🚀 go-pugleaf

I disagree, in today's environments you probably need to try all the addresses.
This shouldn't be too much of a problem.  MMDF, for instance, caches
information such as whether the machine is up or not in its daemon so it
doesn't keep trying the same thing over and over again.  It also sorts
the list of work by host name so that it can try to do it all in one
connection, and with the case of multiple recipients of the same message,
deliver it sending the DATA only once.

In addition, the concept of a turkey queue is available.  For systems that
deal with a lot of mailing list traffic, a list of hosts frequently down can
be maintained to avoid tying up the normal SMTP queue with those attempts.

-Ron