Confirming bad heads on Fujitsu MHV2040AH

[guru]

on closer inspection ;o) I think the SW is incorrectly displaying the values from the ID 5 raw field.

Typicaly a NEW fresh from factory Fujitsu will have an ID 5 field of 00 00 00 00 D0 07 = No Grown Defects

as fzabkar said what happening with the other bits :O)

sknopp I suggest trying FJDT from Fuji if you can find the software and see if it produces a better SMART report

[fzabkar]

on closer inspection ;o) I think the SW is incorrectly displaying the values from the ID 5 raw field.

Typicaly a NEW fresh from factory Fujitsu will have an ID 5 field of 00 00 00 00 D0 07 = No Grown Defects

Wrong again.

Try inspecting it even more closely.

[fzabkar]

He who is up ones anus constipates himself Guru 2010

Oh dear, now we're discussing bowel motions. Can I now invoke some corollary of Godwin's Law? In any case, I would have thought that someone who is so enamoured with themselves to award themselves the title of Guru is by definition vanishing up their own nether regions. Pot, kettle, black.

ISTR that my first visit to this forum was greeted with derision. Since then I have seen many other novices suffer the same fate. Many are treated with condescension. We are told to research the information ourselves. No "spoon feeding" here.

See capacity-the-bios-t14701.html#p95910

The problem with "gurus" is that they think they know everything, which of course means that they aren't able to learn anything new. And for a rank outsider, someone who is not a Member of the Club, to actually contribute anything is unthinkable.

So, not meaning to offend, I chose not to present my usual detailed analysis, but instead offered a gentle hint that would spur the process of self discovery that gurus enjoy so much. I personally don't understand why somebody would prefer to trawl through years of SMS-ese noise, but perhaps it's the thrill of the search, or the satisfaction of discovery.

If you would like me to end this little demonstration, and put you out of your misery, just say so. I have my spoon ready.

[guru]

get your spoon out I'm hungry, and now seriously constipated ;o)

[Doomer]

Let's drop some oil into fire

MHV drive with reallocations SMART to Translator relations

SMART

Code:

Offset      0  1  2  3  4  5  6  7   8  9  A  B  C  D  E  F
00000000   10 00 01 2E 00 00 00 00  00 00 00 00 00 00 02 1E   ................
00000010   00 00 00 00 00 00 00 00  00 00 03 19 00 00 00 00   ................
00000020   00 00 00 00 00 00 04 00  00 00 00 00 00 00 00 00   ................
00000030   00 00 05 18 00 00 00 00  00 00 00 00 00 00 07 2F   .............../
00000040   00 00 00 00 00 00 00 00  00 00 08 13 00 00 00 00   ................
00000050   00 00 00 00 00 00 09 00  00 00 00 00 00 00 00 00   ................
00000060   00 00 0A 14 00 00 00 00  00 00 00 00 00 00 0C 00   ................
00000070   00 00 00 00 00 00 00 00  00 00 C0 00 00 00 00 00   ..........À.....
00000080   00 00 00 00 00 00 C1 00  00 00 00 00 00 00 00 00   ......Á.........
00000090   00 00 C2 00 00 00 00 00  00 00 00 00 00 00 C3 00   ..Â...........Ã.
000000A0   00 00 00 00 00 00 00 00  00 00 C4 00 00 00 00 00   ..........Ä.....
000000B0   00 00 00 00 00 00 C5 00  00 00 00 00 00 00 00 00   ......Å.........
000000C0   00 00 C6 00 00 00 00 00  00 00 00 00 00 00 C7 00   ..Æ...........Ç.
000000D0   00 00 00 00 00 00 00 00  00 00 C8 3C 00 00 00 00   ..........È<....
000000E0   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
000000F0   00 00 00 00 00 00 00 00  00 00 00 00 00 00 CB 00   ..............Ë.
00000100   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
00000110   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
00000120   00 00 00 00 00 00 00 00  00 00 00 00 00 00 F0 00   ..............ð.

Translator

Code:

Offset      0  1  2  3  4  5  6  7   8  9  A  B  C  D  E  F
00000000   44 4D 49 00 AA 00 48 00  13 1B 6C 01 48 00 29 02   DMI.ª.H...l.H.).
00000010   FF FF FF FF FF FF FF FF  18 F1 51 00 18 F1 51 00   ÿÿÿÿÿÿÿÿ.ñQ..ñQ.

