Hey, all,

I was under the assumption that the default transfer (read/write) sizes
across NFS were 8k bytes.  Did this change at some point?

Let me explain what I've done:
[My workstation: 800MHz Athlon; fxp0@100Mb/s
 everything server: SPARCstation 5; le0@10Mb/s]

Old configuration:

	- office mac -.
	- office PC #1...-10Mb hub--wall---dsl rtr/fw
	- office PC #2'				|
			       my workstation-._|_.- everything server

	All connections 10Mb

New Configuration:

	- office mac -.
	- office PC #1...-[SWITCH] -- wall -- [SWITCH] -- dsl rtr/fw
	- office PC#2'				  |
			          my workstation-._.- everything server

	The [SWITCH]es are Hawking 8-port nway 10/100 Mb switches.
	The DSL router only talks 10Mb/s.
	The everything server only talks 10Mb/s.
	All other connections are 100Mb/s.

As soon as I made this change, any NFS *writes* from my workstation to the
everything server would lock up my workstation (lots of processes stuck in
vnlock/vfswait/nfs).  Hard, mandating a middle-finger salute.

I remembered the old days in which smaller transfer sizes were needed some-
times on a large network.  So, dutifully, I began to play with that (after
much moaning and wailing and gnashing of teeth), and I found that specifying
-w=8192 fixes the problem.

...are these switches just not the greatest, or is something else at fault
here?

I don't have problems *reading*, just *writing*.  Would this be indicative
of a faulty network card?

				--*greywolf;
--
NetBSD: Agnostics in the Platform Religious Wars.