FCNTL on Solaris

[Jeremy Allison]

On Mon, Apr 22, 2002 at 12:58:48PM -0400, David Collier-Brown wrote:
> 
> 	At the expense of not addressing the Sun side of the
> 	problem, might I suggest that validation operations
> 	shouldn't lock?  

No - we must take the lock. It's the only way we can assure
that the underlying structure is in a safe state to examine.
It's not validation operations, it's iteration over the entire
database.

Jeremy.