Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 Apollo 11/21/85; site apollo.uucp
Path: utzoo!linus!decvax!genrad!mit-eddie!think!harvard!seismo!rochester!bullwinkle!cornell!uw-beaver!apollo!rees
From: rees@apollo.uucp (Jim Rees)
Newsgroups: net.unix-wizards
Subject: Re: Extended file system on UNIX 4.2/4.3 BSD
Message-ID: <2adcce15.1de6@apollo.uucp>
Date: Sat, 21-Dec-85 15:17:05 EST
Article-I.D.: apollo.2adcce15.1de6
Posted: Sat Dec 21 15:17:05 1985
Date-Received: Tue, 14-Jan-86 07:53:56 EST
References: <910@brl-tgr.ARPA>
Organization: Apollo Computer, Chelmsford, Mass.
Lines: 19

ATT/RFS implements unix file system semantics exactly at the expense of not
being stateless and not caching data in the client.  NFS has a stateless
server at the expense of unix file system semantics.  In case it isn't
obvious, the big advantage of a stateless server is that it simplifies
recovery after machine or network failure.

    Besides, bet a dollar the people who know from
    both groups thought it was a bit misguided often (written by Apollo
    folks tho the intent was sincere.)

You owe me a buck.  The only complaint the AT&T folks had was over a typo
(the RFS mount command came out looking like the NFS mount command).

Actually, I thought we were remarkably restrained about plugging the
Apollo file system.  It has the best caching scheme, but falls down on
unix semantics and heterogeneity.  It doesn't require you to mount the
other disks on the network, they are all automatically available, always.
The AT&T folks don't consider that an advantage, but then they haven't
tried to put together a 1000 node network yet.