Solved Bad sector explanation

[Computer0304]

Note: I know what bad sectors are but I don't fully understand how they are created.
1. What causes bad sectors to be created? Is it a physical problem or a error in the OS?
2. Is there a way to prevent them from being created other than just using chkdsk?
3. Could external drives get bad sectors?
4. Do SSDs get bad sectors?

 

[TVeblen]

1. A physical problem.
Typically it is a block or sector of the platter that cannot be written to and/or read. A spec of dust or tiny scratch will do it.
Sometimes it can be a sector that has been corrupted but can be revived. But that is rare.

2. Not really. Don't drop the drive, no sudden power outages, no hot/cold environment extremes.

3. Yes.

4. Trick question. A SSD does not have spinning disks and so does not technically have blocks/sectors, but the individual NAND cells can go bad, so it is pretty much the same thing.

 

[Computer0304]

Don't hard drives get soft bad sectors too which is a cluster of storage that is not working but can be recovered? Also, does the answer to #4 mean that chkdsk is not effective for ssds?

 

[TVeblen]

My (limited) understanding of the function on chkdsk is that it focuses on the File System through the MBR and File Table, finding and correcting pointers to nowhere, fragments without a home, and other lost and mangled file system artifacts.

So running chkdsk in basic mode should be fine on an SSD.

What you would not want to do on an SSD is check the box that says "find and recover bad sectors". That is the job of the drive's firmware and TRIM. I would think that chkdsk could screw that up.

On a spinning disk, "find and recover bad sectors" does exactly what you suggest: it finds blocks/sectors that have been marked bad but are not really bad ("soft bad sectors") and remarks them OK.
A SF member explained this process in detail a past post, but I can't recall the specifics off the top of my head.

 

[GeneO]

Don't hard drives get soft bad sectors too which is a cluster of storage that is not working but can be recovered? Also, does the answer to #4 mean that chkdsk is not effective for ssds?

Hard drives and other storage devices have error correcting codes that can correct multiple bit errors to a extent. These are soft errors and are normal but not reported - they are not an indication that the a sector is bad, but if they are high could indicate an imminent failure.

However, I think you are talking about reallocated sectors. These are sectors that could still be read but there is some indication that they are bad, so the data is copied to a new sector (from a pool of spares) and the bad sector gets put on the reallocated sector list so it will not be reused. Logical addressing maps logical to physical so this happens transparently.

If you have any significant number of reallocated sectors, it is usually a sign the disk is about to go south.

You can also use chkdsk /R to check for bad sectors and map them out, but the hard drive internal checking should be sufficient.

TRIM doesn't have anything to do with discovering bad sectors on an SSD though.

 

[andrew129260]

A ssd tricks windows into thinking its a ide or sata Mechanical drive. Windows then treats it like such. The controller and trim handle things. Lets say windows tells the hard drive to write to this certain cell block. What trim and the controller do is lie to windows and say yes we have done that when in reality they have not as there is no reason to because its not a physical disk with platters.

http://www.pcworld.com/article/2038...-ssds-what-makes-these-speedy-drives-hum.html

http://arstechnica.com/information-...revolution-how-solid-state-disks-really-work/ <----- A lot of info

"Having a chunk of cache sitting there lets the SSD quickly receive data that it needs to write, even if it's too busy to actually write it at the moment; the data sits in the SDRAM cache until the controller is able to find time to send it down and actually commit it to NAND. All this happens transparently to the computer and you, the end user—regardless of whether or not the data has actually been written, the SSD controller reports back to the operating system that the write was completed successfully."

 

[Computer0304]

Thank you for replying.

 

[GeneO]

SSD isn't really that different than hard drives, there are no "tricks" and it has nothing to do with being mechanical. The OS addresses either by Logical address, and the SSD or HDD maps these to physical address. Under the hood they are a lot different though. In fact, the next generation of hard drives that use shingling are going to be more similar to SSD under the hood in that the drive won't be able to rewrite data without first reading out adjacent data, writing the data, then rewriting the adjacent data. Maybe they will use the same TRIM function as SSD to avoid this wherever possible.