From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Tue, 25 Mar 2008 18:54:16 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with SMTP id m2Q1s6s4025217 for ; Tue, 25 Mar 2008 18:54:08 -0700 Message-ID: <47E9AC67.8070809@sgi.com> Date: Wed, 26 Mar 2008 12:52:39 +1100 From: Mark Goodwin Reply-To: markgw@sgi.com MIME-Version: 1.0 Subject: Re: REVIEW: Write primary superblock info to ALL secondaries during mkfs References: <47E8F5BD.7000601@sandeen.net> In-Reply-To: <47E8F5BD.7000601@sandeen.net> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: Eric Sandeen Cc: Barry Naujok , "xfs@oss.sgi.com" Eric Sandeen wrote: > Barry Naujok wrote: >> Secondaries should contain redundant information from the primary >> superblock. It does this for the filesystem geometry information, >> but not inode values (rootino, rt inos, quota inos). >> >> This patch updates all the secondaries from the primary just before >> it marks the filesystem as good to go. >> >> Unfortunately, this also affects the output of xfs_repair during >> QA 030 and 178 which restores the primary superblock from the >> secondaries. >> >> Now that the secondaries have valid inode values, xfs_repair >> does not have to restore them to the correct values after copying >> the secondary into the primary. >> >> Attached is the mkfs.xfs patch and also the updated golden >> outputs for QA 030 and 178. >> >> The next step after this is to enhance xfs_repair to be more >> thorough in checking the secondaries during Phase 1. > > One related thing I'd always wondered about was stamping a secondary at > the very end of the device (and therefore shrinking the fs by just a > bit) - repair could then do a quick check at the end of the device > before resorting to scanning for the 2nd backup... would this make any > sense? I guess it might, Barry what do you think? Probably makes grow a bit more complicated. What would repair do if it doesn't find the backup SB at the end of the device? We'd need a new SB flag to indicate it's supposed to be there, which seems a bit chicken-and-egg'ish ... Cheers > > -Eric > > -- Mark Goodwin markgw@sgi.com Engineering Manager for XFS and PCP Phone: +61-3-99631937 SGI Australian Software Group Cell: +61-4-18969583 -------------------------------------------------------------