ReiserFS is the one that has the stability, the others are still
"experimental". In our tests ReiserFS also is the performance champ; I do
remember seeing IBM papers on this subject too. I also recall there was
some issue with NFS, you might try looking at www.namesys.com and
searching for NFS posts. It might also be the version of ReiserFS you use.
In Yast it places ver 1 on the drive, I mean partition, while mkreiserfs
-v 2 /dev/sda1 will give you the 3.6 format that also supports LFS.
Regards,
Jon
Jon R. Doyle
Sendmail Inc.
6425 Christie Ave
Emeryville, Ca. 94608
(o_
(o_ (o_ //\
(/)_ (\)_ V_/_
On Tue, 8 Jan 2002, Frank Warzecha wrote:
> Hi,
>
> I was testing different filesystems, and ran into problems with most of
> them:
>
> on 2.4.7:
>
> - ReiserFS worked fine and fast, but an NFS-client (NFS3 or NFS2) gave me
> stale NFS handles.
> - JFS worked correctly, also with NFS, but is terribly slow.
> - Ext2 was fast and worked with NFS, but it is not jounaling!
>
> I applied SGI's XFS-patch from http://oss.sgi.com to the 2.4.16 Kernel
> tree, and tested the following:
>
> - four dasds were formatted with blocksize of 256
> - XFS works fine and very fast on a single dasd
> - I ran into problems with XFS on a logical volume on all four devices
> (kernel-hang)
>
> Did anyone have any experiences with XFS (or Ext3) on S/390-Linux? Any
> patches that work?
>
> Mit freundlichen Gr��en,
>
> Frank Warzecha
> -------------------------------------
> Frank Warzecha,
> Team Betriebssysteme
> RZNet AG, Kerpen, Germany
> >>>>>>>>>>>>>>>Neue Telefon Nummer: >>>>>>>>Tel:02273-603172
> http://www.rznet.de
>
