HARD DRIVE FAILURE, BAD CLUSTERS

[_Shadow_]

So, it happens like this:


Someone comes to you, because you are very experienced, thus a "wizard" when it comes to computers.


"Hey <instert your name here with sweet cute voice>. How is it going? I need your help! My computer will not boot start".


And you know that this is not gonna be good. But you are the only one that this person really trusts and stuff like that, so you have to help out that person.


So you realize soon that the hard drive has failed.


So you ask: "Do you have any backups of your data?"


And you get the following answer with 67% probability: "What is a backup?"


Nah just kidding! The answer with 67% probability is probably: "No"


We always shout out loud: Keep backup copies of your important files. Photos, personal videos, files related to your work or business, all must be kept as a backup safe. I found DVD not reliable at all. CDs are so - so and very expensive as an option, not to mention that it is not a 100% reliable solution and it demands a lot of space to store your data. SSDs on the other hand are very nice as a solution, but I have never ever seen one buing an ssd for backup purposes. Everyone uses good quality usb sticks. Everyone keeps backup, right? Everyone except the friend that will have a failing HDD!


So what would you do? Maybe try to recover data instantly if possible.  After you salvage everything you could, you reach at a point that you might have to recreate partitions, format the HDD or try something more "exotic" as a command line set of tests-fixes.


And here comes the real deal.


Case:

• WD HDD failing (500GB capacity).
  
• WD Lifeguard Diagnostics, Fail returning "Too many bad sectors".
  
• chkdsk recognizes and logs 2328 bad clusters at Stage 5
  

Question:


Can an HDD with SO many bad clusters be reliable after the chkdsk finishes successfuly its tasks?


I would like to hear from IT people but also from experienced users that had similar problems in the past.


Is there a number of bad clusters that determines an HDD useless?


I really don't know what to tell to my friend. Buy a new HDD to be 100% safe or keep this one? What is the risk level here?

 

[AceOfAces_Mod]

This happened to me when my laptop had it's stock HDD. I couldn't rescue it at all. Chkdsk found errors but it did jack. I had to buy a new one. Luckily most of my data were on my phone and my Walkman, so I could restore them (and Immortal Sins was, thankfully backed up properly on the Steam Cloud, with minor loss of progress).

 

[_Shadow_]

This happened to me when my laptop had it's stock HDD. I couldn't rescue it at all. Chkdsk found errors but it did jack. I had to buy a new one. Luckily most of my data were on my phone and my Walkman, so I could restore them (and Immortal Sins was, thankfully backed up properly on the Steam Cloud, with minor loss of progress).

Speaking of luck WOW!


Steam Cloud and cloud in general, can go wrong in many ways!

 

[djDarkX]

Sorry to say, but that hard drive is basically on life support.  Have them back up whatever information they can via external enclosure or setting it as a secondary and then toss it.  It will only degrade further.

 

[Andar]

@Dreadshadow


sadly - no. Not when chkdsk already reports bad sectors, that is too far.


You have to know that all harddrives have an internal reserve of sectors that are never given out to the operating system (or in this case chkdsk).


This is because of production tolerances - if an HDD is supposed to have 50GB (fictive numbers as example), then internally it is usually produced with 51GB (and that does not even count the differences between 1000 and 1024-based calculations, I'm talking about real magnetic space here).


This is because due to production tolerances, a drive produced with 51 GB space will result in a more or less random number between 50.2 and 50.8 GB of intact space.


Then the controller is told to display 50GB to the operating system and keep the difference of let's say 0.5GB as an internal reserve.


And whenever the controller detects another bad sector due to transport shocks, overheating or failing age, then that bad sector is hidden and remapped to a reserve sector. That is one of the reasons for Window crashes when an computer gets older - a file was stored on a bad sector, the computer crashed due to file failure but the user won't find any cause for that failure because the bad sector on the HDD had already been remapped.


However, if chkdsk begins to report bad sectors, that means that the HDD controller already used up the entire internal reserve for other bad sectors and can no longer hide additional sector failures. Which is what the poster above meant when he said that that HDD is already on life support - either it is too old or it was aged prematurely by overheating, but nothing will stop that HDD from getting more bad sectors the longer it works.

 

[LxCharon]

It has how many bad clusters!? On a 500 gb HDD? I'm sorry but that HDD needs to be taken out behind the shed and put out of it's misery. Hopefully you guys can save most of their files 

 

[_Shadow_]

Thanks everyone. ALL important files were rescued. It was a miracle but I rescued EVERYTHING.


The case was exactly what @Andar said. A failed sector that had a system file, made Windows unbootable.


I was almost 100% sure that the HDD is a goner. But there is always a possibility that there might be a solution I never heard of.


It seems not for this case.  


