Summary of Recent Data Recovery Methods; Spinrite 5.0

G

Guest

Guest
Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

My experience with Spinrite 5.0 and failing HD/floppies.

(I have no grudge to bear, purpose of post is payback for good info
gotten here; represents one user's experience, but is also meant to
counteract all the Sprinrite hype)

I had a 3 gig hd with about 5-10% bad sectors. Other HD recovery
programs stalled or would take forever to scan disk.

I had already recovered using simple xcopy most of the hd's data, with
guestimate of 5% of files that gave read errors.

After a review of most of the DR software out there, I chose not to
clone with dispatch due to it's long processing time, already recovered
most data and DP had no special recovery capabilities other than
skipping bad sectors.

Tried Drive Rescue; pcirecovery-both stalled. Get Data Back gave 24 hour
estimate- extremely slow in dealing with bad sectors.

Since I had pretty much nothing to lose, I tried Sprinrite 5.0, which I
got for free.

First I tried it on several floppies. Out of about 10 floppies I was
able to get the data back on about 6. One floppy was made totally
unusable by SR-system would not even recognize floppy was in drive. 3
others scrambled data so bad, no sense could be made of it-so much for
it's statistical processing of bad sectors, ha. Most of the floppies,
even those that I was able to recover data from, went belly up after
trying to use them first time. But about 60%, at least I got the data.

On the hard drive, totally different story. SR started going fairly
rapidly on the first portion (first 5% of drive) which, according to the
program contained no bad sectors. It got increasingly slower and slower,
with estimated finish times increasing steadily from 9 hours to 36
hours,when it then froze up (locked up my system) and I had to reboot.
The drive was toast after using SR 5.0 to "recover" the data. This drive
could have been on the verge of dieing anyways, who is to say.

Conclusion: xcopy data off failing drives, then clone if important data
still remains. ONly use Spinrite on drives that there is nothing to lose
on. My experience also leads me to believe that the extremely long
processing time of Spinrite and it's numerous rereads probably put more
stress on the drive than is acceptable even for routine maintenance.
It's data recovery abilities aren't any better than Norton's Disk Doctor
in my experience and it MAY kill off drives that are marginal during the
recovery process.
 

joeP

Distinguished
Dec 31, 2007
264
0
18,780
Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

"lexluther" <lexluther@krypton.gov> wrote in message
news:GNU969991810C027lexlutherkryptongov@213.155.197.138...
> My experience with Spinrite 5.0 and failing HD/floppies.
>
> (I have no grudge to bear, purpose of post is payback for good info
> gotten here; represents one user's experience, but is also meant to
> counteract all the Sprinrite hype)

As far as I know there's no Spinrite hype here.

>
> I had a 3 gig hd with about 5-10% bad sectors. Other HD recovery
> programs stalled or would take forever to scan disk.
>
> I had already recovered using simple xcopy most of the hd's data, with
> guestimate of 5% of files that gave read errors.
>
> After a review of most of the DR software out there, I chose not to
> clone with dispatch due to it's long processing time, already recovered
> most data and DP had no special recovery capabilities other than
> skipping bad sectors.

It's called DiskPatch - 3 Gb shouldn't take that much time at all - and
there's a bit more DiskPatch can do than skipping sectors it can't read
immediately. But you need to make up your mind: You either clone a disk as
fast as possible (as, as you will later the say the disk may die if you
spent too much time on it) - or - you try to actually recover data from the
problem sectors which takes time.

>
> Tried Drive Rescue; pcirecovery-both stalled. Get Data Back gave 24 hour
> estimate- extremely slow in dealing with bad sectors.

Yes, all that software uses Windows API to read a disk and basically they're
waiting for Windows to discover it can not read the disk. Bottomline is that
this type of software is not really equiped to deal with 'bad' disks and
that they're primarely written with file system issues in mind.

>
> Since I had pretty much nothing to lose, I tried Sprinrite 5.0, which I
> got for free.
>
> On the hard drive, totally different story. SR started going fairly
> rapidly on the first portion (first 5% of drive) which, according to the
> program contained no bad sectors. It got increasingly slower and slower,
> with estimated finish times increasing steadily from 9 hours to 36
> hours,when it then froze up (locked up my system) and I had to reboot.
> The drive was toast after using SR 5.0 to "recover" the data. This drive
> could have been on the verge of dieing anyways, who is to say.

