[comp.protocols.appletalk] NFS access v CAP Software

jf@ap.co.umist.ac.uk (John Forrest) (02/08/91)

This is a similar vain to the current requests regarding
Gatorbox NFS v alternatives. Can someone fill me in on a
comparison on using NFS from the Mac (be it via Gatorboxes or
direct from the Macs) and using CAP software on Unix boxes. I
assume that if you start using NFS the file type/creator
information of Mac files dissapears - or have I missed
something? How do external (Unix created) files appear in both
these environments.

John Forrest
Dept of Computation
UMIST

PS. Thanks to all those people who sent me info about MacX
loadings.

wiechman@athos.rutgers.edu (NightMeower) (02/10/91)

When using GatorShare, files do not loose their attributes.  All of
the bits and stuff are saved for you.  I know this for a fact because
I use it regularly.  For all intensive purposes, the GatorShare volume
appears to be like any other disk or floppy.  Setup is also fairly
simple, no need to have to be a guru to use it.  I am also fairly
certain that Aufs does not loose this information either.

I am also impressed with the tech support that we have received from
the Cayman people.

Kevin
-- 
===========================================================================
Kevin S. Wiechmann			arpa:  wiechman@cs.rutgers.edu

             This is only a test... for the next sixty seconds...

geo@moldev.com (George Pontis) (02/11/91)

There seems to be a detail in the installation of Appleshare services that's
getting away from a lot of people, and leads to the problems mentioned here
regarding files losing their attributes. The secret is to have a a pair of
world-writeable directories named .finderinfo and .resource in the directory
ABOVE the directory that gets mounted by the Mac.
-- 
George Pontis ( geo@moldev.com )

jqj@duff.uoregon.edu (JQ Johnson) (02/12/91)

I observed recently what appeared to be a rather severe problem with CAP
performance.  The configuration was a CAP server running on SunOS4.1.1
with enet (packet filter) support.  This Sun was an NFS client of a file
server that held the actual files i.e. the CAP server was acting as an
AppleShare to NFS translator.  Performance in this configuration was
dismal, with download file transfer rates in the 100 bytes/sec range even
to an Ethertalk Mac.  Perhaps worse, it occasionally tickled some bug in
rpc.lockd that caused lockd to generate a stream of UDP packets which were
answered only by an ICMP "port unreachable", even after the aufs process
was killed.

It turned out that CAP was doing an OSTestLock() for each packet to make
sure that nobody else had locked the file.  Each OSTestLock() translated
into a lockf(fd,F_TEST,len) system call.  lockf() in turn was implemented
as a context switch and call to rpc.lockd, which in turn had to query the
remote system (first, since we run with resolver routines in libc.so, a
remote DNS lookup to turn the hostname in the mount point into an address,
then an RPC call to the rpc.lockd on the NFS server).  Rebuilding CAP
without X_LOCKF in m4.setup increased file transfer performance by about
100x.

Given that lockf() is very expensive for NFS-mounted files (in SunOS 4.1),
how do other Aufs-to-NFS gateways (Pacer, Mt Xinu, Gatorbox, etc.) deal
with the Appleshare file byte range locking and consistency requirements?

Given that flock() [which is all I have left in CAP after I turn off
lockf()] is only local to a particular machine, how can CAP guarantee to
lock out writers on read-only files if the file is NFS-mounted and the writer is on a different system?

Given that Sun's rpc.lockd is so fragile that it can get wedged by heavy
use, why would anyone in his right mind trust it to guarantee file system
consistency?

-- 
JQ Johnson
Director of Network Services		Internet: jqj@oregon.uoregon.edu
University of Oregon			voice:	(503) 346-4394
250E Computing Center			BITNET: jqj@oregon
Eugene, OR  97403-1212			fax: (503) 346-4397

kdb@macaw.intercon.com (Kurt Baumann) (02/13/91)

In article <Feb.9.13.00.33.1991.9626@athos.rutgers.edu>, wiechman@athos.rutgers.edu (NightMeower) writes:
> Path: intercon!uupsi!rpi!zaphod.mps.ohio-state.edu!swrinde!cs.utexas.edu!sun-barr!newstop!sundc!seismo!dimacs.rutgers.edu!aramis.rutgers.edu!athos.rutgers.edu!wiechman
> From: wiechman@athos.rutgers.edu (NightMeower)
> Newsgroups: comp.protocols.appletalk
> Subject: Re: NFS access v CAP Software
> Message-ID: <Feb.9.13.00.33.1991.9626@athos.rutgers.edu>
> Date: 9 Feb 91 18:00:35 GMT
> References: <1991Feb7.222518.18726@cns.umist.ac.uk>
> Organization: Rutgers Univ., New Brunswick, N.J.
> Lines: 16
> 
> When using GatorShare, files do not loose their attributes.  All of
> the bits and stuff are saved for you.  I know this for a fact because
> I use it regularly.  For all intensive purposes, the GatorShare volume
> appears to be like any other disk or floppy.  Setup is also fairly
> simple, no need to have to be a guru to use it.  I am also fairly
> certain that Aufs does not loose this information either.
> 

You don't loose attributes when you use any of the other NFS products either, doing so would be a big loose, if you know what I mean. :-)

> I am also impressed with the tech support that we have received from
> the Cayman people.
> 

The Cayman people are excellent with their tech support.  We have always gotten good responses from them.


Kurt Baumann                       InterCon Systems Corporation
703.709.9890                      Creators of fine TCP/IP products
703.709.9896 FAX               for the Macintosh.