From the second line of log here:
2006-01-07 04:00:35 H=outbound-res.frontbridge.com (outbound1-res-
R.bigfish.com) [63.161.60.49] Warning: F=<sender> delayed 100s RCPT
<recipient>: SPAM FLAGS=0.4.0.0.0.0.0.0.0.0 SCORE=8 RBLS=0 EXTRA: 0
2006-01-07 04:00:35 SMTP connection from outbound-res.frontbridge.com
(outbound1-res-R.bigfish.com) [63.161.60.49] lost while reading
message data (header)
it is not evident what connection/transaction this refers to. One has
to refer to the line immediately above (from the an RCPT ACL warn +
delay) to work out the sender and (last) recipient, but I suppose it
could happen that another connection writes a log line between the two.
It looks like I cannot control what is written, as this is hard coded
in handle_lost_connection(). I should really replace
smtp_get_connection_info() there with something more specific, but
handle_lost_connection can clearly happen at any phase, when the
sender or (a) recipient are not yet established.
I am raising this point because my daily reports do not identify
failed deliveries caused by delay/disconnections. One would wish that
large companies knew how to properly configure smtp servers...
Thanks
Giuliano
--
## List details at http://www.exim.org/mailman/listinfo/exim-users
## Exim details at http://www.exim.org/
## Please use the Wiki with this list - http://www.exim.org/eximwiki/
