On Mon, 6 Jan 1997, Christoph Lameter wrote:
> If an NFS volume becomes active again then delivery of all messages must
> be triggered somehow. If the userforward director was failing and is now
> working again then all pending messages because of that userforward
> failure need to be delivered.
A single userforward director may be referencing any number of different
NFS volumes. This is probably quite common. Suppose you have several
different NFS volumes containing user home directories and you look for
.forwards in home directories in the usual way - then the fact that you
can see one user's .forward does not mean you can see another's.
What is needed is to notice when a delivery to a particular local-part
succeeds, and to retry any messages that are waiting for that user
(similar to what it does for hosts). I have an item on the Wish List to
do something about this. So far I haven't got around to thinking about
it, other things seeming to be more important. One day...
--
Philip Hazel University Computing Service,
ph10@??? New Museums Site, Cambridge CB2 3QG,
P.Hazel@??? England. Phone: +44 1223 334714
