On Sun, Mar 04, 2007 at 09:05:52PM -0700, Brad Hodges wrote:
>> You've configured your mail system to send your outbound mail
>> through dspam, as well as incoming mail. It's actually creating a
>> dspam user for each address you send mail to, and storing token
>> data and possibly signatures etc. for each of them.
> 
> Thank you for you quick reply.  I understand what you are saying,
> but am unsure how to implement it.  Please indulge me a little bit
> to make sure I fully understand.
> 
> In my postfix master.cf file I have:
> 
> #Part A:
> 
> smtp      inet  n       -       n       -       -        smtpd
>             -o content_filter=lmtp:unix:/tmp/dspam.sock
> 
> 
> #Part B:
> 
> localhost:10026 inet  n -       n       -       -        smtpd
>   -o content_filter=
>     [...]
>   -o smtpd_recipient_restrictions=permit_mynetworks,reject
>   -o mynetworks=127.0.0.0/8
>   -o smtpd_authorized_xforward_hosts=127.0.0.0/8
> 
> This is as suggested in the postfix.txt file.  Part A being where
> mail comes in and Part B being the part where dspam re-injects
> mail back into postfix.  Are you saying that other than port 25 I
> should send my mail out on port 10026 (the same port that DSPAM
> re-injects email)?

That's pretty much it, although your :10026 will only relay for the
system itself (mynetworks) and only listens on localhost so you
would need to change the options there.

Better still would be to create another listener specifically for
accepting outbound mail from clients.

You might want to add another IP to the server to save from having
to adjust port settings in clients. Another option, if you're behind
some sort of NAT device, is to have incoming requests (from the
internet) to port 25 redirected to a different port on your server
(leaving port 25 available internally for outbound mail).

Really depends whether you find additional virtual IPs or strange
port numbers uglier.

Reply via email to