[drc]

[hddguy]

Where did you manage to find this photo of fzabkar?

[fzabkar]

get your spoon out I'm hungry, and now seriously constipated ;o)

Here is your laxative. :)

Clearly the drive cannot have 13566324 reallocated sectors. A number that large would have tripped SMART. Instead the normalised value of the attribute has dropped to 92, which is still bad, but not calamitous.

What you need to do when you see a raw value that appears nonsensical is to view it in hexadecimal.

Reallocated sectors count = 13566324 = 0x00cf0174
Reallocation event count = 435028340 = 0x19ee0174

Here you can see that the raw value consists of three 16-bit words. The uppermost 16-bits are not displayed. For some inexplicable reason, some SMART authors deem that the uppermost bits are unimportant, possibly because they don't understand them. HDDScan will show all 48 bits. In fact the actual reallocated sectors count is probably 0x070100cf0174. A good drive will show 0x07D000000000. The uppermost 16 bits indicate the number of remaining spares. For a good drive, that number will be 0x07D0 (= 2000), as stated in the publicly available product manual (page 6-9, page 249 of PDF).

MHV2100AH, MHV2080AH, MHV2060AH, MHV2040AH, DISK DRIVE PRODUCT MANUAL:
http://de.sparrer-online.de/hardware/po ... manual.pdf

The mistake you made was to view the number in big-endian format instead of little endian.

AIUI, in the OP's case the actual number of reallocated sectors is 0x00CF (= 207), and reallocation event counts is 0x0174 (= 372), although I'm not sure of the latter.

The normalised current and worst values of the Read Error Rate are 75 and 55, respectively. AFAICT, that's not good. However, the raw values of the Read Error Rate, Seek Error Rate, and Write Error Rate attributes appear to be sector counts, not error counts. Seagate uses these values in a similar way.

I have posted my test results in the following discussion:

http://groups.google.com/group/comp.sys ... 5959145d44

If the Reallocated Sector Count threshold for the subject drive is 24, as in the abovementioned discussion, then it is clear that the drive would need to lose 76 points before it fails SMART. At the moment it is at 8. Now if 76 points represents 2000 reallocations, 8 points will equate to 210.5, which is close enough to 207.

[drc]

.

[Doomer]

good work but don't overdo things
you also can see how number of reallocations affect DM module (which drive uses for translation) in my example

OMG I posted thresholds instead of attributes you cannot see a thing
Here they are

Code:

