Subject: Re: df output with a really large filesystems
To: Chuck Cranor <firstname.lastname@example.org>
From: Luke Mewburn <lukem@NetBSD.org>
Date: 03/26/2004 12:06:38
Content-Type: text/plain; charset=us-ascii
On Thu, Mar 25, 2004 at 09:13:02AM -0500, Chuck Cranor wrote:
| On Thu, Mar 25, 2004 at 09:53:47PM +0900, enami tsugutomo wrote:
| > > the issue was addressed fairly reently - what is the revisions you=
| > > using?
| > Actually, this issue is slightly different than the previous one. If
| > we don't touch any kernel code, the following patch may help some
| > cases (but not all case. e.g., when # of reserved blocks + total
| > blocks doesn't fit in 32 bit value and disk is nearly full).
| > enami.
| yup, enami's patch did the trick:
| % df /filer
| Filesystem 1K-blocks Used Avail Capacity Mounted on
| filer:/ 1203059592 136500 1202923092 0% /filer
| % df -h /filer
| Filesystem Size Used Avail Capacity Mounted on
| filer:/ 1.1T 133M 1.1T 0% /filer
| BTW, our filer only has one shelf of disks (to make the 1.1T fs).
| i guess if i had more shelves i would have a problem?
Not specifically, because last I looked filers still had a 1.4 TB raw
limit per volume. That may have been fixed recently.
It's still worth looking at, because 2 TB is achievable for under U$2K,
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (NetBSD)
-----END PGP SIGNATURE-----