bad sectors - seeking advice

Last week my PC (specs in signature below) began performing erratically, with some files and folders needing much longer times to open. I ran a full Malwarebytes scan which took much longer than usual to complete, but completed apparently successfully and detected and cleaned one infected file. I tried several times to run a full Microsoft Security Essentials (MSE) scan, but it seems to hang up around the 300,000th file. For what it's worth, MSE successfully completes the quick version of the scan with no detected problems.

I considered a clean re-install of my Vista OS, and tried to copy some files to an external hard drive. A few files would not copy with an I/O message error, though. I am currently running a CHKDSK function and at 45% completion, it indicates about 30-35 bad clusters have been detected and replaced. Is this in the average range for hard drives, or should I be considering replacing my HD (currently 13 months old with a 5-year warranty)? Could these bad clusters causing MSE scans to be incomplete and failure of some files to copy successfully to an external HD?

Thanks,

 

Stop messing with the drive now. Get a backup made of all your data ASAP and replace the drive.

 

Generally, you would prefer to have a drive that has zero bad sectors especially when it's your main hard drive.

+1
Backup and then replace.

 

Turn off the system with the bad drive. Get an external backup (two recommended).

COPY all your data to the first external - MOVE all your data to the second external.

Throw old drive away (okay, RMA it - if you don't care about the data it contained) - buy a new drive (Hitachi, while still available, highly recommended).

Good luck.

 

After you are done making two backups you can zero-fill the hard drive (Eraser for Windows works well, or alternatively command line utilities like 'dd' for Linux) if you're worried about the data on the drive. It might run the drive into the ground if it is slowing failing, but at least you gave it your best shot to protect your data. Also note that you can make a copy onto the 2nd external from the new copy on the 1st; you don't want to read the disk twice and risk the 2nd copy being inconsistent with the 1st.

Tiller is 100% right though...you need to stop using that drive NOW and immediately back up your data. Chances are some of it is already lost with those bad sectors; each time you access the drive you are risking further damage so only do what is absolutely necessary to pull the data off the drive.

With the Thailand floods making hard drives difficult and pricey to come by I'd definitely RMA if you can rather than buy a new drive.

 

I agree on the backup recommendation and getting a replacement drive.
As for the old drive... wipe it completely after you backed up and replaced it, and put it up for recycling.

 

I greatly appreciate the strong and unequivocal assessments and recommendations to save what I can of the data on the errant HD. I have a closely-related question--when I posted the original question, I had already initiated the CHKDSK sequence (recommendation of family member). Should I abort the sequence? It has been in progress for approximately 15 hours now and resumed when changed the PC's power source to AC after I inadvertently allowed the the PC's battery drain (although I think it started over instead of picking up where it left off) and the PC shut off accordingly. A brief Google search revealed strong recommendations to let the CHKDSK complete its sequence, but I realize that a HD with bad sectors could be damaging itself further with continued accessing. If I should stop the CHKDSK dequence, how can I abort it? Based on the same Google search, I could not find directions to abort the CHKDSK sequence--any suggestions? This may well be wishful thinking--does the CHKDSK's replacement of bad sectors help repair the HD and perhaps even prolong the HD's life? I also hope Western Digital (WD) will honor the 5-year warranty and replace the HD after I salvage what I can--anyone have experience with WD HD's and their warranties?

Thanks much again,
Ken

 

At this point in time, I would simply pull the power plug (or hold down the power button) to shut it down.

The drive is obviously failing and leaving it running/scanning/repairing will do more harm than good imo.

I would rather have a few corrupt files that chkdsk may have been able to recover, rather than risk a completely dead drive that also risks all my data.

The next time you boot this up should be when you have two or more external HDD's to copy/move your data to. Don't power cycle it again until this is done fully and completely.

Good luck.

 

Tilleroftheearth-

Thanks, I just shut down the PC with the errant HD. I have two external HDs with the capacity to accomodate the contents of the errant HD. Another question occurred to me--before I power up the HD again and copy its contents, would using an external HD enclosure have any advantage over keeping the HD in the original PC?

Thanks again,
Ken

 

Yes, the advantage is that it will simply be used/accessed to get your data if it was in an external enclosure.

The (possible) disadvantage is that taking 'ownership' of the data files and folders you want to backup could corrupt the data even more.

I think the less 'manipulation' involved the better: boot it up one last time and get your files.

Good luck.

