This is about a WD800JD SATA hard drive I have been using since 2007. A couple of days ago, I used WD Data Lifeguard Diagnostics for Windows v1.37 (release date: 2/27/2020) to first test the hard drive (it did pass both the quick and extended tests), and then I had the maladroit idea to do a full erase (with the same software). At about 90% of the full erase completed, there were numerous errors (the full erase process basically came to a stop). At that point, I opted for ending the full erase process, and that’s when the hard drive stopped being recognized by my OS (I was running the WD Diagnostics software in Windows XP). Finally, the hard drive is now again recognized by both Windows XP and Linux, but only as a 9 GB hard drive with serial number <0>. I would like to restore it to the original factory setting of 80 GB. How can I do this? Thanks…
Is the partition 9 GBs or is the entire drive being recognized as 9 GBs?
What does Disk Management show for the drive?
This hard drive was never partitioned. Before this happened three days ago, it had always shown in the system as an 80 GB hard drive and performed perfectly.
I gave it a try again this morning: MS Windows now does not show anything at all for the drive (It does not recognize it as present). But when I run WD Digital Data Lifeguard Diagnostics for Windows ver.1.37, it does show in this software as:
Drive# 2
Model Number Config AL25744_12345678
Serial Number o
Capacity 9 GB
SMART Status ? Not Available
Again, before running WD Full Erase on it, it showed as a different Model Number and with a proper Serial Number in the same WD software, and it was also normally recognized as a 80 GB hard drive in MS Windows for about 13 years (until three days ago when I ran full erase).
When this happened three days ago, WD Full Erase ran normally until about 90% of the way (for the first 72 GBs in the drive), at which point it got stuck with numerous consecutive bad sectors errors. At that point I opted to cancel the process by clicking on “Cancel,” when asked whether I wanted to continue or cancel the process.
Now, when I attempt to run WD Quick Test on it (same version of WD Diagnostics as aforementioned), it immediately stops with error: “Too many bad sectors detected.” However, as I mentioned before, this hard drive successfully passed both Quick Test and Extended Test three days ago, before I ran Full Erase on it.
Any thoughts on this?
EDIT: I just tried with a different desktop computer I own which runs Linux. Under Linux, the drive is also now not recognized at all, but when I run Linux’s fdisk on the hardware address where the drive should show as installed (for drive information only), I get:
fdisk: cannot open /dev/sdb: input/output error
Unfortunately, it sounds like the drive has failed. 13 years of regular use is pretty good considering.
You may want to try blanking the drive using DBAN (Darik’s Boot And Nuke)*. It may clear things up, but if the drive isn’t being recognized correctly, it sounds like the service area on the disk is corrupt or can’t be read.
*Disclaimer: If you’re unfamiliar with DBAN, please disconnect all additional drives, internal and external, from your system before preceding. DBAN can permanently wipe data from all drives that it detects. I’m not responsible for loss of data if accidental or misused.
Sure 13 years of use is pretty good, but would it be reasonable to consider it a mere coincidence that the hard drive failed right in the middle of running WD Full Erase on it? In other words, would it be advisable to NOT run WD Full Erase on drives that have been more than just a few years in service? If so, then it was a mistake on my part to run it on the drive - but then, again, there was no warning from the WD software that this might happen. Maybe there should be a clear indication that running this software on well-used HDDs may cause their demise.
Anyway, no complaints, as the HDD served me well and I did not lose any data when it failed.
I downloaded DBAN and burned a CD with it. Question: Have you done this before, while attempting to fix a similar issue, and what was the result? Also, what is the rationale for its use in a situation such as this one? I mean, if DBAN fixes it, what is it supposed to do to the drive in order to fix it? (restore remapped sectors and hidden areas to original? etc…)
Thank you for your suggestions…
There was probably a bad spot on one of the platters, and when the head came across it while doing a full erase, it may have damaged the head to the point where it couldn’t read correctly anymore. That would be my guess. You can try the WD full erase again, but you’ll probably won’t get far this time around.
DBAN wipes the drive completely, removing anything that could be corrupt and possibly cause problems. Basically it’s like a full erase with the WD software, but uses algorithms and multiple passes to ensure that data left behind is unrecoverable. If DBAN succeeds, then try reformating the drive again and see if you regain the original size of the drive. If it fails as well, then the drive is likely toast. Also, if the spindle motor produces a loud high pitch grinding sound, that would be another indication that the drive is possibly failing.
As for SMART status, did you try Crystal Disk Info? (May not display anything if your drive is connected via USB).
An an update.
