From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Darrick J. Wong" Subject: Re: [PATCH 06/18] dumpe2fs: add switch to disable checksum verification Date: Mon, 28 Jul 2014 00:48:39 -0700 Message-ID: <20140728074839.GC404@birch.djwong.org> References: <20140726003339.28334.54447.stgit@birch.djwong.org> <20140726003416.28334.11698.stgit@birch.djwong.org> <20140726205808.GP6725@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-ext4@vger.kernel.org To: "Theodore Ts'o" Return-path: Received: from userp1040.oracle.com ([156.151.31.81]:28662 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750945AbaG1Hsp (ORCPT ); Mon, 28 Jul 2014 03:48:45 -0400 Content-Disposition: inline In-Reply-To: <20140726205808.GP6725@thunk.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Sat, Jul 26, 2014 at 04:58:08PM -0400, Theodore Ts'o wrote: > On Fri, Jul 25, 2014 at 05:34:16PM -0700, Darrick J. Wong wrote: > > Add a -n switch to turn off checksum verification. > > Instead of adding a -n flag, I wonder if the better thing to do is if > the various functions that might return a checksum error error out, we > print a warning message indicating checksum failure occured, and then > retry with EXT2_FLAG_IGNORE_CSUM_ERRORS. That is, either retry the > ext2fs_open with the IGNORE_CSUM_ERRORS, or if the file system is > already open, or in EXT2_FLAG_IGNORE_CSUM_ERRORS into fs->flags and > then retry the ext2fs_read_bitmaps() or whatever. > > What do you think? My reason for this approach is that forcing the user to tack on "-n" makes it less likely that a checksum error will be buried in the output and forgotten. That said I don't have any objection to this approach either. The checksum complaint could be printed at the end. I think the error message should also tell the user to run e2fsck. --D > > - Ted