Bad Blocks

Bad Blocks

[Abiy,Mike [Edm]]

I would like to apologize, first, if this happens to be a simple question.

1. How does one test for bad blocks on a hard drive in linux.

2. More important, how does one render a bad block on hard drive unreadable,
so that the bad block utility that was used( whatever it may be)  and/or
program does not try to write or read to this same bad block, resulting in
the same errors happening again and again.

3. This was necessitated by trips to a remote site from remote power bootup
( sometime it is absolutely necessary to do that) to do a manual fsck,
because the linux box stops the normal bootup process awaiting manual
intervention to do manual fsck.

any information on this topic would be greatly appreciated.

thanks
Michael

Re: Bad Blocks

[Jorge R . Csapo]

assim falou Abiy,Mike [Edm] (em 19/09/2002):
> I would like to apologize, first, if this happens to be a simple question.
> 
> 1. How does one test for bad blocks on a hard drive in linux.
> 
> 2. More important, how does one render a bad block on hard drive unreadable,
> so that the bad block utility that was used( whatever it may be)  and/or
> program does not try to write or read to this same bad block, resulting in
> the same errors happening again and again.

mkfs -c does just that, addressing both 1. and 2. 

> 
> 3. This was necessitated by trips to a remote site from remote power bootup
> ( sometime it is absolutely necessary to do that) to do a manual fsck,
> because the linux box stops the normal bootup process awaiting manual
> intervention to do manual fsck.

This is totally configurable, meaning the necessity for a manual fsck can
simply be removed. You can either prevent Linux from fsck'ing at boot (not a
really good idea) or force fsck at every boot but with options that don't
require manual intervention. The way to do this depends on your distro, but
it may involve editing /etc/inittab, a number of /etc/rc.d files, re-creating
your filesystems or all of the above...

-- 
Jorge R. Csapo
--------------------------------------------------
 /"\
 \ / CAMPANHA DA FITA ASCII - CONTRA MAIL HTML
  X  ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL
 / \
--------------------------------------------------
http://www.completo.com.br/~jorge
===========================================
With a PC, I always felt limited
by the software available.
On Unix, I am limited only by my knowledge.
--Peter J. Schoenster

Re: Bad Blocks

[Scott Taylor]

At 07:48 AM 19/09/2002, Abiy,Mike [Edm] wrote:
>I would like to apologize, first, if this happens to be a simple question.
>
>1. How does one test for bad blocks on a hard drive in linux.
>
>2. More important, how does one render a bad block on hard drive unreadable,
>so that the bad block utility that was used( whatever it may be)  and/or
>program does not try to write or read to this same bad block, resulting in
>the same errors happening again and again.
>
>3. This was necessitated by trips to a remote site from remote power bootup
>( sometime it is absolutely necessary to do that) to do a manual fsck,
>because the linux box stops the normal bootup process awaiting manual
>intervention to do manual fsck.
>
>any information on this topic would be greatly appreciated.

Hello Michael,

That could depend on your hardware and file system.  Probably a good idea 
to read 'man fsck'

RE: Bad Blocks

[Abiy,Mike [Edm]]

Continuing on the same subject, How does the linux box decide that it
requires a manual fsck, during bootup, is there a specified number of bad
blocks that it has to come across before it decides to halt the bootup
process and ( fsck) and wait for a manual fsck; if so, is there a way to
change that. again, the whole reason is to avoid having to drive over six
and half hours just to do a manual fsck and the loss of the linux box(
system) during that time.

assim falou Abiy,Mike [Edm] (em 19/09/2002):
> I would like to apologize, first, if this happens to be a simple question.
> 
> 1. How does one test for bad blocks on a hard drive in linux.
> 
> 2. More important, how does one render a bad block on hard drive
unreadable,
> so that the bad block utility that was used( whatever it may be)  and/or
> program does not try to write or read to this same bad block, resulting in
> the same errors happening again and again.

mkfs -c does just that, addressing both 1. and 2. 

> 
> 3. This was necessitated by trips to a remote site from remote power
bootup
> ( sometime it is absolutely necessary to do that) to do a manual fsck,
> because the linux box stops the normal bootup process awaiting manual
> intervention to do manual fsck.

