From: Chuck Lever Subject: Re: [PATCH 4/4] NFS: Show most mount options via nfs_show_options() Date: Wed, 30 Jan 2008 16:58:46 -0500 Message-ID: References: <20080130001333.14637.74159.stgit@manray.1015granger.net> Mime-Version: 1.0 (Apple Message framework v753) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Cc: trond.myklebust@fys.uio.no, linux-nfs@vger.kernel.org To: Miklos Szeredi Return-path: Received: from rgminet01.oracle.com ([148.87.113.118]:20480 "EHLO rgminet01.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751836AbYA3V7S (ORCPT ); Wed, 30 Jan 2008 16:59:18 -0500 In-Reply-To: Sender: linux-nfs-owner@vger.kernel.org List-ID: Hi Miklos- On Jan 30, 2008, at 7:50 AM, Miklos Szeredi wrote: >> Display all mount options in /proc/mount which may be needed to >> reconstruct >> a previous mount. > > Thanks, looks good to me. The only thing I see still missing is > "mounthost". What about that one? User space (the mount.nfs command) converts "mounthost=" to "mountaddr=". /proc/mounts shows "mountaddr=" so redriving the mount request will just pass that back in. Currently, none of the DNS resolutions are redone on an NFS remount. The original "addr=" "clientaddr=" and "mountaddr=" options are reused, even with /etc/mtab. (I'm not editorializing about whether that's right or wrong, but that is my understanding of how it works today). -- Chuck Lever chuck[dot]lever[at]oracle[dot]com