Page 1 of 2
| [ 25 posts ] | Go to page1, 2Next |
Previous topic | Next topic |
|
Author | Message |
---|
Post subject: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | Hello guys, I have an SSHD Seagate ST1000LM014 which is not recognized anymore by the PC (not in BIOS, not by seatools). There is not spinning sound, the current consumption is around 200mA, even if I desconnect the platters and I keep only the PCB. I found a donor hdd, with some bads on it on ebay. It's the same model, same capacitace, same FW and same PCB number. My plan is to buy it, and using the serial port I want to make a secure erase in order to clear the NAND. The seller told me that it's not able to secure erase it using the seatools, probably because of the bads. After erasing the NAND, I would like to change the ROM, am swap the PCB to my faulty hdd. Do you think it's possible? Thanks
|
|
Top |
Click Erase and start to erase data off the device; click Cancel and return to the main program window. When you click the button, irreversible process of zero-filling all the sectors on the selected device is started. In some cases you can get one more warning telling that Lowvel cannot lock the device for exclusive access.
- Use Tool, Erase HDD to zero-fill your hard disk. Seagate (including Maxtor) SeaTools for DOS (see website for specific models supported). This is a modal.
- Here is an example. One easy way to zero out your hard drive, i.e., writing zeros to your hard drive to wipe out any data is to use Diskpart.exe, a utility. Buy Seagate Expansion 4TB Desktop External Hard Drive USB 3.0 (STEB4000100): Data Storage. Reformatting the drive will erase.
- Secure Erase overwrites every single track on the hard drive. That includes the data on 'bad blocks', the data left at the end of partly overwritten blocks, directories, everything.
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: September 8th, 2009, 18:21 Posts: 13257 Location: Australia | See https://forum.hddguru.com/viewtopic.php?f=1&t=35895
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | yes I saw that, but I have some other questions. 1. Is the serial port always on? (I saw some older posts, where people said that segate turned off the serial port) 2. Will I be able to erase the NAND using the serial port? I think yes, but what if the HDD has some bads? Does it matter? 3. Which are the commands? I did not found any documentation or some other things on the internet. The single information is that post If I managed to save my data successfully. I will create a how to, step by step post. For a clear overview of the people from the community. Thanks
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: September 8th, 2009, 18:21 Posts: 13257 Location: Australia | AFAICT from that other thread, the terminal is not locked in this firmware. So, switch to level O (that's the letter 'oh', not zero) and type 'I'. F3 T>/O F3 O>I Note that commands are Case Sensitive.
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | As far as I understand I can used a lever shifter like the MAX323 in order to communicate with the HDD. I check on the oscilloscope the TX pin from SSHD, and it's seems that this hdd is a 2V communication not 3.3V (Like MAX323). I have to found another integrated circuit. Can somebody confirm this? Still some questions: The donor sshd found on ebay has some bads, will I be able to erase the NAND? Should I buy it? Or should a brand new SSHD? (which way more expensive)
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | Hello guys, I managed to make it work with MAX3232. One hint for other people. MAX3232 will work with 2V communication. Just supply it from an external 2V power supply. My broken HDD report the following on the terminal, at START UP: Start_up_diag.png [ 20.91 KiB | Viewed 10158 times ]
Any hints ? what is broken? What shall I do next? Should I buy a donor PCB? Thanks For others, I have tried first with MobaXterm, and I was not able to make it worked, I received: test1.png [ 72.96 KiB | Viewed 10152 times ]
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: October 3rd, 2005, 0:40 Posts: 3592 Location: Hungary | NAND problem. You need to patch the fw or use a pc3k to do it for you. My recommendation is contacting a reputable DR shop. My pick would be Ciprian Enaru at Quartz data recovery (www.qdr.ro). pepe _________________ Adatmentés - Data recovery No bitcoin donations
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | Ok, thanks, I sent an e-mail. There are several hdd services in Romania, my problem is the price but let's see. Let's get back to the topic, I don't have a pc3k , how much does it cost? Is there another way? What do you mean by 'patch the fw' ? the fw is not inside the rom? and by swaping the rom I can solve this. As far as I understand I can buy a donor pcb, clear the NAND and swap to ROM, and I will be able to save my data.
