On Wed, Mar 30, 2005 at 09:23:56PM -0800, David Kewley wrote:
>As you may know, RHEL comes without xfs enabled in the kernel. My
>understanding is that RH chose this path because a) they have in-house
>expertise in ext3 but not for xfs, and b) they believe that xfs doesn't offer
>any advantages to their customers that ext3 cannot provide.
a) fair enough
b) I think RedHat are wrong
eg. bonnie++'s over GigE and NFS to SMP server with 11-disk 3ware raid5
sees significantly better block reads and writes with XFS:
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
xfs nfs 64 4G 35820 98 80227 32 4540 91 37642 97 78269 15 469.0 1
ext3 nfs 64 4G 35294 97 56381 18 4094 92 37269 97 71096 14 675.6 2
kernel is recompiled AS4 2.6.9-5.0.3.EL.rootsmp. filesystems are
aligned to the raid with
mkfs -t xfs -d sunit=128,swidth=1280 /dev/sda1
mkfs -t ext3 -m 1 -R stride=128 /dev/sda1
and default mount options.
>Bottom line I want to know is, if I simply take the RHEL 4 kernel source rpm,
>enable XFS, and rebuild the kernel (I am fluent enough in these steps), am I
See attached patch to apply after you have installed the .src.rpm. Then
edit the spec file for any other changes you want, and
rpmbuild -ba --target=i686 SPECS/kernel-2.6.spec
>likely to have an xfs that is production-quality? If not, what are the
production quality? who knows...
it hasn't been through RedHat's testing, but seems to work ok for me.
Description: Text document