Welcome, Guest. Please login or register.
Did you miss your activation email?
September 26, 2017, 11:55:05 AM
Home Help Search Login Register
News:

+  DIY DataRecovery.nl Support forum
|-+  Support
| |-+  General Data Recovery (Moderators: Tom, Joep)
| | |-+  Problem - Hard Disk Drive not being recognised
0 Members and 1 Guest are viewing this topic. « previous next »
Pages: [1] 2 Go Down Print
Author Topic: Problem - Hard Disk Drive not being recognised  (Read 1364 times)
3cabbage
member

Posts: 9


« on: June 03, 2017, 11:34:13 AM »

I have a problem that appears to be caused by a corrupt MBR.  I have tried using the trial version of Diskpatch, but it couldn't find the faulty Seagate disk to test or write test results to.

For around 3 years I have been running my system with just a 2000GB Seagate Hard Drive connected.  I run normal houskeeping (defrag, registry cleaning etc) and disk backups about once a week.

A couple of weeks ago my computer started going crazy and the lights on my router were running up and down so I had to switch everything off, but when I tried to restart, it just sulked and wouldn't go into Windows.  I tried running windows recovery disk but that didn't recognise the hard disk.

I then decided to use an old Samsung disk as master to boot up with and use this Seagate one as slave to try and read the data on the disk (and possibly copy from it).  I connected the old Samsung Disk to SATA1 and moved the newer Seagate disk to SATA2, then rebooted and it went into Windows 7 using the old disk bootup.

Since it went faulty I've done a lot of command line tests including Diskpart Listdisk where it recognised the 1st Drive as SATA:PM (SAMSUNG) and 2nd Drive as SATA:SM (ST2000DM00 - [Seagate])

So far so good - but that was the last of the good news.

I tried running Seagate Tools for Windows which showed:
                                                                   SATA - PATA:  HD154UI (the Samsung)
                                                                     USB - 13G4:  Unknown (the Seagate)

I have tried various partition commands and partition software but the system only recognises and allows me into the old Samsung Disk.  As far as the newer Seagate disk is concerned it thinks it is now a USB Mass Storeage disk and shows the partitions as Removeable Generic USB devices of 0 length and won't let me access them.

As the computer thought the hard disk was a USB Mass Storeage device, I decided to try connecting it via a SATA to USB interface cable and experimented with booting up that way by changing BIOS sequence.  It recognised the disk as a Seagate (but would not boot) and again recognised the partitions as generic devices on SCSI4, Channel 0, ID 0 and LUNs 0 to 3.  Connecting to my laptops (Win XP & Win 10) to try and access through explorer got the same result i.e. can see that there are partitions but cannot access them.

My conclusion is that all my data is still there mainly intact and I should be able to recover it if I can force the system to recognise the Seagate Hard Drive as a fixed disk. I am convinced that the Master Boot record has been corrupted and if I can restore it I will have sufficient time to recover the important data that I hadn't backed up since the previous weekly backup.

I have tried a couple of MBR editing programs from Ultimate Startup disk, changing the hex interrupt values using one of them but that hasn't helped either because the different software does not recognise the Seagate HDD.
 
My little brain is getting addled now, so does anyone else have any ideas what I can try in order to change the MBR for the Seagate HDD back to a normal SATA fixed disk please?  I am fairly comfortable with working in DOS (or hexadecimal) providing I have explicit instructions of what to do.

Thanks in advance.


           Relevant Hardware:

               Motherboard:  ASRock G31M-S (with four non-removable SATA ports)
               SATA HDD 1:   Seagate ST2000DM00 - partitioned into:-
                                                                     C:Win 7 & other programs
                                                                     D:Data
                                                                     E:Old Vista Programs (don't mind it they are lost)
                                                                     F:Win 10 Technical Preview (don't mind it they are lost)
               SATA HDD 2:   Samsung HD154UI (only connected since Seagate failed to be recognised by motherboard

           Software:  Windows 7 Home Premium
Logged
Joep
Developer and Support Tech
Administrator
member
*****
Posts: 1457


WWW
« Reply #1 on: June 03, 2017, 01:03:32 PM »

Can you post a screenshot of disk management, see how that sees the disk.

Also: Does HD Sentinel see the disk? Works with most USB disks too. And what does it say about the disk's health?

And, DiskPatch (in DOS) does not show the disk at all, whether it's connected via USB or SATA? Are there any BIOS setup settings that would allow you to switch disk access/detection to legacy, compatibility mode or similar?

Is it possible to try the disk in another PC (not booting from it, but see if it can be accessed)?
Logged

--
Kind regards,
Joep
3cabbage
member

Posts: 9


« Reply #2 on: June 03, 2017, 06:38:32 PM »

HI Joep,

Thanks for the quick reply.

