Does "zeroing" hard drive repair it?
Does "zeroing" a hard drive repair bad sectors? How does it repair bad sectors?
tomatthe said:It does not repair them, things that say they repair them basically just mark them as bad sectors so they are no longer used. If you have a drive with bad sectors, replace it or use it under the expectation all data on it may be lost at any point in time.
I've got an IBM Thinkpad laptop with a Hitachi Travelstar 7K60 hard drive which stopped booting Vista. So I booted into safe mode cmd prompt with the Vista disc and ran "chkdsk /f" which reported and repaired errors. But I still could not boot from the hard drive, so I booted off the disc and ran "chkdsk /r" which also reported and repaired errors. I think it said it detected bad sectors and moved the data from these sectors. When I tried to boot from the hard drive once more, no success. So I booted of the disc and let it try to solve startup problems which reported errors with the hard drive. I repeated this process a few times as the laptop would not start from the hard drive. Eventually, chkdsk told me:
"The disk does not have enough space to replace bad clusters detected in file 5349 of name ."
Did some online research and the best explanation I could find was that the unallocated sectors reserved as replacement for bad sectors on the drive was all used up. Too many bad sectors it seem! So I ran Hitachi's Drive Fitness Test utility which reported the drive as defective. I then tried Spinrite at level 5 which also detected bad errors and tried to recover - scanned about 1% in 8 hours. With nothing to lose, I used the Hitachi program to erase the drive which according to the manual "write zeroes to every sector of your hard drive including the boot sector". Then ran Spinrite again at level 5 which scanned the entire drive and reported no errors in about 3 hours. I was amazed....no errors!
So I tried installing XP allowing XP's installer to repartition and full format the drive. After XP installed, I ran XP's check disk (with the GUI) with the "scan for bad sectors" option. Amazingly, no errors reported! How is this possible? Is the drive really repaired?
I hope you don't mind my long story but I'm curious to know how the Hitachi disk zeroing worked this miracle.
ID Description Raw Value Status Value Worst Threshold TEC
1 Raw Read Error Rate 0 OK 100 82 62 N.A.
2 Throughput Performance 3569 OK 111 58 40 N.A.
3 Spin Up Time 1ms OK 179 100 33 N.A.
4 Start/Stop Count 4241 OK 98 98 0 N.A.
5 Reallocated Sector Count 26804788 OK 94 74 5 N.A.
7 Seek Error Rate 0 OK 100 100 67 N.A.
8 Seek Time Performance 39 OK 114 100 40 N.A.
9 Power On Time 8402 OK 81 81 0 N.A.
A Spin Retry Count 0 OK 100 100 60 N.A.
C Power Cycle Count 2255 OK 99 99 0 N.A.
BF Gsense Error Rate 0 OK 100 80 0 N.A.
C0 Power off Retract Count 55 OK 100 100 0 N.A.
C1 Load/Unload Cycle Count 262581 OK 74 74 0 N.A.
C2 Temperature 32 C OK 171 100 0 N.A.
C4 Reallocation Event Count 633 OK 92 92 0 N.A.
C5 Current Pending Sector Count 147 OK 100 84 0 N.A.
C6 Uncorrectable Sector Count 0 OK 100 100 0 N.A.
C7 UltraDMA CRC Error Count 2 OK 200 200 0 N.A.
I agree. I wouldn't trust a drive that needed that many passes to fix. I'd be curious to see what the SMART values are for the current pending sectors and reallocated sector count is.
As you can see from my last post, the raw Reallocated Sector Count is 26804788. The total number of addressable sectors is 117210240, so I've got 23% of the drive reallocated! So yes, I agree with you all - a new drive is in order. However, I'm only using this for surfing, facebooking and watching an occasional movie (original stored on desktop) or online game. So I'm not gonna rush out and buy one tomorrow. And I've psyched myself for this drive to fail at anytime.
What I was really after was an explanation of how the drive manufacturer's zero-fill tool was able to "revive" the drive. How exactly does the zero write work?
Hawkeye22 said:There is nothing magical about it. It simply writes a zero to every sector on the drive. I don't think it fixed the error. I think it hid the error as there are still bad sectors on the drive.
I figured the Hitachi utility reassessed sectors which were previously labelled bad and reallocated those which it found to be bad. But a while back, chkdsk gave me an error about "The disk does not have enough space to replace bad clusters detected" which I thought meant the reserved or unallocated sectors on the drive were all used up in reallocating bad sectors. So I'm wondering if the unallocated sectors were all used up previously, where did the Hitachi utility reallocate the bad sectors confirmed on zero-fill?
I figure the zero-fill may have "zeroed" the unallocated sectors freeing them from chkdsk previous reallocation. And when the zero-fill process confirmed bad sectors it may have found less than was previously marked bad. So there was enough unallocated sectors to use for reallocation.
But then SMART's Reallocated Sector Count was so high, it's difficult to believe that the zero-fill process actually found less bad sectors than were previously marked. So I'm thinking SMART's counts are consecutive and its Reallocated Sector Count was not reset after the zero-fill.
Any thoughts anyone? Prior to this experience, I thought zero-fill utilities just set all sectors to zero without any re-examination of bad sectors.
bfb said:I figured the Hitachi utility reassessed sectors which were previously labelled bad and reallocated those which it found to be bad. But a while back, chkdsk gave me an error about "The disk does not have enough space to replace bad clusters detected" which I thought meant the reserved or unallocated sectors on the drive were all used up in reallocating bad sectors. So I'm wondering if the unallocated sectors were all used up previously, where did the Hitachi utility reallocate the bad sectors confirmed on zero-fill?
I don't think the hitachi utility reassesed the sectors. It probably left those alone since they were already flagged as bad. It may have found more bad sectors and although it had no where to relocate them to, it can still flag them as bad so that they aren't written to in the future.
Checking the SMART values may confirm this. Check it every few days and see if the current pending sector count is still growing.
As stated above, these utilities do not repair bad sectors in the sense that they become good sectors that can be written to. More correctly they partially repair the drive by placing markers that prevent the bad sectors from being used (written to). Once the quarantine is done utilities like check disk see the disk as good and not having bad sectors. The bad sectors are still there but are hidden by the excluding markers. Replace the disk. You have spent more time already than the risk is worth. In the very least your disk now has reduced capacity. It is not prudent to trust your data and OS to a disk that has exhibited behavior you have described.