Quoting Marc Perkel (marc@???):
> It's hardly a solution in that it doesn't do what I want. I want to be
> able to look at the reason for the 550 rejection. So if it's "unknown
> user" it gets treated differently from "relaying denied". If I had that
> I could make a choice to refusing an email or accepting it and storing it.
You're really looking for a solution with an external script that tails
your logfile, scans for 5xx replies, parses out the reason, creates a
(dbm?)file with servers-that-fail-while-they-shouldn't and add some
lookup in your router or acls to see if mail to that domain should be
instantly frozen in the queue.
You could add ETRN support to start a rerun of frozen mail for $domain,
do periodic checks (with an external script) to see if
server-that-failed in (dbm?)file still rejects the legit mails, etc...
As you can see this is way more flexible than just an Exim codechange to
make it interpret (certain) 5xx messages as 4xx and thus breaking the
SMTP protocol.
-Sndr.
-- | Experience is a wonderful thing.
| It enables you to recognize a mistake when you make it again.
| 1024D/08CEC94D - 34B3 3314 B146 E13C 70C8 9BDB D463 7E41 08CE C94D
