Last week my SMART diagnostics utility, CrystalDiskInfo, reported that the external hard drive that I was saving my backups to had suddenly reported 900+ reallocated sectors. I double-checked to confirm, then ordered a replacement drive.

I spent all of this week copying data from that drive to the new drive. But toward the end of the copy, something peculiar happened. CrystalDiskInfo popped up an alert that the reallocated sector count had gone back down to 0.

I know that when SMART detects a read error on a block, it adds that block to the current pending reallocation list. If it subsequently is successfully written or read later, it is removed from the list and assumed to be fine, but if a subsequent write fails, it is marked bad and added to the reallocated sector count.

What concerns me most is that I've never read anywhere that a sector can be recovered as "good" after it has been marked as a bad sector and remapped.

I've just finished running an extended SMART diagnostic, and it found no surface errors. Now I'm doubtful that the manufacturer will honor a warranty claim if the SMART info does not report any problems.

Has anyone had this happen? If so, then is the drive, indeed, okay, or should I be concerned about an imminent failure?

3 Answers 3


If it were me, I wouldn't trust it. It's already been marked as bad. Not as a warning, but as flat out "This sector is bad, no doubt about it."

By nature of SMART, it's programmed the way it is to find real problems vs. good drive (duh). If it says they were bad, I'd stick with that.

Ask yourself one question: Is it easier to replace a drive or the data?

  • Thanks for the cautionary recommendation. I guess I was hoping for something a bit more definitive regarding what actually happened, or an anecdote from someone else who has witnessed this same type of issue, since the drive first reported a huge leap in errors, then later reported no record of the errors. The scary part is that, if I hadn't been at my desk when the two notifications ("0 --> 990+ reallocated sectors" and "990+ --> 0 reallocated sectors") were triggered, I never would have known.
    – rob
    Nov 7, 2012 at 19:03

I contacted Seagate regarding the issue and they did not provide any further information, but they didn't have any problem honoring the warranty.

I just feel bad for the sucker who gets the drive that I sent in, if Seagate just ends up running SMART diagnostics, and resetting the SMART counters and boxing the drive back up when they fail to find any problems.


I had the same problem some days ago with a drive with 100+ errored sectors. I used dd to check the surface... and magically the errored sectors where gone.

So I read Hitachi's documentations. One of them garanty their profesionnal drives don't do this (they use a special sentence for that, but I forgot which). Nothing about the consumer drives, so I guess they hide those sectors in certain circumstances.
I didn't read other companies datasheets in detail, but they probably do the same thing.

My personnel conclusion is : manufacturers no longer (if they already did) report correct error sector count on consumer grade drives. This is probably to let their drive appear better than in reallity.

  • Just to clarify, were the "errored sectors" listed in the "pending sector count" or in the "reallocated sector count"? The reason I ask is that a sector marked as pending will be marked good upon the next successful read or write.
    – rob
    Oct 10, 2012 at 18:52
  • 1
    If you can find any references to support this theory that SMART reporting on consumer drives is less trustworthy than that on enterprise drives, please add this to your answer and I'll mark it as the accepted answer. But as it stands, your answer seems too speculative since it's unclear to me whether "errored sectors" in your case referred to Current Pending Sectors, Reallocated Sectors, or just filesystem errors.
    – rob
    Nov 7, 2012 at 18:56

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .