From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexandre Oliva Subject: Re: [PATCH] enable mds rejoin with active inodes' old parent xattrs Date: Sat, 24 Aug 2013 07:38:17 -0300 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from linux-libre.fsfla.org ([208.118.235.54]:45823 "EHLO linux-libre.fsfla.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754092Ab3HXKic (ORCPT ); Sat, 24 Aug 2013 06:38:32 -0400 In-Reply-To: (Gregory Farnum's message of "Fri, 23 Aug 2013 13:11:14 -0700") Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Gregory Farnum Cc: "Yan, Zheng" , ceph-devel On Aug 23, 2013, Gregory Farnum wrote: > On Fri, Aug 23, 2013 at 4:00 AM, Alexandre Oliva wrote: >> On Aug 22, 2013, "Yan, Zheng" wrote: >> >>> This is not bug. Only the tail entry of the path encoded in the parent xattrs >>> need to be updated. (the entry for inode's parent directory) >> >> Why store the others, if they're not usable, then? IMHO it just >> introduces a risk of their being accidentally misused. > We want to write a correct path whenever we touch the file, but we > don't want to have to go out to each file's object in order to > complete a directory rename. So for a given file object we consider > only the immediate parent to be authoritative, but keep around the > full path for disaster recovery. Ok, so all this extra info is for disaster recovery only. I guess that makes some sense to me, as long as that's clear to everone who might want to tap on that info. > If we are touching the file's ancestor xattr and not updating the > whole thing, that would be a bug, but I don't think that's what you're > describing? It was not, indeed. Thanks for the clarification. -- Alexandre Oliva, freedom fighter http://FSFLA.org/~lxoliva/ You must be the change you wish to see in the world. -- Gandhi Be Free! -- http://FSFLA.org/ FSF Latin America board member Free Software Evangelist Red Hat Brazil Compiler Engineer