G'day

Got a panic using:

NetBSD 1.2 (VME147) #0: Mon Sep 16 16:41:20 CDT 1996
    chuck@vme147.ccrc.wustl.edu:/usr/src/sys/arch/mvme68k/compile/VME147
Motorola MVME147: 25MHz MC68030 CPU+MMU, MC68882 FPU
real mem = 8388608
avail mem = 6705152


... while building a kernel:

-------------------------------------------------------------------
cc  -O2 -msoft-float -Werror -nostdinc -I. -I../../../../arch -I../../../.. -und
ef -D__NetBSD__ -Dm68k -Dmc68000 -D_KERNEL -Dmc68020 -Dmvme68k -DSWAPPAGER -DVNO
DEPAGER -DDEVPAGER -DEXEC_AOUT -DEXEC_SCRIPT -DMVME147 -DFPCOPROC -DDIAGNOSTIC -
DDEBUG -DKTRACE -DCOMPAT_43 -DFFS -DNFS -DFIFO -DINET -DTCP_COMPAT_42 -DCOMPAT_0
9 -DCOMPAT_10 -DCOMPAT_11 -DCOMPAT_12 -DCOMPAT_SUNOS -DMAXUSERS=2  -c ../../../.
./nfs/nfs_node.c
/usr: bad dir ino 2 at offset 28: mangled entry
panic: bad dir
syncing disks... 5 5 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 giving up

dumping to dev 409, offset 49144
dump device bad

[...]

Copyright Motorola Inc. 1989, All Rights Reserved

VME147 Monitor/Debugger Release 2.2 - 8/7/89

FPC  passed test
MMU  passed test

COLD Start

Onboard  RAM start = $00000000,  stop = $007FFFFF

147-Bug>
-------------------------------------------------------------------

The result is a corrupted /usr partition (sd1d). Sigh.
After "fixing" the disk with fsck many of the required
programs for multi-user (like getty) had bitten the
dust. Enough to ruin your whole day! 8-)

The entire source tree was nfs mounted, so the only thing
that the system was using /usr for was the build programs.

Am I just "lucky" or is this a known problem? I'm used to
NetBSD/Alpha and it doesn't do this sort of thing. 8-)

Regards,

-- 
John Birrell - jb@cimlogic.com.au; jb@netbsd.org
CIMlogic Pty Ltd, 119 Cecil Street, South Melbourne Vic 3205, Australia
Tel +61 3 9690 6900 Fax +61 3 9690 6650 Mob +61 418 353 137