Problem with my HDD

Problem with my HDD

[Cookie-Monster]

Hi everybody

I hope you can help me. I formated on of my Server-HDDs with ReiserFS
version, which has been distributed in SuSE Linux 7.3. Sorry, but I
don't know the version, cause I'm know running SuSE Linux 8.0.

My Problem is, that the HDD cannot be mounted anymore:

1.   fileserver:/ # mount /dev/sda1 /ftp/apps/
2.   mount: Not a directory

The first line is the command and the second one is what "mount" tells
me about my HDD.
After this problem I ran reiserfsck:

      fileserver:/ # reiserfsck /dev/sda1
      reiserfsck 3.x.1b (2002)
      Will read-only check consistency of the filesystem on /dev/sda1
      Will put log info to 'stdout'

      Do you want to run this program?[N/Yes] (note need to type Yes):Yes
      ###########
      reiserfsck --check started at Wed May 29 19:22:53 2002
      ###########
      Replaying journal..
      0 transactions replayed
      Checking S+tree..

      Bad root block 4294967295. (--rebuild-tree did not complete)



      Aborted

So then I started reiserfsck with option --rebuil-tree as the manpage
says:

     fileserver:/ # reiserfsck /dev/sda1 --rebuild-tree
     reiserfsck 3.x.1b (2002)
       **********************************************************
       ** This  is  an  experimental  version  of  reiserfsck, **
       **              !! MAKE A BACKUP FIRST !!               **
       ** Don't run this program unless something  is  broken. **
       ** Some types of random FS damage can be recovered from **
       ** by  this  program,   which  basically   throws  away **
       ** the internal nodes of the tree and then reconstructs **
       ** them. This program is for use only by the desperate, **
       ** and is  of only beta quality.  If you are using  the **
       ** latest  reiserfsprogs  and  it  fails  please  email **
       ** bug reports to reiserfs-list@namesys.com.            **
       **********************************************************

     Will rebuild the filesystem (/dev/sda1) tree
     Will put log info to 'stdout'

     Do you want to run this program?[N/Yes] (note need to type Yes):Yes
     Replaying journal..
     0 transactions replayed
     ###########
     reiserfsck --rebuild-tree started at Wed May 29 19:27:02 2002
     ###########

     Pass 0:
     ####### Pass 0 #######
     Loading on-disk bitmap .. ok, 693365 blocks marked used
     Skipping 8486 blocks (super block, journal, bitmaps) 684879 blocks will be read
     0%....20%....40%....60%...                               left 171398, 4754 /sec


The check goes on til there are 171398 (more than 60%) blocks left then it hangs
up for a while before continueing. From this point on it continues
very slow only. From 150xxx it continues fast til 13039.
At this point the program quits without saying anything and the HDD is
hung up. The old version of reiserfsck, in the SuSe Linux 7.3 Edition
gave out a lot of "Block could not be read" or something like this and
the HDD hung up.

So I can't get the files on this HDD and my life depends on them, so
can someone HELP ME PLEASE?
I really need the data! Is there any possibilty of deleating the
unreadable blocks and rebuuilding the tree then? When this works, I
can get SOME of the information on the HDD back, because there are
only 2-3GB data on it and only 900MB of the data is really important.
But this 900MB are some of the last files I saved on the HDD.

I would be very happy, if someone has any idea of what I can do.

Jesse Schlüter

p.s.: The HDD is a IBM DRHS36V (36GB U2W-SCSI) if that is important.

--
Cookie-Monster
cookie.monster@starfall.de
http://www.starfall.de/

Re: Problem with my HDD

[Anders Widman]

Have  you tried a later version of reiserfs tools? Still, I'd wait for
some second or third suggestions.

//Anders

> Hi everybody

> I hope you can help me. I formated on of my Server-HDDs with ReiserFS
> version, which has been distributed in SuSE Linux 7.3. Sorry, but I
> don't know the version, cause I'm know running SuSE Linux 8.0.