I ran DBAN on the drive. DBAN had no problems finding the drive (while WinXP’s Disk Management and Linux could not). Surprisingly enough, DBAN ran in standard configuration (1 round, 3 passes, verify last pass) with “0 errors” and “finished with non-fatal errors” (after about 10 hours of wiping with an average throughput of 1,230 KB/s). However, DBAN detected the drive as an 8GB HD (8,589 MB), when it should be 80GB.
Also, its final message after the wipe was:
- FAIL SCSI Disk Config AL25744-12345678 8 GiB (8589MB) o
After the DBAN wipe I tested the drive again with the WD software as above.
Now, it immediately detected it as a 9 GB hard drive and the drive passed both the QUICK and EXTENDED tests!
The drive is now detected under Linux (fdisk) as:
64 heads, 32 sectors/track, 8192 cylinders, total 16777216 sectors, sector size (logical/physical): 512 bytes
but something is wrong here, as it is recognized as an 8GB drive under Linux, too, instead of 80GB.
The drive does not show in WinXP My Computer, but it does in Disk Management as: Disk 1 - Unknown - 8.00 GB Not Initialized (8GB Unallocated) - Device Status: this device is working properly.
However, it cannot be formatted nor a partition created on it. In Disk Management: Action → All Tasks: “New Partition” cannot be selected.
Linux’s Gparted (ver. 0.30.0) shows it as:
8GB /dev/sdb, unallocated
Partition Table: unrecognized
Heads: 255
Sectors/Track: 63
Cylinders: 1044
Total sectors: 16777216
Sector size: 512
File system: unallocated Size:8.00 GB
Partition Path: unallocated
First sector: 0
Last sector: 16777215
Total sectors: 16777216
Warning: /dev/sdb: unrecognized disk label
I attempted creating a new partition table (with Gparted):
Selected partition table type: msdos
Libparted Error: Input/output error during write on /dev/sdb
Any thoughts? It seems odd that after the DBAN wipe now WD Diagnostics passes the drive in both the Quick and Extended tests, but the drive still shows as a 8GB one and is still not usable…
The drive still runs very quietly, just as when it was first bought (no grinding sounds whatsoever). And you are correct, this drive is in an external enclosure connected via USB.
This would make me think the drive actually is in good physical health. It seems that when I first ran WD full erase on it, some info the drive supplied internally about its partition table, capacity, number of heads, sectors, cylinders, name, serial number, etc. for some reason got erased, and now the drive is not capable of being identified with its correct original parameters, and has thus become dysfunctional. I am not an expert, but could this be the case here?
The main issue seems to be the absence of a partition table: The software I have so far run on this device does not find one and is unable to create one due to the “Input/Output error” when trying to write to drive. Is there some software that can force write a partition table to this drive to restore it to the original factory setting? Or is the drive actually toast?
Any feedback based on the new information provided above would be much appreciated.
EDIT: The drive’s original specs:
https://www.cnet.com/products/wd-blue-wd800jd-hard-drive-80-gb-sata-150-series/
So, this problem has finally been (more or less) solved. I am posting my results here for others facing similar hard drive failures, in the hope they may find this information helpful.
If someone knowledgeable gets to read this, I would like to have your expert opinion of the following.
First of all, a big “thank you” to kflash08 for suggesting DBAN as a possible solution - without DBAN, bringing this drive back to life would not have been possible.
So, the first step I took, after writing my previous post, was to run DBAN for the second time on the drive. The drive was detected by DBAN again as a 8GB HDD (though it is an 80GB drive) and, as in the first run, DBAN went through the first 8GB (taking 9 hours to do it) with “0 errors” and “finished with non-fatal errors,” before giving again a FAIL message just like the previous time at the end of the run.
The BIG difference, after this second DBAN run, was that this time WinXP Disk Management recognized the HDD as an 80GB drive. However, partitions still could not be created both in MS Windows and Linux. The odd thing again is that Disk Management at this point detected the drive as a “working device,” but the drive did not appear as present in My Computer. I also ran the WD Quick Test (as done previously) and it failed with “too many bad sectors,” but the positive thing is that the WD software, at this point, also recognized the HDD as an 80GB hard drive.
The “nothing-short-of-miraculous” change happened when I ran DBAN for a third time on it. To begin with, now DBAN recognized the HDD, for the first time, as an 80GB hard drive. DBAN ran this third time for 56 hours, failing at about 70% of the way with the usual FAIL error. Because of this, I figured that the sectors in the HDD were free of errors for at least the first 56GB of the drive.
Linux GParted (ver. 0.30.0) now recognized the drive as an 80GB (74GB) drive missing a partition table, but this time around it was able to easily create a partition table on the drive, without getting Input/Output error messages as before. After that, I proceeded to create a 48GB EXT4 (Linux) partition (this is the first part of the HDD without bad sectors) and a second 26GB NTFS partition (where the bad sectors are supposed to be). GParted did not complain about any errors and created said partitions and formatted them accordingly.