Probably.

>
> Conclusion: xcopy data off failing drives, then clone if important data
> still remains.

Yes, that's common sense. If you can still copy files using normal methods
then do so. Using file copying software, other than sector by sector disk
cloning software which copies all sectors regardless what's in them, allows
you to prioritize and get the most important files first.

> ONly use Spinrite on drives that there is nothing to lose
> on. My experience also leads me to believe that the extremely long
> processing time of Spinrite and it's numerous rereads probably put more
> stress on the drive than is acceptable even for routine maintenance.

You're mixing things up. Using SR on a disk with 10% bad sectors isn't
routine maintenance. It's the 10% bad sector causing the delay. So, when
used for routine maintenance you probably wont get those times.

> It's data recovery abilities aren't any better than Norton's Disk Doctor
> in my experience

That's because you didn't look close enough. NDD and SR do fundametally
different things.

> and it MAY kill off drives that are marginal during the
> recovery process.

That's pretty obvious, but can happen with any software.

--
Joep
 
G

Guest

Guest
Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

on 20 Jul 2005, "Joep" <available@request.nl> wrote in
news:be2a4$42dedfd0$3eddca68$29524@nf1.news-service.com:

> "lexluther" <lexluther@krypton.gov> wrote in message
> news:GNU969991810C027lexlutherkryptongov@213.155.197.138...
>> My experience with Spinrite 5.0 and failing HD/floppies.
>>
>> (I have no grudge to bear, purpose of post is payback for good
>> info gotten here; represents one user's experience, but is also
>> meant to counteract all the Sprinrite hype)
>
> As far as I know there's no Spinrite hype here.

I was alluding to the SR web page, not this group.
>
>>
>> I had a 3 gig hd with about 5-10% bad sectors. Other HD
>> recovery programs stalled or would take forever to scan disk.
>>
>> I had already recovered using simple xcopy most of the hd's
>> data, with guestimate of 5% of files that gave read errors.
>>
>> After a review of most of the DR software out there, I chose
>> not to clone with dispatch due to it's long processing time,
>> already recovered most data and DP had no special recovery
>> capabilities other than skipping bad sectors.
>
> It's called DiskPatch - 3 Gb shouldn't take that much time at
> all - and there's a bit more DiskPatch can do than skipping
> sectors it can't read immediately. But you need to make up your
> mind: You either clone a disk as fast as possible (as, as you
> will later the say the disk may die if you spent too much time
> on it) - or - you try to actually recover data from the problem
> sectors which takes time.

So I made a typo, shoot me? DiskPatch according to it's own
manual, simply tries to reread the bad sectors and skips them if
they are unreadable. If I am wrong about that let me know. IOW,
what does DP do that is unique to recovering bad sectors? Also,I
find rather suspect the claims that once a drive is cloned,
together with it's skipped sectors, that it "often" is repaired
such that it is like it was before all those bad sectors became
unreadable, or so their docs. imply.

>
>>
>> Tried Drive Rescue; pcirecovery-both stalled. Get Data Back
>> gave 24 hour estimate- extremely slow in dealing with bad
>> sectors.
>
> Yes, all that software uses Windows API to read a disk and
> basically they're waiting for Windows to discover it can not
> read the disk. Bottomline is that this type of software is not
> really equiped to deal with 'bad' disks and that they're
> primarely written with file system issues in mind.

Yes, these are examples of more software that makes big claims but
cannot deliver. They all talk the talk, but haven't found a one
that can walk the walk, although I haven't tested that many.
>
>>
>> Since I had pretty much nothing to lose, I tried Sprinrite 5.0,
>> which I got for free.
>>
>> On the hard drive, totally different story. SR started going
>> fairly rapidly on the first portion (first 5% of drive) which,
>> according to the program contained no bad sectors. It got
>> increasingly slower and slower, with estimated finish times
>> increasing steadily from 9 hours to 36 hours,when it then froze
>> up (locked up my system) and I had to reboot. The drive was
>> toast after using SR 5.0 to "recover" the data. This drive
>> could have been on the verge of dieing anyways, who is to say.
>
> Probably.