> My Problem is, that the HDD cannot be mounted anymore:

> 1.   fileserver:/ # mount /dev/sda1 /ftp/apps/
> 2.   mount: Not a directory

> The first line is the command and the second one is what "mount" tells
> me about my HDD.
> After this problem I ran reiserfsck:

>       fileserver:/ # reiserfsck /dev/sda1
>       reiserfsck 3.x.1b (2002)
>       Will read-only check consistency of the filesystem on /dev/sda1
>       Will put log info to 'stdout'

>       Do you want to run this program?[N/Yes] (note need to type Yes):Yes
>       ###########
>       reiserfsck --check started at Wed May 29 19:22:53 2002
>       ###########
>       Replaying journal..
>       0 transactions replayed
>       Checking S+tree..

>       Bad root block 4294967295. (--rebuild-tree did not complete)



>       Aborted

> So then I started reiserfsck with option --rebuil-tree as the manpage
> says:

>      fileserver:/ # reiserfsck /dev/sda1 --rebuild-tree
>      reiserfsck 3.x.1b (2002)
>        **********************************************************
>        ** This  is  an  experimental  version  of  reiserfsck, **
>        **              !! MAKE A BACKUP FIRST !!               **
>        ** Don't run this program unless something  is  broken. **
>        ** Some types of random FS damage can be recovered from **
>        ** by  this  program,   which  basically   throws  away **
>        ** the internal nodes of the tree and then reconstructs **
>        ** them. This program is for use only by the desperate, **
>        ** and is  of only beta quality.  If you are using  the **
>        ** latest  reiserfsprogs  and  it  fails  please  email **
>        ** bug reports to reiserfs-list@namesys.com.            **
>        **********************************************************

>      Will rebuild the filesystem (/dev/sda1) tree
>      Will put log info to 'stdout'

>      Do you want to run this program?[N/Yes] (note need to type Yes):Yes
>      Replaying journal..
>      0 transactions replayed
>      ###########
>      reiserfsck --rebuild-tree started at Wed May 29 19:27:02 2002
>      ###########

>      Pass 0:
>      ####### Pass 0 #######
>      Loading on-disk bitmap .. ok, 693365 blocks marked used
>      Skipping 8486 blocks (super block, journal, bitmaps) 684879 blocks will be read
>      0%....20%....40%....60%...                               left 171398, 4754 /sec


> The check goes on til there are 171398 (more than 60%) blocks left then it hangs
> up for a while before continueing. From this point on it continues
> very slow only. From 150xxx it continues fast til 13039.
> At this point the program quits without saying anything and the HDD is
> hung up. The old version of reiserfsck, in the SuSe Linux 7.3 Edition
> gave out a lot of "Block could not be read" or something like this and
> the HDD hung up.

> So I can't get the files on this HDD and my life depends on them, so
> can someone HELP ME PLEASE?
> I really need the data! Is there any possibilty of deleating the
> unreadable blocks and rebuuilding the tree then? When this works, I
> can get SOME of the information on the HDD back, because there are
> only 2-3GB data on it and only 900MB of the data is really important.
> But this 900MB are some of the last files I saved on the HDD.

> I would be very happy, if someone has any idea of what I can do.

> Jesse Schlüter

> p.s.: The HDD is a IBM DRHS36V (36GB U2W-SCSI) if that is important.

> --
> Cookie-Monster
> cookie.monster@starfall.de
> http://www.starfall.de/

Re: Problem with my HDD

[Kuba Ober]

On ¶roda 29 maj 2002 01:49 pm, Cookie-Monster wrote:
> Hi everybody
>
> I hope you can help me. I formated on of my Server-HDDs with ReiserFS
> version, which has been distributed in SuSE Linux 7.3. Sorry, but I
> don't know the version, cause I'm know running SuSE Linux 8.0.
>
> My Problem is, that the HDD cannot be mounted anymore:
>
> 1.   fileserver:/ # mount /dev/sda1 /ftp/apps/
> 2.   mount: Not a directory

try
# ls -la /ftp/apps
# mkdir /ftp/apps
and say what happens

Also, try
# dmesg
and see if there are any read errors on your drive. if there are, you're out 
of luck... get a new drive

Cheers, Kuba

Re: Problem with my HDD

[Robert Brockway]

On Wed, 29 May 2002, Cookie-Monster wrote:

> So I can't get the files on this HDD and my life depends on them, so
> can someone HELP ME PLEASE?

