From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtpq2.tb.mail.iss.as9143.net ([212.54.42.165]:38867 "EHLO smtpq2.tb.mail.iss.as9143.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932416AbdGSWZc (ORCPT ); Wed, 19 Jul 2017 18:25:32 -0400 Received: from [212.54.42.135] (helo=smtp11.tb.mail.iss.as9143.net) by smtpq2.tb.mail.iss.as9143.net with esmtp (Exim 4.86_2) (envelope-from ) id 1dXxPO-0003QL-3w for linux-xfs@vger.kernel.org; Thu, 20 Jul 2017 00:25:30 +0200 Received: from dhcp-077-251-156-033.chello.nl ([77.251.156.33] helo=[192.168.178.3]) by smtp11.tb.mail.iss.as9143.net with esmtp (Exim 4.86_2) (envelope-from ) id 1dXxPO-0003nH-0Z for linux-xfs@vger.kernel.org; Thu, 20 Jul 2017 00:25:30 +0200 Message-ID: <1500510327.1939.36.camel@versatel.nl> Subject: is there a useful purpose to keep "xfs meta-data" when mkfs.xfs has created a new file system From: Richard Waterbeek Date: Thu, 20 Jul 2017 00:25:27 +0000 Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: linux-xfs@vger.kernel.org I wonder, since I have decided to not follow proper advice, that for example ext2 is meant for a usb removable volume, while journalling fs aren't [benchmarking wasn't real slow, but I don't seem to appreciate i-nodes], nor that the operating system says it has finished copying and unmount the device seems impossible/ give a long waiting time, and the hardware blinker on this usb stick I have isn't "lying". That was enough annoyance to drop the whole ext thing for me. Now that I've came to the conclusion that jfs somehow didn't want to be fsck'd at all, I again went with xfs for usb [only this time I attempt to learn about what I can do and do not. Because instead of backing up, restoring is just as important as the backup itself. Now I've learnt that xfs has to be treated well, I saw the following output with the command "mkfs.xfs /dev/sd" --- meta-data=/dev/sdc isize=256 agcount=4, agsize=244928 blks = sectsz=512 attr=2, projid32bit=1 = crc=0 finobt=0 data = bsize=4096 blocks=979712, imaxpct=25 = sunit=0 swidth=0 blks naming =version 2 bsize=4096 ascii-ci=0 ftype=0 log =internal log bsize=4096 blocks=2560, version=2 = sectsz=512 sunit=0 blks, lazy-count=1 realtime =none extsz=4096 blocks=0, rtextents=0 --- I am not only curious but also interested, what xfs is trying to tell me, when I just asked it to create a new xfs. Since I couldn't get jfs to do a simple 'fsck -n', and fat and ntfs are of no importance to me, and that btrfs simply isn't meant that way and on top of that, that ext2 is a bit dated, ext3 isn't clear to me, and ext4 doesn't seem to add much [I find], the only conclusion that I have now, is to stick with xfs, and most of the internet advices otherwise. I wonder if someone on this list can tell me more on the meta-data of xfs, not my bias to xfs while I wasn't able to give jfs a proper chance. It is kind of important for a usb backup to know what could go wrong before anything goes wrong with the data trusted to it. That is why I am asking here, in the hope that a knowledgeable/ experienced person could shed some light on this. [for my main purpose, having a "pre-prepared" disaster recovery plan, instead of just backups].