Long story short, my wife can’t fully boot into Windows. She went into the Recovery Console and picked “Startup Repair”. It ran forever, so I stopped it and instead chose the Command Prompt and did a chkdsk /R. The recovery console is on the hard drive so I had to unmount it to run the check disk.
We started around 10:00pm 6/19/12. It’s 1:55pm 6/20/12 and is only on stage 4. It’s been running the whole time, it never froze. It’s just running very, very, slowly.
It’s a 500gb SATA drive. I’m thinking it’s a problem with the cable or the hard drive itself, but Googling some people say they’ve had chkdsk run for hours, so I’m not sure if this is a problem or not.
Both issues are quite likely due to a (mostly) dead drive. As soon as chkdsk finishes do a backup before anything else. If you still can’t get into Windows copy the contents to a new drive somehow.
Maybe something like Recovery Console off of the Windows CD will work. Otherwise, do something like slave the drive to another computer and see if you can’t back up all your data, or to another drive on the same computer. Then since it’s a new drive, you can likely get it replaced under warranty.
Even the best drives can fail prematurely. At least mechanical platter hard drives are still a crapshoot at times.
Well, I shouldn’t have said new. I think it’s about a year old or so, so not old, but not new either. A few days ago my wife’s computer fell about three or four feet onto the floor (she has a mini-tower). That could have caused the damage. Although I would think that she would have experienced problems the same day, not almost a week later.
Anyway, after the chkdsk is done, or else I might just interrupt it, I’ll try plugging the drive into a different sata port on the motherboard. Wishful thinking, but still.
If that doesn’t work I have a new hard drive I helped my mother order for her computer, only to find out her problem wasn’t the hard drive. I’ll just buy it off of her and transfer (or at least try to) everything from my wife’s drive over to the new drive.
yes, chkdsk takes literally hours to run on modern, large hard disks. My experience has been something like 1-1½ hours per 100GB of disk space. so your 500 GB drive should be 5-7½ hours, which seems about what it took.
From a quick look on Newegg, I believe most HD manufacturers give a 1 year warranty, while Western Digital gives 5 or sometimes 2 or 3 years. My memory says that Seagate used to give longer periods (3 years?), but I may be wrong.
I got confused. The newer drive was actually a secondary device. The main drive is older. And I don’t have the original packaging or receipts for either drive.
The newer drive has enough space that I’ve created a new partition and will try to copy the bad drive onto it.
For warranty, you shouldn’t need your receipts I don’t think, it goes by serial whenever I’ve done it. Western Digital at least has a “Warranty Check.” You just enter your serials and it will look up if you’re eligible. Once you have it backed up and out, see who made it (or check on BIOS) and see if you’re still good.
Even if the older drive is still within warranty, I think that the fact it started malfunctioning a few days after getting dropped probably disqualifies it. It’s more than likely our fault it broke, and not a covered defect.
Say you run a chkdsk and it marks some sectors bad. Then you copy the drive to another an re-run chkdsk again. Since there are no bad sectors (hopefully) then the sectors should no longer be marked bad, right?
Right, assuming there were enough good sectors available to remap the data to. But when a lot of sectors go bad due to physical damage there’s usually a snowballing effect as more and more physical damage is caused by the damaged head and debris from the platters.
I just meant if a scan disk program went over a healthy partition that was a clone of a bad partition then would it remove the bad sectors that it originally marked?
It’s still worth a shot. I’ve never been asked if the drive was dropped and they usually send me the drive first (along with a return label for the defective unit).
You can have Windows re-evaluate any bad sectors by running chkdsk with the /B option.
Scandisk moves the data it can from bad sectors to good sectors, then marks the bad sectors in a table so that they’re not accessed again. If you do a sector-by-sector clone it won’t even see the bad sectors so they aren’t copied. If any data got corrupted it’s copied, but from good-to-good sectors.
Where you can run into a problem is if between the time you ran scandisk and the time the cloning finishes you got a lot more bad sectors. That will copy corrupted data, but still to good sectors on the target drive. You don’t copy the “badness”, just the 1s and 0s, which may or may not be good
