This whole thing does make one wonder, however, about how a system
like this behaves normally in a crash. We aren't *supposed* to need to 
do special stuff in order to assure that stuff like metadata is
cleanly out after a sync()....

.pm

Jason Thorpe writes:
> On Wed, 10 Jun 1998 18:12:18 -0700 
>  "Erik E. Fair" <fair@clock.org> wrote:
> 
>  > It worked - I compiled a new kernel, shut the previous one down to single
>  > user mode, unmounted everything (except root, of course), and halted with 
a
>  > power off (after sync'ing myself silly), and fsck found damage on / when
>  > the new kernel came up.
>  > 
>  > I shut down the new kernel to single user, umounted everything, created a
>  > file, deleted the file, sync'd, and halt -p. No damage found on reboot.
>  > 
>  > So, for at least the particular combination of a SPARC Classic (sun4m), an
d
>  > a Quantum 1080S, the SCSI SYNCHRONIZE CHACHE command at shutdown apparentl
y
>  > had the desired result.
> 
> Excellent!
> 
> Jason R. Thorpe                                       thorpej@nas.nasa.gov
> NASA Ames Research Center                            Home: +1 408 866 1912
> NAS: M/S 258-5                                       Work: +1 650 604 0935
> Moffett Field, CA 94035                             Pager: +1 650 428 6939