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: October 3rd, 2005, 0:40 Posts: 3592 Location: Hungary | Hello, to be honest i am not sure how that pcb swap thing works, coz i have other methods, as i mentioned. In fact, having some clue about how the fw is loaded, i doubt those steps are sufficient, something is still missing from the cake. For you, as an end user it is perfectly useless to think about patching a firmware or investing into pc3k (several thousand eurs), moreover, it requires quite some knowledge (experience) to work with it, so believe me, it is out of scope. Also, the recovery gets very difficult if you happen to damage the rom chip or its contents during the swap, so i would be very careful with that if you have no experience. pepe _________________ Adatmentés - Data recovery No bitcoin donations
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | I'm looking at this post: https://forum.hddguru.com/viewtopic.php?f=1&t=35895 No worry, I'm able to swap a ROM I'm a hardware engineer with some years of experience in this field. I'm able to solder more complex component package than this one. Don't understand me wrong. I'm not saying that I'm able to bring back to life my sshd, but at least I'm trying to find a solution here. If it's a secure one and easy to understand. I don't want to assume any risk to make more damage to my sshd. Some specifications from my SSHD: Seagate 1TB ST1000LM014 PN: 1EJ164-071 FW: LVD3 PCB Version: 100705349 REV D
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | AFAICT from that other thread, the terminal is not locked in this firmware. So, switch to level O (that's the letter 'oh', not zero) and type 'I'. F3 T>/O F3 O>I Note that commands are Case Sensitive. This is made on the faulty SSHD after you clean the NAND,swap the ROM and PCB, right?
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: September 8th, 2009, 18:21 Posts: 13257 Location: Australia | I'm not a data recovery professional, and I don't know exactly what transpired in that other thread, but ISTM that you are unable to reach the F3> prompt. I expect that Ctrl-Z won't invoke the prompt either. I don't know if it will work, but I would first try to dump your serial flash memory chip ('ROM'). Then I would procure a donor HDD and secure erase it. That should initialise its NAND and clear its cached data. Then I would reprogram the donor's ROM with the dump from the patient, and install the donor PCB on the patient HDA. If this doesn't work, then we could analyse the ROM contents and determine if there are any drive specific data that need to be modified. There is a tool called F3RomExplorer which could help with this, otherwise I have my own ROM parsing methods. You must not damage the ROM. I expect that it will be a 1.8V or 2.5V part, so you will need an appropriate programmer to read it. As for the ROM patch, I suspect that PC3000 injects a JMP instruction to bypass the NAND check. BTW, welcome to Seagate's sh*tty firmware. No intelligent firmware designer would kill a drive just because of a bad NAND cache. At the very least, the drive should continue functioning in cache-less read-only mode so that users can recover their data. Better still, the firmware should initialise the NAND, retire any bad blocks, and keep using it, if at all possible. Some users may even be satisfied if the drive continues to function in read/write mode without cache, especially if it has outlasted the warranty period.
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: August 15th, 2006, 3:01 Posts: 3012 Location: CDRLabs @ Chandigarh [ India ] | Roots , You Can Read ROM Using a EPROM Programmer And Share ROM Here ,Someone can patch rom and give back and the terminal stuff you can handle ,Then you will need to clone disk if its also having some issues _________________ Regards Amarbir S Dhillon , Chandigarh Data Recovery Labs Logical,Semi Physical And Physical Data Recovery Website-> http://www.chandigarhdatarecovery.com
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | @fzabkar: Should I try to press Ctrl-Z ? I have WINBOND 25Q80BW10 it seems to be at 1.8V, you were right. 20200214_075618.jpg [ 1.69 MiB | Viewed 9964 times ]
I'm seeking for a EEPROM Programmer, and I found this one: https://www.ebay.com/itm/352894753009?ul_noapp=true As far as I understand it's a 5V or 3.3V and I will get a 1.8V adapter. I think it should work, do you have other recommendations? Regarding the donor HDD, should I have the HDA also? (or only the PCB it's ok) I found one full sshd but with bads, and the seller told me that it was not able to secure erase it using Seatools. Can I make a secure erase via serial port? I don't have a PC3000 and it's quite expensive I think. Thanks for the warm welcome, this will be my last Seagate product! ) @Amarbir[CDR-Labs]: It will take a while until I will find a EEPROM programmer, seems that I cannot found one in my country but when I will have it, I will share here the ROM content here. As far as I understand somebody must patch the ROM, in order to grant access to the terminal? Am I right? I was not aware of that. Thanks What do you mean by 'clone disk'? If everything works fine, I would expect to see my drive in Windows/Linux, and I can save my data. Or do I need a special tool to clone it?
