[HD103SJ] Weird behavior - recognized properly, 100% UNRC.

[kotperski]

Hi!

I've got quite unusual problem with samsung hd103sj 1tb drive. It worked for long time without any flaws. SMART attributes were good, monitored regularly. Suddenly the computer just hung and after reboot the drive looked awry. Symptoms are:

- recognized by bios in normal way. Quick and properly, but with message: SMART Capable but Command Failed.

- can't boot from the drive, while booting from another one, dvd or pendrive, everything is greatly slowed down, almost unusable. So tried some diagnostic tools.

- MHDD - SMART attribs look generally OK, but every block of surface scan gives UNRC error.

- Victoria - the same, no chance to scan anything, but SMART looks good. Or at least drive doesn't seem to be dead yet according to SMART.

- ESTOOL can't pass any test, results are:

Read DMA - Undefined Error

Check SMART - nothing?

Spin up/down - Undefined Error

Simple Surface Scan - Command timeout

Write verify - ECC error

Random surface testing gives error on each LBA tested.



- drive doesn't make any unusual noises, no ticks or clicks or any other suspicious sounds. Everything seem to be OK, but it is not.

I tried to change cables and moved the drive to another machine. Same results unfortunately. I guess that the firmware got messed up so I tried to flash the drive using F3 utility available on Samsung/Seagate website but the program said that "the program isn't for this drive" or something like this.

Do you have any advices about this issue? I think that the problem could lie in the firmware but I couldn't find any possibility to force flash it... maybe I'm wrong though.

Any help would be appreciated, thanks in advance.

Greetings from Poland!

Attachment: SMART attributes logged from mhdd.

Code:

HDD: SAMSUNG HD103SJ; FW: 1AJ10001; SN: S246J9KB516980
--------------------------------------------------------
            Name                        Val Worst Raw
Att #   1 : Read error rate           : 100  100  436 
Att #   2 : Throughput performance    : 252  252  0 
Att #   3 : Spin up time              :   1    1  49486 
Att #   4 : Number of spin-up times   :  99   99  1150 
Att #   5 : Reallocated sectors count : 252  252  0 
Att #   7 : Seek error rate           : 252  252  0 
Att #   8 : Seek time performance     : 252  252  0 
Att #   9 : Power-on time             : 100  100  4264 
Att #  10 : Spin-up retries           : 252  252  0 
Att #  11 : Calibration retries       : 252  252  0 
Att #  12 : Start/stop count          :  99   99  1216 
Att # 191 : Unknown                   : 100  100  995 
Att # 192 : Power-off retract count   : 252  252  0 
Att # 194 : HDA Temperature           :  63   51  37 
Att # 195 : Hardware ECC recovered    : 100  100  0 
Att # 196 : Reallocate event count    : 252  252  0 
Att # 197 : Current pending sectors   : 100  100  54 
Att # 198 : Offline scan UNC sectors  : 252  252  0 
Att # 199 : Ultra ATA CRC Error Rate  : 100  100  1 
Att # 200 : Write error rate          : 100  100  151 
Att # 223 : Load retry count          : 252  252  0 
Att # 225 : Load cycle count          : 100  100  1219 


[BlackST]

You simply have "some" pending bad blocks hence the behaviour.

The only thing YOU can try is zero fill the drive but the problem can come back with a vengeance in the future (nobody knows when).

A pro can fix it once for all, hopefully for a small fee.

P.S. is the drive running hot or 24/7 like used for downloading all day and all night , by a chance ?

[Vulcan]

@kotperski,

Since you are asking exactly the same question on the Samsung/Seagate forum, I'll just give a short reply here, so that I don't waste too much time duplicating effort with readers there.

Personally, my biggest concern from that SMART data (and your test results) is SMART attribute #3 having a normalised value of 1, which is the extreme low limit of such values (and the related test failure from ESTOOL). It would be interesting to see the SMART threshold for that attribute on your drive (your data doesn't show the thresholds). Some other drives of that model have a threshold of 25 for that attribute (so that may apply for your drive too), and a raw value around 6000-8000 seems typical. I doubt this is a firmware problem, but I can think of a couple of other possibilities. Let's see what the folks on the Seagate forum reply to you...

http://forums.seagate.com/t5/Samsung-In ... m-p/182448

Do you need the data recovered from this drive? If so, is the data "vital" for you, or not worth paying for professional help? Is the drive still within its warranty period?

[kotperski]

You simply have "some" pending bad blocks hence the behaviour.
P.S. is the drive running hot or 24/7 like used for downloading all day and all night , by a chance ?

No, it was used in home desktop computer - typical usage, some gaming, some internet browsing, documents editing etc. Nothing too serious. Good ventilation, no excessive G's and so on.

@kotperski,

Since you are asking exactly the same question on the Samsung/Seagate forum, I'll just give a short reply here, so that I don't waste too much time duplicating effort with readers there.

You are right - I didn't get any answer for couple of days so I decided to look somewhere else for a solution

Personally, my biggest concern from that SMART data (and your test results) is SMART attribute #3 having a normalised value of 1, which is the extreme low limit of such values (and the related test failure from ESTOOL). It would be interesting to see the SMART threshold for that attribute on your drive (your data doesn't show the thresholds).

I'll check these SMART data again using other tool - but now I'm quite far from home and do not have direct access to this drive. Please give me two days more...
Anyway, this attribute looks weird for me also - I'm not sure but I think that when the drive was working properly, #3 has quite typical values. I think at least I would notice if something would be unusual.

Do you need the data recovered from this drive? If so, is the data "vital" for you, or not worth paying for professional help? Is the drive still within its warranty period?

No, there's nothing important. Of course it would be nice if I manage to "resurrect" this drive but I won't get depressed if it will be impossible. I consider it rather as a testbed for something I didn't seen so far and I try to learn something new from it.

This drive can be RMA'ed so this is the final option.

Thank you both guys for your help and please excuse me any language errors I made.

[BlackST]

If there is no important data try zerofilling the drive with ERASE or FASTERASE with ATA password lock. It will take a loooooong time so be patient ! After erasure, cycle power and recheck smart for pending blocks.