Ok, it's too late to point out the virtue of backups (this time).  From
your description I'm personally suspecting a hardware issue with the drive 
but it's orth ruling out all other alternastives before accepting this.

If you can I'd consider getting an image of the drive.  It may help with 
data recovery later.  something like:

dd if=/dev/sda1 of=/bigfilesystem/diskimage

I don't believe it will be necessary to set a particular blocksize for 
the dd here but perhaps others might comment.

Keep this image aside (put it on another box infact).

You may find it is actually impossible to get much of an image from the 
drive.

Now the main reason I posted was this:  If you find that all else has
failed (and if I were you I'd wait for a few more suggestions from the
learned people here to come in) then there are companies who will attempt
data recovery even on physically broken drives.  These companies have
decent success rates with respect to data recovery but be prepared to pay
$$$.  The things they do to the drive are often time consuming and costly
resource-wise.

Ask local computer experts for a local establishment which would do this.
Rob

-- Robert Brockway B.Sc. email: robert@timetraveller.org  ICQ: 104781119
   Linux counter project ID #16440 (http://counter.li.org)
   blake: up 1 day,  3:50,  4 users,  load average: 0.49, 0.47, 0.61
   "The earth is but one country and mankind its citizens" -Baha'u'llah

Re[2]: Problem with my HDD

[Anders Widman]

> On Wed, 29 May 2002, Cookie-Monster wrote:

>> So I can't get the files on this HDD and my life depends on them, so
>> can someone HELP ME PLEASE?

> Ok, it's too late to point out the virtue of backups (this time).  From
> your description I'm personally suspecting a hardware issue with the drive 
> but it's orth ruling out all other alternastives before accepting this.

> If you can I'd consider getting an image of the drive.  It may help with 
> data recovery later.  something like:

> dd if=/dev/sda1 of=/bigfilesystem/diskimage

I've  heard  of  something  called  dd_rescue,  but  "dd  if=/dev/sda1
of=/path/image  conv=sync,noerror"  should  do the trick. The sync and
noerror   causes  the disk-copy to continue when there are read errors
and  fill in with blanks instead. If you don't do this you will end up
with  offset  errors  -  the  image  will be shorter than the original
drive, and it would be impossible to fix.

IF  you  have damaged sectors you will need alot of time, perhaps many
hours, or even days to copy a bigger disk.

//Anders

> I don't believe it will be necessary to set a particular blocksize for 
> the dd here but perhaps others might comment.

> Keep this image aside (put it on another box infact).

> You may find it is actually impossible to get much of an image from the 
> drive.

> Now the main reason I posted was this:  If you find that all else has
> failed (and if I were you I'd wait for a few more suggestions from the
> learned people here to come in) then there are companies who will attempt
> data recovery even on physically broken drives.  These companies have
> decent success rates with respect to data recovery but be prepared to pay
> $$$.  The things they do to the drive are often time consuming and costly
> resource-wise.

> Ask local computer experts for a local establishment which would do this.
> Rob

> -- Robert Brockway B.Sc. email: robert@timetraveller.org  ICQ: 104781119
>    Linux counter project ID #16440 (http://counter.li.org)
>    blake: up 1 day,  3:50,  4 users,  load average: 0.49, 0.47, 0.61
>    "The earth is but one country and mankind its citizens" -Baha'u'llah

Re: Problem with my HDD

[Oleg Drokin]

Hello!

On Thu, May 30, 2002 at 02:09:09AM +0200, Anders Widman wrote:

> > If you can I'd consider getting an image of the drive.  It may help with 
> > data recovery later.  something like:
> > dd if=/dev/sda1 of=/bigfilesystem/diskimage
> I've  heard  of  something  called  dd_rescue,  but  "dd  if=/dev/sda1
> of=/path/image  conv=sync,noerror"  should  do the trick. The sync and

Our attempts show that dd (even with conv=sync,noerror,notrunc) does not
help and produces broken images, where no data inserted in image for sectors
that cannot be read.

dd_rescue is shipped with SuSE.
Also it can be obtained from here:
http://www.garloff.de/kurt/linux/ddrescue/

> noerror   causes  the disk-copy to continue when there are read errors
> and  fill in with blanks instead. If you don't do this you will end up

No, for some reason it does not fills the blanks.
At least for us.

