long enough, the drive stops clicking. I can then e.g. go to /2, and spin down the drive:

Code:

F3 2>Z

Spin Down Complete
Elapsed Time 12.583 secs

Spinning up again will take over one minute until the drive stops clicking.

Very interesting point: If I detach the PCB from the drive and plug in power, I don't get any output at all, neither log messages, nor any response to Ctrl-z, regardless how long I wait after power on.

What's wrong with this disk? PCB, EEPROM, head damage?

Just for the reference, the other ST31000340NS shows expected behaviour (i.e., producing log output at power on). It is identified as

Code:

F3 T>
M13TNLQC TetonST 2.0 SATA Moose Gen3.0(RAP10)QNRdisc(RV)
Product FamilyId: 27, MemberId: 01
HDA SN: 9QJ56AP6, RPM: 7206, Wedges: 108, Heads: 8, Lbas: 74706DB0, PreampType: 47 08
PCBA SN: 0000C008E201, Controller: TETONST_2(639D)(3-0E-3-5), Channel: AGERE_COPPERHEAD_LITE, PowerAsic: MCKINLEY ENTERPRISE Rev 51, BufferBytes: 2000000
Package Version: MS031B.SDN1.BG090G.SN06    , Package P/N: 100549981, Package Builder ID: F1,
Package Build Date: 12/04/2008, Package Build Time: 17:19:06, Package CFW Version: MS03.SDN1.00130245.F100,
Package SFW1 Version: B102, Package SFW2 Version: C0FE, Package SFW3 Version: ----, Package SFW4 Version: ----
Controller FW Rev: 12041719, CustomerRel: SN06, Changelist: 00130245, ProdType: MS03.SDN1, Date: 12/04/2008, Time: 171906, UserId: 00391559
Servo FW Rev: B102
RAP FW Implementation Key: FF, Format Rev: 0092, Contents Rev: 15 0A 00 11
Features:
- Quadradic Equation AFH enabled
- VBAR with adjustable zone boundaries enabled
- Volume Based Sparing enabled
- IOEDC enabled
- IOECC enabled
- DERP Read Retries enabled
- LTTC-UDR2 compiled off

best regards

[arztt]

Firstly too much reallocations and now clicking is most likely a weak head, and you can't do anything against it.

Well, but you could try to "repair" the other drive.

First, clear the G-List with F3 T>i4,1,22 command, then do a full media scan with F3 T>m0,8,2,,,,,22 command. It will take some hours and moves the bad sectors into the P-List (if there are not too much bad sectors or a weak head).
After this, do a SMART reset under level 1 with F3 1>N1 command.

Then a full erase and a sector scan under Windows/DOS...

But don't use the drive for important data, even if all the actions succeed!

Good luck

[arztt]

Well, forget about my first post.

I just checked the S/Ns of your drives and they are still under warranty by Seagate!

One drive till 01/2014, the other till 09/2014.

So go to Seagate.com and create a RMA. Send the drives to Seagate (you have to pay the shipping) and you'll get fully working repaired drives.

Btw, a very nice finding out of the trash

[hnch]

I just checked the S/Ns of your drives and they are still under warranty by Seagate!

One drive till 01/2014, the other till 09/2014.

So go to Seagate.com and create a RMA. Send the drives to Seagate (you have to pay the shipping) and you'll get fully working repaired drives.

Btw, a very nice finding out of the trash

I'm quite astonished about that, since the disks are labeled as OEM drives. Actually, I know the sysadmin out of whose scrap box I picked the drives, and I also know that the OEM (a small IT warehouse that builds our servers) only gives 3 years warranty. Since the 3 years were already over, he didn't attempt to address the issue with the OEM. Actually, I know that there were 3 more identical drives which failed earlier (within the 3 years period) and were exchanged by the OEM.

If I could be sure to get repaired drives, I'd give it a try, but I don't want to pay the shipping in vain.

best regards

[arztt]

If the online check says, the drives are covered under a Seagate warranty and you are able to create a RMA, Seagate will accept it.

I sent in some OEM labeled drives too in the past without any problems.

[hnch]

Ok, thanks for your answer, although I have to admit it's not what I wanted to hear.

Having written off the disk, I already was looking forward to some good old hacking, you know, soldering off EEPROMs, reflashing with some makeshift circuit, and so on. Because I'm really wondering why the PCB doesn't show any output when it's powered up without heads and motor connected. To me that sounds like a problem with the bootstrap firmware in the EEPROM.

Knowing I could RMA the drive is a real killjoy at the moment...
Actually I'm afraid that Seagate will refuse the exchange because they can see that I had the PCB removed (screws already a little worn out, traces from polishing head and motor contacts).

best regards

[BlackST]

... or reading this board with all that serial numbers exposed....

BESIDE THAT, assuming they were NS drives running 24/7 they are most probably at their EOL.

[hnch]

... or reading this board with all that serial numbers exposed....

BESIDE THAT, assuming they were NS drives running 24/7 they are most probably at their EOL.

That's definitely true, that was another reason why they were scrapped (and why I didn't put valuable data on it). The working drive has following SMART values:

Code:

Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x000f   081   060   044    Pre-fail  Always       -       146804842
  3 Spin_Up_Time            0x0003   098   098   000    Pre-fail  Always       -       0
  4 Start_Stop_Count        0x0032   100   100   020    Old_age   Always       -       108
  5 Reallocated_Sector_Ct   0x0033   092   092   036    Pre-fail  Always       -       169
  7 Seek_Error_Rate         0x000f   082   060   030    Pre-fail  Always       -       54035686850
  9 Power_On_Hours          0x0032   075   075   000    Old_age   Always       -       22238
10 Spin_Retry_Count        0x0013   100   100   097    Pre-fail  Always       -       1
12 Power_Cycle_Count       0x0032   100   037   020    Old_age   Always       -       108
184 End-to-End_Error        0x0032   100   100   099    Old_age   Always       -       0
187 Reported_Uncorrect      0x0032   001   001   000    Old_age   Always       -       169
188 Command_Timeout         0x0032   100   098   000    Old_age   Always       -       4295098379
189 High_Fly_Writes         0x003a   100   100   000    Old_age   Always       -       0
190 Airflow_Temperature_Cel 0x0022   072   043   045    Old_age   Always   In_the_past 28 (0 32 28 20 0)
194 Temperature_Celsius     0x0022   028   057   000    Old_age   Always       -       28 (0 17 0 0 0)
195 Hardware_ECC_Recovered  0x001a   048   034   000    Old_age   Always       -       146804842
197 Current_Pending_Sector  0x0012   100   043   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0010   100   043   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always       -       12


Meanwhile, there were over 1000 Reported Uncorrectable, which I have eliminated by repeated zero'ing (using dd under Linux).

But coming back to my original question: How can it be that the PCB doesn't produce any output when powered up without heads/motor attached?

One thing I'd be interested in is what would happen if I copied the working disk's EEPROM into the one on the other PCB and powered it up? I am well aware that the EEPROM also contains calibration data, so I'd try without heads and motor connected.
The problem with this approach is that the working disk is SN06, while the faulty one is SN05.
BTW: Does anyone have the memory layout of the EEPROM at hand, where is the bootstrap code located, where the calibration data?

All of that is more of academic interest, I don't need necessarily need the drives working.

best regards