Re: Copying what I can from bit rotted disks?

["Jayson Smith" <ihatespamratguynospamplease@insightbb.spamsucks.com>]

Hi,

     I just discovered a bit more information. The files that are unreadable 
seem to be on the innermost tracks on both sides of the disk. I can tell by 
the distance the head travels as it recalibrates after a read error. I have 
taken the step of copying off all files that can be copied without errors 
onto a good disk, and imaging that. Are there any disk cleaning procedures 
that might let the data be read again, or is it more likely to just be a 
defect in the magnetic material on the innermost tracks? The errors happen 
on two different drives hooked up to the same system.
Jayson

"Michael J. Mahon" <mjmahon@aol.com> wrote in message 
DJSdncPwB6hrjBbWnZ2dnUVZ_tmdnZ2d@giganews.com">news:DJSdncPwB6hrjBbWnZ2dnUVZ_tmdnZ2d@giganews.com...
> Jayson Smith wrote:
>> Hi,
>>
>>      I have one disk in particular that has sadly become a victim of bit 
>> rot. It's a flippy with various music programs on both sides. It loads 
>> the Textalker software, and was part of a set of talking disks with lots 
>> of public domain software on them. It is DOS 3.3. I have no idea who put 
>> together this collection of disks, or I'd ask if anyone has a 
>> replacement, as well as any other disks in the set I don't have. There 
>> were probably over fifty of these disks. In any event, on this particular 
>> disk, both sides have been victimized by bit rot, and some of the 
>> programs are not copyable. I assume, once bit rot has occurred, there's 
>> virtually nothing that can be done to try to bring those files back from 
>> the grave?
>
> You don't mention whether you have tried reading this disk on at least
> two different drives--sometimes poor margins on one drive cause read
> errors that another drive can read fine.  Margins can be degraded by
> many things, including dirty heads, out-of-tolerance speed, and even
> poor alignment (rare).
>
> Generally, a controller card will not affect drive margins, so changing
> controller cards is less likely to make a difference.
>
> If the disk(s) are still bad on two or more drives, then a disk verify
> will let you know which sectors are unreadable.  If they (or a subset of
> them) are the same on two drives, then they are likely bad on the disk.
>
> Even then, the disk may be recoverable either completely (if the bad
> sectors are unallocated) or partially (the files containing the bad
> sectors will, of course, be incomplete).
>
> If the errors are in the file system metadata, like the directory, then
> the disk may still be recoverable by a "repair" program that will find
> the track/sector lists and rebuild the directory.  Such repairs are best
> done on a sector copy of the damaged disk, so that no writes are done to
> the damaged disk at all.
>
> Before getting deep into recovering a disk with errors, it's worth
> trying a "persistent" copy program, since the sectors may read correctly
> every once in a while.  ProSel's "Volume Copy" is such a copier--it will
> try repeatedly to read a bad area and, upon failure, will give an option
> to continue copying substituting a block of zeros for the bad block
> (it's a ProDOS-based copier, so it deals in blocks, not sectors).
>
> If nothing works, I'd be happy to try to recover the disk for you.
>
>>                          If that's true, than I'd love to find a program 
>> that would take the disks, and copy all the files it can to another disk. 
>> Possibly retry a few times if an error occurs, but if all its efforts are 
>> for naught, don't stop the process entirely, just go on to the next file, 
>> and make a notation of which files could not be copied. Obviously, FID 
>> will copy all files, but when it hits an I/O error, it stops. I can 
>> manually copy each file, making a list of those that can't be copied, but 
>> surely there must be a more automatic way. Something you can just set and 
>> forget, and when you come back, it says, "I copied this disk, I could not 
>> get the following files..." Any thoughts?
>
> I'm sure there's a modification of COPYA that will do that, but I don't
> have it myself.  The "driver" for COPYA is Applesoft, so you might take
> a look at it and see if you can modify it to just print out the bad
> filename and go on to the next file instead of quitting.
>
> -michael
>
> NadaNet and AppleCrate II: parallel computing for Apple II computers!
> Home page: http://home.comcast.net/~mjmahon
>
> "The wastebasket is our most important design
> tool--and it's seriously underused."