Date: Thu, 13 Jan 2005 19:25:59 -0800 (PST) From: Don Lewis <truckman@FreeBSD.org> To: current@FreeBSD.org Subject: Re: repeatable panic - ffs_blkfree: freeing free block Message-ID: <200501140326.j0E3PxDf031145@gw.catspoiler.org> In-Reply-To: <200501140106.j0E16atP030900@gw.catspoiler.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 13 Jan, Don Lewis wrote:
> I'm getting a repeatable panic when trying to build openoffice with a
> 2005-01-12 kernel and world. I manually fsck'ed the file system
> immediately before the build. The first panic happened during the
> patching phase of the build. I tried again and this time the panic
> happened during configuration.
>
> dev = da0s2a, block = 5090392, fs = /mnt
> panic: ffs_blkfree: freeing free block
> cpuid = 0
> KDB: enter: panic
> [thread pid 55 tid 100043 ]
> Stopped at kdb_enter+0x2c: leave
> db> tr
> Tracing pid 55 tid 100043 td 0xc22fa450
> kdb_enter(c0848a40,100,c22fa450,1361b80,4b5) at kdb_enter+0x2c
> panic(c085f39d,c085f37d,c26cc280,4dac58,0) at panic+0x17f
> ffs_blkfree(c2863800,c2987114,4dac58,0,4000) at ffs_blkfree+0x3a0
> indir_trunc(136b060,0,0,c,0,e4d5bc4c) at indir_trunc+0x1bf
> handle_workitem_freeblocks(0,0,0,1,0) at handle_workitem_freeblocks+0x2e6
> process_worklist_item(41e7112d,0,0,c247d2bc,23) at process_worklist_item+0x1c0
> softdep_process_worklist(0,29,41e7112d,1,0) at softdep_process_worklist+0x66
> sched_sync(0,e4d5bd48,0,c0677b60,0) at sched_sync+0x3b2
> fork_exit(c0677b60,0,e4d5bd48) at fork_exit+0x7e
> fork_trampoline() at fork_trampoline+0x8
> --- trap 0x1, eip = 0, esp = 0xe4d5bd7c, ebp = 0 ---
>
> The kernel was built with INVARIANTS, WITNESS, and DEBUG_VFS_LOCKS.
>
> I was able to get a crash dump from the first panic, so at least doadump
> works for me.
>
> After the crash, fsck reports some unexpected damage:
> PARTIALLY TRUNCATED INODE I=127552
> but this inode doesn't seem to be associated with the block number in
> the panic message.
> % fsdb /dev/da0s2a
> ** /dev/da0s2a (NO WRITE)
> Editing file system `/dev/da0s2a'
> Last Mounted on /mnt
> current inode: directory
> I=2 MODE=40755 SIZE=512
> MTIME=Feb 14 11:55:43 2004 [0 nsec]
> CTIME=Feb 14 11:55:43 2004 [0 nsec]
> ATIME=Jan 13 16:15:58 2005 [0 nsec]
> OWNER=root GRP=wheel LINKCNT=7 FLAGS=0 BLKCNT=4 GEN=53557245
> fsdb (inum: 2)> inode 127552
> current inode: regular file
> I=127552 MODE=100644 SIZE=1932
> MTIME=Dec 22 07:47:38 2003 [0 nsec]
> CTIME=Dec 22 07:47:38 2003 [0 nsec]
> ATIME=Dec 22 07:57:30 2003 [0 nsec]
> OWNER=root GRP=wheel LINKCNT=1 FLAGS=0 BLKCNT=4 GEN=1442a6af
> fsdb (inum: 127552)> blocks
> Blocks for inode 127552:
> Direct blocks:
> 533086 (1 frag)
Some other interesting bits of data:
The file in question is
/mnt/usr/ports/editors/openoffice-1.1/work/config_office/configure
The inode number of this file is same as the inum parameter to
ffs_blkfree().
The panic occurs while this script is being executed.
