ah I see the miscommunitcation.. Majordomo is submitting locally and via
port 25.. port 10025 is for the off box amavis process to submit mail
back in to..
On Mon, 2006-03-13 at 09:59 +0100, Phil Pennock wrote: > On 2006-03-12 at 20:06 -0600, David Byte wrote:
> > I think we misunderstand each other.. I know that the router IS
> > grabbing majordomo emails and is sending majordomo traffic off to amavis.
>
> Yes, I did understand. The problem is that as far as I can see, your
> condition is good (provided that Majordomo really is submitting on port
> 10025 and not using 25; the Exim mainlog should show which port it was
> actually received on).
>
> Well, okay, the other problem is that I was too tired to explain this
> coherently. Sorry.
>
> So I'm suggesting debugging what you have, to see why it's not working.
> One way is to tag the emails with a header which shows, _each_ time that
> it's used by the Amavis router, what the actual data was; you should end
> up with two such headers, if it's being rescanned.
>
> The other way is a debugging run; with -d<something> and -bs you can
> type SMTP in and see Exim print debugging information about what it
> expands and what the results are. By using -oMi you specify the
> Interface that Exim should pretend the mail was received on, which is an
> IP and optional port number. So by using the real IP address and .10025
> on the end, and using -d+route+expand, you should see what routers were
> tried and why they were tried, including the results of expanding
> strings such as used in a condition rule.
>
> Be aware that any complete emails sent whilst during this will actually
> be in the queue and sent on.
>
> > >The other thing you can try is a debugging run, specifying a fake local
> > >port:
> > > exim -d+route+expand -oMi 1.2.3.4.10025 -bs
> > >(replacing 1.2.3.4 with the IP address that your server listens on);
> > >speak SMTP to that.
>
> Sorry for the confusion,
> -Phil