 

Since your last post (below), I've kept my PC with the errant HD shut down and have not yet booted it up again, much less tried to copy its files yet. Admittedly I am somewhat fearful of mucking up the transfer process before the HD crashes for good. Compared to many of you, I am admittedly not very PC-proficient and PC-fluent. Consequently I am approaching this HD rescue process gingerly and accordingly have a few more questions, and ask your patience and pardon:

1. Does using an external enclosure for the errant HD involve more manipulation than booting it up in its home PC for file copy/transfer process?

2. When my PC first showed its sluggish behavior, I tried copying a few folders of files, and every 4th-10th file elicited an I/O copy error. Does this mean the file has been irreversibly corrupted?

3. The general consensus seemed to be to abort the CHKDSK process, which I had initiated early in this thread and continued for about 12 hours, and instead to start copying and transferring as many files as possible. Does this mean the CHKDSK bad sector repair process does not effectively repair the bad sectors in the HD and does not prolong the life of a failing HD?

4. Any other advice/suggestions to rescue the files on a failing HD?

Thanks much for your help.

Ken

 

1) An external drive attached to another PC would definitely be best. You avoid any un-necessary reads/writes that way since the OS isn't booting off that bad disk.

2) Probably, unfortunately. All the more reason to limit I/O to recovery operations only.

3) The bad sector scan does exactly what it is supposed to do, but when a drive is so badly damaged like this one you can actually create more bad sectors as the scan reads/writes over sectors previously gone bad. Think of it as hitting a ramp while skateboarding...the head of the drive lifts up a bit and then slams back down on the disk and damages it further. Scans like this only work well if the bad sectors are not surface damaged sectors but are bad for some other reason.

4) After you make that initial copy, try a program called SpinRite to try and recover the files that were corrupted (and save those copies as a 2nd backup). I completely forgot about that program. It probably will kill your disk in the process of recovering those files based on what you've told us so far, but SpinRite is going to keep hammering at those files trying to read them until the hard drive reports a non-erroneous read. I've seen a few IT people at Purdue University use that program to good effect so it is definitely worth trying as a *final* step to saving your data.

Also invest in a backup solution from here on in, please. You're learning now how painful not having a backup can be since you've only got one copy of your data. Generally speaking I try to run with no fewer than three copies (with at least one copy on a completely separate computer) to ensure that even in a disaster situation I probably won't lose anything. We can talk a bit more about that in this thread when you're ready...let's focus on getting your data back from the dead first.

As far as the ownership thing I read about, just copy the files into a backup and then modify the permissions of the backup. I cannot stress how important it is to limit use of that bad hard drive to critical recovery tasks only. I don't think you'll need to steal ownership to get at critical files but maybe others will have a bit more experience with that.

One more thing...make sure you're focusing on copying personal data, and not the OS and programs. Screw the stuff you can re-install later; focus on the data and files that are important to you. Usually most of that is going to be in C:\Users\<your username>.

 

1. Normally, if it was simply a data drive it would be best to not boot off of it and simply put it in an external enclosure to get the files off.

However, because it is an O/S drive, to see any of those 'Users' folders and be able to copy your files, you will first need to 'take ownership' of those files/folders - this is what may be detrimental to your data - each and every file will need to be 'touched' (accessed, modified, saved) to change the owner permission to the new system's (the one you're using to copy the files from) owner.

2. I can second Greg's recommendation to try SpinRite after you have recovered all files you can first.

3. The reason more and more bad sectors (and errors) keep coming up is because all the sectors used in reserve (standard for all drives) for such error scanning have all been used up (and, the platter seems to be disintegrating on you, causing more bad sectors as time passes...). Chkdsk can only tell the drive to mark the sector as 'bad' and map its location to a good sector from the reserve. When no more sectors are available in the reserve (as in this case) running chkdsk will simply run the drive into the ground.

4. What I would do when you boot it up is keep it as cool as possible - use a desktop/floor standing fan directly on the system and if you can open a window (assuming it is cooler outside than the room's ambient temps) it will go a long way to reduce the errors to as few as possible while you're warming it up copying your data over.

If you can (safely) lift the notebook off the desk (use two thick books on each end) and let the airflow from the fan go over and under it, this will be ideal (especially if the fan is pushing cooler (outside) air around the computer.


Hope this gives you the information you need?

 

Greg and Tilleroftheearth-

I very much appreciate your detailed and thoughtful--and rapid!--responses to my questions and situation. I think I understand the situation better conceptually. I'll implement your recommendations as I try to rescue my data and I'll let you know the results. I have a dedicated cooler fan for the notebook PC with the errant HD and will keep it running at maximum during the operation.

