There is one other way round this problem... that scates rather close
to the edge of legality... but gets round this sort of problem.
If you route all the remote addresses you will normally find a
*large* number will have common MX hosts - normally the mail hubs for
their ISP. Unfortunately these common MX hosts will normally be well
down the list - typically third or fourth in the list. You could
weight these hosts by the number of recipients that could go via them
and then shove out the majority of your deliveries in a small number
of SMTP connections! This is nasty, a hack, and the precise opposite
of qmail's approach (had to add that since Dan came into the
conversation).
However its a different approach. If you did this (and I'm not
saying I would!), then you would need to have some algorithm for
determining where the greater good outweighs the individual recipient
downside of more hops.
djb@??? said:
} (Some people use bulk_mailer, which achieves parallelism at the
} expense of disk space by sending separate copies of the message.)
I have to add that this is a very similar approach to qmail, which
acheives parallelism at the expense of bandwidth. However this is a
long running argument - and depends how much you pay for bandwidth!
djb@??? said:
} (Btw, I find the which-messages-is-it-handling discussion a bit
} strange: surely it's possible to extract this information from
} syslog?)
After the event it is - however exim does not currently log "I am
locking and beginning work on message X" - whether it should is a
good point. Finding this data in the log (without a "I have finished
message X") is of course a pain!
Nigel.
--
[ Nigel.Metheringham@??? - Unix Applications Engineer ]
[ *Views expressed here are personal and not supported by PLAnet* ]
[ PLAnet Online : The White House Tel : +44 113 251 6012 ]
[ Melbourne Street, Leeds LS2 7PS UK. Fax : +44 113 2345656 ]
