Have a Seagate ST2000DM001 that has been operational for a few years as a Windows 7 OS drive. On an error the PC was rebooted and it got as far as the splash screen and that was about it. I installed a new HDD and can get the faulty one to detect in BIOS/UEFI but it will not automatically mount inside an OS or on an identical PC (other HDDs work fine). In other words I'm confidant it's isolated to the HDD.

I inspected the PCB (I have electronics experience) and noted the PCB had some oxidization so removed it and cleaned up the two sets of contacts, didn't seem to make a difference. I have another HDD of the same batch, tools, multimeter, oscilloscope and hotair rework (can test and replace components if required).

I would like to know what to look for in isolating the fault to either the Data, PCB, or Media.
** Note: I will be recovering the data myself, please don't waste time suggesting otherwise **

More information: The faulty HDD seems to spins up, and seems to have head movement from the initial flutter of activity then a some occasional moving.

Windows System Event Logs report error code 9 "The device, XYZ, did not respond within the timeout period" then followed by error code 51's and 57's which report they are unable to access the disk.

Hi Click,

if drive is spinning there is not problem with pcb. Can you post SMART / SMART log and terminal log? If drive is spinning & detected with right specs (S/N, capacity, model) the problem most likely is bad sectors, media damage, weak / bad heads.

Have you tried mounting the drive in linux? You can try ddrecsue under linux do make an image or dmde to extract userdata.

Do _not_ open the drive, it's a Grenada and these are sometimes hard to recover even for a pro. As long as drive is detected there should be no need to open it!

best regards,
pcn

_________________
030 Datenrettung Berlin GmbH (Germany)
HDD, SSD, NAS, RAID Datenrettung - 030-datenrettung.de

Thanks for your help, whats your recommended way to dump the smart log and terminal log info with minimum impact to the disk, is there a linux tool for this as well? I've used the odd Windows app but it's been a while.



No I haven't is there a reliable bootable image made up with these tools or can you recommend a portable distro to use?
I'll wait until I have the SMART info first but is it generally better to just go for a one pass at getting an image or just go for the files?



No problem, not going to open it .

OK makes sense, will MHDD version 4.5 do the job at getting the SMART info and terminal logs?
Suggestions on better tools than ddrescue in order?
Does the reallocation by TTL attempt to move/rewrite bad sectors?

I like this strategy I would like to avoid hammering the disk if at all possible.

Stumbling block

MHDD 4.5 and 4.6 (boot floppy) will not detect any drives the on the ASUS P8P67 PRO motherboard I'm using (even working disks). This is a known good motherboard (8 ports).

They are detected in the BIOS fine it seems. Is there any tricks to getting it going? I used shift F3 and F2 in MHDD and also tried loading drivers at the menu option.

Try to switch BIOS from AHCI to IDE.

_________________
030 Datenrettung Berlin GmbH (Germany)
HDD, SSD, NAS, RAID Datenrettung - 030-datenrettung.de

Thanks, setting it to IDE worked fine.

Here is the ID/EID and SMART info. Let me know if you need anything else and what you think.

ST2000DM001-9YN164
LBA 3,907,029,168 BIOS
SN: W240XXXX FW: CC4H
SIZE 1907729MB
LBA48 HPA LBA MS16 DMA (UDMA, MWDMA)
SMART ENABLED
SELFTEST SUPPORTED
ERRORLOG SUPPORTED
SECURITY HIGH, OFF

[PRE]HDD: ST2000DM001-9YN164; FW: CC4H; SN: CC4H
--------------------------------------------------------
Name Val Worst Raw
Att # 1 : Read error rate : 88 88 25725091
Att # 3 : Spin up time : 92 92 0
Att # 4 : Number of spin-up times : 100 100 166
Att # 5 : Reallocated sectors count : 100 100 0
Att # 7 : Seek error rate : 66 60 4258070
Att # 9 : Power-on time : 64 64 31884
Att # 10 : Spin-up retries : 100 100 0
Att # 12 : Start/stop count : 100 100 161
Att # 183 : Unknown : 100 100 0
Att # 184 : Unknown : 100 100 0
Att # 187 : Unknown : 1 1 168
Att # 188 : Unknown : 100 100 0
Att # 189 : Unknown : 100 100 0
Att # 190 : Unknown : 65 47 589430819
Att # 191 : Unknown : 100 100 0
Att # 192 : Power-off retract count : 100 100 51
Att # 193 : Load/unload cycle count : 90 90 21650
Att # 194 : HDA Temperature : 35 53 35
Att # 197 : Current pending sectors : 100 100 8
Att # 198 : Offline scan UNC sectors : 100 100 8
Att # 199 : Ultra ATA CRC Error Rate : 200 200 0
Att # 240 : Unknown : 100 253 31779
Att # 241 : Unknown : 100 253 3936467293
Att # 242 : Unknown : 100 253 2597294617

