Re: [nfsv4] Mandatory

New Message Reply About this list Date view Thread view Subject view Author view Attachment view

From: Brent Callaghan (brent@eng.sun.com)
Date: 01/22/04-01:02:19 PM Z


Message-ID: <40101E3B.9020702@eng.sun.com>
From: Brent Callaghan <brent@eng.sun.com>
Subject: Re: [nfsv4] Mandatory
Date: Thu, 22 Jan 2004 11:02:19 -0800


COMPOUNDs were supposed to allow clients to assemble NFS calls
tailored more precisely to their needs, i.e. a Java client with
a different API might use a quite different set of COMPOUNDs than
a Linux kernel client.  There was also a (perhaps) naive hope that
in building the protocol from a simple set of primitives, we could
actually make it simpler (heh).  The expectation that COMPOUNDs might
accomplish more on the server per RPC call was secondary.

However, many protocol features later conflicted with the purpose
of COMPOUND such as: OPEN, stateids, non-atomicity, and RESOURCE
errors.  Also, now that the protocol is so semantically heavy, I
doubt it will be interesting to those non-UNIX, non-Windows
clients and servers that may want to use COMPOUND in novel ways.

The protocol, with its current semantics, would likely be
simpler without COMPOUND.  Case in point: the DAFS protocol
retained many of the NFSv4 semantics, but did away with
COMPOUND.

	Brent


_______________________________________________
nfsv4 mailing list
nfsv4@ietf.org
https://www1.ietf.org/mailman/listinfo/nfsv4


New Message Reply About this list Date view Thread view Subject view Author view Attachment view

This archive was generated by hypermail 2.1.2 : 03/04/05-02:13:14 AM Z CST