Re: [Exim] pri/sec MX on same machine to catch spam

> > Some spammers will hit the primary once and never try again. Since the host
> > hasn't been seen before, the message won't come through. Some spammers will
> > only hit the secondary server. If they do, the secondary will check the DB
> > and if it hasn't seen this host, assume it's a spam and blacklist.
>
> ... and when your primary interface hardware breaks?
>
> ... and when your secondary inferface hardware breaks?

The idea I had was assuming 2 IPs close together (like .1 and .2) and the
interface was aliased.

> ... and when there's a screwup in somebody's router so that only the
> secondary IP address is reachable?

OUCH!

> OK, relatively unlikely, especially if there's only one interface with 2
> IP addresses. But I would be scared of this kind of thing.

I have not implemented this idea anywhere so it remains an idea. How would
2 systems share the same db w/o involving something like db replication
(like mysql supports) and not having a 3rd machine to do the DB lookups.

Basically, ATM, this idea is for people who have only 1 mailserver but can
get 2 IPs. I can't do this because I have 2 IPs, but not at the same
location or the same type.

I'm stumped right now as how to do this across the internet. The basic idea
involves 2 MXs that access the same DB (read and write).

Am I correct in assuming that most mailservers (exim, sendmail, qmail,
smail, postfix, ...) will try the backup MX if the primary is down or
returns a temporary failure code? If so and having 2 MXs on seperate
machines, I'm not sure how to ensure that the 2nd has the information before
the mailserver gives up on #1 and tries #2.

--
Lab tests show that use of micro$oft causes cancer in lab animals