On Wed, 24 Jan 2001, Konrad Schroder wrote:
> On Wed, 24 Jan 2001, Brian Chase wrote:

> > Maybe this is just a fundamental misunderstanding of how LFS works on my
> > part, but I'm seeing some significant discrepancies between the disk usage
> > reported by `df' and the actual disk usage reported by `du' on an LFS
>
> Oreo runs NetBSD 1.5?

Yup. Straight from the NetBSD/i386 1.5 install CD-ROM.

> I can explain the change over time, which will be (mostly) fixed for
> 1.5.1; the gross discrepancy between df's Used value and du's value is
> more difficult to explain unless certain things went wrong with the
> filesystem up to this point.

I don't think so.  I did pull the power plug on the box during an initial
attempt at a copy because I wanted to see how well LFS recovered.  Only
about 10 megabytes had been copied when I did this.

I was a little surprised that none of the copied data was in place once
the system came back up.  I'm guessing the nature of LFS is more for
keeping a filesystem consistent and not so much geared towards being able
to replay the most recent changes to the filesystem after an ungraceful
shutdown.  Though not ideal, that behavior is fine with me.  XFS under
IRIX really spoiled me quite badly.

According to Jesse Off's reply, this all looks normal on mostly empty and
large filesystems.  I'm fine with that.  I've been around traditional
filesystems for too long I guess. :-)

And thanks for the LFS documentation pointers.  I'll go educate myself a
bit more.

-brian.
--- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----