Subject: Re: filesystem health
From: Christopher E. Brown (cbrown@woods.net)
Date: Wed Jul 03 2002 - 09:59:30 AKDT
On Tue, 2 Jul 2002 bryan@ak.net wrote:
>
> On Tue, Jun 18, 2002 at 10:56:30PM -0800, civileme <civileme@mandrakesoft.com> wrote:
> >
> > However, to do something, point your browser here:
> >
> > http://www.civileme.com/Software/defrag-0.70.tar.gz
> >
> > It will work for ext2, though I have used it only once, for test purposes. I
> > have never managed to reach Arthur's level of fragmentation--got to 5.4% once.
>
> I tried compiling defrag 0.70, but got this error:
>
> I found the header file that defines EINTR, but I'm not sure whether
> including it would fix the problem properly. Do you know of any bug
> patches for defrag?
There is a reason that the only ext2 defrag tool is a one off item
produced my a single guy a very long time ago. ext2 does *not* need
to be defragmented...
There are 2 reasons that mke2fs defaults to 10% reserved for root.
One is to make sure users cannot fill the device totally (root
*should* have enough space left over to log in and try to fix things),
though this is a usually item, and not a absolute.
The main reason is fragmentation control. ext2 is designed such that
if you maintain at least 10% free it is almost impossible to get any
meaningful fragmentation. There are a very few corner cases, and easy
fixes (but *nothing* that requires defragging, and they can be done on
a live system).
Keep in mind that defrag 0.70 was written well before 1.2.13 was
released, and while ext2s has not really changed, the way the kernel
autocleans and organizes the directory trees has.
--
I route, therefore you are.
---------
To unsubscribe, send email to <aklug-request@aklug.org>
with 'unsubscribe' in the message body.
This archive was generated by hypermail 2a23 : Wed Jul 03 2002 - 10:09:59 AKDT