I then installed the drive in the WinXP computer. Now Disk Management detects the drive as a healthy HDD (with the Linux partition detected as Healthy - Unknown Partition and the NTFS one normally detected) and the Disk Management options for deleting partitions, creating them, formatting them, etc. are all available. The NTFS partition (where the bad sectors are supposed to be) appears normally in My Computer and I even ran Windows Disk Check on it without any errors at all.
So it now would appear as a perfectly healthy and functional hard drive (I have also tested it under Linux with no problems or errors at all) except for:
-
It still provides a serial number of “o”.
-
When running WD diagnostics on it, both Quick and Extended tests fail with “too many bad sectors” (which are very likely detected in the second, NTFS partition).
Aside from the above two points, it behaves as a perfectly healthy and functional hard drive.
What do you make of all this? Feedback on it would be much appreciated.
THANK YOU!
Some additional information:
I just connected this drive via USB to a brand new Windows 10 laptop (latest Win10 install). With the “This PC” window open, the drive (the NTFS partition on it, the one presumed to contain bad sectors) was immediately recognized and its icon appeared within a few short seconds of being connected to the laptop.
Drive information shows as:
External AL25744_12345678 USB Device
Manufacturer: (Standard disk drives)
Device status: This device is working properly
Clicking on Check (Error checking):
“You don’t need to scan this drive. We haven’t found any errors on this drive. You can still scan the drive for errors if you want.”
Clicking on Scan Drive:
“Error checking: Your drive was successfully scanned. Windows successfully scanned the drive. No errors were found.”
Therefore, as far as I am concerned, this hard drive has been restored to perfect health and functionality. It performs and sounds as a brand-new hard drive.
My unanswered question, which I hope will eventually be answered in this forum, is: Why, then, do both Quick Test and Extended Test from WD Data Lifeguard Diagnostics for Windows v1.37 (release date: 2/27/2020) still fail the drive with error message “too many bad sectors,” after bad sectors are diagnosed in both tests and I choose the option “attempt to fix bad sectors” (not the exact wording)?
This is quite baffling…
That’s great that you got the drive working again. However, I strongly suggest that you do not rely on the drive as it can potentially fail again, so don’t store anything important on it.
56 hours is an awfully long time to perform 3 passes on an 80 GB hard drive with DBAN. However, if you connected the drive directly to the motherboard via the SATA interface, the process would have been less time consuming.
Note that the Disk Check in Windows only scans for file system errors by default. Unless you specify it to check all sectors, or you perform a full format on that partition, Windows will not be aware of any bad sectors and can potentially store data where they are located. Once you perform any of those options, you should see the capacity of that partition shrink as it marks all the bad sectors as unusable. However, if you fix those bad sectors using the WD Data Lifeguard Diagnostics, then the bad sectors will be recorded in the drive’s firmware. So even if you wipe the drive again with DBAN, the bad sector info will be preserved, and such programs should not report any errors as the marked sectors are being skipped.
While it is common for a few bad sectors to develop on the drive over time and be marked without triggering SMART or alerting the user, multiple bad sectors suggest an inevitable drive failure, or possible manufacturing defect that may qualify for replacement if the drive was still under warranty. That’s probably why WD Lifeguard Diagnostics gives up after x amount of bad sectors have been detected. You can try the DOS version of Lifeguard Diagnostics as I believe that I was able to scan and mark all the sectors with it.
Thank you so very much for your explanation, which I found to be very helpful! I tested it by completing filling the drive with large files and, yes, at about 90% of capacity there were errors in copying additional files and a few of the files which had (apparently successfully) been copied turned out to be corrupted, which proves your explanation as correct.
I could not connect the HDD directly to the motherboard via a SATA interface, as the only desktop PC I have now available with SATA is not functional at this time.
I will scan again with Lifeguard Diagnostics for DOS soon. Thank you.
Question: If the DOS version of LD succeeded in properly marking all the bad sectors as such so that they would be skipped during normal drive operation, would you still consider the drive to be unreliable for storing anything important on it?
I would still be wary about storing anything important or irreplaceable on the drive. Always keep up-to-date backups of your files on at least 2 external devices.
Running a complete sector scan on a regular basis for a suspected failing drive is important. If, after marking all the bad sectors, you see new ones pop-up over time, you should just completely stop using the drive and replace it. Even in its current state, I wouldn’t rely on it as there’s no telling when the drive will just outright quit altogether.
Thank you for another very useful response! You have provided a satisfactory, problem-solving answer to every question I asked, and at this point there are none left. Hopefully, others will find this thread to be as useful as I have.
Best wishes for a great and happy holiday season to both you and the other forum members!
