1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Storage Remapping Bad Sectors (more quickly than Chkdsk)

Discussion in 'Hardware' started by boiled_elephant, 4 Aug 2017.

  1. boiled_elephant

    boiled_elephant Merom Celeron 4 lyfe

    Joined:
    14 Jul 2004
    Posts:
    6,626
    Likes Received:
    864
    A quick disclaimer: if your instinct is to say "don't bother, just replace the drive", please keep calm and carry on. This is not a thread about what you should do when a drive develops bad sectors. It's a thread exploring ways to remap bad sectors, as a technical question. Thank you.

    The internet is awash with historical threads in which the only response anybody ever gave to this question is, "if the drive is developing bad sectors, it's toast. It's only a matter of time. Get your data off it and replace it."

    This is usually well-intended advice but it's totally misaimed. If you know what a bad sector is, you know that a drive developing them is not to be trusted - but they can still be useful, for redundant backups, mirror pairs, ferrying data, testing clones, and so on. In a way, they're perfect testing tools, because they have no inherent value but still function.

    And to dispel a common myth - not all drives with bad sectors are doomed. I'm currently using a 2TB reconditioned drive for storage that developed over 100 bad sectors in its first year, all of which I painstakingly repaired with Chkdsk; it's functioned fine for another 5 years to date. A 500GB 2.5" drive I inherited had 200 bad sectors, remapped fine, and now works perfectly. People claiming that bad sectors are a death sentence are revealing their ignorance of the multiple and distinct possible causes of bad sectors, which I won't go into here because it's boring and long-winded.

    (Also - saying "that drive will eventually die" is the infamous, and fallacious, economist's trick: constantly restate a prediction until it comes true. All hard drives eventually die, so it's a safe - though vacuous - observation.)

    So here's the rub. You can remap bad sectors easily enough; Chkdsk does it if you use the /B argument. But, for obscure reasons, it tends to only do 3 per pass, meaning that if a large, modern drive (1TB-4TB) has a few hundred bad sectors (not unusual or damning relative to the size of the drive), remapping them all will take weeks, because each pass of Chkdsk /B takes several hours and only remaps 1-3 at a time.

    The question is: is there a better, quicker tool for remapping all the pending bad sectors on a drive in one pass?

    Answers:

    HDDregenerator, $100. Remaps and also has claimed data recovery functionality. - goldstar0011

    edit 27/04/2018: I finally got around to trying HDD Regenerator and can confirm that it's the mammaries. Does the job. A very straightforward and practical GUI (albeit a command-line-based one) that's idiot-proof:

     
    Last edited: 27 Apr 2018
  2. goldstar0011

    goldstar0011 Multimodder

    Joined:
    2 Sep 2007
    Posts:
    3,409
    Likes Received:
    303
    I've actually just salvaged a HDD from bad sectors which is now being repurposed as a low important backup drive.

    I used HDDregenerator which scans for bad sectors and attempts a repair, after 1 run I clear the data the program saves and run again and it hasn't shown any bad sectors, if it does I repeat the stage and if still there I bin the drive. For peace of mind I also run a surface check using any other program.
     
  3. Corky42

    Corky42 Where's walle?

    Joined:
    30 Oct 2012
    Posts:
    9,648
    Likes Received:
    386
    Chkdsk /B does more than just remap bad sectors, it "Clears the list of bad clusters on the volume and rescans all allocated and free clusters for errors" You'd only really need to run /B after imaging a volume with bad clusters and transferring that image to a new drive.

    I agree that people seem to over-react when a drive starts developing bad sectors, in the olden days some HDD used have stickers on them listing the bad sectors from new, they're only a worry IMO if the number of bad sectors start increasing over time.
     
    Last edited: 4 Aug 2017
  4. boiled_elephant

    boiled_elephant Merom Celeron 4 lyfe

    Joined:
    14 Jul 2004
    Posts:
    6,626
    Likes Received:
    864
    Based on my limited fledgling understanding of hard drive failure, it depends what is actually going on to generate the bad sectors. If they're the result of a physical knock causing a head to skip against a platter's surface and scratch through the protective film, you've just got a scratched area on the drive that can't be used. Once that area's marked off and redirected to the reserve sectors, the drive will function as normal. If, however, the bad sectors are the result of deteriorating heads, motors, magnets or spindle, the drive really is mechanically wearing out and it's just a matter of time.

    As you both say, the latter is easily recognised by the pattern: more and more bad sectors emerge over successive scans.

    People haven't adjusted their perception of bad sectors to reflect the increasing density of platters in recent years. My old boss would condemn a drive that had more than 20 lifetime bad sectors, because back when they were 60-80GB across two or three platters, that was a significant amount of damage. Now, with 1TB or more per platter, the same proportion of damage would be 500-1000 bad sectors. But people think in absolute numbers (I'm guilty of it myself) and 500 bad sectors just feels 'scarier' because it's a bigger number, even though it may only represent the same little surface scuff that would've generated 20 bad sectors a decade ago.

    Edit - I'm also interested to know whether there's any logic behind the idea that newer, bigger drives are more likely to develop unimportant bad sectors than smaller drives were. I feel like 2005-2007 drives of 40-160GB capacity would more often end up dying following a small number of bad sectors, while larger >=1TB drives often seem to develop a few bad sectors but almost none of mine have died. One possible explanation, of course, would be that manufacturing standards have improved and that drives now recover from and work around shock damage better than they used to...
     
  5. edzieba

    edzieba Virtual Realist

    Joined:
    14 Jan 2009
    Posts:
    3,909
    Likes Received:
    591
    Nowadays that list is too big to fit on a sticker so it's just stored on the drive controller as the P-list.

    Per-sector probability of a failure (usually a URE) has stayed the same or even dropped due to the sectors getting physically smaller and closer together, while the number of sectors has increased. Also debris: a few nm wide chunk of crud getting plinked into the platter surface as the R/W head whizzes over might not have immediately killed a physically larger sector on an older drive, but now that may take out tens of sectors in one go.
    Conversely, the damage threshold that starts to have a measurable effect on a newer drive may be unnoticeable on an older drive. And the level of damage you'd see on an older drive once it starts reporting bad sectors would be seen on a newer drive as "Blimey, half the bloody disc is gone!".
     
    Last edited: 4 Aug 2017
    boiled_elephant likes this.

Share This Page