Hi,
maybe someone can help me with some architecture design.

I have some prerequisites given by our infrastructure:
* A LOT of virtual machines with installed and configured postfix relay, so our 
web applications can just use localhost to send emails.
* All those servers are managed by us. No root access for customers or 
developers.
* A relay access to a "giant company which is our main client" mailsystem. We 
use it to send mails based on sender or receiver through this network, but only 
the company related ones. Simplest way was to use transport maps.

Works awesome.

Now we got some new customers and we want to offer the same service, but we 
don't want them to use the "giant company mailsystem". Is it possible to define 
other transport maps based on the client IP address? I thought about something 
like:
* 10.234.0.0/16 uses transport maps through relay.behrens.io
* 10.212.5.0/24 uses transport maps through mybigmailcluster.example.org
* 172.19.66.4/32 does not use transport maps at all

If this is a crap setup in you opinion, enlighten me please. 

please excuse my terrible english, it's the worst when *random excuse*.


Boris

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to