This is totally configurable, meaning the necessity for a manual fsck can
simply be removed. You can either prevent Linux from fsck'ing at boot (not a
really good idea) or force fsck at every boot but with options that don't
require manual intervention. The way to do this depends on your distro, but
it may involve editing /etc/inittab, a number of /etc/rc.d files,
re-creating
your filesystems or all of the above...

-- 
Jorge R. Csapo
--------------------------------------------------
 /"\
 \ / CAMPANHA DA FITA ASCII - CONTRA MAIL HTML
  X  ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL
 / \
--------------------------------------------------
http://www.completo.com.br/~jorge
===========================================
With a PC, I always felt limited
by the software available.
On Unix, I am limited only by my knowledge.
--Peter J. Schoenster

RE: Bad Blocks

[Glynn Clements]

Abiy,Mike [Edm] wrote:

> Continuing on the same subject, How does the linux box decide that it
> requires a manual fsck, during bootup

If fsck returns an error code other than 0 (no errors) or 1 (some
errors, but they were all fixed), the boot sequence will normally be
interrupted before the root filesystem is mounted read-write.

-- 
Glynn Clements <glynn.clements@virgin.net>

RE: Bad Blocks

[Aleksander Kujbida]

Also, if a specified period of time has elapsed since the last fsck (or is 
it last boot?), it will fsck. Can't remember where the time period is set.

Aleksander


>From: Glynn Clements <glynn.clements@virgin.net>
>To: "Abiy,Mike [Edm]" <Mike.Abiy@EC.gc.ca>
>CC: "'Jorge R . Csapo'" <jorge@completo.com.br>,linux-admin@vger.kernel.org
>Subject: RE: Bad Blocks
>Date: Mon, 23 Sep 2002 21:33:10 +0100
>
>
>Abiy,Mike [Edm] wrote:
>
> > Continuing on the same subject, How does the linux box decide that it
> > requires a manual fsck, during bootup
>
>If fsck returns an error code other than 0 (no errors) or 1 (some
>errors, but they were all fixed), the boot sequence will normally be
>interrupted before the root filesystem is mounted read-write.
>


_________________________________________________________________
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com

RE: Bad Blocks

[Glynn Clements]

Aleksander Kujbida wrote:

> > > Continuing on the same subject, How does the linux box decide that it
> > > requires a manual fsck, during bootup
> >
> > If fsck returns an error code other than 0 (no errors) or 1 (some
> > errors, but they were all fixed), the boot sequence will normally be
> > interrupted before the root filesystem is mounted read-write.
> 
> Also, if a specified period of time has elapsed since the last fsck (or is 
> it last boot?), it will fsck. Can't remember where the time period is set.

The boot sequence *always* runs fsck; but fsck itself won't actually
perform the check if the filesystem was cleanly unmounted and neither
the maximum mount count nor the check interval have been reached.

-- 
Glynn Clements <glynn.clements@virgin.net>

RE: Bad Blocks

[Aleksander Kujbida]

Thanks for the clarification, Glynn.
Aleksander


>From: Glynn Clements <glynn.clements@virgin.net>
>To: "Aleksander Kujbida" <akujbida@hotmail.com>
>CC: linux-admin@vger.kernel.org
>Subject: RE: Bad Blocks
>Date: Tue, 24 Sep 2002 05:54:28 +0100
>
>
>Aleksander Kujbida wrote:
>
> > > > Continuing on the same subject, How does the linux box decide that 
>it
> > > > requires a manual fsck, during bootup
> > >
> > > If fsck returns an error code other than 0 (no errors) or 1 (some
> > > errors, but they were all fixed), the boot sequence will normally be
> > > interrupted before the root filesystem is mounted read-write.
> >
> > Also, if a specified period of time has elapsed since the last fsck (or 
>is
> > it last boot?), it will fsck. Can't remember where the time period is 
>set.
>
>The boot sequence *always* runs fsck; but fsck itself won't actually
>perform the check if the filesystem was cleanly unmounted and neither
>the maximum mount count nor the check interval have been reached.
>
>--
>Glynn Clements <glynn.clements@virgin.net>




_________________________________________________________________
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com