Subject: Re: NFS problems
To: Frank van der Linden <fvdl@wasabisystems.com>
From: Greywolf <greywolf@starwolf.com>
List: tech-kern
Date: 07/22/2002 16:30:34
Incredible.

You mean we're still doing 8k R/W on Sparc, while the intel is doing 32k
R/W?

No *wonder* I started running into problems between the two.

[well, that and a miniswitch that probably doesn't have enough memory to
rub together to make a spark.]

On Mon, 22 Jul 2002, Frank van der Linden wrote:

# The larger default size is present in 1.5.2 and 1.5.3 as well, and
# is only used on i386. Very few problems were reported with 1.5.2,
# except for some old ethernet cards, which was more or less expected.
# The fix for this is documented in the mount_nfs manual page.
#
# More recently, there seems to be a stream of reports from 1.6BETA
# systems, which somehow makes me think that UBC has a different write
# pattern (more bursty perhaps), which makes more network hardware
# get into trouble. In fact, problems have been reported even with
# 8k packet size, which had been the old default writesize forever.

I had problems before UBC entered the picture.

Discovered, by the way, that AIX won't play nice with UDP NFS clients
that are not on the same network as the server.

				--*greywolf;
--
NetBSD: it'll be there when you're ready for it.