Well, sad news for my friend, I just wanted to make sure that bad clusters are 100% guaranteed bad news and there is nothing we can do about it.

 

[mogwai]

I'm probably on warm ice here with my HD, but about 5 years ago my iMac was restarting because of HD errors and the boot mode Snow Leopard DVD equivalent of your thing, the Disk Utility said there was bad sectors or whatever. I can't remember what happened to ever fix it or what the point of my story is, but I'm still on that computer today and Disk Utility says the HD is ship shape. A perverted Radio Shack guy probably swapped my HD out for his own personal inspection and left me a good copy. It took him way too long to air dust my iMac and for some reason I wasn't allowed in the store while he was doing it. Or else it's like they say, Mac don't crack. Posts various cracked screen iPod/iPad pics...

 

[_Shadow_]

 my iMac

Unix operating system's tool might refer to bad sectors differently than Windows chkdsk.


Thus we might never know.


I ran a second check and the bad clusters are now even more.


So,  I think I will suggest a new HDD.

 

[Andar]

@mogwai, @Dreadshadow


there are soft sector errors that are caused by a crash for unrelated reasons (for example power failure or reset).


When an operating system crashes at the same moment as it was writing on the HDD, then that specific sector might get wrong data and automatically be remapped as a soft bad sector.


It is called a "soft" error because those errors can be repaired by formatting - the formatting is nothing other than writing the sector structure on the magnetic disc, and that rewrites the damaged structure of such soft errors.


However, that is a rare case of no more than a handful of such sectors after a really bad crash - there is no way several thousand sectors can be soft errors.


But in mogwais case it probably was a soft error, or a non-aging related single bad sector (as said, there is an internal reserve to handle a few bad sectors that are inevitable due to production tolerances and transport shocks)


Because there is no way that an HDD with bad sectors due to age or overheating can survive another five years.

 

[_Shadow_]

I turst you @Andar and to be honest I was expecting your  reply since you are experienced in this topic.

 

[Andar]

Master's Degree in computer engineering and fifteen years of professional computer support do leave their traces ;-)

 

[_Shadow_]

Master's Degree in computer engineering and fifteen years of professional computer support do leave their traces ;-)

Oh! Explains a LOT I suppose.

 

[Reapergurl]

Oh, but here's a mega-dilemma.


What happens when three runs of chkdsk by different users causes three different results for the same HDD.


And I mean, three drastically different, unrelatable results.


Thus, this was the case of my first computer. I decided to SLAVE the 'failing' drive and get a new one for MASTER.


Two weeks later, start having the same problem with the brand new drive.


I found out something; if x86 system files get infected with a certain type of virus, it makes chkdsk (and other critical components) unreliable.


Make sure to always keep system files and registry clean, because false scan results can make things seem far worse than they are.


I'm glad that I learned these things, though the computer in question failed after twelve years due to too many upgrades (and it would of never handled anything past XP anyway).

 

[bgillisp]

@Reapergurl: I actually had that issue with my old x386 computer when I was in junior high. We later found out that something in the motherboard was causing the floppy drive to malfunction, so all chkdsk runs on that floppy also were unreliable. It only went away when we replaced the motherboard completely and upgraded to an x486.

 

[_Shadow_]

@Reapergurl I always check for any kind of malware, it's always a case, but thanks for mentioning this. I never thought that this would ever be an issue.


If the tests though follow a certain pattern, meaning bad clusters become more and more everytime, well it is obvious. The HDD is dying. Oh well, I will open it and use it as a show and tell class for hardware. I had an older one anyways and my friend told me that if it can't be fixed, I can keep it.

 

[Reapergurl]

@bgillisp I know now that if I had the x 64 version of the OS, I'd of never had these issues (the virus itself attacked only 32 bit OS' - go figure).


But back then, I wasn't about to pay five hundred bucks to upgrade the motherboard, processor and RAM to handle 64 bit.


@Dreadshadow Yep. I actually had a computer I built, with custom mish-mosh parts. Within a year the HDD started to slow, and had many, many bad sectors. I now know to be careful when buying refurbed parts, especially HDD. Thankfully, I was able to make three partitions and save my data by SLAVING it and getting a new one for MASTER.


Hey, I oughta lend you the HDD I shot with a double barrel twelve gauge shotgun because of that nasty malware called Dregol. Tell your pupils all about the nasty things malware can do, and can make people do. lol 

 

[_Shadow_]

Thankfully, I was able to make three partitions and save my data by SLAVING it and getting a new one for MASTER.

You really do BDSM your HDDs don't ya?

 

[Reapergurl]

@Dreadshadow lol

 

[_Shadow_]

One thing is for certain though. I know how to treat misbehaving components.

Got ya!