On Jan 16, 2004, at 5:16 PM, Thor Lancelot Simon wrote:
> Vnodes are recycled.  You're almost certainly looking at some kind of
> locking or synchronization bug.

That is what the manpage says, so is this a known bug?  Is there a PR 
associated with it?
What sort of information would I need to collect in order to help the 
developers nail this?

I've already isolated to which lockinit() (there are two) (line 538) in 
the vfs_subr.c but I have
no way to finding the vnode address from the debugger.

What would be a good price to put  on this bug's head?  $1000? $1500?  
With access to a machine
with crash dumps.  Its not an easy job, I know.  There are very few 
people who could actually
identify the problem let alone code a fix.