>
> > That said, it COULD be that I am a ningcompoop with things not
> > configured right... Any NFS tuning hints/suggestions?
> >
>
> mount the nfs volumes with -o rsize=8192,wsize=8192. The default is 1024,
> but 8192 is much much faster. You should be able to beat samba. Also look
> into getting the updated knfsd, and apply the kernel patches. Esp if you
> need locking to work, eg reading mail across nfs is unsafe without the
> patches.
Well, this might work as long as poor Mike doesn't need to communicate
with non-Linux NFS-clients. However, it is (virtually) impossible to use
Linux v2.2.x NFS with w/rsize > 2048 against other clients. There is a bug
in Sun's nfs-client, which causes corruption, that only Linux triggers.
This bug has been fixed by Sun, but to little avail; lots of other 'ix:es
license Sun's (old, buggy) code.
I would REALLY appreciate if someone could look into this and see if it
would be possible to fix this on the Linux end. I know too little of NFS
myself to look into it, but I know that a lot of people using Linux in
bigger solutions needs this problem solved. Sure, in an ideal world, all
other OS's would bugfix their clients instead, but in the real world, this
won't happen, and that inflicts pain to us.
/David Weinehall
_ _
// David Weinehall <tao@acc.umu.se> /> Northern lights wander \\
// Project MCA Linux hacker // Dance across the winter sky //
\> http://www.acc.umu.se/~tao/ </ Full colour fire </
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/