I usually don't put all my eggs in one basket with DF. I use it as a
generic tool, but for accuracy, I use du. Perhaps you can perform du -sk
and compare that to the DF output. You should see good similarities.
Systems Architect, CCNA
On Mon, 14 May 2001, Federico Sevilla III wrote:
> Hi everyone,
> While migrating a ReiserFS partition to XFS, I noticed that df reported a
> pretty significant space utilization difference between the two
> filesystems. I transferred my files using two methods, and both had
> identical results.
> First I used tar from /xfs: "tar -cl /reiserfs | tar xv", then I used cp:
> "cp -a /reiserfs /xfs". Here is the output when I ran "df -h" after each
> file transfer operation:
> Filesystem Size Used Avail Use% Mounted on
> /dev/hda5 2.0G 692M 1.4G 33% /reiserfs
> /dev/hda3 952M 746M 206M 78% /xfs
> If this is accurate (although it doesn't seem accurate as I will later
> elaborate), then XFS uses 52MB more than ReiserFS does for a relatively
> small partition (this is actually my root partition, and I only exclude a
> small boot and the home partition).
> I decided to check using du by running "du -csm" and found that ReiserFS
> uses 734MB and XFS uses 733MB. This looks more accurate. Would anyone know
> what's wrong with df? It's a much handier tool to use than du, but if it's
> not accurate then it doesn't look like a good tool to use, handy or not.
> Thanks in advance for your comments and suggestions.
> --> Jijo
> Linux, MS-DOS, and Windows NT ...
> ... also known as the Good, the Bad, and the Ugly