If it has worked there should be a screenshot of the disk management report. Note Disk 0 has been partitioned into C:Programs and I:Data.  D: is my DVD and F:G: H: and T: are all partitions of the Seagate.

I hadn't heard of HD Sentinal, but I downloaded a trial version and ran it.  It showed health of 48, but I think that was for the Samsung that I had replaced three years ago when it started failing - see screenshot attached.

I've checked BIOS and in Advanced - USB it already has Legacy USB support [enabled].  In ATA/IDE Configuration [enabled] it shows SATAII_1 Hard Disk, SATAII_2 Hard Disk, SATAII_3 Not Detected and SATAII_4 Not Detected.

I've tried interrogating the disk via the USB/SATA interface from both of my laptops, but couldn't.  The only other desktop PC I have is in the loft and pre-dates SATA so I wouldn't be able to plug it in.  I could try booting to the Samsung with the Seagate disconnected, and connect up later to see if plug and play detects it - would that help?

Kind regards

Alan
Logged
Joep
Developer and Support Tech
Administrator
member
*****
Posts: 1457


WWW
« Reply #3 on: June 04, 2017, 11:08:17 AM »

Hi,

Well, long story short, if the disk isn't detected anywhere when there's nothing we can do.

Logged

--
Kind regards,
Joep
3cabbage
member

Posts: 9


« Reply #4 on: June 04, 2017, 11:57:24 AM »

Hi Joep,

I was afraid that was the answer I would get - but you can't blame me for trying. 

I had hoped it was just a few corrupt bits that could be edited, like I was able to do in the old days with Norton, but if there isn't any software that can get in to read the MBR I guess that is it. 

I just can't understand why the motherboard is sometimes able to read what manufacturer the hard drive is and other times it can't.  I'll try accessing it again in a few weeks in the hope it is an intermittent fault, but I don't hold much hope.

Thanks for your all time and effort.

Alan
Logged
Joep
Developer and Support Tech
Administrator
member
*****
Posts: 1457


WWW
« Reply #5 on: June 04, 2017, 06:51:22 PM »

Hi,

Yes, but before we can edit anything we at least need to be able to access the physical disk ..

What are all the other disks I see in Disk management?

Logged

--
Kind regards,
Joep
3cabbage
member

Posts: 9


« Reply #6 on: June 05, 2017, 12:39:29 PM »

Hi,

You ask what the other disks are.  I only have two physical disks, the Samsung you see and the Seagate that you don't.  Disk 1 (F:), Disk 2 (G:), Disk 3 (H:) and Disk 4 (T:) are the four partitions of the same Seagate Hard Drive and although can be partially seen, the computer is misreading what they actually are. It thinks that they are USB disks - see Screenshot of Computer Management page attached

I believe that if I could only access the MBR then the computer would be able to interpret the information properly.

Kind regards

Alan
Logged
Joep
Developer and Support Tech
Administrator
member
*****
Posts: 1457


WWW
« Reply #7 on: June 07, 2017, 01:34:08 PM »

Yeah, I mean all the USB, SD, compact flash and what not 'disks'.

The HD Sentinel warning for the Samsung indicates some issues with the communication between the disk and the host. So, just an idea/hunch, but this may be the underlying cause for the issues with the Seagate as well.
Logged

--
Kind regards,
Joep
3cabbage
member

Posts: 9


« Reply #8 on: June 18, 2017, 07:08:16 PM »

Hi Joel,

Sorry I didn't reply earlier, I've been away.

Yes, my first thoughts were with communications so before I jumped into it being an MBR corruption fault I did the usual basic tests.

I checked that the spindle was spinning and that there were no head crash noises - everything seemed OK.

I swapped out the SATA cable on the faulty Seagate with the SATA cable on the working Samsung, the Samsung kept working and the Seagate still failed.

I plugged them into each of the four different SATA sockets on the motherboard in turn.  Whichever they were plugged into, the Samsung always came up and the Seagate always thought it was a USB drive.

I decided it couldn't be memory or BIOS because they shared the same memory and BIOS, and the system reported them both as being OK anyway.

All my tests brought me to the same conclusion, it was either the HDD controller board or the MBR.  I couldn't see any dry joints or burn marks on the HDD controller board hence my decision it was probably a corrupt MBR.

I know that in the old style (pre-LBA-logical block addressing) configuration the MBR is the first 512 byte sector of the disk with the bootstrap in the first 446 bytes, followed by 4 lots of 16 byte partition info (size etc) finished off with a 2 byte signature. As this is a 2 Terabyte disk, it could use either old type or LBA configuration.

In the old days ('84 to early '90's) there would have been a copy of everything (system and data) on the other side of the disk offset by a set number of bytes (16 I think) to prevent a head crash destroying both copies - it was easier to recover data then.  When heads became more reliable they changed the use of the space used by the backup copy to double the size of storeage - shame they didn't keep a small segment for a backup MBR.