Does not look too bad. How much important data is on the drive? If it is not that much I would boot up linux live-cd, download dmde and try to read mft, select important data and extact it. If mft does not read clone the drive with dmde or ddrescue and run dr software on the clone.

_________________
030 Datenrettung Berlin GmbH (Germany)
HDD, SSD, NAS, RAID Datenrettung - 030-datenrettung.de

Had backups, they weren't perfect so I still need to recover a few things if possible. What do you think the failure is?

I'm going to try a few random small scans in MHDD to destroy the 99% of the data I don't want if I can figure out the options curious to see what happens to the smart error count.

Did about 4 small scans of a second each, no errors, here is the second SMART info after the scan.
The drive has been running CC4H firmware since deployed.

Is it important at all to backup the ROMs on these models? Can this be done in software? I'm thinking of going through the other disks looking for anything that may indicate a premature failure, and would make sense to back up ROMs at the same time if it prevented that sort of failure.

[PRE]HDD: ST2000DM001-9YN164; FW: CC4H; SN: CC4H
--------------------------------------------------------
Name Val Worst Raw
Att # 1 : Read error rate : 89 88 29933435
Att # 3 : Spin up time : 92 92 0
Att # 4 : Number of spin-up times : 100 100 167
Att # 5 : Reallocated sectors count : 100 100 0
Att # 7 : Seek error rate : 66 60 4258123
Att # 9 : Power-on time : 64 64 31884
Att # 10 : Spin-up retries : 100 100 0
Att # 12 : Start/stop count : 100 100 162
Att # 183 : Unknown : 100 100 0
Att # 184 : Unknown : 100 100 0
Att # 187 : Unknown : 1 1 168
Att # 188 : Unknown : 100 100 0
Att # 189 : Unknown : 100 100 0
Att # 190 : Unknown : 65 47 589103139
Att # 191 : Unknown : 100 100 0
Att # 192 : Power-off retract count : 100 100 52
Att # 193 : Load/unload cycle count : 90 90 21657
Att # 194 : HDA Temperature : 35 53 35
Att # 197 : Current pending sectors : 100 100 8
Att # 198 : Offline scan UNC sectors : 100 100 8
Att # 199 : Ultra ATA CRC Error Rate : 200 200 0
Att # 240 : Unknown : 100 253 31779
Att # 241 : Unknown : 100 253 3936467293
Att # 242 : Unknown : 100 253 1587228227

Success!
Thank you both for the help.

I made a Linux Live CD and it was mounting the test disk I was using automatically. So I put in the problem disk and it read the mft and I managed to copy the important files off with clean reads.

I'll will end up taking a proper image but I'm past the worst of it. Do you have any idea what type of repair I could perform on the imaged disk to make it accessible in Windows?

Cheers.

Just checking before I proceed...

I assume I have some bad sectors somewhere near the beginning of the disk causing enough problems to kill windows and not let it be booted and also accessed under another windows host. Linux works however and critical data is recovered.

If I now do a ddrescue clone (jumping bad sectors) to get non critical data on to a recovery disk, would the resulting clone typically be windows mountable? I'm not sure of what constitutes as "good enough" to get a disk accessible under windows when it can be accessed under linux.

If I had to do a repair on that cloned disk what is the preferred tool? (partition repair, not necessarily OS repair).

Again really appreciate your help.

yes windows can sometimes 'repair' itself to a 'functional' state on the clone. best to copy data off clone before allowing changes or attempting booting.