I either own or am in charge of (at work) a lot of HDDs that are either part of raid arrays, or there are backups of the data elsewhere.

When a disk is starting to fail and going slow because it's repeatedly trying to recover data out of an area is very annoying. I don't need the data and I'd like to either fix or force some errors on the failing drives so that hdd manufacturers are more likely replace them under warranty.

I realise there are other QAs about bad sectors but they are typically talking about recovering data. I'm not interested in what data is on the drive. Essentially I'd like to rapidly force the SMART reallocated sector count up over the failure threshold as quickly as possible or have the disk stabilise with all bad sectors reallocated and no more occurring. Currently the steps I take are

  1. quick format the drive
  2. if there was any errors go to step 1
  3. use a powershell script to copy files to fill the entire disk
  4. if there were any errors delete the problem files and goto step 3
  5. verify contents of all files are the same as the source file
  6. if there were any errors go to step 1

Usually the disk stops working completely or I become fed up with how many days this takes before the reallocated sector count reaches its failure threshold.

Sometimes I try the technique of leaving a file in a special bad files folder I make that I know has a bad sector in the file as my own marker to not use those sectors.

Are there any other tools or methods that could be faster?

Edit: I should also point out that I'm not concerned about securely wiping the drive either. If someone physically obtains the disk and can read any data off it good for them.

  • Once a drive starts losing sectors it belongs in the trash. Why are you trying to salvage it?? – Loren Pechtel Feb 25 '14 at 4:36
  • 2
    @LorenPechtel The OP is actually trying to just make the drive's reallocated sector count increase over the failure threshold, in the hope that will increase the likelihood that the manufacture will replace it under warranty, which I find rather unlikely and pointless. – Dan D. Feb 25 '14 at 5:34
  • Doesn't modern operating systems already have functionalities to deep scan a drive and mark bad sectors so they are no longer used? Why this manual route anyway? – ADTC Feb 25 '14 at 8:05
  • 1
    @ADTC "deep scan" is painfully slow and will often just fail or never complete. I want fast. I want at the first hint a sector is bad, forget what data was there, have the hdd reallocate it and move on. – BeowulfNode42 Feb 25 '14 at 8:15
  • Windows for example has "chkdsk" (Scan Disk) which can look for bad sectors. On a drive with several hundred GB this only takes some 4-5 hours, and it's not going to 'never complete' because it marks the unreadable sectors as bad and moves on (not linger around trying to recover and hang). I'm sure unix and mac has similar tools. So again, why manually? – ADTC Feb 25 '14 at 8:19

You will need to find an older computer you can hide in a corner to let these programs run. If the drive is lightly damaged it can finish in <2 hours depending on the size of the hard drive.

If you set dban (dariks boot and nuke) to 2 or 3 passes it might be faster. Maybe mhdd (with remap turned ON)

Depending on the size of the hard drive, it does not like multi-TB hard drives. Spinrite from grc.com.

All 3 methods are set and forget. Check it a couple times a day to see if they are done.


Your time is worth more than the drive, if your employed by someone im sure youve cost your employer more in salary Time than the cost of a new HD. Just get a new drive and copy over whatever data you can. If you have a backup (and you should) just restore the drive via that.


Tools I use with similar purpose:

  1. HDAT2
  2. MHDD (look for "MHDD" in the list), also see FAQ Q: How can I fix delays on my HDD (red and brown blocks in MHDD)?
  3. HDDScan

1 and 2 are both low-level DOS utilities, work on drive level, ignore filesystem, and can use custom timeouts (like "if there is 300 ms delay, rewrite the sector"). Most of the time the process is following (it's automatic): read data until there is a BAD or delay, rewrite bad/delayed sectors, continue. During rewrite, HDD either fixes the sector, or remaps it to reserve area. But as you don't need the data, you may choose to just erase the HDD. It will be slightly faster. Both utilities support repeat after end, so if drive slowly dies and throws more and more BADs, you'll be able to automatically finish drive' agony.

HDDScan is Windows utility, it's from same author as MHDD, but it has limited functionality comparing to MHDD.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.