I believe that the partition info is intact because the system reports 4 lots of USB drives which I think are the 4 partitions. It sometimes reports the Disk Identity, so I believe that quite a lot of the bootstrap info is intact.

What I need is some sort of "sniffer" (Protocol Analyser) like I used to use 20 plus years ago on network comms racks where I could identify each individual bit as it goes through the equipment.  Unfortunately I no longer have access to that physical equipment.  I have found some free software based Network Analysers, but they don't let me check the SATA bus.  However, I have just found a USB analyser (https://freeusbanalyzer.com/ ) and tomorrow I plan to load a copy onto my old PC to try and find out what is coming through the SATA/USB interface cable to see if I can work out the sizes of the partitions etc. 

I think that if I could read the sizes of the partitions, then it would enable me to recreate the MBR using some software I saw a few weeks ago.  However, without knowing the exact sizes then to use the re-creation software would destroy my data.  I know that I have to be extremely careful and I will not overwrite anything until I have checked and double checked.

I will let you know how I get on.

Kind Regards

Alan

Logged
Joep
Developer and Support Tech
Administrator
member
*****
Posts: 1457


WWW
« Reply #9 on: June 19, 2017, 11:59:08 PM »

Hello,

If the disk fails to ID correctly most of the time then MBR is something that is not relevant at this point. If the MBR was the problem then we could have addressed that using DiskPatch.

You can access Seagates via a terminal port (http://atola.com/products/insight/manual/serial-port-connection.html) but I can not help you with that. And more http://fillwithcoolblogname.blogspot.nl/2011/02/fixing-seagate-720011-bsy-0-lba-fw-bug.html

There's loads of stuff like that. Sounds like you're adventures so maybe this will help you.
Logged

--
Kind regards,
Joep
3cabbage
member

Posts: 9


« Reply #10 on: June 20, 2017, 10:25:59 PM »

Hi Joep,

I have used a few software analysers and got some weird results. 

With one analyser it showed the correct manufacturer and model in the hex listing, but I couldn't capture it.  It
also said there was a problem with the drive.  Each other time I plugged it in, it didn't show the manufacturer and model, it showed a row of zeros where the manufacturer and model number was.  On latest batch of runs it shows "USBS" starting at byte 37, then repeating "USBS" every 13 bytes.

With another analyser, it recognises the device but I cannot get it to output anything.

With a third analyser it recognises a header length of 128 bytes and reports success.  It also lists all messages between the device and my laptop showing direction (in/out) and status (Pending/Success).  With all the messages of 13 bytes, 18 bytes and 31 bytes there is always a matching pair.  However it keeps sending a 512 byte message (MBR?) in the "Out" direction which shows up as "Pending", but it never has a success.

My conclusion is that the MBR is well and truely corrupt, so at some stage my only option is to try and overwrite it and hope.

In the meantime, I saw your links and I have ordered a USB RS232 TTL adapter cable from eBay.  It should arrive in 7 to 10 days when I will try out the solution you offered. 

You are correct in your assumption about me, I am a retired telecoms engineer.  My first City & Guilds computer exam in the early 1970's covered 50% analogue computers and 50% digital computers (I really am that old) and my last course was the CCNA about 8 years ago, so I have the basic skills and broad knowledge required to prod about. I just need help from experts like yourself to get the detailed knowledge required to delve into the necessary areas.

I'll let you know more how I get on after the cable arrives.

Kind regards

Alan
Logged
Joep
Developer and Support Tech
Administrator
member
*****
Posts: 1457


WWW
« Reply #11 on: June 21, 2017, 08:41:13 AM »

Hi Alan,

As far as my knowledge is concerned, it's mainly logical disk structures which is my field of expertise. I follow a few boards that deal with hardware that also deal with firmware related stuff, but I have no hands on experience there.

To me it sounds like this disk has an issue that's either hardware or firmware related. The MBR has nothing to do with a system detecting the physical disk or not. It only comes into play after the firmware 'booted' correctly and after this firmware was able to read all disk parameters from the, let's call it, the service area from the disk, and then reported this back to the system asking the disk to identify itself. This area can not be accessed by your typical Windows or DOS based software without the help of additional hardware.

With this, often very expensive, specialized software/hardware (Google for PC 3000 for example and you'll see what I mean) the service areas can be accessed, modified and even emulated, after which the disk can be cloned. And only after that logical damage/corruption in for example the MBR can be addressed.

For Saegates DIY-ish solutions may be available because of the presence of this serial port on the disks (example http://www.msfn.org/board/topic/128807-the-solution-for-seagate-720011-hdds/). I would have to look for it, but I think I once even came across a piece of (inexpensive) software that knows the Seagate commands to communicate with the disk over a serial connection. Found it again: http://www.hddserialcommander.com/.