> with  offset  errors  -  the  image  will be shorter than the original
> drive, and it would be impossible to fix.

This is also not true, it will be possible to fix, but that would be
more expensive ;)

Bye,
    Oleg

Re: Problem with my HDD

[Billy Transue]

On Thu, 2002-05-30 at 00:17, Oleg Drokin wrote:
> Hello!
> 
> On Thu, May 30, 2002 at 02:09:09AM +0200, Anders Widman wrote:
> 
> > > If you can I'd consider getting an image of the drive.  It may help with 
> > > data recovery later.  something like:
> > > dd if=/dev/sda1 of=/bigfilesystem/diskimage
> > I've  heard  of  something  called  dd_rescue,  but  "dd  if=/dev/sda1
> > of=/path/image  conv=sync,noerror"  should  do the trick. The sync and
> 
> Our attempts show that dd (even with conv=sync,noerror,notrunc) does not
> help and produces broken images, where no data inserted in image for sectors
> that cannot be read.
> 
> dd_rescue is shipped with SuSE.
> Also it can be obtained from here:
> http://www.garloff.de/kurt/linux/ddrescue/
> 
> > noerror   causes  the disk-copy to continue when there are read errors
> > and  fill in with blanks instead. If you don't do this you will end up
> 
> No, for some reason it does not fills the blanks.
> At least for us.
> 
> > with  offset  errors  -  the  image  will be shorter than the original
> > drive, and it would be impossible to fix.
> 
> This is also not true, it will be possible to fix, but that would be
> more expensive ;)
> 
> Bye,
>     Oleg

I've suffered a very similar problem and posted to the list earlier, I
did, dd_resuce -l dd.log /dev/hdd5 ./hdd5 and winded up w/ this:
dd_rescue: (info): ipos:         0.0k, opos:         0.0k,
xferd:         0.0k
                   errs:      0, errxfer:         0.0k,
succxfer:         0.0k
             +curr.rate:        0kB/s, avg.rate:        0kB/s,
avg.load:  0.0%

it never changed, the file created hdd5 was 65536 bytes large.  The log
file was empty.  Am I doing something wrong?  Any info would be
appreciated.  Thanks in advance.

Re: Problem with my HDD

[Oleg Drokin]

Hello!

On Fri, May 31, 2002 at 12:15:48AM -0500, Billy Transue wrote:
 
> I've suffered a very similar problem and posted to the list earlier, I
> did, dd_resuce -l dd.log /dev/hdd5 ./hdd5 and winded up w/ this:
> dd_rescue: (info): ipos:         0.0k, opos:         0.0k,
> xferd:         0.0k
>                    errs:      0, errxfer:         0.0k,
> succxfer:         0.0k
>              +curr.rate:        0kB/s, avg.rate:        0kB/s,
> avg.load:  0.0%
> it never changed, the file created hdd5 was 65536 bytes large.  The log
> file was empty.  Am I doing something wrong?  Any info would be
> appreciated.  Thanks in advance.

I'd suggest you to write to Kurt Garloff, since this looks like a bug.

Bye,
    Oleg

Re: Problem with my HDD

[Oleg Drokin]

Hello!

On Thu, May 30, 2002 at 09:45:50AM +1000, Robert Brockway wrote:

> If you can I'd consider getting an image of the drive.  It may help with 
> data recovery later.  something like:
> dd if=/dev/sda1 of=/bigfilesystem/diskimage

That does not work on bad hard drives, resulting image would be almost useless.
BTW, this way it stops on first read error.

Bye,
    Oleg

Re: Problem with my HDD

[Robert Brockway]

On Thu, 30 May 2002, Oleg Drokin wrote:

> Hello!
> 
> On Thu, May 30, 2002 at 09:45:50AM +1000, Robert Brockway wrote:
> 
> > If you can I'd consider getting an image of the drive.  It may help with 
> > data recovery later.  something like:
> > dd if=/dev/sda1 of=/bigfilesystem/diskimage
> 
> That does not work on bad hard drives, resulting image would be almost useless.
> BTW, this way it stops on first read error.

Yeah, I know but I figure that a partial recovery is better than none at 
all.  You could always go through a binary dump or pay $$$ for a 
professional data recovery group to do a professional job of it.
Cheers,
	-Rob

-- Robert Brockway B.Sc. email: robert@timetraveller.org  ICQ: 104781119
   Linux counter project ID #16440 (http://counter.li.org)
   blake: up 1 day,  3:50,  4 users,  load average: 0.49, 0.47, 0.61
   "The earth is but one country and mankind its citizens" -Baha'u'llah

Re[2]: Problem with my HDD

[Cookie-Monster]

>> If you can I'd consider getting an image of the drive.  It may help with
>> data recovery later.  something like:
>> dd if=/dev/sda1 of=/bigfilesystem/diskimage

> That does not work on bad hard drives, resulting image would be almost useless.
> BTW, this way it stops on first read error.

If it stops at the first read error, can I put this image back onto
the formatted disk, so I have left the working blocks?



--
Cookie-Monster
cookie.monster@starfall.de
http://www.starfall.de/

Re[2]: Problem with my HDD

[Cookie-Monster]

Hi,

>> If you can I'd consider getting an image of the drive.  It may help with 
>> data recovery later.  something like:
>> dd if=/dev/sda1 of=/bigfilesystem/diskimage

> That does not work on bad hard drives, resulting image would be almost useless.
> BTW, this way it stops on first read error.

I made an image of the disk. It's 3GB. Can I do anything with this
thing now? i.e.: formatting the HDD and putting the image back on it.
Then there would be no unreadable blocks anymore.

--
Cookie-Monster
cookie.monster@starfall.de
http://www.starfall.de/

Re[3]: Problem with my HDD

[Anders Widman]

> Hi,

>>> If you can I'd consider getting an image of the drive.  It may help with 
>>> data recovery later.  something like:
>>> dd if=/dev/sda1 of=/bigfilesystem/diskimage

>> That does not work on bad hard drives, resulting image would be almost useless.
>> BTW, this way it stops on first read error.

> I made an image of the disk. It's 3GB. Can I do anything with this
> thing now? i.e.: formatting the HDD and putting the image back on it.
> Then there would be no unreadable blocks anymore.

Formatting  would  not  repair the broken sectors. If the drive itself
has spare sectors left, then they would be used instead.

If  there  are  bad sectors you should really have this disk replaced.
IBM  has  5  years warranty on all SCSI drives. You can fill out a RMA
form at http://www.storage.ibm.com/hdd/support/index.htm

And,  as  Oleg  wrote,  use ddrescue to make a complete backup of your
harddrive.  Make  a  backup of that image too, so you can recover data
from one of the images. 

//Anders

> --
> Cookie-Monster
> cookie.monster@starfall.de
> http://www.starfall.de/

Re: Re[2]: Problem with my HDD

[Chris Mason]

On Thu, 2002-05-30 at 07:58, Cookie-Monster wrote:
> Hi,
> 
> >> If you can I'd consider getting an image of the drive.  It may help with 
> >> data recovery later.  something like:
> >> dd if=/dev/sda1 of=/bigfilesystem/diskimage
> 
> > That does not work on bad hard drives, resulting image would be almost useless.
> > BTW, this way it stops on first read error.
> 
> I made an image of the disk. It's 3GB. Can I do anything with this
> thing now? i.e.: formatting the HDD and putting the image back on it.
> Then there would be no unreadable blocks anymore.

I would keep the original disk in its current state until you get all
the data back you can.  If you think you've got a good copy, take the
original out of the machine and put it somewhere safe.

I'd start by running the copy through the latest version of reiserfsck.

-chris

Re: Re[2]: Problem with my HDD

[Robert Brockway]

On 30 May 2002, Chris Mason wrote:

> > I made an image of the disk. It's 3GB. Can I do anything with this
> > thing now? i.e.: formatting the HDD and putting the image back on it.
> > Then there would be no unreadable blocks anymore.
> 
> I would keep the original disk in its current state until you get all
> the data back you can.  If you think you've got a good copy, take the
> original out of the machine and put it somewhere safe.
> 
> I'd start by running the copy through the latest version of reiserfsck.

This is addresses mainly to the original poster:  One of the great things 
about this approach is that you can mount the image loopback & test your 
progress at recovering data, while not risking further damage to the 
original data source.

Good luck with the recovery & keep us posted.

Btw, at this point I'd like to thank all the reiserfs developers.  It is a 
great filesystem, and I personally believe will end up being the defacto 
standard for Linux within a few years despite all of the competing 
journalling filesystems out there at the moment.  Diversity is great of 
course - keeps everyone on their toes, and there will be room for all of 
the filesystems in the Linux distros of the future.  I was a bit sad when 
xiafs was dropped from the kernel, but I didn't worry too much about the 
passing of extfs :)

Cheers,
	-Rob

-- Robert Brockway B.Sc. email: robert@timetraveller.org  ICQ: 104781119
   Linux counter project ID #16440 (http://counter.li.org)
   blake: up 1 day,  3:50,  4 users,  load average: 0.49, 0.47, 0.61
   "The earth is but one country and mankind its citizens" -Baha'u'llah

Re: Problem with my HDD

[Oleg Drokin]

Hello!

On Wed, May 29, 2002 at 07:49:29PM +0200, Cookie-Monster wrote:

> I hope you can help me. I formated on of my Server-HDDs with ReiserFS
> version, which has been distributed in SuSE Linux 7.3. Sorry, but I
> don't know the version, cause I'm know running SuSE Linux 8.0.
> My Problem is, that the HDD cannot be mounted anymore:
> 1.   fileserver:/ # mount /dev/sda1 /ftp/apps/
> 2.   mount: Not a directory
>       Bad root block 4294967295. (--rebuild-tree did not complete)

Well, so you in fact run reiserfsck --rebuild-tree before.
What caused you to run it for the first time?

> The check goes on til there are 171398 (more than 60%) blocks left then it hangs
> up for a while before continueing. From this point on it continues
> very slow only. From 150xxx it continues fast til 13039.
> At this point the program quits without saying anything and the HDD is
> hung up. The old version of reiserfsck, in the SuSe Linux 7.3 Edition
> gave out a lot of "Block could not be read" or something like this and
> the HDD hung up.

what does 'dmesg' output says after this?
If you have read errors, then namesys can help you only based
on http://namesys.com/support.html terms.

> I really need the data! Is there any possibilty of deleating the
> unreadable blocks and rebuuilding the tree then? When this works, I

This is possible, but all the data that was contained in these blocks would be
lost.

> p.s.: The HDD is a IBM DRHS36V (36GB U2W-SCSI) if that is important.

Hm, I thought that only IBM DTLA-307040 and IC40... were subject to quick
death.

Bye,
    Oleg

Re[2]: Problem with my HDD

[Cookie-Monster]

I saved 1.7 Gig from the HDD, the proiblem is solved.

I made an image with "dd" and put it on another partition, ran
reiserfsck and I got the data back.

Thx for your help

--
Cookie-Monster
cookie.monster@starfall.de
http://www.starfall.de/

Re: Problem with my HDD

[Dieter Nützel]

On Friday 31 May 2002 04:20,  Billy Transue wrote:

[-]
> I've suffered a very similar problem and posted to the list earlier, I
> did, dd_resuce -l dd.log /dev/hdd5 ./hdd5 and winded up w/ this:
> dd_rescue: (info): ipos:         0.0k, opos:         0.0k,
> xferd:         0.0k
>                    errs:      0, errxfer:         0.0k,
> succxfer:         0.0k
>              +curr.rate:        0kB/s, avg.rate:        0kB/s,
> avg.load:  0.0%
>
> it never changed, the file created hdd5 was 65536 bytes large.  The log
> file was empty.  Am I doing something wrong?  Any info would be
> appreciated.  Thanks in advance.

