On Tue, 23 Feb 1999, mark david mcCreary wrote:
> a_thru_l_domains:
> driver = domainlist,
> transport = smtp,
> self = fail_soft,
> host_find_failed = fail_soft,
> pass_on_timeout = true,
> route_list = "^[a-l].* delivery.com bydns";
The .* at the end of your regex achieves nothing other than to waste a
little bit of processor time.
> It looks like the envelope sender remains what it would be, if not routed
> thru delivery.com, which is good, but wasn't what I was expecting. That
> is, delivery .com is noted in the Received Headers, but not the Return_Path.
Exim does not modify the return path when relaying mail. Source routing
is deprecated these days. The draft of the revised RFC 821 (colloquially
called 821bis, actual name draft-ietf-drums-smtpupd-09.txt) says:
The <reverse-path> contains the source mailbox
(between "<" and ">" brackets, which can be used to report errors
...
Historically, the <reverse-path> can contain more than just a mailbox,
however, contemporary systems SHOULD NOT use source routing (see
appendix C).
> If I did that, what might be a reasonable number for max_rcpt on the
> transport. The bigger the better, but I don't want to crash Exim with too
> many names at one time. I would be sending from one Exim machine to
> another.
Exim has no "reasonable" limit. However a message with thousands of
recipients is going to use up more memory during delivery, and take
longer. If you used a limit of 100-500, say, you would get more
parallelism.
> Another option would seem to be the BSMTP (Batched SMTP) option provided by
> Exim, but that is the extent of my knowledge about BSMTP.
BSMTP is for delivery into files and pipes, not over the network.
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
--
*** Exim information can be found at
http://www.exim.org/ ***