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: September 8th, 2009, 18:21 Posts: 13257 Location: Australia | It won't hurt to press Ctrl-Z, but I doubt you will reach the F3> prompt. I don't have a programmer (I gave my old one away), so I can't recommend any, at least none that is inexpensive enough for a one-off job. As for the other HDD, if you can reach the F3> prompt, you should be able to initialise the NAND, if the terminal is not locked. If it is not too expensive, try to obtain the whole drive. I'm surprised that it was unable to be secure erased, though. Still, if you want to experiment, we will at least find out which procedures don't work. ;-) There is one other you could try. It didn't work in a different model (no damage done), but you could try disabling the supply voltage to the NAND IC. There should be a 3.3V LDO regulator nearby. Shorting its Enable pin to ground while powering on the PCB should render the NAND invisible. Hopefully you will find the regulator's datasheet here: http://www.users.on.net/~fzabkar/Datasheets/DATAURLS.HTM
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: October 3rd, 2005, 0:40 Posts: 3592 Location: Hungary | 'disabling' the NAND won't work that way. It needs the nand to boot. Getting F3 prompt is also not possible in that stage coz the firmware already hangs. pepe _________________ Adatmentés - Data recovery No bitcoin donations
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | Today I ordered the donor SSHD(PCB+HDA), which has 'End to end Error Detection', I hope I will be able to secure erase it. I also ordered the EEPROM Programmer ch341a with 1.8V adapter. (I found my serial flash memory in the list 25Q80BW). I hope I will be able to share with you soon some results and memory dump. Thanks.
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | Hello guys, Two topics: 1. CH341A EEPROM Programmer (software V1.3): I check it with a 3.3V memory, it does work (but it's not able to detect the memory, I have to manual select it). I was able to read/write etc.. I tried to read my 25Q80BW flash memory (not with the clamp) from the donor hdd (with 1.8V adapter). It does not auto detect it and even if I manual select it's still not working, I get the following errors (please ignore the device state, I made the print screen after I disconnect the programmer from USB): File comment: CH341A_Programmer - manual select 25Q80BW CH341A_Programmer.png [ 51.48 KiB | Viewed 9543 times ]
File comment: CH341A_Error_at_Detect of 25Q80BW CH341A_Error_at_Detect.png [ 4.56 KiB | Viewed 9543 times ]
The memory is still ok, I solder it back to the PCB and the SSHD it's working. 2. Donor SSHD and seatools: - I have checked the s.m.a.r.t. first, I get 'End To End Error Detection'. I checked the Seatools (windows version and usb bootable version), and I did not found the secure erase: In usb bootable stick version I have: File comment: Seagate bootable USB 20200219_215457.jpg [ 6.61 MiB | Viewed 9543 times ]
In windows version I forgot to make a print screen, and the memory is soldered on the programmer adapter (I will come back with a print screen it you need it). But were 3 options, as far as I remember were something like: - fw update - boot partition erase ? - erase track zero?
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | After debugging this programmer, I managed to dumb the memory. Here it is: Raul_HDD.7z [585.46 KiB] Downloaded 340 times
Is someone kindly to help me? Retarding the donor SSHD, do you have any idea how to secure erase the SSHD ? 'Erase track ZERO' or 'OVERWRITE FULL' would clear the NAND? Thanks p.s. the option from Seatools (Windows versions): Firmware update, Overwrite Erase, Erase Boot Sectors.
|
|
Top |
Post subject: Re: Faulty SSHD Seagate ST1000LM014 (NAND and PCB swap) |
|
Joined: February 10th, 2020, 8:59 Posts: 27 Location: Romania | I have tried to run 'Overwrite Full', it takes around 3 hours, and at 99.x% it fails. 20200224_064820.jpg [ 6.19 MiB | Viewed 9301 times ]
After that I written into the memory my firmware, swapped the PCB. In Windows I see it like this: HDD_Raul_Partitions_edited.png [4.36 KiB] Not downloaded yet
Pretty strange to me that terminal is working, I have tried to use '/o' and 'I', but it fails with the following message: HDD_RAUL_Terminal_1.png [ 103.87 KiB | Viewed 9301 times ]
After a couple of minutes, I safely removed the SSHD from the PC and get: HDD_RAUL_Terminal_try2.png [ 119.12 KiB | Viewed 9301 times ]
Do you guys have any ideas? It seems that it's a mechanical problem, am I right? Thanks
|
|
Top |
Page 1 of 2
| [ 25 posts ] | Go to page1, 2Next |
Okay, this is another one that annoys me: The claim, the myth that certain three letter agencies can recover data from erased or zero filled hard drives. The claim that those can magically ‘read’ and reconstruct data from ‘latent magnetic residue’ (I am not making this up, others did) using special machines and whatnot.