Bill, have you ever tried the (r)evers mode?
A valid run on a 859.412k partition should look like this:

SunWave1 /tmp# time dd_rescue -r -l dd_rescue-sdb8.log /dev/sdb8 sdb8
dd_rescue: (info): ipos:    859446.0k, opos:    859446.0k, xferd:         0.0k
             -  *  errs:      0, errxfer:         0.0k, succxfer:         0.0k
             +curr.rate:        0kB/s, avg.rate:        0kB/s, avg.load: 96.1%
dd_rescue: (warning): /dev/sdb8 (859446.0k): Input/output error!
dd_rescue: (info): ipos:    859445.5k, opos:    859445.5k, xferd:         0.5k
             -  *  errs:      1, errxfer:         0.5k, succxfer:         0.0k
             +curr.rate:      197kB/s, avg.rate:        0kB/s, avg.load: 96.0%
dd_rescue: (warning): /dev/sdb8 (859445.5k): Input/output error!
dd_rescue: (info): ipos:    859445.0k, opos:    859445.0k, xferd:         1.0k
             -  *  errs:      2, errxfer:         1.0k, succxfer:         0.0k
             +curr.rate:      252kB/s, avg.rate:        0kB/s, avg.load: 95.9%
dd_rescue: (warning): /dev/sdb8 (859445.0k): Input/output error!
dd_rescue: (info): ipos:    859444.5k, opos:    859444.5k, xferd:         1.5k
             -  *  errs:      3, errxfer:         1.5k, succxfer:         0.0k
             +curr.rate:      261kB/s, avg.rate:        1kB/s, avg.load: 95.8%
dd_rescue: (warning): /dev/sdb8 (859444.5k): Input/output error!
dd_rescue: (info): ipos:        54.0k, opos:        54.0k, xferd:    859392.0k
             -     errs:      4, errxfer:         2.0k, succxfer:    859390.0k
             +curr.rate:     2413kB/s, avg.rate:     2219kB/s, avg.load:  4.5%
Summary for /dev/sdb8 -> sdb8:
dd_rescue: (info): ipos:         0.0k, opos:         0.0k, xferd:    859446.0k
             -     errs:      4, errxfer:         2.0k, succxfer:    859444.0k
             +curr.rate:     2201kB/s, avg.rate:     2217kB/s, avg.load:  4.5%
0.530u 16.960s 6:27.67 4.5%     0+0k 0+0io 122pf+0w

SunWave1 /tmp# l dd_rescue-sdb8.log sdb8
-rw-r-----    1 root     root         1713 Mai 31 18:15 dd_rescue-sdb8.log
-rw-r-----    1 root     root     880072704 Mai 31 18:15 sdb8

Are you sure that you had write permission and enough space (!!!) on your 
destination (in your case the current working directory)?

dd_rescue seems to allocate the whole needed disk space at the beginning!

Good luck!

-Dieter

BTW I've had "recovered" several (Winbloze) disks of our customers with 
dd_rescue. But *nix FSes are much better to recover because of the better FS 
redundancy and fsck tools after the dd_rescue run.

-- 
Dieter Nützel
Graduate Student, Computer Science

University of Hamburg
Department of Computer Science
@home: Dieter.Nuetzel@hamburg.de

Re: Problem with my HDD

[Billy Transue]

Thanks, I had no idea it allocated all the space right away, but it
still didn't work, the file got a bit bigger, not much.  It was 
73728 bytes big, I let the process run til my computer suddenly crashed,
here is an excert from syslog, there was ton of lines just like the
first 3 shown, from 12:01 til this last one then the kernel panic.  It
should be noted I've been getting kernel panics before too.  I think my
2nd chip of memory is bad.  I'm taking it out after I type this,
hopefully no kernel panics til then.  Any advice?

