OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
Postfix Archives: Re: 2 SMTP Relays, depending on the From: add

Re: 2 SMTP Relays, depending on the From: address

Subject: Re: 2 SMTP Relays, depending on the From: address
From: Martin Schulze (joeyfinlandia.Infodrom.North.DE)
Date: Tue Jan 04 2000 - 04:16:09 CST

Craig Sanders wrote:
> On Tue, Jan 04, 2000 at 01:14:08AM +0100, Martin Schulze wrote:
> > How do you want to distinguish if:
> >
> > . You read mail locally on the mail host
> > . You use the same MUA for both commercial and non-commercial mail
> > . The MUA does not use an IP address for submitting mail but uses
> > something like /usr/sbin/sendmail -t
>
> ok, i see where the confusion is coming from now. when i say "relay
> control", i am talking about preventing 3rd party relay so you don't get
> hijacked by spammers. you're talking about controlling which external
> smtp relay to use.

Oh.

> to do what you want, you don't even really need two instances of
> postfix. just set up a transport table to route mail via different
> transports (e.g. smtp to relayA, smtp to relayB, uucp to relayC, or
> direct smtp to the destination...etc) depending on the destination
> address.

I already said that depending on the destination is not possible since
mail is sent to the guys both privately and commercially.

> this is not exactly what you seem to be asking for (transport selection
> based on source address) but is or can be a workable solution. AFAIK,
> what you want isn't supported by postfix right now.

That's what Wietse told me as well..., though I'm thinking about pushing
postfix to be the nr. 1 MTA and Smail being a shadow MTA for outgoing
commercial traffic.

> fact is, if you don't like my answers or can't use them then you are
> perfectly free to ignore them...but i object to your attitude which

Sure. But... For about two years I'm trying to get away from Smail...
These tiny little things keep me from doing it.

> seems to carry the implication that i'm obliged to find a perfect

It was a result of our misunderstanding apparently. I apologize if
I have offended you, that was not what I wanted to do.

Regards,

        Joey 

-- 
There are lies, statistics and benchmarks.

This archive was generated by hypermail 2b27 : Tue Jan 04 2000 - 04:16:47 CST