Seagate 7200.11 SD1B: Sort of BUSY

[romsom]

Hello everyone,

I own one ST31500341AS drive which has failed circa two months ago. I'm not quite sure if that happened when powering on or during use.

The drive

HAS been updated to SD1B and
HAS worked after the update.

However, for about two months it

IS recognized only, when hotplugged, but
NOT by BIOS during POST (due to timeout).

Then it it kind of accessable but it takes ages even to mount a partition (ext3, ca 1.3TiB) and it IS possible to view its content (file structure etc.) but not the files themselves (except very small ones which aren't stored in any subdirectories).
When connected to a external case, the access LED of the case is permanently ON. Connected to the mainboard's S-ATA controller this is quite the same with the PC's HDD LED, until some time (~10 minutes).

I managed to get access via RS232 but have done NOTHING so far. I only attached the drive and checked if a connection is possible.
The first result: There is NO "LED:000000CC", just the "F3 T>" Terminal.
I wanted an experts opinion on it first.
So here we are. The data stored on the drive is important to me, but I surely cannot afford a professional recovery service.

What kind of failure could this be? And is there something I can du?
I am familiar with electronics, soldering and the use of software without a GUI.
Thanks in advance even if the answer(s) mean bad news.
Regards,

Roman

[BlackST]

99,99% firmware problem or something that force the drive into T (need to see the drive)

Did you try by a chance now or before the firmware update the "internet do it yourself solution" for busy or LBA0 ? In that case...

[derp]

Do MHDD scan of drive and report here.

I suspect weak head/s.

[BlackST]

Derp :

I managed to get access via RS232 but have done NOTHING so far. I only attached the drive and checked if a connection is possible.
The first result: There is NO "LED:000000CC", just the "F3 T>" Terminal.

And I asked :

Did you try by a chance now or before the firmware update the "internet do it yourself solution" for busy or LBA0 ?

MHDD in this situation is useful like a tit on a bull.

[derp]

BlackST,

Roman wrote

and it IS possible to view its content (file structure etc.)

MHDD will be just fine with this drive - weak or possibly bad heads will show up in scan

Probably fw is just fine, fine.

[BlackST]

If FW is fine, then why F3 T> ?

[derp]

If FW is fine, then why F3 T> ?

Ah, these romans

[romsom]

Did you try by a chance now or before the firmware update the "internet do it yourself solution" for busy or LBA0 ? In that case...

No. There was no need to do that. The drive worked fine, I updated firmware, it continued to work fine, some months later: it failed.

Do MHDD scan of drive and report here.

I suspect weak head/s.

I'll do that.
By the way: What does "weak" mean in this context? Probably something like "damaged", "broken" (not literally of course). Or has it some "special" meaning?

If FW is fine, then why F3 T> ?

Er.. why not?

@all:
Thanks for your replies so far!

[romsom]

MHDD is now running and I have some first results:




With this nice pattern weaving all through (there are some bigger "clean" parts, too), I think derp might be right.
I going to stop the scan now, as I don't have the ~400h, MHDD estimates. If a complete scan is necessary, I will start a new one.
So, I'm already saying Goodbye to the idea of getting back all my data. It would be nice, though, if I could save at least the accessable parts.
Could you recommend me a tool to do so?

[derp]

Roman,

No, don't scan the whole drive.
You should scan beginning (done), middle Sector 300 000 000 and then near end Sector 1000 000 000,
just to see if you have a head fault pattern or a media fault pattern. You can work it out.
I think these drives scan about 110 000 sectors per head, so if this is so, you can even figure out
how many heads are giving trouble.

The whole process should not take you longer than about 2 minutes, if that.

Once you are done you can deduce which method is better to get data.
a. make image of drive.
b. use some program which can quickly create file structure and allow you to select important files to recover.

Note : both options above may be risky, in that bad head/s might just break or something.

All you do is at your own risk.

[felibo]

Roman

All you do is at your own risk.

So all I can say is ur drive is running on life support. U just have that tinnnnny bit but of life left so be very careful with what u do. My advice would be to get an imaging software that would image the drive bitwise or buy a(n) Imaging device/Data Compass/PC3K to do the work or finally send it to a pro. In all cases the data is important to u I guess

[romsom]

Well, my data is important but not essential.
So my attempt would have been to copy the sectors bitwise to a new drive (I have two empty non-seagate 1.5TB drives right around here) with a software that quickly leaves those out, that can't be read.
And if the drive breaks during that process I can still return it to Seagate, can't I?

I've tried some tools now, but they either could not get access to the drive (linux) or were demo versions of commercial tools that had some limitations (copy only the first 30GB etc.)
As I asked before: Can you recommend me some software for that purpose?

[dick]

Yes try this freeware.
You can copy in reverse or forward mode.
If you get any problems try both modes and then try a software recovery on the cloned drive.

Roadkil's Raw Copy

Code:

http://www.roadkil.net/program.php?ProgramID=22

[romsom]

Yes, I tried this one before.
The problem was, the drive was not recognized in Windows (even when detected by BIOS). Also I'm not sure about the numbering (Physical Disk 0,1,2 etc.)
But the tool itself looks quite promising.

[pcimage]

dd_rescue