On Thu, Nov 30, 2006 at 09:32:09PM +0100, Tobias Nygren wrote:
> 
> You might want to try a fsck_lfs -f on the filesystem.
> I had a similiar problem on todays current + chs@'s vnlock patch.
> Turns out the filesystem had a corrupt inode even though it was marked 
> clean.

Yeah, I've run a few fsck's over it - but it's clean:

ksh$ sudo fsck -yf /local
** /dev/rld0g
** File system is already clean
** Last Mounted on /local
** Phase 0 - Check Inode Free List
** Phase 1 - Check Blocks and Sizes
** Phase 2 - Check Pathnames
** Phase 3 - Check Connectivity
** Phase 4 - Check Reference Counts
** Phase 5 - Check Segment Block Accounting
** Phase 6 - Roll Forward
109586 files, 1134725 used, 37202721 free

Just updated kernel to sources about 3 hrs old, same crash, same
backtrace.

-- 
stix