From owner-freebsd-current Tue Mar 18 22:17:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA08858 for current-outgoing; Tue, 18 Mar 1997 22:17:45 -0800 (PST) Received: from sumatra.americantv.com (sumatra.americantv.com [199.184.181.250]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA08853 for ; Tue, 18 Mar 1997 22:17:41 -0800 (PST) Received: from right.PCS (right.pcs. [148.105.10.31]) by sumatra.americantv.com (8.7.6/8.7.3) with ESMTP id AAA23904 for ; Wed, 19 Mar 1997 00:22:57 -0600 (CST) Received: (jlemon@localhost) by right.PCS (8.6.13/8.6.4) id GAA13221; Wed, 19 Mar 1997 06:18:21 GMT Message-ID: <19970319001821.63996@right.PCS> Date: Wed, 19 Mar 1997 00:18:21 -0600 From: Jonathan Lemon To: freebsd-current@freebsd.org Subject: Panics on reboot/halt/shutdown with MFS References: <199703131421.GAA28883@freefall.freebsd.org> <19970318081322.59650@klemm.gtn.com> <19970318212528.28756@klemm.gtn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.61.1 In-Reply-To: <19970318212528.28756@klemm.gtn.com>; from Andreas Klemm on Mar 03, 1997 at 09:25:28PM +0100 Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Mar 03, 1997 at 09:25:28PM +0100, Andreas Klemm wrote: > > I noticed that if I used shutdown I got a message that some process > > refused to die. The only process which might cause it was mount_mfs. > > I just disabled mounting an mfs on /tmp and now I can halt cleanly. > > Are you using mfs ??? If so try disabling it. > > Well, yes, I use mfs ... but ... who or what broke it in 2.2.0 and 3.0 ? > I'm using mfs for months and had no trouble. I spent some time digging into this. The symptoms are that the system panics/locks up when executing reboot() when there is a MFS mounted filesystem. (freebsd bugs #3012, #3026) When MFS is not mounted, the system shuts down without a problem. The culprit appears to be this line in spec_strategy(): (*bdevsw[major(ap->a_bp->b_dev)]->d_strategy)(ap->a_bp); major(ap->a_bp->b_dev) == 0xff. d_strategy for this bdevsv[] == 0. #0 spec_strategy (ap=0xefbffe64) at ../../miscfs/specfs/spec_vnops.c:509 509 { #1 0xf012e2a8 in bwrite (bp=0xf3a23024) at vnode_if.h:1116 1116 return (VCALL((bp)->b_vp, VOFFSET(vop_strategy), &a)); #2 0xf019599e in ffs_sbupdate (mp=0xf1a20c00, waitfor=1) at ../../ufs/ffs/ffs_vfsops.c:1057 1057 error = bwrite(bp); #3 0xf019524c in ffs_unmount (mp=0xf1a0d200, mntflags=524288, p=0xf1a20800) at ../../ufs/ffs/ffs_vfsops.c:672 672 ffs_sbupdate(ump, MNT_WAIT); #4 0xf0134f1c in dounmount (mp=0xf1a0d200, flags=524288, p=0xf1a20800) at ../../kern/vfs_syscalls.c:309 309 error = VFS_UNMOUNT(mp, flags, p); #5 0xf0132f79 in vfs_unmountall () at ../../kern/vfs_subr.c:245 245 error = dounmount(mp, MNT_FORCE, initproc); #6 0xf0114e1f in boot (howto=8) at ../../kern/kern_shutdown.c:217 217 vfs_unmountall(); Perhaps someone with a CVS tree available could look at the recent changes that were made to 2_2_RELENG? This only just started happening. -- Jonathan