Thanks by the way, I feel young again (49 yrs old)

 Cheesy

Logged

--
Kind regards,
Joep
3cabbage
member

Posts: 9


« Reply #12 on: July 03, 2017, 08:52:34 PM »

Hi young Joep,

Nothing is as straightforward as it should be is it?  In preparation:-

I printed out the instructions you linked to so that when I turn my PC off I can still refer to the MML commands I need to type in.

I loaded RealTerm software onto my laptop ready to view any data flows.

I looked at the printed circuit board on the base of the faulty Hard Drive and it was fastened by really tiny star shaped screws.  I finally remembered they were Torx screws, and that when I was repairing my vacuum cleaner a couple of years ago I had bought a set from Maplins.  I dug them out and found a Torx T6 size undid them easily

The USB RS232 TTL adapter cable finally from came from eBay, but there was no documentation with it.  From other web sites I found out that the white wire was "Receive/RX", the green wire was "Transmit/TX", the black wire was "Ground/Earth" and red wire was "VCC/Power supply pin".

I thought I was ready to go, but when I tried connecting the RS232 connectors, although they were tiny (less than an eighth of an inch [1.5mm]) they were too big to go into the 4 pin jumper block socket.  I now have to either cut away some of the block side wall (which may mean I cannot fully format and re-use the disk later), or get some fine sandpaper from the local hardware shop in the morning and carefully rub down each individual cover.  That is what I will have to try first.

I'm also a little wary that on the third page of instructions it shows a quick format.   I know a quick format only deletes the list or library of files and does not touch the data, but I am still cautious.  I will try all the instructions without that line first, then if it doesn't work as a last resort I will try again with that quick format command in.

At the end of that I will again try some data recovery software to see if it can bring the data back (these companies advertise that they can bring hard drives back from a format).  Hopefully one of these things will work, otherwise the data has gone forever.

I should get time to play with it tommorrow evening.

Kind regards

Alan
Logged
Joep
Developer and Support Tech
Administrator
member
*****
Posts: 1457


WWW
« Reply #13 on: July 05, 2017, 09:01:01 AM »

Quote
I'm also a little wary that on the third page of instructions it shows a quick format.   I know a quick format only deletes the list or library of files and does not touch the data, but I am still cautious.  I will try all the instructions without that line first, then if it doesn't work as a last resort I will try again with that quick format command in.

Where did you read that? A quick format is never required to recover data. And could very well ruin the chances of recovery on SSD drives due to trimming. On a spinning disk, quick formatting isn't required to do file recovery. On NTFS only the first few MFT entries are affected. On FAT(32) the FATs are destroyed meaning recovery of fragmented files will be problematic.

Edit: Oh I see. "Flash", is he talking about a flash drive? Weird, don't even see him specifying a file system. I am really wondering if he's actually talking about a file system format ...
Logged

--
Kind regards,
Joep
3cabbage
member

Posts: 9


« Reply #14 on: July 08, 2017, 09:01:26 AM »

Hi Joel,

First my comment on quick format - no I understand that quick format is never required to recover data, but I meant as a last resort if the command did format the disk then many companies advertise something like "Restore Lost Data after Format Quickly and Easily".

I tried the RealTerm solution and when I switched on I got green lights for CTS (Clear to Send), DSR (Data Set Ready) and DCD (Data Carrier Detect) and a red Error light.  CTS and DSR are from the laptop, but DCD should be returned from the HDD. When I tried <Ctrl Z> I had no response from the HDD whatsoever.

I tried the same commands using Hyperterminal and PuTTY software and still nothing.  I set Echo and line feed then shorted out TX and RX with a paper clip.  I typed "Hello" and received back "HHeelllloo" so I knew the software at the laptop end was OK.

That took me into a whole lot of research into controller cards for HDDs.  Apparently they are classed as Cat 1 (replace and go), Cat 2 (replace, but have to unsolder BIOS and transfer) and Cat 3 (BIOS part of main chip and has to be transferred by speciallist equipment).  I am OK with Cat 1 & 2 because I am trained in soldering.

I can buy a controller card for my model for just under 40, or I can buy a HDD complete with controller card for less than half, but I have to check for the right software and revision number.  It is still cheaper than sending the disk away for someone else to fix it.

My next step is to make sure it really is the controller card and wait for the right combination to come along on eBay.  In the meantime I need to find out if my controller card is Cat 1,2 or 3 which I should be able to do next week.

It looks like it will be several weeks before I can move forward on this now, but I thank you for all your help, you pointed me in the right direction when I was going round in circles.

Kind regards

Alan
Logged
Pages: [1] 2 Go Up Print 
« previous next »
Jump to:  


Login with username, password and session length

Powered by MySQL Powered by PHP Powered by SMF 1.1.20 | SMF © 2013, Simple Machines Valid XHTML 1.0! Valid CSS!