Some more of the fsck output:
** /dev/da0s2a (NO WRITE)
** Last Mounted on /mnt
** Phase 1 - Check Blocks and Sizes
PARTIALLY TRUNCATED INODE I=1275522
SALVAGE? no
INCORRECT BLOCK COUNT I=1275522 (416 should be 704)
CORRECT? no
INCORRECT BLOCK COUNT I=1282133 (704 should be 384)
CORRECT? no
It looks like inode 1282133 has been deleted because I can't
find it in the file system tree. If I use fsdb to find the
first few block numbers and dump them out, I see the following:
# Generated by GNU Autoconf 2.59.
#
# Copyright (C) 2003 Free Software Foundation, Inc.
# This configure script is free software; the Free Software Foundation
# gives unlimited permission to copy, distribute and modify it.
## --------------------- ##
## M4sh Initialization. ##
## --------------------- ##
# Be Bourne compatible
if test -n "${ZSH_VERSION+set}" && (emulate sh) >/dev/null 2>&1; then
emulate sh
NULLCMD=:
# Zsh 3.x and 4.x performs word splitting on ${1+"$@"}, which
# is contrary to our usage. Disable this feature.
alias -g '${1+"$@"}'='"$@"'
elif test -n "${BASH_VERSION+set}" && (set -o posix) >/dev/null 2>&1; then
set -o posix
fi
DUALCASE=1; export DUALCASE # for MKS sh
# Support unset when possible.
if ( (MAIL=60; unset MAIL) || exit) >/dev/null 2>&1; then
as_unset=unset
else
as_unset=false
fi
# Work around bugs in pre-3.0 UWIN ksh.
$as_unset ENV MAIL MAILPATH
PS1='$ '
PS2='> '
PS4='+ '
# NLS nuisances.
for as_var in \
LANG LANGUAGE LC_ADDRESS LC_ALL LC_COLLATE LC_CTYPE LC_IDENTIFICATION \
LC_MEASUREMENT LC_MESSAGES LC_MONETARY LC_NAME LC_NUMERIC LC_PAPER \
LC_TELEPHONE LC_TIME
do
if (set +x; test -z "`(eval $as_var=C; export $as_var) 2>&1`"); then
eval $as_var=C; export $as_var
else
$as_unset $as_var
fi
done
[snip]
whereas .../work/config_office/configure starts out:
#! /bin/sh
# From configure.in Revision: 1.55.6.10 .
# Guess values for system-dependent variables and create Makefiles.
# Generated by GNU Autoconf 2.59.
#
# Copyright (C) 2003 Free Software Foundation, Inc.
# This configure script is free software; the Free Software Foundation
# gives unlimited permission to copy, distribute and modify it.
## --------------------- ##
## M4sh Initialization. ##
## --------------------- ##
# Be Bourne compatible
if test -n "${ZSH_VERSION+set}" && (emulate sh) >/dev/null 2>&1; then
emulate sh
NULLCMD=:
# Zsh 3.x and 4.x performs word splitting on ${1+"$@"}, which
# is contrary to our usage. Disable this feature.
alias -g '${1+"$@"}'='"$@"'
elif test -n "${BASH_VERSION+set}" && (set -o posix) >/dev/null 2>&1; then
set -o posix
fi
DUALCASE=1; export DUALCASE # for MKS sh
# Support unset when possible.
if ( (MAIL=60; unset MAIL) || exit) >/dev/null 2>&1; then
as_unset=unset
else
as_unset=false
fi
# Work around bugs in pre-3.0 UWIN ksh.
$as_unset ENV MAIL MAILPATH
PS1='$ '
PS2='> '
PS4='+ '
# NLS nuisances.
for as_var in \
LANG LANGUAGE LC_ADDRESS LC_ALL LC_COLLATE LC_CTYPE LC_IDENTIFICATION \
LC_MEASUREMENT LC_MESSAGES LC_MONETARY LC_NAME LC_NUMERIC LC_PAPER \
LC_TELEPHONE LC_TIME
do
if (set +x; test -z "`(eval $as_var=C; export $as_var) 2>&1`"); then
eval $as_var=C; export $as_var
else
$as_unset $as_var
fi
done
It looks like the configure script has just been regenerated and a bug
is being triggered by executing the script shortly after the script file
was regenerated and before softupdates has flushed the changes to disk.
I cleaned things up and ran "make extract". I don't see anything that
has the contents seen in inode 1282133, so I suspect it is a temporary
file. The configure script starts off:
#! /bin/sh
# From configure.in Revision: 1.55.6.9.4.2 .
[snip]
so the configure script is definitely being regenerated (and the
configure.in script is $Revision: 1.55.6.10).
The file system in question is UFS2 with the default 16K/2K
block/fragment sizes.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200501140326.j0E3PxDf031145>