Wobbly heads
Idea is that the positioning from read/write heads is not exact. so new data (the zeros) may be written slightly off-track compared to the original data that can thus be recovered. My common sense tells me that IF exact head positioning is so difficult, then reading exactly those latent tracks is very difficult. And that IF head positioning is so difficult then there’s a chance that indeed some of the previous data may survive while other parts are actually overwritten. So at best you’d have partial data without the context of a for example a file system. You’d have binary blobs at best.
Now, assume a specific area was overwritten with new data several times as data was deleted or modified and rewritten by the same wobbly read/write heads. And finally the zero fill. Now results will be even more confusing as we’d have several layers of imperfectly overwritten data.
Zeros and ones
By the way, modern hard drives may depend to a degree on error correction when reading data. Data reads may be imperfect, but this imperfection is detected and corrected using ECC error correction. Each sector is ‘guarded’ by an ECC checksum that is computed as data is written, and data read at a later time is checked against that checksum and if needed corrected. Using ECC we can detect exactly which bits are off, the ones that were supposed to be zeros and vice versa. These ECC codes will not be available when we’re reading the latent magnetic data so we read (again) uncorrected fragmented tiny binary blobs at best.
Anyway I don’t have a science degree and will not claim that I know how hard drives work at that level, but these guys do and have examined the claims: https://www.researchgate.net/publication/221160815_Overwriting_Hard_Drive_Data_The_Great_Wiping_Controversy.
Let’s jump right to the conclusion: Filmora for mac full version.
This study has demonstrated that correctly wiped data cannot reasonably
be retrieved even if it is of a small size or found only over small parts of the hard
drive. Not even with the use of a MFM or other known methods. The belief that a tool
can be developed to retrieve gigabytes or terabytes of information from a wiped drive
So, if I zero fill my drive, no data can be recovered, right? WRONG!
Hah! You didn’t see this coming I bet! Kingdom rush vengeance jugar pc. What you are correct about is that data can not be recovered from zero-filled LBA space. No one can, not you, not me, not the CIA.
Apowermirror pc 2020. All zero filling tools that I know of can only write (their zeros) to LBA space. LBA space is the space on a hard drive an OS, a tool can address to write to or read from. As far as the OS is concerned, LBA space is all the space that exists on a hard drive.
But that does not necessarily mean LBA space covers all space that exists on a hard drive. You may have heard of HPA (host protected area) or DCO (device configuration overlay). These are areas on a hard drive that exist outside LBA space. Using ATA commands a ‘wiper’ could add these areas to LBA space. There’s also plenty of tools that can do that for you.
So, if taken notice of, HPA and DCO can be wiped. However, there may still be space lurking on the hard drive, that contains user data but exists outside LBA space. Modern drives may reserve space for caching purposes for example. Seagates refer to this as ‘Media Cache’. This is low level information obtained from a 8 TB Seagate drive:
User Partition
LBAs 000000000000-0000756080F9
PBAs 000000000000-000076893477
System Partition
LBAs 000000000000-00000013497F
PBAs 000000000000-000000146F3F
Media Cache Partition
LBAs 000074702556-0000756080F9
PBAs 0000759486D0-000076893477
Spare pool
PBAs: 00007578F548-00007586BDF5 RST Available: 8000 SCT Available: EF
Spare pool (Multi-IOEDC Region)
PBAs: 00007687B32C-0000768872C1 RST Available: 400 SCT Available: 1A
The Media Cache partition is not inside LBA space. It is not a partition that will pop up in Windows Disk Management! Even if we completely zero fill the drive, potentially 60 GB of recently accessed data remains untouched in the Media Cache! This data can probably be recovered by a capable data recovery lab.
The only way to wipe this space is using the ATA Enhanced Secure Erase command.