We are in the process of implementing a set of mail servers based on the
idea of having multiple identically configured systems and increasing the
number of systems we have as load increases...
We would normally expect the majority of outgoing mail to be sent onward
pretty much immediately, maybe depending on load hitting the queue and
having one successful attempt to send it to the final destination (or next
hop).
However there will be some messages that cannot be sent in one delivery
attempt. It would be neat if I could transfer all of those to a single
system which would handle intractable mail - the idea being that this box
would build a good set of retry data, would be in a good position to
efficiently forward all mail queued on a stuck domain when it becomes
unstuck, and be an good place to monitor for problems!
Any ideas on whether this can be easily done, and would it really be a
good idea... Should this be done should the initial boxes send mail on to
the intractable mail box if retry would kick in...
Nigel.
--
[ Nigel.Metheringham@??? - Systems Software Engineer ]
[ Tel : +44 113 251 6012 Fax : +44 113 224 0003 ]
[ Friends don't let friends use sendmail! ]
