Path: utzoo!attcan!uunet!munnari.oz.au!uhccux!uhunix1.uhcc.Hawaii.Edu
From: pilger@uhunix1.uhcc.Hawaii.Edu (Eric Pilger)
Newsgroups: comp.protocols.nfs
Subject: Enduring PC-NFS Annoyances
Message-ID: <10083@uhccux.uhcc.Hawaii.Edu>
Date: 30 Oct 90 18:14:20 GMT
Sender: news@uhccux.uhcc.Hawaii.Edu
Organization: University of Hawaii
Lines: 25

From: A Dedicated User of PC-NFS
To: The Fine, Upstanding Programmers at Sun
Re: When Will It Be Fixed?

- Why do I, as often as not, have to use <Ctrl-Break> to unstick a
"net ypset *" and continue?

- When will I be able to open more than 6 Windows on my X server
software (eg. fix the socket overload bug AND provide more sockets)?

- Will you (or do you) get together with some of these people writing
X servers and help them optimize and debug their code (and yours?)
This would only further my commitment to PC-NFS.  If PC-NFS doesn't
keep up, I'll be forced to find a solution that does, as much as I
like what PC-NFS has to offer.

- Why is there such a stiff penalty for reads and writes on a socket
(order of 1/5 second of overhead?)  Is this the fault of using a DOS
device driver?  Would send() and rcv() be faster? (Note: rpc's display
none of this overhead.  Do they use send() and rcv()?)

P.S.  Keep up the good work.  I've always been quite happy with almost
all aspects of your product.  I certainly wouldn't be able to survive
today without it, and nobody else has all the features you provide,
YET.  They're catching upppp...