From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ipmail07.adl2.internode.on.net ([150.101.137.131]:8622 "EHLO ipmail07.adl2.internode.on.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752158AbeEOF16 (ORCPT ); Tue, 15 May 2018 01:27:58 -0400 Date: Tue, 15 May 2018 15:27:55 +1000 From: Dave Chinner Subject: Re: [PATCH] xfs_db: add blockget -L option Message-ID: <20180515052755.GI23861@dastard> References: <20180510213636.GA5045@deer-run.com> <3ae25107-7242-a645-c0f7-784f35ed3e79@sandeen.net> <1C5C502C-D536-4C2C-B7CA-9B13F9B2EB20@deer-run.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <1C5C502C-D536-4C2C-B7CA-9B13F9B2EB20@deer-run.com> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Hal Pomeranz Cc: Eric Sandeen , linux-xfs@vger.kernel.org On Tue, May 15, 2018 at 12:27:38AM -0400, Hal Pomeranz wrote: > I get the argument Dave is making. But the reality of forensics is > that many of our tools are not reliable in all cases. We do a lot > of cross-validation for crucial results. > > Having multiple tools helps. Right now, we basically have nothing > for XFS. So adding this option to xfs_db makes things better. > > We work with underplayed file systems constantly, and it’s > less of a worry than you might think. Try my patched version of > xfs_db on a live file system of your choice. See if you can make > it blow up. When it doesn’t, please consider folding in my > patch. And the source code to your new and improved xfs_db can be found where? I'm interested to know how you've solved the problem of detecting stale metadata from userspace on a live filesystem... Cheers, Dave. -- Dave Chinner david@fromorbit.com