checking filesystem

How can I schedule a disk check for bad sectors and fragmentation? Thanks
for your help.

Cross

  • Cross_AM wrote, On 08/25/2008 10:45 PM:
    > How can I schedule a disk check for bad sectors and fragmentation? Thanks
    > for your help.

Opensuse has a scheduled check upon boot (I think every 60 days or so) which will find bad blocks.
As for fragmentation, it depends on the file system you use. AFAIK there is no defragmentation tool for ext3.

Uwe

Strictly speaking fsck checks for filesystem inconsistencies and not bad blocks, which could be inside a file and not visible to fsck. For that it would be better to enable the SMART daemon which will monitor the health of the disk.

Defrag? Linux is not Windows. :stuck_out_tongue:

ken yap wrote:
> better to enable the SMART daemon

sorry, just spent close to an hour trying to figure out how to do
that in 10.3 (or if it is already)…tried several different search
strings in the suse.forum, in the suse.wiki, in the
suse.documentation and at google…

could ya gimme a hint on a working search string, or just point to a
how-to…

THANKS in advance…i admit my search patience today is short!

ps: noticed without the ability to search with case, it is really
difficult to sift “SMART” (as in the daemon needed) out from “Smart”
(as a replacement for YaST), or “smart” (as in brains), or or

and, is it possible to get the wiki search engine to return hits on a
phrase (like “SMART daemon” or SMART+daemon or SMART*daemon) instead
of return hits on the individual words??


see caveat: http://tinyurl.com/6aagco
DenverD (Linux Counter 282315) via NNTP, Thunderbird 2.0.0.14, KDE
3.5.7, SUSE Linux 10.3, 2.6.22.18-0.2-default #1 SMP i686 athlon

On 08/26/2008 ken yap wrote:
> Strictly speaking fsck checks for filesystem inconsistencies and not
> bad blocks, which could be inside a file and not visible to fsck.

So, under which conditions would the scheduled fsck find a bad block? Because it did when the HD in my Acer failed (or was that the Toshiba? Don’t recall, both had bad HDs…)

Uwe

Hi
smartctl for the CLI command, smartd is the daemon and the config file
is in /etc/smartd.conf. You can enable the daemon (once you have added
the drive(s) to monitor in the config) via YaST runlevels or chkconfig
smartd on.


Cheers Malcolm °¿° (Linux Counter #276890)
openSUSE 11.0 x86 Kernel 2.6.25.11-0.1-default
up 20:30, 1 user, load average: 0.02, 0.03, 0.05
GPU GeForce 6600 TE/6200 TE - Driver Version: 173.14.12

If the bad block affects a directory or an inode, which would cause fsck to report an inconsistency. If the bad block develops inside the data blocks of a file, you wouldn’t know about until you tried to read it.

  • ken yap,

thank you!

Uwe

> smartctl for the CLI command, smartd is the daemon and the config file
> is in /etc/smartd.conf. You can enable the daemon (once you have added
> the drive(s) to monitor in the config) via YaST runlevels or chkconfig
> smartd on.

mmmmmmmmmmm…THANKS…unfortunately i am short on patience today and
the man appeared too daunting…

i’ll save that for a rainy day project…


see caveat: http://tinyurl.com/6aagco
DenverD (Linux Counter 282315) via NNTP, Thunderbird 2.0.0.14, KDE
3.5.7, SUSE Linux 10.3, 2.6.22.18-0.2-default #1 SMP i686 athlon

Hi
This is the quick way to check :slight_smile:


sudo smartctl --all /dev/<your_drive>

Where <your_drive> equals sda hdb etc.


Cheers Malcolm °¿° (Linux Counter #276890)
openSUSE 11.0 x86 Kernel 2.6.25.11-0.1-default
up 22:05, 2 users, load average: 0.08, 0.09, 0.09
GPU GeForce 6600 TE/6200 TE - Driver Version: 173.14.12

> sudo smartctl --all /dev/sda

wow, KEWL…thanks…

but, i learned that i’ve already had one error :frowning: (bought this
drive in about May 08, and had 700+ hours use)


see caveat: http://tinyurl.com/6aagco
DenverD (Linux Counter 282315) via NNTP, Thunderbird 2.0.0.14, KDE
3.5.7, SUSE Linux 10.3, 2.6.22.18-0.2-default #1 SMP i686 athlon

Somebody can beat me with a stick if I’m wrong, but I’ve always been under the impression that scanning an active partition was a baaaad thing to do as it can destroy it. So what do you do? Plunk
your install DVD in the drive and reboot into rescue mode (you could also use a live cd distro like Knoppix) and running fsck from there, or if your in a LVM vgscan

ken yap wrote:

> Linux is not Windows
I know that. That’s why I changed my base system from Windows to Linux.

On Tue, 2008-08-26 at 19:46 +0000, jplindy wrote:
> Somebody can beat me with a stick if I’m wrong, but I’ve always been
> under the impression that scanning an active partition was a baaaad
> thing to do as it can destroy it. So what do you do? Plunk
> your install DVD in the drive and reboot into rescue mode (you could
> also use a live cd distro like Knoppix) and running fsck from there, or
> if your in a LVM vgscan
>
>

But using smart just reads the data from the tables in the drive
itself.

Well ain’t that cool! Guess this old dog is going to have to sit down and learn some new tricks :).

> Guess this old dog is going to have to sit down
> and learn some new tricks :).

EXACTLY!! seems progress is happening all around us!!!


see caveat: http://tinyurl.com/6aagco
DenverD (Linux Counter 282315) via NNTP, Thunderbird 2.0.0.14, KDE
3.5.7, SUSE Linux 10.3, 2.6.22.18-0.2-default #1 SMP i686 athlon

> I know that. That’s why I changed my base system from Windows to Linux.

you missed his point, maybe…

he was not trying to Poke You In the Eye…or otherwise harm pr make
fun of you…

instead (ihmo), we all understand that you expected to need to defrag
Linux hard drives because you learned that this is a VERY good
practice when using Windows[tm], becuase it needs to be defragged…

his point was, since “Linux is not Windows” we don’t need to defrag
(because our file system is different, and it doesn’t fragment)…

see?

if you have not yet read Linux is Not Windows, though it is longer
than some might wanna read, if you do read it all i think you will
find it informative, enlightening and useful: http://tinyurl.com/8b9s6


see caveat: http://tinyurl.com/6aagco
DenverD (Linux Counter 282315) via NNTP, Thunderbird 2.0.0.14, KDE
3.5.7, SUSE Linux 10.3, 2.6.22.18-0.2-default #1 SMP i686 athlon

On 08/27/2008 DenverD wrote:
> (because our file system is different, and it doesn’t fragment)…

Linux filesystems try to keep fragmentation at a low level with different methods, nevertheless you’ll probably see some fragmentation, just do an fsck.

Uwe

DenverD wrote:

> he was not trying to Poke You In the Eye…or otherwise harm pr make
> fun of you…
>
Sorry if my response was derogatory in any sense. I do understand he was
making a sincere attempt to help me. Sincere apologies Ken.

Nah, don’t worry about it. I didn’t think your response was in the slightest peeved or annoyed or negative in any way. Enjoy Linux; that’s one less thing you have to worry about, coming from Windows.