May 31 12:19:09 threshold kernel: hdd: read_intr: status=0x59 {
DriveReady SeekComplete DataRequest Error }
May 31 12:19:09 threshold kernel: hdd: read_intr: error=0x40 {
UncorrectableError }, LBAsect=11310269, sector=446
May 31 12:19:09 threshold kernel: end_request: I/O error, dev 16:45
(hdd), sector 446
May 31 12:19:14 threshold kernel: CPU 0: Machine Check Exception:
0000000000000004

Love,
Billy

On Fri, 2002-05-31 at 11:28, Dieter Nützel wrote:
> On Friday 31 May 2002 04:20,  Billy Transue wrote:
> 
> [-]
> > I've suffered a very similar problem and posted to the list earlier, I
> > did, dd_resuce -l dd.log /dev/hdd5 ./hdd5 and winded up w/ this:
> > dd_rescue: (info): ipos:         0.0k, opos:         0.0k,
> > xferd:         0.0k
> >                    errs:      0, errxfer:         0.0k,
> > succxfer:         0.0k
> >              +curr.rate:        0kB/s, avg.rate:        0kB/s,
> > avg.load:  0.0%
> >
> > it never changed, the file created hdd5 was 65536 bytes large.  The log
> > file was empty.  Am I doing something wrong?  Any info would be
> > appreciated.  Thanks in advance.
> 
> Bill, have you ever tried the (r)evers mode?
> A valid run on a 859.412k partition should look like this:
> 
> SunWave1 /tmp# time dd_rescue -r -l dd_rescue-sdb8.log /dev/sdb8 sdb8
> dd_rescue: (info): ipos:    859446.0k, opos:    859446.0k, xferd:         0.0k
>              -  *  errs:      0, errxfer:         0.0k, succxfer:         0.0k
>              +curr.rate:        0kB/s, avg.rate:        0kB/s, avg.load: 96.1%
> dd_rescue: (warning): /dev/sdb8 (859446.0k): Input/output error!
> dd_rescue: (info): ipos:    859445.5k, opos:    859445.5k, xferd:         0.5k
>              -  *  errs:      1, errxfer:         0.5k, succxfer:         0.0k
>              +curr.rate:      197kB/s, avg.rate:        0kB/s, avg.load: 96.0%
> dd_rescue: (warning): /dev/sdb8 (859445.5k): Input/output error!
> dd_rescue: (info): ipos:    859445.0k, opos:    859445.0k, xferd:         1.0k
>              -  *  errs:      2, errxfer:         1.0k, succxfer:         0.0k
>              +curr.rate:      252kB/s, avg.rate:        0kB/s, avg.load: 95.9%
> dd_rescue: (warning): /dev/sdb8 (859445.0k): Input/output error!
> dd_rescue: (info): ipos:    859444.5k, opos:    859444.5k, xferd:         1.5k
>              -  *  errs:      3, errxfer:         1.5k, succxfer:         0.0k
>              +curr.rate:      261kB/s, avg.rate:        1kB/s, avg.load: 95.8%
> dd_rescue: (warning): /dev/sdb8 (859444.5k): Input/output error!
> dd_rescue: (info): ipos:        54.0k, opos:        54.0k, xferd:    859392.0k
>              -     errs:      4, errxfer:         2.0k, succxfer:    859390.0k
>              +curr.rate:     2413kB/s, avg.rate:     2219kB/s, avg.load:  4.5%
> Summary for /dev/sdb8 -> sdb8:
> dd_rescue: (info): ipos:         0.0k, opos:         0.0k, xferd:    859446.0k
>              -     errs:      4, errxfer:         2.0k, succxfer:    859444.0k
>              +curr.rate:     2201kB/s, avg.rate:     2217kB/s, avg.load:  4.5%
> 0.530u 16.960s 6:27.67 4.5%     0+0k 0+0io 122pf+0w
> 
> SunWave1 /tmp# l dd_rescue-sdb8.log sdb8
> -rw-r-----    1 root     root         1713 Mai 31 18:15 dd_rescue-sdb8.log
> -rw-r-----    1 root     root     880072704 Mai 31 18:15 sdb8
> 
> Are you sure that you had write permission and enough space (!!!) on your 
> destination (in your case the current working directory)?
> 
> dd_rescue seems to allocate the whole needed disk space at the beginning!
> 
> Good luck!
> 
> -Dieter
> 
> BTW I've had "recovered" several (Winbloze) disks of our customers with 
> dd_rescue. But *nix FSes are much better to recover because of the better FS 
> redundancy and fsck tools after the dd_rescue run.
> 
> -- 
> Dieter Nützel
> Graduate Student, Computer Science
> 
> University of Hamburg
> Department of Computer Science
> @home: Dieter.Nuetzel@hamburg.de