On Thu, Jun 17, 2010 at 06:01:33PM +0200, Carlos Velasco wrote:
> The "dummy" instance has myhostname changed and relayhost set to:
> relayhost = [1.1.2.1]:25
The ":25" is not needed and best avoided.
> Problem is that "dummy" believes that destination is itself and "loops
> back to myself" is logged. It is wrong because destination is really
> "work" instance (port 25) not itself (port 12025).
Loop detection is on by default when the destination port is 25.
Loop detection matches on either banner hostnames or interfaces
or IP addresses found in inet_interfaces or proxy_addresses.
Remove 192.0.2.1 (better example address than 1.1.2.1) from
inet_interfaces in the "dummy" instance. e.g.:
inet_interfaces = 127.0.0.1
and in master.cf use
192.0.2.1:12025 smtp inet ... smtpd
to explicitly bind the IP owned (on port 25) by the other instance.
Then the IP address in question will not trigger loop detection.
Alternatively, you can override "inet_interfaces" for just the
smtp(8) delivery agent:
smtp unix ... smtp
-o inet_interfaces=127.0.0.1
Ideally, avoid port 25 forwarding between instances on the same host
where multiple instances use the same IP. So IMHO the non-default
forwarding port is the cleaner solution.
--
Viktor.
