Xref: utzoo comp.protocols.nfs:482 comp.protocols.misc:677
Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!decwrl!sgi!vjs@rhyolite.wpd.sgi.com
From: vjs@rhyolite.wpd.sgi.com (Vernon Schryver)
Newsgroups: comp.protocols.nfs,comp.protocols.misc
Subject: Re: Has anyone tried NFS over SLIP?
Summary: its possible
Message-ID: <43038@sgi.sgi.com>
Date: 14 Oct 89 19:07:54 GMT
References: <731@anagld.UUCP>
Sender: vjs@rhyolite.wpd.sgi.com
Distribution: na
Organization: Silicon Graphics, Inc., Mountain View, CA
Lines: 17

In article <731@anagld.UUCP>, rcsmith@anagld.UUCP (Ray Smith) writes:
> 
> ...[is NFS possible over SLIP]...

Yes, provided you either (1) get NFS clients and servers that turn
on UDP checksums or (2) use a SLIP varient that does end-to-end checksumming
(or equivalent).

However, it's not pretty.  Think about the sizes of "small" NFS traffic
like attribute requests.  If you value your lunch, do not think about
distant hosts doing 8KB UDP's without knowing there is a SLIP 9.6Kb gateway
in the route.  It stops being funny at <9.6.


Vernon Schryver
Silicon Graphics
vjs@sgi.com