Up until using SR, however, I had no problem accessing the drive
ANd there are other several complaints by others in past posts
that SR killed their drives. The jury is out, and until someone
attempts comprehensive independent tests (which may be
unlikely/difficult), it will continue to be out.

>
>>
>> Conclusion: xcopy data off failing drives, then clone if
>> important data still remains.
>
> Yes, that's common sense. If you can still copy files using
> normal methods then do so. Using file copying software, other
> than sector by sector disk cloning software which copies all
> sectors regardless what's in them, allows you to prioritize and
> get the most important files first.

The big mistake I made was not giving xcopy a command line option
to make a log of the files it could not get. That's assuming there
is such an option, I never checked.
>
>> ONly use Spinrite on drives that there is nothing to lose
>> on. My experience also leads me to believe that the extremely
>> long processing time of Spinrite and it's numerous rereads
>> probably put more stress on the drive than is acceptable even
>> for routine maintenance.
>
> You're mixing things up. Using SR on a disk with 10% bad sectors
> isn't routine maintenance. It's the 10% bad sector causing the
> delay. So, when used for routine maintenance you probably wont
> get those times.

NO, I clearly understand the diff. between using SR on a failing
drive and on a good drive. I just think it really doesn't do much
in either case and is potentially harmful on good drives or drives
that have some life left, by overexercising the drive and thus
increasing the odds it will fail. Besides "smart" technology if it
works and I am no expert does what SR is supposed to do
automatically in the case of giving you warnings a drive is about
to fail. As I used it it was ok, since I had already gotten most
everything I could from a drive that may have been faulty and the
size of which was such that I wasn't going to use it again
anyways.
>
>> It's data recovery abilities aren't any better than Norton's
>> Disk Doctor in my experience
>
> That's because you didn't look close enough. NDD and SR do
> fundametally different things.

Yet, in my limited experience the results, amount of data
recovered, are about the same. To think of it, that might be one
of the big faults of SR, that it does nothing to address logical
errors on the drive whilst it is repeatedly attempting to read,
again and again "bad" areas on the drive. When SG says in his
manual for SR 5 to use checkdisk /f FIRST, before using SR, that
tells me he either doesn't know how to write a manual or knows
little about recovering data from failing drives. And, before you
say it, I disagree that he specifies that that is for maintenance
scans only. The way the manual reads the user is directed to use
checkdisk /f, which doesn't even work on most of todays systems
(no mention is made of scandisk), PRIOR to using SR. That, imo, is
clearly a bad direction/error instruction.

>
>> and it MAY kill off drives that are marginal during the
>> recovery process.
>
> That's pretty obvious, but can happen with any software.

Not so obvious to SG fans. They think that SR is some kind of
magic elixir, that will make their bunny rabbit keep going and
going and going, when in fact it is making their bunny rabbit run
marathons to see if he will develop arrythmia.

>
> --
> Joep
>
>
>
 

joeP

Distinguished
Dec 31, 2007
264
0
18,780
Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

"Compulsive Rod" <rod_speed@yahoo.com> wrote in message
news:3kfhbbFu4lk3U1@individual.net...
> >>>>>> (I have no grudge to bear, purpose of post is payback for
> >>>>>> good info gotten here; represents one user's experience,
> >>>>>> but is also meant to counteract all the Sprinrite hype)
>
> >>>>> As far as I know there's no Spinrite hype here.
>
> >>>> I was alluding to the SR web page, not this group.
>
> >>> Then take your business there ...
>
> >> Why should he ? And its unlikely they
> >> would let him post there anyway.
>
> > Well, for starters they are actually letting him post there. And sure,
> > lexie can post here but sofar he didn't seem to stir a lot of ghosts
yet.
>
> Irrelevant to whether he is welcome to post here.

As always your mangled mind manages to take a thread into a completely
different and irrelevant direction. I will not follow.

Bye Rod.

--
Joep