Greg, I also appreciate your offer to discuss backup solutions after the rescue operation and I'll certainly get back to you about that in this thread at the appropriate time.

Thanks much again and here's hoping I can save the failing HD's critical files, if not all of them.

Ken

 

This is both an update on the bad sector situation I described several weeks ago and a request for additional advice. Firstly, my hard drive (HD) failed completely about a week after my last post. Fortunately I was able to transfer essentially all of my critical files to two external HDs before the hard drive failed completely. Some entertainment files were lost, but I should be able to replace those soon. The most problematic data lost were the most recent bookmarks for my Google Chrome browser, and I may not be able to completely replace these; fortunately I was able to save the bookmarks from late June 2011.

The failed hard drive is a Western Digital (WD) model still under warranty. It no longer boots up and, when I install it as an external HD in external enclosure, my back-up PC (another Dell M1330) does not recognize it. Perhaps obviously, WD will replace the HD contingent on my returning the defective one. The defective HD failed completely before I could complete transfer of all files and before I could eradicate its files. Since I can no longer access the defective HD, I cannot perform any eradication procedures on it. My question--is returning this defective HD to Western Digital safe? My concern is someone could attempt advanced data extraction on the HD, although it is obviously not accessible by conventional methods.

Thanks much again for previous advice, especially Greg and Tilleroftheearth, and for any advice regarding returning my defective HD to Western Digital under warranty.

Ken

 

Xmarks addon for chrome, ff and ie has saved me from many troubles. It's too late for now but prepare for the future.

You need to consider how important the data is, because if it's too sensitive just smash the drive with a hammer and buy a new one

 

Since it's a bad sector issue, you might still be able to wipe it with a variety of tools even if it refuses to boot. DBAN is always a favorite, but try WD's utilities first to do a low-level format. That wipe will not affect the bad sectors themselves, but they're bad sectors, so in theory they're already corrupt. Modern drives map bad sectors within the drive itself, so you probably won't be able to do anything about that.

Based on the model number in your sig, it's a scorpio black, so check here: WD Support / Downloads / WD Scorpio & AV-25 / WD Scorpio Black

FWIW, I have a torx bit set that I purchased solely for the ability to dismantle every hard drive I've ever owned after wiping them. I still have every HDD platter I've ever owned hanging off a chain, except one I accidentally shattered (1.8" drives have absurdly fragile platters).

 

Thanks to KLF and Woofer00 for the suggestions. Woofer00, you are correct about the make/model of my defective HD (WDC5000BEKT Scorpio Black) and I accessed the Western Digital Lifeguard Diagnostic tool from your URL. I installed the defective HD as an external HD (using external enclosure) and the Lifeguard tool detected the drive, but showed it as "not available" with remaining memory of 0 MB. I apologize for perhaps belaboring the topic, but I wanted to clarify one aspect. Based on the defective HD not being accessible as both the primary boot drive and an external HD, as well as the Lifeguard tool indicating it as "unavailable", I presume the defective HD is essentially unusable and dead. I also surmise that, with intense effort, data could still be recovered from the defective HD, but that most people who strip-mine data from discarded drives would not find the effort worthwhile in this case? I am very mindful of KLF's suggestion that the safest route is to hammer the drive into an amorphous mass and buy a new drive. Of course, I know I'd be taking a risk of unknown magnitude if I sent the drive back to Western Digital for a replacement that someone would take the time and effort to search for usable data from the defective drive. Being a frugal person, I am admittedly favoring replacing the HD under warranty, and I'm just curious if anyone has an opinion about the relative risk in this situation.

Thanks much again,
Ken




I am also wondering if the probability of people my defective drive being

 

In the interim, I also inquired of Western Digital regarding the data on HDs returned for warranty replacement. This morning, WD sent me an e-mail reply stating, perhaps unsurprisingly, that they are aware of the sensitive nature of data on customers' returned HDs and consequently the returned drives undergo a series of tests that irrevocably destroys the data on the HDs. I belatedly discovered this information is also available on the WD site. I surmise that given WD is presumably protective of their reputation, they strive to prevent data on customers' HDs from being hacked. I know the risk is not zero, but is low enough that I am comfortable returning my defective HD to them for replacement.

 

I believe all of the major HDD manufacturers have similar policies in effect. The question is, how secure is the drive on its way to them?

I know that in some cases, if it's only the controller board that's become unresponsive, it's possible to swap them out and access data on the drive, regardless of what the bad one is reporting. Not easy, but possible.

Ultimately, as long as you make an effort that is sufficient to satisfy your personal level of security paranoia, you're good to go.