Hi there, 

I appreciate everyone's trying to help, that is what I like about this
group. Here is the context of what I am trying to do. 

We are doing system integration for two software system to exchange
messages. Our industry standards requires the use of emails for message
exchange, so that a human could use an email client to exchange messages
with a software system. The software will process messages with
recipient address across many different domains. We frequently need to
add new domains and addresses. We do not use emails in traditional
sense, i.e., the messages do not got to a user's mail directory, rather,
they are processed by the software.

A straight-forward solution requires two Postfix instances. 

One Postfix (backend) receives messages from another software using
SMTP. Once it gets message and after address validation, Postfix will
call a script that forwards to us for processing. 

The other Postfix (front-end) accepts messages from an email client
using SMTP. The process is similar: once it gets message and after
address validation, Postfix will call another script that forwards to us
for processing, but the processing itself is different from the above
flow. 

Depends on where the message is from (backend or front-end), the
processing are different, that is why we cannot use one single Postfix
instance. I am trying to find out if we can accomplish this with one
single Postfix instance with two IPs (instead of two separate Postfix).

We do not want having to modify files every time a new domain is
created. That is why everything is in LDAP so that it can be done by the
software with no human intervention for adding domains and addresses.
This is why virtual domains do not apply to us, because it seems to
involve several file modification every time a domain is created. 

The following is what I have tried. If you have idea on either of the
following, that would be what I need.

All our messages are delivered by the local delivery agent, since all
domains and addresses are defined in our LDAP. Relay and virtual
delivery do not apply to us. One Postfix instance can only define one
single local delivery agent. I do not think I can define two local
delivery agents, depends on where the message is from. Because only
SMTPD has knowledge of IP, but the queue manager does not.  If there is
a way to define two local delivery agents depends on incoming IP, I
would love to hear it. 

I also tried content-filter approach as some suggested. But the problem
is that SMTPD is handing over all messages to my script, now I lose the
ability for Postfix to reject any invalid address. This is because
content filter gets called before address validation happens. If there
is a way to have address validation first, and then content filter
follows, I would love to hear it, too.

Thanks, 
Yan





Confidentiality Notice: The information contained in this electronic 
transmission is confidential and may be legally privileged. It is intended only 
for the addressee(s) named above. If you are not an intended recipient, be 
aware that any disclosure, copying, distribution or use of the information 
contained in this transmission is prohibited and may be unlawful. If you have 
received this transmission in error, please notify us by telephone (513) 
229-5500 or by email (postmas...@medplus.com). After replying, please erase it 
from your computer system.

Reply via email to