Hi there,
I'm not sure if what I want to do is already available, so this may be a
query or a feature request, depending on the outcome! :)
We are currently using a MySQL database for delivery instructions. The
problem we have here is that if the MySQL database fails for any reason,
a mail server will then lose delivery instructions (and start to bounce
mail). I am aware the 'mysql_servers = ' option is a list, but from
what I can see in the .spec, these are tried in order until a query
succeeds.
The way I interpret this is:
Mail comes in, query is done on server1.
Server1 says "nothing found"
Query retries on server2.
This isn't what I want because the data may not be in server1 on purpose
(e.g., a customer doesn't have that mailbox name), so the way it works
above would suggest that a backup server gets hammered when data isn't
in the first database (e.g., a spammer just sent a mail to
234832assad47028@???). What would be ideal for me is
where if the connection itself fails to the database, to try the second
machine.
The existing option would be great if it somehow knew that the data
isn't meant to be there, but I'm not sure how that is handled internally
by Exim.
I can clarify if needed, has anyone done anything similar?
Cheers,
Mike.
--
| Mike Grice Broadband Solutions for
| Systems Engineer Home & Business @
| PlusNet plc. www.plus.net
+ ----- PlusNet - The smarter way to broadband ------
