Re: [SLE] 9.0 Device checling on ext3 fs

• Previous message: Carlos E. R.: "Re: [SLE] Re: Identifizierung (verification check)"
• In reply to: Hylton Conacher (ZR1HPC): "Re: [SLE] 9.0 Device checling on ext3 fs"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Date: Wed, 28 Jul 2004 14:01:35 +0200 (CEST)
To: SLE <suse-linux-e@suse.com>

The Saturday 2004-07-17 at 20:17 +0200, Hylton Conacher (ZR1HPC) wrote:

> > If a hard disk is so bad that you need to check for bad blocks on every
> > boot, dump it immediately.
> DOH!! Incorrect syntax by me.
>
> What I would like to do is make sure that the filesystems on all the
> partitions are as error free as possible ie no errors.

Then, run the badbloc program on all partitions once. Just notice that if
there is an IO error on a reiser partition, reiserfs can not handle that
(it relies on the HD hardware relocating it).

>
> > If what you want is to keep an eye on your disk, have a look at smart
> > instead (man smartctl).
> Letting the system keep them in top top shape IS exactly what I want to do.
>
> As per your advice I had a look for smartctl and didn't find it installed. On
> checking on Yast I was warned that the package is buggy and that I should test
> it first etc and then enable it. With something as critical as disk
> filesystems, I need an app that has been tested and just works, preferably
> with a GUI interface too of course. :) with my luck it works perfectly on your
> system but for some unknown reason bombs out mine. Without decent backups,
> this is a NO-NO

I'm not aware of that warning :-? What I don't find "perfect" is the
automated daemon, but the manual program is very nice.

SMART stands for "Self-Monitoring, Analysis and Reporting Technology", and
it is a feature of "modern" hard disks. The HD itself is responsible for
monitoring some parameters (temperature, rpms, wear, error rates etc),
that can sometimes give warning of impending failure. It has to be enabled
in BIOS.

Also, the HD can test itself; there are a long test and a short test.
Notice that all this is done by the HD on its own, not by the operating
system: the OS only requests the tests being started, and then reads the
results some time later. They can be run while the system is working,
transparently.

If there is a bug on those tests, it is inside the HD firmware or ROM, not
in the monitor program. But I think that if your HD logs smart parameters
(they are saved somewhere on the HD), you should use a program capable of
displaying that data (smartmon is not the only one). Whether you trust
your HD firmware to auto test itself, that's another issue.

Also, you can usually download test software from the maker of your HD.
Sometimes (Seagate) that program simply runs the internal auto test, or an
external equivalent. Usually they come as a file to create an MsDos boot
diskette.

-- 
Cheers,
       Carlos Robinson
-- 
Check the headers for your unsubscription address
For additional commands send e-mail to suse-linux-e-help@suse.com
Also check the archives at http://lists.suse.com
Please read the FAQs: suse-linux-e-faq@suse.com

• Previous message: Carlos E. R.: "Re: [SLE] Re: Identifizierung (verification check)"
• In reply to: Hylton Conacher (ZR1HPC): "Re: [SLE] 9.0 Device checling on ext3 fs"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]