From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: linux-nfs-owner@vger.kernel.org Received: from fieldses.org ([174.143.236.118]:46810 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753780Ab3CSPQf (ORCPT ); Tue, 19 Mar 2013 11:16:35 -0400 Date: Tue, 19 Mar 2013 11:16:34 -0400 From: "J. Bruce Fields" To: "Myklebust, Trond" Cc: "linux-nfs@vger.kernel.org" Subject: Re: [PATCH 10/11] NFSv4.1: Enable open-by-filehandle Message-ID: <20130319151634.GF7912@fieldses.org> References: <1363698463-3681-6-git-send-email-Trond.Myklebust@netapp.com> <1363698463-3681-7-git-send-email-Trond.Myklebust@netapp.com> <1363698463-3681-8-git-send-email-Trond.Myklebust@netapp.com> <1363698463-3681-9-git-send-email-Trond.Myklebust@netapp.com> <1363698463-3681-10-git-send-email-Trond.Myklebust@netapp.com> <20130319140901.GB7912@fieldses.org> <1363702570.7515.6.camel@leira.trondhjem.org> <20130319143236.GD7912@fieldses.org> <1363704100.7515.12.camel@leira.trondhjem.org> <1363704300.7515.16.camel@leira.trondhjem.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1363704300.7515.16.camel@leira.trondhjem.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Tue, Mar 19, 2013 at 02:45:00PM +0000, Myklebust, Trond wrote: > On Tue, 2013-03-19 at 10:41 -0400, Trond Myklebust wrote: > > On Tue, 2013-03-19 at 10:32 -0400, J. Bruce Fields wrote: > > > (Sorry for the digression, I was just trying to decide whether I can get > > > away with turning on 4.1 by default before implementing SP4_MACH_CRED or > > > GSS on the backchannel....) > > > > If you don't, could you please at least offer up a module parameter that > > allows me to do that? The current thing is a PITA, since it isn't > > supported by any distro scripts, and it requires you to switch off the > > server in order to change the version pseudofile contents. > > > BTW: you may recall that SPKM3 and UTF-8 support was mandatory in > RFC3530, Yes. > as was RPCSEC_GSS support on the back channel. I don't agree there--maybe the spec says it's mandatory, I don't remember, but in practice 4.0 degrades gracefully without callbacks. In the 4.1 case the create_session's just going to fail with an unspecified error. > That didn't stop anybody from shipping clients and servers that didn't > support them... Yeah, I may just resign myself to erroring out in some more cases. --b.