smtpd_recipient_restrictions =
       reject_invalid_hostname,
       reject_non_fqdn_hostname,
       reject_non_fqdn_sender,
       reject_non_fqdn_recipient,
       reject_unknown_sender_domain,
       reject_unknown_recipient_domain,
       reject_unauth_pipelining,
       reject_rbl_client sbl-xbl.spamhaus.org,
       reject_rbl_client zombie.dnsbl.sorbs.net,
       reject_rbl_client relays.ordb.org,
       reject_rbl_client opm.blitzed.org,
       reject_rbl_client list.dsbl.org,
       reject_rbl_client blackholes.mail-abuse.org,
       reject_rbl_client relays.mail-abuse.org,
       permit_mynetworks,
       permit_sasl_authenticated,
       reject_unauth_destination,
       check_helo_access
               hash:/etc/postfix/helo_access,
       check_sender_access
               hash:/etc/postfix/sender_access,
       check_recipient_access
               hash:/etc/postfix/recipient_access,
       permit

On 5/3/06, Marc Dirix <[EMAIL PROTECTED]> wrote:
But it seems to have one big flaw, postfix.

Spam prevention is only done after a message has been received, while
sendmail milters can reject a message during negotation with the sender.

Did you solve this in some way or the other?

/Marc

On Wed, May 03, 2006 at 01:38:42PM +0200, Matija Grabnar wrote:
> Marc Dirix wrote:
>
> >What are your experiences? How scalable is it?
> >
> >
> Postfix is _very_ scalable. In my previous job I managed servers which
> pushed milions of messages per day through amavis/spamassassin/clamav.
> It was all postfix based (this after our mail guru did a thorough
> paralel testing of sendmai, qmail, postfix and other candidates that I
> no longer remember).
>
> I seem to remember postfix beat sendmail by an order of magnitude in
> throughput.
_______________________________________________
Dbmail mailing list
[email protected]
https://mailman.fastxs.nl/mailman/listinfo/dbmail



--

Demi

Reply via email to