Seagate GoFlex Desk - Automatically unallocated and partitioned

[vasukinv]

I had in my Seagate GoFlex External HDD partitions before I went to sleep last night. I don't know what happened to it when I woke up this morning. First, I noticed uTorrent showing an error saying "Drive error".

In the "Computer" window, I found this

Disk Manager showed this

I know, the possibility of getting everything back with not much effort is slim. I want to know what happened and what causes it to happen, so that it can be prevented from happening in future.

I just noticed frequent power cut occurring in the area I live in. Could frequent power supply cut off have cause this?

 

[phil22]

In 90% of external hard drives it's the SATA-USB bridge (a small PCB) inside the enclosure which fails first, whilst the drive itself is fine.

Removing the drive and putting it into a third-party enclosure usually fixes that problem, though it's sometimes not possible because some of these drives have a proprietary interface which is nether SATA nor IDE as we know it.

 

[vasukinv]

In 90% of external hard drives it's the SATA-USB bridge (a small PCB) inside the enclosure which fails first, whilst the drive itself is fine.

Removing the drive and putting it into a third-party enclosure usually fixes that problem, though it's sometimes not possible because some of these drives have a proprietary interface which is nether SATA nor IDE as we know it.

Oh, alright, thank you. How come the failure in that bridge cause unallocation of space?

 

[fzabkar]

IMHO it sounds like a logical fault rather than a hardware fault.

Since the partition information begins at sector 0, I would start by examining this sector with a disc editor in readonly mode. If you could upload its contents (in hexadecimal mode), then this will help us to narrow down the problem.

Roadkil's Sector Editor:
http://www.roadkil.net/program.php/P24/Sector%20Editor

HxD - Freeware Hex Editor and Disk Editor:
http://mh-nexus.de/en/hxd

DMDE (DM Disk Editor and Data Recovery):
http://softdm.com/download.html

 

[vasukinv]

IMHO it sounds like a logical fault rather than a hardware fault.

Since the partition information begins at sector 0, I would start by examining this sector with a disc editor in readonly mode. If you could upload its contents (in hexadecimal mode), then this will help us to narrow down the problem.

Roadkil's Sector Editor:
http://www.roadkil.net/program.php/P24/Sector%20Editor

HxD - Freeware Hex Editor and Disk Editor:
http://mh-nexus.de/en/hxd

DMDE (DM Disk Editor and Data Recovery):
http://softdm.com/download.html

Is this what you said you wanted?

 

[fzabkar]

You need to understand the difference between logical and physical sectors. You have shown us logical sector 0 of the first logical volume. It appears to correspond to physical sector 2048 of the physical drive.

Logical sector 0 is an NTFS boot sector.

Physical sector 0 should contain a partition table.

That said, the information in your boot sector is telling us that the sector size is 512 bytes (= 0x200), the starting sector of the NTFS volume is 2048 (= 0x800), and the size of the volume is 1,953,517,560 sectors (= 0x74704FF8). However Roadkil is indicating a sector range from 0 to 1,953,517,566, so there is a slight mismatch. Roadkil's information must be coming from the partition table, so that would be the next sector to examine.

At the moment it is looking like only physical sector 0 may be corrupt, in which case the fix will be easy.

 

[vasukinv]

You need to understand the difference between logical and physical sectors. You have shown us logical sector 0 of the first logical volume. It appears to correspond to physical sector 2048 of the physical drive.

Logical sector 0 is an NTFS boot sector.

Physical sector 0 should contain a partition table.

That said, the information in your boot sector is telling us that the sector size is 512 bytes (= 0x200), the starting sector of the NTFS volume is 2048 (= 0x800), and the size of the volume is 1,953,517,560 sectors (= 0x74704FF8). However Roadkil is indicating a sector range from 0 to 1,953,517,566, so there is a slight mismatch. Roadkil's information must be coming from the partition table, so that would be the next sector to examine.

At the moment it is looking like only physical sector 0 may be corrupt, in which case the fix will be easy.

Thank you. I always love the technical information. If it will be easy, how can I fix it?

 

[fzabkar]

If you can't access physical sector 0, then try DMDE instead (select Mode -> Hexadecimal).

After you upload the contents of this sector, then we will be able to compare it against your NTFS boot sector. The partition table will identify all the partitions and tell us their starting sectors and partition sizes. If necessary, we can then examine each boot sector for consistency, and make any changes to sector 0.

