From owner-freebsd-hackers Wed Jul 29 03:25:30 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id DAA26076 for freebsd-hackers-outgoing; Wed, 29 Jul 1998 03:25:30 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from elvis.vnet.net (elvis.vnet.net [166.82.1.5]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id DAA26071 for ; Wed, 29 Jul 1998 03:25:28 -0700 (PDT) (envelope-from rivers@dignus.com) Received: from dignus.com (ponds.vnet.net [166.82.177.48]) by elvis.vnet.net (8.8.8/8.8.4) with ESMTP id GAA05922; Wed, 29 Jul 1998 06:24:49 -0400 (EDT) Received: from lakes.dignus.com (lakes [10.0.0.3]) by dignus.com (8.8.8/8.8.5) with ESMTP id GAA03466; Wed, 29 Jul 1998 06:57:20 -0400 (EDT) Received: (from rivers@localhost) by lakes.dignus.com (8.8.8/8.6.9) id GAA00451; Wed, 29 Jul 1998 06:28:52 -0400 (EDT) Date: Wed, 29 Jul 1998 06:28:52 -0400 (EDT) From: Thomas David Rivers Message-Id: <199807291028.GAA00451@lakes.dignus.com> To: dg@root.com, freebsd-hackers@FreeBSD.ORG, rob@perihelion.co.uk Subject: 'daily panic' reproduction [was re: fsck errors] In-Reply-To: <199807290908.KAA42650@betty.perihelion.co.uk> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG See below... "Rob McIntyre" writes: > > Hi, > > I'm currently trying to debug a version of FreeBSD that has been > ported onto our Helios operating system. > After running newfs on an ide device, fsck produces errors indicating > that some number of inodes are partially allocated and others are of > an unknown file type (shown below). Could you tell me more about > what these messages actually mean and maybe give me some > advice on tracking down such file system errors during the fast file > system creation in newfs. > Any help would be grately appreciated. > > Rob. > > R.A.McIntyre MSc, > > Perihelion Distributed Software > Tel: 44 (0) 1749 344345 > Fax: +44 (0) 1749 344977 > http://www.perihelion.co.uk > > > PARTIALLY ALLOCATED INODE INODE # 0x68 > +++ > +++ INODE DISK BLOCK ADDRESS = 0x342cd0 > +++ > +++ > +++ > +++ CLEAR? yes > +++ UNKNOWN FILE TYPE INODE # 0x6d > +++ > +++ > +++ > +++ CLEAR? yes > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-questions" in the body of the message > Rob - I've redirected this to -hackers. But; if you'll look in the mail archives for subjects containing "daily panic", and "freeing free block", you'll see a long running discussing of this problem. Just f.y.i. - a newfs'd partition should fsck completely clean assuming you haven't mounted it and changed it in any way. What you describe is exactly my stand-alone reproduction of the problem. I can write 0xff to a disk partition, newfs that partition and then run fsck on that newfs'd partition to find out some of the bytes didn't actually make it. I believe that in a running system; this particular bug causes the spurious problems some people have seen. I don't know exactly what Helios is... but I hope it can help find this problem... - Dave Rivers - To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message