CHKDSK - discuss: convert lost chains to files

[Dumb_Question]

I have run CHKDSK on my 80GB C:\ HD, and found some lost chains, but no bad sectors (hooray !)

Thus there is no need to run CHKDSK /R

If I run CHKDSK /F, I will get the message "Convert lost chains into files ? (Y/N).  What should I answer ?

If "Y", do I  want many FILEXXXX.CHK files which might possibly contain something useful ? (I haven't missed anything that I can remember in the last 10 years), but that I know I can delete *.CHK files, and thus I know I can make that disk space available. 

Or do I answer "N" in which case what happens to the HD space occupied by the lost chains ? 

Are there any other considerations ?

After I complete this operation, I will defrag this disk, then I feel it be reay for some more installations.

Dumb_Question
8.June.2013
Compaq Presario S5160UK DT261A under XP/SP3
Processor - Celeron 2.7 GHz
Motherboard - MSI MS-6577 v2.1
RAM - 1GB + 512MB (1GB +1GB max) DDR PC2700

[Allan]

chkdsk is a useless command. You always want to run chkdsk /r

[Geek-9pm]

chkdsk is a useless command. You always want to run chkdsk /r

Not useless. It is a command that informs with no action.

Type CHKDSK without parameters to check the current disk.

http://www.computerhope.com/chkdskh.htm

What is not documented is that running CHKDSK without parameters will help reduce errors on some hard drives. So it really is a useful command.

[patio]

Without parameters chkdsk does nothing except examine the HDD and create a logfile so how would it reduce errors on a drive ? ?

[Geek-9pm]

Without parameters chkdsk does nothing except examine the HDD and create a logfile so how would it reduce errors on a drive ? ?

Exactly. I said it was not documented.
Some, even many,  had drives have a firmware algorithm that counts the number of read errors on a drive by sector. Any program that reads all of the drive will cause the error count to go up if there are any weak sectors. Next time a disk write is made, the HDD firmware will force a re-assignment of weak sectors and thus eliminate further errors for those few bad sectors that had not been timed out previously.

I had to say this, because every time that is a discussion about CHKDSK, somebody will say it does nothing useful.  Not so!

[Dumb_Question]

People, thank you for bothering to read my post and offer your comments in so short a time.  I hope there will be more responses.

My request for a discussion about the detail of what I should do with lost chains/clusters, however, remains pertinent.

I, too, thought that running 'CHKDSK' without parameters changed nothing, but was informative.

Dumb_Question
9.June.2013

[Allan]

Exactly. I said it was not documented.
Some, even many,  had drives have a firmware algorithm that counts the number of read errors on a drive by sector. Any program that reads all of the drive will cause the error count to go up if there are any weak sectors. Next time a disk write is made, the HDD firmware will force a re-assignment of weak sectors and thus eliminate further errors for those few bad sectors that had not been timed out previously.

I had to say this, because every time that is a discussion about CHKDSK, somebody will say it does nothing useful.  Not so!

If you don't mind, please post SOME documentation to back up what you're saying.

Until such time, please let me reiterate - running checkdisk with no parameters serves no useful purpose.

[BC_Programmer]

Exactly. I said it was not documented.

if it is not documented, how did you learn about it? divine revelation?

Some, even many,  had drives have a firmware algorithm that counts the number of read errors on a drive by sector.

Most drives have firmware that reallocated weak or unreadable sectors without the knowledge of external software/hardware components. This is true.

Any program that reads all of the drive will cause the error count to go up if there are any weak sectors.

Makes sense, however- chkdsk doesn't read the entire drive, it only reads the File system. In order to read the entire drive, you need to use /r. Which also invokes the FS repair mechanisms in earlier passes. (The surface scan would result in the logic you mention).

Next time a disk write is made, the HDD firmware will force a re-assignment of weak sectors and thus eliminate further errors for those few bad sectors that had not been timed out previously.

Actually the typical algorithm for reallocated sectors reallocates the sector when it reads the sector; such reallocation is not deferred to write operations on that sector. The reallocation occurs then and there, it has trouble reading the sector but can get the data still; so it copies that data into one of the spare sectors, marks off the sector mapping (sector XXX is reallocated sector YYY) and continues on.

I had to say this, because every time that is a discussion about CHKDSK, somebody will say it does nothing useful.  Not so!

chkdsk without any arguments will do the same amount of useful work in the capacity you mention as TYPE-ing the contents of every text file on the disk would. by default it only deals with FS structures.

[Geek-9pm]

Thank you, BC_Programmer.
I meant it is not documented in this context. In is documented elsewhere, if papers about HID firmware.
But now it is documented. Is this a kind of recursive documentary?

[Allan]

chkdsk doesn't read the entire drive, it only reads the File system. In order to read the entire drive, you need to use /r. Which also invokes the FS repair mechanisms in earlier passes. (The surface scan would result in the logic you mention).

chkdsk without any arguments will do the same amount of useful work in the capacity you mention as TYPE-ing the contents of every text file on the disk would. by default it only deals with FS structures.

EXACTLY!!!

[patio]

Now what is documented ? ?

[Allan]

It's from the school of "If you say it loud enough and often enough, people will believe it's true" documentation

[patio]

Ahhh ...OK.