John Henders writes ("Exim scanning all local interfaces before sending mail"):
> I'm bringing up exim on all our local machines and just ran across this
> on our web server. We bind all the virtual ip addresses to the local
> interface, and it appears exim scans them all before sending mail,
> leading to the following in the debug logs.
> Local interface address is 127.0.0.1
> Local interface address is 204.174.16.5
...
> and so on for 160 more addresses.
>
> Is there a reason it needs to do this? Isn't 127.0.0.1 sufficient?
RFC1123 (Host Requirements, Application and Support):
5.2.17 Domain Literals: RFC-822 Section 6.2.3
A mailer MUST be able to accept and parse an Internet domain
literal whose content ("dtext"; see RFC-822) is a dotted-
decimal host address. This satisfies the requirement of
Section 2.1 for the case of mail.
An SMTP MUST accept and recognize a domain literal for any of
its own IP addresses.
So, either you'd have to list these in the config file (silly), or
Exim has to get a list of the local interfaces so that it can
recognise postmaster@??? &c.
Of course people shouldn't generally use this notation, but it can be
very useful to be able to get through to a host which is up and
listening when the DNS is screwed (particularly if the DNS contact is
in the affected zone).
Ian.
