NFS performance issue

Using SuSE 9.2 with kernel 2.6.8-25.5, over a gigabit connection
(tried both thru a Netgear switch and a crossover cable with no
switch) to a second system configured identically, NFS appears to be
incredibly slow -I was seeing MUCH faster transfers on the same
hardware with SuSE 9.1. Did SuSE muck with NFS? If so, how do I undo
it?

Mike-

--
If you can keep your head while those around you are losing theirs...
You may have a great career as a network administrator ahead!
--
Please note - Due to the intense volume of spam, we have installed
site-wide spam filters at catherders.com. If email from you bounces,
try non-HTML, non-encoded, non-attachments,

Michael W Cocke <(E-Mail Removed)> wrote:
> Using SuSE 9.2 with kernel 2.6.8-25.5, over a gigabit connection
> (tried both thru a Netgear switch and a crossover cable with no
> switch) to a second system configured identically, NFS appears to be
> incredibly slow -I was seeing MUCH faster transfers on the same
> hardware with SuSE 9.1. Did SuSE muck with NFS? If so, how do I
> undo it?

Well, first,gigiabit does not need cross-over for pt-pt - it (the PHY
I guess?) figures it out for itself.

Are there packet losses or NFS retries in the statistics?

Are your mounts TCP or UDP? IIRC, there was a change in defaults for
TSO (large send) along the way there for capable GbE NICs. Which
NIC(s) are you using?

Do you have comparison data not involving filesystems and discs?
Netperf TCP_RR latency figures or TCP_STREAM bulk transfer perhaps?

rick jones
--
The glass is neither half-empty nor half-full. The glass has a leak.
The real question is "Can it be patched?"
these opinions are mine, all mine; HP might not want them anyway...
feel free to post, OR email to raj in cup.hp.com but NOT BOTH...