Offset      0  1  2  3  4  5  6  7   8  9  A  B  C  D  E  F

00000000   10 00 01 0F 00 64 64 69  F5 00 00 00 00 00 02 05   .....ddiõ.......
00000010   00 64 64 00 00 52 01 00  00 00 03 03 00 64 64 01   .dd..R.......dd.
00000020   00 00 00 00 00 00 04 32  00 63 63 9B 01 00 00 00   .......2.cc›....
00000030   00 00 05 33 00 61 61 48  00 48 00 88 07 00 07 0F   ...3.aaH.H.ˆ....
00000040   00 64 64 19 0B 00 00 00  00 00 08 05 00 64 64 00   .dd..........dd.
00000050   00 00 00 00 00 00 09 32  00 64 64 23 00 00 00 00   .......2.dd#....
00000060   00 00 0A 13 00 64 64 00  00 00 00 00 00 00 0C 32   .....dd........2
00000070   00 64 64 8D 01 00 00 00  00 00 C0 32 00 63 63 1D   .dd......À2.cc.
00000080   01 00 00 00 00 00 C1 32  00 64 64 16 03 00 00 00   ......Á2.dd.....
00000090   00 00 C2 22 00 64 64 18  00 13 00 30 00 00 C3 1A   ..Â".dd....0..Ã.
000000A0   00 64 64 01 01 00 00 00  00 00 C4 32 00 61 61 48   .dd.......Ä2.aaH
000000B0   00 13 1B 05 00 00 C5 12  00 64 64 00 00 00 00 00   ......Å..dd.....
000000C0   00 00 C6 10 00 64 64 00  00 00 00 00 00 00 C7 3E   ..Æ..dd.......Ç>
000000D0   00 C8 C8 00 00 00 00 00  00 00 C8 0F 00 64 64 6D   .ÈÈ.......È..ddm
000000E0   49 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   I...............
000000F0   00 00 00 00 00 00 00 00  00 00 00 00 00 00 CB 02   ..............Ë.
00000100   00 64 64 D7 01 D2 02 64  01 00 00 00 00 00 00 00   .dd×.Ò.d........
00000110   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
00000120   00 00 00 00 00 00 00 00  00 00 00 00 00 00 F0 3E   ..............ð>
00000130   00 C8 C8 00 00 00 00 00  00 00 00 00 00 00 00 00   .ÈÈ.............
00000140   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
00000150   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
00000160   00 00 00 00 00 00 00 00  00 00 00 79 D7 01 00 5B   ...........y×..[
00000170   03 00 01 12 02 37 00 00  00 00 00 00 00 00 00 00   .....7..........
00000180   00 00 C0 46 C5 01 00 00  00 00 00 00 00 00 00 00   ..ÀFÅ...........
00000190   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
000001A0   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
000001B0   00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00   ................
000001C0   00 00 00 00 00 00 00 00  00 00 00 00 00 00 85 F1   ..............…ñ
000001D0   01 00 37 07 00 00 1B 00  1D 00 0F 0A 00 00 5A 07   ..7...........Z.
000001E0   D3 07 00 00 00 00 00 00  00 00 00 00 00 0E 00 00   Ó...............
000001F0   00 00 00 00 00 00 80 06  95 06 A6 04 75 00 00 36   ......€.•.¦.u..6

[fzabkar]

good work but don't overdo things :)

I'm sorry, I don't understand what you mean.

[Doomer]

good work but don't overdo things

I'm sorry, I don't understand what you mean.

That's too bad

[guru]

aha I see it now. I am used to seeing bytes in other order :O)

So I agree and feel relieved after my Senakot.

As for Seagate RAW Error Rate, something for you to chew on.

Seagate SMART Raw Error Rate = 10 * log10(NumberOfSectorsTransferredToOrFromHost * 512 * 8 / (Number of sectors requiring retries)) Where the factor of 512*8 is used to convert from sectors to bits. The attribute value is only computed when the number of bits in the "transferred bits" count is in the range 1010 to 1012

This may not be true for the latest platform...... And I suggest IMHO ;o) that you start another thread for this subject

[BlackST]

Why Senakot ? There are better solutions

[guru]

linseed oil? :O)

[guru]

it's because of my endianness :O) tokaneS

[Alexii]

Where did you manage to find this photo of fzabkar?

Bad troll is bad.

[fzabkar]

good work but don't overdo things :)

I'm sorry, I don't understand what you mean.

That's too bad

My behaviour in this thread was well within the bounds set by your own colleagues. I had intended to hold a mirror up to the gurus. In fact my posting style was [temporarily] remodelled on BlackST and others.

What's too bad is that Guru's ad hominem attacks were not deemed inappropriate. However, the issue of your double standards can best be addressed in my own special thread.

[fzabkar]

aha I see it now. I am used to seeing bytes in other order :O)

So I agree and feel relieved after my Senakot.

As for Seagate RAW Error Rate, something for you to chew on.

Seagate SMART Raw Error Rate = 10 * log10(NumberOfSectorsTransferredToOrFromHost * 512 * 8 / (Number of sectors requiring retries)) Where the factor of 512*8 is used to convert from sectors to bits. The attribute value is only computed when the number of bits in the "transferred bits" count is in the range 1010 to 1012

This may not be true for the latest platform...... And I suggest IMHO ;o) that you start another thread for this subject

So there are "proper manuals", at least for Seagate. My humble apologies.
I do in fact have a few questions and observations, and I will take you up on your generous offer, but in another thread.

In deference to the OP, I'm through with this pi$$ing contest. Now perhaps someone could help him with his problems. ;-)

[sknopp]

Update - I purchased a matching donor drive, swapped the heads, and imaged the data with Atola Insight (93% recovered). All but 6 of the customers pictures were recovered. Many thanks to all of the good information and free entertainment.