Return-Path: linux-nfs-owner@vger.kernel.org Received: from mx1.redhat.com ([209.132.183.28]:25745 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757587AbaISSPp (ORCPT ); Fri, 19 Sep 2014 14:15:45 -0400 Date: Fri, 19 Sep 2014 14:15:36 -0400 From: Simo Sorce To: "J. Bruce Fields" Cc: Steve Dickson , Linux NFS Mailing list Subject: Re: [PATCH 0/2] Use the gssproxy damon for GSSAPI credentials Message-ID: <20140919141536.36b0a83a@willson.usersys.redhat.com> In-Reply-To: <20140919181122.GC16658@fieldses.org> References: <1411131103-17254-1-git-send-email-steved@redhat.com> <20140919155733.GA16658@fieldses.org> <541C5A55.7010403@RedHat.com> <20140919163617.GB16658@fieldses.org> <20140919130752.2ac3372e@willson.usersys.redhat.com> <20140919181122.GC16658@fieldses.org> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Sender: linux-nfs-owner@vger.kernel.org List-ID: On Fri, 19 Sep 2014 14:11:22 -0400 "J. Bruce Fields" wrote: > On Fri, Sep 19, 2014 at 01:07:52PM -0400, Simo Sorce wrote: > > On Fri, 19 Sep 2014 12:36:17 -0400 > > "J. Bruce Fields" wrote: > > > > > On Fri, Sep 19, 2014 at 12:31:17PM -0400, Steve Dickson wrote: > > > > > > > > > > > > On 09/19/2014 11:57 AM, J. Bruce Fields wrote: > > > > > On Fri, Sep 19, 2014 at 08:51:41AM -0400, Steve Dickson wrote: > > > > >> The gssproxy(8) daemon is now how the managing of GSSAPI > > > > >> credentials should be done. > > > > >> > > > > >> These to patches has the nfs-server systemd script start > > > > >> gssproxy instead of rpc.svcgssd and removes the building > > > > >> and installation of rpc.svcgssd. > > > > >> > > > > >> For now I'm leaving the source files... Just in case... ;-) > > > > > > > > > > If nothing else, we'd like to keep support for older kernels > > > > > for a while. > > > > Good point... I should probably make this a configurable... > > > > something like --enable_svcgssd and have it off by default... > > > > > > Sounds fine. > > > > +1, it will be very useful for transitions in distributions, not > > all of them will align identically I guess. > > If we wanted to be *very* nice, we could try to detect and run one or > the other at run-time. It's nice for people trying to boot old > kernels to track down (probably unrelated) regressions. But I think > it would be too complicated. > > It might be simple enough for gss-proxy to log some kind of warning > when it starts up on an old kernel. Excellent idea, I opened a ticket to track that: https://fedorahosted.org/gss-proxy/ticket/126 Simo. -- Simo Sorce * Red Hat, Inc * New York