|
Home > Archive > Linux Debian support > January 2008 > check disk under linux for bad clusters / sectors
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
check disk under linux for bad clusters / sectors
|
|
| xubunt6 2008-01-03, 7:12 pm |
| reposted in ... debian, because my tb message filters somehow screw up
posting and receiving in ... ubuntu.
Under ms (difficult to type ;-) ) windows chdsk checks for bad clusters.
Rule of thumb is: when you get bad clusters, your disk should be
considered to be replaced.
Xubuntu checks (with e2fsck, I thought) every 30 starts for file
allocation errors. And perhaps for disk errors.
Question: is there a (simple) way to get some kind of report with
something similar as chkdsk that shows bad clusters or bad sectors or so?
looking at man e2fsck, could it be: e2fsck -cc /dev/hda5 ?
--
Xubunt6
"Xubuntu 6 - (bubbling around, sometimes ...)."
Loads inexpensive, nice usable xoftware ... (^-^)
| |
| Kai-Martin Knaak 2008-01-03, 7:12 pm |
| On Thu, 03 Jan 2008 21:59:36 +0100, xubunt6 wrote:
> is there a
> (simple) way to get some kind of report with something similar as chkdsk
> that shows bad clusters or bad sectors or so?
Modern hard disks know about their faults and will tell you about them if
you ask kindly. Some clever marketing team decided to call the
corresponding protocol "smart" (Self-Monitoring, Analysis and Reporting
Technology System).
The debian packages to access smart data are
smartmontools
and
smart-notifier
> looking at man e2fsck,
> could it be: e2fsck -cc /dev/hda5 ?
The on board controller of the disk will do its best to compensate any
faults. So e2fsck will only detect something if the problems are so grave
that they cannot be hidden any more. In other words, at this moment you
already have lost whatever data the affected sectors contained. Smart
allows you to see the tide of problems rising before anything serious
happens.
---<(kaimartin)>---
--
Kai-Martin Knaak
http://lilalaser.de/blog
| |
|
| On Thu, 03 Jan 2008 21:59:36 +0100, xubunt6 wrote:
> reposted in ... debian, because my tb message filters somehow screw up
> posting and receiving in ... ubuntu.
>
>
> Under ms (difficult to type ;-) ) windows chdsk checks for bad clusters.
> Rule of thumb is: when you get bad clusters, your disk should be
> considered to be replaced.
>
> Xubuntu checks (with e2fsck, I thought) every 30 starts for file
> allocation errors. And perhaps for disk errors.
> Question: is there a (simple) way to get some kind of report with
> something similar as chkdsk that shows bad clusters or bad sectors or so?
> looking at man e2fsck, could it be: e2fsck -cc /dev/hda5 ?
'badblocks'
| |
| Trevor Best 2008-01-04, 1:13 pm |
| On Thu, 03 Jan 2008 18:30:21 -0700
ray <ray@zianet.com> wrote:
> On Thu, 03 Jan 2008 21:59:36 +0100, xubunt6 wrote:
>
>
> 'badblocks'
>
should the partition be unmounted to run it?
--
Have you ever imagined a world with no hypothetical situations?
| |
| Peter Terpstra 2008-01-04, 1:13 pm |
| Trevor Best in <20080104150454.3c37f6e3@voyager> :
> should the partition be unmounted to run it?
No, it only searches for the bad blocks.
P.
--
mailto:peter@dharma.dyndns.biz
| |
|
| On Fri, 04 Jan 2008 15:04:54 +0000, Trevor Best wrote:
> On Thu, 03 Jan 2008 18:30:21 -0700
> ray <ray@zianet.com> wrote:
>
>
> should the partition be unmounted to run it?
Not necessary. By default it does a simple read check.
| |
| Trevor Best 2008-01-04, 1:14 pm |
| On Fri, 04 Jan 2008 17:02:01 +0100
Peter Terpstra <peter@dharma.dyndns.biz> wrote:
> Trevor Best in <20080104150454.3c37f6e3@voyager> :
>
>
> No, it only searches for the bad blocks.
Thanks for the clarification.
--
Have you ever imagined a world with no hypothetical situations?
| |
| xubunt6 2008-01-04, 1:14 pm |
| ray wrote:
> On Thu, 03 Jan 2008 21:59:36 +0100, xubunt6 wrote:
>
>
> 'badblocks'
>
Thanks.
I saw "badblocks" mentioned under man e2fsck. It sais, more or less: use
e2fsck -cc <device>
Are you having an opinion about that?
--
Xubunt6
"Xubuntu 6 - (bubbling around, sometimes ...)."
Loads inexpensive, nice usable xoftware ... (^-^)
| |
|
| On Fri, 04 Jan 2008 18:19:30 +0100, xubunt6 wrote:
> ray wrote:
> Thanks.
> I saw "badblocks" mentioned under man e2fsck. It sais, more or less: use
> e2fsck -cc <device>
>
> Are you having an opinion about that?
As I understand, the fsck programs will check file system integrity - that
is something which should be done periodically. badblocks is lower level
and checks to see if the disk is going bad. It is entirely possible to
have file system errors without having a bad disk - for example if the
power goes out while the computer is on and you don't have an UPS. If it
starts showing bad blocks, it is definitely time to replace. The 'smart'
tool disk monitor is also helpful.
| |
| xubunt6 2008-01-04, 1:14 pm |
| Kai-Martin Knaak wrote:
> On Thu, 03 Jan 2008 21:59:36 +0100, xubunt6 wrote:
>
>
> Modern hard disks know about their faults and will tell you about them if
> you ask kindly. Some clever marketing team decided to call the
> corresponding protocol "smart" (Self-Monitoring, Analysis and Reporting
> Technology System).
>
> The debian packages to access smart data are
> smartmontools
> and
> smart-notifier
>
>
> The on board controller of the disk will do its best to compensate any
> faults. So e2fsck will only detect something if the problems are so grave
> that they cannot be hidden any more. In other words, at this moment you
> already have lost whatever data the affected sectors contained. Smart
> allows you to see the tide of problems rising before anything serious
> happens.
>
> ---<(kaimartin)>---
Thanks. I installed (smartmontools) it. Let's see what it does.
--
Xubunt6
"Xubuntu 6 - (bubbling around, sometimes ...)."
Loads inexpensive, nice usable xoftware ... (^-^)
|
|
|
|
|