Your NTFS boot sector looks like this one (Windows XP):
http://thestarman.pcministry.com/asm/mbr/NTFSbrHexEd.htm

Therefore sector 0 may look like this:
http://thestarman.pcministry.com/asm/mbr/Win2kmbr.htm#CHS

Alternatively, the boot code may be blank, in which case you will only see data in the bottom 5 lines.

 

[vasukinv]

@fzabkar
When I selected Physical Drive, it showed this window.

And then after clicking on Close button and choosing "Mode -> Hexadecimal", as you said, it showed this:

 

[fzabkar]

Sector 0 is showing a single NTFS partition beginning at sector 2048, with a size of 1TB.

Sector 2048 is the boot sector of a 1TB NTFS partition. It looks OK.

Therefore the information displayed in Disk Management appears to be bogus.

 

[vasukinv]

So do you think the whole HDD is OK? Can I use it for data storage now?

 

[fzabkar]

You could test the drive using SeaTools.

You could also use a tool such as CrystalDiskInfo or HD Sentinel to examine the SMART data. Look for reallocated, pending, or uncorrectable sectors.

You might also like to try a Ubuntu Live CD. Perhaps Linux might be able to access your file system.

 

[vasukinv]

I've already formatted it on the same day after taking the screenshots (in the first post) and didn't use it since then. I was afraid to store any more data in the drive with the possibility of loss.

Linux disk managers always showed SMART error, and the disk could fail anytime. Maybe I had to take it seriously and get it replaced.

Results of SeaTools:
1. Short Drive Self Test - FAILED.
2. Short Generic Test - PASSED.
3. Long Generic Test - PASSED.

SeaTools doesn't S.M.A.R.T. test option for this drive, but shows for the internal WD drive.

 

[fzabkar]

SeaTools doesn't show the raw SMART data. You need something like CrystalDiskInfo or HD Sentinel.

 

[vasukinv]

Oh, alright. SeaTools utility showed S.M.A.R.T. Check for my laptop's internal WD drive.

Anyway, here's the report of CrystalDiskInfo. Is reallocated sectors problem a valid reason to get Seagate HDD replaced under warranty? If the warranty has expired, how trustworthy is this HDD in the future?

<!-- Detailed description removed for privacy after a credible response. -->

 

[fzabkar]

IMO your drive cannot be trusted. It should be replaced.

Currently it has 4015 (= 0xFAF) reallocated sectors.

Attribute 05 has fallen below its threshold value. SeaTools should give you a failing test code that you can submit with your RMA.

 

[vasukinv]

The warranty is over. Linux used to show since last two years that there's SMART error. Would accept that fact? It has been around 3 and a half years since I bought it.

Would they detect if I create an invoice to pretend it was purchased within last 3 years (around 30-32 months ago). I think it contains a log file MHDD.log, right?

 

[vasukinv]

Update to the previous post:
Ah! Just gone through my bank statement and found that it was purchased on October 03, 2010. Feeling highly relieved. I'll get it replaced ASAP!

I paid ₹4511.55 for that. Although it's not a hefty amount, it isn't something small.

 

[fzabkar]

http://support.seagate.com/customer/en-us/warranty_validation.jsp

 

[vasukinv]

Yes, I tried it, but it showed the following:

Component
The product you identified was sold as a system component. Please contact your place of purchase for service. Seagate sells many drives to direct OEM (Original Equipment Manufacturer) customers. These products are usually configured for the OEMs only, as components for their systems. You must contact your place of purchase for any warranty support on these drives.
If you purchased this unit directly from Seagate, please return to the previous page and verify your Seagate customer number, because the customer number you entered ( ) does not match our records.

I bought it from a retailer on eBay. I still have the confirmation e-mail.

Here's a possibility:
The serial number I checked for is shown in SeaTools and other software. But the serial number printed under the dock is different. That number isn't being identified.

When I entered the serial number I see in the bottom of the dock, it says

Unknown
Warranty could not be determined. Please check the serial and product numbers entered and try again. If the problem persists, contact your nearest Customer Support center.

Update:
Spoke to an executive on their customer care centre. They said the warranty ended on July 29, 2013. But were very co-operative when I said I have confirmation mail from eBay and bank statement as the proof of purchase. They immediately sent an e-mail asking for proofs. It seems like they provide great service.